GENETICALLY MODIFIED (R)-LACTIC ACID PRODUCING THERMOPHILIC BACTERIA

Abstract

The invention relates to a genetically engineered thermophilic bacterial cell that is facultative anaerobic comprising: a) inactivation or deletion of the endogenous (S)-lactate dehydrogenase gene; b) introduction of a (R)-lactate dehydrogenase gene; c) inactivation or deletion of the endogenous pyruvate formate lyase A and/or B gene.

Claims

1. A genetically engineered thermophilic bacterial cell that is facultative anaerobic comprising: a) inactivation or deletion of the endogenous (S)-lactate dehydrogenase gene; b) introduction of a (R)-lactate dehydrogenase gene; c) inactivation or deletion of the endogenous pyruvate formate lyase A and/or B gene.

2. The cell according to claim 1 wherein in addition the endogenous methylglyoxal synthase gene mgsA is inactivated or deleted.

3. The cell according to claim 1 wherein the (R)-lactate dehydrogenase is the hdhD gene from Lactobacillus delbrueckii encoding the amino acid sequence of SEQ ID NO:38 or an amino acid sequence having at least 90% identity.

4. The cell according to claim 1 wherein the (R)-lactate dehydrogenase is the ldhA gene from Lactobacillus delbrueckii encoding the amino acid sequence of SEQ ID NO:36 or an amino acid sequence having at least 90% identity.

5. The cell according to claim 3 wherein the hdhD gene encodes the amino acid sequence of SEQ ID NO:38.

6. The cell according to claim 4 wherein the ldhA gene encodes the amino acid sequence of SEQ ID NO:36.

7. The cell according to claim 1 wherein in addition the endogenous phosphotransacetylase gene (pta) is inactivated or deleted.

8. The cell according to claim 1 which is a sporulation deficient derivative due to inactivation or deletion of an endogenous sporulation gene.

9. The cell according to claim 8 wherein the sporulation gene is sigF.

10. The cell according to claim 1 wherein the pyruvate formate lyase A and/or B gene is inactivated by inactivation or deletion of the endogenous pyruvate formate lyase/alcohol dehydrogenase locus pflBA-adhE.

11. The cell according to claim 1 which produces (R)-lactic acid with an enantiomeric purity of at least 98%.

12. The bacterial cell according to claim 1 wherein the genes are modified by homologous recombination.

13. The cell according to claim 1 which is a gram positive bacterial cell.

14. The cell according to claim 13 which belongs to the genus Geobacillus.

15. The cell according to claim 14 wherein the Geobacillus species is Geobacillus thermoglucosidans.

16. A method to produce enantiomeric pure (R)-lactic acid, said method comprising culturing a thermophilic bacterial cell according to claim 1 using suitable fermentable carbon containing feedstock and isolating the (R)-lactic acid.

17. The method according to claim 16 wherein the carbon containing feedstock comprises xylose, glucose or sucrose.

18. The method according to claim 16 wherein the culturing is performed at a temperature of between 50° C. and 70° C.

19. The method according to claim 16 wherein no more than 15% (w/w) of by-products are formed, based on the total weight of by-products over the total weight of lactic acid produced.

20. The method according to claim 16 wherein the formed amount of at least one of formic acid, ethanol and acetic acid is no more than 5% (w/w), based on the total weight of formic acid, ethanol or acetic acid over the total weight of lactic acid produced.

Description

EXAMPLES

Materials and Methods

Strains and Plasmids

[0066] Strains and plasmids used in this study are listed in Table 1.

[0067] Escherichia coli was routinely cultured in LB broth (Sambrook & Russell, 2001, Molecular Cloning, a laboratory manual. 3rd edition. Cold Spring Harbor Laboratory Press, New York) at 37° C. under aerobic conditions. When appropriate chloramphenicol and/or ampicillin were used at concentrations of 20 mg/L and 100 mg/L, respectively.

[0068] L. lactis was routinely cultured in M17 Broth® (BD Biosciences) supplemented with 0.5% glucose at 30° C. under anaerobic conditions. When appropriate chloramphenicol was used at a concentration of 5 mg/L.

[0069] G. thermoglucosidans was routinely grown in TGP medium at 52° C., 55° C. or 60° C. under aerobic conditions, unless stated otherwise. TGP medium (Taylor et al., 2008, Plasmid 60:45-52) contained 17 g/L trypton, 3 g/L soy peptone, 5 g/L NaCl, 2.5 g/L K.sub.2HPO.sub.4 at pH 7.0, and post-autoclave additions of 4 ml/L glycerol and 4 g/L Na-pyruvate. For TGP plates 10 g/L agar was used. When appropriate, the medium was supplemented with chloramphenicol (8 μg/mL).

TABLE-US-00001 TABLE 1 Strains and plasmids used in this study Strain or plasmid Relevant characteristics Source or reference Strains E. coli TG90 Plasmid-free strain Gonzy-Tréboul, G., Karmzyn-Campelli, C., Stragier, P. 1992. J. Mol. Biol. 224: 967-979 E. coli DH5α Plasmid-free strain ZymoResearch L. lactis MG1363 Plasmid-free strain Gasson, M. J. 1983. J. Bacteriol. 154: 1-9 G. thermoglucosidans G. thermoglucosidans type DSMZ, DSM 2542 strain Braunschweig G. thermoglucosidans Sporulation-deficient This work DSM 2542 ΔsigF G. thermoglucosidans G. thermoglucosidans Sporulation-deficient, (R)-lactic This work DSM 2542 ΔsigF, acid producing G. thermoglucosidans ΔldhL::hdhD G. thermoglucosidans Sporulation-deficient, chiral This work DSM 2542 ΔsigF, pure, and homolactic (R)-lactic ΔldhL::hdhD, ΔpflBA- acid producing G. thermoglucosidans ΔadhE G. thermoglucosidans Sporulation-deficient, chiral This work DSM 2542 ΔsigF, pure, and homolactic (R)-lactic ΔldhL::hdhD, ΔpflBA- acid producing G. thermoglucosidans ΔadhE, ΔmgsA G. thermoglucosidans Sporulation-deficient, homolactic This work DSM 2542 ΔsigF, (R)-lactic acid producing ΔldhL::ldhA, ΔpflBA- G. thermoglucosidans ΔadhE G. thermoglucosidans Sporulation-deficient, chiral This work DSM 2542 ΔsigF, pure, and homolactic (R)-lactic ΔldhL::ldhA, ΔpflBA- acid producing G. thermoglucosidans ΔadhE, ΔmgsA Plasmids pNW33N 4.2 kb, Cm.sup.R, E. coli/Geobacillus Bacillus Genetic shuttle vector Stock Centre pNZ124 2.8 kb, Cm.sup.R, E. coli/Gram- Platteeuw, C., G. Simons, positive shuttle vector and W. M. de Vos. 1994. Appl. Environ. Microbiol. 60: 587-593 pRM3 6.2 kb, Cm.sup.R, pNW33n derivative This work with the upstream and downstream regions of G. thermoglucosidans sigF pRM12 6.4 kb, Cm.sup.R, pNW33n derivative This work with upstream and downstream regions of G. thermoglucosidans pflBA-adhE locus pJS65 6.3 kb, Cm.sup.R, pNZ124 derivative This work with L. delbrueckii ldhA flanked by upstream and downstream regions of G. thermoglucosidans ldhL pFS3 7.9 kb, Cm.sup.R, pNW33n derivative This work with L. delbrueckii hdhD flanked by upstream and downstream regions of G. thermoglucosidans ldhL pJS43 6.4 kb, Cm.sup.R, pNW33n derivative This work with upstream and downstream regions of G. thermoglucosidans mgsA

DNA Manipulation Techniques

[0070] Standard DNA manipulation techniques were performed as described by Sambrook and Russell (Sambrook & Russell, 2001, Molecular Cloning, a laboratory manual. 3rd edition. Cold Spring Harbor Laboratory Press, New York).

[0071] Construction pNW33N derivatives was performed in E. coll.

[0072] Large-scale E. coli plasmid DNA isolation from 100 mL culture was performed using the Jetstar 2.0 Plasmid Maxiprep Kit® (Genomed) following the instructions of the manufacturer. Small-scale E. coli plasmid DNA isolation from 1 mL culture was performed using the Nucleospin Plasmid Quick Pure® (Macherey-Nagel) kit following the instructions of the manufacturer.

[0073] E. coli competent cells were prepared using calcium chloride and transformed by heat shock as described by Sambrook and Russell (Sambrook & Russell, 2001, Molecular Cloning, a laboratory manual. 3rd edition. Cold Spring Harbor Laboratory Press, New York).

[0074] Construction of pNZ124 derivatives was performed in L. lactis.

[0075] L. lactis plasmid DNA isolation from 100 mL culture was performed using the Jetstar 2.0 Plasmid Maxiprep Kit® (Genomed). The cell pellet was resuspended in 10 ml modified E1 buffer (10 mM Tris/HCL pH8; 50 mM NaCl; 10 mM EDTA; 20% sucrose; 4 g/L lysozyme (Sigma Aldrich)) and incubated at 50° C. for 1 hour. Subsequently, the instructions of the manufacturer were followed from cell lysis onwards.

[0076] L. lactis was transformed by electroporation as described by Holo and Nes (Holo, H., and I. F. Nes. 1989. Appl. Environ. Microbiol. 55:3119-3123).

[0077] PCR reactions for cloning purposes were performed with the high-fidelity Pwo polymerase (Roche) following the instructions of the manufacturer.

[0078] For colony-PCR analysis colonies were picked with a tooth pick and a little cell material was transferred to a PCR reaction tube. The cells were disrupted by 1 min incubation at 1000 W in a microwave oven. PCR reaction mixtures of 25 μL with DreamTaq™ DNA Polymerase (Thermo Scientific™) were prepared as recommended by the manufacturer and added to the reaction tubes with the disrupted cells.

Electroporation of G. thermoglucosidans

[0079] G. thermoglucosidans was transformed by electroporation, based on the protocol described by Cripps et al. (Cripps, et al., 2009, Metab. Eng. 11:398-408). G. thermoglucosidans was grown overnight at 55° C. and 1 mL was used to inoculate 50 ml pre-warmed TGP medium in a 250 ml conical flask with baffles. Cells were incubated at 60° C. (180 rpm) until the OD.sub.600 was ≈1.0. The flask was cooled on ice for 10 min. and the cells were pelleted by centrifugation (4° C.). Next, the cells were washed four times with ice cold electroporation buffer (0.5 M sorbitol, 0.5 M mannitol, 10% (v/v) glycerol). The volumes of the washing steps were 50 ml, 25 ml, 10 ml, and 10 ml. The final pellet was resuspended in 1.3 ml of ice cold electroporation buffer and 60 μl aliquots of electrocompetent cells were stored at −80° C. or directly used for electroporation.

[0080] A 60 μl aliquot of electrocompetent cells (defrosted) was mixed with 1-2 μg plasmid DNA and subsequently transferred to a chilled electroporation cuvet (gap width 0.1 cm). The electroporation conditions using a Bio-Rad gene pulser electroporator were 2.5 kV, 10 ρF and 6000. After electroporation the cells were transferred to 1 ml of pre-warmed (52° C.) TGP in a 50 ml plastic tube and recovered at 52° C., 180 rpm for two hours. The recovered cell suspension was pelleted and all but 150 μl supernatant was discarded. The pellet was resuspended in the remaining supernatant. Volumes of 1/10 and 9/10 were plated onto TGP plates containing 8 μg/L chloramphenicol. The plates were incubated at 52° C. for 24-48 hours. Colonies which appeared on the plates were transferred to a fresh TGP plate containing 8 μg/L chloramphenicol and incubated at 55° C. overnight. Those that grew were tested for the presence of the plasmid by colony PCR using primers 1 and 2 (Table 2).

Integration

[0081] The Geobacillus-E. coli shuttle vector pNW33n was used as integration vector in G. thermoglucosidans as previously described (Cripps et al., 2009, Metab. Eng. 11:398-408). 20 mL TGP containing 8 μg/mL chloramphenicol was inoculated with transformed strains from a glycerol stock. After overnight growth at 55° C., 180 rpm, appropriate dilutions were plated on TGP plates containing 8 μg/mL chloramphenicol. These plates were then incubated at 68° C. for 24 h. Single colonies were streaked to a fresh plate (incubated at 52° C.) and a colony PCR was conducted on these colonies to identify a colony with a single crossover. The appropriate primer combinations were used to identify single crossovers via the upstream or downstream fragment (Table 2; primer combinations 655-170 and 656-571 for integration of pRM3, primer combinations 744-170 and 808-571 for integration of pRM12, primer combinations 629-170 and 630-571 for integration of pFS3, primer combinations 754-170 and 991-571 for integration of pJS43, respectively). Next, chromosomal DNA of positive colonies was isolated using the Masterpure Gram Positive DNA Purification Kit (Epicentre Biotechnologies) and to confirm the results of the colony PCR, the PCR described above was repeated on the isolated chromosomal DNA. A single crossover via the upstream flanking region and a single crossover via the downstream flanking region were selected for the second recombination step.

[0082] To obtain a double crossover, the primary integrants were sub-cultured several times in TGP without chloramphenicol. Appropriate dilutions (10.sup.−4, 10.sup.−5, 10.sup.−6) were plated on TGP plates. Isolated colonies were transferred to a TGP plate with and one without 8 μg/mL chloramphenicol. Double crossover mutants are chloramphenicol sensitive. PCR analysis using the appropriate primer combinations (Table 2; primer combinations 655-656 for ΔsigF, 744-808 for ΔpflBA-adhE, and 754-991 for ΔmgsA) was used to discriminate wild-type from deletion mutants and to verify the absence of the plasmid. All modifications were confirmed by sequencing of the PCR products.

[0083] The Lactococcus cloning vector pNZ124 was used as integration vector in G. thermoglucosidans for ldhA. Freshly prepared G. thermoglucosidans competent cells with relatively high transformation efficiency (at least 10.sup.3 CFU/μg pNW33n) were transformed with 2 μg of pJS65 plasmid DNA. The transformation plates were incubated 16 hours at 60° C., 8 hours at 68° C. and 20 hours at 55° C. Single colonies were streaked to a fresh plate (incubated at 52° C.) and colony PCR was conducted on these colonies to identify a colony with a single crossover. The appropriate primer combinations were used to identify single crossovers via the upstream or downstream fragment (Table 2; primer combinations 1539-205 and 204-630).

TABLE-US-00002 TABLE 2 Primers used in this study SEQ. Primer ID No ID Sequence (5′-3′).sup.1  1    1 TCGCCTTCTTCTGTGTCATC  2    2 CTGGAGGAGAGCAATGAAAC  3  651 GCGCGGGTACCCAGCAAACCGAGCGGAATCAG  4  652 GCGCGGTCGACGGATGGGTAGGCATCCATTC  5  653 GCGCGGTCGACGTCTCCCTTAGTTACATAACGC  6  654 GCGCGAAGCTTGCTTCGCAGTCCAATCGTCGC  7  739 GCGCGGGATCCCCCAAATGGCATTACCGGTGTG  8  805 TGTTATTGCTGGCAGTTTCCCTCCCATGCATCTG  9  806 GGAGGGAAACTGCCAGCAATAACACCAACAGGCTC 10  807 GCGCGCTGCAGCGAAAGCGAACGAAATTGCCAAC 11  624 GCGCGGTCGACCTGACTTTGAATACAACAAGGTGA AC 12  631 GCGCGGCATGCCGGCAAACAGAGCTTTAAAACCAG GC 13 1200 CCCGCATGCTTAGCCAACCTTAACTGGAGTTTCAG 14  676 TTTAGTCATCGCTGTCTGTCATCCTTTCC 15  675 GATGACAGACAGCGATGACTAAAATTTTTGCTTAC GCAATTCG 16  564 GCGCGGTCGACTTAGCCAACCTTAACTGGAGTTTC AG 17 1057 GCGCGGGATCCCTCGTTGTATTTGGGCATACGTCG 18 1203 CTGACATTATACATGGCAATTTTAGTCATCGCTGT CTGTCATCCTTTCC 19 1202 GGAAAGGATGACAGACAGCGATGACTAAAATTGCC ATGTATAATGTCAG 20 1189 CGGCTCGAGTTACAGGTTAACGATGCTTCTTGGC 21  750 GCGCGGGATCCGCTTTCCGTTTGCCATTTGCCG 22  999 TATGCGACGGGCGCGTGGAGGAATATTGTCCGC 23 1000 ATTCCTCCACGCGCCCGTCGCATACAGTTCATGTT G 24  753 GCGCGCTGCAGGGCAAGACTGACAGAAGAGCTTGG 25  170 GCCCTCGAGAGGGCTCGCCTTTGGGAAG 26  571 GCTCGTTATAGTCGATCGGTTC 27  655 GCTAAGATCGGCCATACGTTAAGC 28  656 GGAGACGAGCTTGGCGTCCTG 29  744 GCCAAGATGGATATGGGCGTTAGC 30  808 CCGGAGATGGACGGAATTGAAG 31  629 GACTGGGCGCAAGCGGTGATG 32  630 CCTGTTGCTGATACAAGGTCTAGC 33  754 CAGCAGTAACGGCATCCGATTG 34  991 GCGGATATGATTGAATTTGTGACTGCC 53  204 CTGCAAGCTTTGGCAGACAACGGCATCAC 54  205 TTGCGTAACCGAAGACCTTGCCTGAGTCC 55  957 CCTCGAGCGGCAAACAGAGCTTTAAAACCAGGC 56 1537 GGGTCTAGAGCCGCTTCGTTTTCCAACTGATGC 57 1539 TCTTTCGCTTCCAGGGCTGTTC 58 1589 GCGCGGAGCTCGTCGACCTGACTTTGAATACAACA AGGTGAAC .sup.1Restriction sites are underlined

Fermentation

[0084] TMM medium was modified from Fong et al. (Fong et al., 2006) and contained per L: 60 g/L glucose; 30 g/L xylose; 8.37 g MOPS, 0.23 g K.sub.2HPO.sub.4; 0.51 g NH.sub.4Cl; 0.50 g NaCl; 1.47 g Na.sub.2SO.sub.4; 0.08 g NaHCO.sub.3; 0.25 g KCl; 1.87 g MgCl.sub.2.6H.sub.2O; 0.41 g CaCl.sub.2.2H.sub.2O; 16.0 mg MnCl.sub.2.4H.sub.2O; 1.0 mg ZnSO.sub.4.7H.sub.2O; 2.0 mg H.sub.3BO.sub.3; 0.1 mg CuSO.sub.4.5H.sub.2O; 0.1 mg Na.sub.2MoO.sub.4.2H.sub.2O; 1.0 mg CoCl.sub.2.6H.sub.2O; 7.0 mg FeSO.sub.4.7H.sub.2O; 0.1 mg thiamine; 0.1 mg riboflavin; 0.5 mg nicotinic acid; 0.1 mg panthothenic acid; 0.5 mg pyridoxamine, HCl; 0.5 mg pyridoxal, HCl; 0.1 mg D-biotin; 0.1 mg folic acid; 0.1 mg p-aminobenzoic acid; 0.1 mg cobalamin. pH was adjusted to pH 7.2. Glucose, xylose, metals and vitamins were filter sterilized. Medium was autoclaved. TMM1, TMM2.5, and TMMS were supplemented with 1 g/L, 2.5 g/L, and 5 g/L yeast extract (Oxoid), respectively.

[0085] STMM, differed from TMM in concentrations of K.sub.2HPO.sub.4 (1.00 g/L), NH.sub.4Cl (2.50 g/L), NaCl (5.00 g/L), and CaCl.sub.2.2H.sub.2O (50 mg/L) and was supplemented with D,L-methionine (68.5 mg/L) and betaine (0.14 g/L). Sucrose (90 g/L) was used instead of glucose and xylose. STMMS was supplemented with 5 g/L yeast extract (Biospringer).

[0086] A 100 mL preculture in TMMS or STMMS was used to inoculate (10% v/v) 400 mL TMM1, TMM2.5, or STMMS in a 0.75 L Multifors fermentor (Infors) equipped with a condenser (cooled with running tap water of approximately 15° C.). The pH was controlled at pH 7.2 by addition of sterile 2.5 M KOH or sterile 75 g/L Ca(OH).sub.2. Temperature was 60° C. Stirrer speed was 300 rpm

[0087] Samples were withdrawn from the fermentation for measurement of (R)- and (S)-lactic acid, and possible by-products. Samples were centrifuged and remaining debris was removed by filtration using a Millex GP 0.22 μm Filter® (Millipore). Filtrate was stored at −21° C. until further analysis.

[0088] Sugars were measured by HPLC using a Thermo CarboPac SA-10 column (Dionex). Formic acid was measured by HPLC using a Bio-Rad Aminex HPX-87C column (Bio-Rad). Other organic acids (lactic acid, acetic acid, succinic acid, fumaric acid, pyruvic acid) and ethanol were measured using a derivatisation and gas-liquid chromatography (GLC). (R)- and (S)-lactates were methylated to methyl-lactate and measured by headspace analysis on a chiral column.

Example 1

[0089] Homolactic Lactic Acid Production with G. thermoglucosidans

[0090] Integration plasmid pRM3 was constructed to delete the sigF gene in G. thermoglucosidans. The upstream and downstream flanking regions of the sigF gene were generated by PCR using genomic DNA of DSM 2542 as template and primer combinations 653 and 654 (Table 2) to obtain the upstream fragment, and the primers 651 and 652 (Table 2) to obtain the downstream fragment. First, the downstream fragment was cloned as KpnI-SalI fragment into pNW33n, digested with the same enzymes. Next, the upstream fragment was cloned as SalI-HindIII fragment into this construct, digested with the same enzymes resulting in plasmid pRM3. Construction of pRM3 was done in E. coli TG90. The integrity of the pRM3 sequence was confirmed by DNA sequencing.

[0091] Plasmid pRM3 was electroporated to G. thermoglucosidans DSM 2542. A single transformant colony was selected and used to obtain single crossover mutants as described in Materials and Methods. Two colonies were selected for further work, one with a single crossover via the upstream flanking region and one with a single crossover via the downstream flanking region.

[0092] A double crossover mutant was obtained following the procedure described in Materials and Methods. Sixty colonies, obtained after subculturing of the single crossover integrants in TGP without chloramphenicol, were transferred to TGP plates with and without chloramphenicol. Fifteen colonies were sensitive to chloramphenicol. Twelve colonies had the desired modification and three had reverted to wild-type. One colony was selected and designated G. thermoglucosidans DSM 2542 ΔsigF. The deletion was confirmed by sequencing.

TABLE-US-00003 TABLE 3 Fermentations with G. thermoglucosidans DSM 2542 ΔsigF a glucose/xylose mixture on TMM2.5. Chiral purity Total (S)- lactic lactic Acetic Formic Time Glucose Xylose acid acid acid acid Ethanol (h) (g/L) (g/L) (g/kg) (%) (g/kg) (g/kg) (g/kg) 24 18.5 11.4 29 89.5 <0.1 1.2 2.2 48 15.2 7.0 33 89.4 <0.1 1.2 2.2

[0093] G. thermoglucosidans DSM 2542 ΔsigF was evaluated in pH-controlled (KOH) fermentation using TMM2.5. Fermentations were transferred 4 times and the final fermentations were analysed. The results are summarized in Table 3. G. thermoglucosidans DSM 2542 ΔsigF consumed xylose and glucose simultaneously.

Example 2

[0094] Construction of (R)-Lactic Acid-Producing G. thermoglucosidans Derivative Using hdhD.

[0095] Plasmid pFS3 was constructed to facilitate the gene replacement of the native ldhL gene with the hdhD gene originating from L. delbrueckii and encoding D-lactate dehydrogenase activity. Construction was such that hdhD start and stop codons replace the positions of the original ldhL start and stop codons and result in a translational fusion of hdhD to the ldhL promoter. The downstream flanking region of the ldhL gene was generated by PCR using genomic DNA of DSM 2542 as template and primer combination 624 and 631. The product was digested with SalI and SphI and ligated into pNW33n digested with SalI and SphI. The resulting plasmid was designated pFS2. Construction of pFS2 was done in E. coli DH5a.

[0096] The upstream flanking region of the ldhL gene was generated by PCR using genomic DNA of DSM 2542 as template and primer combination 1057 and 1203. The hdhD gene (SEQ ID NO: 37) was generated by PCR using L. delbrueckii genomic DNA as template and primer combination 1202 and 1189. The gene can also be synthesized based on SEQ ID NO: 37. The resulting two PCR-products are subsequently used as template in an overlap-PCR using primer combination 1057 and 1189 to fuse them together. The product was digested with BamHI and XhoI and ligated in pFS2 digested with BamHI and SalI. The resulting plasmid was designated pFS3. Construction of pFS3 was done in E. coli TG90. Integrity of the pFS3 nucleotide sequence was confirmed by sequencing.

[0097] Plasmid pFS3 was electroporated to G. thermoglucosidans DSM 2542 ΔsigF. A single transformant colony was selected and used to obtain single crossover mutants as described in Materials and Methods. In the number of colonies tested only single crossover mutants via the downstream flanking region were obtained. One of these was selected for further work.

[0098] A double crossover mutant was obtained following the procedure described in Materials and Methods. Colonies, obtained after subculturing of the single crossover integrant in TGP without chloramphenicol, were transferred to TGP plates with and without chloramphenicol. Seventeen colonies sensitive to chloramphenicol were checked by PCR. One colony had the desired modification and 16 had reverted to wild-type. The colony having ldhL exchanged by hdhD was selected and designated G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL::hdhD.

[0099] To further optimize G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL::hdhD there was a wish to eliminate formic acid, acetic acid, and ethanol byproduct formation. Although mutations of pflA and/or pflB and adhE are known to impact formic acid and ethanol production in many bacteria, the side effects of disrupting those genes are unpredictable.

[0100] In order to evaluate the effect of the disruption of these genes, a plasmid (pRM12) was constructed to delete the genes pflB, pflA and adhE (partially) in G. thermoglucosidans. The upstream flanking region of pflBA and the upstream flanking region of the convergently oriented adhE were generated by PCR using genomic DNA of DSM 2542 as template and primer combinations 739 and 805 to obtain the upstream pflBA fragment and the primers 806 and 807 to acquire the upstream adhE fragment. The resulting two PCR-products were subsequently used as template in an overlap-PCR using primer combination 739 and 807 to fuse them together. The product was cloned as BamHI-PstI fragment into plasmid pNW33n digested with BamHI and PstI, resulting in plasmid pRM12. Construction of pRM12 was done in E. coli TG90. Integrity of the pRM12 nucleotide sequence was confirmed by sequencing.

[0101] Plasmid pRM12 was electroporated to G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL::hdhD. A single transformant colony was selected and used to obtain single crossover mutants as described in Materials and Methods. Two colonies were selected for further work, one with a single crossover via the upstream pflBA flanking region and one with a single crossover via the upstream adhE flanking region.

[0102] A double crossover mutant was obtained following the procedure described in Materials and Methods. 120 colonies, obtained after subculturing of the single crossover integrants in TGP without chloramphenicol, were transferred to TGP plates with and without chloramphenicol. Five colonies sensitive to chloramphenicol were checked by PCR. Two colonies had the desired modification and three had reverted to wild-type. One colony with the desired modification was selected and was designated G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL::hdhD, ΔpflBA-ΔadhE.

[0103] G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL::hdhD, ΔpflBA-ΔadhE was evaluated in pH-controlled (Ca(OH).sub.2) fermentations using STMMS medium containing 5 g/L yeast extract and 90 g/L sucrose. Fermentations were transferred and the second fermentation was analysed for the production of homolactic (R)-lactic acid. G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL::hdhD, ΔpflBA-ΔadhE was able to produce homolactic (R)-lactic acid with a limited amount of ethanol, formic acid, and acetic acid by-products. Thus, introduction of the HdhD D-lactate dehydrogenase in combination with disruption of the pyruvate-formate lyase and alcohol dehydrogenase complex genes results in a homolactic (R)-lactic acid fermentation at 60° C. with a limited amount of ethanol, formic acid, and acetic acid by-products.

Example 3

[0104] Construction of (R)-Lactic Acid-Producing G. thermoglucosidans Derivative Using ldhA.

[0105] Cloning of ldhA genes originating from Lactobacillus species in E. coli is known to be problematic (Bernard et al., 1991. FEBS Lett. 290:61-64). To circumvent possible cloning issues we decided to use L. lactis as intermediate host and pNZ124 as cloning vector. Plasmid pJS65 was constructed to facilitate the gene replacement of the native ldhL gene with the ldhA originating from L. delbrueckii and encoding D-lactate dehydrogenase activity.

[0106] Construction was such that ldhA start and stop codons replace the positions of the original ldhL start and stop codons and result in a translational fusion of ldhA to the ldhL promoter.

[0107] The downstream flanking region of the ldhL gene is generated by PCR using genomic DNA of DSM 2542 as template and primer combination 1589 and 957. The product is digested with SacI and XhoI and ligated into pNZ124 digested with SacI and XhoI. The resulting plasmid is designated pJS64. Construction of pJS64 is done in L. lactis MG1363.

[0108] The upstream flanking region of the ldhL gene is generated by PCR using genomic DNA of DSM 2542 as template and primer combination 1537 and 676. The ldhA gene (SEQ ID NO: 35) is generated by PCR using L. delbrueckii genomic DNA as template and primer combination 675 and 564. The gene can also be synthesized based on SEQ ID NO: 35 taking into account that the synthetic gene should, preferably, be cloned in pNZ124 and in L. lactis. The resulting two PCR-products are subsequently used as template in an overlap-PCR using primer combination 1537 and 564 to fuse them together. The product is digested with XbaI and SalI and ligated in pJS64 digested with XbaI and partially digested with SalI. The resulting plasmid is designated pJS65. Construction of pJS65 is done in L. lactis MG1363.

[0109] Plasmid pJS65 was electroporated to G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL::hdhD, ΔpflBA-ΔadhE for direct integration. A single transformant colony was obtained with a single crossover via the upstream ldhL flanking region. Achieving direct integration in the G. thermoglucosidans genome required using freshly prepared competent cells with a relatively high transformation efficiency of at least 10.sup.3 CFU/μg pNW33n.

[0110] A double crossover mutant was obtained following the procedure described in Materials and Methods. 240 colonies, obtained after subculturing of the single crossover integrants in TGP without chloramphenicol, were transferred to TGP plates with and without chloramphenicol. Thirteen colonies sensitive to chloramphenicol were checked by PCR. Ten colonies had the desired modification and three had reverted to wild-type. One colony having ldhL exchanged by ldhA was selected and designated G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL:ldhA, ΔpflBA-ΔadhE.

TABLE-US-00004 TABLE 4 Fermentation with G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL::ldhA, ΔpflBA-ΔadhE on STMM5 Chiral Total purity Acetic Formic Sucrose lactic acid (R)-lactic acid acid Ethanol Time (h) (g/kg) (g/kg) acid (%) (g/kg) (g/kg) (g/kg) 0 72.1 3.3 98.9 0.2 <0.05 <0.2 24 22.0 36 98.9 0.4 <0.05 <0.2 48 1.6 48 98.9 0.5 <0.05 <0.2

[0111] G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL:ldhA, ΔpflBA-ΔadhE was evaluated in pH-controlled (Ca(OH).sub.2) fermentations using STMM5 medium containing 5 g/L yeast extract and 90 g/L sucrose. Fermentations were transferred and the second fermentation was analysed. The results are summarized in Table 4. These data clearly demonstrate that introduction of the LdhA D-lactate dehydrogenase in combination with disruption of the pyruvate-formate lyase and alcohol dehydrogenase complex genes results in a homolactic (R)-lactic acid fermentation with a limited amount of ethanol, formic acid, and acetic acid by-products.

Example 4

[0112] Enantiopure Homolactic Acid Production with G. thermoglucosidans

[0113] Plasmid pJS43 was constructed to delete 267 bp of the mgsA gene (423 bp) in G. thermoglucosidans. The upstream and downstream flanking regions of the mgsA gene were generated by PCR using genomic DNA of DSM 2542 as template and primer combinations 750 and 999 to obtain the mgsA downstream fragment, and the primers 1000 and 753 to acquire the upstream mgsA fragment. The resulting two PCR-products were subsequently used as template in an overlap-PCR using primer combination 750 and 753 to fuse them together. The product was cloned as BamHI-PstI fragment into plasmid pNW33n digested with BamHI and PstI, resulting in plasmid pJS43. Construction of pJS43 was done in E. coli TG90. Integrity of the pJS43 nucleotide sequence was confirmed by sequencing.

[0114] Plasmid pJS43 was electroporated to G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL::hdhD, ΔpflBA-ΔadhE. Single transformant colonies were selected and used to obtain single crossover mutants as described in Materials and Methods. Two colonies were selected for further work, one with a single-crossover via the upstream flanking region and one with a single-crossover via the downstream flanking region.

[0115] Double crossover mutants were obtained following the procedure described in Materials and Methods. 400 colonies, obtained after subculturing of the single crossover integrants in TGP without chloramphenicol, were transferred to TGP plates with and without chloramphenicol. Of these 213 colonies were sensitive to chloramphenicol. 39 of the chloramphenicol-sensitive colonies were checked by PCR for double crossovers. Eight colonies had the desired modification and 31 had reverted to wild-type. A single colony with the desired modification was selected and designated G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL::hdhD, ΔpflBA-ΔadhE, ΔmgsA. The deletion was confirmed by sequencing.

[0116] Plasmid pJS43 was electroporated to G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL:ldhA, ΔpflBA-ΔadhE. Single transformant colonies were selected and used to obtain single crossover mutants as described in Materials and Methods. Two colonies were selected for further work, both with a single-crossover via the upstream flanking region. Single-crossovers via the downstream flanking region were not obtained.

[0117] Double crossover mutants were obtained following the procedure described in Materials and Methods. 240 colonies, obtained after subculturing of the single crossover integrants in TGP without chloramphenicol, were transferred to TGP plates with and without chloramphenicol. 239 colonies were sensitive to chloramphenicol, of which 134 colonies were checked by PCR for double crossovers. One had the desired modification and 133 reverted back to wild-type. The single colony with the desired modification was selected and designated G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL:ldhA, ΔpflBA-ΔadhE, ΔmgsA. The deletion was confirmed by sequencing.

TABLE-US-00005 TABLE 5 Fermentation with G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL::ldhA, ΔpflBA-ΔadhE, ΔmgsA on STMM5 Chiral Total purity Acetic Formic Sucrose lactic acid (R)-lactic acid acid Ethanol Time (h) (g/kg) (g/kg) acid (%) (g/kg) (g/kg) (g/kg) 0 74.3 2.3 >99.1 0.2 <0.05 <0.2 24 43.9 24 99.9 0.2 <0.05 <0.2 48 28.2 32 99.8 0.3 <0.05 <0.2

[0118] G. thermoglucosidans DSM 2542 ΔsigF, ΔldhL:ldhA, ΔpflBA-ΔadhE, ΔmgsA was evaluated in pH-controlled (Ca(OH).sub.2) fermentations using STMM5 medium containing 5 g/L yeast extract and 90 g/L sucrose. Fermentations were transferred and the second fermentation was analysed. The results are summarized in Table 5. Chiral purity of the (R)-lactic acid produced was >99.0% for low concentrations of lactic acid (<5 g/kg) and >99.7 for higher concentrations of lactic acid (>20 g/kg), which is more pure than lactic acid from strains without disruption of mgsA (Table 5). These data clearly show that despite the apparent incompleteness of the methylglyoxal pathway in G. thermoglucosidans, disruption of mgsA results in the ability to produce chiral pure (R)-lactic acid resulting in a homolactic and chiral pure (R)-lactic acid fermentation.