MICROORGANISMS AND PROCESS FOR PRODUCING GLYCOLIC ACID FROM PENTOSES AND HEXOSES

Abstract

The present invention relates to a recombinant microorganism which exhibits i) a conversion activity from D-ribulose-5-phosphate into D-arabinose-5-phosphate, increased in comparison with the same, non-modified microorganism; ii) a cleavage catalysis activity from D-arabinose-5-phosphate into D-glyceraldehyde-3-phosphate and glycolaldehyde, increased in comparison with the same, non-modified microorganism; iii) an oxidation activity from glycolaldehyde into glycolate, increased in comparison with the same, non-modified microorganism; and iv) an oxidation activity from glyceraldehyde-3-phosphate into 1,3-bisphosphoglycerate, decreased in comparison with the same, non-modified microorganism. The present invention also relates to a process for preparing glycolic acid from pentoses and/or hexoses, using such a recombinant microorganism. The present invention also relates to a process for producing glycolic acid involving a biomass production phase and a bioconversion phase from hexoses and/or pentoses into glycolic acid.

Claims

1) A recombinant microorganism which exhibits i) a conversion activity from D-ribulose-5-phosphate into D-arabinose-5-phosphate, increased in comparison with the same, non-modified microorganism; ii) an aldolic cleavage activity from D-arabinose-5-phosphate into D-glyceraldehyde-3-phosphate and glycolaldehyde, increased in comparison with the same, non-modified microorganism; iii) an oxidation activity from glycolaldehyde into glycolate, increased in comparison with the same, non-modified microorganism; and iv) an oxidation activity from glyceraldehyde-3-phosphate into 1,3-bisphosphoglycerate, decreased in comparison with the same, non-modified microorganism, said recombinant microorganism producing glycolic acid from pentoses and hexoses.

2) The recombinant microorganism according to claim 1, characterised in that said microorganism exhibits an overexpression of the E. coli kdsD gene or a homologue thereof.

3) The recombinant microorganism according to claim 1 or 2, characterised in that said microorganism exhibits an overexpression of the E. coli fsa gene or a homologue thereof.

4) The recombinant microorganism according to any of claims 1 to 3, characterised in that said microorganism exhibits an overexpression of the E. coli aldA gene or a homologue thereof.

5) The recombinant microorganism according to any of the preceding claims, characterised in that said microorganism comprises: α) a first plasmid in which the sequence of the E. coli kdsD gene or a homologue thereof and the sequence of the E. coli fsa gene or a homologue thereof lie, said sequences being cloned as an operon and under the control of a first inducible or constitutive promoter and β) a second plasmid in which the sequence of the E. coli aldA gene or a homologue thereof under the control of a second inducible or constitutive promoter lies, said first and second promoters being identical or different.

6) The recombinant microorganism according to any of the preceding claims, characterised in that the expression of the E. coli gapA gene or a homologue thereof is decreased but not inactivated with respect to the non-modified microorganism.

7) The recombinant microorganism according to any of claims 1 to 5, characterised in that the expression of the E. coli gapA gene or a homologue thereof is inactivated with respect to the non-modified microorganism.

8) The recombinant microorganism according to any of the preceding claims, characterised in that the phosphotransferase system (PTS), which depends on phosphoenolpyruvate (PEP) is inactivated, whereas a glucose transport activity coded by E. coli galP or Zymomonas mobilis g/f or a homologue thereof and a transformation activity from glucose into glucose-6-phosphase are increased in comparison with the same, non-modified microorganism.

9) The recombinant microorganism according to any of the preceding claims, characterised in that said microorganism exhibits at least one of the following characteristics: v) an oxidation activity from glycolate into glyoxylate, decreased in comparison with the same, non-modified microorganism; vi) a repression of the genes involved in regulating the aerobic respiratory metabolism, decreased in comparison with the same, non-modified microorganism; vii) a glycolate internalisation, decreased in comparison with the same, non-modified microorganism; viii) an irreversible formation activity of methylglyoxal from dihydroxyacetone, decreased in comparison with the same, non-modified microorganism; ix) a conversion activity from fructose-6-phosphate into fructose-1,6-biphosphate, decreased in comparison with the same, non-modified microorganism; x) a production activity of D-ribose-1-phosphate from dihydroxyacetone phosphate and glycolaldehyde, decreased in comparison with the same, non-modified microorganism; and xi) an oxidation activity from D-glucose-6-phosphate into 6-phospho D-glucono-1,5-lactone, modified in comparison with the same, non-modified microorganism.

10) The recombinant microorganism according to claim 9, characterised in that said production activity of 6-phospho-D-glucono-1,5-lactone from D-glucose-6-phosphate is decreased in comparison with the same, non-modified microorganism.

11) The recombinant microorganism according to claim 9, characterised in that said recombinant microorganism exhibits the following characteristics: xi) an oxidation activity from D-glucose-6-phosphate into 6-phospho D-glucono-1,5-lactone, increased in comparison with the same, non-modified microorganism; xii) a formation activity of 2-dehydroxy-3-deoxy-D-gluconate-6-phosphate from D-gluconate-6-phosphate, decreased in comparison with the same, non-modified microorganism; and xiii) a formation activity of glyceraldehyde-3-phosphate and pyruvate from 2 dehydroxy-3-deoxy-D-gluconate-6-phosphate, decreased in comparison with the same, non-modified microorganism.

12) A process for producing glycolic acid comprising the steps of: a) culturing a recombinant microorganism as defined in any of the preceding claims in a culture medium comprising, as a carbon source, at least one pentose and/or at least one hexose; and b) recovering glycolic acid from the microorganism and/or in the culture medium.

13) The process according to claim 12, characterised in that said cultured recombinant microorganism is a recombinant microorganism as defined in claim 6 and in that the implemented carbon source only comprises one element chosen from D-glucose, D-xylose, L-arabinose and a mixture thereof.

14) The process according to claim 12, characterised in that said cultured recombinant microorganism is a recombinant microorganism as defined in claim 7 and in that the carbon source comprises, in addition to D-xylose and/or L-arabinose and/or D-glucose, one or more C2, C3 and C4 compounds chosen from malate, pyruvate, succinate, acetate and a mixture thereof.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0234] FIG. 1 already mentioned shows the three key reactions for glycolic acid production, isolated from the E. coli central carbon metabolism.

[0235] FIG. 2 already mentioned shows the non-natural pathway for glycolic acid production from D-glucose, D-xylose and L-arabinose.

[0236] FIG. 3 is a schematic representation of the enzymatic test for checking arabinose-5P isomerase (KdsD), fructose-6P aldolase (FSA) and aldehyde dehydrogenase (aldA) activity. KdsD, FSA and aldA have been purified, triose phosphate isomerase (Tpi) and glycerol-3P dehydrogenase (G3PDH) have been ordered to Sigma.

[0237] FIG. 4 shows the system comprising 7 purified enzymes (AraA, AraB, AraD, RPE, KdsD, FSA, AldA) which catalyse the conversion of L-arabinose into glycolic acid (FIG. 4A). The glycolic acid production with this system in the presence of L-arabinose has been compared with that of a system not having AraA; the enzymatic reaction is based on the NADH assay at 340 nm (FIG. 4B).

[0238] FIG. 5 shows the system comprising 6 purified enzymes (XylA, XylB, RPE, KdsD, FSA, AldA) which are capable of converting D-xylose into glycolic acid (FIG. 5A). The glycolic acid production with this system in the presence of D-xylose has been compared with that with a system not having XylA; the enzymatic reaction is based on the NADH assay at 340 nm (FIG. 5B).

[0239] FIG. 6 shows the system comprising 7 purified enzymes (Hxk, Pgi, Tkt, RPE, KdsD, FSA, AldA) which are capable of converting D-glucose into glycolic acid (FIG. 6A). The glycolic acid production with this system in the presence of D-glucose has been compared with that with a system not having Hxk; the enzymatic assay is based on the NADH assay at 340 nm (FIG. 6B).

[0240] FIG. 7 shows the highlighting of the in vivo functionality of the non-natural pathway according to the invention in E. coli MG1655 ΔtktA ΔtktB ΔglcD, with in bold: the non-natural pathway according to the invention, in dotted lines, the deletions. Xu5P: xylulose-5-phosphate, Ru5P: ribulose-5-phosphate, R5P: ribose-5-phosphate, S7P: sedoheptulose-7-phosphate, F6P: fructose-6-phosphate, F16BP: Fructose-1,6-bisphosphate, G6P: glucose-6-phosphate, DHAP: dihydroxyacetone phosphate, Glyald: glycolaldehyde, E4P: erythrose-4-phosphate, G3P: glyceraldehyde-3-phosphate.

[0241] FIG. 8 shows the glycolic acid production of the E. coli strain MG1655 ΔtktA ΔtktB ΔglcD expressing the kdsD-fsa-aldA dependent non-natural pathway according to the invention from D-xylose or L-arabinose, at 37° C., 100h.

[0242] FIG. 9 shows the glycolic acid production of the strain of E. coli WC3G ΔgapA ΔglcD ΔarcA ΔmgsA ΔfucA Δpkf proD-galP expressing the kdsD-fsa-aldA dependent non-natural pathway according to the invention from glucose, xylose and arabinose, at 37° C., 50h.

DETAILED DISCLOSURE OF PARTICULAR EMBODIMENTS

[0243] I. Proof of In Vitro Feasibility.

[0244] I.1. Material and Methods.

[0245] A. Construction of Plasmids.

[0246] To construct a system in vitro, the inventors have cloned the ORFs (“Open Reading Frames”) of the xylose isomerase (XylA), xylulokinase (XylB), arabinose isomerase (AraA), ribulose kinase (AraB), L-ribulose-5-phosphate 4-epimerase (AraD), ribulose-5-phosphate-3-epimerase (RPE), D-arabinose-5-phosphate isomerase (KdsD), fructose-6-phosphate aldolase (FSA corresponding to FSAA such as previously defined) and aldehyde dehydrogenase (AdA) of E. coli K12 MG1655 with a polyhistidine tag for a facilitated purification. The oligonucleotides used for amplifying ORFS of the E. coli enzymes are listed in Table 3 below, the sequences SEQ ID NO: making reference to the appended sequence listing. The E. coli genomic DNA which served as on template is that of the strain K12 MG1655.

[0247] The plasmids for the expression of the C-terminal polyhistidine tagged proteins XylA, XylB, AraA, AraB, AraD, AraB, Rpe, KdsD, FSA, AdA have been constructed by HiFi Assembly® (NEB) with pET28a as a receiving vector. pET28a has been linearised beforehand with the restriction enzymes HindIII and BamHI (NEB) (Table 4). The E. coli strain NEB5®, derived from DH5 alpha, has been used for cloning and storing the different plasmids.

TABLE-US-00003 TABLE 3 Gene Primer Primer sequence xylA Sense ctggtgccgcgcggcagccatATGCAAGCCTATTTTGAC (SEQ ID NO: 26) Antisense gtcgacggagctcgaattcgTTATTTGTCGAACAGATAATGG (SEQ ID NO: 27) xylB Sense ctggtgccgcgcggcagccatATGTATATCGGGATAGATCTTG (SEQ ID NO: 28) Antisense gtcgacggagctcgaattcgTTACGCCATTAATGGCAG (SEQ ID NO: 29) araA Sense ctggtgccgcgcggcagccatATGACGATTTTTGATAATTATGAAG (SEQ ID NO: 30) Antisense gtcgacggagctcgaattcgTTAGCGACGAAACCCGTAATAC (SEQ ID NO: 31) araB Sense ctggtgccgcgcggcagccatATGGCGATTGCAATTGG (SEQ ID NO: 32) Antisense gtcgacggagctcgaattcgTTATAGAGTCGCAACGGCC (SEQ ID NO: 33) araD Sense ctggtgccgcgcggcagccatATGTTAGAAGATCTCAAACG (SEQ ID NO: 34) Antisense gtcgacggagctcgaattcgTTACTGCCCGTAATATGC (SEQ ID NO: 35) rpe Sense ctggtgccgcgcggcagccatATGAAACAGTATTTGATTGC (SEQ ID NO: 36) Antisense gtcgacggagctcgaattcgTTATTCATGACTTACCTTTGC (SEQ ID NO: 37) kdsD Sense gccgcgcggcagccatatATGTCGCACGTAGAGTTAC (SEQ ID NO: 38) Antisense gtcgacggagctcgaattcgTTACACTACGCCTGCACG (SEQ ID NO: 39) fsa Sense gccgcgcggcagccatatATGGAACTGTATCTGGATACTTC (SEQ ID NO: 40) Antisense gtcgacggagctcgaattcgTTAAATCGACGTTCTGCC (SEQ ID NO: 41) aldA Sense ctggtgccgcgcggcagccatATGTCAGTACCCGTTCAAC (SEQ ID NO: 42) Antisense gtcgacggagctcgaattcgTTAAGACTGTAAATAAACCACC (SEQ ID NO: 43)

TABLE-US-00004 TABLE 4 plasmids used for the proof of in vitro concept Name Description Reference pET28 Kan .sup.R, ColE1 ori Novagen ® pVT-FSA fsa-bearing pET28 This study pVT-KDSD kdsD-bearing pET28 This study pVT-ALDA aldA-bearing pET28 This study pVT-RPE rpe-bearing pET28 This study pVT-XYLA xylA-bearing pET28 This study pVT-XYLB xylB-bearing pET28 This study pVT-ARAA araA-bearing pET28 This study pVT-ARAB araB-bearing pET28 This study pVT-ARAD araC-bearing pET28 This study

[0248] B. Construction of the Strains.

[0249] E. coli-competent cells BL21 (DE3) have been used for the expression of tagged proteins since these cells express RNA polymerase T7, and are thus compatible with the expression system T7 of the vector pET28a. The plasmids pET28a checked beforehand by sequencing have been transformed in the BL21(DE3) according to the NEB protocol. The strains obtained are stored in 50% glycerol at −80° C. (Table 5).

TABLE-US-00005 TABLE 5 Escherichia coli strains used for the proof of in vitro concept Strain Genotype Reference MG1655 F-λ-ilvG-rfb-50 rph-1 ATCC 47076 BL21 (DE3) fhuA2 [Ion] ompT Invitrogen ® gal (λ DE3) [dcm] ΔhsdS prodFSA pVT-FSA-containing BL21 This study prodKDSD pVT-KDSD-containing BL21 This study prodALDA pVT-ALDA-containing BL21 This study prodRPE pVT-RPE-containing BL21 This study prodXYLA pVT-XYLA-containing BL21 This study prodXYLB pVT-XYLB-containing BL21 This study prodARA pVT-ARAA-containing BL21 This study prodARAB pVT-ARAB-containing BL21 This study prodARAC pVT-ARAC-containing BL21 This study

[0250] C. Expression and Purification of Polyhistidine-Tagged Proteins.

[0251] All the proteins expressed are soluble and have been produced from the expression vector pET28a transformed in an E. coli strain BL21 (DE3). A pre-culture in a LB (for “Luria-Bertani”) medium added with the antibiotic kanamycin is made overnight at 37° C. The pre-culture is used to seed a fresh culture of 200 mL of LB-Kanamycin at an optical density at 600 nm (OD.sub.600) of 0.1 (37° C., 200 rpm). When the OD.sub.600 is between 0.6 and 0.8, the expression of the protein of interest is induced by adding finally 1 mM IPTG. The proteins are expressed overnight at 16° C. The cells are collected as 50 ml fractions and centrifuged at 4 800 rpm for 15 min at 4° C. The cell pellets are preserved at −20° C.

[0252] The purification of proteins is made from the cell pellets obtained during the production step. All the steps are made while cold to avoid degradation of proteins by proteases. The cell pellets are re-suspended in 1.5 mL washing buffer (50 mM HEPES, pH 7.5; 0.3 M NaCl) and then sonicated on ice. A centrifugation step at 13 000 rpm, for 15 min at 4° C. enables the cell debris to be separated from the cytoplasmic liquid. The clarified lysates are deposited onto 600 ul cobalt resin (Clontech) previously balanced with the washing buffer. After 20 min at room temperature in contact with the resin, the tubes are centrifuged (700 rcf, 3 min, 4° C.). The supernatant is removed, the resin is contacted with 3 mL washing buffer for 10 min in order to remove non-specific interactions. After centrifugation (700 rcf, 3 min, 4° C.) and removing the supernatant, 3 mL of a solution of 15 mM imidazole are contacted with the resin for 5 min. The supernatant is separated from the resin by centrifugation, replaced by 500 μl of 200 mM imidazole. Imidazole causes elution of polyhistidine-tag bearing proteins. To promote protein stability at their optimum pH, the buffer has been modified.

[0253] The method used to measure the concentration of proteins in solution is based on the Bradford method. The Protein assay reagent sold by BioRad is diluted to %, the reaction mixture comprises 160 reagent and 40 diluted eluate (to 1/10.sup.th and to 1/20.sup.th). A standard range is made with BSA from 12.5 to 100 μg/ml.

[0254] D. Enzymatic Tests.

[0255] All the enzymes have been tested in 100 mM Tris, pH 7.5, 10 mM MgCl.sub.2, at 37° C. The co-factors and activators have been added if necessary.

[0256] The enzymes Saccaromyces cerevisiae hexokinase (Hxk), Escherichia coli transketolase (Tkt), Escherichia coli pyruvate kinase (PK) and Escherichia coli lactate dehydrogenase (LDH), triose-phosphate isomerase (Tpi) and glycerol-3-phosphate dehydrogenase (G3PDH) have been ordered to Sigma. The Escherichia coli phosphoglucose isomerase (Pgi) is from Megazyme.

[0257] All the enzymatic substrates have been bought to Sigma-Aldrich. The enzymatic tests are coupled to a redox reaction with as a co-factor, NADH which absorbs ultraviolet with a 340 nm peack with a coefficient of molar extinction of 6 220 M.sup.−1.Math.cm.sup.−1. The Bio Tek Epoch 2 spectrophotometer has been used for UV reaction monitoring.

[0258] Measurement of the aldehyde dehydrogenase (AldA) activity: this activity has been measured in the presence of 5 mM glycoaldehyde, 3 mM NAD.sup.+ and 3 mM ATP. For one molecule of glycoaldehyde oxidised into glycolate, one NAD.sup.+ molecule is reduced into NADH.

[0259] Measurement of the fructose-6P aldolase (FSA) activity: the enzyme FSA cleaves D-arabinose-5-phosphate into glycolaldehyde and glyceraldehyde-3-phosphate (GAP). The glycolaldehyde rate of appearance has been measured in the presence of AldA with 2 mM NAD.sup.+. To monitor GAP appearance, TPI and G3PDH have been added, in the presence of 0.4 mM NADH.

[0260] Measurement of the D-arabinose-5-phosphate isomerase (KdsD) activity: D-arabinose-5-phosphate isomerase catalyses the interconversion of D-ribulose-5-phosphate into D-arabinose-5-phosphate. The KdsD activity on D-ribulose-5-phosphate has been determined by adding FSA, aldA in excess in the presence of 3 mM NAD.sup.+.

[0261] In vitro conversion of D-xylose into glycolic acid: the conversion of D-xylose into glycolic acid requires XylA, XylB, Rpe, KdsD, FSA and aldA (FIG. 5A). The conversion of D-xylose into glycolic acid is demonstrated in a reaction mixture containing 100 mM Tris HCl pH 7.5, 10 mM MgCl.sub.2, 0.5 mM MnCl.sub.2, 3 mM ATP, 2 mM NAD.sup.+, 1-10 μg of each enzyme and 5 mM D-xylose.

[0262] In vitro conversion of L-arabinose into glycolic acid: the conversion of L-arabinose into glycolic acid requires AraA, AraB, AraD, Rpe, KdsD, FSA and aldA (FIG. 4A). The conversion of L-arabinose into glycolic acid is demonstrated in 100 mM Tris HCl, pH7.5, 10 mM MgCl.sub.2, 0.5 mM MnCl.sub.2, 3 mM ATP, 2 mM NAD.sup.+, 1-10 μg of each enzyme and 5 mM L-arabinose.

[0263] In vitro conversion of D-glucose into glycolic acid: The conversion of D-glucose into glycolic acid requires the enzymes Hxk, Pgi, Tkt, Rpe, KdsD, FSA and aldA (FIG. 6A). The proof of concept is made in 100 mM Tris HCl, pH 7.5, 10 mM MgCl.sub.2, 0.5 mM MnCl.sub.2, 1 mM TPP (“Thiamine Pyrophosphate”), 3 mM ATP, 2 mM NAD, 5 mM Glyceraldehyde-3P, 1-10 μg of each enzyme and 5 mM D-glucose.

[0264] I.2. Results.

[0265] A. In vitro conversion of D-ribulose-5P into glycolic acid.

[0266] Ribulose-5P is a metabolite common to the catabolism of arabinose, xylose and glucose in Escherichia coli. The non-natural conversion pathway of D-ribulose-5P into glycolic acid being the object of the present invention consists of the enzymes KdsD (D-arabinose 5P isomerase), FSA (Fructose-6P aldolase) and AldA (glycolaldehyde dehydrogenase). The in vitro proof of concept consisted in reconstructing this pathway with purified enzymes in an adapted buffer. First, aldA has been characterised on glycolaldehyde, and then the FSA activity on D-arabinose-5P has been checked by coupling with aldA in order to check glycolaldehyde formation and by coupling with triose phosphate isomerase and glycerol-3P dehydrogenase to highlight glyceraldehyde-3P synthesis (FIG. 3). The FSA activity measurements obtained by these 2 methods are identical. Finally, the KdsD activity on D-ribulose-5P has been measured by coupling with the excess enzymes FSA and aldA. This last enzymatic test enabled KdsD functionality to be checked and the entire glycolic acid synthesis pathway to be reconstituted. NADH is produced in an equimolar amount with glycolic acid. In order to check that NADH is really a glycolic acid production witness, the glycolic acid production has been measured by HPLC.

[0267] The catalytic constants of these 3 enzymes are shown in Table 6. Polyhistidine-tagged proteins KdsD, FSA and aldA are active. The reaction mixture containing KdsD, FSA and AdA in the presence of 5 mM of D-ribulose-5P has been analysed by HPLC, 1.5 mM of glycolic acid has been quantified. The glycolic acid production measured corresponds to the initial NAD.sup.+ amount in the reaction mixture, indicating that the reaction has been completed.

TABLE-US-00006 TABLE 6 catalytic constants of polyhistidine-tagged arabinose-5P isomerase (KdsD), fructose-6P aldolase (FSA), aldehyde dehydrogenase (AldA) Vmax/Km Vmax (.10.sup.−6) Enzyme Substrate Km (mM) (U mg.sup.−1) (min.sup.−1 mg.sup.−1 ) KdsD-His D-ribulose-5P  1.34 ± 0.12 1.14 ± 0.17 850 FSA-His D-arabinose-5P 0.652 ± 0.15 0.26 ± 0.09 398 aldA-His glycolaldehyde  0.21 ± 0.05 1.16 ± 0.31 5 552

[0268] B. Conversion of L-Arabinose or D-Xylose or D-Glucose into Glycolic Acid.

[0269] The enzymes catalyzing the conversion of L-arabinose, D-xylose and D-glucose into glycolic acid have been purified (AraA, AraB, AraD, XyA, XyB, Rpe, KdsD, Fsa, AldA) or ordered (Tkt, Glk, Pgi) in order to proof the concept of the glycolic acid production from these 3 substrates in vitro.

[0270] i. Conversion of L-Arabinose into Glycolic Acid.

[0271] The conversion of L-arabinose into glycolic acid and glyceraldehyde 3P is catalysed by AraA, araB, araD, Rpe, KdsD, FSA and aldA (FIG. 4A). The NADH synthesis observed during the enzymatic test in the presence of L-arabinose is the indirect witness of glycolic acid production (Table 7, FIG. 4B).

[0272] The conversion of L-arabinose into glycolic acid is demonstrated and this metabolic pathway is thus thermodynamically possible.

TABLE-US-00007 TABLE 7 enzymatic activity measured for the conversion of L-arabinose into glycolic acid at 340 nm, 37° C.. An enzymatic activity unit (U) is defined as the conversion of one micromole of substrate per minute. The negative control is devoid of arabinose isomerase activity catalysed by AraA. Substrate (5 mM) Enzymes Activity (U) L-arabinose AraA, AraB, AraD, Rpe, KdsD, FSA, 11.59 AldA AraB, AraD, Rpe, KdsD, FSA, AldA 0.00

[0273] ii. Conversion of D-Xylose into Glycolic Acid.

[0274] The conversion of D-xylose into glycolic acid and glyceraldehyde 3P is catalysed by XylA, XylB, Rpe, KdsD, FSA and AldA (FIG. 5A). When these enzymes are contacted with D-xylose, D-xylose is converted into glycolic acid in vitro as evidenced by NADH production (FIG. 5B, Table 8).

TABLE-US-00008 TABLE 8 enzymatic activity measured for the conversion of D-xylose into glycolic acid at 340 nm, 37° C.. An enzymatic activity unit (U) is defined as the conversion of a micromole of substrate per minute. The negative control is devoid of xylulose isomerase activity catalysed by XylA. Substrate (5 mM) Enzyme(s) Activity (U) D-xylose XylA, XylB, Rpe, KdsD, FSA, AldA 15.07 XylB, Rpe, KdsD, FSA, AldA 0.00

[0275] The conversion pathway of D-xylose into glycolate is functional and thermodynamically possible.

[0276] iii. Conversion of D-Glucose into Glycolic Acid.

[0277] The conversion of D-glucose into glycolic acid has been demonstrated by contacting D-glucose and DL-glyceraldehyde-3P (GAP) with Hxk, Pgi, Tkt, Rpe, KdsD, FSA and AdA (FIG. 6A). The negative control shows a significant activity (FIG. 6B, Table 9). GAP is a relatively instable molecule, it is thereby dephosphorylated spontaneously into glyceraldehyde. Baldoma et al. (1987) have demonstrated that Escherichia coli AldA was active on glycolaldehyde, lactaldehyde, methylglyoxal and L-glyceraldehyde [13]. NADH synthesis in the negative control is therefore not correlated with synthesis of glycolate but of glycerate from glyceraldehyde oxidation. This secondary activity is not relevant in vivo, especially as L-glyceraldehyde is not formed in Escherichia coli.

TABLE-US-00009 TABLE 9 enzymatic activity measured for the conversion of D-glucose into glycolic acid at 340 nm, 37° C.. An enzymatic activity unit is defined as the conversion of 1 micromole of substrate per minute. The negative control is devoid of hexokinase (Hxk) activity. Activity Corrected Substrates (5 mM) Enzymes (U) activity (U) D-glucose, GAP Hxk, Pgi, Tkt, Rpe, 23.5 6.2 KdsD, Fsa, AldA Pgi, Tkt, Rpe, KdsD, Fsa, AldA 17.3

[0278] From these results, the conversion pathway of D-glucose into glycolic acid is functional in vitro and is thus thermodynamically favourable.

[0279] The carbohydrates L-arabinose, D-xylose and D-glucose are naturally assimilated and converted into D-ribulose-5P in Escherichia coli, the non-natural conversion of D-ribulose-5P into glycolic acid allowed by the overexpression of KdsD, FSA and aldA has been demonstrated. The implementation of the non-natural pathway KdsD-FSA-aldA is thermodynamically favourable, the complete assimilation and conversion pathways of the reconstructed pentoses and hexose enabled glycolic acid to be synthetised in vitro.

[0280] However, the in vitro demonstration is by definition isolated from the E. coli natural metabolism, these results do not take into account reactions involving intermediates, transmembrane transport efficiency of the substrates and glycolic acid excretion, co-factor availability, . . . . A complementary in vivo proof of concept has been made to enhance these preliminary results.

[0281] II. In Vivo Proof of Feasibility.

[0282] II.1. Material and Methods.

[0283] A. Construction of Plasmids.

[0284] i. Choice of the Vectors

[0285] The series of vectors pZ (Expressys) has the advantage of being modulable: it is easy to change the replication origin, resistance marker and vector promoter by restriction/ligation.

[0286] Vectors pZA23, pZA33, pZE23 and pZS23 have the promoter PA1/ac0-1 which is a promoter derived from the lactose operon promoter comprising the operator o. PA1lac0-1 is under the control of the repressor lacI: in its active form, the repressor lacI is linked to the operator o and inhibits the transcription whereas, when complexed with IPTG, it changes conformation and is no longer capable to be bound to the site o, whereby the transcription becomes possible. The promoter PA1lac0-1 is said to be IPTG-inducible. Even if E. coli naturally has a lacI gene copy in its genome upstream of the operon lac, most of the IPTG-inducible bacterial expression vectors bear the lac gene in order to ensure full inhibition of the transcription of genes which are under its dependence. The vectors pZ have the feature to have a lightened structure of the lacI gene which provides them with a small size (2 358 to 3 764 bp).

[0287] Modifications have been provided to the vector pZA33, the promoter PA1/ac0-1 has been replaced by the constitutive promoter proD and by the inducible promoter Ptac, generating the vectors pZA36 and pZA38 respectively. The promoter PA1/ac0-1 of pZS23 has been modified by the promoter proD generating the vectors pZS27.

[0288] ii. Cloning Method: HiFi Assembly

[0289] The HiFi assembly (NEB) method has been retained to construct the vectors used hereinafter. This method enables several fragments to be assembled. It has been validated for fragments with different sizes with variable overlapping regions (15-80 bp). In a single step, the fragments can be assembled, it is a method commonly used for its simplicity and flexibility.

[0290] The commercial mixture provided by New England Biolabs contains different enzymes: (a) an exonuclease which creates 3′ single strand ends, which facilitates assembly of the fragments which share a sequence complementarity; (b) a polymerase which fills the empty spaces after the fragments have been assembled; and (c) a ligase which links fragments together.

[0291] The genes of the glycolic acid production synthetic pathway according to the invention kdsD, fsa and aldA have been amplified by PCR from the genome DNA extracted from E. coli K12 MG1655 and inserted by HiFi Assembly® into the vectors pZ, linearised beforehand by PCR with primers hybridising on either side of the MCS. All the plasmids have been checked by sequencing before use.

[0292] iii. Expression Vectors for the Overexpression of kdsD, Fsa, aldA

[0293] The vectors used for the overexpression of kdsD,fsa and aldA are shown in Table 10 hereinafter.

TABLE-US-00010 TABLE 10 Name Description Source pZA23 Kan .sup.R, ori p15A, PA1lac0-1 Expressys pZA33 Chm.sup.R, ori p15A, PA1lac0-1 Expressys pZS23 Kan .sup.R, ori pSC101, PA1lac0-1 Expressys pZA36 Chm .sup.R, ori p15A, Ptac This study pZA37 Chm .sup.R, ori p15A, proC This study pZS28 Kan .sup.R, ori pSC101, proD This study pEXT20 AmpR, ori ColE1, Ptac [33] pET28a Kan 11, ori ColE1, Ptac Novagen pKF3 pZA36 kdsD fsa This study pA4 pZS23 aldA This study pKF5 pZA37 kdsD fsa This study pA8 pZS28 aldA This study

[0294] B. Construction of the Strains.

[0295] i. Deletions of the Genes

[0296] The deletions have been made by transduction, using the phage P1vir. The preparation of the lysates P1vir and the transduction procedures have been made as described previously with little modifications [34].

[0297] Thus, the strains KEIO bearing a single deletion and a kanamycin antibiotic-resistance cassette (donor strain) have been infected with P1vir and high-titer lysates A have been obtained [35]. The donor strains (KEIO) have been cultured overnight in LB at 37° C. The day after, 5 ml of LB containing 0.2% glucose and 5 mM CaCl.sub.2) have been inoculated with 200 of the donor strain and cultured for 30 min at 37° C. Then, 100 of P1vir lysate (.sup.˜5×10.sup.8 phages/ml) have been added to each donor culture and incubated again at 37° C. for 2 to 3 hours until the culture was clear and the cells were completely lysed. The lysates have been filtered by using 25 mm sterile syringe filters with a 0.2 m support membrane (Pall) and preserved at 4° C.

[0298] The strain to be deleted (receiving strain) has been infected with P1vir containing a donor gene deletion cassette having a kanamycin resistance. For this, the receiving strain has been cultured in 5 ml LB medium at 37° C. The cells have been centrifuged at 1 500 g for 10 min and re-suspended in 1.5 ml of 10 mM MgSO.sub.4 and 5 mM CaCl.sub.2. Lysate from the donor strain (0.1 ml) is added to the cellular suspension which is incubated for 30 min. Then, 0.1 ml of 1M sodium citrate is added to the cell and P1vir mixture. Then, 1 mL LB is added to the homogenised suspension before an incubation of 1h at 37° C., 200 rpm.

[0299] The cellular suspensions are spread on a solid LB medium with the appropriate antibiotic then the colonies are screened by PCR to highlight successful transduction events.

[0300] To remove the antibiotic cassette, the cells have been transformed with a plasmid pCP20 bearing the FLP recombinase. Each step has been checked by PCR. When the deleted strain is sensitive to kanamycin after removing the cassette, it can again be used as a receiving strain in order to add a new deletion from a new phage lysate.

[0301] ii. Preparation of Competent Bacteria and Transformation

[0302] The competent non-commercial strains are prepared according to the protocol of Chung et al, 1989, which is slightly modified [36]. A pre-culture is made in LB overnight to inoculate the day after a fresh LB culture at a OD.sub.600 of 0.1. When the OD.sub.600 reaches 0.3-0.5 (the bacteria can be made competent up to a OD.sub.600 of 1), an amount of cellular culture equivalent to a OD.sub.600 unit is sampled and centrifuged (8 000 rpm, 2 min). The supernatant is removed whereas the pellet is up taken in 300 TSS buffer (2.5%.sub.(wt/vol) PEG 3350, 1 M MgCl.sub.2, 5%.sub.(vol/vol) DMSO). The mixture is incubated for 10 min on ice. The plasmid can then be added to the competent cells. After 30 further minutes of ice incubation, a heat shock is made at 42° C. for 90 seconds. The cells transformed are put on ice for 10 min. 400 LB are added and the culture is incubated at 200 rpm for 1 h, at an adapted temperature (the temperature can not exceed 30° C. in the case of a transformation with a plasmid the replication origin of which is thermosensitive). The bacterial culture is centrifuged at 8 000 rpm for 2.5 min. 600 of supernatant are removed, the remaining volume is inoculated on a solid LB dish with the appropriate antibiotic.

[0303] iii. Strains for Glycolic Acid Production

[0304] The E. coli strains used in this study are listed in Table 11 hereinafter.

TABLE-US-00011 TABLE 11 Strain Genotype Reference MG1655 F-λ-ilvG- rfb-50 rph-1 ATCC 407076 NEB5 fhuA2 Δ(argF-lacZ)U169 phoA NEB glnV44 ∅80 Δ(lacZ)M15 gyrA96 recA1 relAl endA1 thi-1 hsdR17 BL21 (DE3) huA2 [Ion] ompT gal (λ DE3) NEB [dcm] ΔhsdS) Screen00 MG1655 ΔtktA ΔtktB ΔglcD This study Screen09 Screen00 containing pZA36 This study kdsD fsa (pKF3) and pZS23 aldA (pA4) Screen23 Screen00 containing pZA37 kdsD This study fsa (pKF5) and pZA28 aldA (pA8) WC3G W3CG F-, LAM-, gapA 10::Tn10, [37] gapA- IN(rrnD-rrnE), rph-1 BW25113 F-, Δ(araD-araB)567, [26] ΔlacZ4787(::rrnB-3), λ.sup.-, yrph-1, Δ(rhaD- rhaB)568, hsdR514 JW2771 BW25113 ΔfucA [26] JW4364 BW25113 ΔarcA [26] JW5129 BW25113 ΔmgsA [26] JW2946 BW25113 ΔgIcD [26] JW2771 BW25113 ΔfucA [26] JW3887 BW25113 ΔpfkA [26] GA00 WC3G gapA- ΔglcD ΔarcA This study ΔmgsA ΔfucA ΔpfkA proD galP GA09 GA00 containing pZA36 kdsD This study fsa (pKF3) and pZS23 aldA (pA4) GA23 GA00 containing pZA37 kdsD This study fsa (pKF5) and pZA28 aldA (pA8)

[0305] C. Media and Culture Conditions.

[0306] i. Composition of the Media

[0307] The cells are cultured on the LB medium for the molecular biology (cloning, deletions, transformation) steps. This rich medium is comprised of 10 g/L trypton, 5 g/L yeast extracts and 5 g/L NaCl. 15 g/L agar are added for obtaining a solid medium. The LB, with or without agar, is sterilised by autoclaving for 20 min at 110° C. before use.

[0308] The cultures for glycolic acid production are made in a mineral medium M9 (Table 12) containing a carbon source (glucose, xylose or arabinose) at a 10 g/L or 20 g/L concentration and LB traces in order to reduce the latency phase (2 g/L trypton, 1 g/L yeast extracts and 1 g/L NaCl).

TABLE-US-00012 TABLE 12 composition of the medium M9 Compound Final concentration (g/L) Na.sub.2HPO.sub.4 * 12 H.sub.2O 18 KH.sub.2PO.sub.4 3 NaCl 0.5 NH.sub.4Cl 2 MgSO.sub.4 * 7 H.sub.2O 0.5 CaCl.sub.2 * 2 H.sub.2O 0.015 FeCl.sub.3 0.010 Thiamine HCl 6.10.sup.−3 NaEDTA 0.49.10.sup.−3 CoCl.sub.2 * 6 H.sub.2O 1.8.10.sup.−3 ZnCl.sub.2SO.sub.4 * 7 H.sub.2O 1.8.10.sup.−3 Na.sub.2MoO.sub.4 * 2 H.sub.2O 0.4 .10.sup.−3 H.sub.3BO.sub.3 0.1.10.sup.−3 MnSO.sub.4 * H.sub.2O 1.2.10.sup.−3 1.2 mg/L CuCl.sub.2 * 2 H.sub.2O 1.2.10.sup.−3

[0309] For the strain MG1655 ΔtktA ΔtktB ΔglcD, the medium M9 with LB traces is complemented with 500 M L-phenylalanine, 250 μM L-tyrosine, 200 μM L-tryptophane, 6 μM p-aminobenzoate, 6 μM p-hydroxydenzoate and 280 μM shikimate. For the strains the glyceraldehyde-3-phosphate dehydrogenase activity of which has been removed, the medium has been completed with 0.4 g/L malic acid adjusted at pH 7 with KOH beforehand. The medium is buffered by adding 20 g/L 3-(N-morpholino)propanesuphonic acid (MOPS) at pH 7 and then filtered through 0.2 μm membranes to obtain a sterile medium. If need be, the appropriate antibiotics have been added to the medium (100 g/mL ampicillin, 50 μg/mL kanamycin, 25 μg/mL chloramphenicol). For the strains containing an IPTG inducible vector, a 0.1 mM IPTG final concentration is used. All the products have been ordered to Sigma. The cultures are placed in a stirrer (Infors) at 200 rpm, 37° C. for the experiment time. The culture growth is monitored by measuring the 600 nm optical density (OD.sub.600) with a spectrophotometer (Biochrom Libra S11).

[0310] ii. Fermentation Process in 250 mL Baffle Erlenmeyer Flasks

[0311] Strains Containing Constituted Promoters

[0312] Protocol 1

[0313] The strains are uptaken from of a glycerol stock preserved at −80° C. in 10 mL LB, at 37° C. overnight. The pre-cultures are centrifuged the day after at 4 000 rpm for 5 min and re-suspended in 20 mL of M9 medium with 1% xylose or arabinose in 100 mL Erlenmeyer flasks. An adaptation phase with a 24h duration enables the strains to adapt to pentose use. The cells are then centrifuged (4 000 rpm, 5 min) and then uptaken under the final culture conditions: at an initial OD.sub.600 of 0.5 in 50 mL medium with a composition identical to that used during the adaptation phase (M9 with 1% xylose or arabinose). 250 mL baffle culture flasks are used for an optimum oxygenation. The cultures are stirred at 200 rpm, at 37° C.

[0314] Protocol 2

[0315] The strain GA23 is uptaken from of a glycerol stock preserved at −80° C. in 10 mL LB containing the following antibiotics: tetracyclin (10 μg/mL), kanamycin (50 μg/ml) and chloramphenicol (25 g/ml). The pre-culture is incubated at 37° C., 200 rpm overnight. The process is decoupled into two phases: a growth phase dedicated to biomass production and a production phase dedicated to the production of glycolic acid. The growth phase is achieved in 50 mL of medium M9 to pH 7 containing 1.5 g/L xylose, 5 g/L succinate, 73 mg/L L L-methionine, 73 mg/L L-tryptophan and 1 g/L casaminoacids inoculated with pre-culture at an OD.sub.600 of 0.2 in a 250 ml erlenmeyer. After 24h of culture at 37° C., 200 rpm, the cells are centrifuged, washed and recovered in 50 mL M9 at pH 7 containing 10 g/L lignocellulosic sugar (glucose, xylose or arabinose), 73 mg/L L-methionine, 73 mg/L L-tryptophan and 1 g/L casaminoacids in a 250 ml erlenmeyer. The medium used for production is devoid of succinate and does not allow growth, the source of carbon (glucose, xylose or arabinose) is used for the production of glycolic acid. The production phase lasts 48h at 37° C., 200 rpm, samples are taken regularly to measure pH, optical density and track sugar consumption and GA production by HPLC.

[0316] Strains Containing Inducible Promoters

[0317] The LB pre-cultures of the strains containing a vector with an inducible promoter are cultured at an initial OD.sub.600 of 0.1 in 30 mL of M9 with 1% glucose. When the OD.sub.600 reaches 0.6, 0.1 mM IPTG is added to induce gene expression under the control of the IPTG inducible promoter (plac or ptac). 24h after, the cells are centrifuged, a sterile water washing enables glucose traces to be removed, and they are re-suspended in 20 mL of the medium chosen from the study (M9 with 1% xylose or arabinose, 0.1 mM IPTG) for the adaptation phase. The rest of the culturing protocol is identical to that for the strains with constitutive promoters.

[0318] iii. Fermentation Process in a 2 L BIOSTAT® B Startorius Stedim Biotech Bioreactor

[0319] Cultures of the previously described strain GA23 have been made in a 2 L bioreactor. The pre-cultures have been made in LB added with kanamycin, chloramphenicol and tetracyclin, cultured at 37° C., 200 rpm. A first 3 mL pre-culture has been made overnight, the same has been used for inoculating a second 15 mL pre-culture cultured for 24h. The same has been used for inoculating a third and last pre-culture in a 150 mL Erlenmeyer flask cultured for 8h. In the exponential phase, this last pre-culture has been used to inoculate the fermenter culture at a OD of 0.3. The M9 culture medium (Table 12) contains 10 g/L xylose or glucose, as well as LB (10%) and 0.4 g/L malic acid. The cultures have been made at 37° C., with stirring (300-1500 rpm) and venting maintaining dissolved oxygen above 20% of the air flow rate. The pH has been maintained at 7 with a KOH base solution. The culture has been conducted in a batch mode for 40h.

[0320] D. Measurements of the Concentrations of Extracellular Metabolites.

[0321] Sugar consumption and glycolic acid production are monitored by regularly sampling samples under sterile conditions which are centrifuged at 13 000 rpm for 5 min in a laboratory centrifuge (Eppendorf 5415D), filtered through a 0.2 μm filter and stored at −20° C. The measurement of extracellular metabolites is made by high performance liquid chromatography (HPLC) with an Ultimate 3000 chromatograph (Dionex, Sunnyvale, USA). The HPLC system is equipped with a cation exchange column (Aminex, HPX87H-300×7.8 mm, 9 μm, BioRad), an automatic injector (WPS-3000RS, Dionex), an IR detector (RID 10A, Shimadzu) and a UV detector (SPD-20A, Shimadzu). The mobile phase is a 1.25 mM sulfuric acid solution at a 0.5 mL/min flow rate. The samples are maintained at 4° C. and the injection of 20 is made in the column at 35° C.

[0322] II.2. Results Obtained with a Baffle Erlenmeyer Flask.

[0323] The inventors have demonstrated that the overexpression in E. coli of the 3 non-natural pathway enzymes according to the invention, i.e. arabinose-5-phosphate isomerase (KdsD), fructose-6-phosphate aldolase (FSA) and aldehyde dehydrogenase (AldA) is necessary and sufficient for synthesising glycolic acid from lignocellulosic monosaccharides.

[0324] A. With Pentoses as a Carbon Source.

[0325] To that end, a strain of E. coli K12 MG1655 deleted from genes tktA, tktB coding for transketolase has been constructed to highlight the functionality of the pathway according to the invention. Transketolase are major enzymes of the pentose phosphate pathway. In their absence, the growth on pentose is impossible, the intermediates pentose phosphate (D-ribose-5-phosphate and D-Xylulose-5-phosphate) are accumulated and can not be converted into glyceraldehyde-3phosphate for growth [38]. This strain needs the non-natural pathway according to the invention to grow from pentoses. Indeed, only the coupled activity of KdsD and FSA can generate glyceraldehyde-3-phosphate involved in glycolysis for the production of precursors necessary for growth [39]. Thus, a functional screen per growth test is contemplatable in which growth would be an indicator of the in vivo synthetic pathway efficiency (FIG. 7).

[0326] The 3 genes have been cloned as an operon and expressed in a strain of E. coli MG1655 ΔtktA ΔtktB ΔglcD. To that end, a two-vector expression system has been used, the latter has been accepted by the strain without generating detrimental modifications for the expression of the synthetic pathway according to the invention. The expression system of the synthetic pathway according to the invention thus includes a vector having a medium copy number bearing kdsD-fsa co-expressed with a vector with a low copy number bearing aldA.

[0327] Two expression systems have been studied, the first one with vectors with IPTG inducible promoters (pZA36 kdsD fsa and pZS23 aldA, expression system n° 9) and the other one with vectors with constitutive promoters (pZA37 kdsD fsa and pZS28 aldA, expression system n° 23).

[0328] The cultures of the strains MG1655 ΔtktA ΔtktB ΔglcD expressing the systems 9 (Screen09) and 23 (Screen23) have been made at 37° C. and have shown a significant growth indicating that the overexpression of the kdsD and fsa genes is functional. HPLC analyses of the exometabolome of the bacterial cultures on L-arabinose and D-xylose have confirmed the presence of glycolic acid with both expression systems. This result demonstrates the feasibility of glycolic acid synthesis from pentoses by the in vivo synthetic pathway (FIG. 8). Table 13 hereinafter shows the glycolic acid yield from xylose and arabinose of the strains Screen09 and Screen23.

TABLE-US-00013 TABLE 13 Glycolic Glycolic acid/xylose acid/arabinose Strain (mol/mol) (mol/mol) Screen00 0.14 ± 0.03 0.14 ± 0.03 Screen09 0.43 ± 0.02 0.32 ± 0.01 Screen23 0.42 ± 0.02 0.20 ± 0.03 Glycolic Glycolic acid/xylose acid/arabinose Strain (g/g) (g/g) Screen00 0.07 ± 0.03 0.07 ± 0.03 Screen09 0.22 ± 0.02 0.16 ± 0.01 Screen23 0.21 ± 0.02 0.10 ± 0.03

[0329] B. With Pentoses and Hexoses as a Carbon Source.

[0330] The expression systems 9 and 23 have been transformed in the strain of E. coli WC3G ΔgapA ΔglcD ΔarcA ΔmgsA ΔfucA Δpkf proD-galP (GA00) designed for the glycolic acid production from hexoses and pentoses with an optimum carbon preservation, generating strains GA09 and GA23, respectively.

[0331] The bacterial cultures have been made in an Erlenmeyer flask at 37° C. on D-glucose, L-arabinose and D-xylose for 46h. The overexpression of the kdsD,fsa and aldA genes is necessary for glycolic acid production.

[0332] With Protocol 1 Such as Previously Defined

[0333] The strain GA00 with the constitutive expression system n° 23 that is GA23 has shown a glycolic acid production from D-glucose (0.29 g/L), D-xylose (0.41 g/L) and L-arabinose (0.07 g/L). The glycolic acid production with the inducible expression system n° 9 that is GA09 is less significant on D-glucose (0.1 g/L), D-xylose (0.05 g/L) and L-arabinose (0.03 g/L) (FIG. 9). The expression system including constitutive promoters is favourable. The yield of the strain GA23 in glycolic acid is 0.09 g/g (0.21 mol/mol) from glucose, 0.18 g/g (0.36 mol/mol) from xylose and 0.16 g/g (0.32 mol/mol) from arabinose.

[0334] With protocol 2 such as previously defined

[0335] The yield of the strain GA23 in glycolic acid is 0.096 g/g (0.19 mol/mol) from glucose, 0.3 g/g (0.60 mol/mol) from xylose and 0.34 g/g (0.68 mol/mol) from arabinose.

REFERENCES

[0336] [1] Gädda et al, 2014, «The industrial potential of bio-based glycolic acid and polyglycolic acid», Appita Journal, vol. 67, pages 1-2. [0337] [2] Kunze et al, 2006, «A central role for the peroxisomal membrane in glyoxylate cycle function», Biochimica et Biophysica Acta—Molecular Cell Research, vol. 1763, pages 1441-1452. [0338] [3] International application WO 2007/141316 filed on behalf of Metabolic Explorer, published on 13 Dec. 2007. [0339] [4] Deng, et al, 2018, «Balancing the carbon flux distributions between the TCA cycle and glyoxylate shunt to produce glycolate at high yield and titer in Escherichia coli», Metabolic Engineering, vol. 46, pages 28-34. [0340] [5] International application WO 2017/09236 filed on behalf of Massachusetts Institute of Technology, published on 6 Apr. 2017. [0341] [6] Pereira et al, 2016, «Efficient utilization of pentoses for bioproduction of the renewable two-carbon compounds ethylene glycol and glycolate», Metabolic Engineering, vol. 34, pages 80-87. [0342] [7] International application WO 2016/079440 field on behalf of INRA, INSA and CNRS, published on 26 may 2016. [0343] [8] Alkim et al, 2016, «The synthetic xylulose-1 phosphate pathway increases production of glycolic acid from xylose-rich sugar mixtures», Biotechnology for Biofuels, 9. doi: 10.1186/s13068-016-0610-2. [0344] [9] Dugar and Stephanopoulos, 2011, «Relative potential of biosynthetic pathways for biofuels and bio-based products», Nat Biotechnol, vol. 29, pages 1074-1078. [0345] [10] Bogorad et al, 2013, «Synthetic non-oxidative glycolysis enables complete carbon conservation», Nature, vol. 502, pages 693-697. [0346] [11] Meredith and Woodard, 2006, «Characterization of Escherichia coli D-arabinose 5-phosphate isomerase encoded by kpsF: implications for group 2 capsule biosynthesis», Biochemical Journal. doi: 10.1042/BJ20051828. [0347] [12] Garrabou et al, 2009, «Asymmetric self- and cross-aldol reactions of glycolaldehyde catalysed by d-fructose-6-phosphate aldolase», Angewandte Chemie—International Edition, vol. 48, pages 5521-5525. [0348] [13] Baldomh and Aguilar, 1987, «Involvement of lactaldehyde dehydrogenase in several metabolic pathways of Escherichia coli K12», The Journal of biological chemistry, vol. 262, pages 13991-13996. [0349] [14] Schurmann and Sprenger, 2001, «Fructose-6-phosphate aldolase is a novel class I aldolase from Escherichia coli and is related to a novel group of bacterial transaldolases». Journal of Biological Chemistry, vol. 276, pages 11055-11061. [0350] [15] Szekrenyi et al, 2014, «Engineering the donor selectivity of D-fructose-6-phosphate aldolase for biocatalytic asymmetric cross-aldol additions of glycolaldehyde», Chemistry—A European Journal, vol. 20, pages 12572-12583. [0351] [16] Ward et al, 1999, «Expression of prokaryotic membrane transport proteins in Escherichia coli», Biochemical Society Transactions,vol. 27, pages 893-899. [0352] [17] Lu et al, 2012, «Combinatorial modulation of galP and glk gene expression for improved alternative glucose utilization», Applied Microbiology and Biotechnology, vol. 93, pages 2455-2462. [0353] [18] Hernandez-Montalvo et al, 2003, «Expression of galP and glk in a Escherichia coli PTS mutant restores glucose transport and increases glycolytic flux to fermentation products», Biotechnology and Bioengineering, vol. 83, pages 687-694. [0354] [19] Snoep et al, 1994, «Reconstitution of glucose uptake and phosphorylation in a glucose-negative mutant of Escherichia coli by using Zymomonas mobilis genes encoding the glucose facilitator protein and glucokinase», J. Bacteriol., vol. 176, pages 2133-2135. [0355] [20] Garcia et al, 2001, «Characterization of sugar mixtures utilization by an Escherichia coli mutant devoid of the phosphotransferase system», Applied Microbiology and Biotechnology, vol. 57, pages 186-191. [0356] [21] Lord, 1972, «Glycolate oxidoreductase in Escherichia coli», BBA—Bioenergetics, vol. 267, pages 227-237. [0357] [22] Alexeeva et al, 2003, «Requirement of ArcA for redox regulation in Escherichia coli under microaerobic but not anaerobic or aerobic conditions», J Bacteriol., vol. 185, pages 204-209. [0358] [23] Pellicer et al, 1999, «A mutational study of the ArcA-P binding sequences in the aldA promoter of Escherichia coli», Molecular and General Genetics, vol. 261, pages 170-176. [0359] [24] Liu and De Wulf, 2004, «Probing the ArcA-P Modulon of Escherichia coli by Whole Genome Transcriptional Analysis and Sequence Recognition Profiling», Journal of Biological Chemistry, vol. 279, pages 12588-12597. [0360] [25] Cho et al, 2006, «Transcriptional regulation of the fad regulon genes of Escherichia coli by ArcA», Microbiology, vol. 152, pages 2207-2219. [0361] [26] Saadat and Harrison, 1998, «Identification of catalytic bases in the active site of Escherichia coli methylglyoxal synthase: Cloning, expression, and functional characterization of conserved aspartic acid residues», Biochemistry, vol. 37, pages 10074-10086. [0362] [27] Atkinson and Walton, 1965, «Kinetics of regulatory enzymes Escherichia coli phosphofructokinase», The Journal of Biological Chemistry, vol. 240, pages 757-764. [0363] [28] Ghalambor and Heath, 1962, «The metabolism of L-Fucose», The Journal of Biological Chemistry, vol. 237, pages 2427-2433. [0364] Banerjee and Fraenkel, 1972, «Glucose-6-phosphate dehydrogenase from Escherichia coli and from “high level” mutant», J. bacteriol., vol. 110, pages 155-160. [0365] [30] Gardner and Fridovich, 1991, «Superoxide sensitivity of the Escherichia coli aconitase», Journal of Biological Chemistry, vol. 266, pages 19328-19333. [0366] [31] Fong et al, 2000, «Directed evolution of D-2-keto-3-deoxy-6-phosphogluconate aldolase to new variants for the efficient synthesis of D- and L-sugars», Chemistry and Biology, vol. 7, pages 873-883. [0367] [32] Jørgensen et al, 2007, «Enzymatic conversion of lignocellulose into fermentable sugars: Challenges and opportunities», Biofuels, Bioproducts and Biorefining, vol. 1, pages 119-134. [0368] [33] Dykxhoorn et al, 1996, «A set of compatible tac promoter expression vectors», Gene, vol. 177, pages 133-136. [0369] [34] Bremer et al, 1984, «λ placMu: A transposable derivative of bacteriophage lambda for creating lacZ protein fusions in a single step», Journal of Bacteriology, vol. 158, pages 1084-1093. [0370] [35] Baba et al, 2006, «Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: The Keio collection», Molecular Systems Biology, 2. doi: 10.1038/msb4100050. [0371] [36] Chung et al, 1989, «One-step preparation of competent Escherichia coli: transformation and storage of bacterial cells in the same solution», Proceedings of the National Academy of Sciences, vol. 86, pages 2172-2175. [0372] [37] Ganter et Pliickthun, 1990, «Glycine to alanine substitutions in helices of glyceraldehyde-3-phosphate dehydrogenase: effects on stability», Biochemistry, vol. 29, pages 9395-2402. [0373] [38] Teleman et al, 1999, «Identification and quantitation of phosphorus metabolites in yeast neutral pH extracts by nuclear magnetic resonance spectroscopy», Analytical Biochemistry, vol. 272, pages 71-79. [0374] [39] Zhao and Winkler, 1994, «An Escherichia coli K-12 tktA tktB mutant deficient in transketolase activity requires pyridoxine (vitamin B6) as well as the aromatic amino acids and vitamins for growth», J. Bacteriol., vol. 176, pages 6134-6138