Method for improving yield and yield of 2, 4-dihydroxybutyric acid

Abstract

The present invention relates to an improved process for the preparation of 2, 4-dihydroxybutyrate from a carbon source via homoserine, comprising a two-step route: a first step of catalytically converting the primary amino group of homoserine to a carbonyl group by an enzyme having homoserine transaminase activity as defined by E.C.2. 6. 1.1 or E.C.2. 6. 1.42 or E.C.2. 6. 1.57 to obtain 2-oxo-4-hydroxybutyrate, and a second step of catalytically converting the primary amino group of homoserine to a carbonyl group by an enzyme having homoserine transaminase activity as defined by E.C.2. 6. 1.1 or E.C.2. 6. 1.57 to obtain 2-oxo-4-hydroxybutyrate. And-a second step of catalytically reducing the obtained 2-oxo-4-hydroxybutyrate (OHB) to 2, 4-dihydroxybutyrate by an enzyme having OHB reductase activity, the enzyme having OHB reductase activity being a lactic dehydrogenase as defined by E.C. 1.1. 1.27 or E.C. 1.1. 1.28, or a malate dehydrogenase as defined by E.C. 1.1. 1.37, E.C. 1.1. 1.82, or E.C. 1.1. 1.299, and-a second step of catalytically reducing the obtained 2-oxo-4-hydroxybutyrate (OHB) to 2, 4-dihydroxybutyrate by an enzyme having OHB reductase activity, the enzyme having OHB reductase activity being a lactic dehydrogenase as defined by E.C. 1.1. 1.28, or a malate dehydrogenase as defined by and wherein at least one enzyme having aldolase activity, as defined by E.C.4.1. 2.28, is inactivated to prevent degradation of OHB.

Description

Technical Field

[0001] This invention relates to improvements to a prior art method for preparing 2,4-dihydroxybutyrate (2,4-DHB) from a carbon source via homoserine, comprising a two-step pathway, wherein the first step comprises converting the primary amino group of homoserine to a carbonyl group to obtain 2-oxo-4-hydroxybutyrate (OHB), and the second step comprises reducing the obtained 2-oxo-4-hydroxybutyrate to produce 2,4-DHB. The improvements to this method include reducing or even eliminating the transfer of OHB to optimize the pathway flux from the carbon source to the target compound (i.e., DHB). The DHB thus produced is particularly useful for the production / synthesis of 2-hydroxy-4-(meththio)-butyrate (HMTB) and / or hydroxy analogs of selenomethionine (HMSeBA or HMSeB) according to the invention. The invention also relates to modified microorganisms for preparing 2,4-DHB from a carbon source via homoserine, said preparation comprising a two-step pathway wherein the pathway flux from the carbon source to the target compound (i.e., DHB) is optimized; and to a method for producing 2,4-DHB by culturing said modified microorganisms. Background Technology

[0002] 2,4-Dihydroxybutyric acid (equivalent to 2,4-DHB or DHB) is a compound with considerable economic benefits. By adjusting the appropriate pH, DHB can be readily converted to α-hydroxy-γ-butyrolactone in an aqueous medium. α-Hydroxy-γ-butyrolactone is an important precursor for the production of methionine-substituted 2-hydroxy-4-(meththio)-butyrate (HMTB) (US2009 / 318715) or hydroxy analogs of selenomethionine (HMSeBA or HMSeB; EP1778706B1), which have a large market in animal nutrition. Currently, α-hydroxy-γ-butyrolactone can be prepared industrially using two different chemical methods:

[0003] i) Derived from γ-butyrolactone through a multi-stage chemical process, which involves the halogenation of γ-butyrolactone at the α-position and subsequent substitution of the halogen atom with a hydroxyl group in an alkaline medium (Deck et al., 2008), and

[0004] ii) Two-step chemical reaction from HMTB, namely alkylation followed by hydrolysis (EP2627645 B1).

[0005] Rising oil prices necessitate the production of DHB from renewable resources. Microorganisms are capable of converting biomass-derived feedstocks, such as sugars or organic acids, into a variety of chemical compounds (Werpy and Petersen, 2004). With the increasing volume of biochemical and genomic information, it is possible to modify microorganisms to overproduce naturally occurring metabolic intermediates in high yields and productivity (Bailey, 1991). Optimization of producing microorganisms typically requires the rational engineering of metabolic networks, which, among other things, ensures the overexpression of enzymes required for the biosynthetic target metabolite and the mitigation of product feedback inhibition. Another possibility is the implementation of novel enzyme systems for the catalytic production of industrially relevant non-natural metabolites.

[0006] Like DHB, OHB is a non-natural product in microorganisms. The first description in document WO 2014 / 009435 of the conversion of homoserine to 2,4-DHB in modified microorganisms using a two-step enzymatic conversion involving the reduction of 2-oxo-4-hydroxybutyrate (OHB) catalyzed by an engineered malate dehydrogenase is found in this DHB production pathway. However, this DHB production pathway is not optimized, and carbon fluxes, particularly OHB as an enzyme substrate, can be diverted from the DHB production pathway to other competing pathways, thus reducing the final DHB yield. However, since OHB is a non-natural metabolite in microorganisms, the emergence of these potential competitors is not immediately apparent to those skilled in the art and must be identified and ultimately removed. In other words, optimizing the DHB production yield via the two-step enzymatic conversion of 2-oxo-4-hydroxybutyrate (OHB) as described in the prior art would require:

[0007] - Optimizing pathway flux from carbon source to target compound by overexpressing restriction enzymes in the pathway; and / or

[0008] - Eliminate and attenuate unwanted reactions that cause carbon to be lost as byproducts (such as organic acids or CO2); and / or

[0009] - Eliminate competing pathways for common intermediates between secondary and production pathways.

[0010] Therefore, identifying potential pathways to avoid OHB loss and / or facilitate its transfer from DHB production is crucial. However, as mentioned above, since OHB is a non-natural product of microbial metabolism, hypothetical OHB consumption pathways may be expected to exist or be activated in response to the presence of this unusual metabolite, which has neither been proposed nor described in previous prior art concerning DHB production or other applications. Therefore, how to ensure optimal DHB yield is not immediately apparent to those skilled in the art.

[0011] As mentioned above, a two-step pathway for the production of 2,4-DHB from homoserine has been observed (i.e., according to WO2014 / 009435), specifically involving the co-production of formaldehyde and formate with DHB. However, formaldehyde is known to be highly cytotoxic and mutagenic, while formate can cause problems in downstream DHB processes. Therefore, preventing the production of these byproducts would enhance the DHB production pathway and consequently enhance the production of HMTB and / or HMSeB, particularly the chemical production of HMTB and / or HMSeB (i.e., outside of microorganisms).

[0012] Therefore, in order to optimize the production of DHB and its byproducts such as HMTB and / or HMSeB, while suppressing the production of secondary products that are particularly harmful / toxic to the producing microorganisms, it is necessary to reduce or even eliminate carbon flux transfer. The purpose of this invention is to meet these needs. Summary of the Invention

[0013] Therefore, one object of the present invention is an improved method for preparing 2,4-DHB from a carbon source, particularly homoserine, via homoserine, comprising a two-step approach (see [link to original text]). Figure 1 ):

[0014] The first step involves converting the primary amino group of homoserine to a carbonyl group to obtain OHB, and

[0015] The second step involves reducing the obtained OHB to 2,4-DHB.

[0016] The pathway flux from carbon source to target compound was optimized, particularly ensuring that OHB, as an enzymatic substrate, is not transferred from the DHB production pathway, and the production of HMTB and / or HMSeB was also optimized, especially the chemical production of HMTB and / or HMSeB (i.e., outside of microorganisms).

[0017] The inventors unexpectedly and surprisingly observed that inactivating, deleting, or altering the enzyme activity of so-called aldolases, or preferably pyruvate aldolases of classes I and II, using EC 4.1.2.28 resulted in higher DHB yields, and consequently, higher HMTB and / or HMSeB yields, particularly in chemical production methods (i.e., outside of microorganisms) than in methods described in the prior art (particularly in the literature WO2014 / 009435 or by implementing the microorganisms described in WO 2014 / 009435). Furthermore, inactivating, deleting, or altering the enzyme activity of aldolases using EC 4.1.2.28 reduced and / or prevented the production of toxic byproducts such as formaldehyde.

[0018] Therefore, the present invention relates to a method for preparing 2,4-dihydroxybutyrate and / or relatedly preparing HMTB and / or HMSeB, particularly a method for the chemical production of HMTB and / or HMSeB from a carbon source via homoserine, particularly from homoserine, comprising a two-step approach:

[0019] - The first step involves catalyzing the conversion of the primary amino group of homoserine to a carbonyl group using an enzyme with homoserine transaminase activity as defined in EC2.6.1.1, EC2.6.1.42, or EC2.6.1.57 to obtain 2-oxo-4-hydroxybutyrate.

[0020] The second step involves reducing the obtained 2-oxo-4-hydroxybutyrate (OHB) to 2,4-dihydroxybutyrate using an enzyme with OHB reductase activity. This enzyme is a lactate dehydrogenase as defined in EC 1.1.1.27 or EC 1.1.1.28, or a malate dehydrogenase as defined in EC 1.1.1.37, EC 1.1.1.82, or EC 1.1.1.299.

[0021] At least one of the enzymes having aldolase activity, as defined in EC4.1.2.28, is inactivated to prevent OHB degradation.

[0022] In fact, Bouzon reported the use of pyruvate-dependent aldolases to produce OHB via the condensation reaction of pyruvate and formaldehyde (Bouzon et al., 2017). Although it has been reported that various hybrid aldolases can catalyze the condensation of pyruvate and formaldehyde to form OHB, a non-natural intermediate in the bacterial CO2 assimilation pathway (Bouzon et al., 2017), it is not evident whether the loss of these hybrid enzymes would increase DHB production.

[0023] The method for producing 2,4-DHB and / or relatedly producing HMTB and / or HMSeB according to the present invention, particularly the chemical production method of HMTB and / or HMSeB, provides an improved / optimized preparation of 2,4-dihydroxybutyrate and an improved / optimized preparation of HMTB and / or HMSeB, wherein significant degradation of OHB to formaldehyde and subsequently to formate is prevented (see...). Figure 2 Therefore, it is the reason for the significant increase in yield, volumetric productivity, and DHB concentration and / or related HMTB and / or HMSeB concentration.

[0024] Therefore, the inactivation of at least one enzyme having aldolase activity as defined in EC4.1.2.28 results in a significant increase in the initial substrate yield (e.g., glucose) and also avoids the accumulation of formate, an organic acid that may potentially hinder the purification of 2,4-DHB and / or, consequently, the production of HMTB and / or HMSeB, particularly chemical production methods of HMTB and / or HMSeB.

[0025] Therefore, the present invention significantly increases carbon yield and thus significantly reduces the cost of any biotechnological process operated by the method and / or modified microorganisms according to the present invention.

[0026] The carboxylic acids referenced in this application are named equivalently by their salts (e.g., 2,4-dihydroxybutyrate) or acid forms (e.g., 2,4-dihydroxybutyric acid).

[0027] In the specification, enzyme activity also refers to the gene encoding the enzyme that has this activity. The use of gene nomenclature is not limited to a specific organism, but covers all corresponding genes and proteins in other organisms (such as microorganisms, functional analogs, functional variants and their functional fragments, as long as they retain the said enzyme activity).

[0028] In the context of this invention, the term "genetically modified microorganism" refers to a microorganism according to the invention that is not found in nature but is modified by introducing new genetic elements and / or by deleting and / or modifying endogenous genetic elements of a microorganism. Such modification can be produced by genome editing, in vivo evolution, genome recombination, directed or random mutagenesis, or any other strain engineering techniques known to those skilled in the art.

[0029] In the context of this invention, the terms “inactivation,” “alteration,” and “deletion” are used interchangeably and refer to the absence of a corresponding protein or the production of a nonfunctional corresponding protein, which is achieved, for example, by implementing gene silencing techniques corresponding to transcriptional gene silencing or post-transcriptional gene silencing, by introducing mutations in the promoter sequence controlling the expression of the gene encoding the protein and / or in the coding sequence of the protein, by altering the reading frame to introduce an early stop codon in the coding sequence of the protein, or by deleting the gene encoding the promoter controlling the expression of the gene encoding the protein and / or deleting the gene encoding the protein.

[0030] In the context of this invention, the terms “improved,” “optimized,” “enhanced,” and “increased” are used interchangeably and refer to higher activity (including enzyme activity and / or higher compound yield) compared to the same “unimproved” microorganism.

[0031] In the context of this invention, the terms "identical" and "homologous" are used interchangeably and refer to sequence identity between two polypeptides. When a position in each of two compared sequences is occupied by the same monomeric amino acid subunit, then the molecule is homologous or identical at that position. The percentage of identity between two sequences is a function of the number of corresponding positions shared by the two sequences, and is that number divided by the number of positions compared and multiplied by 100. For example, if 6 out of 10 positions in two matching sequences are the same, then the two sequences are 60% identical. As a general rule, the highest possible homology / identity is given by aligning two sequences for comparison. Various bioinformatics tools known to those skilled in the art can be used to align nucleic acid sequences, such as BLAST or FASTA.

[0032] The terms "code" and "code for" are used interchangeably and refer to the inherent properties of a specific nucleotide sequence in a polynucleotide (such as a gene, cDNA, or mRNA) that serves as a template for the synthesis of other polymers having a defined amino acid sequence and the resulting biological properties. Thus, if in a cell or other biological system, the transcription of a gene and the translation of the mRNA corresponding to that gene produce a protein, then the gene encodes that protein. The coding strand, whose nucleotide sequence is identical to the mRNA sequence and is generally described in sequence listings and databases, and the non-coding strand, which serves as a template for gene or cDNA transcription, can both be designated as encoding the protein or other product of that gene or cDNA. Therefore, in the context of this invention, the expression "fundamental homology" encompasses homology in terms of structure and / or amino acid composition and / or biological activity. Methods for performing sequence alignment and determining sequence identity are known to those skilled in the art, can be performed without inappropriate experiments, and allow for the definitive calculation of identity values. See, for example, Ausubel et al., ed. (1995); and the ALIGN program (Dayhoff (1978)). Many algorithms can be used to align sequences and determine sequence identity, including, for example, the homology alignment algorithm of Needleman et al. (1970); the local homology algorithm of Smith et al. (1981); the similarity search method of Pearson et al. (1988); the Smith-Waterman algorithm (Shpaer (1997)); and the BLASTP, BLASTN, and BLASTX algorithms (see Altschul et al.). Human, (1990)). Computerized programs using these algorithms are also available, and include, but are not limited to: ALIGN or Megalign (DNASTAR) software, or WU-BLAST-2 (Altschul et al., (1996)); or GAP, BESTFIT, BLAST (Altschul et al., (1997)), ibid., FASTA and TFASTA, available in the Genetic Computation Group (GCG) software package, 8th edition, Madison, Wis., USA; and CLUSTAL in the PC / Gene program of Intelligenetics, Mountain View, Calif. Those skilled in the art can determine suitable parameters for measuring alignment, including the algorithms required to achieve maximum alignment over the lengths of the sequences being compared. Preferably, default parameters determined by the program are used to determine sequence identity.Specifically, sequence identity can be determined using the Smith-Waterman homology search algorithm (Shpaer, 1997) implemented in the MSPRCH program (Oxford Molecular), which uses an affine vacancy search with the following search parameters: a vacancy opening penalty of 12 and a vacancy extension penalty of 1. Preferably, paired amino acid comparisons can be performed using the GAP program of the GCG sequence analysis software package from Genetics Computer Group, Inc., Madison, Wis., using a blosum 62 amino acid substitution matrix with a vacancy weight of 12 and a length weight of 2. For optimal alignment of two amino acid sequences, consecutive segments of the variant amino acid sequence may have additional or missing amino acid residues relative to the reference amino acid sequence. Consecutive segments used for comparison with the reference amino acid sequence will include at least 20 consecutive amino acid residues, and may be 30, 40, 50, or more. By setting the vacancy penalty, corrections can be made for increases in sequence identity associated with vacancy inclusions in the derivative amino acid sequence.

[0033] In the context of this invention, the term "carbon source" refers to a nutrient that provides the carbon necessary for the production of the enzymes according to the invention. This so-called "carbon source" is a low-molecular-weight molecule, i.e., with a molecular weight of 500 g / mol or lower, such as monosaccharides, dihalogenated sugars, or polyols (e.g., glucose).

[0034] In the context of this invention, the expression "via homoserine" means that homoserine is an intermediate substrate in the DHB production pathway, and / or relatedly an intermediate substrate in the chemical production methods of HMTB and / or HMSeB.

[0035] In the context of this invention, any amino acid substitution is a “conserved amino acid substitution” using L-amino acids, where one amino acid is replaced by another biologically similar amino acid. A conserved amino acid substitution is one that retains the total charge, hydrophobicity / hydrophilicity, and / or steric volume of the substituted amino acid. Examples of conserved substitutions are substitutions between the following groups: Gly / Ala, Val / Ile / Leu, Lys / Arg, Asn / Gln, Glu / Asp, Ser / Cys / Thr, and Phe / Trp / Tyr. For example, derivatives can differ by as few as 1 to 10 amino acid residues, such as 6-10, as few as 5, as few as 4, 3, 2, or even 1 amino acid residue.

[0036] In the context of this invention, the term "functional variant" refers to an enzyme that may have substantial sequence modifications compared to the sequence specifically described in this application, but still retains the original enzymatic activity. This also means that the enzyme sequence may contain fewer amino acids than the original sequence, but the truncated enzyme still retains the original enzymatic activity.

[0037] In the context of this invention, the expression "improved activity and / or substrate affinity" means that the enzyme before mutation could not utilize the substrate, and / or synthesized the reaction product at a maximum specific rate at least three times lower, and / or had an affinity for homoserine or OHB at least three times lower, and / or had a maximum specific activity at least three times higher than that of the natural substrate, and / or had an affinity for the natural substrate at least three times higher.

[0038] In the context of this invention, "nucleic acid sequence" refers to a DNA or RNA molecule in single-stranded or double-stranded form, preferably a DNA molecule. As used herein, "isolated DNA" refers to DNA that is not naturally occurring or no longer present in its original natural environment, such as DNA coding sequences associated with other regulatory elements in a chimeric gene, DNA transferred to another host cell, or artificially synthesized DNA sequences that have a different nucleotide sequence compared to any naturally occurring DNA sequence.

[0039] In the context of this invention, the term "microorganism" refers to any lower single-celled organism in which a chimeric gene, nucleic acid, or vector according to the invention can be introduced to produce 2,4-DHB.

[0040] In the context of this invention, the expression "functionally linked to each other" means that the elements of a chimeric gene are linked to each other such that the function of one element is influenced by the function of another. For example, a promoter is functionally linked to a coding sequence when it can affect the expression of that coding sequence. The construction of the chimeric gene according to the invention and the assembly of its various elements can be performed using techniques well known to those skilled in the art. The selection of the regulatory elements constituting the chimeric gene depends primarily on the host organism in which they must function, and those skilled in the art can select regulatory elements that are functional in a given host organism.

[0041] In the context of this invention, the term "functional" means being able to perform a function in a given host organism.

[0042] Preferably, the present invention relates to a method for preparing 2,4-dihydroxybutyrate from a carbon source, particularly from homoserine, via homoserine, comprising a two-step pathway, wherein at least one enzyme having aldolase activity as defined in EC4.1.2.28 is inactivated to prevent OHB degradation, defined as having the following characteristics, alone or in combination:

[0043] - The enzyme implemented in the first step is catalyzed by an enzyme encoded by an endogenous or heterologous gene;

[0044] - The enzyme implemented in the second step is catalyzed by an enzyme encoded by an endogenous or heterologous gene;

[0045] - The enzyme having aldolase activity is catalyzed by an enzyme encoded by an endogenous or heterologous gene;

[0046] - An enzyme having aldolase activity is encoded by the sequence shown in SEQ ID NO:1 or SEQ ID NO:3, or has the sequence shown in SEQ ID NO:2 or SEQ ID NO:4; or any sequence having at least 50%, advantageously at least 60%, at least 70%, or even at least 80%, preferably at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, or even at least 99% homology with said sequence SEQ ID NO:1, 3, 2, or 4;

[0047] - The enzyme that converts the primary amino group of homoserine to a carbonyl group to obtain OHB can be homoserine transaminase, homoserine dehydrogenase or homoserine oxidase.

[0048] Enzymes with high serine transaminase activity can be identified among enzymes with aspartate transaminase (EC2.6.1.1) activity, branched-chain amino acid transaminase (EC2.6.1.42) activity, or aromatic amino acid transaminase (EC2.6.1.57) activity.

[0049] -The carbon source is selected from the group consisting of monosaccharides, dihalogenated sugars, polyols and their mixtures;

[0050] - The carbon source is selected from the group consisting of glucose, sucrose, lactose, glycerol and mixtures thereof;

[0051] -The carbon source is glucose;

[0052] - Enzymes with high serine transaminase activity are encoded by genes iIvE, tyrB, aspC, araT, bcaT, alaC, and ARO8; preferably, enzymes with high serine transaminase activity can be branched-chain amino acid transaminases from Escherichia coli Ec-IlvE and Lactococcus lactis Ll-BcaT, and enzymes with high serine transaminase activity can be alanine-α-ketoglutarate transaminase from Escherichia coli Ec-AlaC, more preferably variant Ec-AlaCA142PY275D, and aromatic amino acid transaminases from Escherichia coli Ec-TyrB, Lactococcus lactis LI-AraT, and Saccharomyces cerevisiae Sc-Aro8, or aspartate transaminase from Escherichia coli Ec-AspC; or homologs thereof;

[0053] - An enzyme having high serine transaminase activity is encoded by the sequences shown in SEQ ID NO:5, SEQ ID NO:7, SEQ ID NO:9, SEQ ID NO:11, SEQ ID NO:13, SEQ ID NO:15; SEQ ID NO:126, SEQ ID NO:128, or corresponding to SEQ ID NO:6, SEQ ID NO:8, SEQ ID NO:10, SEQ ID NO:12, SEQ ID NO:14, SEQ ID NO:16, SEQ ID NO:127, SEQ ID NO:129; or any sequence having at least 50%, advantageously at least 60%, at least 70%, or even at least 80%, preferably at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, or even at least 99% homology with the sequences SEQ ID NO:5 to 16;

[0054] - The enzyme containing OHB reductase is (D)-lactate dehydrogenase, preferably from *Escherichia coli*; (L)-lactate dehydrogenase, preferably from *Lactococcus lactis*, *Oryctalagus cuniculus*, *Geobacillus stearothermophilus*, or *Bacillus subtilis*; (L)-malate dehydrogenase, preferably from *Escherichia coli*; or homologs thereof.

[0055] - An enzyme containing OHB reductase is a lactate dehydrogenase containing at least one mutation at positions V17, Q85, E89, I226, or A222, as defined in *Lactococcus lactis* tA (SEQ ID NO:17 or SEQ ID NO:18); and / or a malate dehydrogenase containing at least one mutation at positions I12, R81, M85, D86, V93, G179, T211, or M227, as defined in *Escherichia coli* Mdh (SEQ ID NO:19 or SEQ ID NO:20), wherein the naturally occurring amino acid at said position is replaced by any of the other 19 naturally occurring proteogenic amino acids, namely, alanine, arginine, asparagine, aspartic acid, cysteine, glutamic acid, glutamine, glycine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine, or valine.

[0056] - An enzyme containing OHB reductase is a lactate dehydrogenase containing at least one mutant at position V17, Q85N, or I226V, as defined in *Lactococcus lactis* tA (SEQ ID NO:17 or SEQ ID NO:18); and / or a malate dehydrogenase containing at least one mutant at position I12V, R81A, M85Q, D86S, or G179D, as defined in *Escherichia coli* Mdh (SEQ ID NO:19 or SEQ ID NO:20);

[0057] - An enzyme having OHB reductase is encoded by a sequence shown in SEQ ID NO:21, SEQ ID NO:23 or SEQ ID NO:25, or a lactate dehydrogenase corresponding to SEQ ID NO:22, SEQ ID NO:24 or SEQ ID NO:26; or any sequence having at least 50%, advantageously at least 60%, at least 70% or even at least 80%, preferably at least 90%, at least 95%, at least 96%, at least 97%, at least 98% or even at least 99% homology with said sequences SEQ ID NO:21 to 26;

[0058] - An enzyme having OHB reductase is encoded by the sequence shown in SEQ ID NO:27, SEQ ID NO:29, SEQ ID NO:31, SEQ ID NO:33, SEQ ID NO:35, SEQ ID NO:37, SEQ ID NO:39, SEQ ID NO:41, SEQ ID NO:43, or corresponding to malate dehydrogenases of SEQ ID NO:28, SEQ ID NO:30, SEQ ID NO:32, SEQ ID NO:34, SEQ ID NO:36, SEQ ID NO:38, SEQ ID NO:40, SEQ ID NO:42, SEQ ID NO:44; or any sequence having at least 50%, advantageously at least 60%, at least 70%, or even at least 80%, preferably at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, or even at least 99% homology to the sequences SEQ ID NO:27 to 44;

[0059] - Enzymes possessing OHB reductase activity are encoded by the sequences shown in SEQ ID NO:45, SEQ ID NO:47, SEQ ID NO:49, SEQ ID NO:51, SEQ ID NO:53, SEQ ID NO:55, SEQ ID NO:57, SEQ ID NO:59, SEQ ID NO:61, SEQ ID NO:63, SEQ ID NO:65, SEQ ID NO:67, SEQ ID NO:69, SEQ ID NO:71, SEQ ID NO:73, SEQ ID NO:75, SEQ ID NO:77, SEQ ID NO:79 or SEQ ID NO:81; or correspond to the sequences shown in SEQ ID NO:46, SEQ ID NO:48, SEQ ID NO:50, SEQ ID NO:52, SEQ ID NO:54, SEQ ID NO:56, SEQ ID NO:58, SEQ ID NO:60, SEQ ID NO:62, SEQ ID NO:64, SEQ ID NO:66, SEQ ID NO:68, SEQ ID NO:70, SEQ ID NO:81. SEQ ID NO:72, SEQ ID NO:74, SEQ ID NO:76, SEQ ID NO:78, SEQ ID NO:80 or SEQ ID NO:82; or any sequence having at least 50%, advantageously at least 60%, at least 70% or even at least 80%, preferably at least 90%, at least 95%, at least 96%, at least 97%, at least 98% or even at least 99% homology with the sequences SEQ ID NO:45 to 82;

[0060] - An enzyme with the same activity (any OHB reductase) according to the present invention, or an enzyme that converts the primary amino group of homoserine to a carbonyl group to obtain OHB, or an enzyme having at least about 50%, 70% or 85% amino acid sequence identity, preferably at least about 85% amino acid sequence identity, more preferably at least about 90% amino acid sequence identity, even more preferably at least about 95% amino acid sequence identity, and even more preferably an enzyme with aldolase activity having 98% or even 99% amino acid sequence identity;

[0061] - Enhancement of the activity of the enzyme catalyzing the first and / or second steps of the method of the present invention. This enhancement can be measured by the enzyme assays described in the examples. The improvement of the enzyme can be obtained by at least one mutation, said mutation i) improving the activity and / or substrate affinity of the mutant enzyme to homoserine or OHB, and / or ii) decreasing the activity and / or substrate affinity of the mutant enzyme to its natural substrate.

[0062] This invention also relates to a method for preparing hydroxymethionine (HMTB) and / or hydroxyselenomethionine (HMSeB) from a carbon source, particularly homoserine, via homoserine. This method comprises a two-step approach, as defined above, wherein at least one enzyme having aldolase activity as defined in EC 4.1.2.28 is inactivated to prevent OHB degradation. Indeed, as described in documents EP 2054 382 and EP 1 778 706, 2,4-DHB is converted to HMTB or HMSeB by cyclization followed by the addition of sodium thiolate or lithium methaneselenate, respectively.

[0063] This invention also relates to chimeric genes comprising at least one functional promoter in a host organism, a polynucleotide encoding any enzyme that catalyzes the first and second steps of the method defined according to the invention, and a terminator element functional in the same host organism, all functionally linked to each other. The various elements that a chimeric gene may contain are primarily elements that regulate protein transcription, translation, and maturation, such as promoters, sequences encoding signal peptides or transport peptides, or terminator elements constituting polyadenylation signals, and secondarily, polynucleotides encoding proteins.

[0064] The chimeric gene according to the invention may contain a constitutive or inducible promoter. For example, the promoter for expression in bacteria may be selected from the following promoters. For expression in *Escherichia coli*, lac, trp, lpp, phoA, recA, araBAD, prou, cst-I, tetA, cadA, nar, tac, trc, lpp-lac, Psyn, cspA, PL, PL-9G-50, PR-PL, T7, [λ]PL-PT7, T3-lac, T5-lac, T4 gene 32, nprM-lac, VHb and protein A promoters or the Ptrp promoter (WO 99 / 64607) may be mentioned. For expression in Gram-positive bacteria such as *Corynebacterium* or *Streptomyces*, PtipA or PS1 and PS2 (FR91 / 09870) promoters or those described in application EP0629699A2 may be mentioned. For expression in yeast and fungi, the following promoters can be mentioned: Kluyveromyces lactis PLAC4 promoter or Kluyveromyces lactis Ppgk promoter (FR 91 / 05294), Trichoderma reesei tef1 or cbh1 promoter (WO 94 / 04673), Penicillium funiculosum his, csl or apf promoter (WO00 / 68401), and Aspergillus niger gla promoter.

[0065] According to the present invention, chimeric genes may also contain other regulatory sequences located between the promoter and the coding sequence, such as transcriptional activators (enhancers).

[0066] Therefore, in one specific embodiment, the chimeric gene of the present invention comprises, at least in the transcriptional direction, a functionally linked promoter regulatory sequence in the host organism, a nucleic acid sequence encoding a polynucleotide encoding any enzyme that catalyzes the first and second steps of the method defined according to the present invention, and a terminator regulatory sequence in the host organism.

[0067] This invention also relates to cloning and / or expression vectors comprising a chimeric gene according to the invention or a nucleic acid sequence according to the invention. The vector according to the invention is used to transform a host organism and express in that organism any enzyme catalyzing the first and / or second steps of the method of the invention. The vector can be a plasmid, granule, bacteriophage, or virus. Preferably, the transformation vector according to the invention is a plasmid. Generally, the main characteristic of such a vector should be its ability to self-sustain and self-replicate within the cells of the host organism, particularly due to the presence of the origin of replication, and its ability to express in that organism any enzyme catalyzing the first and / or second steps of the method of the invention. To stably transform the host organism, the vector can also be integrated into the genome. The selection of such a vector and the technique for inserting the chimeric gene according to the invention into the vector are part of the common knowledge of those skilled in the art. Advantageously, in addition to the chimeric gene according to the invention, the vector used in this invention also contains a chimeric gene encoding a selectability marker. This selectability marker makes it possible to select host organisms that are effectively transformed, i.e., those host organisms that have integrated the vector. According to a particular embodiment of the invention, the host organism to be transformed is bacteria, yeast, or fungi. Among the optional biomarkers that can be used, biomarkers containing antibiotic resistance genes, such as the hygromycin phosphotransferase gene, may be mentioned. Other biomarkers may be genes complementary to auxotrophs, such as pyrA, pyrB, pyrG, pyr4, arg4, argB, and trpC genes, molybdopterin synthase genes, or acetamase genes. Genes encoding easily identifiable enzymes such as GUS enzymes, or genes encoding pigments or enzymes regulating pigment production in transformed cells, may also be mentioned. In particular, such optional biomarker genes are described in patent applications WO 91 / 02071, WO 95 / 06128, WO 96 / 38567, and WO 97 / 04103.

[0068] The improved / optimized methods for the production of DHB and / or related HMTB and / or HMSeB according to the present invention, particularly the chemical production methods of HMTB and / or HMSeB, can be implemented in wild-type microorganisms (i.e., non-genetically modified microorganisms) or in genetically modified microorganisms that have been genetically modified, i.e., different from those genetically modified according to the present invention. Such genetic modification can lead to an increase in the yield / concentration of intermediates involved in the production of DHB and / or relatedly the chemical production of HMTB and / or HMSeB. For example, it can be microorganisms genetically modified to produce homoserine in higher amounts / yields than wild-type microorganisms.

[0069] More specifically, modified microorganisms for preparing 2,4-DHB and / or relatedly HMTB and / or HMSeB from a carbon source via homoserine, particularly homoserine, the preparation of which comprises a two-step pathway including:

[0070] - The first step involves catalyzing the conversion of the primary amino group of homoserine to a carbonyl group using an enzyme with homoserine transaminase activity as defined in EC2.6.1.1, EC2.6.1.42, or EC2.6.1.57 to obtain 2-oxo-4-hydroxybutyrate.

[0071] The second step involves reducing the obtained 2-oxo-4-hydroxybutyrate with an enzyme possessing OHB reductase activity to obtain 2,4-dihydroxybutyrate. The enzyme possessing OHB reductase activity is a lactate dehydrogenase as defined in EC 1.1.1.27 or EC 1.1.1.28, or a malate dehydrogenase as defined in EC 1.1.1.37, EC 1.1.1.82, or EC 1.1.1.299.

[0072] At least one of the enzymes having aldolase activity, as defined in EC4.1.2.28, is inactivated to prevent OHB degradation.

[0073] Preferably, the present invention relates to modified microorganisms for improving / optimizing the production of DHB from a carbon source via homoserine and / or relatedly producing HMTB and / or HMSeB (particularly chemical production methods of HMTB and / or HMSeB), said production comprising a two-step pathway, wherein at least one enzyme having aldolase activity as defined in EC4.1.2.28 is inactivated to prevent OHB degradation, defined as having, alone or in combination, the following characteristics:

[0074] - The enzymes involved in the two steps are those mentioned above;

[0075] - The microorganisms mentioned are fungi, such as Penicillium, Aspergillus, and more particularly Aspergillus flavus, Chrysosporium or Trichoderma.

[0076] - The microorganisms are yeasts, especially yeasts of the families Saccharomycetaceae, Pichiaceae, or Schizosaccharomycetaceae, with the most preferred being Saccharomyces cerevisiae, Schizosaccharomyces pombe, Kluyveromyces lactis, Kluyveromyces marxianus, or Pichia jadinii, Pichia stipitis, or Pichia pastoris.

[0077] The microorganism is a bacterium, preferably selected from the families Enterobacteriaceae, Clostridiaceae, Bacillaceae, Streptomycetaceae, Streptococcaceae, Methylobacteriacae, and Corynebacteriaceae, with most preferably Escherichia coli, Bacillus subtilis, Corynebacterium glutamicum, Clostridium acetobutylicum, Methylobacterium extorquens, or Lactococcus lactis.

[0078] According to specific embodiments, the present invention also relates to microorganisms modified with at least one chimeric gene according to the present invention, said chimeric gene being integrated into their genome or carried on extrachromosomal genetic elements, such as plasmids. In a more specific aspect of the invention, the transformed host organism comprises a nucleic acid encoding a polypeptide of the present invention that converts a primary amino group of homoserine to a carbonyl group to obtain OHB, and / or a nucleic acid encoding a polypeptide that reduces OHB in 2,4-DHB, or a chimeric gene comprising a nucleic acid encoding a polypeptide that converts a primary amino group of homoserine to a carbonyl group to obtain OHB, and / or an OHB reductase or an expression vector comprising a polypeptide encoding a primary amino group of homoserine to a carbonyl group to obtain OHB, or a polypeptide having OHB reductase activity.

[0079] In another aspect of the invention, a synthetic pathway for converting homoserine to DHB and / or consequently producing HMTB and / or HMSeB (particularly through chemical methods of HMTB and / or HMSeB production) is expressed in microorganisms with enhanced homoserine production (i.e., homoserine production higher than that of the parent or wild-type organism). Enhanced homoserine production in microorganisms can be achieved by: i) overexpressing aspartate kinase, aspartate semialdehyde dehydrogenase, and homoserine dehydrogenase; ii) insensitizing aspartate kinase to product inhibition induced by lysine, methionine, or threonine; and iii) deleting a branch of the homoserine biosynthetic pathway. Overexpression of aspartate kinase, aspartate semialdehyde dehydrogenase, and homoserine dehydrogenase can be achieved by expressing these enzymes from a multicopy plasmid under the control of a suitable constitutive or inducible promoter. Alternatively, overexpression of the enzymes can be achieved by deleting transcriptional repressors that restrict transcription of the genes encoding aspartate kinase, aspartate semialdehyde dehydrogenase, and homoserine dehydrogenase. By introducing appropriate mutations into the amino acid sequence of aspartate kinase, the aspartate kinase can be made insensitive to the inhibition of aspartate-derived amino acids. The metabolic pathway entry point of the homoserine biosynthesis pathway is catalyzed by enzymes with O-succinylhomoserine or O-acetylhomoserine synthase activity (entering methionine biosynthesis), homoserine kinase activity (entering threonine biosynthesis), or diaminopimelic acid decarboxylase activity (entering lysine biosynthesis). Deletion of the gene encoding the protein with the aforementioned enzyme activity avoids the formation of aspartate-derived amino acids, thus facilitating the formation of homoserine (…). Figure 1 ).

[0080] Therefore, the deletion of genes metA, thrB, and lysA in *E. coli* weakens the branched homoserine biosynthetic pathway. Increased activity of homoserine pathway enzymes in *E. coli* can be achieved, for example, through overexpression of the bifunctional aspartate kinase-homoserine dehydrogenase mutant thrA S345F (insensitive to threonine inhibition) and asd (both genes originating from *E. coli*); or through overexpression of the monofunctional aspartate kinase mutant lysC E250K (insensitive to lysine), asd (both genes originating from *E. coli*), and the homoserine dehydrogenase gene HOM6 from *Saccharomyces cerevisiae*.

[0081] The microorganisms of the present invention may also possess a reduced ability to export homoserine, which increases the intracellular availability of this amino acid. To achieve reduced homoserine export from the cell, a permease capable of exporting homoserine may be omitted. Such a permease can be identified by overexpressing a genomic library in the microorganism and culturing the microorganism at an inhibited concentration of homoserine or structurally similar amino acids such as threonine, leucine, or aspartic acid (Zakataeva et al., (1999)). At increased concentrations of any of these amino acids, overexpression of genes that promote microbial growth may be involved in homoserine export.

[0082] On the other hand, the microorganism of the present invention is an Escherichia coli carrying the homoserine efflux transporters rhtA, rhtB and / or rhtC without these transporters.

[0083] Therefore, the microorganisms of the present invention may also exhibit increased enzyme activity selected from phosphoenolpyruvate carboxylase, phosphoenolpyruvate carboxylkinase, isocitrate lyase, pyruvate carboxylase, and hexose cotransporter permease, and / or decreased enzyme activity selected from at least one of lactate dehydrogenase, alcohol dehydrogenase, acetate kinase, phosphoacetyltransferase, pyruvate oxidase, isocitrate lyase, fumarate, 2-oxoglutarate dehydrogenase, pyruvate kinase, malic acid kinase, phosphoglucose isomerase, phosphoenolpyruvate carboxylase, phosphoenolpyruvate carboxylkinase, pyruvate formate lyase, succinate semialdehyde dehydrogenase, sugar transport phosphotransferase, ketohydroxyglutarate aldolase, homoserine-O-succinyltransferase, homoserine kinase, homoserine efflux transporter, diaminopimelic acid decarboxylase, and / or methylglyoxal synthase.

[0084] On the other hand, the microorganism of the present invention is an Escherichia coli that overexpresses at least one gene selected from *Escherichia coli*, namely ppc, pck, aceA, galP, asd, thrA, metL, lysC; *Lactococcus lactis*, namely pycA; and / or has at least one gene deletion selected from *ldhA*, *adhE*, *ackA*, *pta*, *poxB*, *focA*, *pfIB*, *sad*, *gabABC*, *sfcA*, *maeB*, ppc, *pykA*, *pykF*, *mgsA*, *sucAB*, *ptsL*, *ptsG*, *pgi*, *fumABC*, *aldA*, *lldD*, *icIR*, *metA*, *thrB*, lysA*, *eda*, *rhtA*, *rhtB*, *rhtC*.

[0085] The present invention also includes methods for producing 2,4-DHB and / or relatedly producing HMTB and / or HMSeB, particularly methods for producing HMTB and / or HMSeB from 2,4-DHB produced according to the present invention, comprising the following steps:

[0086] - Culturing the modified microorganisms of the present invention in a suitable culture medium; and

[0087] - and / or recover 2,4-DHB from the culture medium. The 2,4-DHB can be further purified. Furthermore, HMTB and / or HMSeB can be produced according to the method of the present invention by adding methyl sulfide or methyl selenide donors to 2,4-DHB.

[0088] Product separation and purification are crucial factors that significantly impact overall process efficiency and product cost. Product recovery methods typically include cell separation, as well as product purification, concentration, and drying steps.

[0089] Cell separation Ultrafiltration and centrifugation can be used to separate cells from fermentation media. Cell separation from fermentation media is often complicated by the high viscosity of the medium. Therefore, additives such as inorganic acids or alkali metal salts, or heating the culture broth, can be added to optimize cell separation.

[0090] Product recyclingVarious ion-exchange chromatography methods can be used to separate DHB and / or HMTB and / or HMSeB before or after biomass removal. These include the use of primary cation exchange resins that facilitate product separation based on the isoelectric point of the products. Typically, the resin is packed with solution, and the retained products are eluted separately after the pH of the eluent is increased (e.g., by the addition of ammonium hydroxide). Another possibility is the use of ion-exchange chromatography utilizing stationary or simulated moving bed resins. Different chromatographic steps may need to be combined to obtain sufficient product purity. These purification methods are more economical than expensive crystallization steps and also offer additional advantages and flexibility regarding the final product form.

[0091] Product concentration and drying The purification process may also include a drying step, which can involve any suitable drying apparatus, such as a spray granulator, spray dryer, drum dryer, rotary dryer, and tunnel dryer. Concentrated DHB and / or HMTB and / or HMSeB solutions can be obtained by heating broth with steam at 130°C under reduced pressure using a multipurpose concentrator or thin-film evaporator. Attached Figure Description

[0092] Figure 1 Methods for preparing 2,4-DHB from homoserine and relatedly preparing HMTB and HMSeB include a two-step approach, employing a first step of converting the primary amino group of homoserine to a carbonyl group to obtain OHB, and a second step of reducing the obtained OHB to 2,4-DHB.

[0093] Figure 2 OHB degradation pathway

[0094] Figure 3 In vivo DHB consumption and formic acid production after incubation of wild-type E. coli cells and "monoaldolase-deficient" E. coli cells with DHB.

[0095] Figure 4 In vivo DHB consumption and formic acid production after incubation of wild-type E. coli cells and "polyaldolase-deficient" E. coli cells with DHB.

[0096] Figure 5 In vivo DHB and formate production during glucose fermentation of "aldolase-deficient" Escherichia coli "DHB-producing" strains. Detailed Implementation

[0097] Example

[0098] 1-Materials and Methods

[0099] 1.1 Strains and Culture Conditions

[0100] The strains used in the examples are detailed in Table 1.

[0101] Table 1: Escherichia coli strains used in this study

[0102]

[0103]

[0104] Unless otherwise specified (as described below), Escherichia coli strain K-12 substrain MG1655 (ATCC n°47046) was cultured in M9 medium containing 0.2% D-xylose as a carbon source. When needed (as described below), the medium also contained 50 mM DHB (a racemic mixture of L and D DHB ammonium).

[0105] 1.2-Gene Deletion

[0106] A modified P1 transduction protocol using yagE, yjhH, garL, dgoA, yfaU, and eda (Thomason et al., 2007) was used to delete the gene encoding pyruvate aldolase (i.e., the aldolase according to the present invention) in Escherichia coli strain K-12 substrain MG1655, and the mhpE was processed using CrispR-Cas9 technology (Jiang et al., 2015).

[0107] 1.3-P1 transduction

[0108] Preparation of pyrolysis products

[0109] Add 25 μL of sterile glucose solution (20% w / v) and 25 μL of sterile CaCl2 1M to 5 mL of lysate broth (LB), and inoculate the mixture with 100 μL of a saturated culture containing the desired deletion strain from the KEIO collection (Baba et al., 2006), and incubate at 37 °C with shaking. After 45 min, add 100 μL of P1 stock solution, and then incubate for 3 h or until lysis is complete. Centrifuge the lysate at 4000 g, 4 °C for 10 min, and filter to sterilize the supernatant. The lysate can be stored at 4 °C for several years without loss of performance.

[0110] Transduction

[0111] Centrifuge 1 mL of saturated culture of the desired target strain and resuspend in 100 μL of sterile P1 salt solution (10 mM CaCl2, 5 mM MgSO4) and 100 μL of P1 lysate carrying the desired mutation. Incubate the cells at 30 °C without stirring for 30 min, then add 100 μL of sterile 1 M sodium citrate solution to stop transduction, followed by the addition of 1 mL of LB. Incubate the cells at 37 °C for 1 h, then plate them on LB kanamycin (50 μg / mL).

[0112] Kanamycin resistance Express Box elimination

[0113] Given that the kanamycin resistance expression cassette in the KEIO library contains FRT sites on each side, this cassette can be easily disrupted using the Flp enzyme from Saccharomyces cerevisiae (Cox, 1983), which is encoded on the pCP20 plasmid (Cherepanov and Wackernagel, 1995). Cells carrying the kanamycin resistance expression cassette, but not the target gene, can be transformed with pCP20 using a protocol described elsewhere (Chung et al., 1989).

[0114] Multiple clones that had grown due to transduction were picked from LB Kan plates and seeded with 1 mL of LB. When the culture reached an OD600 of approximately 0.3, the cells were centrifuged and resuspended in TSS solution (10 g / L tryptone, 5 g / L yeast extract, 10 g / L NaCl, 10 g / L PEG 10 kDa, 5 g / L DMSO, and 20 mM MgCl2). The cells were placed on ice for 15 minutes, followed by the addition of 100–150 ng of pCP20 plasmid formulation. After 15 minutes on ice, the cells were subjected to heat shock at 42 °C for 1 min, followed by the addition of 1 mL of LB. Phenotypic expression of the pCP20-encoded β-lactamase was induced by stirring at 30 °C for 1–2 h, followed by plating on LB ampicillin (100 μg / mL).

[0115] Pick multiple clones from LB ampicillin plates and inoculate them with 1 mL of LB, then incubate at 42°C for 6–8 h. Streak a few μL of the culture onto an LB plate and incubate overnight at 37°C.

[0116] Selected clones were individually picked from LB plates and streaked onto LB Amp, LB kan, and LB plates to counterselect clones that had lost resistance to both ampicillin and kanamycin. Following the manufacturer's instructions and using the primers listed in Table 1, [the process was carried out]. 2X Master Mix and standard buffer (BioLabs) were used to verify deletions on antibiotic-sensitive clones by colony PCR. In all strains with one of multiple deletions, gene deletions were verified by PCR amplification using primers described in Table 3.

[0117] Table 2 - Plasmids used

[0118]

[0119] Table 3 - Primers used for PCR verification of gene deletions

[0120]

[0121]

[0122] 1.4-CrispR-Cas9 missing

[0123] As described elsewhere, mphE deletion was performed using CrispR-Cas9 (Jiang et al., 2015). First, the target strain was transformed with a pCas plasmid carrying the Cas9 encoding gene and the λred recombination mechanism, as previously described. pTarMHPE (a gRNA-producing plasmid) was constructed based on pTargetF and amplified using primers TM60 and TM34. Correct amplification was verified using primer TM35 via the Sanger sequencing service provided by EurofinsGenomics.

[0124] Construction of donor DNA

[0125] The repair fragment was obtained from the genomic Escherichia coli substrain MG1655 (K-12). In short, the mhpE region was amplified from the genomic DNA preparation of wild-type E. coli MG1655 using primers TM32 and TM33. The resulting fragment was 1873 bp, gel-purified (BioBasic EZ-10 centrifugal column DNA gel extraction kit), and phosphorylated at its 5' end using T4 polynucleotide kinase (BioLabs). The vector used was pZE13 (Expressys) digested with EcoRV (BioLabs) and dephosphorylated with Antarctic phosphatase (BioLabs). The vector and insert fragment were then ligated using HiT4 DNA ligase (BioLabs). The quality of the insert fragment was checked by sequencing using primers TM61 and TM62.

[0126] The resulting circular plasmid (pMHPE) was amplified using primers TM36 and TM37 to linearize and delete the target gene. It was then gel-purified, phosphorylated, and ligated to obtain pMHPEΔ containing the mhpE genomic region with the missing coding sequence. Finally, CrispR-Cas9 donor DNA was amplified from pMHPEΔ using primers TM32 and T33.

[0127] Genetic disruption using CRISPR-Cas9

[0128] The strain carrying the pCas plasmid was electroporated with 100 ng pTarget and 400 ng donor DNA. Briefly, 50 μL of overnight culture of the target strain was inoculated with 5 mL of LB Kan (50 μg / mL) arabinose (10 mM) and allowed to grow at 30 °C for 3–4 h or until the OD600 reached 0.5–0.7. Then, 2 mL was harvested and placed on ice, the pellet was washed once with 1 mL of ice-cold ddH2O, and once with 1 mL of ice-cold 10% glycerol. The cells were then resuspended in 50 μL of ice-cold 10% glycerol and transformed with 100 ng pTarget and 400 ng donor DNA under the following conditions: 2 mm cuvette, 2.5 kV, 200 Ω, 25 μF, and immediately resuspended at SOC at 30 °C overnight. All transformation reactions were plated on LB kanamycin-specific zirconia. If necessary, use primers mhpE_cas_F and mhpE_cas_R to verify correct genome editing before solidifying the plasmid and validating any remaining mutations.

[0129] Cloning and characterization of the gene encoding 1,5-pyruvate aldolase

[0130] Cloning methods

[0131] Seven genes encoding pyruvate aldolase were cloned into the pET28 vector (Addgene). Briefly, they were amplified from the start codon to a stop codon with a non-flanking 3' end, as shown in Table 4. This base extension, except for eda and mhpE, allowed the addition of an AvrII restriction site at the 5' end and an MfeI site at the 3' end of the genes. In the cases of eda and mhpE, different restriction site pairs were used, namely NdeI / XhoI and BamHI / HindIII, respectively. The amplicons were gel purified (BioBasic Gel Extraction Kit) before digestion with the corresponding restriction enzymes (BioLabs). The pET28 vector was amplified with primers that could be ligated accordingly, and the amplicons were digested with 1 μL of DpnI (BioLabs), then cleared (BioBasic Gel Extraction Kit) and digested with the cloning restriction enzyme. The digested product was purified and dephosphorylated using Antarctic phosphatase (BioLabs). Vector amplification allows for the addition of MfeI sites, as well as the deletion of the T7 tag and part of the MCS carried by the vector. For mhpE insertion, the pET28 vector is used directly for digestion.

[0132] The insert fragment and vector were ligated using HiT4 DNA ligase (BioLabs), and the results were transformed into... In 5-α cells (Biolabs), positive clones were amplified by PCR using primers TM89 and TM90 to check the correct length of the inserted fragment, which was then verified by sequencing.

[0133] Table 4 - List of primers used for cloning the gene encoding pyruvate aldolase in the pET28+ vector

[0134]

[0135]

[0136] Production and purification of recombinant proteins

[0137] The *E. coli* strain BL21(DE3) was transformed with a pET-28a(+) plasmid containing an aldolase gene (yagE, yjhH, yfaU, garL, dgoA, or eda). The isolated colonies were then cultured overnight at 37°C and 200 rpm in 14 mL culture tubes containing 5 mL of LB medium with 50 μg / mL kanamycin. The cells were then transferred to 500 mL Erlenmeyer flasks containing 100 mL of fresh LB medium with 50 μg / mL kanamycin, initially with an OD600 nm of approximately 0.1. The cultures were incubated at 37°C and 200 rpm until the OD600 nm reached approximately 0.6. At this stage, 1 mM IPTG was added to the medium, and the cells were incubated at 37°C and 200 rpm for 3 h. The cells were then centrifuged. The precipitate was resuspended in lysis buffer (Tris-HCl 50 mM pH 7.5, NaCl 300 mM, imidazole 5 mM) and sonicated for four cycles at 30% power for 30 seconds each, followed by intermittent cooling on an ice-ethanol mixture (Fisherbrand) for 30 seconds. Cell lysates were centrifuged at 14800 rpm for 15 min, and the supernatant was collected for purification. HIS-tagged proteins were purified using a cobalt TALON His-tagged purification resin (GE Healthcare). Proteins were eluted in 400 μL of Tris-HCl 50 mM pH 7.5, NaCl 300 mM, imidazole 200 mM. The purified enzyme was concentrated in 50 mM Tris-HCl buffer using a 10 kDa Amicon Ultra (Millipore) filter. Protein purification was verified by SDS-PAGE and further validated by spectrophotometry (Epoch Times). 2 Biotek) measured protein concentration, taking into account that 1 mg corresponds to 1 OD at 260 nm.

[0138] 1.5-Synthesis of 2-oxo-4-hydroxybutyrate (OHB)

[0139] As previously described, OHB was synthesized from D-homoserine (Walther et al., 2018). Briefly, 125 mM D-homoserine was incubated with porcine kidney D-amino acid oxidase (.25 U / mL, Sigma Aldrich) and bovine liver catalase (4400 U / mL, Sigma Aldrich) in 100 mM Tris buffer at pH 7.8 and 37°C for 90 min. (Amicon...) TM The reactants were purified on an ultracentrifuge filter (cutoff value 10 kDa, Millipore) to remove enzymes.

[0140] OHB was quantified by mixing 100 μL of purified reactant with 1 mL of 1 M boric acid and 1 M sodium arsenate at pH 6.5. The reactant was incubated at room temperature for 30 min, and the absorbance was read at 325 nm. The relationship between the absorbance and concentration of the ketone was calibrated using a pyruvate solution of known concentration (Wellner and Lichtenberg, 1971).

[0141] Kinetic characteristics of pyruvate aldolase on 1,6-OHB

[0142] The assay was performed by oxidizing NADH and using an excess LdhA-coupled enzyme on an Epoch reader. 2 Aldolase activity was measured at 340 nm at 37 °C in Biotek. The assay was performed in 250 μL of 50 mM Tris-HCl buffer (pH 7.5) containing 0.2 mM NADH in 5 mM MgCl2, using different concentrations of OHB from 0 to 13 mM. Since *E. coli* LDH is not commercially available, the enzyme was produced by the BL21(DE3) *E. coli* strain, which was transformed as described above using a pET-28a(+) plasmid containing *E. coli* ldhA. The transformed somatic cells were cultured overnight at 37 °C and 200 rpm in 500 mL Erlenmeyer flasks containing 100 mL of auto-inducible ZYM-5052 medium (Studier, 2005) containing 50 μg / mL kanamycin.

[0143] Evidence for the 2-OHB degradation pathway

[0144] The results are as follows Figure 3 As shown.

[0145] Data showed that incubation of *E. coli* in M9 medium containing 0.2% D-xylose in the presence of 50 mM DHB led to a significant accumulation of formic acid. The fact that over 70% of the consumed DHB was in the form of formate underscores the potential importance of the OHB degradation pathway. Figure 3 ).

[0146] 3-Identification of OHB aldolase

[0147] Since we hypothesized the presence of OHB aldolase activity in *E. coli*, we searched databases (particularly *Ecocyc*) and the literature for potential so-called pyruvate aldolases. This study allowed us to identify at least seven such aldolases that share some common versatility in reacting with OHB. He et al. (2020) constitutively expressed these aldolases to condense pyruvate and formaldehyde into OHB. Although aldolases catalyze reversible reactions, their in vivo functions may differ. Furthermore, although the genes encoding these pyruvate aldolases are present in the *E. coli* genome, it is not obvious whether they are expressed, or even if they are expressed, whether they can cleave OHB into pyruvate and formaldehyde in vivo. Therefore, we list seven pyruvate aldolases (Table 5) from previous literature studies (He, H et al., 2020; Clapés et al., 2010) that can cleave OHB into pyruvate and formaldehyde in *E. coli* or another microorganism.

[0148] Table 5 - OHB aldolases from Escherichia coli

[0149]

[0150] Transcriptome analysis during DHB fermentation identified four genes encoding aldolases listed in Table 5, which may be involved in OHB degradation (see Table 6). In particular, YagE and YjhH, encoding aldolases, were the most upregulated, and this upregulation coincided with the highest formic acid production during fermentation.

[0151] Table 6 - Gene expression of OHB aldolase in the presence of high levels of DHB

[0152] Aldolase gene Minimal Expression Maximum expression rhmA 0.98 1.14 garL 1.06 1,32 yagE 1.82 13,33 YjhH 1.25 3.33 yfaU 0.9 1.09

[0153] Therefore, we performed the required systematic deletions one by one and evaluated the impact of these single and combined deletions on the results of DHB formic acid production.

[0154] Kinetic characteristics of pyruvate aldolase on 4-OHB

[0155] To target the most potent intracellular OHB aldolases, we characterized five aldolases. We determined the OHB affinity (Km) and catalytic efficiency (Kcat / Km) of the YagE, YjhH, RhmA, DgoA, and GarL aldolases.

[0156] The results are shown in Table 7. It can be seen that the aldolase encoded by yfaU has the highest catalytic efficiency (kcat / Km), but this enzyme exhibits a ten-fold higher affinity than the YagE and YjhH aldolases. Furthermore, yfaU expression remained unchanged during DHB fermentation. These results suggest that YagE and YjhH are likely the best OHB aldolase candidates under our DHB production conditions.

[0157] Table 7. Kinetic characteristics of pyruvate aldolase on OHB.

[0158]

[0159] 5- The ability of the body to degrade OHB by lacking the gene encoding pyruvate aldolase

[0160] To assess the in vivo degradation capacity of OHB and DHB, we conducted experiments culturing DHB and *E. coli* cells in M9 medium containing 0.2% glucose under aerobic conditions. We measured formic acid production in strains lacking one of the genes encoding these aldolases.

[0161] The results are as follows Figure 3 As shown.

[0162] The combined absence of yagE and yjhH significantly reduced DHB consumption and formic acid production (see [link]). Figure 4 This indicates that these enzymes are the main cause of OHB loss during the fermentation production of DHB (see [link]). Figure 5 However, the absence of the seven aldolases listed in Table 1 resulted in a complete loss of OHB production from DHB (see Table 1). Figure 4 ).

[0163] Because these results showed a strong reduction in formic acid, we then assessed whether the deletion of yagE and yjhH in engineered E. coli expressing the DHB production pathway increased DHB production during glucose fermentation. These engineered strains (i.e., strains 1160, 1526, 1527, and 1528, see Table 1) contain pZA23-thrA S345F _alaC A142P Y275D _mdh R81A M85E I12V G179D D86S The plasmid, in which the mutation in thrA alleviates the threonine feedback inhibition on ThrA, has had its alaC mutation obtained through in vivo evolutionary engineering, resulting in AlaC. A142PY275D The variant exhibits 100-fold catalytic activity toward homoserine (Bouzon et al., 2017), and mutations in E. coli mdh are derived from protein engineering (Frazao et al., 2018) to produce optimized MDH mutants with reduced OHB.

[0164] Inactivation of yjhH (strain 1526) resulted in a significant increase in DHB production (>10%) by increasing glucose yield, while eliminating formic acid production (see [link]). Figure 5 This effect was less pronounced after yagE inactivation (strain 1527). These results are consistent with aldolase biochemical data showing that YjhH has higher catalytic efficiency than YagE.

[0165] However, the combination of yjhH and yagE deletions (strain 1528) resulted in a better increase in DHB yield (>15%). We summarized the synergistic effect caused by dual inactivation, which may be because the loss of one aldolase can be compensated for by the increase of the other. It should be noted that only DHB yield is a relevant performance indicator because formic acid can be consumed by the cell as a carbon source under DHB production conditions (i.e., glucose-limited).

[0166] 6-Conclusion

[0167] Of all the OHB aldolases found in E. coli, we clearly identified YagE and YjhH aldolases as the most efficient at degrading OHB in vivo and thus partially removing it from the DHB pathway. Conversely, other aldolases such as RhmA and GarL do not have any function against OHB in vivo, but they appear to be very active in condensing formaldehyde and pyruvate to OHB.

[0168] Finally, in the application of DHB production from OHB, we state that the inactivation of yjhH and / or yagE (both EC4.1.2.28) is the reason for the significant increase in DHB yield (more specifically, glucose yield, more precisely, production by eliminating formic acid).

[0169] References

[0170] Altschul,SF,Gish,W.,Miller,W.,Myers,EW&Lipman DJ,1990.Basiclocal alignment search tool.J Mol Biol 215,403-10.

[0171] Altschul, SF & Gish, W., 1996. Local alignment statistics. Methods Enzymol 266, 460-80.

[0172] Altschul,SF,Madden,TL A.A.,Zhang,J.,Zhang,Z.,Miller,W.&Lipman,D.J.,1997.Gapped BLAST and PSI-BLAST:a new generation of proteindatabase search programs.Nucleic Acids Res 25,3389-402.

[0173] Ausubel,M.,Brent,R.,Kingston,R.E,Moore,D.D.,Seidman,J.G.,Smith,J.A.,&Struhl,K.,1995.Current Protocols in Molecular Biology,Chapter 19.GreenePublishing and Wiley-Interscience,New York.

[0174] Baba,T.,Ara,T.,Hasegawa,M.,Takai,Y.,Okumura,Y.,Baba,M.,Datsenko,K.A.,Tomita,M.,Wanner,B.L.,Mori,H.,2006.Construction of Escherichia coli K-12in-frame,single-gene knockout mutants:the Keio collection.Mol Syst Biol 2.

[0175] Bailey,J.E.,1991.Toward a science of Metabolic Engineering.Science252,1668-1975.

[0176] Bouzon,M.,Perret,A.,Loreau,O.,Delmas,V.,Perchat,N.,Weissenbach,J.,Taran,F.&Marlière,P.,2017.Asynthetic alternative to canonical one-carbonmetabolism.Acs Synthetic Biology.61520–1533.

[0177] Cherepanov,P.P.,Wackernagel,W.,1995.Gene disruption in Escherichiacoli:TcR and KmR cassettes with the option of Flp-catalyzed excision of theantibiotic-resistance determinant.Gene 158,9–14.

[0178] Chung,C.T.,Niemela,S.L.,Miller,R.H.,1989.One-step preparation ofcompetent Escherichia coli:transformation and storage of bacterial cells inthe same solution.Proc.Natl.Acad.Sci.U.S.A.86,2172–2175.

[0179] Clapés P.,Fessner W.D.,Sprenger G.A.&Samland A.K.2010.Recent progressin stereoselective synthesis with aldolases.Curr Opin Chem Biol.14,154-67.

[0180] Cox,M.M.,1983.The FLP protein of the yeast 2-eum plasmid:Expressionof a eukaryotic genetic recombination system in Escherichiacoli.Proc.Natl.Acad.Sci.USA5.

[0181] Dayhoff,M.O.,Schwartz,R.M.&Orcutt,B.C.,1978.Amodel of evolutionarychange in proteins.Atlas of Protein Sequence and Structure.

[0182] Deck,P.,Exner,K.&Buschhaus,B.2008.Method for the production of D,L-hydroxy-4-alkylthio butyric acid.Edited by B.AG.

[0183] C.,Topham C.,Malbert Y., J.M.&Walther T.2018.Rationalengineering of a malate dehydrogenase for microbial production of2,4-dihydroxybutyric acid via homoserine pathway Biochem J 475,3887-3901.

[0184] He H., R., M.,Marlière P.&Bar-Even A.2020.An optimizedmethanol assimilation pathway relying on promiscuous formaldehyde-condensingaldolases in E.coli.Metab Eng.60,1-13.

[0185] Jiang,Y.,Chen,B.,Duan,C.,Sun,B.,Yang,J.&Yang,S.,2015.MultigeneEditing in the Escherichia coli Genome via the CRISPR-Cas9 System.ApplEnviron Microbiol 81,2506–2514.

[0186] Needleman,S.B.&Wunsch,C.D.,1970.A general method applicable to thesearch for similarities in the amino acid sequence of two proteins.J Mol Biol48,443-53.

[0187] Pearson,W.R&Lipman,D.J.,1988.Improved tools for biological sequencecomparison.Proc Natl Acad Sci U S A85,2444-8.

[0188] Shpaer,E.G.,1997.GeneAssist.Smith-Waterman and other databasesimilarity searches and identification of motifs.Methods Mol Biol 70,173-87.

[0189] Smith,T.F.&Waterman M.S.,1981.Adv.Appl.Math.2,482.

[0190] Studier FW.(2005)Protein production by auto-induction in high-densityshaking cultures.Prot.Exp.Pur.41,207-234.

[0191] Thomason,L.C.,Costantino,N.,Court,D.L.,2007.E.coli GenomeManipulation by P1 Transduction.CP Molecular Biology 79.

[0192] Walther,T.,Calvayrac,F.,Malbert,Y.,Alkim,C.,Dressaire,C.,Cordier,H., J.M.,2018.Construction of a synthetic metabolic pathway for theproduction of 2,4-dihydroxybutyric acid from homoserine.Metabolic Engineering45,237–245.

[0193] Wellner,D.,Lichtenberg,L.A.,1971.[218a]Assay of amino acid oxidase,in:Methods in Enzymology.Elsevier,pp.593–596.

[0194] Werpy,T.&Petersen,G.,2004.Top Value Added Chemicals from Biomass:Volume I--Results of Screening for Potential Candidates from Sugars andSynthesis Gas.Pacific Northwest National Laboratory(PNNL)and NationalRenewable Energy Laboratory(NREL).

[0195] Zakataeva N.P.,Aleshin V.V.,Tokmakova I.L.,Troshin P.V.&LivshitsV.A.,1999.The novel transmembrane Escherichia coli proteins involved in theamino acid efflux.FEBS Lett.452,228-32.

Claims

1. A method for preparing 2,4-dihydroxybutyrate from a carbon source via homoserine, comprising a two-step approach: - The first step involves catalyzing the conversion of the primary amino group of homoserine to a carbonyl group using an enzyme with homoserine transaminase activity as defined in EC2.6.1.1, EC2.6.1.42, or EC2.6.1.57 to obtain 2-oxo-4-hydroxybutyrate. - The second step involves reducing the obtained 2-oxo-4-hydroxybutyrate (OHB) to 2,4-dihydroxybutyrate using an enzyme with OHB reductase activity. This enzyme is a lactate dehydrogenase as defined in EC1.1.1.27 or EC1.1.1.28, or a malate dehydrogenase as defined in EC1.1.1.37, EC1.1.1.82, or EC1.1.1.

299. At least one of the enzymes having aldolase activity, as defined in EC4.1.2.28, is inactivated to prevent OHB degradation.

2. The method according to claim 1, wherein the enzyme having aldolase activity is encoded by the sequence shown in SEQ ID NO:1 or SEQ ID NO:3, or by the sequence shown in SEQ ID NO:2 or SEQ ID NO:4, or by any sequence encoding an enzyme having aldolase activity and having at least 50% homology with said sequence.

3. The method according to any one of the preceding claims, wherein the enzyme having high serine transaminase activity is encoded by the genes iIvE, tyrB, aspC, araT, bcaT, alaC, and ARO8.

4. The method according to claim 3, wherein the enzyme having high serine transaminase activity: -A sequence shown in SEQ ID NO:5, SEQ ID NO:7, SEQ ID NO:9, SEQ ID NO:11, SEQ ID NO:13, SEQ ID NO:15, SEQ ID NO:126, SEQ ID NO:128, or any sequence encoding an enzyme having high serine transaminase activity and having at least 50% homology with said sequences; or - Corresponding to SEQ ID NO:6, SEQ ID NO:8, SEQ ID NO:10, SEQ ID NO:12, SEQ ID NO:14, SEQ ID NO:16, SEQ ID NO:127, SEQ ID NO:129, or any sequence having at least 50% homology with the said sequence.

5. The method according to any one of the preceding claims, wherein the enzyme having OHB reductase is (D)-lactate dehydrogenase from Escherichia coli, (L)-lactate dehydrogenase from Lactococcus lactis, from rabbits, from Bacillus stearothermophilus, or from Bacillus subtilis, (L)-malate dehydrogenase from Escherichia coli, or a homolog thereof.

6. The method according to claim 5, wherein the enzyme having OHB reductase is: - Contains at least one mutated lactate dehydrogenase at positions V17, Q85, E89, I226, or A222, as defined in Lactococcus lactis tA (SEQ.ID NO:18); - Contains at least one mutated malate dehydrogenase at positions I12, R81, M85, D86, V93, G179, T211, or M227, as defined in Escherichia coli Mdh (SEQ ID NO:20).

7. The method according to claim 5 or 6, wherein the enzyme having OHB reductase activity is: - Encoded by a sequence shown in SEQ ID NO:45, SEQ ID NO:47, SEQ ID NO:49, SEQ ID NO:51, SEQ ID NO:53, SEQ ID NO:55, SEQ ID NO:57, SEQ ID NO:59, SEQ ID NO:61, SEQ ID NO:63, SEQ ID NO:65, SEQ ID NO:67, SEQ ID NO:69, SEQ ID NO:71, SEQ ID NO:73, SEQ ID NO:75, SEQ ID NO:77, SEQ ID NO:79 or SEQ ID NO:81, or any sequence having at least 50% homology with said sequence; or - Corresponding to SEQ ID NO:46, SEQ ID NO:48, SEQ ID NO:50, SEQ ID NO:52, SEQ ID NO:54, SEQ ID NO:56, SEQ ID NO:58, SEQ ID NO:60, SEQ ID NO:62, SEQ ID NO:64, SEQ ID NO:66, SEQ ID NO:68, SEQ ID NO:70, SEQ ID NO:72, SEQ ID NO:74, SEQ ID NO:76, SEQ ID NO:78, SEQ ID NO:80 or SEQ ID NO:82, or any sequence having at least 50% homology with the said sequence.

8. A modified microorganism for preparing 2,4-DHB from a carbon source via homoserine, the preparation comprising a two-step approach, the approach comprising: - The first step involves catalyzing the conversion of the primary amino group of homoserine to a carbonyl group using an enzyme with homoserine transaminase activity as defined in EC2.6.1.1, EC2.6.1.42, or EC2.6.1.57 to obtain 2-oxo-4-hydroxybutyrate. - The second step involves reducing the obtained 2-oxo-4-hydroxybutyrate using an enzyme with OHB reductase activity to obtain 2,4-dihydroxybutyrate. The enzyme with OHB reductase activity is a lactate dehydrogenase as defined in EC1.1.1.27 or EC1.1.1.28, or a malate dehydrogenase as defined in EC1.1.1.37, EC1.1.1.82, or EC1.1.1.

299. At least one of the enzymes having aldolase activity, as defined in EC4.1.2.28, is inactivated to prevent OHB degradation.

9. The modified microorganism according to claim 8, wherein the enzyme having high serine transaminase activity, the enzyme having OHB reductase activity, and the enzyme having aldolase activity are those enzymes according to any one of claims 2 to 8.

10. The modified microorganism according to claim 8 or 9, wherein the production of homoserine is enhanced compared to the same non-transformed microorganism.

11. The modified microorganism according to any one of claims 8 to 10, wherein the activities of aspartate kinase, aspartate semialdehyde dehydrogenase and / or homoserine dehydrogenase are enhanced.

12. The modified microorganism according to any one of claims 8 to 11, wherein the production of 2,4-DHB is enhanced compared to the following microorganisms: -The same non-transformed microorganism; and / or - A modified microorganism comprising a first step of converting the primary amino group of homoserine to a carbonyl group to obtain 2-oxo-4-hydroxybutyrate, and a second step of reducing the obtained 2-oxo-4-hydroxybutyrate to obtain 2,4-dihydroxybutyrate.

13. The modified microorganism according to any one of claims 8 to 12, wherein it is a bacterium, preferably selected from Enterobacteriaceae, Clostridiumceae, Bacillusaceae, Streptomycetes, Streptococci, Methylbacteriaceae, and Corynebacteriumceae, most preferably Escherichia coli, Bacillus subtilis, Corynebacterium glutamicum, Clostridium acetobutyricum, Methylobacterium thiocyanate, or Lactococcus lactis; a yeast, preferably selected from Saccharomycetes, Pichia pastoris, and Schizomycetes, most preferably Saccharomyces cerevisiae, Schizomyces cerevisiae, Kluyveromyces lactis, Kluyveromyces marx, or Pichia pastoris, Pichia jetting, Pichia trunk, or Pichia pastoris; or a fungus, preferably selected from Penicillium, Aspergillus, Aureosulcata, or Trichoderma.

14. The modified microorganism according to any one of claims 8 to 13, wherein the following is expressed: - Increased activity of at least one of the following enzymes: phosphoenolpyruvate carboxylase, phosphoenolpyruvate carboxylkinase, isocitrate lyase, pyruvate carboxylase, and hexose cotransporter permease; and / or Decreased activity of at least one of the following enzymes: lactate dehydrogenase, alcohol dehydrogenase, acetate kinase, phosphoacetyltransferase, pyruvate oxidase, isocitrate lyase, fumarate, 2-oxoglutarate dehydrogenase, pyruvate kinase, malic acid kinase, phosphoglucose isomerase, phosphoenolpyruvate carboxylase, phosphoenolpyruvate carboxylkinase, pyruvate formate lyase, succinate semialdehyde dehydrogenase, glucose transport phosphotransferase, ketohydroxyglutarate aldolase, homoserine-O-succinyltransferase, homoserine kinase, homoserine efflux transporter, diaminopimelic acid decarboxylase, and / or methylglyoxal synthase.

15. The modified microorganism according to claim 14, wherein it is an *E. coli* strain overexpressing at least one of the following genes: *ppc*, *pck*, *aceA*, *galP*, *asd*, *thrA*, *metL*, *lysC*, all from *E. coli*; *pycA* from *Lactococcus lactis*; and / or an *E. coli* strain having at least one gene deletion selected from *IdhA*, *adhE*, *ackA*, *pta*, *poxB*, *focA*, *pflB*, *sad*, *gabABC*, *sfcA*, *maeB*, *ppc*, *pykA*, *pykF*, *mgsA*, *sucAB*, *ptsL*, *ptsG*, *pgi*, *fumABC*, *aldA*, *lldD*, *iclR*, *metA*, *thrB*, *lysA*, *eda*, *rthA*, *rthB*, *rthC*.

16. A method for producing 2,4-DHB, comprising the following steps: - Cultivate the modified microorganism according to any one of claims 8 to 15 in a suitable culture medium. - 2,4-DHB was recovered from the culture medium.

17. The method of claim 16, wherein the 2,4-DHB is further purified.

Images