MICROORGANISMS FOR THE PRODUCTION OF 1,4-BUTANEDIOL AND RELATED METHODS

This application is a continuation of application Ser. No. 12/794,700, filed Jun. 4, 2010, which claims the benefit of priority of U.S. application Ser. No. 61/184,311, filed Jun. 4, 2009, the entire contents of each of which is incorporated herein by reference.

Incorporated herein by reference is the Sequence Listing being concurrently submitted via EFS-Web as an ASCII text file named 12207-177-999 SeqList.txt, created Jan. 27, 2012, and being 150,347 bytes in size.

This invention relates generally to in silico design of organisms and engineering of organisms, more particularly to organisms having 1,4-butanediol biosynthesis capability.

The compound 4-hydroxybutanoic acid (4-hydroxybutanoate, 4-hydroxybutyrate, 4-HB) is a 4-carbon carboxylic acid that has industrial potential as a building block for various commodity and specialty chemicals. In particular, 4-HB has the potential to serve as a new entry point into the 1,4-butanediol family of chemicals, which includes solvents, resins, polymer precursors, and specialty chemicals. 1,4-Butanediol (BDO) is a polymer intermediate and industrial solvent with a global market of about 3 billion lb/year. BDO is currently produced from petrochemical precursors, primarily acetylene, maleic anhydride, and propylene oxide.

For example, acetylene is reacted with 2 molecules of formaldehyde in the Reppe synthesis reaction (Kroschwitz and Grant, Encyclopedia of Chem. Tech., John Wiley and Sons, Inc., New York (1999)), followed by catalytic hydrogenation to form 1,4-butanediol. It has been estimated that 90% of the acetylene produced in the U.S. is consumed for butanediol production. Alternatively, it can be formed by esterification and catalytic hydrogenation of maleic anhydride, which is derived from butane. Downstream, butanediol can be further transformed; for example, by oxidation to γ-butyrolactone, which can be further converted to pyrrolidone and N-methyl-pyrrolidone, or hydrogenolysis to tetrahydrofuran. These compounds have varied uses as polymer intermediates, solvents, and additives, and have a combined market of nearly 2 billion lb/year.

It is desirable to develop a method for production of these chemicals by alternative means that not only substitute renewable for petroleum-based feedstocks, and also use less energy- and capital-intensive processes. The Department of Energy has proposed 1,4-diacids, and particularly succinic acid, as key biologically-produced intermediates for the manufacture of the butanediol family of products (DOE Report, “Top Value-Added Chemicals from Biomass”, 2004). However, succinic acid is costly to isolate and purify and requires high temperatures and pressures for catalytic reduction to butanediol.

Thus, there exists a need for alternative means for effectively producing commercial quantities of 1,4-butanediol and its chemical precursors. The present invention satisfies this need and provides related advantages as well.

The invention provides non-naturally occurring microbial organisms containing a 1,4-butanediol (BDO) pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO and further optimized for expression of BDO. The invention additionally provides methods of using such microbial organisms to produce BDO.

The present invention is directed to the design and production of cells and organisms having biosynthetic production capabilities for 4-hydroxybutanoic acid (4-HB), γ-butyrolactone and 1,4-butanediol (BDO). The invention, in particular, relates to the design of microbial organisms capable of producing BDO by introducing one or more nucleic acids encoding a BDO pathway enzyme.

In one embodiment, the invention utilizes in silico stoichiometric models of Escherichia coli metabolism that identify metabolic designs for biosynthetic production of 4-hydroxybutanoic acid (4-HB) and 1,4-butanediol (BDO). The results described herein indicate that metabolic pathways can be designed and recombinantly engineered to achieve the biosynthesis of 4-HB and downstream products such as 1,4-butanediol in Escherichia coli and other cells or organisms. Biosynthetic production of 4-HB, for example, for the in silico designs can be confirmed by construction of strains having the designed metabolic genotype. These metabolically engineered cells or organisms also can be subjected to adaptive evolution to further augment 4-HB biosynthesis, including under conditions approaching theoretical maximum growth.

In certain embodiments, the 4-HB biosynthesis characteristics of the designed strains make them genetically stable and particularly useful in continuous bioprocesses. Separate strain design strategies were identified with incorporation of different non-native or heterologous reaction capabilities into E. coli or other host organisms leading to 4-HB and 1,4-butanediol producing metabolic pathways from either CoA-independent succinic semialdehyde dehydrogenase, succinyl-CoA synthetase and CoA-dependent succinic semialdehyde dehydrogenase, or glutamate: succinic semialdehyde transaminase. In silico metabolic designs were identified that resulted in the biosynthesis of 4-HB in both E. coli and yeast species from each of these metabolic pathways. The 1,4-butanediol intermediate γ-butyrolactone can be generated in culture by spontaneous cyclization under conditions at pH<7.5, particularly under acidic conditions, such as below pH 5.5, for example, pH<7, pH<6.5, pH<6, and particularly at pH<5.5 or lower.

Strains identified via the computational component of the platform can be put into actual production by genetically engineering any of the predicted metabolic alterations which lead to the biosynthetic production of 4-HB, 1,4-butanediol or other intermediate and/or downstream products. In yet a further embodiment, strains exhibiting biosynthetic production of these compounds can be further subjected to adaptive evolution to further augment product biosynthesis. The levels of product biosynthesis yield following adaptive evolution also can be predicted by the computational component of the system.

In other specific embodiments, microbial organisms were constructed to express a 4-HB biosynthetic pathway encoding the enzymatic steps from succinate to 4-HB and to 4-HB-CoA. Co-expression of succinate coenzyme A transferase, CoA-dependent succinic semialdehyde dehydrogenase, NAD-dependent 4-hydroxybutyrate dehydrogenase and 4-hydroxybutyrate coenzyme A transferase in a host microbial organism resulted in significant production of 4-HB compared to host microbial organisms lacking a 4-HB biosynthetic pathway. In a further specific embodiment, 4-HB-producing microbial organisms were generated that utilized α-ketoglutarate as a substrate by introducing nucleic acids encoding α-ketoglutarate decarboxylase and NAD-dependent 4-hydroxybutyrate dehydrogenase.

In another specific embodiment, microbial organisms containing a 1,4-butanediol (BDO) biosynthetic pathway were constructed that biosynthesized BDO when cultured in the presence of 4-HB. The BDO biosynthetic pathway consisted of a nucleic acid encoding either a multifunctional aldehyde/alcohol dehydrogenase or nucleic acids encoding an aldehyde dehydrogenawse and an alcohol dehydrogenase. To support growth on 4-HB substrates, these BDO-producing microbial organisms also expressed 4-hydroxybutyrate CoA transferase or 4-butyrate kinase in conjunction with phosphotranshydroxybutyrlase. In yet a further specific embodiment, microbial organisms were generated that synthesized BDO through exogenous expression of nucleic acids encoding a functional 4-HB biosynthetic pathway and a functional BDO biosynthetic pathway. The 4-HB biosynthetic pathway consisted of succinate coenzyme A transferase, CoA-dependent succinic semialdehyde dehydrogenase, NAD-dependent 4-hydroxybutyrate dehydrogenase and 4-hydroxybutyrate coenzyme A transferase. The BDO pathway consisted of a multifunctional aldehyde/alcohol dehydrogenase. Further described herein are additional pathways for production of BDO (see FIGS. 8-13).

As used herein, the term “non-naturally occurring” when used in reference to a microbial organism or microorganism of the invention is intended to mean that the microbial organism has at least one genetic alteration not normally found in a naturally occurring strain of the referenced species, including wild-type strains of the referenced species. Genetic alterations include, for example, modifications introducing expressible nucleic acids encoding metabolic polypeptides, other nucleic acid additions, nucleic acid deletions and/or other functional disruption of the microbial genetic material. Such modification include, for example, coding regions and functional fragments thereof, for heterologous, homologous or both heterologous and homologous polypeptides for the referenced species. Additional modifications include, for example, non-coding regulatory regions in which the modifications alter expression of a gene or operon. Exemplary metabolic polypeptides include enzymes or proteins within a biosynthetic pathway for a BDO family of compounds.

A metabolic modification refers to a biochemical reaction that is altered from its naturally occurring state. Therefore, non-naturally occurring microorganisms having genetic modifications to nucleic acids encoding metabolic polypeptides or, functional fragments thereof. Exemplary metabolic modifications are disclosed herein.

As used herein, the term “isolated” when used in reference to a microbial organism is intended to mean an organism that is substantially free of at least one component as the referenced microbial organism is found in nature. The term includes a microbial organism that is removed from some or all components as it is found in its natural environment. The term also includes a microbial organism that is removed from some or all components as the microbial organism is found in non-naturally occurring environments. Therefore, an isolated microbial organism is partly or completely separated from other substances as it is found in nature or as it is grown, stored or subsisted in non-naturally occurring environments. Specific examples of isolated microbial organisms include partially pure microbes, substantially pure microbes and microbes cultured in a medium that is non-naturally occurring.

As used herein, the terms “microbial,” “microbial organism” or “microorganism” is intended to mean any organism that exists as a microscopic cell that is included within the domains of archaea, bacteria or eukarya. Therefore, the term is intended to encompass prokaryotic or eukaryotic cells or organisms having a microscopic size and includes bacteria, archaea and eubacteria of all species as well as eukaryotic microorganisms such as yeast and fungi. The term also includes cell cultures of any species that can be cultured for the production of a biochemical.

As used herein, the term “4-hydroxybutanoic acid” is intended to mean a 4-hydroxy derivative of butyric acid having the chemical formula C₄H₈O₃and a molecular mass of 104.11 g/mol (126.09 g/mol for its sodium salt). The chemical compound 4-hydroxybutanoic acid also is known in the art as 4-HB, 4-hydroxybutyrate, gamma-hydroxybutyric acid or GHB. The term as it is used herein is intended to include any of the compound's various salt forms and include, for example, 4-hydroxybutanoate and 4-hydroxybutyrate. Specific examples of salt forms for 4-HB include sodium 4-HB and potassium 4-HB. Therefore, the terms 4-hydroxybutanoic acid, 4-HB, 4-hydroxybutyrate, 4-hydroxybutanoate, gamma-hydroxybutyric acid and GHB as well as other art recognized names are used synonymously herein.

As used herein, the term “monomeric” when used in reference to 4-HB is intended to mean 4-HB in a non-polymeric or underivatized form. Specific examples of polymeric 4-HB include poly-4-hydroxybutanoic acid and copolymers of, for example, 4-HB and 3-HB. A specific example of a derivatized form of 4-HB is 4-HB-CoA. Other polymeric 4-HB forms and other derivatized forms of 4-HB also are known in the art.

As used herein, the term “γ-butyrolactone” is intended to mean a lactone having the chemical formula C₄H₆O₂and a molecular mass of 86.089 g/mol. The chemical compound γ-butyrolactone also is know in the art as GBL, butyrolactone, 1,4-lactone, 4-butyrolactone, 4-hydroxybutyric acid lactone, and gamma-hydroxybutyric acid lactone. The term as it is used herein is intended to include any of the compound's various salt forms.

As used herein, the term “1,4-butanediol” is intended to mean an alcohol derivative of the alkane butane, carrying two hydroxyl groups which has the chemical formula C₄H₁₀O₂and a molecular mass of 90.12 g/mol. The chemical compound 1,4-butanediol also is known in the art as BDO and is a chemical intermediate or precursor for a family of compounds referred to herein as BDO family of compounds.

As used herein, the term “tetrahydrofuran” is intended to mean a heterocyclic organic compound corresponding to the fully hydrogenated analog of the aromatic compound furan which has the chemical formula C₄H₈O and a molecular mass of 72.11 g/mol. The chemical compound tetrahydrofuran also is known in the art as THF, tetrahydrofuran, 1,4-epoxybutane, butylene oxide, cyclotetramethylene oxide, oxacyclopentane, diethylene oxide, oxolane, furanidine, hydrofuran, tetra-methylene oxide. The term as it is used herein is intended to include any of the compound's various salt forms.

As used herein, the term “CoA” or “coenzyme A” is intended to mean an organic cofactor or prosthetic group (nonprotein portion of an enzyme) whose presence is required for the activity of many enzymes (the apoenzyme) to form an active enzyme system. Coenzyme A functions in certain condensing enzymes, acts in acetyl or other acyl group transfer and in fatty acid synthesis and oxidation, pyruvate oxidation and in other acetylation.

As used herein, the term “substantially anaerobic” when used in reference to a culture or growth condition is intended to mean that the amount of oxygen is less than about 10% of saturation for dissolved oxygen in liquid media. The term also is intended to include sealed chambers of liquid or solid medium maintained with an atmosphere of less than about 1% oxygen.

The non-naturally occurring microbal organisms of the invention can contain stable genetic alterations, which refers to microorganisms that can be cultured for greater than five generations without loss of the alteration. Generally, stable genetic alterations include modifications that persist greater than 10 generations, particularly stable modifications will persist more than about 25 generations, and more particularly, stable genetic modifications will be greater than 50 generations, including indefinitely.

Those skilled in the art will understand that the genetic alterations, including metabolic modifications exemplified herein are described with reference to a suitable source organism such as E. coli, yeast, or other organisms disclosed herein and their corresponding metabolic reactions or a suitable source organism for desired genetic material such as genes encoding enzymes for their corresponding metabolic reactions. However, given the complete genome sequencing of a wide variety of organisms and the high level of skill in the area of genomics, those skilled in the art will readily be able to apply the teachings and guidance provided herein to essentially all other organisms. For example, the E. coli metabolic alterations exemplified herein can readily be applied to other species by incorporating the same or analogous encoding nucleic acid from species other than the referenced species. Such genetic alterations include, for example, genetic alterations of species homologs, in general, and in particular, orthologs, paralogs or nonorthologous gene displacements.

An ortholog is a gene or genes that are related by vertical descent and are responsible for substantially the same or identical functions in different organisms. For example, mouse epoxide hydrolase and human epoxide hydrolase can be considered orthologs for the biological function of hydrolysis of epoxides. Genes are related by vertical descent when, for example, they share sequence similarity of sufficient amount to indicate they are homologous, or related by evolution from a common ancestor. Genes can also be considered orthologs if they share three-dimensional structure but not necessarily sequence similarity, of a sufficient amount to indicate that they have evolved from a common ancestor to the extent that the primary sequence similarity is not identifiable. Genes that are orthologous can encode proteins with sequence similarity of about 25% to 100% amino acid sequence identity. Genes encoding proteins sharing an amino acid similarity less that 25% can also be considered to have arisen by vertical descent if their three-dimensional structure also shows similarities. Members of the serine protease family of enzymes, including tissue plasminogen activator and elastase, are considered to have arisen by vertical descent from a common ancestor.

Orthologs include genes or their encoded gene products that through, for example, evolution, have diverged in structure or overall activity. For example, where one species encodes a gene product exhibiting two functions and where such functions have been separated into distinct genes in a second species, the three genes and their corresponding products are considered to be orthologs. For the growth-coupled production of a biochemical product, those skilled in the art will understand that the orthologous gene harboring the metabolic activity to be disrupted is to be chosen for construction of the non-naturally occurring microorganism. An example of orthologs exhibiting separable activities is where distinct activities have been separated into distinct gene products between two or more species or within a single species. A specific example is the separation of elastase proteolysis and plasminogen proteolysis, two types of serine protease activity, into distinct molecules as plasminogen activator and elastase. A second example is the separation of mycoplasma 5′-3′ exonuclease and Drosophila DNA polymerase III activity. The DNA polymerase from the first species can be considered an ortholog to either or both of the exonuclease or the polymerase from the second species and vice versa.

In contrast, paralogs are homologs related by, for example, duplication followed by evolutionary divergence and have similar or common, but not identical functions. Paralogs can originate or derive from, for example, the same species or from a different species. For example, microsomal epoxide hydrolase (epoxide hydrolase I) and soluble epoxide hydrolase (epoxide hydrolase II) can be considered paralogs because they represent two distinct enzymes, co-evolved from a common ancestor, that catalyze distinct reactions and have distinct functions in the same species. Paralogs are proteins from the same species with significant sequence similarity to each other suggesting that they are homologous, or related through co-evolution from a common ancestor. Groups of paralogous protein families include HipA homologs, luciferase genes, peptidases, and others.

A nonorthologous gene displacement is a nonorthologous gene from one species that can substitute for a referenced gene function in a different species. Substitution includes, for example, being able to perform substantially the same or a similar function in the species of origin compared to the referenced function in the different species. Although generally, a nonorthologous gene displacement will be identifiable as structurally related to a known gene encoding the referenced function, less structurally related but functionally similar genes and their corresponding gene products nevertheless will still fall within the meaning of the term as it is used herein. Functional similarity requires, for example, at least some structural similarity in the active site or binding region of a nonorthologous gene compared to a gene encoding the function sought to be substituted. Therefore, a nonorthologous gene includes, for example, a paralog or an unrelated gene.

Therefore, in identifying and constructing the non-naturally occurring microbial organisms of the invention having 4-HB, GBL and/or BDO biosynthetic capability, those skilled in the art will understand with applying the teaching and guidance provided herein to a particular species that the identification of metabolic modifications can include identification and inclusion or inactivation of orthologs. To the extent that paralogs and/or nonorthologous gene displacements are present in the referenced microorganism that encode an enzyme catalyzing a similar or substantially similar metabolic reaction, those skilled in the art also can utilize these evolutionally related genes.

Orthologs, paralogs and nonorthologous gene displacements can be determined by methods well known to those skilled in the art. For example, inspection of nucleic acid or amino acid sequences for two polypeptides will reveal sequence identity and similarities between the compared sequences. Based on such similarities, one skilled in the art can determine if the similarity is sufficiently high to indicate the proteins are related through evolution from a common ancestor. Algorithms well known to those skilled in the art, such as Align, BLAST, Clustal W and others compare and determine a raw sequence similarity or identity, and also determine the presence or significance of gaps in the sequence which can be assigned a weight or score. Such algorithms also are known in the art and are similarly applicable for determining nucleotide sequence similarity or identity. Parameters for sufficient similarity to determine relatedness are computed based on well known methods for calculating statistical similarity, or the chance of finding a similar match in a random polypeptide, and the significance of the match determined. A computer comparison of two or more sequences can, if desired, also be optimized visually by those skilled in the art. Related gene products or proteins can be expected to have a high similarity, for example, 25% to 100% sequence identity. Proteins that are unrelated can have an identity which is essentially the same as would be expected to occur by chance, if a database of sufficient size is scanned (about 5%). Sequences between 5% and 24% may or may not represent sufficient homology to conclude that the compared sequences are related. Additional statistical analysis to determine the significance of such matches given the size of the data set can be carried out to determine the relevance of these sequences.

Exemplary parameters for determining relatedness of two or more sequences using the BLAST algorithm, for example, can be as set forth below. Briefly, amino acid sequence alignments can be performed using BLASTP version 2.0.8 (Jan. 5, 1999) and the following parameters: Matrix: 0 BLOSUM62; gap open: 11; gap extension: 1; x_dropoff: 50; expect: 10.0; wordsize: 3; filter: on. Nucleic acid sequence alignments can be performed using BLASTN version 2.0.6 (Sep. 16, 1998) and the following parameters: Match: 1; mismatch: −2; gap open: 5; gap extension: 2; x_dropoff: 50; expect: 10.0; wordsize: 11; filter: off Those skilled in the art will know what modifications can be made to the above parameters to either increase or decrease the stringency of the comparison, for example, and determine the relatedness of two or more sequences.

Disclosed herein are non-naturally occurring microbial biocatalyst or microbial organisms including a microbial organism having a 4-hydroxybutanoic acid (4-HB) biosynthetic pathway that includes at least one exogenous nucleic acid encoding 4-hydroxybutanoate dehydrogenase, CoA-independent succinic semialdehyde dehydrogenase, succinyl-CoA synthetase, CoA-dependent succinic semialdehyde dehydrogenase, glutamate: succinic semialdehyde transaminase, alpha-ketoglutarate decarboxylase, or glutamate decarboxylase, wherein the exogenous nucleic acid is expressed in sufficient amounts to produce monomeric 4-hydroxybutanoic acid (4-HB). 4-hydroxybutanoate dehydrogenase is also referred to as 4-hydroxybutyrate dehydrogenase or 4-HB dehydrogenase. Succinyl-CoA synthetase is also referred to as succinyl-CoA synthase or succinyl-CoA ligase.

Also disclosed herein is a non-naturally occurring microbial biocatalyst or microbial organism including a microbial organism having a 4-hydroxybutanoic acid (4-HB) biosynthetic pathway having at least one exogenous nucleic acid encoding 4-hydroxybutanoate dehydrogenase, succinyl-CoA synthetase, CoA-dependent succinic semialdehyde dehydrogenase, or α-ketoglutarate decarboxylase, wherein the exogenous nucleic acid is expressed in sufficient amounts to produce monomeric 4-hydroxybutanoic acid (4-HB).

The non-naturally occurring microbial biocatalysts or microbial organisms can include microbial organisms that employ combinations of metabolic reactions for biosynthetically producing the compounds of the invention. The biosynthesized compounds can be produced intracellularly and/or secreted into the culture medium. Exemplary compounds produced by the non-naturally occurring microorganisms include, for example, 4-hydroxybutanoic acid, 1,4-butanediol and γ-butyrolactone.

In one embodiment, a non-naturally occurring microbial organism is engineered to produce 4-HB. This compound is one useful entry point into the 1,4-butanediol family of compounds. The biochemical reactions for formation of 4-HB from succinate, from succinate through succinyl-CoA or from α-ketoglutarate are shown in steps 1-8 of FIG. 1.

It is understood that any combination of appropriate enzymes of a BDO pathway can be used so long as conversion from a starting component to the BDO product is achieved. Thus, it is understood that any of the metabolic pathways disclosed herein can be utilized and that it is well understood to those skilled in the art how to select appropriate enzymes to achieve a desired pathway, as disclosed herein.

In another embodiment, disclosed herein is a non-naturally occurring microbial organism, comprising a microbial organism having a 1,4-butanediol (BDO) pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, the BDO pathway comprising 4-aminobutyrate CoA transferase, 4-aminobutyryl-CoA hydrolase, 4-aminobutyrate-CoA ligase, 4-aminobutyryl-CoA oxidoreductase (deaminating), 4-aminobutyryl-CoA transaminase, or 4-hydroxybutyryl-CoA dehydrogenase (see Example VII Table 17). The BDO pathway further can comprise 4-hydroxybutyryl-CoA reductase (alcohol forming), 4-hydroxybutyryl-CoA reductase, or 1,4-butanediol dehydrogenase.

It is understood by those skilled in the art that various combinations of the pathways can be utilized, as disclosed herein. For example, in a non-naturally occurring microbial organism, the nucleic acids can encode 4-aminobutyrate CoA transferase, 4-aminobutyryl-CoA hydrolase, or 4-aminobutyrate-CoA ligase; 4-aminobutyryl-CoA oxidoreductase (deaminating) or 4-aminobutyryl-CoA transaminase; and 4-hydroxybutyryl-CoA dehydrogenase. Other exemplary combinations are specifically describe below and further can be found in FIGS. 8-13. For example, the BDO pathway can further comprise 4-hydroxybutyryl-CoA reductase (alcohol forming), 4-hydroxybutyryl-CoA reductase, or 1,4-butanediol dehydrogenase.

Additionally disclosed herein is a non-naturally occurring microbial organism, comprising a microbial organism having a BDO pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, the BDO pathway comprising 4-aminobutyrate CoA transferase, 4-aminobutyryl-CoA hydrolase, 4-aminobutyrate-CoA ligase, 4-aminobutyryl-CoA reductase (alcohol forming), 4-aminobutyryl-CoA reductase, 4-aminobutan-1-ol dehydrogenase, 4-aminobutan-1-ol oxidoreductase (deaminating) or 4-aminobutan-1-ol transaminase (see Example VII and Table 18), and can further comprise 1,4-butanediol dehydrogenase. For example, the exogenous nucleic acids can encode 4-aminobutyrate CoA transferase, 4-aminobutyryl-CoA hydrolase, or 4-aminobutyrate-CoA ligase; 4-aminobutyryl-CoA reductase (alcohol forming); and 4-aminobutan-1-ol oxidoreductase (deaminating) or 4-aminobutan-1-ol transaminase. In addition, the nucleic acids can encode. 4-aminobutyrate CoA transferase, 4-aminobutyryl-CoA hydrolase, or 4-aminobutyrate-CoA ligase; 4-aminobutyryl-CoA reductase; 4-aminobutan-1-ol dehydrogenase; and 4-aminobutan-1-ol oxidoreductase (deaminating) or 4-aminobutan-1-ol transaminase.

Also disclosed herein is a non-naturally occurring microbial organism, comprising a microbial organism having a BDO pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, the BDO pathway comprising 4-aminobutyrate kinase, 4-aminobutyraldehyde dehydrogenase (phosphorylating), 4-aminobutan-1-ol dehydrogenase, 4-aminobutan-1-oloxidoreductase (deaminating), 4-aminobutan-1-ol transaminase, [(4-aminobutanolyl)oxy]phosphonic acid oxidoreductase (deaminating), [(4-aminobutanolyl)oxy]phosphonic acid transaminase, 4-hydroxybutyryl-phosphate dehydrogenase, or 4-hydroxybutyraldehyde dehydrogenase (phosphorylating) (see Example VII and Table 19). For example, the exogenous nucleic acids can encode 4-aminobutyrate kinase; 4-aminobutyraldehyde dehydrogenase (phosphorylating); 4-aminobutan-1-ol dehydrogenase; and 4-aminobutan-1-ol oxidoreductase (deaminating) or 4-aminobutan-1-ol transaminase. Alternatively, the exogenous nucleic acids can encode 4-aminobutyrate kinase; [(4-aminobutanolyl)oxy]phosphonic acid oxidoreductase (deaminating) or [(4-aminobutanolyl)oxy]phosphonic acid transaminase; 4-hydroxybutyryl-phosphate dehydrogenase; and 4-hydroxybutyraldehyde dehydrogenase (phosphorylating).

Additionally disclosed herein is a non-naturally occurring microbial organism, comprising a microbial organism having a BDO pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, the BDO pathway comprising alpha-ketoglutarate 5-kinase, 2,5-dioxopentanoic semialdehyde dehydrogenase (phosphorylating), 2,5-dioxopentanoic acid reductase, alpha-ketoglutarate CoA transferase, alpha-ketoglutaryl-CoA hydrolase, alpha-ketoglutaryl-CoA ligase, alpha-ketoglutaryl-CoA reductase, 5-hydroxy-2-oxopentanoic acid dehydrogenase, alpha-ketoglutaryl-CoA reductase (alcohol forming), 5-hydroxy-2-oxopentanoic acid decarboxylase, or 5-hydroxy-2-oxopentanoic acid dehydrogenase (decarboxylation) (see Example VIII and Table 20). The BDO pathway can further comprise 4-hydroxybutyryl-CoA reductase (alcohol forming), 4-hydroxybutyryl-CoA reductase, or 1,4-butanediol dehydrogenase. For example, the exogenous nucleic acids can encode alpha-ketoglutarate 5-kinase; 2,5-dioxopentanoic semialdehyde dehydrogenase (phosphorylating); 2,5-dioxopentanoic acid reductase; and 5-hydroxy-2-oxopentanoic acid decarboxylase. Alternatively, the exogenous nucleic acids can encode alpha-ketoglutarate 5-kinase; 2,5-dioxopentanoic semialdehyde dehydrogenase (phosphorylating); 2,5-dioxopentanoic acid reductase; and 5-hydroxy-2-oxopentanoic acid dehydrogenase (decarboxylation). Alternatively, the exogenous nucleic acids can encode alpha-ketoglutarate CoA transferase, alpha-ketoglutaryl-CoA hydrolase, or alpha-ketoglutaryl-CoA ligase; alpha-ketoglutaryl-CoA reductase, 5-hydroxy-2-oxopentanoic acid dehydrogenase; and 5-hydroxy-2-oxopentanoic acid decarboxylase. In another embodiment, the exogenous nucleic acids can encode alpha-ketoglutarate CoA transferase, alpha-ketoglutaryl-CoA hydrolase, or alpha-ketoglutaryl-CoA ligase; alpha-ketoglutaryl-CoA reductase, 5-hydroxy-2-oxopentanoic acid dehydrogenase, and 5-hydroxy-2-oxopentanoic acid dehydrogenase (decarboxylation). Alternatively, the exogenous nucleic acids can encode alpha-ketoglutarate CoA transferase, alpha-ketoglutaryl-CoA hydrolase, or alpha-ketoglutaryl-CoA ligase; alpha-ketoglutaryl-CoA reductase (alcohol forming); and 5-hydroxy-2-oxopentanoic acid decarboxylase. In yet another embodiment, the exogenous nucleic acids can encode alpha-ketoglutarate CoA transferase, alpha-ketoglutaryl-CoA hydrolase, or alpha-ketoglutaryl-CoA ligase; alpha-ketoglutaryl-CoA reductase (alcohol forming); and 5-hydroxy-2-oxopentanoic acid dehydrogenase (decarboxylation).

Further disclosed herein is a non-naturally occurring microbial organism, comprising a microbial organism having a BDO pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, the BDO pathway comprising glutamate CoA transferase, glutamyl-CoA hydrolase, glutamyl-CoA ligase, glutamate 5-kinase, glutamate-5-semialdehyde dehydrogenase (phosphorylating), glutamyl-CoA reductase, glutamate-5-semialdehyde reductase, glutamyl-CoA reductase (alcohol forming), 2-amino-5-hydroxypentanoic acid oxidoreductase (deaminating), 2-amino-5-hydroxypentanoic acid transaminase, 5-hydroxy-2-oxopentanoic acid decarboxylase, 5-hydroxy-2-oxopentanoic acid dehydrogenase (decarboxylation) (see Example 1× and Table 21). For example, the exogenous nucleic acids can encode glutamate CoA transferase, glutamyl-CoA hydrolase, or glutamyl-CoA ligase; glutamyl-CoA reductase; glutamate-5-semialdehyde reductase; 2-amino-5-hydroxypentanoic acid oxidoreductase (deaminating) or 2-amino-5-hydroxypentanoic acid transaminase; and 5-hydroxy-2-oxopentanoic acid decarboxylase or 5-hydroxy-2-oxopentanoic acid dehydrogenase (decarboxylation). Alternatively, the exogenous nucleic acids can encode glutamate 5-kinase;

• glutamate-5-semialdehyde dehydrogenase (phosphorylating); glutamate-5-semialdehyde reductase; 2-amino-5-hydroxypentanoic acid oxidoreductase (deaminating) or 2-amino-5-hydroxypentanoic acid transaminase; and 5-hydroxy-2-oxopentanoic acid decarboxylase or 5-hydroxy-2-oxopentanoic acid dehydrogenase (decarboxylation). In still another embodiment, the exogenous nucleic acids can encode glutamate CoA transferase, glutamyl-CoA hydrolase, or glutamyl-CoA ligase; glutamyl-CoA reductase (alcohol forming); 2-amino-5-hydroxypentanoic acid oxidoreductase (deaminating) or 2-amino-5-hydroxypentanoic acid transaminase; and 5-hydroxy-2-oxopentanoic acid decarboxylase or 5-hydroxy-2-oxopentanoic acid dehydrogenase (decarboxylation). In yet another embodiment, the exogenous nucleic acids can encode glutamate 5-kinase; glutamate-5-semialdehyde dehydrogenase (phosphorylating); 2-amino-5-hydroxypentanoic acid oxidoreductase (deaminating) or 2-amino-5-hydroxypentanoic acid transaminase; and 5-hydroxy-2-oxopentanoic acid decarboxylase or 5-hydroxy-2-oxopentanoic acid dehydrogenase (decarboxylation).

Also disclosed herein is a non-naturally occurring microbial organism, comprising a microbial organism having a BDO pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, the BDO pathway comprising 3-hydroxybutyryl-CoA dehydrogenase, 3-hydroxybutyryl-CoA dehydratase, vinylacetyl-CoA A-isomerase, or 4-hydroxybutyryl-CoA dehydratase (see Example X and Table 22). For example, the exogenous nucleic acids can encode 3-hydroxybutyryl-CoA dehydrogenase; 3-hydroxybutyryl-CoA dehydratase; vinylacetyl-CoA A-isomerase; and 4-hydroxybutyryl-CoA dehydratase.

Further disclosed herein is a non-naturally occurring microbial organism, comprising a microbial organism having a BDO pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, the BDO pathway comprising homoserine deaminase, homoserine CoA transferase, homoserine-CoA hydrolase, homoserine-CoA ligase, homoserine-CoA deaminase, 4-hydroxybut-2-enoyl-CoA transferase, 4-hydroxybut-2-enoyl-CoA hydrolase, 4-hydroxybut-2-enoyl-CoA ligase, 4-hydroxybut-2-enoate reductase, 4-hydroxybutyryl-CoA transferase, 4-hydroxybutyryl-CoA hydrolase, 4-hydroxybutyryl-CoA ligase, or 4-hydroxybut-2-enoyl-CoA reductase (see Example XI and Table 23). For example, the exogenous nucleic acids can encode homoserine deaminase; 4-hydroxybut-2-enoyl-CoA transferase, 4-hydroxybut-2-enoyl-CoA hydrolase, 4-hydroxybut-2-enoyl-CoA ligase; 4-hydroxybut-2-enoyl-CoA reductase. Alternatively, the exogenous nucleic acids can encode homoserine CoA transferase, homoserine-CoA hydrolase, or homoserine-CoA ligase; homoserine-CoA deaminase; and 4-hydroxybut-2-enoyl-CoA reductase. In a further embodiment, the exogenous nucleic acids can encode homoserine deaminase; 4-hydroxybut-2-enoate reductase; and 4-hydroxybutyryl-CoA transferase, 4-hydroxybutyryl-CoA hydrolase, or 4-hydroxybutyryl-CoA ligase. Alternatively, the exogenous nucleic acids can encode homoserine CoA transferase, homoserine-CoA hydrolase, or homoserine-CoA ligase; homoserine-CoA deaminase; and 4-hydroxybut-2-enoyl-CoA reductase.

Further disclosed herein is a non-naturally occurring microbial organism, comprising a microbial organism having a BDO pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BOD, the BDO pathway comprising succinyl-CoA reductase (alcohol forming), 4-hydroxybutyryl-CoA hydrolase, 4-hydroxybutyryl-CoA ligase, 4-hydroxybutanal dehydrogenase (phosphorylating) (see Table 15). Such a BDO pathway can further comprise succinyl-CoA reductase, 4-hydroxybutyrate dehydrogenase, 4-hydroxybutyryl-CoA transferase, 4-hydroxybutyrate kinase, phosphotrans-4-hydroxybutyrylase, 4-hydroxybutyryl-CoA reductase, 4-hydroxybutyryl-CoA reductase (alcohol forming), or 1,4-butanediol dehydrogenase.

Additionally disclosed herein is a non-naturally occurring microbial organism, comprising a microbial organism having a BDO pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, the BDO pathway comprising glutamate dehydrogenase, 4-aminobutyrate oxidoreductase (deaminating), 4-aminobutyrate transaminase, glutamate decarboxylase, 4-hydroxybutyryl-CoA hydrolase, 4-hydroxybutyryl-CoA ligase, 4-hydroxybutanal dehydrogenase (phosphorylating)(see Table 16). Such a BDO pathway can further comprise alpha-ketoglutarate decarboxylase, 4-hydroxybutyrate dehydrogenase, 4-hydroxybutyryl-CoA transferase, 4-hydroxybutyrate kinase, phosphotrans-4-hydroxybutyrylase, 4-hydroxybutyryl-CoA reductase, 4-hydroxybutyryl-CoA reductase (alcohol forming), or 1,4-butanediol dehydrogenase.

The pathways described above are merely exemplary. One skilled in the art can readily select appropriate pathways from those disclosed herein to obtain a suitable BDO pathway or other metabolic pathway, as desired.

The invention provides genetically modified organisms that allow improved production of a desired product such as BDO by increasing the product or decreasing undesirable byproducts. As disclosed herein, the invention provides a non-naturally occurring microbial organism, comprising a microbial organism having a 1,4-butanediol (BDO) pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO. In one embodiment, the microbial organism is genetically modified to express exogenous succinyl-CoA synthetase (see Example XII). For example, the succinyl-CoA synthetase can be encoded by an Escherichia coli sucCD genes.

In another embodiment, the microbial organism is genetically modified to express exogenous alpha-ketoglutarate decarboxylase (see Example XIII). For example, the alpha-ketoglutarate decarboxylase can be encoded by the Mycobacterium bovis sucA gene. In still another embodiment, the microbial organism is genetically modified to express exogenous succinate semialdehyde dehydrogenase and 4-hydroxybutyrate dehydrogenase and optionally 4-hydroxybutyryl-CoA/acetyl-CoA transferase (see Example XIII). For example, the succinate semialdehyde dehydrogenase (CoA-dependent), 4-hydroxybutyrate dehydrogenase and 4-hydroxybutyryl-CoA/acetyl-CoA transferase can be encoded by Porphyromonas gingivalis W83 genes. In an additional embodiment, the microbial organism is genetically modified to express exogenous butyrate kinase and phosphotransbutyrylase (see Example XIII). For example, the butyrate kinase and phosphotransbutyrylase can be encoded by Clostridium acetobutilicum buk1 and ptb genes.

In yet another embodiment, the microbial organism is genetically modified to express exogenous 4-hydroxybutyryl-CoA reductase (see Example XIII). For example, the 4-hydroxybutyryl-CoA reductase can be encoded by Clostridium beijerinckii ald gene. Additionally, in an embodiment of the invention, the microbial organism is genetically modified to express exogenous 4-hydroxybutanal reductase (see Example XIII). For example, the 4-hydroxybutanal reductase can be encoded by Geobacillus thermoglucosidasius adhl gene. In another embodiment, the microbial organism is genetically modified to express exogenous pyruvate dehydrogenase subunits (see Example XIV). For example, the exogenous pyruvate dehydrogenase can be NADH insensitive. The pyruvate dehydrogenase subunit can be encoded by the Klebsiella pneumonia lpdA gene. In a particular embodiment, the pyruvate dehydrogenase subunit genes of the microbial organism can be under the control of a pyruvate formate lyase promoter.

In still another embodiment, the microbial organism is genetically modified to disrupt a gene encoding an aerobic respiratory control regulatory system (see Example XV). For example, the disruption can be of the arcA gene. Such an organism can further comprise disruption of a gene encoding malate dehydrogenase. In a further embodiment, the microbial organism is genetically modified to express an exogenous NADH insensitive citrate synthase (see Example XV). For example, the NADH insensitive citrate synthase can be encoded by gltA, such as an R163L mutant of gltA. In still another embodiment, the microbial organism is genetically modified to express exogenous phosphoenolpyruvate carboxykinase (see Example XVI). For example, the phosphoenolpyruvate carboxykinase can be encoded by an Haemophilus influenza phosphoenolpyruvate carboxykinase gene.

It is understood that any of a number of genetic modifications, as disclosed herein, can be used alone or in various combinations of one or more of the genetic modifications disclosed herein to increase the production of BDO in a BDO producing microbial organism. In a particular embodiment, the microbial organism can be genetically modified to incorporate any and up to all of the genetic modifications that lead to increased production of BDO. In a particular embodiment, the microbial organism containing a BDO pathway can be genetically modified to express exogenous succinyl-CoA synthetase; to express exogenous alpha-ketoglutarate decarboxylase; to express exogenous succinate semialdehyde dehydrogenase and 4-hydroxybutyrate dehydrogenase and optionally 4-hydroxybutyryl-CoA/acetyl-CoA transferase; to express exogenous butyrate kinase and phosphotransbutyrylase; to express exogenous 4-hydroxybutyryl-CoA reductase; and to express exogenous 4-hydroxybutanal reductase; to express exogenous pyruvate dehydrogenase; to disrupt a gene encoding an aerobic respiratory control regulatory system; to express an exogenous NADH insensitive citrate synthase; and to express exogenous phosphoenolpyruvate carboxykinase. Such strains for improved production are described in Examples XII-XIX. It is thus understood that, in addition to the modifications described above, such strains can additionally include other modifications disclosed herein. Such modifications include, but are not limited to, deletion of endogenous lactate dehydrogenase (ldhA), alcohol dehydrogenase (adhE), and/or pyruvate formate lyase (pflB)(see Examples XII-XIX and Table 28).

Additionally provided is a microbial organism in which one or more genes encoding the exogenously expressed enzymes are integrated into the fimD locus of the host organism (see Example XVII). For example, one or more genes encoding a BDO pathway enzyme can be integrated into the fimD locus for increased production of BDO. Further provided is a microbial organism expressing a non-phosphotransferase sucrose uptake system that increases production of BDO.

Although the genetically modified microbial organisms disclosed herein are exemplified with microbial organisms containing particular BDO pathway enzymes, it is understood that such modifications can be incorporated into any microbial organism having a BDO pathway suitable for enhanced production in the presence of the genetic modifications. The microbial organisms of the invention can thus have any of the BDO pathways disclosed herein. For example, the BDO pathway can comprise 4-hydroxybutanoate dehydrogenase, succinyl-CoA synthetase, CoA-dependent succinic semialdehyde dehydrogenase, 4-hydroxybutyrate:CoA transferase, 4-butyrate kinase, phosphotransbutyrylase, alpha-ketoglutarate decarboxylase, aldehyde dehydrogenase, alcohol dehydrogenase or an aldehyde/alcohol dehydrogenase (see FIG. 1). Alternatively, the BDO pathway can comprise 4-aminobutyrate CoA transferase, 4-aminobutyryl-CoA hydrolase, 4-aminobutyrate-CoA ligase, 4-aminobutyryl-CoA oxidoreductase (deaminating), 4-aminobutyryl-CoA transaminase, or 4-hydroxybutyryl-CoA dehydrogenase (see Table 17). Such a BDO pathway can further comprise 4-hydroxybutyryl-CoA reductase (alcohol forming), 4-hydroxybutyryl-CoA reductase, or 1,4-butanediol dehydrogenase

Additionally, the BDO pathway can comprise 4-aminobutyrate CoA transferase, 4-aminobutyryl-CoA hydrolase, 4-aminobutyrate-CoA ligase, 4-aminobutyryl-CoA reductase (alcohol forming), 4-aminobutyryl-CoA reductase, 4-aminobutan-1-ol dehydrogenase, 4-aminobutan-1-ol oxidoreductase (deaminating) or 4-aminobutan-1-ol transaminase (see Table 18). Also, the BDO pathway can comprise 4-aminobutyrate kinase, 4-aminobutyraldehyde dehydrogenase (phosphorylating), 4-aminobutan-1-ol dehydrogenase, 4-aminobutan-1-oloxidoreductase (deaminating), 4-aminobutan-1-ol transaminase, [(4-aminobutanolyl)oxy]phosphonic acid oxidoreductase (deaminating), [(4-aminobutanolyl)oxy]phosphonic acid transaminase, 4-hydroxybutyryl-phosphate dehydrogenase, or 4-hydroxybutyraldehyde dehydrogenase (phosphorylating) (see Table 19). Such a pathway can further comprise 1,4-butanediol dehydrogenase.

The BDO pathway can also comprise alpha-ketoglutarate 5-kinase, 2,5-dioxopentanoic semialdehyde dehydrogenase (phosphorylating), 2,5-dioxopentanoic acid reductase, alpha-ketoglutarate CoA transferase, alpha-ketoglutaryl-CoA hydrolase, alpha-ketoglutaryl-CoA ligase, alpha-ketoglutaryl-CoA reductase, 5-hydroxy-2-oxopentanoic acid dehydrogenase, alpha-ketoglutaryl-CoA reductase (alcohol forming), 5-hydroxy-2-oxopentanoic acid decarboxylase, or 5-hydroxy-2-oxopentanoic acid dehydrogenase (decarboxylation)(see Table 20). Such a BDO pathway can further comprise 4-hydroxybutyryl-CoA reductase (alcohol forming), 4-hydroxybutyryl-CoA reductase, or 1,4-butanediol dehydrogenase. Additionally, the BDO pathway can comprise glutamate CoA transferase, glutamyl-CoA hydrolase, glutamyl-CoA ligase, glutamate 5-kinase, glutamate-5-semialdehyde dehydrogenase (phosphorylating), glutamyl-CoA reductase, glutamate-5-semialdehyde reductase, glutamyl-CoA reductase (alcohol forming), 2-amino-5-hydroxypentanoic acid oxidoreductase (deaminating), 2-amino-5-hydroxypentanoic acid transaminase, 5-hydroxy-2-oxopentanoic acid decarboxylase, 5-hydroxy-2-oxopentanoic acid dehydrogenase (decarboxylation) (see Table 21). Such a BDO pathway can further comprise 4-hydroxybutyryl-CoA reductase (alcohol forming), 4-hydroxybutyryl-CoA reductase, or 1,4-butanediol dehydrogenase.

Additionally, the BDO pathway can comprise 3-hydroxybutyryl-CoA dehydrogenase, 3-hydroxybutyryl-CoA dehydratase, vinylacetyl-CoA A-isomerase, or 4-hydroxybutyryl-CoA dehydratase (see Table 22). Also, the BDO pathway can comprise homoserine deaminase, homoserine CoA transferase, homoserine-CoA hydrolase, homoserine-CoA ligase, homoserine-CoA deaminase, 4-hydroxybut-2-enoyl-CoA transferase, 4-hydroxybut-2-enoyl-CoA hydrolase, 4-hydroxybut-2-enoyl-CoA ligase, 4-hydroxybut-2-enoate reductase, 4-hydroxybutyryl-CoA transferase, 4-hydroxybutyryl-CoA hydrolase, 4-hydroxybutyryl-CoA ligase, or 4-hydroxybut-2-enoyl-CoA reductase (see Table 23). Such a BDO pathway can further comprise 4-hydroxybutyryl-CoA reductase (alcohol forming), 4-hydroxybutyryl-CoA reductase, or 1,4-butanediol dehydrogenase.

The BDO pathway can additionally comprise succinyl-CoA reductase (alcohol forming), 4-hydroxybutyryl-CoA hydrolase, 4-hydroxybutyryl-CoA ligase, or 4-hydroxybutanal dehydrogenase (phosphorylating) (see Table 15). Such a pathway can further comprise succinyl-CoA reductase, 4-hydroxybutyrate dehydrogenase, 4-hydroxybutyryl-CoA transferase, 4-hydroxybutyrate kinase, phosphotrans-4-hydroxybutyrylase, 4-hydroxybutyryl-CoA reductase, 4-hydroxybutyryl-CoA reductase (alcohol forming), or 1,4-butanediol dehydrogenase. Also, the BDO pathway can comprise glutamate dehydrogenase, 4-aminobutyrate oxidoreductase (deaminating), 4-aminobutyrate transaminase, glutamate decarboxylase, 4-hydroxybutyryl-CoA hydrolase, 4-hydroxybutyryl-CoA ligase, or 4-hydroxybutanal dehydrogenase (phosphorylating) (see Table 16). Such a BDO pathway can further comprise alpha-ketoglutarate decarboxylase, 4-hydroxybutyrate dehydrogenase, 4-hydroxybutyryl-CoA transferase, 4-hydroxybutyrate kinase, phosphotrans-4-hydroxybutyrylase, 4-hydroxybutyryl-CoA reductase, 4-hydroxybutyryl-CoA reductase (alcohol forming), or 1,4-butanediol dehydrogenase.

The invention is described herein with general reference to the metabolic reaction, reactant or product thereof, or with specific reference to one or more nucleic acids or genes encoding an enzyme associated with or catalyzing the referenced metabolic reaction, reactant or product. Unless otherwise expressly stated herein, those skilled in the art will understand that reference to a reaction also constitutes reference to the reactants and products of the reaction. Similarly, unless otherwise expressly stated herein, reference to a reactant or product also references the reaction and that reference to any of these metabolic constitutes also references the gene or genes encoding the enzymes that catalyze the referenced reaction, reactant or product. Likewise, given the well known fields of metabolic biochemistry, enzymology and genomics, reference herein to a gene or encoding nucleic acid also constitutes a reference to the corresponding encoded enzyme and the reaction it catalyzes as well as the reactants and products of the reaction.

The production of 4-HB via biosynthetic modes using the microbial organisms of the invention is particularly useful because it can produce monomeric 4-HB. The non-naturally occurring microbial organisms of the invention and their biosynthesis of 4-HB and BDO family compounds also is particularly useful because the 4-HB product can be (1) secreted; (2) can be devoid of any derivatizations such as Coenzyme A; (3) avoids thermodynamic changes during biosynthesis; (4) allows direct biosynthesis of BDO, and (5) allows for the spontaneous chemical conversion of 4-HB to γ-butyrolactone (GBL) in acidic pH medium. This latter characteristic also is particularly useful for efficient chemical synthesis or biosynthesis of BDO family compounds such as 1,4-butanediol and/or tetrahydrofuran (THF), for example.

Microbial organisms generally lack the capacity to synthesize 4-HB and therefore any of the compounds disclosed herein to be within the 1,4-butanediol family of compounds or known by those in the art to be within the 1,4-butanediol family of compounds. Moreover, organisms having all of the requisite metabolic enzymatic capabilities are not known to produce 4-HB from the enzymes described and biochemical pathways exemplified herein. Rather, with the possible exception of a few anaerobic microorganisms described further below, the microorganisms having the enzymatic capability use 4-HB as a substrate to produce, for example, succinate. In contrast, the non-naturally occurring microbial organisms of the invention can generate 4-HB or BDO as a product. As described above, the biosynthesis of 4-HB in its monomeric form is not only particularly useful in chemical synthesis of BDO family of compounds, it also allows for the further biosynthesis of BDO family compounds and avoids altogether chemical synthesis procedures.

The non-naturally occurring microbial organisms of the invention that can produce 4-HB or BDO are produced by ensuring that a host microbial organism includes functional capabilities for the complete biochemical synthesis of at least one 4-HB or BDO biosynthetic pathway of the invention. Ensuring at least one requisite 4-HB or BDO biosynthetic pathway confers 4-HB biosynthesis capability onto the host microbial organism.

Five 4-HB biosynthetic pathways are exemplified herein and shown for purposes of illustration in FIG. 1. Additional 4-HB and BDO pathways are described in FIGS. 8-13. One 4-HB biosynthetic pathway includes the biosynthesis of 4-HB from succinate (the succinate pathway). The enzymes participating in this 4-HB pathway include CoA-independent succinic semialdehyde dehydrogenase and 4-hydroxybutanoate dehydrogenase. In this pathway, CoA-independent succinic semialdehyde dehydrogenase catalyzes the reverse reaction to the arrow shown in FIG. 1. Another 4-HB biosynthetic pathway includes the biosynthesis from succinate through succinyl-CoA (the succinyl-CoA pathway). The enzymes participating in this 4-HB pathway include succinyl-CoA synthetase, CoA-dependent succinic semialdehyde dehydrogenase and 4-hydroxybutanoate dehydrogenase. Three other 4-HB biosynthetic pathways include the biosynthesis of 4-HB from α-ketoglutarate (the α-ketoglutarate pathways). Hence, a third 4-HB biosynthetic pathway is the biosynthesis of succinic semialdehyde through glutamate: succinic semialdehyde transaminase, glutamate decarboxylase and 4-hydroxybutanoate dehydrogenase. A fourth 4-HB biosynthetic pathway also includes the biosynthesis of 4-HB from α-ketoglutarate, but utilizes α-ketoglutarate decarboxylase to catalyze succinic semialdehyde synthesis. 4-hydroxybutanoate dehydrogenase catalyzes the conversion of succinic semialdehyde to 4-HB. A fifth 4-HB biosynthetic pathway includes the biosynthesis from α-ketoglutarate through succinyl-CoA and utilizes α-ketoglutarate dehydrogenase to produce succinyl-CoA, which funnels into the succinyl-CoA pathway described above. Each of these 4-HB biosynthetic pathways, their substrates, reactants and products are described further below in the Examples. As described herein, 4-HB can further be biosynthetically converted to BDO by inclusion of appropriate enzymes to produce BDO (see Example). Thus, it is understood that a 4-HB pathway can be used with enzymes for converting 4-HB to BDO to generate a BDO pathway.

The non-naturally occurring microbial organisms of the invention can be produced by introducing expressible nucleic acids encoding one or more of the enzymes participating in one or more 4-HB or BDO biosynthetic pathways. Depending on the host microbial organism chosen for biosynthesis, nucleic acids for some or all of a particular 4-HB or BDO biosynthetic pathway can be expressed. For example, if a chosen host is deficient in one or more enzymes in a desired biosynthetic pathway, for example, the succinate to 4-HB pathway, then expressible nucleic acids for the deficient enzyme(s), for example, both CoA-independent succinic semialdehyde dehydrogenase and 4-hydroxybutanoate dehydrogenase in this example, are introduced into the host for subsequent exogenous expression. Alternatively, if the chosen host exhibits endogenous expression of some pathway enzymes, but is deficient in others, then an encoding nucleic acid is needed for the deficient enzyme(s) to achieve 4-HB or BDO biosynthesis. For example, if the chosen host exhibits endogenous CoA-independent succinic semialdehyde dehydrogenase, but is deficient in 4-hydroxybutanoate dehydrogenase, then an encoding nucleic acid is needed for this enzyme to achieve 4-HB biosynthesis. Thus, a non-naturally occurring microbial organism of the invention can be produced by introducing exogenous enzyme or protein activities to obtain a desired biosynthetic pathway or a desired biosynthetic pathway can be obtained by introducing one or more exogenous enzyme or protein activities that, together with one or more endogenous enzymes or proteins, produces a desired product such as 4-HB or BDO.

In like fashion, where 4-HB biosynthesis is selected to occur through the succinate to succinyl-CoA pathway (the succinyl-CoA pathway), encoding nucleic acids for host deficiencies in the enzymes succinyl-CoA synthetase, CoA-dependent succinic semialdehyde dehydrogenase and/or 4-hydroxybutanoate dehydrogenase are to be exogenously expressed in the recipient host. Selection of 4-HB biosynthesis through the α-ketoglutarate to succinic semialdehyde pathway (the α-ketoglutarate pathway) can utilize exogenous expression for host deficiencies in one or more of the enzymes for glutamate:succinic semialdehyde transaminase, glutamate decarboxylase and/or 4-hydroxybutanoate dehydrogenase, or α-ketoglutarate decarboxylase and 4-hydroxybutanoate dehydrogenase. One skilled in the art can readily determine pathway enzymes for production of 4-HB or BDO, as disclosed herein.

Depending on the 4-HB or BDO biosynthetic pathway constituents of a selected host microbial organism, the non-naturally occurring microbial organisms of the invention will include at least one exogenously expressed 4-HB or BDO pathway-encoding nucleic acid and up to all encoding nucleic acids for one or more 4-HB or BDO biosynthetic pathways. For example, 4-HB or BDO biosynthesis can be established in a host deficient in a pathway enzyme or protein through exogenous expression of the corresponding encoding nucleic acid. In a host deficient in all enzymes or proteins of a 4-HB or BDO pathway, exogenous expression of all enzyme or proteins in the pathway can be included, although it is understood that all enzymes or proteins of a pathway can be expressed even if the host contains at least one of the pathway enzymes or proteins. For example, 4-HB biosynthesis can be established from all five pathways in a host deficient in 4-hydroxybutanoate dehydrogenase through exogenous expression of a 4-hydroxybutanoate dehydrogenase encoding nucleic acid. In contrast, 4-HB biosynthesis can be established from all five pathways in a host deficient in all eight enzymes through exogenous expression of all eight of CoA-independent succinic semialdehyde dehydrogenase, succinyl-CoA synthetase, CoA-dependent succinic semialdehyde dehydrogenase, glutamate: succinic semialdehyde transaminase, glutamate decarboxylase, α-ketoglutarate decarboxylase, α-ketoglutarate dehydrogenase and 4-hydroxybutanoate dehydrogenase.

Given the teachings and guidance provided herein, those skilled in the art will understand that the number of encoding nucleic acids to introduce in an expressible form will, at least, parallel the 4-HB or BDO pathway deficiencies of the selected host microbial organism. Therefore, a non-naturally occurring microbial organism of the invention can have one, two, three, four, five, six, seven, eight or up to all nucleic acids encoding the enzymes disclosed herein constituting one or more 4-HB or BDO biosynthetic pathways. In some embodiments, the non-naturally occurring microbial organisms also can include other genetic modifications that facilitate or optimize 4-HB or BDO biosynthesis or that confer other useful functions onto the host microbial organism. One such other functionality can include, for example, augmentation of the synthesis of one or more of the 4-HB pathway precursors such as succinate, succinyl-CoA, α-ketoglutarate, 4-aminobutyrate, glutamate, acetoacetyl-CoA, and/or homoserine.

Generally, a host microbial organism is selected such that it produces the precursor of a 4-HB or BDOpathway, either as a naturally produced molecule or as an engineered product that either provides de novo production of a desired precursor or increased production of a precursor naturally produced by the host microbial organism. For example, succinyl-CoA, α-ketoglutarate, 4-aminobutyrate, glutamate, acetoacetyl-CoA, and homoserine are produced naturally in a host organism such as E. coli. A host organism can be engineered to increase production of a precursor, as disclosed herein. In addition, a microbial organism that has been engineered to produce a desired precursor can be used as a host organism and further engineered to express enzymes or proteins of a 4-HB or BDO pathway.

In some embodiments, a non-naturally occurring microbial organism of the invention is generated from a host that contains the enzymatic capability to synthesize 4-HB or BDO. In this specific embodiment it can be useful to increase the synthesis or accumulation of a 4-HB or BDO pathway product to, for example, drive 4-HB or BDO pathway reactions toward 4-HB or BDO production. Increased synthesis or accumulation can be accomplished by, for example, overexpression of nucleic acids encoding one or more of the 4-HB or BDO pathway enzymes disclosed herein. Over expression of the 4-HB or BDO pathway enzyme or enzymes can occur, for example, through exogenous expression of the endogenous gene or genes, or through exogenous expression of the heterologous gene or genes. Therefore, naturally occurring organisms can be readily generated to be non-naturally 4-HB or BDO producing microbial organisms of the invention through overexpression of one, two, three, four, five, six and so forth up to all nucleic acids encoding 4-HB or BDO biosynthetic pathway enzymes. In addition, a non-naturally occurring organism can be generated by mutagenesis of an endogenous gene that results in an increase in activity of an enzyme in the 4-HB or BDO biosynthetic pathway.

In particularly useful embodiments, exogenous expression of the encoding nucleic acids is employed. Exogenous expression confers the ability to custom tailor the expression and/or regulatory elements to the host and application to achieve a desired expression level that is controlled by the user. However, endogenous expression also can be utilized in other embodiments such as by removing a negative regulatory effector or induction of the gene's promoter when linked to an inducible promoter or other regulatory element. Thus, an endogenous gene having a naturally occurring inducible promoter can be up-regulated by providing the appropriate inducing agent, or the regulatory region of an endogenous gene can be engineered to incorporate an inducible regulatory element, thereby allowing the regulation of increased expression of an endogenous gene at a desired time. Similarly, an inducible promoter can be included as a regulatory element for an exogenous gene introduced into a non-naturally occurring microbial organism (see Examples).

“Exogenous” as it is used herein is intended to mean that the referenced molecule or the referenced activity is introduced into the host microbial organism. The molecule can be introduced, for example, by introduction of an encoding nucleic acid into the host genetic material such as by integration into a host chromosome or as non-chromosomal genetic material such as a plasmid. Therefore, the term as it is used in reference to expression of an encoding nucleic acid refers to introduction of the encoding nucleic acid in an expressible form into the microbial organism. When used in reference to a biosynthetic activity, the term refers to an activity that is introduced into the host reference organism. The source can be, for example, a homologous or heterologous encoding nucleic acid that expresses the referenced activity following introduction into the host microbial organism. Therefore, the term “endogenous” refers to a referenced molecule or activity that is present in the host. Similarly, the term when used in reference to expression of an encoding nucleic acid refers to expression of an encoding nucleic acid contained within the microbial organism. The term “heterologous” refers to a molecule or activity derived from a source other than the referenced species whereas “homologous” refers to a molecule or activity derived from the host microbial organism. Accordingly, exogenous expression of an encoding nucleic acid of the invention can utilize either or both a heterologous or homologous encoding nucleic acid.

Sources of encoding nucleic acids for a 4-HB or BDO pathway enzyme can include, for example, any species where the encoded gene product is capable of catalyzing the referenced reaction. Such species include both prokaryotic and eukaryotic organisms including, but not limited to, bacteria, including archaea and eubacteria, and eukaryotes, including yeast, plant, insect, animal, and mammal, including human. Exemplary species for such sources include, for example, Escherichia coli, Saccharomyces cerevisiae, Saccharomyces kluyveri, Clostridium kluyveri, Clostridium acetobutylicum, Clostridium beijerinckii, Clostridium saccharoperbutylacetonicum, Clostridium perfringens, Clostridium difficile, Clostridium botulinum, Clostridium tyrobutyricum, Clostridium tetanomorphum, Clostridium tetani, Clostridium propionicum, Clostridium aminobutyricum, Clostridium subterminale, Clostridium sticklandii, Ralstonia eutropha, Mycobacterium bovis, Mycobacterium tuberculosis, Porphyromonas gingivalis, Arabidopsis thaliana, Thermus thermophilus, Pseudomonas species, including Pseudomonas aeruginosa, Pseudomonas putida, Pseudomonas stutzeri, Pseudomonas fluorescens, Homo sapiens, Oryctolagus cuniculus, Rhodobacter spaeroides, Thermoanaerobacter brockii, Metallosphaera sedula, Leuconostoc mesenteroides, Chloroflexus aurantiacus, Roseiflexus castenholzii, Erythrobacter, Simmondsia chinensis, Acinetobacter species, including Acinetobacter calcoaceticus and Acinetobacter baylyi, Porphyromonas gingivalis, Sulfolobus tokodaii, Sulfolobus solfataricus, Sulfolobus acidocaldarius, Bacillus subtilis, Bacillus cereus, Bacillus megaterium, Bacillus brevis, Bacillus pumilus, Rattus norvegicus, Klebsiella pneumonia, Klebsiella oxytoca, Euglena gracilis, Treponema denticola, Moorella thermoacetica, Thermotoga maritima, Halobacterium salinarum, Geobacillus stearothermophilus, Aeropyrum pernix, Sus scrofa, Caenorhabditis elegans, Corynebacterium glutamicum, Acidaminococcus fermentans, Lactococcus lactis, Lactobacillus plantarum, Streptococcus thermophilus, Enterobacter aerogenes, Candida, Aspergillus terreus, Pedicoccus pentosaceus, Zymomonas mobilus, Acetobacter pasteurians, Kluyveromyces lactis, Eubacterium barkeri, Bacteroides capillosus, Anaerotruncus colihominis, Natranaerobius thermophilusm, Campylobacter jejuni, Haemophilus influenzae, Serratia marcescens, Citrobacter amalonaticus, Myxococcus xanthus, Fusobacterium nuleatum, Penicillium chrysogenum marine gamma proteobacterium, butyrate-producing bacterium, and others disclosed herein (see Examples). For example, microbial organisms having 4-HB or BDO biosynthetic production are exemplified herein with reference to E. coli and yeast hosts. However, with the complete genome sequence available for now more than 550 species (with more than half of these available on public databases such as the NCBI), including 395 microorganism genomes and a variety of yeast, fungi, plant, and mammalian genomes, the identification of genes encoding the requisite 4-HB or BDO biosynthetic activity for one or more genes in related or distant species, including for example, homologues, orthologs, paralogs and nonorthologous gene displacements of known genes, and the interchange of genetic alterations between organisms is routine and well known in the art. Accordingly, the metabolic alterations enabling biosynthesis of 4-HB or BDO and other compounds of the invention described herein with reference to a particular organism such as E. coli or yeast can be readily applied to other microorganisms, including prokaryotic and eukaryotic organisms alike. Given the teachings and guidance provided herein, those skilled in the art will know that a metabolic alteration exemplified in one organism can be applied equally to other organisms.

In some instances, such as when an alternative 4-HB or BDO biosynthetic pathway exists in an unrelated species, 4-HB or BDO biosynthesis can be conferred onto the host species by, for example, exogenous expression of a paralog or paralogs from the unrelated species that catalyzes a similar, yet non-identical metabolic reaction to replace the referenced reaction. Because certain differences among metabolic networks exist between different organisms, those skilled in the art will understand that the actual genes usage between different organisms may differ. However, given the teachings and guidance provided herein, those skilled in the art also will understand that the teachings and methods of the invention can be applied to all microbial organisms using the cognate metabolic alterations to those exemplified herein to construct a microbial organism in a species of interest that will synthesize 4-HB, such as monomeric 4-HB, or BDO.

Host microbial organisms can be selected from, and the non-naturally occurring microbial organisms generated in, for example, bacteria, yeast, fungus or any of a variety of other microorganisms applicable to fermentation processes. Exemplary bacteria include species selected from Escherichia coli, Klebsiella oxytoca, Anaerobiospirillum succiniciproducens, Actinobacillus succinogenes, Mannheimia succiniciproducens, Rhizobium etli, Bacillus subtilis, Corynebacterium glutamicum, Gluconobacter oxydans, Zymomonas mobilis, Lactococcus lactis, Lactobacillus plantarum, Streptomyces coelicolor, Clostridium acetobutylicum, Pseudomonas fluorescens, and Pseudomonas putida. Exemplary yeasts or fungi include species selected from Saccharomyces cerevisiae, Schizosaccharomyces pombe, Kluyveromyces lactis, Kluyveromyces marxianus, Aspergillus terreus, Aspergillus niger and Pichia pastoris. E. coli is a particularly useful host organisms since it is a well characterized microbial organism suitable for genetic engineering. Other particularly useful host organisms include yeast such as Saccharomyces cerevisiae.

Methods for constructing and testing the expression levels of a non-naturally occurring 4-HB- or BDO-producing host can be performed, for example, by recombinant and detection methods well known in the art. Such methods can be found described in, for example, Sambrook et al., Molecular Cloning: A Laboratory Manual, Third Ed., Cold Spring Harbor Laboratory, New York (2001); Ausubel et al., Current Protocols in Molecular Biology, John Wiley and Sons, Baltimore, Md. (1999). 4-HB and GBL can be separated by, for example, HPLC using a Spherisorb 5 ODS1 column and a mobile phase of 70% 10 mM phosphate buffer (pH=7) and 30% methanol, and detected using a UV detector at 215 nm (Hennessy et al. 2004, J. Forensic Sci. 46(6):1-9). BDO is detected by gas chromatography or by HPLC and refractive index detector using an Aminex HPX-87H column and a mobile phase of 0.5 mM sulfuric acid (Gonzalez-Pajuelo et al., Met. Eng. 7:329-336 (2005)).

Exogenous nucleic acid sequences involved in a pathway for production of 4-HB or BDO can be introduced stably or transiently into a host cell using techniques well known in the art including, but not limited to, conjugation, electroporation, chemical transformation, transduction, transfection, and ultrasound transformation. For exogenous expression in E. coli or other prokaryotic cells, some nucleic acid sequences in the genes or cDNAs of eukaryotic nucleic acids can encode targeting signals such as an N-terminal mitochondrial or other targeting signal, which can be removed before transformation into prokaryotic host cells, if desired. For example, removal of a mitochondrial leader sequence led to increased expression in E. coli (Hoffmeister et al., J. Biol. Chem. 280:4329-4338 (2005)). For exogenous expression in yeast or other eukaryotic cells, genes can be expressed in the cytosol without the addition of leader sequence, or can be targeted to mitochondrion or other organelles, or targeted for secretion, by the addition of a suitable targeting sequence such as a mitochondrial targeting or secretion signal suitable for the host cells. Thus, it is understood that appropriate modifications to a nucleic acid sequence to remove or include a targeting sequence can be incorporated into an exogenous nucleic acid sequence to impart desirable properties. Furthermore, genes can be subjected to codon optimization with techniques well known in the art to achieve optimized expression of the proteins.

An expression vector or vectors can be constructed to harbor one or more 4-HB biosynthetic pathway and/or one or more BDO biosynthetic encoding nucleic acids as exemplified herein operably linked to expression control sequences functional in the host organism. Expression vectors applicable for use in the microbial host organisms of the invention include, for example, plasmids, phage vectors, viral vectors, episomes and artificial chromosomes, including vectors and selection sequences or markers operable for stable integration into a host chromosome. Additionally, the expression vectors can include one or more selectable marker genes and appropriate expression control sequences. Selectable marker genes also can be included that, for example, provide resistance to antibiotics or toxins, complement auxotrophic deficiencies, or supply critical nutrients not in the culture media. Expression control sequences can include constitutive and inducible promoters, transcription enhancers, transcription terminators, and the like which are well known in the art. When two or more exogenous encoding nucleic acids are to be co-expressed, both nucleic acids can be inserted, for example, into a single expression vector or in separate expression vectors. For single vector expression, the encoding nucleic acids can be operationally linked to one common expression control sequence or linked to different expression control sequences, such as one inducible promoter and one constitutive promoter. The transformation of exogenous nucleic acid sequences involved in a metabolic or synthetic pathway can be confirmed using methods well known in the art. Such methods include, for example, nucleic acid analysis such as Northern blots or polymerase chain reaction (PCR) amplification of mRNA, or immunoblotting for expression of gene products, or other suitable analytical methods to test the expression of an introduced nucleic acid sequence or its corresponding gene product. It is understood by those skilled in the art that the exogenous nucleic acid is expressed in a sufficient amount to produce the desired product, and it is further understood that expression levels can be optimized to obtain sufficient expression using methods well known in the art and as disclosed herein.

The non-naturally occurring microbial organisms of the invention are constructed using methods well known in the art as exemplified herein to exogenously express at least one nucleic acid encoding a 4-HB or BDO pathway enzyme in sufficient amounts to produce 4-HB, such as monomeric 4-HB, or BDO. It is understood that the microbial organisms of the invention are cultured under conditions sufficient to produce 4-HB or BDO. Exemplary levels of expression for 4-HB enzymes in each pathway are described further below in the Examples. Following the teachings and guidance provided herein, the non-naturally occurring microbial organisms of the invention can achieve biosynthesis of 4-HB, such as monomeric 4-HB, or BDO resulting in intracellular concentrations between about 0.1-200 mM or more, for example, 0.1-25 mM or more. Generally, the intracellular concentration of 4-HB, such as monomeric 4-HB, or BDO is between about 3-150 mM or more, particularly about 5-125 mM or more, and more particularly between about 8-100 mM, for example, about 3-20 mM, particularly between about 5-15 mM and more particularly between about 8-12 mM, including about 10 mM, 20 mM, 50 mM, 80 mM or more. Intracellular concentrations between and above each of these exemplary ranges also can be achieved from the non-naturally occurring microbial organisms of the invention. In particular embodiments, the microbial organisms of the invention, particularly strains such as those disclosed herein (see Examples XII-XIX and Table 28), can provide improved production of a desired product such as BDO by increasing the production of BDO and/or decreasing undesirable byproducts. Such production levels include, but are not limited to, those disclosed herein and including from about 1 gram to about 25 grams per liter, for example about 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, or even higher amounts of product per liter.

In some embodiments, culture conditions include anaerobic or substantially anaerobic growth or maintenance conditions. Exemplary anaerobic conditions have been described previously and are well known in the art. Exemplary anaerobic conditions for fermentation processes are described herein and are described, for example, in U.S. patent application Ser. No. 11/891,602, filed Aug. 10, 2007. Any of these conditions can be employed with the non-naturally occurring microbial organisms as well as other anaerobic conditions well known in the art. Under such anaerobic conditions, the 4-HB or BDO producers can synthesize 4-HB or BDO at intracellular concentrations of 5-10 mM or more as well as all other concentrations exemplified herein. It is understood that, even though the above description refers to intracellular concentrations, 4-HB or BDO producing microbial organisms can produce 4-HB or BDO intracellularly and/or secrete the product into the culture medium.

The culture conditions can include, for example, liquid culture procedures as well as fermentation and other large scale culture procedures. As described herein, particularly useful yields of the biosynthetic products of the invention can be obtained under anaerobic or substantially anaerobic culture conditions.

As described herein, one exemplary growth condition for achieving biosynthesis of 4-HB or BDO includes anaerobic culture or fermentation conditions. In certain embodiments, the non-naturally occurring microbial organisms of the invention can be sustained, cultured or fermented under anaerobic or substantially anaerobic conditions. Briefly, anaerobic conditions refers to an environment devoid of oxygen. Substantially anaerobic conditions include, for example, a culture, batch fermentation or continuous fermentation such that the dissolved oxygen concentration in the medium remains between 0 and 10% of saturation. Substantially anaerobic conditions also includes growing or resting cells in liquid medium or on solid agar inside a sealed chamber maintained with an atmosphere of less than 1% oxygen. The percent of oxygen can be maintained by, for example, sparging the culture with an N₂/CO₂mixture or other suitable non-oxygen gas or gases.

The invention also provides a non-naturally occurring microbial biocatalyst including a microbial organism having 4-hydroxybutanoic acid (4-HB) and 1,4-butanediol (BDO) biosynthetic pathways that include at least one exogenous nucleic acid encoding 4-hydroxybutanoate dehydrogenase, CoA-independent succinic semialdehyde dehydrogenase, succinyl-CoA synthetase, CoA-dependent succinic semialdehyde dehydrogenase, 4-hydroxybutyrate:CoA transferase, glutamate: succinic semialdehyde transaminase, glutamate decarboxylase, CoA-independent aldehyde dehydrogenase, CoA-dependent aldehyde dehydrogenase or alcohol dehydrogenase, wherein the exogenous nucleic acid is expressed in sufficient amounts to produce 1,4-butanediol (BDO). 4-Hydroxybutyrate:CoA transferase also is known as 4-hydroxybutyryl CoA:acetyl-CoA transferase. Additional 4-HB or BDO pathway enzymes are also disclosed herein (see Examples and FIGS. 8-13).

The invention further provides non-naturally occurring microbial biocatalyst including a microbial organism having 4-hydroxybutanoic acid (4-HB) and 1,4-butanediol (BDO) biosynthetic pathways, the pathways include at least one exogenous nucleic acid encoding 4-hydroxybutanoate dehydrogenase, succinyl-CoA synthetase, CoA-dependent succinic semialdehyde dehydrogenase, 4-hydroxybutyrate:CoA transferase, 4-butyrate kinase, phosphotransbutyrylase, α-ketoglutarate decarboxylase, aldehyde dehydrogenase, alcohol dehydrogenase or an aldehyde/alcohol dehydrogenase, wherein the exogenous nucleic acid is expressed in sufficient amounts to produce 1,4-butanediol (BDO).

Non-naturally occurring microbial organisms also can be generated which biosynthesize BDO. As with the 4-HB producing microbial organisms of the invention, the BDO producing microbial organisms also can produce intracellularly or secret the BDO into the culture medium. Following the teachings and guidance provided previously for the construction of microbial organisms that synthesize 4-HB, additional BDO pathways can be incorporated into the 4-HB producing microbial organisms to generate organisms that also synthesize BDO and other BDO family compounds. The chemical synthesis of BDO and its downstream products are known. The non-naturally occurring microbial organisms of the invention capable of BDO biosynthesis circumvent these chemical synthesis using 4-HB as an entry point as illustrated in FIG. 1. As described further below, the 4-HB producers also can be used to chemically convert 4-HB to GBL and then to BDO or THF, for example. Alternatively, the 4-HB producers can be further modified to include biosynthetic capabilities for conversion of 4-HB and/or GBL to BDO.

The additional BDO pathways to introduce into 4-HB producers include, for example, the exogenous expression in a host deficient background or the overexpression of one or more of the enzymes exemplified in FIG. 1 as steps 9-13. One such pathway includes, for example, the enzyme activies necessary to carryout the reactions shown as steps 9, 12 and 13 in FIG. 1, where the aldehyde and alcohol dehydrogenases can be separate enzymes or a multifunctional enzyme having both aldehyde and alcohol dehydrogenase activity. Another such pathway includes, for example, the enzyme activities necessary to carry out the reactions shown as steps 10, 11, 12 and 13 in FIG. 1, also where the aldehyde and alcohol dehydrogenases can be separate enzymes or a multifunctional enzyme having both aldehyde and alcohol dehydrogenase activity. Accordingly, the additional BDO pathways to introduce into 4-HB producers include, for example, the exogenous expression in a host deficient background or the overexpression of one or more of a 4-hydroxybutyrate:CoA transferase, butyrate kinase, phosphotransbutyrylase, CoA-independent aldehyde dehydrogenase, CoA-dependent aldehyde dehydrogenase or an alcohol dehydrogenase. In the absence of endogenous acyl-CoA synthetase capable of modifying 4-HB, the non-naturally occurring BDO producing microbial organisms can further include an exogenous acyl-CoA synthetase selective for 4-HB, or the combination of multiple enzymes that have as a net reaction conversion of 4-HB into 4-HB-CoA. As exemplified further below in the Examples, butyrate kinase and phosphotransbutyrylase exhibit BDO pathway activity and catalyze the conversions illustrated in FIG. 1 with a 4-HB substrate. Therefore, these enzymes also can be referred to herein as 4-hydroxybutyrate kinase and phosphotranshydroxybutyrylase respectively.

Exemplary alcohol and aldehyde dehydrogenases that can be used for these in vivo conversions from 4-HB to BDO are listed below in Table 1.

Other exemplary enzymes and pathways are disclosed herein (see Examples). Furthermore, it is understood that enzymes can be utilized for carry out reactions for which the substrate is not the natural substrate. While the activity for the non-natural substrate may be lower than the natural substrate, it is understood that such enzymes can be utilized, either as naturally occurring or modified using the directed evolution or adaptive evolution, as disclosed herein (see also Examples).

BDO production through any of the pathways disclosed herein are based, in part, on the identification of the appropriate enzymes for conversion of precursors to BDO. A number of specific enzymes for several of the reaction steps have been identified. For those transformations where enzymes specific to the reaction precursors have not been identified, enzyme candidates have been identified that are best suited for catalyzing the reaction steps. Enzymes have been shown to operate on a broad range of substrates, as discussed below. In addition, advances in the field of protein engineering also make it feasible to alter enzymes to act efficiently on substrates, even if not a natural substrate. Described below are several examples of broad-specificity enzymes from diverse classes suitable for a BDO pathway as well as methods that have been used for evolving enzymes to act on non-natural substrates.

A key class of enzymes in BDO pathways is the oxidoreductases that interconvert ketones or aldehydes to alcohols (1.1.1). Numerous exemplary enzymes in this class can operate on a wide range of substrates. An alcohol dehydrogenase (1.1.1.1) purified from the soil bacterium Brevibacterium sp KU 1309 (Hirano et al., J. Biosc. Bioeng. 100:318-322 (2005)) was shown to operate on a plethora of aliphatic as well as aromatic alcohols with high activities. Table 2 shows the activity of the enzyme and its K_mon different alcohols. The enzyme is reversible and has very high activity on several aldehydes also (Table 3).

Lactate dehydrogenase (1.1.1.27) from Ralstonia eutropha is another enzyme that has been demonstrated to have high activities on several 2-oxoacids such as 2-oxobutyrate, 2-oxopentanoate and 2-oxoglutarate (a C5 compound analogous to 2-oxoadipate) (Steinbuchel and Schlegel, Eur. J. Biochem. 130:329-334 (1983)). Column 2 in Table 4 demonstrates the activities of ldhA from R. eutropha (formerly A. eutrophus) on different substrates (Steinbuchel and Schlegel, supra, 1983).

Oxidoreductases that can convert 2-oxoacids to their acyl-CoA counterparts (1.2.1) have been shown to accept multiple substrates as well. For example, branched-chain 2-keto-acid dehydrogenase complex (BCKAD), also known as 2-oxoisovalerate dehydrogenase (1.2.1.25), participates in branched-chain amino acid degradation pathways, converting 2-keto acids derivatives of valine, leucine and isoleucine to their acyl-CoA derivatives and CO₂. In some organisms including Rattus norvegicus (Paxton et al., Biochem. J. 234:295-303 (1986)) and Saccharomyces cerevisiae (Sinclair et al., Biochem. Mol. Biol. Int. 32:911-922 (1993), this complex has been shown to have a broad substrate range that includes linear oxo-acids such as 2-oxobutanoate and alpha-ketoglutarate, in addition to the branched-chain amino acid precursors.

Members of yet another class of enzymes, namely aminotransferases (2.6.1), have been reported to act on multiple substrates. Aspartate aminotransferase (aspAT) from Pyrococcus fursious has been identified, expressed in E. coli and the recombinant protein characterized to demonstrate that the enzyme has the highest activities towards aspartate and alpha-ketoglutarate but lower, yet significant activities towards alanine, glutamate and the aromatic amino acids (Ward et al., Archaea 133-141 (2002)). In another instance, an aminotransferase indentified from Leishmania mexicana and expressed in E. coli (Vernal et al., FEMS Microbiol. Lett. 229:217-222 (2003)) was reported to have a broad substrate specificity towards tyrosine (activity considered 100% on tyrosine), phenylalanine (90%), tryptophan (85%), aspartate (30%), leucine (25%) and methionine (25%), respectively (Vernal et al., Mol. Biochem. Parasitol 96:83-92 (1998)). Similar broad specificity has been reported for a tyrosine aminotransferase from Trypanosoma cruzi, even though both of these enzymes have a sequence homology of only 6%. The latter enzyme can accept leucine, methionine as well as tyrosine, phenylalanine, tryptophan and alanine as efficient amino donors (Nowicki et al., Biochim. Biophys. Acta 1546: 268-281 (2001)).

CoA transferases (2.8.3) have been demonstrated to have the ability to act on more than one substrate. Specifically, a CoA transferase was purified from Clostridium acetobutylicum and was reported to have the highest activities on acetate, propionate, and butyrate. It also had significant activities with valerate, isobutyrate, and crotonate (Wiesenborn et al., Appl. Environ. Microbiol. 55:323-329 (1989)). In another study, the E. coli enzyme acyl-CoA:acetate-CoA transferase, also known as acetate-CoA transferase (EC 2.8.3.8), has been shown to transfer the CoA moiety to acetate from a variety of branched and linear acyl-CoA substrates, including isobutyrate (Matthies and Schink, App. Environm. Microbiol. 58:1435-1439 (1992)), valerate (Vanderwinkel et al., Biochem. Biophys. Res Commun. 33:902-908 (1968b)) and butanoate (Vanderwinkel et al., Biochem. Biophys. Res Commun. 33:902-908(1968a).

Other enzyme classes additionally support broad substrate specificity for enzymes. Some isomerases (5.3.3) have also been proven to operate on multiple substrates. For example, L-rhamnose isomerase from Pseudomonas stutzeri catalyzes the isomerization between various aldoalses and ketoses (Yoshida et al., J. Mol. Biol. 365:1505-1516 (2007)). These include isomerization between L-rhamnose and L-rhamnulose, L-mannose and L-fructose, L-xylose and L-xylulose, D-ribose and D-ribulose, and D-allose and D-psicose.

In yet another class of enzymes, the phosphotransferases (2.7.1), the homoserine kinase (2.7.1.39) from E. coli that converts L-homoserine to L-homoserine phosphate, was found to phosphorylate numerous homoserine analogs. In these substrates, the carboxyl functional group at the R-position had been replaced by an ester or by a hydroxymethyl group (Huo and Viola, Biochemistry 35:16180-16185 (1996)). Table 5 demonstrates the broad substrate specificity of this kinase.

Another class of enzymes useful in BDO pathways is the acid-thiol ligases (6.2.1). Like enzymes in other classes, certain enzymes in this class have been determined to have broad substrate specificity. For example, acyl CoA ligase from Pseudomonas putida has been demonstrated to work on several aliphatic substrates including acetic, propionic, butyric, valeric, hexanoic, heptanoic, and octanoic acids and on aromatic compounds such as phenylacetic and phenoxyacetic acids (Fernandez-Valverde et al., Appl. Environ. Microbiol. 59:1149-1154 (1993)). A related enzyme, malonyl CoA synthetase (6.3.4.9) from Rhizobium trifolii could convert several diacids, namely, ethyl-, propyl-, allyl-, isopropyl-, dimethyl-, cyclopropyl-, cyclopropylmethylene-, cyclobutyl-, and benzyl-malonate into their corresponding monothioesters (Pohl et al., J. Am. Chem. Soc. 123:5822-5823 (2001)). Similarly, decarboxylases (4.1.1) have also been found with broad substrate ranges. Pyruvate decarboxylase (PDC), also termed keto-acid decarboxylase, is a key enzyme in alcoholic fermentation, catalyzing the decarboxylation of pyruvate to acetaldehyde. The enzyme isolated from Saccharomyces cerevisiae has a broad substrate range for aliphatic 2-keto acids including 2-ketobutyrate, 2-ketovalerate, and 2-phenylpyruvate (L1 and Jordan, Biochemistry 38:10004-10012 (1999)). Similarly, benzoylformate decarboxylase has a broad substrate range and has been the target of enzyme engineering studies. The enzyme from Pseudomonas putida has been extensively studied and crystal structures of this enzyme are available (Polovnikova et al., Biochemistry 42:1820-1830 (2003); Hasson et al., Biochemistry 37:9918-9930 (1998)). Branched chain alpha-ketoacid decarboxylase (BCKA) has been shown to act on a variety of compounds varying in chain length from 3 to 6 carbons (Oku and Kaneda, J. Biol. Chem. 263:18386-18396 (1998); Smit et al., Appl. Environ. Microbiol. 71:303-311 (2005b)). The enzyme in Lactococcus lactis has been characterized on a variety of branched and linear substrates including 2-oxobutanoate, 2-oxohexanoate, 2-oxopentanoate, 3-methyl-2-oxobutanoate, 4-methyl-2-oxobutanoate and isocaproate (Smit et al., Appl. Environ. Microbiol. 71:303-311 (2005a).

Interestingly, enzymes known to have one dominant activity have also been reported to catalyze a very different function. For example, the cofactor-dependent phosphoglycerate mutase (5.4.2.1) from Bacillus stearothermophilus and Bacillus subtilis is known to function as a phosphatase as well (Rigden et al., Protein Sci. 10:1835-1846 (2001)). The enzyme from B. stearothermophilus is known to have activity on several substrates, including 3-phosphoglycerate, alpha-napthylphosphate, p-nitrophenylphosphate, AMP, fructose-6-phosphate, ribose-5-phosphate and CMP.

In contrast to these examples where the enzymes naturally have broad substrate specificities, numerous enzymes have been modified using directed evolution to broaden their specificity towards their non-natural substrates. Alternatively, the substrate preference of an enzyme has also been changed using directed evolution. Therefore, it is feasible to engineer a given enzyme for efficient function on a natural, for example, improved efficiency, or a non-natural substrate, for example, increased efficiency. For example, it has been reported that the enantioselectivity of a lipase from Pseudomonas aeruginosa was improved significantly (Reetz et al., Agnew. Chem. Int. Ed Engl. 36:2830-2832 (1997)). This enzyme hydrolyzed p-nitrophenyl 2-methyldecanoate with only 2% enantiomeric excess (ee) in favor of the (S)-acid. However, after four successive rounds of error-prone mutagenesis and screening, a variant was produced that catalyzed the requisite reaction with 81% ee (Reetz et al., Agnew. Chem. Int. Ed Engl. 36:2830-2832 (1997)).

Directed evolution methods have been used to modify an enzyme to function on an array of non-natural substrates. The substrate specificity of the lipase in P. aeruginosa was broadened by randomization of amino acid residues near the active site. This allowed for the acceptance of alpha-substituted carboxylic acid esters by this enzyme (Reetz et al., Agnew. Chem. Int. Ed Engl. 44:4192-4196 (2005)). In another successful modification of an enzyme, DNA shuffling was employed to create an Escherichia coli aminotransferase that accepted (3-branched substrates, which were poorly accepted by the wild-type enzyme (Yano et al., Proc. Nat. Acad. Sci. U.S.A. 95:5511-5515 (1998)). Specifically, at the end of four rounds of shuffling, the activity of aspartate aminotransferase for valine and 2-oxovaline increased by up to five orders of magnitude, while decreasing the activity towards the natural substrate, aspartate, by up to 30-fold. Recently, an algorithm was used to design a retro-aldolase that could be used to catalyze the carbon-carbon bond cleavage in a non-natural and non-biological substrate, 4-hydroxy-4-(6-methoxy-2-naphthyl)-2-butanone (Jiang et al., Science 319:1387-1391 (2008)). These algorithms used different combinations of four different catalytic motifs to design new enzyme, and 20 of the selected designs for experimental characterization had four-fold improved rates over the uncatalyzed reaction (Jiang et al., Science 319:1387-1391 (2008)). Thus, not only are these engineering approaches capable of expanding the array of substrates on which an enzyme can act, but they allow the design and construction of very efficient enzymes. For example, a method of DNA shuffling (random chimeragenesis on transient templates or RACHITT) was reported to lead to an engineered monooxygenase that had an improved rate of desulfurization on complex substrates as well as 20-fold faster conversion of a non-natural substrate (Coco et al., Nat. Biotechnol. 19:354-359 (2001)). Similarly, the specific activity of a sluggish mutant triosephosphate isomerase enzyme was improved up to 19-fold from 1.3 fold (Hermes et al., Proc. Nat. Acad. Sci. U.S.A. 87:696-700 1990)). This enhancement in specific activity was accomplished by using random mutagenesis over the whole length of the protein and the improvement could be traced back to mutations in six amino acid residues.

The effectiveness of protein engineering approaches to alter the substrate specificity of an enzyme for a desired substrate has also been demonstrated in several studies. Isopropylmalate dehydrogenase from Thermus thermophilus was modified by changing residues close to the active site so that it could now act on malate and D-lactate as substrates (Fujita et al., Biosci. Biotechnol. Biochem. 65:2695-2700 (2001)). In this study as well as in others, it was pointed out that one or a few residues could be modified to alter the substrate specificity. For example, the dihydroflavonol 4-reductase for which a single amino acid was changed in the presumed substrate-binding region could preferentially reduce dihydrokaempferol (Johnson et al., Plant. J. 25:325-333 (2001)). The substrate specificity of a very specific isocitrate dehydrogenase from Escherichia coli was changed form isocitrate to isopropylmalate by changing one residue in the active site (Doyle et al., Biochemistry 40:4234-4241 (2001)). Similarly, the cofactor specificity of a NAD⁺-dependent 1,5-hydroxyprostaglandin dehydrogenase was altered to NADP⁺ by changing a few residues near the N-terminal end (Cho et al., Arch. Biochem. Biophys. 419:139-146 (2003)). Sequence analysis and molecular modeling analysis were used to identify the key residues for modification, which were further studied by site-directed mutagenesis.

Numerous examples exist spanning diverse classes of enzymes where the function of enzyme was changed to favor one non-natural substrate over the natural substrate of the enzyme. A fucosidase was evolved from a galactosidase in E. coli by DNA shuffling and screening (Zhang et al., Proc. Natl. Acad. Sci. U.S.A. 94:4504-4509 (1997)). Similarly, aspartate aminotransferase from E. coli was converted into a tyrosine aminotransferase using homology modeling and site-directed mutagenesis (Onuffer and Kirsch, Protein Sci., 4:1750-1757 (1995)). Site-directed mutagenesis of two residues in the active site of benzoylformate decarboxylase from P. putida reportedly altered the affinity (K_m) towards natural and non-natural substrates (Siegert et al., Protein Eng Des Sel 18:345-357 (2005)). Cytochrome c peroxidase (CCP) from Saccharomyces cerevisiae was subjected to directed molecular evolution to generate mutants with increased activity against the classical peroxidase substrate guaiacol, thus changing the substrate specificity of CCP from the protein cytochrome c to a small organic molecule. After three rounds of DNA shuffling and screening, mutants were isolated which possessed a 300-fold increased activity against guaiacol and up to 1000-fold increased specificity for this substrate relative to that for the natural substrate (Iffland et al., Biochemistry 39:10790-10798 (2000)).

In some cases, enzymes with different substrate preferences than either of the parent enzymes have been obtained. For example, biphenyl-dioxygenase-mediated degradation of polychlorinated biphenyls was improved by shuffling genes from two bacteria, Pseudomonas pseudoalcaligens and Burkholderia cepacia (Kumamaru et al., Nat. Biotechnol. 16:663-666 (1998)). The resulting chimeric biphenyl oxygenases showed different substrate preferences than both the parental enzymes and enhanced the degradation activity towards related biphenyl compounds and single aromatic ring hydrocarbons such as toluene and benzene which were originally poor substrates for the enzyme.

In addition to changing enzyme specificity, it is also possible to enhance the activities on substrates for which the enzymes naturally have low activities. One study demonstrated that amino acid racemase from P. putida that had broad substrate specificity (on lysine, arginine, alanine, serine, methionine, cysteine, leucine and histidine among others) but low activity towards tryptophan could be improved significantly by random mutagenesis (Kino et al., Appl. Microbiol. Biotechnol. 73:1299-1305 (2007)). Similarly, the active site of the bovine BCKAD was engineered to favor alternate substrate acetyl-CoA (Meng and Chuang, Biochemistry 33:12879-12885 (1994)). An interesting aspect of these approaches is that even if random methods have been applied to generate these mutated enzymes with efficacious activities, the exact mutations or structural changes that confer the improvement in activity can be identified. For example, in the aforementioned study, the mutations that facilitated improved activity on tryptophan was traced back to two different positions.

Directed evolution has also been used to express proteins that are difficult to express. For example, by subjecting horseradish peroxidase to random mutagenesis and gene recombination, mutants were identified that had more than 14-fold higher activity than the wild type (Lin et al., Biotechnol. Prog. 15:467-471 (1999)).

Another example of directed evolution shows the extensive modifications to which an enzyme can be subjected to achieve a range of desired functions. The enzyme lactate dehydrogenase from Bacillus stearothermophilus was subjected to site-directed mutagenesis, and three amino acid substitutions were made at sites that were believed to determine the specificity towards different hydroxyacids (Clarke et al., Biochem. Biophys. Res. Commun. 148:15-23 (1987)). After these mutations, the specificity for oxaloacetate over pyruvate was increased to 500 in contrast to the wild type enzyme that had a catalytic specificity for pyruvate over oxaloacetate of 1000. This enzyme was further engineered using site-directed mutagenesis to have activity towards branched-chain substituted pyruvates (Wilks et al., Biochemistry 29:8587-8591 (1990)). Specifically, the enzyme had a 55-fold improvement in K_catfor alpha-ketoisocaproate. Three structural modifications were made in the same enzyme to change its substrate specificity from lactate to malate. The enzyme was highly active and specific towards malate (Wilks et al., Science 242:1541-1544 (1988)). The same enzyme from B. stearothermophilus was subsequently engineered to have high catalytic activity towards alpha-keto acids with positively charged side chains, such as those containing ammonium groups (Hogan et al., Biochemistry 34:4225-4230 (1995)). Mutants with acidic amino acids introduced at position 102 of the enzyme favored binding of such side chain ammonium groups. The results obtained proved that the mutants showed up to 25-fold improvements in k_cat/K_mvalues for omega-amino-alpha-keto acid substrates. Interestingly, this enzyme was also structurally modified to function as a phenyllactate dehydrogenase instead of a lactate dehydrogenase (Wilks et al., Biochemistry 31:7802-7806 1992). Restriction sites were introduced into the gene for the enzyme which allowed a region of the gene to be excised. This region coded for a mobile surface loop of the polypeptide (residues 98-110) which normally seals the active site from bulk solvent and is a major determinant of substrate specificity. The variable length and sequence loops were inserted so that hydroxyacid dehydrogenases with altered substrate specificities were generated. With one longer loop construction, activity with pyruvate was reduced one-million-fold but activity with phenylpyruvate was largely unaltered. A switch in specificity (k_at/K_m) of 390.000-fold was achieved. The 1700:1 selectivity of this enzyme for phenylpyruvate over pyruvate is that required in a phenyllactate dehydrogenase. The studies described above indicate that various approaches of enzyme engineering can be used to obtain enzymes for the BDO pathways as disclosed herein.

As disclosed herein, biosynthetic pathways to 1,4-butanediol from a number of central metabolic intermediates are can be utilized, including acetyl-CoA, succinyl-CoA, alpha-ketoglutarate, glutamate, 4-aminobutyrate, and homoserine. Acetyl-CoA, succinyl-CoA and alpha-ketoglutarate are common intermediates of the tricarboxylic acid (TCA) cycle, a series of reactions that is present in its entirety in nearly all living cells that utilize oxygen for cellular respiration and is present in truncated forms in a number of anaerobic organisms. Glutamate is an amino acid that is derived from alpha-ketoglutarate via glutamate dehydrogenase or any of a number of transamination reactions (see FIG. 8B). 4-aminobutyrate can be formed by the decarboxylation of glutamate (see FIG. 8B) or from acetoacetyl-CoA via the pathway disclosed in FIG. 9C. Acetoacetyl-CoA is derived from the condensation of two acetyl-CoA molecules by way of the enzyme, acetyl-coenzyme A acetyltransferase, or equivalently, acetoacetyl-coenzyme A thiolase. Homoserine is an intermediate in threonine and methionine metabolism, formed from oxaloacetate via aspartate. The conversion of oxaloacetate to homoserine requires one NADH, two NADPH, and one ATP.

Pathways other than those exemplified above also can be employed to generate the biosynthesis of BDO in non-naturally occurring microbial organisms. In one embodiment, biosynthesis can be achieved using a L-homoserine to BDO pathway (see FIG. 13). This pathway has a molar yield of 0.90 mol/mol glucose, which appears restricted by the availability of reducing equivalents. A second pathway synthesizes BDO from acetoacetyl-CoA and is capable of achieving the maximum theoretical yield of 1.091 mol/mol glucose (see FIG. 9). Implementation of either pathway can be achieved by introduction of two exogenous enzymes into a host organism such as E. coli, and both pathways can additionally complement BDO production via succinyl-CoA. Pathway enzymes, thermodynamics, theoretical yields and overall feasibility are described further below.

A homoserine pathway also can be engineered to generate BDO-producing microbial organisms. Homoserine is an intermediate in threonine and methionine metabolism, formed from oxaloacetate via aspartate. The conversion of oxaloacetate to homoserine requires one NADH, two NADPH, and one ATP (FIG. 2). Once formed, homoserine feeds into biosynthetic pathways for both threonine and methionine. In most organisms, high levels of threonine or methionine feedback to repress the homoserine biosynthesis pathway (Caspi et al., Nucleic Acids Res. 34:D511-D516 (1990)).

The transformation of homoserine to 4-hydroxybutyrate (4-HB) can be accomplished in two enzymatic steps as described herein. The first step of this pathway is deamination of homoserine by a putative ammonia lyase. In step 2, the product alkene, 4-hydroxybut-2-enoate is reduced to 4-HB by a putative reductase at the cost of one NADH. 4-HB can then be converted to BDO.

Enzymes available for catalyzing the above transformations are disclosed herein. For example, the ammonia lyase in step 1 of the pathway closely resembles the chemistry of aspartate ammonia-lyase (aspartase). Aspartase is a widespread enzyme in microorganisms, and has been characterized extensively (Viola, R. E., Mol. Biol. 74:295-341 (2008)). The crystal structure of the E. coli aspartase has been solved (Shi et al., Biochemistry 36:9136-9144 (1997)), so it is therefore possible to directly engineer mutations in the enzyme's active site that would alter its substrate specificity to include homoserine. The oxidoreductase in step 2 has chemistry similar to several well-characterized enzymes including fumarate reductase in the E. coli TCA cycle. Since the thermodynamics of this reaction are highly favorable, an endogenous reductase with broad substrate specificity will likely be able to reduce 4-hydroxybut-2-enoate. The yield of this pathway under anaerobic conditions is 0.9 mol BDO per mol glucose.

The succinyl-CoA pathway was found to have a higher yield due to the fact that it is more energetically efficient. The conversion of one oxaloacetate molecule to BDO via the homoserine pathway will require the expenditure of 2 ATP equivalents. Because the conversion of glucose to two oxaloacetate molecules can generate a maximum of 3 ATP molecules assuming PEP carboxykinase to be reversible, the overall conversion of glucose to BDO via homoserine has a negative energetic yield. As expected, if it is assumed that energy can be generated via respiration, the maximum yield of the homoserine pathway increases to 1.05 mol/mol glucose which is 96% of the succinyl-CoA pathway yield. The succinyl-CoA pathway can channel some of the carbon flux through pyruvate dehydrogenase and the oxidative branch of the TCA cycle to generate both reducing equivalents and succinyl-CoA without an energetic expenditure. Thus, it does not encounter the same energetic difficulties as the homoserine pathway because not all of the flux is channeled through oxaloacetate to succinyl-CoA to BDO. Overall, the homoserine pathway demonstrates a high-yielding route to BDO.

An acetoacetate pathway also can be engineered to generate BDO-producing microbial organisms. Acetoacetate can be formed from acetyl-CoA by enzymes involved in fatty acid metabolism, including acetyl-CoA acetyltransferase and acetoacetyl-CoA transferase. Biosynthetic routes through acetoacetate are also particularly useful in microbial organisms that can metabolize single carbon compounds such as carbon monoxide, carbon dioxide or methanol to form acetyl-CoA.

A three step route from acetoacetyl-CoA to 4-aminobutyrate (see FIG. 9C) can be used to synthesize BDO through acetoacetyl-CoA. 4-Aminobutyrate can be converted to succinic semialdehyde as shown in FIG. 8B. Succinic semialdehyde, which is one reduction step removed from succinyl-CoA or one decarboxylation step removed from α-ketoglutarate, can be converted to BDO following three reductions steps (FIG. 1). Briefly, step 1 of this pathway involves the conversion of acetoacetyl-CoA to acetoacetate by, for example, the E. coli acetoacetyl-CoA transferase encoded by the atoA and atoD genes (Hanai et al., Appl. Environ. Microbiol. 73: 7814-7818 (2007)). Step 2 of the acetoacetyl-CoA biopathway entails conversion of acetoacetate to 3-aminobutanoate by an w-aminotransferase. The w-amino acid:pyruvate aminotransferase (w-APT) from Alcaligens denitrificans was overexpressed in E. coli and shown to have a high activity toward 3-aminobutanoate in vitro (Yun et al., Appl. Environ. Microbiol. 70:2529-2534 (2004)).

In step 2, a putative aminomutase shifts the amine group from the 3- to the 4-position of the carbon backbone. An aminomutase performing this function on 3-aminobutanoate has not been characterized, but an enzyme from Clostridium sticklandii has a very similar mechanism. The enzyme, D-lysine-5,6-aminomutase, is involved in lysine biosynthesis.

The synthetic route to BDO from acetoacetyl-CoA passes through 4-aminobutanoate, a metabolite in E. coli that's normally formed from decarboxylation of glutamate. Once formed, 4-aminobutanoate can be converted to succinic semialdehyde by 4-aminobutanoate transaminase (2.6.1.19), an enzyme which has been biochemically characterized.

One consideration for selecting candidate enzymes in this pathway is the stereoselectivity of the enzymes involved in steps 2 and 3. The w-ABT in Alcaligens denitrificans is specific to the L-stereoisomer of 3-aminobutanoate, while D-lysine-5,6-aminomutase likely requires the D-stereoisomer. If enzymes with complementary stereoselectivity are not initially found or engineered, a third enzyme can be added to the pathway with racemase activity that can convert L-3-aminobutanoate to D-3-aminobutanoate. While amino acid racemases are widespread, whether these enzymes can function on w-amino acids is not known.

The maximum theoretical molar yield of this pathway under anaerobic conditions is 1.091 mol/mol glucose. In order to generate flux from acetoacetyl-CoA to BDO it was necessary to assume that acetyl-CoA:acetoacetyl-CoA transferase is reversible. The function of this enzyme in E. coli is to metabolize short-chain fatty acids by first converting them into thioesters.

While the operation of acetyl-CoA:acetoacetyl-CoA transferase in the acetate-consuming direction has not been demonstrated experimentally in E. coli, studies on similar enzymes in other organisms support the assumption that this reaction is reversible. The enzyme butyryl-CoA:acetate:CoA transferase in gut microbes Roseburia sp. and F. prasnitzii operates in the acetate utilizing direction to produce butyrate (Duncan et al., Appl. Environ. Microbiol. 68:5186-5190 (2002)). Another very similar enzyme, acetyl:succinate CoA-transferase in Trypanosoma brucei, also operates in the acetate utilizing direction. This reaction has a Δ_rxnG close to equilibrium, so high concentrations of acetate can likely drive the reaction in the direction of interest. At the maximum theoretical BDO production rate of 1.09 mol/mol glucose simulations predict that E. coli can generate 1.098 mol ATP per mol glucose with no fermentation byproducts. This ATP yield should be sufficient for cell growth, maintenance, and production. The acetoacetatyl-CoA biopathway is a high-yielding route to BDO from acetyl-CoA.

Therefore, in addition to any of the various modifications exemplified previously for establishing 4-HB biosynthesis in a selected host, the BDO producing microbial organisms can include any of the previous combinations and permutations of 4-HB pathway metabolic modifications as well as any combination of expression for CoA-independent aldehyde dehydrogenase, CoA-dependent aldehyde dehydrogenase or an alcohol dehydrogenase or other enzymes disclosed herein to generate biosynthetic pathways for GBL and/or BDO. Therefore, the BDO producers of the invention can have exogenous expression of, for example, one, two, three, four, five, six, seven, eight, nine, or up to all enzymes corresponding to any of the 4-HB pathway and/or any of the BDO pathway enzymes disclosed herein.

Design and construction of the genetically modified microbial organisms is carried out using methods well known in the art to achieve sufficient amounts of expression to produce BDO. In particular, the non-naturally occurring microbial organisms of the invention can achieve biosynthesis of BDO resulting in intracellular concentrations between about 0.1-200 mM or more, such as about 0.1-25 mM or more, as discussed above. For example, the intracellular concentration of BDO is between about 3-20 mM, particularly between about 5-15 mM and more particularly between about 8-12 mM, including about 10 mM or more. Intracellular concentrations between and above each of these exemplary ranges also can be achieved from the non-naturally occurring microbial organisms of the invention. As with the 4-HB producers, the BDO producers also can be sustained, cultured or fermented under anaerobic conditions.

The invention further provides a method for the production of 4-HB. The method includes culturing a non-naturally occurring microbial organism having a 4-hydroxybutanoic acid (4-HB) biosynthetic pathway comprising at least one exogenous nucleic acid encoding 4-hydroxybutanoate dehydrogenase, CoA-independent succinic semialdehyde dehydrogenase, succinyl-CoA synthetase, CoA-dependent succinic semialdehyde dehydrogenase, glutamate: succinic semialdehyde transaminase, α-ketoglutarate decarboxylase, or glutamate decarboxylase under substantially anaerobic conditions for a sufficient period of time to produce monomeric 4-hydroxybutanoic acid (4-HB). The method can additionally include chemical conversion of 4-HB to GBL and to BDO or THF, for example.

Additionally provided is a method for the production of 4-HB. The method includes culturing a non-naturally occurring microbial organism having a 4-hydroxybutanoic acid (4-HB) biosynthetic pathway including at least one exogenous nucleic acid encoding 4-hydroxybutanoate dehydrogenase, succinyl-CoA synthetase, CoA-dependent succinic semialdehyde dehydrogenase or α-ketoglutarate decarboxylase under substantially anaerobic conditions for a sufficient period of time to produce monomeric 4-hydroxybutanoic acid (4-HB). The 4-HB product can be secreted into the culture medium.

Further provided is a method for the production of BDO. The method includes culturing a non-naturally occurring microbial biocatalyst or microbial organism, comprising a microbial organism having 4-hydroxybutanoic acid (4-HB) and 1,4-butanediol (BDO) biosynthetic pathways, the pathways including at least one exogenous nucleic acid encoding 4-hydroxybutanoate dehydrogenase, succinyl-CoA synthetase, CoA-dependent succinic semialdehyde dehydrogenase, 4-hydroxybutyrate:CoA transferase, 4-hydroxybutyrate kinase, phosphotranshydroxybutyrylase, α-ketoglutarate decarboxylase, aldehyde dehydrogenase, alcohol dehydrogenase or an aldehyde/alcohol dehydrogenase for a sufficient period of time to produce 1,4-butanediol (BDO). The BDO product can be secreted into the culture medium.

Additionally provided are methods for producing BDO by culturing a non-naturally occurring microbial organism having a BDO pathway of the invention. The BDO pathway can comprise at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, under conditions and for a sufficient period of time to produce BDO, the BDO pathway comprising 4-aminobutyrate CoA transferase, 4-aminobutyryl-CoA hydrolase, 4-aminobutyrate-CoA ligase, 4-aminobutyryl-CoA oxidoreductase (deaminating), 4-aminobutyryl-CoA transaminase, or 4-hydroxybutyryl-CoA dehydrogenase (see Example VII and Table 17).

Alternatively, the BDO pathway can compare at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, under conditions and for a sufficient period of time to produce BDO, the BDO pathway comprising 4-aminobutyrate CoA transferase, 4-aminobutyryl-CoA hydrolase, 4-aminobutyrate-CoA ligase, 4-aminobutyryl-CoA reductase (alcohol forming), 4-aminobutyryl-CoA reductase, 4-aminobutan-1-ol dehydrogenase, 4-aminobutan-1-ol oxidoreductase (deaminating) or 4-aminobutan-1-ol transaminase (see Example VII and Table 18).

In addition, the invention provides a method for producing BDO, comprising culturing a non-naturally occurring microbial organism having a BDO pathway, the pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, under conditions and for a sufficient period of time to produce BDO, the BDO pathway comprising 4-aminobutyrate kinase, 4-aminobutyraldehyde dehydrogenase (phosphorylating), 4-aminobutan-1-ol dehydrogenase, 4-aminobutan-1-oloxidoreductase (deaminating), 4-aminobutan-1-ol transaminase, [(4-aminobutanolyl)oxy]phosphonic acid oxidoreductase (deaminating), [(4-aminobutanolyl)oxy]phosphonic acid transaminase, 4-hydroxybutyryl-phosphate dehydrogenase, or 4-hydroxybutyraldehyde dehydrogenase (phosphorylating) (see Example VII and Table 19).

The invention further provides a method for producing BDO, comprising culturing a non-naturally occurring microbial organism having a BDO pathway, the pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, under conditions and for a sufficient period of time to produce BDO, the BDO pathway comprising alpha-ketoglutarate 5-kinase, 2,5-dioxopentanoic semialdehyde dehydrogenase (phosphorylating), 2,5-dioxopentanoic acid reductase, alpha-ketoglutarate CoA transferase, alpha-ketoglutaryl-CoA hydrolase, alpha-ketoglutaryl-CoA ligase, alpha-ketoglutaryl-CoA reductase, 5-hydroxy-2-oxopentanoic acid dehydrogenase, alpha-ketoglutaryl-CoA reductase (alcohol forming), 5-hydroxy-2-oxopentanoic acid decarboxylase, or 5-hydroxy-2-oxopentanoic acid dehydrogenase (decarboxylation)(see Example VIII and Table 20).

The invention additionally provides a method for producing BDO, comprising culturing a non-naturally occurring microbial organism having a BDO pathway, the pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, under conditions and for a sufficient period of time to produce BDO, the BDO pathway comprising glutamate CoA transferase, glutamyl-CoA hydrolase, glutamyl-CoA ligase, glutamate 5-kinase, glutamate-5-semialdehyde dehydrogenase (phosphorylating), glutamyl-CoA reductase, glutamate-5-semialdehyde reductase, glutamyl-CoA reductase (alcohol forming), 2-amino-5-hydroxypentanoic acid oxidoreductase (deaminating), 2-amino-5-hydroxypentanoic acid transaminase, 5-hydroxy-2-oxopentanoic acid decarboxylase, 5-hydroxy-2-oxopentanoic acid dehydrogenase (decarboxylation)(see Example IX and Table 21).

The invention additionally includes a method for producing BDO, comprising culturing a non-naturally occurring microbial organism having a BDO pathway, the pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, under conditions and for a sufficient period of time to produce BDO, the BDO pathway comprising 3-hydroxybutyryl-CoA dehydrogenase, 3-hydroxybutyryl-CoA dehydratase, vinylacetyl-CoA A-isomerase, or 4-hydroxybutyryl-CoA dehydratase (see Example X and Table 22).

Also provided is a method for producing BDO, comprising culturing a non-naturally occurring microbial organism having a BDO pathway, the pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, under conditions and for a sufficient period of time to produce BDO, the BDO pathway comprising homoserine deaminase, homoserine CoA transferase, homoserine-CoA hydrolase, homoserine-CoA ligase, homoserine-CoA deaminase, 4-hydroxybut-2-enoyl-CoA transferase, 4-hydroxybut-2-enoyl-CoA hydrolase, 4-hydroxybut-2-enoyl-CoA ligase, 4-hydroxybut-2-enoate reductase, 4-hydroxybutyryl-CoA transferase, 4-hydroxybutyryl-CoA hydrolase, 4-hydroxybutyryl-CoA ligase, or 4-hydroxybut-2-enoyl-CoA reductase (see Example XI and Table 23).

The invention additionally provides a method for producing BDO, comprising culturing a non-naturally occurring microbial organism having a BDO pathway, the pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, under conditions and for a sufficient period of time to produce BDO, the BDO pathway comprising succinyl-CoA reductase (alcohol forming), 4-hydroxybutyryl-CoA hydrolase, 4-hydroxybutyryl-CoA ligase, 4-hydroxybutanal dehydrogenase (phosphorylating). Such a BDO pathway can further comprise succinyl-CoA reductase, 4-hydroxybutyrate dehydrogenase, 4-hydroxybutyryl-CoA transferase, 4-hydroxybutyrate kinase, phosphotrans-4-hydroxybutyrylase, 4-hydroxybutyryl-CoA reductase, 4-hydroxybutyryl-CoA reductase (alcohol forming), or 1,4-butanediol dehydrogenase.

Also provided is a method for producing BDO, comprising culturing a non-naturally occurring microbial organism having a BDO pathway, the pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO, under conditions and for a sufficient period of time to produce BDO, the BDO pathway comprising glutamate dehydrogenase, 4-aminobutyrate oxidoreductase (deaminating), 4-aminobutyrate transaminase, glutamate decarboxylase, 4-hydroxybutyryl-CoA hydrolase, 4-hydroxybutyryl-CoA ligase, 4-hydroxybutanal dehydrogenase (phosphorylating).

The invention additionally provides methods of producing a desired product using the genetically modified organisms disclosed herein that allow improved production of a desired product such as BDO by increasing the product or decreasing undesirable byproducts. Thus, the invention provides a method for producing 1,4-butanediol (BDO), comprising culturing the non-naturally occurring microbial organisms disclosed herein under conditions and for a sufficient period of time to produce BDO. In one embodiment, the invention provides a method of producing BDO using a non-naturally occurring microbial organism, comprising a microbial organism having a 1,4-butanediol (BDO) pathway comprising at least one exogenous nucleic acid encoding a BDO pathway enzyme expressed in a sufficient amount to produce BDO. In one embodiment, the microbial organism is genetically modified to express exogenous succinyl-CoA synthetase (see Example XII). For example, the succinyl-CoA synthetase can be encoded by an Escherichia coli sucCD genes.

In another embodiment, the microbial organism is genetically modified to express exogenous alpha-ketoglutarate decarboxylase (see Example XIII). For example, the alpha-ketoglutarate decarboxylase can be encoded by the Mycobacterium bovis sucA gene. In still another embodiment, the microbial organism is genetically modified to express exogenous succinate semialdehyde dehydrogenase and 4-hydroxybutyrate dehydrogenase and optionally 4-hydroxybutyryl-CoA/acetyl-CoA transferase (see Example XIII). For example, the succinate semialdehyde dehydrogenase (CoA-dependent), 4-hydroxybutyrate dehydrogenase and 4-hydroxybutyryl-CoA/acetyl-CoA transferase can be encoded by Porphyromonas gingivalis W83 genes. In an additional embodiment, the microbial organism is genetically modified to express exogenous butyrate kinase and phosphotransbutyrylase (see Example XIII). For example, the butyrate kinase and phosphotransbutyrylase can be encoded by Clostridium acetobutilicum buk1 and ptb genes.

In yet another embodiment, the microbial organism is genetically modified to express exogenous 4-hydroxybutyryl-CoA reductase (see Example XIII). For example, the 4-hydroxybutyryl-CoA reductase can be encoded by Clostridium beijerinckii ald gene. Additionally, in an embodiment of the invention, the microbial organism is genetically modified to express exogenous 4-hydroxybutanal reductase (see Example XIII). For example, the 4-hydroxybutanal reductase can be encoded by Geobacillus thermoglucosidasius adhl gene. In another embodiment, the microbial organism is genetically modified to express exogenous pyruvate dehydrogenase subunits (see Example XIV). For example, the exogenous pyruvate dehydrogenase can be NADH insensitive. The pyruvate dehydrogenase subunit can be encoded by the Klebsiella pneumonia lpdA gene. In a particular embodiment, the pyruvate dehydrogenase subunit genes of the microbial organism can be under the control of a pyruvate formate lyase promoter.

In still another embodiment, the microbial organism is genetically modified to disrupt a gene encoding an aerobic respiratory control regulatory system (see Example XV). For example, the disruption can be of the arcA gene. Such an organism can further comprise disruption of a gene encoding malate dehydrogenase. In a further embodiment, the microbial organism is genetically modified to express an exogenous NADH insensitive citrate synthase (see Example XV). For example, the NADH insensitive citrate synthase can be encoded by gltA, such as an R163L mutant of gltA. In still another embodiment, the microbial organism is genetically modified to express exogenous phosphoenolpyruvate carboxykinase (see Example XVI). For example, the phosphoenolpyruvate carboxykinase can be encoded by an Haemophilus influenza phosphoenolpyruvate carboxykinase gene. It is understood that strains exemplified herein for improved production of BDO can similarly be used, with appropriate modifications, to produce other desired products, for example, 4-hydroxybutyrate or other desired products disclosed herein.

It is understood that, in methods of the invention, any of the one or more exogenous nucleic acids can be introduced into a microbial organism to produce a non-naturally occurring microbial organism of the invention. The nucleic acids can be introduced so as to confer, for example, a 4-HB, BDO, THF or GBL biosynthetic pathway onto the microbial organism. Alternatively, encoding nucleic acids can be introduced to produce an intermediate microbial organism having the biosynthetic capability to catalyze some of the required reactions to confer 4-HB, BDO, THF or GBL biosynthetic capability. For example, a non-naturally occurring microbial organism having a 4-HB biosynthetic pathway can comprise at least two exogenous nucleic acids encoding desired enzymes, such as the combination of 4-hydroxybutanoate dehydrogenase and α-ketoglutarate decarboxylase; 4-hydroxybutanoate dehydrogenase and CoA-independent succinic semialdehyde dehydrogenase; 4-hydroxybutanoate dehydrogenase and CoA-dependent succinic semialdehyde dehydrogenase; CoA-dependent succinic semialdehyde dehydrogenase and succinyl-CoA synthetase; succinyl-CoA synthetase and glutamate decarboxylase, and the like. Thus, it is understood that any combination of two or more enzymes of a biosynthetic pathway can be included in a non-naturally occurring microbial organism of the invention. Similarly, it is understood that any combination of three or more enzymes of a biosynthetic pathway can be included in a non-naturally occurring microbial organism of the invention, for example, 4-hydroxybutanoate dehydrogenase, α-ketoglutarate decarboxylase and CoA-dependent succinic semialdehyde dehydrogenase; CoA-independent succinic semialdehyde dehydrogenase and succinyl-CoA synthetase; 4-hydroxybutanoate dehydrogenase, CoA-dependent succinic semialdehyde dehydrogenase and glutamate:succinic semialdehyde transaminase, and so forth, as desired, so long as the combination of enzymes of the desired biosynthetic pathway results in production of the corresponding desired product.

Similarly, for example, with respect to any one or more exogenous nucleic acids introduced to confer BDO production, a non-naturally occurring microbial organism having a BDO biosynthetic pathway can comprise at least two exogenous nucleic acids encoding desired enzymes, such as the combination of 4-hydroxybutanoate dehydrogenase and α-ketoglutarate decarboxylase; 4-hydroxybutanoate dehydrogenase and 4-hydroxybutyryl CoA:acetyl-CoA transferase; 4-hydroxybutanoate dehydrogenase and butyrate kinase; 4-hydroxybutanoate dehydrogenase and phosphotransbutyrylase; 4-hydroxybutyryl CoA:acetyl-CoA transferase and aldehyde dehydrogenase; 4-hydroxybutyryl CoA:acetyl-CoA transferase and alcohol dehydrogenase; 4-hydroxybutyryl CoA:acetyl-CoA transferase and an aldehyde/alcohol dehydrogenase, 4-aminobutyrate-CoA transferase and 4-aminobutyryl-CoA transaminase; 4-aminobutyrate kinase and 4-aminobutan-1-ol oxidoreductase (deaminating), and the like. Thus, it is understood that any combination of two or more enzymes of a biosynthetic pathway can be included in a non-naturally occurring microbial organism of the invention. Similarly, it is understood that any combination of three or more enzymes of a biosynthetic pathway can be included in a non-naturally occurring microbial organism of the invention, for example, 4-hydroxybutanoate dehydrogenase, α-ketoglutarate decarboxylase and 4-hydroxybutyryl CoA:acetyl-CoA transferase; 4-hydroxybutanoate dehydrogenase, butyrate kinase and phosphotransbutyrylase; 4-hydroxybutanoate dehydrogenase, 4-hydroxybutyryl CoA:acetyl-CoA transferase and aldehyde dehydrogenase; 4-hydroxybutyryl CoA:acetyl-CoA transferase, aldehyde dehydrogenase and alcohol dehydrogenase; butyrate kinase, phosphotransbutyrylase and an aldehyde/alcohol dehydrogenase; 4-aminobutyryl-CoA hydrolase, 4-aminobutyryl-CoA reductase and 4-amino butan-1-ol transaminase; 3-hydroxybutyryl-CoA dehydrogenase, 3-hydroxybutyryl-CoA dehydratase and 4-hydroxybutyryl-CoA dehydratase, and the like. Similarly, any combination of four, five or more enzymes of a biosynthetic pathway as disclosed herein can be included in a non-naturally occurring microbial organism of the invention, as desired, so long as the combination of enzymes of the desired biosynthetic pathway results in production of the corresponding desired product.

Any of the non-naturally occurring microbial organisms described herein can be cultured to produce and/or secrete the biosynthetic products of the invention. For example, the 4-HB producers can be cultured for the biosynthetic production of 4-HB. The 4-HB can be isolated or be treated as described below to generate GBL, THF and/or BDO. Similarly, the BDO producers can be cultured for the biosynthetic production of BDO. The BDO can be isolated or subjected to further treatments for the chemical synthesis of BDO family compounds, as disclosed herein.

The growth medium can include, for example, any carbohydrate source which can supply a source of carbon to the non-naturally occurring microorganism. Such sources include, for example, sugars such as glucose, sucrose, xylose, arabinose, galactose, mannose, fructose and starch. Other sources of carbohydrate include, for example, renewable feedstocks and biomass. Exemplary types of biomasses that can be used as feedstocks in the methods of the invention include cellulosic biomass, hemicellulosic biomass and lignin feedstocks or portions of feedstocks. Such biomass feedstocks contain, for example, carbohydrate substrates useful as carbon sources such as glucose, sucrose, xylose, arabinose, galactose, mannose, fructose and starch. Given the teachings and guidance provided herein, those skilled in the art will understand that renewable feedstocks and biomass other than those exemplified above also can be used for culturing the microbial organisms of the invention for the production of 4-HB or BDO and other compounds of the invention.

Accordingly, given the teachings and guidance provided herein, those skilled in the art will understand that a non-naturally occurring microbial organism can be produced that secretes the biosynthesized compounds of the invention when grown on a carbon source such as a carbohydrate. Such compounds include, for example, 4-HB, BDO and any of the intermediates metabolites in the 4-HB pathway, the BDO pathway and/or the combined 4-HB and BDO pathways. All that is required is to engineer in one or more of the enzyme activities shown in FIG. 1 to achieve biosynthesis of the desired compound or intermediate including, for example, inclusion of some or all of the 4-HB and/or BDO biosynthetic pathways. Accordingly, the invention provides a non-naturally occurring microbial organism that secretes 4-HB when grown on a carbohydrate, secretes BDO when grown on a carbohydrate and/or secretes any of the intermediate metabolites shown in FIG. 1 when grown on a carbohydrate. The BDO producing microbial organisms of the invention can initiate synthesis from, for example, succinate, succinyl-CoA, α-ketogluterate, succinic semialdehyde, 4-HB, 4-hydroxybutyrylphosphate, 4-hydroxybutyryl-CoA (4-HB-CoA) and/or 4-hydroxybutyraldehyde.

In some embodiments, culture conditions include anaerobic or substantially anaerobic growth or maintenance conditions. Exemplary anaerobic conditions have been described previously and are well known in the art. Exemplary anaerobic conditions for fermentation processes are described below in the Examples. Any of these conditions can be employed with the non-naturally occurring microbial organisms as well as other anaerobic conditions well known in the art. Under such anaerobic conditions, the 4-HB and BDO producers can synthesize monomeric 4-HB and BDO, respectively, at intracellular concentrations of 5-10 mM or more as well as all other concentrations exemplified previously.

A number of downstream compounds also can be generated for the 4-HB and BDO producing non-naturally occurring microbial organisms of the invention. With respect to the 4-HB producing microbial organisms of the invention, monomeric 4-HB and GBL exist in equilibrium in the culture medium. The conversion of 4-HB to GBL can be efficiently accomplished by, for example, culturing the microbial organisms in acid pH medium. A pH less than or equal to 7.5, in particular at or below pH 5.5, spontaneously converts 4-HB to GBL.

The resultant GBL can be separated from 4-HB and other components in the culture using a variety of methods well known in the art. Such separation methods include, for example, the extraction procedures exemplified in the Examples as well as methods which include continuous liquid-liquid extraction, pervaporation, membrane filtration, membrane separation, reverse osmosis, electrodialysis, distillation, crystallization, centrifugation, extractive filtration, ion exchange chromatography, size exclusion chromatography, adsorption chromatography, and ultrafiltration. All of the above methods are well known in the art. Separated GBL can be further purified by, for example, distillation.

Another down stream compound that can be produced from the 4-HB producing non-naturally occurring microbial organisms of the invention includes, for example, BDO. This compound can be synthesized by, for example, chemical hydrogenation of GBL. Chemical hydrogenation reactions are well known in the art. One exemplary procedure includes the chemical reduction of 4-HB and/or GBL or a mixture of these two components deriving from the culture using a heterogeneous or homogeneous hydrogenation catalyst together with hydrogen, or a hydride-based reducing agent used stoichiometrically or catalytically, to produce 1,4-butanediol.

Other procedures well known in the art are equally applicable for the above chemical reaction and include, for example, WO No. 82/03854 (Bradley, et al.), which describes the hydrogenolysis of gamma-butyrolactone in the vapor phase over a copper oxide and zinc oxide catalyst. British Pat. No. 1,230,276, which describes the hydrogenation of gamma-butyrolactone using a copper oxide-chromium oxide catalyst. The hydrogenation is carried out in the liquid phase. Batch reactions also are exemplified having high total reactor pressures. Reactant and product partial pressures in the reactors are well above the respective dew points. British Pat. No. 1,314,126, which describes the hydrogenation of gamma-butyrolactone in the liquid phase over a nickel-cobalt-thorium oxide catalyst. Batch reactions are exemplified as having high total pressures and component partial pressures well above respective component dew points. British Pat. No. 1,344,557, which describes the hydrogenation of gamma-butyrolactone in the liquid phase over a copper oxide-chromium oxide catalyst. A vapor phase or vapor-containing mixed phase is indicated as suitable in some instances. A continuous flow tubular reactor is exemplified using high total reactor pressures. British Pat. No. 1,512,751, which describes the hydrogenation of gamma-butyrolactone to 1,4-butanediol in the liquid phase over a copper oxide-chromium oxide catalyst. Batch reactions are exemplified with high total reactor pressures and, where determinable, reactant and product partial pressures well above the respective dew points. U.S. Pat. No. 4,301,077, which describes the hydrogenation to 1,4-butanediol of gamma-butyrolactone over a Ru—Ni—Co—Zn catalyst. The reaction can be conducted in the liquid or gas phase or in a mixed liquid-gas phase. Exemplified are continuous flow liquid phase reactions at high total reactor pressures and relatively low reactor productivities. U.S. Pat. No. 4,048,196, which describes the production of 1,4-butanediol by the liquid phase hydrogenation of gamma-butyrolactone over a copper oxide-zinc oxide catalyst. Further exemplified is a continuous flow tubular reactor operating at high total reactor pressures and high reactant and product partial pressures. And U.S. Pat. No. 4,652,685, which describes the hydrogenation of lactones to glycols.

A further downstream compound that can be produced form the 4-HB producing microbial organisms of the invention includes, for example, THF. This compound can be synthesized by, for example, chemical hydrogenation of GBL. One exemplary procedure well known in the art applicable for the conversion of GBL to THF includes, for example, chemical reduction of 4-HB and/or GBL or a mixture of these two components deriving from the culture using a heterogeneous or homogeneous hydrogenation catalyst together with hydrogen, or a hydride-based reducing agent used stoichiometrically or catalytically, to produce tetrahydrofuran. Other procedures well know in the art are equally applicable for the above chemical reaction and include, for example, U.S. Pat. No. 6,686,310, which describes high surface area sol-gel route prepared hydrogenation catalysts. Processes for the reduction of maleic acid to tetrahydrofuran (THF) and 1,4-butanediol (BDO) and for the reduction of gamma butyrolactone to tetrahydrofuran and 1,4-butanediol also are described.

The culture conditions can include, for example, liquid culture procedures as well as fermentation and other large scale culture procedures. As described further below in the Examples, particularly useful yields of the biosynthetic products of the invention can be obtained under anaerobic or substantially anaerobic culture conditions.

Suitable purification and/or assays to test for the production of 4-HB or BDO can be performed using well known methods. Suitable replicates such as triplicate cultures can be grown for each engineered strain to be tested. For example, product and byproduct formation in the engineered production host can be monitored. The final product and intermediates, and other organic compounds, can be analyzed by methods such as HPLC (High Performance Liquid Chromatography), GC-MS (Gas Chromatography-Mass Spectroscopy) and LC-MS (Liquid Chromatography-Mass Spectroscopy) or other suitable analytical methods using routine procedures well known in the art. The release of product in the fermentation broth can also be tested with the culture supernatant. Byproducts and residual glucose can be quantified by HPLC using, for example, a refractive index detector for glucose and alcohols, and a UV detector for organic acids (Lin et al., Biotechnol. Bioeng. 90:775-779 (2005)), or other suitable assay and detection methods well known in the art. The individual enzyme or protein activities from the exogenous DNA sequences can also be assayed using methods well known in the art.

The 4-HB or BDO product can be separated from other components in the culture using a variety of methods well known in the art. Such separation methods include, for example, extraction procedures as well as methods that include continuous liquid-liquid extraction, pervaporation, membrane filtration, membrane separation, reverse osmosis, electrodialysis, distillation, crystallization, centrifugation, extractive filtration, ion exchange chromatography, size exclusion chromatography, adsorption chromatography, and ultrafiltration. All of the above methods are well known in the art.

The invention further provides a method of manufacturing 4-HB. The method includes fermenting a non-naturally occurring microbial organism having a 4-hydroxybutanoic acid (4-HB) biosynthetic pathway comprising at least one exogenous nucleic acid encoding 4-hydroxybutanoate dehydrogenase, CoA-independent succinic semialdehyde dehydrogenase, succinyl-CoA synthetase, CoA-dependent succinic semialdehyde dehydrogenase, glutamate: succinic semialdehyde transaminase, α-ketoglutarate decarboxylase, or glutamate decarboxylase under substantially anaerobic conditions for a sufficient period of time to produce monomeric 4-hydroxybutanoic acid (4-HB), the process comprising fed-batch fermentation and batch separation; fed-batch fermentation and continuous separation, or continuous fermentation and continuous separation.

The culture and chemical hydrogenations described above also can be scaled up and grown continuously for manufacturing of 4-HB, GBL, BDO and/or THF. Exemplary growth procedures include, for example, fed-batch fermentation and batch separation; fed-batch fermentation and continuous separation, or continuous fermentation and continuous separation. All of these processes are well known in the art. Employing the 4-HB producers allows for simultaneous 4-HB biosynthesis and chemical conversion to GBL, BDO and/or THF by employing the above hydrogenation procedures simultaneous with continuous cultures methods such as fermentation. Other hydrogenation procedures also are well known in the art and can be equally applied to the methods of the invention.

Fermentation procedures are particularly useful for the biosynthetic production of commercial quantities of 4-HB and/or BDO. Generally, and as with non-continuous culture procedures, the continuous and/or near-continuous production of 4-HB or BDO will include culturing a non-naturally occurring 4-HB or BDO producing organism of the invention in sufficient neutrients and medium to sustain and/or nearly sustain growth in an exponential phase. Continuous culture under such conditions can be include, for example, 1 day, 2, 3, 4, 5, 6 or 7 days or more. Additionally, continuous culture can include 1 week, 2, 3, 4 or 5 or more weeks and up to several months. Alternatively, organisms of the invention can be cultured for hours, if suitable for a particular application. It is to be understood that the continuous and/or near-continuous culture conditions also can include all time intervals in between these exemplary periods. It is further understood that the time of culturing the microbial organism of the invention is for a sufficient period of time to produce a sufficient amount of product for a desired purpose.

Fermentation procedures are well known in the art. Briefly, fermentation for the biosynthetic production of 4-HB, BDO or other 4-HB derived products of the invention can be utilized in, for example, fed-batch fermentation and batch separation; fed-batch fermentation and continuous separation, or continuous fermentation and continuous separation. Examples of batch and continuous fermentation procedures well known in the art are exemplified further below in the Examples.

In addition, to the above fermentation procedures using the 4-HB or BDO producers of the invention for continuous production of substantial quantities of monomeric 4-HB and BDO, respectively, the 4-HB producers also can be, for example, simultaneously subjected to chemical synthesis procedures as described previously for the chemical conversion of monomeric 4-HB to, for example, GBL, BDO and/or THF. The BDO producers can similarly be, for example, simultaneously subjected to chemical synthesis procedures as described previously for the chemical conversion of BDO to, for example, THF, GBL, pyrrolidones and/or other BDO family compounds. In addition, the products of the 4-HB and BDO producers can be separated from the fermentation culture and sequentially subjected to chemical conversion, as disclosed herein.

Briefly, hydrogenation of GBL in the fermentation broth can be performed as described by Frost et al., Biotechnology Progress 18: 201-211 (2002). Another procedure for hydrogenation during fermentation include, for example, the methods described in, for example, U.S. Pat. No. 5,478,952. This method is further exemplified in the Examples below.

Therefore, the invention additionally provides a method of manufacturing α-butyrolactone (GBL), tetrahydrofuran (THF) or 1,4-butanediol (BDO). The method includes fermenting a non-naturally occurring microbial organism having 4-hydroxybutanoic acid (4-HB) and/or 1,4-butanediol (BDO) biosynthetic pathways, the pathways comprise at least one exogenous nucleic acid encoding 4-hydroxybutanoate dehydrogenase, CoA-independent succinic semialdehyde dehydrogenase, succinyl-CoA synthetase, CoA-dependent succinic semialdehyde dehydrogenase, 4-hydroxybutyrate:CoA transferase, glutamate: succinic semialdehyde transaminase, α-ketoglutarate decarboxylase, glutamate decarboxylase, 4-hydroxybutanoate kinase, phosphotransbutyrylase, CoA-independent 1,4-butanediol semialdehyde dehydrogenase, CoA-dependent 1,4-butanediol semialdehyde dehydrogenase, CoA-independent 1,4-butanediol alcohol dehydrogenase or CoA-dependent 1,4-butanediol alcohol dehydrogenase, under substantially anaerobic conditions for a sufficient period of time to produce 1,4-butanediol (BDO), GBL or THF, the fermenting comprising fed-batch fermentation and batch separation; fed-batch fermentation and continuous separation, or continuous fermentation and continuous separation.

In addition to the biosynthesis of 4-HB, BDO and other products of the invention as described herein, the non-naturally occurring microbial organisms and methods of the invention also can be utilized in various combinations with each other and with other microbial organisms and methods well known in the art to achieve product biosynthesis by other routes. For example, one alternative to produce BDO other than use of the 4-HB producers and chemical steps or other than use of the BDO producer directly is through addition of another microbial organism capable of converting 4-HB or a 4-HB product exemplified herein to BDO.

One such procedure includes, for example, the fermentation of a 4-HB producing microbial organism of the invention to produce 4-HB, as described above and below. The 4-HB can then be used as a substrate for a second microbial organism that converts 4-HB to, for example, BDO, GBL and/or THF. The 4-HB can be added directly to another culture of the second organism or the original culture of 4-HB producers can be depleted of these microbial organisms by, for example, cell separation, and then subsequent addition of the second organism to the fermentation broth can utilized to produce the final product without intermediate purification steps. One exemplary second organism having the capacity to biochemically utilize 4-HB as a substrate for conversion to BDO, for example, is Clostridium acetobutylicum (see, for example, Jewell et al., Current Microbiology, 13:215-19 (1986)).

In other embodiments, the non-naturally occurring microbial organisms and methods of the invention can be assembled in a wide variety of subpathways to achieve biosynthesis of, for example, 4-HB and/or BDO as described. In these embodiments, biosynthetic pathways for a desired product of the invention can be segregated into different microbial organisms and the different microbial organisms can be co-cultured to produce the final product. In such a biosynthetic scheme, the product of one microbial organism is the substrate for a second microbial organism until the final product is synthesized. For example, the biosynthesis of BDO can be accomplished as described previously by constructing a microbial organism that contains biosynthetic pathways for conversion of one pathway intermediate to another pathway intermediate or the product, for example, a substrate such as endogenous succinate through 4-HB to the final product BDO. Alternatively, BDO also can be biosynthetically produced from microbial organisms through co-culture or co-fermentation using two organisms in the same vessel. A first microbial organism being a 4-HB producer with genes to produce 4-HB from succinic acid, and a second microbial organism being a BDO producer with genes to convert 4-HB to BDO.

Given the teachings and guidance provided herein, those skilled in the art will understand that a wide variety of combinations and permutations exist for the non-naturally occurring microbial organisms and methods of the invention together with other microbial organisms, with the co-culture of other non-naturally occurring microbial organisms having subpathways and with combinations of other chemical and/or biochemical procedures well known in the art to produce 4-HB, BDO, GBL and THF products of the invention.

To generate better producers, metabolic modeling can be utilized to optimize growth conditions. Modeling can also be used to design gene knockouts that additionally optimize utilization of the pathway (see, for example, U.S. patent publications US 2002/0012939, US 2003/0224363, US 2004/0029149, US 2004/0072723, US 2003/0059792, US 2002/0168654 and US 2004/0009466, and U.S. Pat. No. 7,127,379). Modeling analysis allows reliable predictions of the effects on cell growth of shifting the metabolism towards more efficient production of BDO.

One computational method for identifying and designing metabolic alterations favoring biosynthesis of a desired product is the OptKnock computational framework (Burgard et al., Biotechnol. Bioeng. 84:647-657 (2003)). OptKnock is a metabolic modeling and simulation program that suggests gene deletion or disruption strategies that result in genetically stable microorganisms which overproduce the target product. Specifically, the framework examines the complete metabolic and/or biochemical network of a microorganism in order to suggest genetic manipulations that force the desired biochemical to become an obligatory byproduct of cell growth. By coupling biochemical production with cell growth through strategically placed gene deletions or other functional gene disruption, the growth selection pressures imposed on the engineered strains after long periods of time in a bioreactor lead to improvements in performance as a result of the compulsory growth-coupled biochemical production. Lastly, when gene deletions are constructed there is a negligible possibility of the designed strains reverting to their wild-type states because the genes selected by OptKnock are to be completely removed from the genome. Therefore, this computational methodology can be used to either identify alternative pathways that lead to biosynthesis of a desired product or used in connection with the non-naturally occurring microbial organisms for further optimization of biosynthesis of a desired product.

Briefly, OptKnock is a term used herein to refer to a computational method and system for modeling cellular metabolism. The OptKnock program relates to a framework of models and methods that incorporate particular constraints into flux balance analysis (FBA) models. These constraints include, for example, qualitative kinetic information, qualitative regulatory information, and/or DNA microarray experimental data. OptKnock also computes solutions to various metabolic problems by, for example, tightening the flux boundaries derived through flux balance models and subsequently probing the performance limits of metabolic networks in the presence of gene additions or deletions. OptKnock computational framework allows the construction of model formulations that enable an effective query of the performance limits of metabolic networks and provides methods for solving the resulting mixed-integer linear programming problems. The metabolic modeling and simulation methods referred to herein as OptKnock are described in, for example, U.S. publication 2002/0168654, filed Jan. 10, 2002, in International Patent No. PCT/US02/00660, filed Jan. 10, 2002, and U.S. patent application Ser. No. 11/891,602, filed Aug. 10, 2007.

Another computational method for identifying and designing metabolic alterations favoring biosynthetic production of a product is a metabolic modeling and simulation system termed SimPheny®. This computational method and system is described in, for example, U.S. publication 2003/0233218, filed Jun. 14, 2002, and in International Patent Application No. PCT/US03/18838, filed Jun. 13, 2003. SimPheny® is a computational system that can be used to produce a network model in silico and to simulate the flux of mass, energy or charge through the chemical reactions of a biological system to define a solution space that contains any and all possible functionalities of the chemical reactions in the system, thereby determining a range of allowed activities for the biological system. This approach is referred to as constraints-based modeling because the solution space is defined by constraints such as the known stoichiometry of the included reactions as well as reaction thermodynamic and capacity constraints associated with maximum fluxes through reactions. The space defined by these constraints can be interrogated to determine the phenotypic capabilities and behavior of the biological system or of its biochemical components.

These computational approaches are consistent with biological realities because biological systems are flexible and can reach the same result in many different ways. Biological systems are designed through evolutionary mechanisms that have been restricted by fundamental constraints that all living systems must face. Therefore, constraints-based modeling strategy embraces these general realities. Further, the ability to continuously impose further restrictions on a network model via the tightening of constraints results in a reduction in the size of the solution space, thereby enhancing the precision with which physiological performance or phenotype can be predicted.

Given the teachings and guidance provided herein, those skilled in the art will be able to apply various computational frameworks for metabolic modeling and simulation to design and implement biosynthesis of a desired compound in host microbial organisms. Such metabolic modeling and simulation methods include, for example, the computational systems exemplified above as SimPheny® and OptKnock. For illustration of the invention, some methods are described herein with reference to the OptKnock computation framework for modeling and simulation. Those skilled in the art will know how to apply the identification, design and implementation of the metabolic alterations using OptKnock to any of such other metabolic modeling and simulation computational frameworks and methods well known in the art.

The methods described above will provide one set of metabolic reactions to disrupt. Elimination of each reaction within the set or metabolic modification can result in a desired product as an obligatory product during the growth phase of the organism. Because the reactions are known, a solution to the bilevel OptKnock problem also will provide the associated gene or genes encoding one or more enzymes that catalyze each reaction within the set of reactions. Identification of a set of reactions and their corresponding genes encoding the enzymes participating in each reaction is generally an automated process, accomplished through correlation of the reactions with a reaction database having a relationship between enzymes and encoding genes.

Once identified, the set of reactions that are to be disrupted in order to achieve production of a desired product are implemented in the target cell or organism by functional disruption of at least one gene encoding each metabolic reaction within the set. One particularly useful means to achieve functional disruption of the reaction set is by deletion of each encoding gene. However, in some instances, it can be beneficial to disrupt the reaction by other genetic aberrations including, for example, mutation, deletion of regulatory regions such as promoters or cis binding sites for regulatory factors, or by truncation of the coding sequence at any of a number of locations. These latter aberrations, resulting in less than total deletion of the gene set can be useful, for example, when rapid assessments of the coupling of a product are desired or when genetic reversion is less likely to occur.

To identify additional productive solutions to the above described bilevel OptKnock problem which lead to further sets of reactions to disrupt or metabolic modifications that can result in the biosynthesis, including growth-coupled biosynthesis of a desired product, an optimization method, termed integer cuts, can be implemented. This method proceeds by iteratively solving the OptKnock problem exemplified above with the incorporation of an additional constraint referred to as an integer cut at each iteration. Integer cut constraints effectively prevent the solution procedure from choosing the exact same set of reactions identified in any previous iteration that obligatorily couples product biosynthesis to growth. For example, if a previously identified growth-coupled metabolic modification specifies reactions 1, 2, and 3 for disruption, then the following constraint prevents the same reactions from being simultaneously considered in subsequent solutions. The integer cut method is well known in the art and can be found described in, for example, Burgard et al., Biotechnol. Prog. 17:791-797 (2001). As with all methods described herein with reference to their use in combination with the OptKnock computational framework for metabolic modeling and simulation, the integer cut method of reducing redundancy in iterative computational analysis also can be applied with other computational frameworks well known in the art including, for example, SimPheny®.

The methods exemplified herein allow the construction of cells and organisms that biosynthetically produce a desired product, including the obligatory coupling of production of a target biochemical product to growth of the cell or organism engineered to harbor the identified genetic alterations. Therefore, the computational methods described herein allow the identification and implementation of metabolic modifications that are identified by an in silico method selected from OptKnock or SimPheny®. The set of metabolic modifications can include, for example, addition of one or more biosynthetic pathway enzymes and/or functional disruption of one or more metabolic reactions including, for example, disruption by gene deletion.

As discussed above, the OptKnock methodology was developed on the premise that mutant microbial networks can be evolved towards their computationally predicted maximum-growth phenotypes when subjected to long periods of growth selection. In other words, the approach leverages an organism's ability to self-optimize under selective pressures. The OptKnock framework allows for the exhaustive enumeration of gene deletion combinations that force a coupling between biochemical production and cell growth based on network stoichiometry. The identification of optimal gene/reaction knockouts requires the solution of a bilevel optimization problem that chooses the set of active reactions such that an optimal growth solution for the resulting network overproduces the biochemical of interest (Burgard et al., Biotechnol. Bioeng. 84:647-657 (2003)).

An in silico stoichiometric model of E. coli metabolism can be employed to identify essential genes for metabolic pathways as exemplified previously and described in, for example, U.S. patent publications US 2002/0012939, US 2003/0224363, US 2004/0029149, US 2004/0072723, US 2003/0059792, US 2002/0168654 and US 2004/0009466, and in U.S. Pat. No. 7,127,379. As disclosed herein, the OptKnock mathematical framework can be applied to pinpoint gene deletions leading to the growth-coupled production of a desired product. Further, the solution of the bilevel OptKnock problem provides only one set of deletions. To enumerate all meaningful solutions, that is, all sets of knockouts leading to growth-coupled production formation, an optimization technique, termed integer cuts, can be implemented. This entails iteratively solving the OptKnock problem with the incorporation of an additional constraint referred to as an integer cut at each iteration, as discussed above.

The methods exemplified above and further illustrated in the Examples below enable the construction of cells and organisms that biosynthetically produce, including obligatory couple production of a target biochemical product to growth of the cell or organism engineered to harbor the identified genetic alterations. In this regard, metabolic alterations have been identified that result in the biosynthesis of 4-HB and 1,4-butanediol. Microorganism strains constructed with the identified metabolic alterations produce elevated levels of 4-HB or BDO compared to unmodified microbial organisms. These strains can be beneficially used for the commercial production of 4-HB, BDO, THF and GBL, for example, in continuous fermentation process without being subjected to the negative selective pressures.

Therefore, the computational methods described herein enable the identification and implementation of metabolic modifications that are identified by an in silico method selected from OptKnock or SimPheny®. The set of metabolic modifications can include, for example, addition of one or more biosynthetic pathway enzymes and/.or functional disruption of one or more metabolic reactions including, for example, disruption by gene deletion.

It is understood that modifications which do not substantially affect the activity of the various embodiments of this invention are also included within the definition of the invention provided herein. Accordingly, the following examples are intended to illustrate but not limit the present invention.

Any of the non-naturally occurring microbial organisms described herein can be cultured to produce and/or secrete the biosynthetic products of the invention. For example, the BDO producers can be cultured for the biosynthetic production of BDO.

For the production of BDO, the recombinant strains are cultured in a medium with carbon source and other essential nutrients. It is highly desirable to maintain anaerobic conditions in the fermenter to reduce the cost of the overall process. Such conditions can be obtained, for example, by first sparging the medium with nitrogen and then sealing the flasks with a septum and crimp-cap. For strains where growth is not observed anaerobically, microaerobic conditions can be applied by perforating the septum with a small hole for limited aeration. Exemplary anaerobic conditions have been described previously and are well-known in the art. Exemplary aerobic and anaerobic conditions are described, for example, in U.S. publication 2009/0047719, filed Aug. 10, 2007. Fermentations can be performed in a batch, fed-batch or continuous manner, as disclosed herein.

If desired, the pH of the medium can be maintained at a desired pH, in particular neutral pH, such as a pH of around 7 by addition of a base, such as NaOH or other bases, or acid, as needed to maintain the culture medium at a desirable pH. The growth rate can be determined by measuring optical density using a spectrophotometer (600 nm), and the glucose uptake rate by monitoring carbon source depletion over time.

In addition to renewable feedstocks such as those exemplified above, the BDO producing microbial organisms of the invention also can be modified for growth on syngas as its source of carbon. In this specific embodiment, one or more proteins or enzymes are expressed in the BDO producing organisms to provide a metabolic pathway for utilization of syngas or other gaseous carbon source.

Synthesis gas, also known as syngas or producer gas, is the major product of gasification of coal and of carbonaceous materials such as biomass materials, including agricultural crops and residues. Syngas is a mixture primarily of H₂and CO and can be obtained from the gasification of any organic feedstock, including but not limited to coal, coal oil, natural gas, biomass, and waste organic matter. Gasification is generally carried out under a high fuel to oxygen ratio. Although largely H₂and CO, syngas can also include CO₂and other gases in smaller quantities. Thus, synthesis gas provides a cost effective source of gaseous carbon such as CO and, additionally, CO₂.

The Wood-Ljungdahl pathway catalyzes the conversion of CO and H₂to acetyl-CoA and other products such as acetate. Organisms capable of utilizing CO and syngas also generally have the capability of utilizing CO₂and CO₂/H₂mixtures through the same basic set of enzymes and transformations encompassed by the Wood-Ljungdahl pathway. H₂-dependent conversion of CO₂to acetate by microorganisms was recognized long before it was revealed that CO also could be used by the same organisms and that the same pathways were involved. Many acetogens have been shown to grow in the presence of CO₂and produce compounds such as acetate as long as hydrogen is present to supply the necessary reducing equivalents (see for example, Drake, Acetogenesis, pp. 3-60 Chapman and Hall, New York, (1994)). This can be summarized by the following equation:

2CO₂+4H₂+nADP+nPi→CH₃COOH+2H₂O+nATP

Hence, non-naturally occurring microorganisms possessing the Wood-Ljungdahl pathway can utilize CO₂and H₂mixtures as well for the production of acetyl-CoA and other desired products.

The Wood-Ljungdahl pathway is well known in the art and consists of 12 reactions which can be separated into two branches: (1) methyl branch and (2) carbonyl branch. The methyl branch converts syngas to methyl-tetrahydrofolate (methyl-THF) whereas the carbonyl branch converts methyl-THF to acetyl-CoA. The reactions in the methyl branch are catalyzed in order by the following enzymes or proteins: ferredoxin oxidoreductase, formate dehydrogenase, formyltetrahydrofolate synthetase, methenyltetrahydrofolate cyclodehydratase, methylenetetrahydrofolate dehydrogenase and methylenetetrahydrofolate reductase. The reactions in the carbonyl branch are catalyzed in order by the following enzymes or proteins: methyltetrahydrofolate:corrinoid protein methyltransferase (for example, AcsE), corrinoid iron-sulfur protein, nickel-protein assembly protein (for example, AcsF), ferredoxin, acetyl-CoA synthase, carbon monoxide dehydrogenase and nickel-protein assembly protein (for example, CooC). Following the teachings and guidance provided herein for introducing a sufficient number of encoding nucleic acids to generate a BDO pathway, those skilled in the art will understand that the same engineering design also can be performed with respect to introducing at least the nucleic acids encoding the Wood-Ljungdahl enzymes or proteins absent in the host organism. Therefore, introduction of one or more encoding nucleic acids into the microbial organisms of the invention such that the modified organism contains the complete Wood-Ljungdahl pathway will confer syngas utilization ability.

Accordingly, given the teachings and guidance provided herein, those skilled in the art will understand that a non-naturally occurring microbial organism can be produced that secretes the biosynthesized compounds of the invention when grown on a carbon source such as a carbohydrate. Such compounds include, for example, BDO and any of the intermediate metabolites in the BDO pathway. All that is required is to engineer in one or more of the required enzyme or protein activities to achieve biosynthesis of the desired compound or intermediate including, for example, inclusion of some or all of the BDO biosynthetic pathways. Accordingly, the invention provides a non-naturally occurring microbial organism that produces and/or secretes BDO when grown on a carbohydrate or other carbon source and produces and/or secretes any of the intermediate metabolites shown in the BDO pathway when grown on a carbohydrate or other carbon source. The BDO producing microbial organisms of the invention can initiate synthesis from an intermediate in a BDO pathway, as disclosed herein.

To generate better producers, metabolic modeling can be utilized to optimize growth conditions. Modeling can also be used to design gene knockouts that additionally optimize utilization of the pathway (see, for example, U.S. patent publications US 2002/0012939, US 2003/0224363, US 2004/0029149, US 2004/0072723, US 2003/0059792, US 2002/0168654 and US 2004/0009466, and U.S. Pat. No. 7,127,379). Modeling analysis allows reliable predictions of the effects on cell growth of shifting the metabolism towards more efficient production of BDO.

One computational method for identifying and designing metabolic alterations favoring biosynthesis of a desired product is the OptKnock computational framework (Burgard et al., Biotechnol. Bioeng. 84:647-657 (2003)). OptKnock is a metabolic modeling and simulation program that suggests gene deletion strategies that result in genetically stable microorganisms which overproduce the target product. Specifically, the framework examines the complete metabolic and/or biochemical network of a microorganism in order to suggest genetic manipulations that force the desired biochemical to become an obligatory byproduct of cell growth. By coupling biochemical production with cell growth through strategically placed gene deletions or other functional gene disruption, the growth selection pressures imposed on the engineered strains after long periods of time in a bioreactor lead to improvements in performance as a result of the compulsory growth-coupled biochemical production. Lastly, when gene deletions are constructed there is a negligible possibility of the designed strains reverting to their wild-type states because the genes selected by OptKnock are to be completely removed from the genome. Therefore, this computational methodology can be used to either identify alternative pathways that lead to biosynthesis of a desired product or used in connection with the non-naturally occurring microbial organisms for further optimization of biosynthesis of a desired product.

Briefly, OptKnock is a term used herein to refer to a computational method and system for modeling cellular metabolism. The OptKnock program relates to a framework of models and methods that incorporate particular constraints into flux balance analysis (FBA) models. These constraints include, for example, qualitative kinetic information, qualitative regulatory information, and/or DNA microarray experimental data. OptKnock also computes solutions to various metabolic problems by, for example, tightening the flux boundaries derived through flux balance models and subsequently probing the performance limits of metabolic networks in the presence of gene additions or deletions. OptKnock computational framework allows the construction of model formulations that enable an effective query of the performance limits of metabolic networks and provides methods for solving the resulting mixed-integer linear programming problems. The metabolic modeling and simulation methods referred to herein as OptKnock are described in, for example, U.S. publication 2002/0168654, filed Jan. 10, 2002, in International Patent No. PCT/US02/00660, filed Jan. 10, 2002, and U.S. patent application Ser. No. 11/891,602, filed Aug. 10, 2007.

Another computational method for identifying and designing metabolic alterations favoring biosynthetic production of a product is a metabolic modeling and simulation system termed SimPheny®. This computational method and system is described in, for example, U.S. publication 2003/0233218, filed Jun. 14, 2002, and in International Patent Application No. PCT/US03/18838, filed Jun. 13, 2003. SimPheny® is a computational system that can be used to produce a network model in silico and to simulate the flux of mass, energy or charge through the chemical reactions of a biological system to define a solution space that contains any and all possible functionalities of the chemical reactions in the system, thereby determining a range of allowed activities for the biological system. This approach is referred to as constraints-based modeling because the solution space is defined by constraints such as the known stoichiometry of the included reactions as well as reaction thermodynamic and capacity constraints associated with maximum fluxes through reactions. The space defined by these constraints can be interrogated to determine the phenotypic capabilities and behavior of the biological system or of its biochemical components.

These computational approaches are consistent with biological realities because biological systems are flexible and can reach the same result in many different ways. Biological systems are designed through evolutionary mechanisms that have been restricted by fundamental constraints that all living systems must face. Therefore, constraints-based modeling strategy embraces these general realities. Further, the ability to continuously impose further restrictions on a network model via the tightening of constraints results in a reduction in the size of the solution space, thereby enhancing the precision with which physiological performance or phenotype can be predicted.

Given the teachings and guidance provided herein, those skilled in the art will be able to apply various computational frameworks for metabolic modeling and simulation to design and implement biosynthesis of a desired compound in host microbial organisms. Such metabolic modeling and simulation methods include, for example, the computational systems exemplified above as SimPheny® and OptKnock. For illustration of the invention, some methods are described herein with reference to the OptKnock computation framework for modeling and simulation. Those skilled in the art will know how to apply the identification, design and implementation of the metabolic alterations using OptKnock to any of such other metabolic modeling and simulation computational frameworks and methods well known in the art.

The methods described above will provide one set of metabolic reactions to disrupt. Elimination of each reaction within the set or metabolic modification can result in a desired product as an obligatory product during the growth phase of the organism. Because the reactions are known, a solution to the bilevel OptKnock problem also will provide the associated gene or genes encoding one or more enzymes that catalyze each reaction within the set of reactions. Identification of a set of reactions and their corresponding genes encoding the enzymes participating in each reaction is generally an automated process, accomplished through correlation of the reactions with a reaction database having a relationship between enzymes and encoding genes.

Once identified, the set of reactions that are to be disrupted in order to achieve production of a desired product are implemented in the target cell or organism by functional disruption of at least one gene encoding each metabolic reaction within the set. One particularly useful means to achieve functional disruption of the reaction set is by deletion of each encoding gene. However, in some instances, it can be beneficial to disrupt the reaction by other genetic aberrations including, for example, mutation, deletion of regulatory regions such as promoters or cis binding sites for regulatory factors, or by truncation of the coding sequence at any of a number of locations. These latter aberrations, resulting in less than total deletion of the gene set can be useful, for example, when rapid assessments of the coupling of a product are desired or when genetic reversion is less likely to occur.

To identify additional productive solutions to the above described bilevel OptKnock problem which lead to further sets of reactions to disrupt or metabolic modifications that can result in the biosynthesis, including growth-coupled biosynthesis of a desired product, an optimization method, termed integer cuts, can be implemented. This method proceeds by iteratively solving the OptKnock problem exemplified above with the incorporation of an additional constraint referred to as an integer cut at each iteration. Integer cut constraints effectively prevent the solution procedure from choosing the exact same set of reactions identified in any previous iteration that obligatorily couples product biosynthesis to growth. For example, if a previously identified growth-coupled metabolic modification specifies reactions 1, 2, and 3 for disruption, then the following constraint prevents the same reactions from being simultaneously considered in subsequent solutions. The integer cut method is well known in the art and can be found described in, for example, Burgard et al., Biotechnol. Prog. 17:791-797 (2001). As with all methods described herein with reference to their use in combination with the OptKnock computational framework for metabolic modeling and simulation, the integer cut method of reducing redundancy in iterative computational analysis also can be applied with other computational frameworks well known in the art including, for example, SimPheny®.

The methods exemplified herein allow the construction of cells and organisms that biosynthetically produce a desired product, including the obligatory coupling of production of a target biochemical product to growth of the cell or organism engineered to harbor the identified genetic alterations. Therefore, the computational methods described herein allow the identification and implementation of metabolic modifications that are identified by an in silico method selected from OptKnock or SimPheny®. The set of metabolic modifications can include, for example, addition of one or more biosynthetic pathway enzymes and/or functional disruption of one or more metabolic reactions including, for example, disruption by gene deletion.

As discussed above, the OptKnock methodology was developed on the premise that mutant microbial networks can be evolved towards their computationally predicted maximum-growth phenotypes when subjected to long periods of growth selection. In other words, the approach leverages an organism's ability to self-optimize under selective pressures. The OptKnock framework allows for the exhaustive enumeration of gene deletion combinations that force a coupling between biochemical production and cell growth based on network stoichiometry. The identification of optimal gene/reaction knockouts requires the solution of a bilevel optimization problem that chooses the set of active reactions such that an optimal growth solution for the resulting network overproduces the biochemical of interest (Burgard et al., Biotechnol. Bioeng. 84:647-657 (2003)).

An in silico stoichiometric model of E. coli metabolism can be employed to identify essential genes for metabolic pathways as exemplified previously and described in, for example, U.S. patent publications US 2002/0012939, US 2003/0224363, US 2004/0029149, US 2004/0072723, US 2003/0059792, US 2002/0168654 and US 2004/0009466, and in U.S. Pat. No. 7,127,379. As disclosed herein, the OptKnock mathematical framework can be applied to pinpoint gene deletions leading to the growth-coupled production of a desired product. Further, the solution of the bilevel OptKnock problem provides only one set of deletions. To enumerate all meaningful solutions, that is, all sets of knockouts leading to growth-coupled production formation, an optimization technique, termed integer cuts, can be implemented. This entails iteratively solving the OptKnock problem with the incorporation of an additional constraint referred to as an integer cut at each iteration, as discussed above.

It is understood that modifications which do not substantially affect the activity of the various embodiments of this invention are also provided within the definition of the invention provided herein. Accordingly, the following examples are intended to illustrate but not limit the present invention.

This example describes exemplary biochemical pathways for 4-HB production.

Previous reports of 4-HB synthesis in microbes have focused on this compound as an intermediate in production of the biodegradable plastic poly-hydroxyalkanoate (PHA) (U.S. Pat. No. 6,117,658). The use of 4-HB/3-HB copolymers over poly-3-hydroxybutyrate polymer (PHB) can result in plastic that is less brittle (Saito and Doi, Intl. J. Biol. Macromol. 16:99-104 (1994)). The production of monomeric 4-HB described herein is a fundamentally distinct process for several reasons: (1) the product is secreted, as opposed to PHA which is produced intracellularly and remains in the cell; (2) for organisms that produce hydroxybutanoate polymers, free 4-HB is not produced, but rather the Coenzyme A derivative is used by the polyhydroxyalkanoate synthase; (3) in the case of the polymer, formation of the granular product changes thermodynamics; and (4) extracellular pH is not an issue for production of the polymer, whereas it will affect whether 4-HB is present in the free acid or conjugate base state, and also the equilibrium between 4-HB and GBL.

4-HB can be produced in two enzymatic reduction steps from succinate, a central metabolite of the TCA cycle, with succinic semialdehyde as the intermediate (FIG. 1). The first of these enzymes, succinic semialdehyde dehydrogenase, is native to many organisms including E. coli, in which both NADH- and NADPH-dependent enzymes have been found (Donnelly and Cooper, Eur. J. Biochem. 113:555-561 (1981); Donnelly and Cooper, J. Bacteriol. 145:1425-1427 (1981); Marek and Henson, J. Bacteriol. 170:991-994 (1988)). There is also evidence supporting succinic semialdehyde dehydrogenase activity in S. cerevisiae (Ramos et al., Eur. J. Biochem. 149:401-404 (1985)), and a putative gene has been identified by sequence homology. However, most reports indicate that this enzyme proceeds in the direction of succinate synthesis, as shown in FIG. 1 (Donnelly and Cooper, supra; Lutke-Eversloh and Steinbuchel, FEMS Microbiol. Lett. 181:63-71 (1999)), participating in the degradation pathway of 4-HB and gamma-aminobutyrate. Succinic semialdehyde also is natively produced by certain microbial organisms such as E. coli through the TCA cycle intermediate α-ketogluterate via the action of two enzymes: glutamate:succinic semialdehyde transaminase and glutamate decarboxylase. An alternative pathway, used by the obligate anaerobe Clostridium kluyveri to degrade succinate, activates succinate to succinyl-CoA, then converts succinyl-CoA to succinic semialdehyde using an alternative succinic semialdehyde dehydrogenase which is known to function in this direction (Sohling and Gottschalk, Eur. J. Biochem. 212:121-127 (1993)). However, this route has the energetic cost of ATP required to convert succinate to succinyl-CoA.

The second enzyme of the pathway, 4-hydroxybutanoate dehydrogenase, is not native to E. coli or yeast but is found in various bacteria such as C. kluyveri and Ralstonia eutropha (Lutke-Eversloh and Steinbuchel, supra; Sohling and Gottschalk, J. Bacteriol. 178:871-880 (1996); Valentin et al., Eur. J. Biochem. 227:43-60 (1995); Wolff and Kenealy, Protein Expr. Purif. 6:206-212 (1995)). These enzymes are known to be NADH-dependent, though NADPH-dependent forms also exist. An additional pathway to 4-HB from alpha-ketoglutarate was demonstrated in E. coli resulting in the accumulation of poly(4-hydroxybutyric acid) (Song et al., Wei Sheng Wu Xue. Bao. 45:382-386 (2005)). The recombinant strain required the overexpression of three heterologous genes, PHA synthase (R. eutropha), 4-hydroxybutyrate dehydrogenase (R. eutropha) and 4-hydroxybutyrate:CoA transferase (C. kluyveri), along with two native E. coli genes: glutamate:succinic semialdehyde transaminase and glutamate decarboxylase. Steps 4 and 5 in FIG. 1 can alternatively be carried out by an alpha-ketoglutarate decarboxylase such as the one identified in Euglena gracilis (Shigeoka et al., Biochem. J. 282(Pt2):319-323 (1992); Shigeoka and Nakano, Arch. Biochem. Biophys. 288:22-28 (1991); Shigeoka and Nakano, Biochem J. 292(Pt 2):463-467 (1993)). However, this enzyme has not previously been applied to impact the production of 4-HB or related polymers in any organism.

The microbial production capabilities of 4-hydroxybutyrate were explored in two microbes, Escherichia coli and Saccharomyces cerevisiae, using in silico metabolic models of each organism. Potential pathways to 4-HB proceed via a succinate, succinyl-CoA, or alpha-ketoglutarate intermediate as shown in FIG. 1.

A first step in the 4-HB production pathway from succinate involves the conversion of succinate to succinic semialdehyde via an NADH- or NADPH-dependant succinic semialdehyde dehydrogenase. In E. coli, gabD is an NADP-dependant succinic semialdehyde dehydrogenase and is part of a gene cluster involved in 4-aminobutyrate uptake and degradation (Niegemann et al., Arch. Microbiol. 160:454-460 (1993); Schneider et al., J. Bacteriol. 184:6976-6986 (2002)). sad is believed to encode the enzyme for NAD-dependant succinic semialdehyde dehydrogenase activity (Marek and Henson, supra). S. cerevisiae contains only the NADPH-dependant succinic semialdehyde dehydrogenase, putatively assigned to UGA2, which localizes to the cytosol (Huh et al., Nature 425:686-691 (2003)). The maximum yield calculations assuming the succinate pathway to 4-HB in both E. coli and S. cerevisiae require only the assumption that a non-native 4-HB dehydrogenase has been added to their metabolic networks.

The pathway from succinyl-CoA to 4-hydroxybutyrate was described in U.S. Pat. No. 6,117,658 as part of a process for making polyhydroxyalkanoates comprising 4-hydroxybutyrate monomer units. Clostridium kluyveri is one example organism known to possess CoA-dependant succinic semialdehyde dehydrogenase activity (Sohling and Gottschalk, supra; Sohling and Gottschalk, supra). In this study, it is assumed that this enzyme, from C. kluyveri or another organism, is expressed in E. coli or S. cerevisiae along with a non-native or heterologous 4-HB dehydrogenase to complete the pathway from succinyl-CoA to 4-HB. The pathway from alpha-ketoglutarate to 4-HB was demonstrated in E. coli resulting in the accumulation of poly(4-hydroxybutyric acid) to 30% of dry cell weight (Song et al., supra). As E. coli and S. cerevisiae natively or endogenously possess both glutamate:succinic semialdehyde transaminase and glutamate decarboxylase (Coleman et al., J. Biol. Chem. 276:244-250 (2001)), the pathway from AKG to 4-HB can be completed in both organisms by assuming only that a non-native 4-HB dehydrogenase is present.

This example illustrates the construction and biosynthetic production of 4-HB and BDO from microbial organisms. Pathways for 4-HB and BDO are disclosed herein.

There are several alternative enzymes that can be utilized in the pathway described above. The native or endogenous enzyme for conversion of succinate to succinyl-CoA (Step 1 in FIG. 1) can be replaced by a CoA transferase such as that encoded by the cat1 gene C. kluyveri (Sohling and Gottschalk, Eur. J Biochem. 212:121-127 (1993)), which functions in a similar manner to Step 9. However, the production of acetate by this enzyme may not be optimal, as it might be secreted rather than being converted back to acetyl-CoA. In this respect, it also can be beneficial to eliminate acetate formation in Step 9. As one alternative to this CoA transferase, a mechanism can be employed in which the 4-HB is first phosphorylated by ATP and then converted to the CoA derivative, similar to the acetate kinase/phosphotransacetylase pathway in E. coli for the conversion of acetate to acetyl-CoA. The net cost of this route is one ATP, which is the same as is required to regenerate acetyl-CoA from acetate. The enzymes phosphotransbutyrylase (ptb) and butyrate kinase (bk) are known to carry out these steps on the non-hydroxylated molecules for butyrate production in C. acetobutylicum (Cary et al., Appl Environ Microbiol 56:1576-1583 (1990); Valentine, R. C. and R. S. Wolfe, J Biol. Chem. 235:1948-1952 (1960)). These enzymes are reversible, allowing synthesis to proceed in the direction of 4-HB.

BDO also can be produced via α-ketoglutarate in addition to or instead of through succinate. A described previously, and exemplified further below, one pathway to accomplish product biosynthesis is with the production of succinic semialdehyde via α-ketoglutarate using the endogenous enzymes (FIG. 1, Steps 4-5). An alternative is to use an α-ketoglutarate decarboxylase that can perform this conversion in one step (FIG. 1, Step 8; Tian et al., Proc Natl Acad Sci U.S.A 102:10670-10675 (2005)).

For the construction of different strains of BDO-producing microbial organisms, a list of applicable genes was assembled for corroboration. Briefly, one or more genes within the 4-HB and/or BDO biosynthetic pathways were identified for each step of the complete BDO-producing pathway shown in FIG. 1, using available literature resources, the NCBI genetic database, and homology searches. The genes cloned and assessed in this study are presented below in Table 6, along with the appropriate references and URL citations to the polypeptide sequence. As discussed further below, some genes were synthesized for codon optimization while others were cloned via PCR from the genomic DNA of the native or wild-type organism. For some genes both approaches were used, and in this case the native genes are indicated by an “n” suffix to the gene identification number when used in an experiment. Note that only the DNA sequences differ; the proteins are identical.

Expression Vector Construction for BDO Pathway.

Vector backbones and some strains were obtained from Dr. Rolf Lutz of Expressys (expressys.de/). The vectors and strains are based on the pZ Expression System developed by Dr. Rolf Lutz and Prof. Hermann Bujard (Lutz, R. and H. Bujard, Nucleic Acids Res 25:1203-1210 (1997)). Vectors obtained were pZE13luc, pZA33luc, pZS* 13luc and pZE22luc and contained the luciferase gene as a stuffer fragment. To replace the luciferase stuffer fragment with a lacZ-alpha fragment flanked by appropriate restriction enzyme sites, the luciferase stuffer fragment was first removed from each vector by digestion with EcoRI and XbaI. The lacZ-alpha fragment was PCR amplified from pUC19 with the following primers:

This generated a fragment with a 5′ end of EcoRI site, NheI site, a Ribosomal Binding Site, a SalI site and the start codon. On the 3′ end of the fragment contained the stop codon, XbaI, HindIII, and AvrII sites. The PCR product was digested with EcoRI and AvrII and ligated into the base vectors digested with EcoRI and XbaI (XbaI and AvrII have compatible ends and generate a non-site). Because NheI and XbaI restriction enzyme sites generate compatible ends that can be ligated together (but generate a NheI/XbaI non-site that is not digested by either enzyme), the genes cloned into the vectors could be “Biobricked” together (openwetware.org/wiki/Synthetic_Biology:BioBricks). Briefly, this method enables joining an unlimited number of genes into the vector using the same 2 restriction sites (as long as the sites do not appear internal to the genes), because the sites between the genes are destroyed after each addition.

All vectors have the pZ designation followed by letters and numbers indication the origin of replication, antibiotic resistance marker and promoter/regulatory unit. The origin of replication is the second letter and is denoted by E for ColE1, A for p15A and S for pSC101-based origins. The first number represents the antibiotic resistance marker (1 for Ampicillin, 2 for Kanamycin, 3 for Chloramphenicol, 4 for Spectinomycin and 5 for Tetracycline). The final number defines the promoter that regulated the gene of interest (1 for P_LtetO-1, 2 for P_LlacO-1, 3 for P_A1lacO-1, and 4 for P_lac/ara-1). The MCS and the gene of interest follows immediately after. For the work discussed here we employed two base vectors, pZA33 and pZE13, modified for the biobricks insertions as discussed above. Once the gene(s) of interest have been cloned into them, resulting plasmids are indicated using the four digit gene codes given in Table 6; e.g., pZA33-XXXX-YYYY-.

Host Strain Construction.

The parent strain in all studies described here is E. coli K-12 strain MG1655. Markerless deletion strains in adhE, gabD, and aldA were constructed under service contract by a third party using the redET method (Datsenko, K. A. and B. L. Wanner, Proc Natl Acad Sci U.S.A 97:6640-6645 (2000)). Subsequent strains were constructed via bacteriophage P1 mediated transduction (Miller, J. Experiments in Molecular Genetics, Cold Spring Harbor Laboratories, New York (1973)). Strain C600Z1 (laci^q, PN25-tetR, Sp^R, lacY1, leuB6,mcrB+, supE44, thi-1, thr-1, tonA21) was obtained from Expressys and was used as a source of a lacIq allele for P1 transduction. Bacteriophage P1vir was grown on the C600Z1 E. coli strain, which has the spectinomycin resistance gene linked to the lacIq. The P1 lysate grown on C600Z1 was used to infect MG1655 with selection for spectinomycin resistance. The spectinomycin resistant colonies were then screened for the linked lacI^qby determining the ability of the transductants to repress expression of a gene linked to a P_A1lacO-1promoter. The resulting strain was designated MG1655 lacI'. A similar procedure was used to introduce lacI^Qinto the deletion strains.

Production of 4-HB From Succinate.

For construction of a 4-HB producer from succinate, genes encoding steps from succinate to 4-HB and 4-HB-CoA (1, 6, 7, and 9 in FIG. 1) were assembled onto the pZA33 and pZE13 vectors as described below. Various combinations of genes were assessed, as well as constructs bearing incomplete pathways as controls (Tables 7 and 8). The plasmids were then transformed into host strains containing lacI^Q, which allow inducible expression by addition of isopropyl β-D-1-thiogalactopyranoside (IPTG). Both wild-type and hosts with deletions in genes encoding the native succinic semialdehyde dehydrogenase (step 2 in FIG. 1) were tested.

Activity of the heterologous enzymes were first tested in in vitro assays, using strain MG1655 lacI^Qas the host for the plasmid constructs containing the pathway genes. Cells were grown aerobically in LB media (Difco) containing the appropriate antibiotics for each construct, and induced by addition of IPTG at 1 mM when the optical density (OD600) reached approximately 0.5. Cells were harvested after 6 hours, and enzyme assays conducted as discussed below.

In Vitro Enzyme Assays.

To obtain crude extracts for activity assays, cells were harvested by centrifugation at 4,500 rpm (Beckman-Coulter, Allegera X-15R) for 10 min. The pellets were resuspended in 0.3 mL BugBuster (Novagen) reagent with benzonase and lysozyme, and lysis proceeded for 15 minutes at room temperature with gentle shaking. Cell-free lysate was obtained by centrifugation at 14,000 rpm (Eppendorf centrifuge 5402) for 30 min at 4° C. Cell protein in the sample was determined using the method of Bradford et al., Anal. Biochem. 72:248-254 (1976), and specific enzyme assays conducted as described below. Activities are reported in Units/mg protein, where a unit of activity is defined as the amount of enzyme required to convert 1 mmol of substrate in 1 min. at room temperature. In general, reported values are averages of at least 3 replicate assays.

Succinyl-CoA transferase (Cat1) activity was determined by monitoring the formation of acetyl-CoA from succinyl-CoA and acetate, following a previously described procedure Sohling and Gottschalk, J. Bacteriol. 178:871-880 (1996). Succinyl-CoA synthetase (SucCD) activity was determined by following the formation of succinyl-CoA from succinate and CoA in the presence of ATP. The experiment followed a procedure described by Cha and Parks, J. Biol. Chem. 239:1961-1967 (1964). CoA-dependent succinate semialdehyde dehydrogenase (SucD) activity was determined by following the conversion of NAD to NADH at 340 nm in the presence of succinate semialdehyde and CoA (Sohling and Gottschalk, Eur. J. Biochem. 212:121-127 (1993)). 4-HB dehydrogenase (4-HBd) enzyme activity was determined by monitoring the oxidation of NADH to NAD at 340 nm in the presence of succinate semialdehyde. The experiment followed a published procedure Gerhardt et al. Arch. Microbiol. 174:189-199 (2000). 4-HB CoA transferase (Cat2) activity was determined using a modified procedure from Scherf and Buckel, Appl. Environ. Microbiol. 57:2699-2702 (1991). The formation of 4-HB-CoA or butyryl-CoA formation from acetyl-CoA and 4-HB or butyrate was determined using HPLC.

Alcohol (ADH) and aldehyde (ALD) dehydrogenase was assayed in the reductive direction using a procedure adapted from several literature sources (Durre et al., FEMS Microbiol. Rev. 17:251-262 (1995); Palosaari and Rogers, J. Bacteriol. 170:2971-2976 (1988) and Welch et al., Arch. Biochem. Biophys. 273:309-318 (1989). The oxidation of NADH is followed by reading absorbance at 340 nM every four seconds for a total of 240 seconds at room temperature. The reductive assays were performed in 100 mM MOPS (adjusted to pH 7.5 with KOH), 0.4 mM NADH, and from 1 to 50 μl of cell extract. The reaction is started by adding the following reagents: 100 μl of 100 mM acetaldehyde or butyraldehyde for ADH, or 100 μl of 1 mM acetyl-CoA or butyryl-CoA for ALD. The Spectrophotometer is quickly blanked and then the kinetic read is started. The resulting slope of the reduction in absorbance at 340 nM per minute, along with the molar extinction coefficient of NAD(P)H at 340 nM (6000) and the protein concentration of the extract, can be used to determine the specific activity.

The enzyme activity of PTB is measured in the direction of butyryl-CoA to butyryl-phosphate as described in Cary et al. J. Bacteriol. 170:4613-4618 (1988). It provides inorganic phosphate for the conversion, and follows the increase in free CoA with the reagent 5,5′-dithiobis-(2-nitrobenzoic acid), or DTNB. DTNB rapidly reacts with thiol groups such as free CoA to release the yellow-colored 2-nitro-5-mercaptobenzoic acid (TNB), which absorbs at 412 nm with a molar extinction coefficient of 14,140 M cm⁻¹. The assay buffer contained 150 mM potassium phosphate at pH 7.4, 0.1 mM DTNB, and 0.2 mM butyryl-CoA, and the reaction was started by addition of 2 to 50 μL cell extract. The enzyme activity of BK is measured in the direction of butyrate to butyryl-phosphate formation at the expense of ATP. The procedure is similar to the assay for acetate kinase previously described Rose et al., J. Biol. Chem. 211:737-756 (1954). However we have found another acetate kinase enzyme assay protocol provided by Sigma to be more useful and sensitive. This assay links conversion of ATP to ADP by acetate kinase to the linked conversion of ADP and phosphoenol pyruvate (PEP) to ATP and pyruvate by pyruvate kinase, followed by the conversion of pyruvate and NADH to lactate and NAD+ by lactate dehydrogenase. Substituting butyrate for acetate is the only major modification to enable the assay to follow BK enzyme activity. The assay mixture contained 80 mM triethanolamine buffer at pH 7.6, 200 mM sodium butyrate, 10 mM MgCl₂, 0.1 mM NADH, 6.6 mM ATP, 1.8 mM phosphoenolpyruvate. Pyruvate kinase, lactate dehydrogenase, and myokinase were added according to the manufacturer's instructions. The reaction was started by adding 2 to 50 μL cell extract, and the reaction was monitored based on the decrease in absorbance at 340 nm indicating NADH oxidation.

Analysis of CoA Derivatives by HPLC.

An HPLC based assay was developed to monitor enzymatic reactions involving coenzyme A (CoA) transfer. The developed method enabled enzyme activity characterization by quantitative determination of CoA, acetyl CoA (AcCoA), butyryl CoA (BuCoA) and 4-hydroxybutyrate CoA (4-HBCoA) present in in-vitro reaction mixtures. Sensitivity down to low μM was achieved, as well as excellent resolution of all the CoA derivatives of interest.

Chemical and sample preparation was performed as follows. Briefly, CoA, AcCoA, BuCoA and all other chemicals, were obtained from Sigma-Aldrich. The solvents, methanol and acetonitrile, were of HPLC grade. Standard calibration curves exhibited excellent linearity in the 0.01-1 mg/mL concentration range. Enzymatic reaction mixtures contained 100 mM Tris HCl buffer (pH 7), aliquots were taken at different time points, quenched with formic acid (0.04% final concentration) and directly analyzed by HPLC.

HPLC analysis was performed using an Agilent 1100 HPLC system equipped with a binary pump, degasser, thermostated autosampler and column compartment, and diode array detector (DAD), was used for the analysis. A reversed phase column, Kromasil 100 5 um C18, 4.6×150 mm (Peeke Scientific), was employed. 25 mM potassium phosphate (pH 7) and methanol or acetonitrile, were used as aqueous and organic solvents at 1 mL/min flow rate. Two methods were developed: a short one with a faster gradient for the analysis of well-resolved CoA, AcCoA and BuCoA, and a longer method for distinguishing between closely eluting AcCoA and 4-HBCoA. Short method employed acetonitrile gradient (0 min-5%, 6 min-30%, 6.5 min-5%, 10 min-5%) and resulted in the retention times 2.7, 4.1 and 5.5 min for CoA, AcCoA and BuCoA, respectively. In the long method methanol was used with the following linear gradient: 0 min-5%, 20 min-35%, 20.5 min-5%, 25 min-5%. The retention times for CoA, AcCoA, 4-HBCoA and BuCoA were 5.8, 8.4, 9.2 and 16.0 min, respectively. The injection volume was 5 μL, column temperature 30° C., and UV absorbance was monitored at 260 nm.

The results demonstrated activity of each of the four pathway steps (Table 7), though activity is clearly dependent on the gene source, position of the gene in the vector, and the context of other genes with which it is expressed. For example, gene 0035 encodes a succinic semialdehyde dehydrogenase that is more active than that encoded by 0008, and 0036 and 0010n are more active 4-HB dehydrogenase genes than 0009. There also seems to be better 4-HB dehydrogenase activity when there is another gene preceding it on the same operon.

Recombinant strains containing genes in the 4-HB pathway were then evaluated for the ability to produce 4-HB in vivo from central metabolic intermediates. Cells were grown anaerobically in LB medium to OD600 of approximately 0.4, then induced with 1 mM IPTG. One hour later, sodium succinate was added to 10 mM, and samples taken for analysis following an additional 24 and 48 hours. 4-HB in the culture broth was analyzed by GC-MS as described below. The results indicate that the recombinant strain can produce over 2 mM 4-HB after 24 hours, compared to essentially zero in the control strain (Table 8).

An alternate to using a CoA transferase (cat1) to produce succinyl-CoA from succinate is to use the native E. coli sucCD genes, encoding succinyl-CoA synthetase. This gene cluster was cloned onto pZE13 along with candidate genes for the remaining steps to 4-HB to create pZE13-0038-0035-0036.

Production of 4-HB from Glucose.

Although the above experiments demonstrate a functional pathway to 4-HB from a central metabolic intermediate (succinate), an industrial process would require the production of chemicals from low-cost carbohydrate feedstocks such as glucose or sucrose. Thus, the next set of experiments was aimed to determine whether endogenous succinate produced by the cells during growth on glucose could fuel the 4-HB pathway. Cells were grown anaerobically in M9 minimal medium (6.78 g/L Na₂HPO₄, 3.0 g/L KH₂PO₄, 0.5 g/L NaCl, 1.0 g/L NH₄Cl, 1 mM MgSO₄, 0.1 mM CaCl₂) supplemented with 20 g/L glucose, 100 mM 3-(N-morpholino)propanesulfonic acid (MOPS) to improve the buffering capacity, 10 μg/mL thiamine, and the appropriate antibiotics. 0.25 mM IPTG was added when OD600 reached approximately 0.2, and samples taken for 4-HB analysis every 24 hours following induction. In all cases 4-HB plateaued after 24 hours, with a maximum of about 1 mM in the best strains (FIG. 3a), while the succinate concentration continued to rise (FIG. 3b). This indicates that the supply of succinate to the pathway is likely not limiting, and that the bottleneck may be in the activity of the enzymes themselves or in NADH availability. 0035 and 0036 are clearly the best gene candidates for CoA-dependent succinic semialdehyde dehydrogenase and 4-HB dehydrogenase, respectively. The elimination of one or both of the genes encoding known (gabD) or putative (aldA) native succinic semialdehyde dehydrogenases had little effect on performance. Finally, it should be noted that the cells grew to a much lower OD in the 4-HB-producing strains than in the controls (FIG. 3c).

An alternate pathway for the production of 4-HB from glucose is via α-ketoglutarate. We explored the use of an α-ketoglutarate decarboxylase from Mycobacterium tuberculosis Tian et al., Proc. Natl. Acad. Sci. USA 102:10670-10675 (2005) to produce succinic semialdehyde directly from α-ketoglutarate (step 8 in FIG. 1). To demonstrate that this gene (0032) was functional in vivo, we expressed it on pZE13 in the same host as 4-HB dehydrogenase (gene 0036) on pZA33. This strain was capable of producing over 1.0 mM 4-HB within 24 hours following induction with 1 mM IPTG (FIG. 4). Since this strain does not express a CoA-dependent succinic semialdehyde dehydrogenase, the possibility of succinic semialdehyde production via succinyl-CoA is eliminated. It is also possible that the native genes responsible for producing succinic semialdehyde could function in this pathway (steps 4 and 5 in FIG. 1); however, the amount of 4-HB produced when the pZE13-0032 plasmid was left out of the host is the negligible.

Production of BDO from 4-HB.

The production of BDO from 4-HB required two reduction steps, catalyzed by dehydrogenases. Alcohol and aldehyde dehydrogenases (ADH and ALD, respectively) are NAD+/H and/or NADP+/H-dependent enzymes that together can reduce a carboxylic acid group on a molecule to an alcohol group, or in reverse, can perform the oxidation of an alcohol to a carboxylic acid. This biotransformation has been demonstrated in wild-type Clostridium acetobutylicum (Jewell et al., Current Microbiology, 13:215-19 (1986)), but neither the enzymes responsible nor the genes responsible were identified. In addition, it is not known whether activation to 4-HB-CoA is first required (step 9 in FIG. 1), or if the aldehyde dehydrogenase (step 12) can act directly on 4-HB. We developed a list of candidate enzymes from C. acetobutylicum and related organisms based on known activity with the non-hydroxylated analogues to 4-HB and pathway intermediates, or by similarity to these characterized genes (Table 6). Since some of the candidates are multifunctional dehydrogenases, they could potentially catalyze both the NAD(P)H-dependent reduction of the acid (or CoA-derivative) to the aldehyde, and of the aldehyde to the alcohol. Before beginning work with these genes in E. coli, we first validated the result referenced above using C. acetobutylicum ATCC 824. Cells were grown in Schaedler broth (Accumedia, Lansing, Mich.) supplemented with 10 mM 4-HB, in an anaerobic atmosphere of 10% CO₂, 10% H₂, and 80% N₂at 30° C. Periodic culture samples were taken, centrifuged, and the broth analyzed for BDO by GC-MS as described below. BDO concentrations of 0.1 mM, 0.9 mM, and 1.5 mM were detected after 1 day, 2 days, and 7 days incubation, respectively. No BDO was detected in culture grown without 4-HB addition. To demonstrate that the BDO produced was derived from glucose, we grew the best BDO producing strain MG1655 lacI^QpZE13-0004-0035-0002 pZA33-0034-0036 in M9 minimal medium supplemented with 4 g/L uniformly labeled¹³C-glucose. Cells were induced at OD of 0.67 with 1 mM IPTG, and a sample taken after 24 hours. Analysis of the culture supernatant was performed by mass spectrometry.

Gene candidates for the 4-HB to BDO conversion pathway were next tested for activity when expressed in the E. coli host MG1655 lacI^Q. Recombinant strains containing each gene candidate expressed on pZA33 were grown in the presence of 0.25 mM IPTG for four hours at 37° C. to fully induce expression of the enzyme. Four hours after induction, cells were harvested and assayed for ADH and ALD activity as described above. Since 4-HB-CoA and 4-hydroxybutyraldehyde are not available commercially, assays were performed using the non-hydroxylated substrates (Table 9). The ratio in activity between 4-carbon and 2-carbon substrates for C. acetobutylicum adhE2 (0002) and E. coli adhE (0011) were similar to those previously reported in the literature a Atsumi et al., Biochim. Biophys. Acta. 1207:1-11 (1994).

For the BDO production experiments, cat2 from Porphyromonas gingivalis W83 (gene 0034) was included on pZA33 for the conversion of 4-HB to 4-HB-CoA, while the candidate dehydrogenase genes were expressed on pZE13. The host strain was MG1655 lacI^Q. Along with the alcohol and aldehyde dehydrogenase candidates, we also tested the ability of CoA-dependent succinic semialdehyde dehydrogenases (sucD) to function in this step, due to the similarity of the substrates. Cells were grown to an OD of about 0.5 in LB medium supplemented with 10 mM 4-HB, induced with 1 mM IPTG, and culture broth samples taken after 24 hours and analyzed for BDO as described below. The best BDO production occurred using adhE2 from C. acetobutylicum, sucD from C. kluyveri, or sucD from P. gingivalis (FIG. 5). Interestingly, the absolute amount of BDO produced was higher under aerobic conditions; however, this is primarily due to the lower cell density achieved in anaerobic cultures. When normalized to cell OD, the BDO production per unit biomass is higher in anaerobic conditions (Table 10).

As discussed above, it may be advantageous to use a route for converting 4-HB to 4-HB-CoA that does not generate acetate as a byproduct. To this aim, we tested the use of phosphotransbutyrylase (ptb) and butyrate kinase (bk) from C. acetobutylicum to carry out this conversion via steps 10 and 11 in FIG. 1. The native ptb/bk operon from C. acetobutylicum (genes 0020 and 0021) was cloned and expressed in pZA33. Extracts from cells containing the resulting construct were taken and assayed for the two enzyme activities as described herein. The specific activity of BK was approximately 65 U/mg, while the specific activity of PTB was approximately 5 U/mg. One unit (U) of activity is defined as conversion of 1 μM substrate in 1 minute at room temperature. Finally, the construct was tested for participation in the conversion of 4-HB to BDO. Host strains were transformed with the pZA33-0020-0021 construct described and pZE13-0002, and compared to use of cat2 in BDO production using the aerobic procedure used above in FIG. 5. The BK/PTB strain produced 1 mM BDO, compared to 2 mM when using cat2 (Table 11). Interestingly, the results were dependent on whether the host strain contained a deletion in the native adhE gene.

Production of BDO from Glucose. The final step of pathway corroboration is to express both the 4-HB and BDO segments of the pathway in E. coli and demonstrate production of BDO in glucose minimal medium. New plasmids were constructed so that all the required genes fit on two plasmids. In general, cat1, adhE, and sucD genes were expressed from pZE13, and cat2 and 4-HBd were expressed from pZA33. Various combinations of gene source and gene order were tested in the MG1655 lacI^Qbackground. Cells were grown anaerobically in M9 minimal medium (6.78 g/L Na₂HPO₄, 3.0 g/L KH₂PO₄, 0.5 g/L NaCl, 1.0 g/L NH₄C1, 1 mM MgSO₄, 0.1 mM CaCl₂) supplemented with 20 g/L glucose, 100 mM 3-(N-morpholino)propanesulfonic acid (MOPS) to improve the buffering capacity, 10 μg/mL thiamine, and the appropriate antibiotics. 0.25 mM IPTG was added approximately 15 hours following inoculation, and culture supernatant samples taken for BDO, 4-HB, and succinate analysis 24 and 48 hours following induction. The production of BDO appeared to show a dependency on gene order (Table 12). The highest BDO production, over 0.5 mM, was obtained with cat2 expressed first, followed by 4-HBd on pZA33, and cat1 followed by P. gingivalis sucD on pZE13. The addition of C. acetobutylicum adhE2 in the last position on pZE13 resulted in slight improvement. 4-HB and succinate were also produced at higher concentrations.

Analysis of BDO, 4-HB and Succinate by GCMS.

BDO, 4-HB and succinate in fermentation and cell culture samples were derivatized by silylation and quantitatively analyzed by GCMS using methods adapted from literature reports ((Simonov et al., J. Anal Chem. 59:965-971 (2004)). The developed method demonstrated good sensitivity down to 1 μM, linearity up to at least 25 mM, as well as excellent selectivity and reproducibility.

Sample preparation was performed as follows: 100 μL filtered (0.2 μm or 0.45 μm syringe filters) samples, e.g. fermentation broth, cell culture or standard solutions, were dried down in a Speed Vac Concentrator (Savant SVC-100H) for approximately 1 hour at ambient temperature, followed by the addition of 20 μL 10 mM cyclohexanol solution, as an internal standard, in dimethylformamide. The mixtures were vortexed and sonicated in a water bath (Branson 3510) for 15 min to ensure homogeneity. 100 μL silylation derivatization reagent, N,O-bis(trimethylsilyl)trifluoro-acetimide (BSTFA) with 1% trimethylchlorosilane, was added, and the mixture was incubated at 70° C. for 30 min. The derivatized samples were centrifuged for 5 min, and the clear solutions were directly injected into GCMS. All the chemicals and reagents were from Sigma-Aldrich, with the exception of BDO which was purchased from J. T. Baker.

GCMS was performed on an Agilent gas chromatograph 6890N, interfaced to a mass-selective detector (MSD) 5973N operated in electron impact ionization (EI) mode has been used for the analysis. A DB-5MS capillary column (J&W Scientific, Agilent Technologies), 30m×0.25 mm i.d.×0.25 μm film thickness, was used. The GC was operated in a split injection mode introducing 1 μL of sample at 20:1 split ratio. The injection port temperature was 250° C. Helium was used as a carrier gas, and the flow rate was maintained at 1.0 mL/min. A temperature gradient program was optimized to ensure good resolution of the analytes of interest and minimum matrix interference. The oven was initially held at 80° C. for 1 min, then ramped to 120° C. at 2° C./min, followed by fast ramping to 320° C. at 100° C./min and final hold for 6 min at 320° C. The MS interface transfer line was maintained at 280° C. The data were acquired using ‘lowmass’ MS tune settings and 30-400 m/z mass-range scan. The total analysis time was 29 min including 3 min solvent delay. The retention times corresponded to 5.2, 10.5, 14.0 and 18.2 min for BSTFA-derivatized cyclohexanol, BDO, 4-HB and succinate, respectively. For quantitative analysis, the following specific mass fragments were selected (extracted ion chromatograms): m/z 157 for internal standard cyclohexanol, 116 for BDO, and 147 for both 4-HB and succinate. Standard calibration curves were constructed using analyte solutions in the corresponding cell culture or fermentation medium to match sample matrix as close as possible. GCMS data were processed using Environmental Data Analysis ChemStation software (Agilent Technologies).

The results indicated that most of the 4-HB and BDO produced were labeled with¹³C (FIG. 6, right-hand sides). Mass spectra from a parallel culture grown in unlabeled glucose are shown for comparison (FIG. 6, left-hand sides). Note that the peaks seen are for fragments of the derivatized molecule containing different numbers of carbon atoms from the metabolite. The derivatization reagent also contributes some carbon and silicon atoms that naturally-occurring label distribution, so the results are not strictly quantitative.

Production of BDO from 4-HB using Alternate Pathways.

The various alternate pathways were also tested for BDO production. This includes use of the native E. coli SucCD enzyme to convert succinate to succinyl-CoA (Table 13, rows 2-3), use of α-ketoglutarate decarboxylase in the α-ketoglutarate pathway (Table 13, row 4), and use of PTB/BK as an alternate means to generate the CoA-derivative of 4HB (Table 13, row 1). Strains were constructed containing plasmids expressing the genes indicated in Table 13, which encompass these variants. The results show that in all cases, production of 4-HB and BDO occurred (Table 13).

This Example describes the biosynthetic production of 4-hydroxybutanoic acid, γ-butyrolactone and 1,4-butanediol using fermentation and other bioprocesses.

Methods for the integration of the 4-HB fermentation step into a complete process for the production of purified GBL, 1,4-butanediol (BDO) and tetrahydrofuran (THF) are described below. Since 4-HB and GBL are in equilibrium, the fermentation broth will contain both compounds. At low pH this equilibrium is shifted to favor GBL. Therefore, the fermentation can operate at pH 7.5 or less, generally pH 5.5 or less. After removal of biomass, the product stream enters into a separation step in which GBL is removed and the remaining stream enriched in 4-HB is recycled. Finally, GBL is distilled to remove any impurities. The process operates in one of three ways: 1) fed-batch fermentation and batch separation; 2) fed-batch fermentation and continuous separation; 3) continuous fermentation and continuous separation. The first two of these modes are shown schematically in FIG. 7. The integrated fermentation procedures described below also are used for the BDO producing cells of the invention for biosynthesis of BDO and subsequent BDO family products.

Fermentation Protocol to Produce 4-HB/GBL (Batch):

The production organism is grown in a 10 L bioreactor sparged with an N₂/CO₂mixture, using 5 L broth containing 5 g/L potassium phosphate, 2.5 g/L ammonium chloride, 0.5 g/L magnesium sulfate, and 30 g/L corn steep liquor, and an initial glucose concentration of 20 g/L. As the cells grow and utilize the glucose, additional 70% glucose is fed into the bioreactor at a rate approximately balancing glucose consumption. The temperature of the bioreactor is maintained at 30 degrees C. Growth continues for approximately 24 hours, until 4-HB reaches a concentration of between 20-200 g/L, with the cell density being between 5 and 10 g/L. The pH is not controlled, and will typically decrease to pH 3-6 by the end of the run. Upon completion of the cultivation period, the fermenter contents are passed through a cell separation unit (e.g., centrifuge) to remove cells and cell debris, and the fermentation broth is transferred to a product separations unit. Isolation of 4-HB and/or GBL would take place by standard separations procedures employed in the art to separate organic products from dilute aqueous solutions, such as liquid-liquid extraction using a water immiscible organic solvent (e.g., toluene) to provide an organic solution of 4-HB/GBL. The resulting solution is then subjected to standard distillation methods to remove and recycle the organic solvent and to provide GBL (boiling point 204-205° C.) which is isolated as a purified liquid.

Fermentation Protocol to Produce 4-HB/GBL (Fully Continuous):

The production organism is first grown up in batch mode using the apparatus and medium composition described above, except that the initial glucose concentration is 30-50 g/L. When glucose is exhausted, feed medium of the same composition is supplied continuously at a rate between 0.5 L/hr and 1 L/hr, and liquid is withdrawn at the same rate. The 4-HB concentration in the bioreactor remains constant at 30-40 g/L, and the cell density remains constant between 3-5 g/L. Temperature is maintained at 30 degrees C., and the pH is maintained at 4.5 using concentrated NaOH and HCl, as required. The bioreactor is operated continuously for one month, with samples taken every day to assure consistency of 4-HB concentration. In continuous mode, fermenter contents are constantly removed as new feed medium is supplied. The exit stream, containing cells, medium, and products 4-HB and/or GBL, is then subjected to a continuous product separations procedure, with or without removing cells and cell debris, and would take place by standard continuous separations methods employed in the art to separate organic products from dilute aqueous solutions, such as continuous liquid-liquid extraction using a water immiscible organic solvent (e.g., toluene) to provide an organic solution of 4-HB/GBL. The resulting solution is subsequently subjected to standard continuous distillation methods to remove and recycle the organic solvent and to provide GBL (boiling point 204-205° C.) which is isolated as a purified liquid.

GBL Reduction Protocol:

Once GBL is isolated and purified as described above, it will then be subjected to reduction protocols such as those well known in the art (references cited) to produce 1,4-butanediol or tetrahydrofuran (THF) or a mixture thereof. Heterogeneous or homogeneous hydrogenation catalysts combined with GBL under hydrogen pressure are well known to provide the products 1,4-butanediol or tetrahydrofuran (THF) or a mixture thereof. It is important to note that the 4-HB/GBL product mixture that is separated from the fermentation broth, as described above, may be subjected directly, prior to GBL isolation and purification, to these same reduction protocols to provide the products 1,4-butanediol or tetrahydrofuran or a mixture thereof. The resulting products, 1,4-butanediol and THF are then isolated and purified by procedures well known in the art.

Fermentation and Hydrogenation Protocol to Produce BDO or THF Directly (Batch):

Cells are grown in a 10 L bioreactor sparged with an N₂/CO₂mixture, using 5 L broth containing 5 g/L potassium phosphate, 2.5 g/L ammonium chloride, 0.5 g/L magnesium sulfate, and 30 g/L corn steep liquor, and an initial glucose concentration of 20 g/L. As the cells grow and utilize the glucose, additional 70% glucose is fed into the bioreactor at a rate approximately balancing glucose consumption. The temperature of the bioreactor is maintained at 30 degrees C. Growth continues for approximately 24 hours, until 4-HB reaches a concentration of between 20-200 g/L, with the cell density being between 5 and 10 g/L. The pH is not controlled, and will typically decrease to pH 3-6 by the end of the run. Upon completion of the cultivation period, the fermenter contents are passed through a cell separation unit (e.g., centrifuge) to remove cells and cell debris, and the fermentation broth is transferred to a reduction unit (e.g., hydrogenation vessel), where the mixture 4-HB/GBL is directly reduced to either 1,4-butanediol or THF or a mixture thereof. Following completion of the reduction procedure, the reactor contents are transferred to a product separations unit. Isolation of 1,4-butanediol and/or THF would take place by standard separations procedures employed in the art to separate organic products from dilute aqueous solutions, such as liquid-liquid extraction using a water immiscible organic solvent (e.g., toluene) to provide an organic solution of 1,4-butanediol and/or THF. The resulting solution is then subjected to standard distillation methods to remove and recycle the organic solvent and to provide 1,4-butanediol and/or THF which are isolated as a purified liquids.

Fermentation and Hydrogenation Protocol to Produce BDO or THF Directly (Fully Continuous):

The cells are first grown up in batch mode using the apparatus and medium composition described above, except that the initial glucose concentration is 30-50 g/L. When glucose is exhausted, feed medium of the same composition is supplied continuously at a rate between 0.5 L/hr and 1 L/hr, and liquid is withdrawn at the same rate. The 4-HB concentration in the bioreactor remains constant at 30-40 g/L, and the cell density remains constant between 3-5 g/L. Temperature is maintained at 30 degrees C., and the pH is maintained at 4.5 using concentrated NaOH and HCl, as required. The bioreactor is operated continuously for one month, with samples taken every day to assure consistency of 4-HB concentration. In continuous mode, fermenter contents are constantly removed as new feed medium is supplied. The exit stream, containing cells, medium, and products 4-HB and/or GBL, is then passed through a cell separation unit (e.g., centrifuge) to remove cells and cell debris, and the fermentation broth is transferred to a continuous reduction unit (e.g., hydrogenation vessel), where the mixture 4-HB/GBL is directly reduced to either 1,4-butanediol or THF or a mixture thereof. Following completion of the reduction procedure, the reactor contents are transferred to a continuous product separations unit. Isolation of 1,4-butanediol and/or THF would take place by standard continuous separations procedures employed in the art to separate organic products from dilute aqueous solutions, such as liquid-liquid extraction using a water immiscible organic solvent (e.g., toluene) to provide an organic solution of 1,4-butanediol and/or THF. The resulting solution is then subjected to standard continuous distillation methods to remove and recycle the organic solvent and to provide 1,4-butanediol and/or THF which are isolated as a purified liquids.

Fermentation Protocol to Produce BDO Directly (Batch):

The production organism is grown in a 10 L bioreactor sparged with an N₂/CO₂mixture, using 5 L broth containing 5 g/L potassium phosphate, 2.5 g/L ammonium chloride, 0.5 g/L magnesium sulfate, and 30 g/L corn steep liquor, and an initial glucose concentration of 20 g/L. As the cells grow and utilize the glucose, additional 70% glucose is fed into the bioreactor at a rate approximately balancing glucose consumption. The temperature of the bioreactor is maintained at 30 degrees C. Growth continues for approximately 24 hours, until BDO reaches a concentration of between 20-200 g/L, with the cell density generally being between 5 and 10 g/L. Upon completion of the cultivation period, the fermenter contents are passed through a cell separation unit (e.g., centrifuge) to remove cells and cell debris, and the fermentation broth is transferred to a product separations unit. Isolation of BDO would take place by standard separations procedures employed in the art to separate organic products from dilute aqueous solutions, such as liquid-liquid extraction using a water immiscible organic solvent (e.g., toluene) to provide an organic solution of BDO. The resulting solution is then subjected to standard distillation methods to remove and recycle the organic solvent and to provide BDO (boiling point 228-229° C.) which is isolated as a purified liquid.

Fermentation Protocol to Produce BDO Directly (Fully Continuous):

The production organism is first grown up in batch mode using the apparatus and medium composition described above, except that the initial glucose concentration is 30-50 g/L. When glucose is exhausted, feed medium of the same composition is supplied continuously at a rate between 0.5 L/hr and 1 L/hr, and liquid is withdrawn at the same rate. The BDO concentration in the bioreactor remains constant at 30-40 g/L, and the cell density remains constant between 3-5 g/L. Temperature is maintained at 30 degrees C., and the pH is maintained at 4.5 using concentrated NaOH and HCl, as required. The bioreactor is operated continuously for one month, with samples taken every day to assure consistency of BDO concentration. In continuous mode, fermenter contents are constantly removed as new feed medium is supplied. The exit stream, containing cells, medium, and the product BDO, is then subjected to a continuous product separations procedure, with or without removing cells and cell debris, and would take place by standard continuous separations methods employed in the art to separate organic products from dilute aqueous solutions, such as continuous liquid-liquid extraction using a water immiscible organic solvent (e.g., toluene) to provide an organic solution of BDO. The resulting solution is subsequently subjected to standard continuous distillation methods to remove and recycle the organic solvent and to provide BDO (boiling point 228-229° C.) which is isolated as a purified liquid (mpt 20° C.).

This example describes exemplary enzymes and corresponding genes for 1,4-butandiol (BDO) synthetic pathways.

Exemplary BDO synthetic pathways are shown in FIGS. 8-13. The pathways depicted in FIGS. 8-13 are from common central metabolic intermediates to 1,4-butanediol. All transformations depicted in FIGS. 8-13 fall into the 18 general categories of transformations shown in Table 14. Below is described a number of biochemically characterized candidate genes in each category. Specifically listed are genes that can be applied to catalyze the appropriate transformations in FIGS. 9-13 when cloned and expressed in a host organism. The top three exemplary genes for each of the key steps in FIGS. 9-13 are provided in Tables 15-23 (see below). Exemplary genes were provided for the pathways depicted in FIG. 8 are described herein.

Aldehyde to alcohol. Exemplary genes encoding enzymes that catalyze the conversion of an aldehyde to alcohol, that is, alcohol dehydrogenase or equivalently aldehyde reductase, include alrA encoding a medium-chain alcohol dehydrogenase for C2-C14 (Tani et al. Appl. Environ. Microbiol. 66:5231-5235 (2000)), ADH2 from Saccharomyces cerevisiae (Atsumi et al. Nature 451:86-89 (2008)), yqhD from E. coli which has preference for molecules longer than C(3) (Sulzenbacher et al. Journal of Molecular Biology 342:489-502 (2004)), and bdh I and bdh II from C. acetobutylicum which converts butyryaldehyde into butanol (Walter et al. Journal of Bacteriology 174:7149-7158 (1992)). The protein sequences for each of these exemplary gene products, if available, can be found using the following GenBank accession numbers:

Enzymes exhibiting 4-hydroxybutyrate dehydrogenase activity (EC 1.1.1.61) also fall into this category. Such enzymes have been characterized in Ralstonia eutropha (Bravo et al. J. Forensic Sci. 49:379-387 (2004), Clostridium kluyveri (Wolff et al. Protein Expr. Purif. 6:206-212 (1995)) and Arabidopsis thaliana (Breitkreuz et al. J. Biol. Chem. 278:41552-41556 (2003)).

Another exemplary enzyme is 3-hydroxyisobutyrate dehydrogenase which catalyzes the reversible oxidation of 3-hydroxyisobutyrate to methylmalonate semialdehyde. This enzyme participates in valine, leucine and isoleucine degradation and has been identified in bacteria, eukaryotes, and mammals. The enzyme encoded by P84067 from Thermus thermophilus HB8 has been structurally characterized (Lokanath et al. J Mol Biol 352:905-17 (2005)). The reversibility of the human 3-hydroxyisobutyrate dehydrogenase was demonstrated using isotopically-labeled substrate (Manning et al. Biochem J 231:481-484 (1985)). Additional genes encoding this enzyme include 3hidh in Homo sapiens (Hawes et al. Methods Enzymol. 324:218-228 (2000)) and Oryctolagus cuniculus (Chowdhury et al. Biosci. Biotechnol Biochem. 60:2043-2047 (1996); Hawes et al. Methods Enzymol. 324:218-228 (2000)), mmsb in Pseudomonas aeruginosa, and dhat in Pseudomonas putida (Aberhart et al. J Chem. Soc. [Perkin 1] 6:1404-1406 (1979); Chowdhury et al. Biosci. Biotechnol Biochem. 67:438-441 (2003); Chowdhury et al. Biosci. Biotechnol Biochem. 60:2043-2047 (1996)).

Several 3-hydroxyisobutyrate dehydrogenase enzymes have also been shown to convert malonic semialdehyde to 3-hydroxyproprionic acid (3-HP). Three gene candidates exhibiting this activity are mmsB from Pseudomonas aeruginosa PAO1(62), mmsB from Pseudomonas putida KT2440 (Liao et al., US Publication 2005/0221466) and mmsB from Pseudomonas putida E23 (Chowdhury et al., Biosci. Biotechnol. Biochem. 60:2043-2047 (1996)). An enzyme with 3-hydroxybutyrate dehydrogenase activity in Alcaligenes faecalis M3A has also been identified (Gokam et al., U.S. Pat. No. 7,393,676; Liao et al., US Publication No. 2005/0221466). Additional gene candidates from other organisms including Rhodobacter spaeroides can be inferred by sequence similarity.

The conversion of malonic semialdehyde to 3-HP can also be accomplished by two other enzymes: NADH-dependent 3-hydroxypropionate dehydrogenase and NADPH-dependent malonate semialdehyde reductase. An NADH-dependent 3-hydroxypropionate dehydrogenase is thought to participate in beta-alanine biosynthesis pathways from propionate in bacteria and plants (Rathinasabapathi, B. Journal of Plant Pathology 159:671-674 (2002); Stadtman, E. R. J. Am. Chem. Soc. 77:5765-5766 (1955)). This enzyme has not been associated with a gene in any organism to date. NADPH-dependent malonate semialdehyde reductase catalyzes the reverse reaction in autotrophic CO₂-fixing bacteria. Although the enzyme activity has been detected in Metallosphaera sedula, the identity of the gene is not known (Alber et al. J. Bacteriol. 188:8551-8559 (2006)).

Ketone to hydroxyl. There exist several exemplary alcohol dehydrogenases that convert a ketone to a hydroxyl functional group. Two such enzymes from E. coli are encoded by malate dehydrogenase (mdh) and lactate dehydrogenase (ldhA). In addition, lactate dehydrogenase from Ralstonia eutropha has been shown to demonstrate high activities on substrates of various chain lengths such as lactate, 2-oxobutyrate, 2-oxopentanoate and 2-oxoglutarate (Steinbuchel, A. and H. G. Schlegel Eur. J. Biochem. 130:329-334 (1983)). Conversion of alpha-ketoadipate into alpha-hydroxyadipate can be catalyzed by 2-ketoadipate reductase, an enzyme reported to be found in rat and in human placenta (Suda et al. Arch. Biochem. Biophys. 176:610-620 (1976); Suda et al. Biochem. Biophys. Res. Commun. 77:586-591 (1977)). An additional candidate for this step is the mitochondrial 3-hydroxybutyrate dehydrogenase (bdh) from the human heart which has been cloned and characterized (Marks et al. J. Biol. Chem. 267:15459-15463 (1992)). This enzyme is a dehydrogenase that operates on a 3-hydroxyacid. Another exemplary alcohol dehydrogenase converts acetone to isopropanol as was shown in C. beijerinckii (Ismaiel et al. J. Bacteriol. 175:5097-5105 (1993)) and T. brockii (Lamed et al. Biochem. J. 195:183-190 (1981); Peretz and Burstein Biochemistry 28:6549-6555 (1989)).

Exemplary 3-hydroxyacyl dehydrogenases which convert acetoacetyl-CoA to 3-hydroxybutyryl-CoA include hbd from C. acetobutylicum (Boynton et al. Journal of Bacteriology 178:3015-3024 (1996)), hbd from C. beijerinckii (Colby et al. Appl Environ. Microbiol 58:3297-3302 (1992)), and a number of similar enzymes from Metallosphaera sedula (Berg et al. Archaea. Science. 318:1782-1786 (2007)).

Exemplary 2-step oxidoreductases that convert an acyl-CoA to alcohol include those that transform substrates such as acetyl-CoA to ethanol (for example, adhE from E. coli (Kessler et al. FEBS. Lett. 281:59-63 (1991)) and butyryl-CoA to butanol (for example, adhE2 from C. acetobutylicum (Fontaine et al. J. Bacteriol. 184:821-830 (2002)). In addition to reducing acetyl-CoA to ethanol, the enzyme encoded by adhE in Leuconostoc mesenteroides has been shown to oxide the branched chain compound isobutyraldehyde to isobutyryl-CoA (Kazahaya et al. J. Gen. Appl. Microbiol. 18:43-55 (1972); Koo et al. Biotechnol Lett. 27:505-510 (2005)).

Another exemplary enzyme can convert malonyl-CoA to 3-HP. An NADPH-dependent enzyme with this activity has characterized in Chloroflexus aurantiacus where it participates in the 3-hydroxypropionate cycle (Hugler et al., J. Bacteriol. 184:2404-2410 (2002); Strauss and Fuchs, Eur. J. Biochem. 215:633-643 (1993)). This enzyme, with a mass of 300 kDa, is highly substrate-specific and shows little sequence similarity to other known oxidoreductases (Hugler et al., J. Bacteriol. 184:2404-2410 (2002)). No enzymes in other organisms have been shown to catalyze this specific reaction; however there is bioinformatic evidence that other organisms may have similar pathways (Klatt et al., Environ. Microbiol. 9:2067-2078 (2007)). Enzyme candidates in other organisms including Roseiflexus castenholzii, Erythrobacter sp. NAP1 and marine gamma proteobacterium HTCC2080 can be inferred by sequence similarity.

Longer chain acyl-CoA molecules can be reduced by enzymes such as the jojoba (Simmondsia chinensis) FAR which encodes an alcohol-forming fatty acyl-CoA reductase. Its overexpression in E. coli resulted in FAR activity and the accumulation of fatty alcohol (Metz et al. Plant Physiology 122:635-644) 2000)).

Several acyl-CoA dehydrogenases are capable of reducing an acyl-CoA to its corresponding aldehyde. Exemplary genes that encode such enzymes include the Acinetobacter calcoaceticus acyl encoding a fatty acyl-CoA reductase (Reiser and Somerville, J. Bacteriology 179:2969-2975 (1997)), the Acinetobacter sp. M-1 fatty acyl-CoA reductase (Ishige et al. Appl. Environ. Microbiol. 68:1192-1195 (2002)), and a CoA- and NADP-dependent succinate semialdehyde dehydrogenase encoded by the sucD gene in Clostridium kluyveri (Sohling and Gottschalk J Bacteriol 178:871-80 (1996); Sohling and Gottschalk J. Bacteriol. 178:871-880 (1996)). SucD of P. gingivalis is another succinate semialdehyde dehydrogenase (Takahashi et al. J. Bacteriol. 182:4704-4710 (2000)). The enzyme acylating acetaldehyde dehydrogenase in Pseudomonas sp, encoded by bphG, is yet another as it has been demonstrated to oxidize and acylate acetaldehyde, propionaldehyde, butyraldehyde, isobutyraldehyde and formaldehyde (Powlowski et al. J Bacteriol. 175:377-385 (1993)).

An additional enzyme type that converts an acyl-CoA to its corresponding aldehyde is malonyl-CoA reductase which transforms malonyl-CoA to malonic semialdehyde. Malonyl-CoA reductase is a key enzyme in autotrophic carbon fixation via the 3-hydroxypropionate cycle in thermoacidophilic archael bacteria (Berg et al. Science 318:1782-1786 (2007); Thauer, R. K. Science 318:1732-1733 (2007)). The enzyme utilizes NADPH as a cofactor and has been characterized in Metallosphaera and Sulfolobus spp (Alber et al. J. Bacteriol. 188:8551-8559 (2006); Hugler et al. J. Bacteriol. 184:2404-2410 (2002)). The enzyme is encoded by Msed_—0709 in Metallosphaera sedula (Alber et al. J. Bacteriol. 188:8551-8559 (2006); Berg et al. Science 318:1782-1786 (2007)). A gene encoding a malonyl-CoA reductase from Sulfolobus tokodaii was cloned and heterologously expressed in E. coli (Alber et al. J. Bacteriol. 188:8551-8559 (2006)). Although the aldehyde dehydrogenase functionality of these enzymes is similar to the bifunctional dehydrogenase from Chloroflexus aurantiacus, there is little sequence similarity. Both malonyl-CoA reductase enzyme candidates have high sequence similarity to aspartate-semialdehyde dehydrogenase, an enzyme catalyzing the reduction and concurrent dephosphorylation of aspartyl-4-phosphate to aspartate semialdehyde. Additional gene candidates can be found by sequence homology to proteins in other organisms including Sulfolobus solfataricus and Sulfolobus acidocaldarius.

Enzymes in this family include 1) branched-chain 2-keto-acid dehydrogenase, 2) alpha-ketoglutarate dehydrogenase, and 3) the pyruvate dehydrogenase multienzyme complex (PDHC). These enzymes are multi-enzyme complexes that catalyze a series of partial reactions which result in acylating oxidative decarboxylation of 2-keto-acids. Each of the 2-keto-acid dehydrogenase complexes occupies key positions in intermediary metabolism, and enzyme activity is typically tightly regulated (Fries et al. Biochemistry 42:6996-7002 (2003)). The enzymes share a complex but common structure composed of multiple copies of three catalytic components: alpha-ketoacid decarboxylase (E1), dihydrolipoamide acyltransferase (E2) and dihydrolipoamide dehydrogenase (E3). The E3 component is shared among all 2-keto-acid dehydrogenase complexes in an organism, while the E1 and E2 components are encoded by different genes. The enzyme components are present in numerous copies in the complex and utilize multiple cofactors to catalyze a directed sequence of reactions via substrate channeling. The overall size of these dehydrogenase complexes is very large, with molecular masses between 4 and 10 million Da (that is, larger than a ribosome).

Activity of enzymes in the 2-keto-acid dehydrogenase family is normally low or limited under anaerobic conditions in E. coli. Increased production of NADH (or NADPH) could lead to a redox-imbalance, and NADH itself serves as an inhibitor to enzyme function. Engineering efforts have increased the anaerobic activity of the E. coli pyruvate dehydrogenase complex (Kim et al. Appl. Environ. Microbiol. 73:1766-1771 (2007); Kim et al. J. Bacteriol. 190:3851-3858) 2008); Zhou et al. Biotechnol. Lett. 30:335-342 (2008)). For example, the inhibitory effect of NADH can be overcome by engineering an H322Y mutation in the E3 component (Kim et al. J. Bacteriol. 190:3851-3858 (2008)). Structural studies of individual components and how they work together in complex provide insight into the catalytic mechanisms and architecture of enzymes in this family (Aevarsson et al. Nat. Struct. Biol. 6:785-792 (1999); Zhou et al. Proc. Natl. Acad. Sci. U.S.A. 98:14802-14807 (2001)). The substrate specificity of the dehydrogenase complexes varies in different organisms, but generally branched-chain keto-acid dehydrogenases have the broadest substrate range.

Alpha-ketoglutarate dehydrogenase (AKGD) converts alpha-ketoglutarate to succinyl-CoA and is the primary site of control of metabolic flux through the TCA cycle (Hansford, R. G. Curr. Top. Bioenerg. 10:217-278 (1980)). Encoded by genes sucA, sucB and lpd in E. coli, AKGD gene expression is downregulated under anaerobic conditions and during growth on glucose (Park et al. Mol. Microbiol. 15:473-482 (1995)). Although the substrate range of AKGD is narrow, structural studies of the catalytic core of the E2 component pinpoint specific residues responsible for substrate specificity (Knapp et al. J. Mol. Biol. 280:655-668 (1998)). The Bacillus subtilis AKGD, encoded by odhAB (E1 and E2) and pdhD (E3, shared domain), is regulated at the transcriptional level and is dependent on the carbon source and growth phase of the organism (Resnekov et al. Mol. Gen. Genet. 234:285-296 (1992)). In yeast, the LPD1 gene encoding the E3 component is regulated at the transcriptional level by glucose (Roy and Dawes J. Gen. Microbiol. 133:925-933 (1987)). The E1 component, encoded by KGDJ, is also regulated by glucose and activated by the products of HAP2 and HAP3 (Repetto and Tzagoloff Mol. Cell Biol. 9:2695-2705 (1989)). The AKGD enzyme complex, inhibited by products NADH and succinyl-CoA, is well-studied in mammalian systems, as impaired function of has been linked to several neurological diseases (Tretter and dam-Vizi Philos. Trans. R. Soc. Lond B Biol. Sci. 360:2335-2345 (2005)).

Branched-chain 2-keto-acid dehydrogenase complex (BCKAD), also known as 2-oxoisovalerate dehydrogenase, participates in branched-chain amino acid degradation pathways, converting 2-keto acids derivatives of valine, leucine and isoleucine to their acyl-CoA derivatives and CO₂. The complex has been studied in many organisms including Bacillus subtilis (Wang et al. Eur. J. Biochem. 213:1091-1099 (1993)), Rattus norvegicus (Namba et al. J. Biol. Chem. 244:4437-4447 (1969)) and Pseudomonas putida (Sokatch J. Bacteriol. 148:647-652 (1981)). In Bacillus subtilis the enzyme is encoded by genes pdhD (E3 component), bfmBB (E2 component), bfmBAA and bfmBAB (E1 component) (Wang et al. Eur. J. Biochem. 213:1091-1099 (1993)). In mammals, the complex is regulated by phosphorylation by specific phosphatases and protein kinases. The complex has been studied in rat hepatocites (Chicco et al. J. Biol. Chem. 269:19427-19434 (1994)) and is encoded by genes Bckdha (E1 alpha), Bckdhb (E1 beta), Dbt (E2), and Dld (E3). The E1 and E3 components of the Pseudomonas putida BCKAD complex have been crystallized (Aevarsson et al. Nat. Struct. Biol. 6:785-792 (1999); Mattevi Science 255:1544-1550 (1992)) and the enzyme complex has been studied (Sokatch et al. J. Bacteriol. 148:647-652 (1981)). Transcription of the P. putida BCKAD genes is activated by the gene product of bkdR (Hester et al. Eur. J. Biochem. 233:828-836 (1995)). In some organisms including Rattus norvegicus (Paxton et al. Biochem. J. 234:295-303 (1986)) and Saccharomyces cerevisiae (Sinclair et al. Biochem. Mol. Biol. Int. 31:911-922 (1993)), this complex has been shown to have a broad substrate range that includes linear oxo-acids such as 2-oxobutanoate and alpha-ketoglutarate, in addition to the branched-chain amino acid precursors. The active site of the bovine BCKAD was engineered to favor alternate substrate acetyl-CoA (Meng and Chuang, Biochemistry 33:12879-12885 (1994)).

The pyruvate dehydrogenase complex, catalyzing the conversion of pyruvate to acetyl-CoA, has also been extensively studied. In the E. coli enzyme, specific residues in the E1 component are responsible for substrate specificity (Bisswanger, H. J Biol. Chem. 256:815-822 (1981); Bremer, J. Eur. J Biochem. 8:535-540 (1969); Gong et al. J Biol. Chem. 275:13645-13653 (2000)). As mentioned previously, enzyme engineering efforts have improved the E. coli PDH enzyme activity under anaerobic conditions (Kim et al. Appl. Environ. Microbiol. 73:1766-1771 (2007); Kim J. Bacteriol. 190:3851-3858 (2008); Zhou et al. Biotechnol. Lett. 30:335-342 (2008)). In contrast to the E. coli PDH, the B. subtilis complex is active and required for growth under anaerobic conditions (Nakano J. Bacteriol. 179:6749-6755 (1997)). The Klebsiella pneumoniae PDH, characterized during growth on glycerol, is also active under anaerobic conditions (Menzel et al. J. Biotechnol. 56:135-142 (1997)). Crystal structures of the enzyme complex from bovine kidney (Zhou et al. Proc. Natl. Acad. Sci. U.S.A. 98:14802-14807 (2001)) and the E2 catalytic domain from Azotobacter vinelandii are available (Mattevi et al. Science 255:1544-1550 (1992)). Some mammalian PDH enzymes complexes can react on alternate substrates such as 2-oxobutanoate, although comparative kinetics of Rattus norvegicus PDH and BCKAD indicate that BCKAD has higher activity on 2-oxobutanoate as a substrate (Paxton et al. Biochem. J. 234:295-303 (1986)).

As an alternative to the large multienzyme 2-keto-acid dehydrogenase complexes described above, some anaerobic organisms utilize enzymes in the 2-ketoacid oxidoreductase family (OFOR) to catalyze acylating oxidative decarboxylation of 2-keto-acids. Unlike the dehydrogenase complexes, these enzymes contain iron-sulfur clusters, utilize different cofactors, and use ferredoxin or flavodixin as electron acceptors in lieu of NAD(P)H. While most enzymes in this family are specific to pyruvate as a substrate (POR) some 2-keto-acid:ferredoxin oxidoreductases have been shown to accept a broad range of 2-ketoacids as substrates including alpha-ketoglutarate and 2-oxobutanoate (Fukuda and Wakagi Biochim. Biophys. Acta 1597:74-80 (2002); Zhang et al. J. Biochem. 120:587-599 (1996)). One such enzyme is the OFOR from the thermoacidophilic archaeon Sulfolobus tokodaii 7, which contains an alpha and beta subunit encoded by gene ST2300 (Fukuda and Wakagi Biochim. Biophys. Acta 1597:74-80 (2002); Zhang et al. J. Biochem. 120:587-599 (1996)). A plasmid-based expression system has been developed for efficiently expressing this protein in E. coli (Fukuda et al. Eur. J. Biochem. 268:5639-5646 (2001)) and residues involved in substrate specificity were determined (Fukuda and Wakagi Biochim. Biophys. Acta 1597:74-80 (2002)). Two OFORs from Aeropyrum pernix str. K1 have also been recently cloned into E. coli, characterized, and found to react with a broad range of 2-oxoacids (Nishizawa et al. FEBS Lett. 579:2319-2322 (2005)). The gene sequences of these OFOR candidates are available, although they do not have GenBank identifiers assigned to date. There is bioinformatic evidence that similar enzymes are present in all archaea, some anaerobic bacteria and amitochondrial eukarya (Fukuda and Wakagi Biochim. Biophys. Acta 1597:74-80 (2005)). This class of enzyme is also interesting from an energetic standpoint, as reduced ferredoxin could be used to generate NADH by ferredoxin-NAD reductase (Petitdemange et al. Biochim. Biophys. Acta 421:334-337 (1976)). Also, since most of the enzymes are designed to operate under anaerobic conditions, less enzyme engineering may be required relative to enzymes in the 2-keto-acid dehydrogenase complex family for activity in an anaerobic environment.

Exemplary enzymes in this class include glyceraldehyde 3-phosphate dehydrogenase which converts glyceraldehyde-3-phosphate into D-glycerate 1,3-bisphosphate (for example, E. coli gapA (Branlant and Branlant Eur. J. Biochem. 150:61-66 (1985)), aspartate-semialdehyde dehydrogenase which converts L-aspartate-4-semialdehyde into L-4-aspartyl-phosphate (for example, E. coli asd (Biellmann et al. Eur. J. Biochem. 104:53-58 (1980)), N-acetyl-gamma-glutamyl-phosphate reductase which converts N-acetyl-L-glutamate-5-semialdehyde into N-acetyl-L-glutamyl-5-phosphate (for example, E. coli argC (Parsot et al. Gene 68:275-283 (1988)), and glutamate-5-semialdehyde dehydrogenase which converts L-glutamate-5-semialdehyde into L-glutamyl-5-phospate (for example, E. coli proA (Smith et al. J. Bacteriol. 157:545-551 (1984)).

An exemplary enoyl-CoA reductase is the gene product of bcd from C. acetobutylicum (Atsumi et al. Metab Eng (2007); Boynton et al. Journal of Bacteriology 178:3015-3024 (1996), which naturally catalyzes the reduction of crotonyl-CoA to butyryl-CoA. Activity of this enzyme can be enhanced by expressing bcd in conjunction with expression of the C. acetobutylicum etfAB genes, which encode an electron transfer flavoprotein. An additional candidate for the enoyl-CoA reductase step is the mitochondrial enoyl-CoA reductase from E. gracilis (Hoffmeister et al. Journal of Biological Chemistry 280:4329-4338 (2005)). A construct derived from this sequence following the removal of its mitochondrial targeting leader sequence was cloned in E. coli resulting in an active enzyme (Hoffmeister et al., supra, (2005)). This approach is well known to those skilled in the art of expressing eukarytotic genes, particularly those with leader sequences that may target the gene product to a specific intracellular compartment, in prokaryotic organisms. A close homolog of this gene, TDE0597, from the prokaryote Treponema denticola represents a third enoyl-CoA reductase which has been cloned and expressed in E. coli (Tucci and Martin FEBS Letters 581:1561-1566 (2007)).

Exemplary 2-enoate reductase (EC 1.3.1.31) enzymes are known to catalyze the NADH-dependent reduction of a wide variety of α,β-unsaturated carboxylic acids and aldehydes (Rohdich et al. J. Biol. Chem. 276:5779-5787 (2001)). 2-Enoate reductase is encoded by enr in several species of Clostridia (Giesel and Simon Arch Microbiol. 135(1): p. 51-57 (2001) including C. tyrobutyricum, and C. thermoaceticum (now called Moorella thermoaceticum) (Rohdich et al., supra, (2001)). In the recently published genome sequence of C. kluyveri, 9 coding sequences for enoate reductases have been reported, out of which one has been characterized (Seedorf et al. Proc Natl Acad Sci U.S.A. 105(6):2128-33 (2008)). The enr genes from both C. tyrobutyricum and C. thermoaceticum have been cloned and sequenced and show 59% identity to each other. The former gene is also found to have approximately 75% similarity to the characterized gene in C. kluyveri (Giesel and Simon Arch Microbiol 135(1):51-57 (1983)). It has been reported based on these sequence results that enr is very similar to the dienoyl CoA reductase in E. coli (fadH) (163 Rohdich et al., supra (2001)). The C. thermoaceticum enr gene has also been expressed in an enzymatically active form in E. coli (163 Rohdich et al., supra (2001)).

Most oxidoreductases operating on amino acids catalyze the oxidative deamination of alpha-amino acids with NAD+ or NADP+ as acceptor. Exemplary oxidoreductases operating on amino acids include glutamate dehydrogenase (deaminating), encoded by gdhA, leucine dehydrogenase (deaminating), encoded by ldh, and aspartate dehydrogenase (deaminating), encoded by nadX. The gdhA gene product from Escherichia coli (Korber et al. J. Mol. Biol. 234:1270-1273 (1993); McPherson and Wootton Nucleic. Acids Res. 11:5257-5266 (1983)), gdh from Thermotoga maritima (Kort et al. Extremophiles 1:52-60 (1997); Lebbink, et al. J. Mol. Biol. 280:287-296 (1998)); Lebbink et al. J. Mol. Biol. 289:357-369 (1999)), and gdhAl from Halobacterium salinarum (Ingoldsby et al. Gene 349:237-244 (2005)) catalyze the reversible interconversion of glutamate to 2-oxoglutarate and ammonia, while favoring NADP(H), NAD(H), or both, respectively. The ldh gene of Bacillus cereus encodes the LeuDH protein that has a wide of range of substrates including leucine, isoleucine, valine, and 2-aminobutanoate (Ansorge and Kula Biotechnol Bioeng. 68:557-562 (2000); Stoyan et al. J. Biotechnol 54:77-80 (1997)). The nadX gene from Thermotoga maritime encoding for the aspartate dehydrogenase is involved in the biosynthesis of NAD (Yang et al. J. Biol. Chem. 278:8804-8808 (2003)).

The lysine 6-dehydrogenase (deaminating), encoded by lysDH gene, catalyze the oxidative deamination of the ε-amino group of L-lysine to form 2-aminoadipate-6-semialdehyde, which in turn nonenzymatically cyclizes to form Δ1-piperideine-6-carboxylate (Misono and Nagasaki J. Bacteriol. 150:398-401 (1982)). The lysDH gene from Geobacillus stearothermophilus encodes a thermophilic NAD-dependent lysine 6-dehydrogenase (Heydari et al. Appl Environ. Microbiol 70:937-942 (2004)). In addition, the lysDH gene from Aeropyrum pernix K1 is identified through homology from genome projects.

Exemplary phosphate transferring acyltransferases include phosphotransacetylase, encoded by pta, and phosphotransbutyrylase, encoded by ptb. The pta gene from E. coli encodes an enzyme that can convert acetyl-CoA into acetyl-phosphate, and vice versa (Suzuki, T. Biochim. Biophys. Acta 191:559-569 (1969)). This enzyme can also utilize propionyl-CoA instead of acetyl-CoA forming propionate in the process (Hesslinger et al. Mol. Microbiol 27:477-492 (1998)). Similarly, the ptb gene from C. acetobutylicum encodes an enzyme that can convert butyryl-CoA into butyryl-phosphate (Walter et al. Gene 134(1): p. 107-11 (1993)); Huang et al. J Mol Microbiol Biotechnol 2(1): p. 33-38 (2000). Additional ptb genes can be found in butyrate-producing bacterium L2-50 (Louis et al. J. Bacteriol. 186:2099-2106 (2004)) and Bacillus megaterium (Vazquez et al. Curr. Microbiol 42:345-349 (2001)).

Aspartate aminotransferase transfers an amino group from aspartate to alpha-ketoglutarate, forming glutamate and oxaloacetate. This conversion is catalyzed by, for example, the gene products of aspC from Escherichia coli (Yagi et al. FEBS Lett. 100:81-84 (1979); Yagi et al. Methods Enzymol. 113:83-89 (1985)), AAT2 from Saccharomyces cerevisiae (Yagi et al. J Biochem. 92:35-43 (1982)) and ASPS from Arabidopsis thaliana (48, 108, 225 48. de la et al. Plant J 46:414-425 (2006); Kwok and Hanson J Exp. Bot. 55:595-604 (2004); Wilkie and Warren Protein Expr. Purif. 12:381-389 (1998)). Valine aminotransferase catalyzes the conversion of valine and pyruvate to 2-ketoisovalerate and alanine. The E. coli gene, avtA, encodes one such enzyme (Whalen and Berg J. Bacteriol. 150:739-746 (1982)). This gene product also catalyzes the amination of α-ketobutyrate to generate α-aminobutyrate, although the amine donor in this reaction has not been identified (Whalen and Berg J. Bacteriol. 158:571-574 (1984)). The gene product of the E. coli serC catalyzes two reactions, phosphoserine aminotransferase and phosphohydroxythreonine aminotransferase (Lam and Winkler J. Bacteriol. 172:6518-6528 (1990)), and activity on non-phosphorylated substrates could not be detected (Drewke et al. FEBS. Lett. 390:179-182 (1996)).

Cargill has developed a beta-alanine/alpha-ketoglutarate aminotransferase for producing 3-HP from beta-alanine via malonyl-semialdehyde (PCT/US2007/076252 (Jessen et al)). The gene product of SkPYD4 in Saccharomyces kluyveri was also shown to preferentially use beta-alanine as the amino group donor (Andersen et al. FEBS. J. 274:1804-1817 (2007)). SkUGA1 encodes a homologue of Saccharomyces cerevisiae GABA aminotransferase, UGA1 (Ramos et al. Eur. J. Biochem. 149:401-404 (1985)), whereas SkPYD4 encodes an enzyme involved in both β-alanine and GABA transamination (Andersen et al. FEBS. J. 274:1804-1817 (2007)). 3-Amino-2-methylpropionate transaminase catalyzes the transformation from methylmalonate semialdehyde to 3-amino-2-methylpropionate. The enzyme has been characterized in Rattus norvegicus and Sus scrofa and is encoded by Abat (Kakimoto et al. Biochim. Biophys. Acta 156:374-380 (1968); Tamaki et al. Methods Enzymol. 324:376-389 (2000)). Enzyme candidates in other organisms with high sequence homology to 3-amino-2-methylpropionate transaminase include Gta-1 in C. elegans and gabT in Bacillus subtilus. Additionally, one of the native GABA aminotransferases in E. coli, encoded by gene gabT, has been shown to have broad substrate specificity (Liu et al. Biochemistry 43:10896-10905 (2004); Schulz et al. Appl Environ Microbiol 56:1-6 (1990)). The gene product of puuE catalyzes the other 4-aminobutyrate transaminase in E. coli (Kurihara et al. J. Biol. Chem. 280:4602-4608 (2005).).

The X-ray crystal structures of E. coli 4-aminobutyrate transaminase unbound and bound to the inhibitor were reported (Liu et al. Biochemistry 43:10896-10905 (2004)). The substrates binding and substrate specificities were studied and suggested. The roles of active site residues were studied by site-directed mutagenesis and X-ray crystallography (Liu et al. Biochemistry 44:2982-2992 (2005)). Based on the structural information, attempt was made to engineer E. coli 4-aminobutyrate transaminase with novel enzymatic activity. These studies provide a base for evolving transaminase activity for BDO pathways.

2.7.2.a—Phosphotransferase, Carboxyl Group Acceptor

Exemplary kinases include the E. coli acetate kinase, encoded by ackA (Skarstedt and Silverstein J. Biol. Chem. 251:6775-6783 (1976)), the C. acetobutylicum butyrate kinases, encoded by buk1 and buk2 (Walter et al. Gene 134(1):107-111 (1993) (Huang et al. J Mol Microbiol Biotechnol 2(1):33-38 (2000)], and the E. coli gamma-glutamyl kinase, encoded by proB (Smith et al. J. Bacteriol. 157:545-551 (1984)). These enzymes phosphorylate acetate, butyrate, and glutamate, respectively. The ackA gene product from E. coli also phosphorylates propionate (Hesslinger et al. Mol. Microbiol 27:477-492 (1998)).

In the CoA-transferase family, E. coli enzyme acyl-CoA:acetate-CoA transferase, also known as acetate-CoA transferase (EC 2.8.3.8), has been shown to transfer the CoA moiety to acetate from a variety of branched and linear acyl-CoA substrates, including isobutyrate (Matthies and Schink Appl Environ Microbiol 58:1435-1439 (1992)), valerate (Vanderwinkel et al. Biochem. Biophys. Res Commun. 33:902-908 (1968)) and butanoate (Vanderwinkel, supra (1968)). This enzyme is encoded by atoA (alpha subunit) and atoD (beta subunit) in E. coli sp. K12 (Korolev et al. Acta Crystallogr. D Biol Crystallogr. 58:2116-2121 (2002); Vanderwinkel, supra (1968)) and actA and cg0592 in Corynebacterium glutamicum ATCC 13032 (Duncan et al. Appl Environ Microbiol 68:5186-5190 (2002)). Additional genes found by sequence homology include atoD and atoA in Escherichia coli UT189.

Similar transformations are catalyzed by the gene products of cat1, cat2, and cat3 of Clostridium kluyveri which have been shown to exhibit succinyl-CoA, 4-hydroxybutyryl-CoA, and butyryl-CoA acetyltransferase activity, respectively (Seedorf et al. Proc Natl Acad Sci U.S.A. 105(6):2128-2133 (2008); Sohling and Gottschalk J Bacteriol 178(3):871-880 (1996)].

The glutaconate-CoA-transferase (EC 2.8.3.12) enzyme from anaerobic bacterium Acidaminococcus fermentans reacts with diacid glutaconyl-CoA and 3-butenoyl-CoA (Mack and Buckel FEES Lett. 405:209-212 (1997)). The genes encoding this enzyme are gctA and gctB. This enzyme has reduced but detectable activity with other CoA derivatives including glutaryl-CoA, 2-hydroxyglutaryl-CoA, adipyl-CoA and acrylyl-CoA (Buckel et al. Eur. J. Biochem. 118:315-321 (1981)). The enzyme has been cloned and expressed in E. coli (Mac et al. Eur. J. Biochem. 226:41-51 (1994)).

In the CoA hydrolase family, the enzyme 3-hydroxyisobutyryl-CoA hydrolase is specific for 3-HIBCoA and has been described to efficiently catalyze the desired transformation during valine degradation (Shimomura et al. J Biol Chem 269:14248-14253 (1994)). Genes encoding this enzyme include hibch of Rattus norvegicus (Shimomura et al., supra (1994); Shimomura et al. Methods Enzymol. 324:229-240 (2000) and Homo sapiens (Shimomura et al., supra, 2000). Candidate genes by sequence homology include hibch of Saccharomyces cerevisiae and BC 2292 of Bacillus cereus.

The conversion of adipyl-CoA to adipate can be carried out by an acyl-CoA hydrolase or equivalently a thioesterase. The top E. coli gene candidate is tesB (Naggert et al. J Biol. Chem. 266(17):11044-11050 (1991)] which shows high similarity to the human acot8 which is a dicarboxylic acid acetyltransferase with activity on adipyl-CoA (Westin et al. J Biol Chem 280(46): 38125-38132 (2005). This activity has also been characterized in the rat liver (Deana, Biochem Int. 26(4): p. 767-773 (1992)).

Other potential E. coli thiolester hydrolases include the gene products of tesA (Bonner and Bloch, J Biol. Chem. 247(10):3123-3133 (1972)), ybgC (Kuznetsova et al., FEMS Microbiol Rev. 29(2):263-279 (2005); Zhuang et al., FEBS Lett. 516(1-3):161-163 (2002)) paaI (Song et al., J Biol. Chem. 281(16):11028-11038 (2006)), and ybdB (Leduc et al., J Bacteriol. 189(19):7112-7126 (2007)).

Several eukaryotic acetyl-CoA hydrolases (EC 3.1.2.1) have broad substrate specificity. The enzyme from Rattus norvegicus brain (Robinson et al. Biochem. Biophys. Res. Commun. 71:959-965 (1976)) can react with butyryl-CoA, hexanoyl-CoA and malonyl-CoA.

An exemplary carboxy-lyase is acetolactate decarboxylase which participates in citrate catabolism and branched-chain amino acid biosynthesis, converting 2-acetolactate to acetoin. In Lactococcus lactis the enzyme is composed of six subunits, encoded by gene aldB, and is activated by valine, leucine and isoleucine (Goupil et al. Appl. Environ. Microbiol. 62:2636-2640 (1996); Goupil-Feuillerat et al. J. Bacteriol. 182:5399-5408 (2000)). This enzyme has been overexpressed and characterized in E. coli (Phalip et al. FEBS Lett. 351:95-99 (1994)). In other organisms the enzyme is a dimer, encoded by aldC in Streptococcus thermophilus (Monnet et al. Lett. Appl. Microbiol. 36:399-405 (2003)), aldB in Bacillus brevis (Diderichsen et al. J. Bacteriol. 172:4315-4321 (1990); Najmudin et al. Acta Crystallogr. D. Biol. Crystallogr. 59:1073-1075 (2003)) and budA from Enterobacter aerogenes (Diderichsen et al. J. Bacteriol. 172:4315-4321 (1990)). The enzyme from Bacillus brevis was cloned and overexpressed in Bacillus subtilis and characterized crystallographically (Najmudin et al. Acta Crystallogr. D. Biol. Crystallogr. 59:1073-1075 (2003)). Additionally, the enzyme from Leuconostoc lactis has been purified and characterized but the gene has not been isolated (O'Sullivan et al. FEMS Microbiol. Lett. 194:245-249 (2001)).

Aconitate decarboxylase catalyzes the final step in itaconate biosynthesis in a strain of Candida and also in the filamentous fungus Aspergillus terreus (Bonnarme et al. J Bacteriol. 177:3573-3578 (1995); Willke and Vorlop Appl Microbiol Biotechnol 56:289-295 (2001)). Although itaconate is a compound of biotechnological interest, the aconitate decarboxylase gene or protein sequence has not been reported to date.

4-oxalocronate decarboxylase has been isolated from numerous organisms and characterized. Genes encoding this enzyme include dmpH and dmpE in Pseudomonas sp. (strain 600) (Shingler et al. J. Bacteriol. 174:711-724 (1992)), xylII and xylIII from Pseudomonas putida (Kato and Asano Arch. Microbiol 168:457-463 (1997); Lian and Whitman J. Am. Chem. Soc. 116:10403-10411 (1994); Stanley et al. Biochemistry 39:3514 (2000)) and Reut B5691 and Reut B5692 from Ralstonia eutropha JMP134 (Hughes et al. J. Bacteriol. 158:79-83 (1984)). The genes encoding the enzyme from Pseudomonas sp. (strain 600) have been cloned and expressed in E. coli (Shingler et al. J. Bacteriol. 174:711-724 (1992)).

An additional class of decarboxylases has been characterized that catalyze the conversion of cinnamate (phenylacrylate) and substituted cinnamate derivatives to the corresponding styrene derivatives. These enzymes are common in a variety of organisms and specific genes encoding these enzymes that have been cloned and expressed in E. coli are: pad 1 from Saccharomyces cerevisae (Clausen et al. Gene 142:107-112 (1994)), pdc from Lactobacillus plantarum (Barthelmebs et al. Appl Environ Microbiol 67:1063-1069 (2001); Qi et al. Metab Eng 9:268-276 (2007); Rodriguez et al. J. Agric. Food Chem. 56:3068-3072 (2008)), pofK (pad) from Klebsiella oxytoca (Hashidoko et al. Biosci. Biotech. Biochem. 58:217-218 (1994); Uchiyama et al. Biosci. Biotechnol. Biochem. 72:116-123 (2008)), Pedicoccus pentosaceus (Barthelmebs et al. Appl Environ Microbiol 67:1063-1069 (2001)), and padC from Bacillus subtilis and Bacillus pumilus (Lingen et al. Protein Eng 15:585-593 (2002)). A ferulic acid decarboxylase from Pseudomonas fluorescens also has been purified and characterized (Huang et al. J. Bacteriol. 176:5912-5918 (1994)). Importantly, this class of enzymes have been shown to be stable and do not require either exogenous or internally bound co-factors, thus making these enzymes ideally suitable for biotransformations (Sariaslani, Annu. Rev. Microbiol. 61:51-69 (2007)).

Additional decarboxylase enzymes can form succinic semialdehyde from alpha-ketoglutarate. These include the alpha-ketoglutarate decarboxylase enzymes from Euglena gracilis (Shigeoka et al. Biochem. J. 282(Pt 2):319-323 (1992); Shigeoka and Nakano Arch. Biochem. Biophys. 288:22-28 (1991); Shigeoka and Nakano Biochem. J. 292 (Pt 2):463-467 (1993)), whose corresponding gene sequence has yet to be determined, and from Mycobacterium tuberculosis (Tian et al. Proc Natl Acad Sci U.S.A. 102:10670-10675 (2005)). In addition, glutamate decarboxylase enzymes can convert glutamate into 4-aminobutyrate such as the products of the E. coli gadA and gadB genes (De Biase et al. Protein. Expr. Purif. 8:430-438 (1993)).

Pyruvate decarboxylase (PDC, EC 4.1.1.1), also termed keto-acid decarboxylase, is a key enzyme in alcoholic fermentation, catalyzing the decarboxylation of pyruvate to acetaldehyde. This enzyme has a broad substrate range for aliphatic 2-keto acids including 2-ketobutyrate, 2-ketovalerate, 3-hydroxypyruvate and 2-phenylpyruvate (Berg et al. Science 318:1782-1786 (2007)). The PDC from Zymomonas mobilus, encoded by pdc, has been a subject of directed engineering studies that altered the affinity for different substrates (Siegert et al. Protein Eng Des Sel 18:345-357 (2005)). The PDC from Saccharomyces cerevisiae has also been extensively studied, engineered for altered activity, and functionally expressed in E. coli (Killenberg-Jabs et al. Eur. J. Biochem. 268:1698-1704 (2001); L1 and Jordan Biochemistry 38:10004-10012 (1999); ter Schure et al. Appl. Environ. Microbiol. 64:1303-1307 (1998)). The crystal structure of this enzyme is available (Killenberg-Jabs Eur. J. Biochem. 268:1698-1704 (2001)). Other well-characterized PDC candidates include the enzymes from Acetobacter pasteurians (Chandra et al. Arch. Microbiol. 176:443-451 (2001)) and Kluyveromyces lactis (Krieger et al. Eur. J. Biochem. 269:3256-3263 (2002)).

Like PDC, benzoylformate decarboxylase (EC 4.1.1.7) has a broad substrate range and has been the target of enzyme engineering studies. The enzyme from Pseudomonas putida has been extensively studied and crystal structures of this enzyme are available (Hasson et al. Biochemistry 37:9918-9930 (1998); Polovnikova et al. Biochemistry 42:1820-1830 (2003)). Site-directed mutagenesis of two residues in the active site of the Pseudomonas putida enzyme altered the affinity (Km) of naturally and non-naturally occurring substrates (Siegert Protein Eng Des Sel 18:345-357 (2005)). The properties of this enzyme have been further modified by directed engineering (Lingen et al. Protein Eng 15:585-593 (2002)); Lingen Chembiochem 4:721-726 (2003)). The enzyme from Pseudomonas aeruginosa, encoded by mdlC, has also been characterized experimentally (Barrowman et al. FEMS Microbiology Letters 34:57-60 (1986)). Additional gene candidates from Pseudomonas stutzeri, Pseudomonas fluorescens and other organisms can be inferred by sequence homology or identified using a growth selection system developed in Pseudomonas putida (Henning et al. Appl. Environ. Microbiol. 72:7510-7517 (2006)).

The 2-(hydroxymethyl)glutarate dehydratase of Eubacterium barkeri is an exemplary hydro-lyase. This enzyme has been studied in the context of nicotinate catabolism and is encoded by hmd (Alhapel et al. Proc Natl Acad Sci USA 103:12341-12346 (2006)). Similar enzymes with high sequence homology are found in Bacteroides capillosus, Anaerotruncus colihominis, and Natranaerobius thermophilius.

A second exemplary hydro-lyase is fumarate hydratase, an enzyme catalyzing the dehydration of malate to fumarate. A wealth of structural information is available for this enzyme and researchers have successfully engineered the enzyme to alter activity, inhibition and localization (Weaver, T. Acta Crystallogr. D Biol Crystallogr. 61:1395-1401 (2005)). Additional fumarate hydratases include those encoded by fumC from Escherichia coli (Estevez et al. Protein Sci. 11:1552-1557 (2002); Hong and Lee Biotechnol. Bioprocess Eng. 9:252-255 (2004); Rose and Weaver Proc Natl Acad Sci U.S.A 101:3393-3397 (2004)), Campylobacter jejuni (Smith et al. Int. J. Biochem. Cell Biol 31:961-975 (1999)) and Thermus thermophilus (Mizobata et al. Arch. Biochem. Biophys. 355:49-55 (1998)), and fumH from Rattus norvegicus (Kobayashi et al. J. Biochem. 89:1923-1931 (1981)). Similar enzymes with high sequence homology include fuml from Arabidopsis thaliana and fumC from Corynebacterium glutamicum.

Citramalate hydrolyase, also called 2-methylmalate dehydratase, converts 2-methylmalate to mesaconate. 2-Methylmalate dehydratase activity was detected in Clostridium tetanomorphum, Morganella morganii, Citrobacter amalonaticus in the context of the glutamate degradation VI pathway (Kato and Asano Arch. Microbiol 168:457-463 (1997)); however the genes encoding this enzyme have not been sequenced to date.

The gene product of crt from C. acetobutylicum catalyzes the dehydration of 3-hydroxybutyryl-CoA to crotonyl-CoA (Atsumi et al. Metab Eng.; 29 (2007)); Boynton et al. Journal of Bacteriology 178:3015-3024 (1996)). The enoyl-CoA hydratases, phaA and phaB, of P. putida are believed to carry out the hydroxylation of double bonds during phenylacetate catabolism; (Olivera et al. Proc Natl Acad Sci USA 95(11):6419-6424 (1998)). The paaA and paaB from P. fluorescens catalyze analogous transformations (14 Olivera et al., supra, 1998). Lastly, a number of Escherichia coli genes have been shown to demonstrate enoyl-CoA hydratase functionality including maoC (Park and Lee J Bacteriol 185(18):5391-5397 (2003)), paaF (Park and Lee Biotechnol Bioeng. 86(6):681-686 (2004a)); Park and Lee Appl Biochem Biotechnol. 113-116: 335-346 (2004b)); Ismail et al. Eur J Biochem 270(14):p. 3047-3054 (2003), and paaG (Park and Lee, supra, 2004; Park and Lee supra, 2004b; Ismail et al., supra, 2003).

The E. coli genes fadA and fadB encode a multienzyme complex that exhibits ketoacyl-CoA thiolase, 3-hydroxyacyl-CoA dehydrogenase, and enoyl-CoA hydratase activities (Yang et al. Biochemistry 30(27): p. 6788-6795 (1991); Yang et al. J Biol Chem 265(18): p. 10424-10429 (1990); Yang et al. J Biol Chem 266(24): p. 16255 (1991); Nakahigashi and Inokuchi Nucleic Acids Res 18(16): p. 4937 (1990)). The fadI and fadJ genes encode similar functions and are naturally expressed only anaerobically (Campbell et al. Mol Microbiol 47(3): p. 793-805 (2003). A method for producing poly[(R)-3-hydroxybutyrate] in E. coli that involves activating fadB (by knocking out a negative regulator, fadR) and co-expressing a non-native ketothiolase (phaA from Ralstonia eutropha) has been described previously (Sato et al. J Biosci Bioeng 103(1): 38-44 (2007)). This work clearly demonstrates that a β-oxidation enzyme, in particular the gene product of fadB which encodes both 3-hydroxyacyl-CoA dehydrogenase and enoyl-CoA hydratase activities, can function as part of a pathway to produce longer chain molecules from acetyl-CoA precursors.

Aspartase (EC 4.3.1.1), catalyzing the deamination of aspartate to fumarate, is a widespread enzyme in microorganisms, and has been characterized extensively (Viola, R. E. Adv. Enzymol. Relat Areas Mol. Biol. 74:295-341 (2000)). The crystal structure of the E. coli aspartase, encoded by aspA, has been solved (Shi et al. Biochemistry 36:9136-9144 (1997)). The E. coli enzyme has also been shown to react with alternate substrates aspartatephenylmethylester, asparagine, benzyl-aspartate and malate (Ma et al. Ann N.Y. Acad Sci 672:60-65 (1992)). In a separate study, directed evolution was been employed on this enzyme to alter substrate specificity (Asano et al. Biomol. Eng 22:95-101 (2005)). Enzymes with aspartase functionality have also been characterized in Haemophilus influenzae (Sjostrom et al. Biochim. Biophys. Acta 1324:182-190 (1997)), Pseudomonas fluorescens (Takagi et al. J. Biochem. 96:545-552 (1984)), Bacillus subtilus (Sjostrom et al. Biochim. Biophys. Acta 1324:182-190 (1997)) and Serratia marcescens (Takagi and Kisumi J Bacteriol. 161:1-6 (1985)).

3-methylaspartase (EC 4.3.1.2), also known as beta-methylaspartase or 3-methylaspartate ammonia-lyase, catalyzes the deamination of threo-3-methylasparatate to mesaconate. The 3-methylaspartase from Clostridium tetanomorphum has been cloned, functionally expressed in E. coli, and crystallized (Asuncion et al. Acta Crystallogr. D Biol Crystallogr. 57:731-733 (2001); Asuncion et al. J Biol. Chem. 277:8306-8311 (2002); Botting et al. Biochemistry 27:2953-2955 (1988); Goda et al. Biochemistry 31:10747-10756 (1992). In Citrobacter amalonaticus, this enzyme is encoded by BAA28709 (Kato and Asano Arch. Microbiol 168:457-463 (1997)). 3-Methylaspartase has also been crystallized from E. coli YG1002 (Asano and Kato FEMS Microbiol Lett. 118:255-258 (1994)) although the protein sequence is not listed in public databases such as GenBank. Sequence homology can be used to identify additional candidate genes, including CTC 02563 in C. tetani and ECs0761 in Escherichia coli O157:H7.

Ammonia-lyase enzyme candidates that form enoyl-CoA products include beta-alanyl-CoA ammonia-lyase (EC 4.3.1.6), which deaminates beta-alanyl-CoA, and 3-aminobutyryl-CoA ammonia-lyase (EC 4.3.1.14). Two beta-alanyl-CoA ammonia lyases have been identified and characterized in Clostridium propionicum (Herrmann et al. FEBS J. 272:813-821 (2005)). No other beta-alanyl-CoA ammonia lyases have been studied to date, but gene candidates can be identified by sequence similarity. One such candidate is MXAN_—4385 in Myxococcus xanthus.

The 4-hydroxybutyryl-CoA dehydratases from both Clostridium aminobutyrium and C. kluyveri catalyze the reversible conversion of 4-hydroxybutyryl-CoA to crotonyl-CoA and posses an intrinsic vinylacetyl-CoA A-isomerase activity (Scherf and Buckel Eur. J Biochem. 215:421-429 (1993); Scherf et al. Arch. Microbiol 161:239-245 (1994)). Both native enzymes were purified and characterized, including the N-terminal amino acid sequences (Scherf and Buckel, supra, 1993; Scherf et al., supra, 1994). The abfD genes from C. aminobutyrium and C. kluyveri match exactly with these N-terminal amino acid sequences, thus are encoding the 4-hydroxybutyryl-CoA dehydratases/vinylacetyl-CoA A-isomerase. In addition, the abfD gene from Porphyromonas gingivalis ATCC 33277 is identified through homology from genome projects.

Lysine 2,3-aminomutase (EC 5.4.3.2) is an exemplary aminomutase that converts lysine to (3S)-3,6-diaminohexanoate, shifting an amine group from the 2- to the 3-position. The enzyme is found in bacteria that ferment lysine to acetate and butyrate, including as Fusobacterium nuleatum (kamA) (Barker et al. J. Bacteriol. 152:201-207 (1982)) and Clostridium subterminale (kamA) (Chirpich et al. J. Biol. Chem. 245:1778-1789 (1970)). The enzyme from Clostridium subterminale has been crystallized (Lepore et al. Proc. Natl. Acad. Sci. U.S.A 102:13819-13824 (2005)). An enzyme encoding this function is also encoded by yodO in Bacillus subtilus (Chen et al. Biochem. J. 348 Pt 3:539-549 (2000)). The enzyme utilizes pyridoxal 5′-phosphate as a cofactor, requires activation by S-Adenosylmethoionine, and is stereoselective, reacting with the only with L-lysine. The enzyme has not been shown to react with alternate substrates.

A second aminomutase, beta-lysine 5,6-aminomutase (EC 5.4.3.3), catalyzes the next step of lysine fermentation to acetate and butyrate, which transforms (3S)-3,6-diaminohexanoate to (3S,5S)-3,5-diaminohexanoate, shifting a terminal amine group from the 6- to the 5-position. This enzyme also catalyzes the conversion of lysine to 2,5-diaminohexanoate and is also called lysine-5,6-aminomutase (EC 5.4.3.4). The enzyme has been crystallized in Clostridium sticklandii (kamD, kamE) (Berkovitch et al. Proc. Natl. Acad. Sci. U.S.A 101:15870-15875 (2004)). The enzyme from Porphyromonas gingivalis has also been characterized (Tang et al. Biochemistry 41:8767-8776 (2002)).

Ornithine 4,5-aminomutase (EC 5.4.3.5) converts D-ornithine to 2,4-diaminopentanoate, also shifting a terminal amine to the adjacent carbon. The enzyme from Clostridium sticklandii is encoded by two genes, oraE and oraS, and has been cloned, sequenced and expressed in E. coli (Chen et al. J. Biol. Chem. 276:44744-44750 (2001)). This enzyme has not been characterized in other organisms to date.

Tyrosine 2,3-aminomutase (EC 5.4.3.6) participates in tyrosine biosynthesis, reversibly converting tyrosine to 3-amino-3-(4-hdyroxyphenyl)propanoate by shifting an amine from the 2- to the 3-position. In Streptomyces globisporus the enzyme has also been shown to react with tyrosine derivatives (Christenson et al. Biochemistry 42:12708-12718 (2003)). Sequence information is not available.

Leucine 2,3-aminomutase (EC 5.4.3.7) converts L-leucine to beta-leucine during leucine degradation and biosynthesis. An assay for leucine 2,3-aminomutase detected activity in many organisms (Poston, J. M. Methods Enzymol. 166:130-135 (1988)) but genes encoding the enzyme have not been identified to date.

Cargill has developed a novel 2,3-aminomutase enzyme to convert L-alanine to β-alanine, thus creating a pathway from pyruvate to 3-HP in four biochemical steps (Liao et al., U.S. Publication No. 2005-0221466).

6.2.1.a—Acid-Thiol Ligase

An exemplary acid-thiol ligase is the gene products of sucCD of E. coli which together catalyze the formation of succinyl-CoA from succinate with the concaminant consumption of one ATP, a reaction which is reversible in vivo (Buck et al. Biochemistry 24(22): p. 6245-6252 (1985)). Additional exemplary CoA-ligases include the rat dicarboxylate-CoA ligase for which the sequence is yet uncharacterized (Vamecq et al. Biochem J. 230(3): p. 683-693 (1985)), either of the two characterized phenylacetate-CoA ligases from P. chrysogenum (Lamas-Maceiras et al. Biochem J 395(1):147-155 (2006); Wang et al. Biochem Biophys Res Commun, 360(2):453-458 (2007)), the phenylacetate-CoA ligase from Pseudomonas putida (Martinez-Blanco et al. J Biol. Chem. 265(12):7084-7090 (1990)), and the 6-carboxyhexanoate-CoA ligase from Bacillus subtilis (Bower et al. J Bacteriol 178(14):4122-4130 (1996)).

This example describes exemplary BDO pathways from succinyl-CoA.

BDO pathways from succinyl-CoA are described herein and have been described previously (see U.S. application Ser. No. 12/049,256, filed Mar. 14, 2008, and PCT application serial No. US08/57168, filed Mar. 14, 2008, each of which is incorporated herein by reference). Additional pathways are shown in FIG. 8A. Enzymes of such exemplary BDO pathways are listed in Table 15, along with exemplary genes encoding these enzymes.

Briefly, succinyl-CoA can be converted to succinic semialdehyde by succinyl-CoA reductase (or succinate semialdehyde dehydrogenase) (EC 1.2.1.b). Succinate semialdehyde can be converted to 4-hydroxybutyrate by 4-hydroxybutyrate dehydrogenase (EC 1.1.1.a), as previously described. Alternatively, succinyl-CoA can be converted to 4-hydroxybutyrate by succinyl-CoA reductase (alcohol forming) (EC 1.1.1.c). 4-Hydroxybutyrate can be converted to 4-hydroxybutyryl-CoA by 4-hydroxybutyryl-CoA transferase (EC 2.8.3.a), as previously described, or by 4-hydroxybutyryl-CoA hydrolase (EC 3.1.2.a) or 4-hydroxybutyryl-CoA ligase (or 4-hydroxybutyryl-CoA synthetase) (EC 6.2.1.a). Alternatively, 4-hydroxybutyrate can be converted to 4-hydroxybutyryl-phosphate by 4-hydroxybutyrate kinase (EC 2.7.2.a), as previously described. 4-Hydroxybutyryl-phosphate can be converted to 4-hydroxybutyryl-CoA by phosphotrans-4-hydroxybutyrylase (EC 2.3.1.a), as previously described. Alternatively, 4-hydroxybutyryl-phosphate can be converted to 4-hydroxybutanal by 4-hydroxybutanal dehydrogenase (phosphorylating) (EC 1.2.1.d). 4-Hydroxybutyryl-CoA can be converted to 4-hydroxybutanal by 4-hydroxybutyryl-CoA reductase (or 4-hydroxybutanal dehydrogenase) (EC 1.2.1.b). Alternatively, 4-hydroxybutyryl-CoA can be converted to 1,4-butanediol by 4-hydroxybutyryl-CoA reductase (alcohol forming) (EC 1.1.1.c). 4-Hydroxybutanal can be converted to 1,4-butanediol by 1,4-butanediol dehydrogenase (EC 1.1.1.a), as previously described.