WO2014102180A1 - Propanol production by lactobacillus bacterial hosts - Google Patents

Abstract

Provided herein are recombinant Lactobacillus reuteri host cells comprising a heterologous polynucleotide encoding a methylglyoxal synthase, optionally with a heterologous polynucleotides encoding a triosephosphate isomerase. Also described are methods of using the recombinant Lactobacillus reuteri host cells to produce propanol and propylene.

Description

PROPANOL PRODUCTION BY LACTOBACILLUS BACTERIAL HOSTS

Cross -Reference to Related Applications

This application claims priority from U.S. provisional application Serial No. 61/746,512, filed, December 27, 2012, the content of which is fully incorporated herein by reference.

Reference to a Sequence Listing

This application contains a Sequence Listing in computer readable form, which is incorporated herein by reference.

Background

Concerns related to future supply of oil have prompted research in the area of renewable energy and renewable sources of other raw materials. Biofuels, such as ethanol and bioplastics (e.g., particularly polylactic acid) are examples of products that can be made directly from agricultural sources using microorganisms. Additional desired products may then be derived using non-enzymatic chemical conversions, e.g., dehydration of ethanol to ethylene.

Polymerization of ethylene provides polyethylene, a type of plastic with a wide range of useful applications. Ethylene is traditionally produced by refined non-renewable fossil fuels, but dehydration of biologically-derived ethanol to ethylene offers an alternative route to ethylene from renewable carbon sources, i.e., ethanol from fermentation of fermentable sugars. This process has been utilized for the production of "Green Polyethylene" that - save for minute differences in the carbon isotope distribution - is identical to polyethylene produced from oil.

Similarly, isopropanol and n-propanol can be dehydrated to propylene, which in turn can be polymerized to polypropylene. As with polyethylene, using biologically-derived starting material (i.e., isopropanol or n-propanol) would result in "Green Polypropylene." However, unlike polyethylene, the production of the polypropylene starting material from renewable sources has proved challenging. Proposed efforts at propanol production have been reported in WO 2009/049274, WO 2009/103026, WO 2009/131286, WO 2010/071697, WO 201 1/031897, WO 201 1/029166, WO 201 1/022651 , WO 2012/058603. It is clear that successful biological production of propanol for a chosen metabolic pathway in a particular host requires careful selection of heterologous genes in view of the host cell's endogenous gene activity.

WO201 1/022651 proposes n-propanol production using recombinant Clostridium and Thermoanaerobacterium bacteria, since these hosts contain native genes for propanediol production (an n-propanol precursor). However, it is not clear which other host platforms contain one or more of the required enzymatic activities for the production of n- propanol using the methylglyoxal pathway, nor is it clear which additional enzymatic activities are required to produce n-propanol via methylglyoxal in other hosts.

Lactobacillus host cells, such as Lactobacillus reuteri, are capable of tolerating propanol at concentrations of as much as 60 g/L. However, lack of knowledge of the Lactobacillus metabolic network and availability of Lactobacillus genetic tools is significantly lower compared to other well-known metabolic hosts, making metabolic engineering of a propanol pathway in Lactobacillus less appealing.

It would be advantageous in the art to improve n-propanol production, as a result of genetic engineering using recombinant DNA techniques.

Summary

The Applicant has surprisingly found that Lactobacillus reuteri contains all the required native enzymatic activities for producing n-propanol from methylglyoxal. Additionally, the Applicant has found that overexpression of a heterologous methylglyoxal synthase gene alone in the Lactobacillus reuteri host cell is sufficient to significantly enhance n-propanol production from a carbohydrate source.

Accordingly, in one aspect is a recombinant Lactobacillus reuteri host cell comprising a heterologous polynucleotide encoding a methylglyoxal synthase, wherein the cell is capable of producing n-propanol. In some embodiments, the host cell is capable of producing a greater amount of n-propanol when consisting only of the heterologous polynucleotide encoding the methylglyoxal synthase (i.e. without any additional heterologous genes), compared to the cell without the heterologous polynucleotide encoding the methylglyoxal synthase.

The Applicant has also surprisingly found that overexpression of triosephosphate isomerase (TPI) in recombinant Lactobacillus reuteri host cells comprising a methylglyoxal synthase can result in even greater production of n-propanol, despite previous reports that teach TPI gene deletion, not overexpression, to improve carbon flux to dihydroxyacetone phosphate (see Jung et al., J. Microbiol. Biotechnol. 2008, 18, 1797-1802). Accordingly, in some aspects, the recombinant host cell comprises a heterologous polynucleotide encoding a triosephosphate isomerase. In some aspects, the recombinant host cell comprises a disruption to an endogenous adhE gene and/or a disruption to an endogenous pduP gene.

In some aspects, the recombinant host cell comprises an active isopropanol pathway.

Also described are methods of producing n-propanol, comprising: (a) cultivating a recombinant Lactobacillus reuteri host cell described herein in a medium under suitable conditions to produce n-propanol; and (b) recovering the n-propanol. In some aspects, the method also produces isopropanol when the host cell comprises an active isopropanol pathway.

Also described are methods of producing propylene, comprising: (a) cultivating the recombinant Lactobacillus reuteri host cell comprising a heterologous polynucleotide encoding a methylglyoxal synthase described herein in a medium under suitable conditions to produce n-propanol; (b) recovering the n-propanol; (c) dehydrating the n-propanol under suitable conditions to produce propylene; and (d) recovering the propylene.

Brief Description of the Figures

Figure 1 shows a metabolic pathway for the production of n-propanol from glucose. Figure 2 shows a metabolic pathway for the coproduction of n-propanol and isopropanol from glucose.

Figure 3 shows a plasmid map for pSJ10600.

Figure 4 shows a plasmid map for pSJ10603.

Figure 5 shows n-propanol production from fermentation of Lactobacillus reuteri strain BKq579.

Figure 6 shows n-propanol production from fermentation of Lactobacillus reuteri strain BKq577.

Figure 7 shows n-propanol and isopropanol coproduction from fermentation of Lactobacillus reuteri strain BKq776.

Figure 8 shows n-propanol and isopropanol coproduction from fermentation of Lactobacillus reuteri strain BKq776.

Figure 9 shows a plasmid map for pJP042.

Figure 10 shows a plasmid map for pBKQ464.

Figure 1 1 shows a plasmid map for pBKQ466.

Figure 12 shows a plasmid map for pBKQ476.

Figure 13 shows a plasmid map for pBKQ478.

Figure 14 shows a plasmid map for pBKQ535. Figure 15 shows a plasmid map for pBKQ557.

Figure 16 shows a plasmid map for pBKQ559.

Figure 17 shows a plasmid map for pBKQ568.

Figure 18 shows a plasmid map for pBKQ574.

Figure 19 shows a plasmid map for pBKQ643.

Figure 20 shows a plasmid map for pBKQ729.

Figure 21 shows a plasmid map for pBKQ731 .

Figure 22 shows n-propanol production from fermentation of Lactobacillus reuteri strains TRgu1283 and TRgu1321 .

Figure 23 shows n-propanol and isopropanol coproduction from fermentation of Lactobacillus reuteri strain TRgu1378.

Figure 24 shows a plasmid map for pTRGU 1200.

Figure 25 shows a plasmid map for pSJ1 1503.

Figure 26 shows a plasmid map for pBKQ726.

Figure 27 shows a plasmid map for pTRGU 1279.

Figure 28 shows a plasmid map for pTOID55.

Figure 29 shows a plasmid map for pTOID81 .

Figure 30 shows a plasmid map for pTOID79.

Definitions

Methylglyoxal synthase: The term "methylglyoxal synthase" is defined herein as an enzyme that catalyzes the reaction of dihydroxyacetone phosphate (DHAP) to methylglyoxal and phosphate (e.g., 4.2.3.3). The methylglyoxal synthase may be monomeric or in the form of a protein complex comprising two or more subunits under cellular conditions. Methylglyoxal synthase activity may be determined from cell-free extracts as described in the art, e.g., Marks et al., 2004, Biochemistry, 43:3802-3813, or as described in the examples section below. For example, methylglyoxal synthase activity can be assayed by spectrophotometrically monitoring the increase in absorbance at 240 nm when 0.7mM DHAP in 50 mM imidazole (pH 7.0) is converted by methylglyoxal synthase at 25°C to produce methylglyoxal that forms a thiohemiacetal with glutathione (15 mM) which is isomerized by glyoxalase I (2 units) to form (S)-D-lactoylglutathione (absorbs light at 240 nm).

Triosephosphate isomerase: The term "triosephosphate isomerase" (TPI) is defined herein as an enzyme that catalyzes the interconversion of dihydroxyacetone phosphate (DHAP) and glyceraldehyde 3-phosphate (e.g., 5.3.1 .1 ). The triosephosphate isomerase may be monomeric or in the form of a protein complex comprising two or more subunits under cellular conditions. Triosephosphate isomerase activity may be determined from cell-free extracts as described in the art, e.g., Garza-Ramos et al., 1996, Eur J Biochem, 241 :1 14-120. For example, triosephosphate isomerase activity can be assayed by spectrophotometrically monitoring the decrease in absorbance of NADH at 340 nm with a coupled assay at 25°C in a reaction mixture of 100 mM triethanolamine, 10 mM EDTA, 1 mM glyceraldehyde 3-phosphate, glycerol-3-phosphate dehydrogenase (20 pg/ml) and 0.2 mM NADH in a final volume of 1 ml (pH 7.4).

Acetate Kinase: The term "acetate kinase" is defined herein as a transferase enzyme that catalyzes the chemical reaction of acetyl-phosphate and ADP to acetate and ATP (e.g., EC 2.7.2.1 ). The acetate kinase may be monomeric or in the form of a protein complex comprising two or more subunits under cellular conditions. Acetate kinase activity may be determined from cell-free extracts as described in the art, e.g., as described in S. Mukhopadhyay et al., 2008, Bioorg Chem. 36: 65-69.

Active isopropanol pathway: As used herein, a host cell having an "active isopropanol pathway" produces active enzymes necessary to catalyze each reaction in a metabolic pathway from a fermentable sugar to isopropanol, and therefore is capable of producing isopropanol in measurable yields when cultivated under fermentation conditions in the presence of at least one fermentable sugar. A host cell having an active isopropanol pathway comprises one or more isopropanol pathway genes. An "isopropanol pathway gene" as used herein refers to a gene that encodes an enzyme involved in an active isopropanol pathway.

The active enzymes necessary to catalyze each reaction in an active isopropanol pathway may result from activities of endogenous gene expression, activities of heterologous gene expression, or from a combination of activities of endogenous and heterologous gene expression, as described in more detail herein.

Thiolase: The term "thiolase" is defined herein as an acetyltransferase that catalyzes the chemical reaction of two molecules of acetyl-CoA to acetoacetyl-CoA and CoA (e.g., EC 2.3.1 .9). The thiolase may be monomeric or in the form of a protein complex comprising two or more subunits under cellular conditions. Thiolase activity may be determined from cell-free extracts as described in the art, e.g., as described in D. P. Wiesenborn et al., 1988, Appl. Environ. Microbiol. 54:2717-2722. For example, thiolase activity may be measured spectrophotometrically by monitoring the condensation reaction coupled to the oxidation of NADH using 3-hydroxyacyl-CoA dehydrogenase in 100 mM Tris hydrochloride (pH 7.4), 1 .0 mM acetyl-CoA, 0.2 mM NADH, 1 mM dithiothreitol, and 2 U of 3-hydroxyacyl-CoA dehydrogenase. After equilibration of the cuvette contents at 30°C for 2 min, the reaction is initiated by the addition of about 125 ng of thiolase in 10 μΙ_. The absorbance decrease at 340 nm due to oxidation of NADH is measured, and an extinction coefficient of 6.22 mM^"1 cm^"1 used.

CoA-transferase: As used herein, the term "CoA-transferase" is defined as any enzyme that catalyzes the removal of coenzyme A from acetoacetyl-CoA to generate acetoacetate. The CoA-transferase may be monomeric or in the form of a protein complex comprising two or more subunits under cellular conditions. In some aspects, the CoA- transferase is an acetoacetyl-CoA:acetate/butyrate transferase (e.g., a butyrate- acetoacetate CoA transferase of EC 2.8.3.9) that converts acetoacetyl-CoA and acetate to acetoacetate and acetyl-CoA and/or converts acetoacetyl-CoA and butyrate to acetoacetate and butyryl-CoA. In some aspects, the CoA-transferase is an acetoacetyl-CoA hydrolase of EC 3.1 .2.1 1 . In some aspects, the CoA-transferase is a succinyl- CoA:acetoacetate transferase of EC 2.8.3.5 that converts acetoacetyl-CoA and succinate to acetoacetate and succinyl-CoA. It is known in the art that some CoA-transferase enzymes can react on multiple substrates.

HMG-CoA synthase (3-hydroxymethylglutaryl CoA synthase): The term "HMG- CoA synthase" or "3-hydroxymethylglutaryl CoA synthase" is defined herein as an enzyme that catalyzes the condensation of acetyl-CoA with acetoacetyl-CoA to form 3-hydroxy-3- methylglutaryl-CoA (HMG-CoA) (e.g., 2.3.3.10). The HMG-CoA synthase may be monomeric or in the form of a protein complex comprising two or more subunits under cellular conditions. HMG-CoA synthase activity may be determined from cell-free extracts as described in the art, e.g., as described in in Quant et al., 1989, Biochem J., 262:159- 164. For example, HMG-CoA synthase activity can be assayed by spectrophotometrically measuring the disappearance of the enol form of acetoacetyl-CoA by monitoring the change of absorbance at 303 nm in 50 mm-Tris/HCI, 10 mM-MgCI₂ and 0.2 mM- dithiothreitol pH 8.0 at 30°C.

HMG-CoA lyase (3-hydroxymethylglutaryl CoA lyase): The term "3- hydroxymethylglutaryl CoA lyase" or "HMG-CoA lyase" is defined herein as an enzyme that catalyzes the chemical reaction of 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) to Acetyl- CoA and Acetoacetate (e.g., 4.1 .3.4). The HMG-CoA lyase may be monomeric or in the form of a protein complex comprising two or more subunits under cellular conditions. HMG- CoA lyase activity may be determined from cell-free extracts using a citrate synthase- coupled assay as described in the art, e.g., as described in Stegink and Coon, 1968, J. Biol. Chem., 243: 5272-5279, in which the acetyl-coA, produced along with acetoacetate upon the cleavage of HMG-CoA, is coupled to the citrate synthase assay of Ochoa et al, 1951 , J. Biol. C em., 193: 691 -702.

Acetoacetate decarboxylase: The term "acetoacetate decarboxylase" is defined herein as an enzyme that catalyzes the chemical reaction of acetoacetate to carbon dioxide and acetone (e.g., EC 4.1 .1 .4). The acetoacetate decarboxylase may be monomeric or in the form of a protein complex comprising two or more subunits under cellular conditions. Acetoacetate decarboxylase activity may be determined from cell-free extracts as described in the art, e.g., as described in D.J. Petersen, et al., 1990, Appl. Environ. Microbiol. 56, 3491 -3498. For example, acetoacetate decarboxylase activity may be measured spectrophotometrically by monitoring the depletion of acetoacetate at 270 nm in 5 nM acetoacetate, 0.1 M K₂P0₄, pH 5.9 at 26°C.

Isopropanol dehydrogenase: The term "isopropanol dehydrogenase" is defined herein as any suitable oxidoreductase that catalyzes the reduction of acetone to isopropanol (e.g., any suitable enzyme of EC1 .1 .1 .1 or EC 1 .1 .1 .80). The isopropanol dehydrogenase may be monomeric or in the form of a protein complex comprising two or more subunits under cellular conditions. Acetoacetate decarboxylase activity may be determined spectrophotometrically from cell-free extracts as described in the art, e.g., by decrease in absorbance at 340 nm in an assay containing 200 μΜ NADPH and 10 mM acetone in 25 mM potassium phosphate, pH 7.2 at 25°C.

Disruption: The term "disruption" means that a coding region and/or control sequence of a referenced gene is partially or entirely modified (such as by deletion, insertion, and/or substitution of one or more nucleotides) resulting in the absence (inactivation) or decrease in expression, and/or the absence or decrease of enzyme activity of the encoded polypeptide. The effects of disruption can be measured using techniques known in the art such as detecting the absence or decrease of enzyme activity using from cell-free extract measurements referenced herein; or by the absence or decrease of corresponding mRNA (e.g., at least 25% decrease, at least 50% decrease, at least 60% decrease, at least 70% decrease, at least 80% decrease, or at least 90% decrease); the absence or decrease in the amount of corresponding polypeptide having enzyme activity (e.g., at least 25% decrease, at least 50% decrease, at least 60% decrease, at least 70% decrease, at least 80% decrease, or at least 90% decrease); or the absence or decrease of the specific activity of the corresponding polypeptide having enzyme activity (e.g., at least 25% decrease, at least 50% decrease, at least 60% decrease, at least 70% decrease, at least 80% decrease, or at least 90% decrease). Disruptions of a particular gene of interest can be generated by methods known in the art, e.g., by directed homologous recombination (see Methods in Yeast Genetics (1997 edition), Adams, Gottschling, Kaiser, and Stems, Cold Spring Harbor Press (1998)).

Coding sequence: The term "coding sequence" or "coding region" means a polynucleotide sequence, which specifies the amino acid sequence of a polypeptide. The boundaries of the coding sequence are generally determined by an open reading frame, which usually begins with the ATG start codon or alternative start codons such as GTG and TTG and ends with a stop codon such as TAA, TAG, and TGA. The coding sequence may be a sequence of genomic DNA, cDNA, a synthetic polynucleotide, and/or a recombinant polynucleotide.

Sequence Identity: The relatedness between two amino acid sequences or between two nucleotide sequences is described by the parameter "sequence identity".

For purposes described herein, the degree of sequence identity between two amino acid sequences is determined using the Needleman-Wunsch algorithm (Needleman and Wunsch, J. Mol. Biol. 1970, 48, 443-453) as implemented in the Needle program of the EMBOSS package (EMBOSS: The European Molecular Biology Open Software Suite, Rice et al., Trends Genet 2000, 16, 276-277), preferably version 3.0.0 or later. The optional parameters used are gap open penalty of 10, gap extension penalty of 0.5, and the EBLOSUM62 (EMBOSS version of BLOSUM62) substitution matrix. The output of Needle labeled "longest identity" (obtained using the -nobrief option) is used as the percent identity and is calculated as follows:

(Identical Residues x 100)/(Length of the Referenced Sequence - Total Number of Gaps in Alignment)

For purposes described herein, the degree of sequence identity between two deoxyribonucleotide sequences is determined using the Needleman-Wunsch algorithm (Needleman and Wunsch, 1970, supra) as implemented in the Needle program of the EMBOSS package (EMBOSS: The European Molecular Biology Open Software Suite, Rice et al., 2000, supra), preferably version 3.0.0 or later. The optional parameters used are gap open penalty of 10, gap extension penalty of 0.5, and the EDNAFULL (EMBOSS version of NCBI NUC4.4) substitution matrix. The output of Needle labeled "longest identity" (obtained using the -nobrief option) is used as the percent identity and is calculated as follows:

(Identical Deoxyribonucleotides x 100)/(Length of Referenced Sequence - Total Number of Gaps in Alignment)

Heterologous polynucleotide: The term "heterologous polynucleotide" is defined herein as a polynucleotide that is not native to the host cell; a native polynucleotide in which structural modifications have been made to the coding region; a native polynucleotide whose expression is quantitatively altered as a result of a manipulation of the DNA by recombinant DNA techniques, e.g., a different (foreign) promoter; or a native polynucleotide in a host cell having one or more extra copies of the native polynucleotide to quantitatively alter expression. A "heterologous gene" is a gene comprising a heterologous polynucleotide.

Endogenous gene: The term "endogenous gene" means a gene that is native to the referenced Lactobacillus cell. "Endogenous gene expression" means expression of an endogenous gene.

Nucleic acid construct: The term "nucleic acid construct" means a polynucleotide comprising one or more (e.g., two, several) control sequences. The polynucleotide may be single-stranded or double-stranded, and may be isolated from a naturally occurring gene, modified to contain segments of nucleic acids in a manner that would not otherwise exist in nature, or synthetic.

Control sequence: The term "control sequence" means a nucleic acid sequence necessary for polypeptide expression. Control sequences may be native or foreign to the polynucleotide encoding the polypeptide, and native or foreign to each other. Such control sequences include, but are not limited to, a leader sequence, polyadenylation sequence, propeptide sequence, promoter sequence, signal peptide sequence, and transcription terminator sequence. The control sequences may be provided with linkers for the purpose of introducing specific restriction sites facilitating ligation of the control sequences with the coding region of the polynucleotide encoding a polypeptide.

Operably linked: The term "operably linked" means a configuration in which a control sequence is placed at an appropriate position relative to the coding sequence of a polynucleotide such that the control sequence directs the expression of the coding sequence.

Expression: The term "expression" includes any step involved in the production of the polypeptide including, but not limited to, transcription, post-transcriptional modification, translation, post-translational modification, and secretion. Expression can be measured— for example, to detect increased expression— by techniques known in the art, such as measuring levels of mRNA and/or translated polypeptide.

Expression vector: The term "expression vector" means a linear or circular DNA molecule that comprises a polynucleotide encoding a polypeptide and is operably linked to control sequences, wherein the control sequences provide for expression of the polynucleotide encoding the polypeptide. At a minimum, the expression vector comprises a promoter sequence, and transcriptional and translational stop signal sequences. Host cell : The term "host cell" means any cell type that is susceptible to transformation, transfection, transduction, and the like with a nucleic acid construct or expression vector. The term "host cell" encompasses any progeny of a parent cell that is not identical to the parent cell due to mutations that occur during replication. The term "recombinant host cell" is defined herein as a non-naturally occurring host cell comprising one or more (e.g., two, several) heterologous polynucleotides.

Allelic variant: The term "allelic variant" means any of two or more alternative forms of a gene occupying the same chromosomal locus. Allelic variation arises naturally through mutation, and may result in polymorphism within populations. Gene mutations can be silent (no change in the encoded polypeptide) or may encode polypeptides having altered amino acid sequences. An allelic variant of a polypeptide is a polypeptide encoded by an allelic variant of a gene.

Volumetric productivity: The term "volumetric productivity" refers to the amount of referenced product produced (e.g., the amount of n-propanol produced) per volume of the system used (e.g., the total volume of media and contents therein) per unit of time.

Fermentable medium: The term "fermentable medium" or "fermentation medium" refers to a medium comprising one or more (e.g., several) sugars, such as glucose, fructose, sucrose, cellobiose, xylose, xylulose, arabinose, mannose, galactose, and/or soluble oligosaccharides, wherein the medium is capable, in part, of being converted (fermented) by a host cell into a desired product, such as n-propanol. In some instances, the fermentation medium is derived from a natural source, such as sugar cane, starch, or cellulose, and may be the result of pretreating the source by enzymatic hydrolysis (saccharification).

Sugar cane juice: The term "sugar cane juice" refers to the liquid extract from pressed Saccharum grass (sugarcane), such as pressed Saccharum officinarum or Saccharum robustom.

High stringency conditions: The term "high stringency conditions" means for probes of at least 100 nucleotides in length, prehybridization and hybridization at 42°C in 5X SSPE, 0.3% SDS, 200 micrograms/ml sheared and denatured salmon sperm DNA, and 50% formamide, following standard Southern blotting procedures for 12 to 24 hours. The carrier material is finally washed three times each for 15 minutes using 0.2X SSC, 0.2% SDS at 65°C.

Low stringency conditions: The term "low stringency conditions" means for probes of at least 100 nucleotides in length, prehybridization and hybridization at 42°C in 5X SSPE, 0.3% SDS, 200 micrograms/ml sheared and denatured salmon sperm DNA, and 25% formamide, following standard Southern blotting procedures for 12 to 24 hours. The carrier material is finally washed three times each for 15 minutes using 0.2X SSC, 0.2% SDS at 50°C.

Medium stringency conditions: The term "medium stringency conditions" means for probes of at least 100 nucleotides in length, prehybridization and hybridization at 42°C in 5X SSPE, 0.3% SDS, 200 micrograms/ml sheared and denatured salmon sperm DNA, and 35% formamide, following standard Southern blotting procedures for 12 to 24 hours. The carrier material is finally washed three times each for 15 minutes using 0.2X SSC, 0.2% SDS at 55°C.

Medium-high stringency conditions: The term "medium-high stringency conditions" means for probes of at least 100 nucleotides in length, prehybridization and hybridization at 42°C in 5X SSPE, 0.3% SDS, 200 micrograms/ml sheared and denatured salmon sperm DNA, and 35% formamide, following standard Southern blotting procedures for 12 to 24 hours. The carrier material is finally washed three times each for 15 minutes using 0.2X SSC, 0.2% SDS at 60°C.

Very high stringency conditions: The term "very high stringency conditions" means for probes of at least 100 nucleotides in length, prehybridization and hybridization at 42°C in 5X SSPE, 0.3% SDS, 200 micrograms/ml sheared and denatured salmon sperm DNA, and 50% formamide, following standard Southern blotting procedures for 12 to 24 hours. The carrier material is finally washed three times each for 15 minutes using 0.2X SSC, 0.2% SDS at 70°C.

Very low stringency conditions: The term "very low stringency conditions" means for probes of at least 100 nucleotides in length, prehybridization and hybridization at 42°C in 5X SSPE, 0.3% SDS, 200 micrograms/ml sheared and denatured salmon sperm DNA, and 25% formamide, following standard Southern blotting procedures for 12 to 24 hours. The carrier material is finally washed three times each for 15 minutes using 0.2X SSC, 0.2% SDS at 45°C.

Reference to "about" a value or parameter herein includes aspects that are directed to that value or parameter per se. For example, description referring to "about X" includes the aspect "X". When used in combination with measured values, "about" includes a range that encompasses at least the uncertainty associated with the method of measuring the particular value, and can include a range of plus or minus two standard deviations around the stated value.

As used herein and in the appended claims, the singular forms "a," "or," and "the" include plural referents unless the context clearly dictates otherwise. It is understood that the aspects described herein include "consisting" and/or "consisting essentially of" aspects.

Unless defined otherwise or clearly indicated by context, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art.

Detailed Description

Described herein, inter alia, are recombinant Lactobacillus host cells comprising a heterologous polynucleotide encoding a methylglyoxal synthase. As mentioned supra, the Applicant has surprisingly found that Lactobacillus reuteri contains all the required native enzymatic activities for producing n-propanol from methylglyoxal (see Figure 1 ) and that overexpression of a heterologous methylglyoxal synthase gene alone in the Lb. reuteri host cell is sufficient to significantly enhance n-propanol production from a carbohydrate source.

Accordingly, in some aspects, the Lactobacillus reuteri host cells lack an endogenous methylglyoxal synthase gene. In some aspects, the host cells produce a greater amount of n-propanol compared to the cells without the heterologous polynucleotide encoding a methylglyoxal synthase when cultivated under identical conditions. In some embodiments, the host cells produce (or are capable of producing) at least 10% more (e.g., at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 75%, at least 100%, or at least 200% at least 500%, at least 750%, at least 1000%, at least 2000%, at least 5000%, or at least 7500% more) n-propanol compared to the cells without the heterologous polynucleotide encoding the methylglyoxal synthase, when cultivated under identical conditions. In some aspects, the host cells are capable of producing a greater amount of n-propanol when consisting only of the heterologous polynucleotide encoding the methylglyoxal synthase, compared to cells without the heterologous polynucleotide encoding the methylglyoxal synthase.

The methylglyoxal synthase can be any methylglyoxal synthase that is suitable for the host cells and their methods of use described herein, such as a naturally occurring methylglyoxal synthase or a variant thereof that retains methylglyoxal synthase activity. In one aspect, the methylglyoxal synthase is present in the cytosol of the host cells.

In some aspects, the host cells comprising the heterologous polynucleotide that encodes a methylglyoxal synthase have an increased level of methylglyoxal synthase activity compared to the host cells without the heterologous polynucleotide that encodes the methylglyoxal synthase, when cultivated under the same conditions. In some aspects, the host cells have an increased level of methylglyoxal synthase activity of at least 5%, e.g., at least 10%, at least 15%, at least 20%, at least 25%, at least 50%, at least 100%, at least 150%, at least 200%, at least 300%, or at 500% compared to the host cells without the heterologous polynucleotide that encodes the methylglyoxal synthase, when cultivated under the same conditions.

Exemplary methylglyoxal synthases that may be used with the host cells and methods of use described herein include, but are not limited to, those shown in Table 1 .

Table 1 .

Additional polynucleotides encoding suitable methylglyoxal synthases may be obtained from microorganisms of any suitable genus, including those readily available within the UniProtKB database (www.uniprot.org).

Polynucleotides encoding the methylglyoxal synthase may be obtained from microorganisms of any genus. In one aspect, the methylglyoxal synthase may be a bacterial, a yeast, or a filamentous fungal methylglyoxal synthase obtained from the microorganisms described herein.

The methylglyoxal synthase may be a bacterial methylglyoxal synthase. For example, the methylglyoxal synthase may be a Gram-positive bacterial methylglyoxal synthase such as a Bacillus, Streptococcus, Streptomyces, Staphylococcus, Enterococcus, Lactobacillus, Lactococcus, Clostridium, Geobacillus, or Oceanobacillus methylglyoxal synthase, or a Gram-negative methylglyoxal synthase such as an E. coli, Pseudomonas, Salmonella, Campylobacter, Helicobacter, Flavobacterium, Fusobacterium, llyobacter, Neisseria, or Ureaplasma methylglyoxal synthase.

In some aspects, the methylglyoxal synthase is a Bacillus methylglyoxal synthase, such as the Bacillus subtilis methylglyoxal synthase of SEQ ID NO: 133 or 1 14, or the Bacillus licheniformis methylglyoxal synthase of SEQ ID NO: 1 15. In one aspect, the methylglyoxal synthase is a Bacillus alkalophilus, Bacillus amyloliquefaciens, Bacillus brevis, Bacillus circulans, Bacillus clausii, Bacillus coagulans, Bacillus firmus, Bacillus lautus, Bacillus lentus, Bacillus licheniformis, Bacillus megaterium, Bacillus pumilus, Bacillus stearothermophilus, Bacillus subtilis, or Bacillus thuringiensis methylglyoxal synthase.

In some aspects, the methylglyoxal synthase is a Lactobacillus methylglyoxal synthase, such as the Lactobacillus sakei methylglyoxal synthase of SEQ ID NO: 1 16, the Lactobacillus coryniformis methylglyoxal synthase of SEQ ID NO: 124, or the Lactobacillus curvatus methylglyoxal synthase of SEQ ID NO: 125.

In another aspect, the methylglyoxal synthase is a Streptococcus equisimilis, Streptococcus pyogenes, Streptococcus uberis, or Streptococcus equi subsp. Zooepidemicus methylglyoxal synthase. In another aspect, the methylglyoxal synthase is a Streptomyces achromogenes, Streptomyces avermitilis, Streptomyces coelicolor, Streptomyces griseus, or Streptomyces lividans methylglyoxal synthase.

The methylglyoxal synthase may be a fungal methylglyoxal synthase. In one aspect, the fungal methylglyoxal synthase is a yeast methylglyoxal synthase such as a Candida, Kluyveromyces, Pichia, Saccharomyces, Schizosaccharomyces, or Yarrowia methylglyoxal synthase.

In another aspect, the fungal methylglyoxal synthase is a filamentous fungal methylglyoxal synthase such as an Acremonium, Agaricus, Alternaria, Aspergillus, Aureobasidium, Botryospaeria, Ceriporiopsis, Chaetomidium, Chrysosporium, Claviceps, Cochliobolus, Coprinopsis, Coptotermes, Corynascus, Cryphonectria, Cryptococcus, Diplodia, Exidia, Filibasidium, Fusarium, Gibberella, Holomastigotoides, Humicola, Irpex, Lentinula, Leptospaeria, Magnaporthe, Melanocarpus, Meripilus, Mucor, Myceliophthora, Neocallimastix, Neurospora, Paecilomyces, Penicillium, Phanerochaete, Piromyces, Poitrasia, Pseudoplectania, Pseudotrichonympha, Rhizomucor, Schizophyllum, Scytalidium, Talaromyces, Thermoascus, Thielavia, Tolypocladium, Trichoderma, Trichophaea, Verticillium, Volvariella, or Xylaria methylglyoxal synthase.

In another aspect, the methylglyoxal synthase is a Saccharomyces carlsbergensis,

Saccharomyces cerevisiae, Saccharomyces diastaticus, Saccharomyces douglasii,

Saccharomyces kluyveri, Saccharomyces norbensis, or Saccharomyces oviformis methylglyoxal synthase.

In another aspect, the methylglyoxal synthase is an Acremonium cellulolyticus, Aspergillus aculeatus, Aspergillus awamori, Aspergillus flavus, Aspergillus fumigatus, Aspergillus foetidus, Aspergillus japonicus, Aspergillus nidulans, Aspergillus niger, Aspergillus oryzae, Aspergillus sojae, Chrysosporium keratinophilum, Chrysosporium lucknowense, Chrysosporium tropicum, Chrysosporium merdarium, Chrysosporium inops, Chrysosporium pannicola, Chrysosporium queenslandicum, Chrysosporium zonatum, Fusarium bactridioides, Fusarium cerealis, Fusarium crookwellense, Fusarium culmorum, Fusarium graminearum, Fusarium graminum, Fusarium heterosporum, Fusarium negundi, Fusarium oxysporum, Fusarium reticulatum, Fusarium roseum, Fusarium sambucinum, Fusarium sarcochroum, Fusarium sporotrichioides, Fusarium sulphureum, Fusarium torulosum, Fusarium trichothecioides, Fusarium venenatum, Humicola grisea, Humicola insolens, Humicola lanuginosa, Irpex lacteus, Mucor miehei, Myceliophthora thermophila, Neurospora crassa, Penicillium funiculosum, Penicillium purpurogenum, Phanerochaete chrysosporium, Thielavia achromatica, Thielavia albomyces, Thielavia albopilosa, Thielavia australeinsis, Thielavia fimeti, Thielavia microspora, Thielavia ovispora, Thielavia peruviana, Thielavia spededonium, Thielavia setosa, Thielavia subthermophila, Thielavia terrestris, Trichoderma harzianum, Trichoderma koningii, Trichoderma longibrachiatum, Trichoderma reesei, or Trichoderma viride methylglyoxal synthase.

It will be understood that for the aforementioned species, both the perfect and imperfect states, and other taxonomic equivalents, e.g., anamorphs, are encompassed regardless of the species name by which they are known. Those skilled in the art will readily recognize the identity of appropriate equivalents.

Strains of these species are readily accessible to the public in a number of culture collections, such as the American Type Culture Collection (ATCC), Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSM), Centraalbureau Voor Schimmelcultures (CBS), and Agricultural Research Service Patent Culture Collection, Northern Regional Research Center (NRRL).

The methylglyoxal synthase coding sequences, or subsequences thereof; as well as the corresponding amino acid sequence, or fragments thereof; may be used to design nucleic acid probes to identify and clone methylglyoxal synthases from strains of different genera or species according to methods well known in the art. In particular, such probes can be used for hybridization with the genomic or cDNA of the genus or species of interest, following standard Southern blotting procedures, in order to identify and isolate the corresponding gene therein. Such probes can be considerably shorter than the entire sequence, e.g., at least 14 nucleotides, at least 25 nucleotides, at least 35 nucleotides, at least 70 nucleotides in lengths. The probes may be longer, e.g., at least 100 nucleotides, at least 200 nucleotides, at least 300 nucleotides, at least 400 nucleotides, at least 500 nucleotides in lengths. Even longer probes may be used, e.g., at least 600 nucleotides, at least 700 nucleotides, at least 800 nucleotides, or at least 900 nucleotides in length. Both DNA and RNA probes can be used. The probes are typically labeled for detecting the corresponding gene (for example, with ³²P, ³H, ³⁵S, biotin, or avidin).

A genomic DNA or cDNA library prepared from such other strains may be screened for DNA that hybridizes with the probes described above and encodes a polypeptide having methylglyoxal synthase activity. Genomic or other DNA from such other strains may be separated by agarose or polyacrylamide gel electrophoresis, or other separation techniques. DNA from the libraries or the separated DNA may be transferred to and immobilized on nitrocellulose or other suitable carrier material. In order to identify a clone or DNA that is homologous with the methylglyoxal synthase coding sequences, or a subsequence thereof, the carrier material may be used in a Southern blot.

For purposes of the probes described above, hybridization indicates that the polynucleotide hybridizes to a labeled nucleic acid probe, or the full-length complementary strand thereof, or a subsequence of the foregoing; under very low to very high stringency conditions. Molecules to which the nucleic acid probe hybridizes under these conditions can be detected using, for example, X-ray film. For long probes of at least 100 nucleotides in length, very low to very high stringency and washing conditions are defined as described supra. For short probes of about 15 nucleotides to about 70 nucleotides in length, stringency and washing conditions are defined as described supra.

Methylglyoxal synthases may also be identified and obtained from other sources including microorganisms isolated from nature (e.g., soil, composts, water, silage, etc.) or DNA samples obtained directly from natural materials (e.g., soil, composts, water, silage, etc.) using the above-mentioned probes. Techniques for isolating microorganisms and DNA directly from natural habitats are well known in the art. The polynucleotide encoding a methylglyoxal synthase may then be derived by similarly screening a genomic or cDNA library of another microorganism or mixed DNA sample. Once a polynucleotide encoding a methylglyoxal synthase has been detected with a suitable probe as described herein, the sequence may be isolated or cloned by utilizing techniques that are known to those of ordinary skill in the art (see, e.g., J. Sambrook, E.F. Fritsch, and T. Maniatus, 1989, Molecular Cloning, A Laboratory Manual, 2d edition, Cold Spring Harbor, New York).

Techniques used to isolate or clone polynucleotides encoding methylglyoxal synthases include isolation from genomic DNA, preparation from cDNA, or a combination thereof. The cloning of the polynucleotides from such genomic DNA can be effected, e.g., by using the well-known polymerase chain reaction (PCR) or antibody screening of expression libraries to detect cloned DNA fragments with shares structural features. See, e.g., Innis et al., 1990, PCR: A Guide to Methods and Application, Academic Press, New York. Other nucleic acid amplification procedures such as ligase chain reaction (LCR), ligated activated transcription (LAT) and nucleotide sequence-based amplification (NASBA) may be used.

In one aspect, the methylglyoxal synthase has at least 50%, e.g., at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any methylglyoxal synthase described herein (e.g., any methylglyoxal synthase of SEQ ID NO: 1 12, 1 13, 1 14, 1 15, 1 16, 1 17, 1 18, 1 19, 120, 121 , 122, 123, 124, or 125). In one aspect, the methylglyoxal synthase sequence differs by no more than ten amino acids, e.g., by no more than five amino acids, by no more than four amino acids, by no more than three amino acids, by no more than two amino acids, or by one amino acid from any methylglyoxal synthase described herein (e.g., any methylglyoxal synthase of SEQ ID NO: 1 12, 1 13, 1 14, 1 15, 1 16, 1 17, 1 18, 1 19, 120, 121 , 122, 123, 124, or 125). In one aspect, the methylglyoxal synthase comprises or consists of the amino acid sequence of any methylglyoxal synthase described herein (e.g., any methylglyoxal synthase of SEQ ID NO: 1 12, 1 13, 1 14, 1 15, 1 16, 1 17, 1 18, 1 19, 120, 121 , 122, 123, 124, or 125), allelic variant, or a fragment thereof having methylglyoxal synthase activity. In one aspect, the methylglyoxal synthase has an amino acid substitution, deletion, and/or insertion of one or more (e.g., two, several) amino acids. In some aspects, the total number of amino acid substitutions, deletions and/or insertions is not more than 10, e.g., not more than 9, 8, 7, 6, 5, 4, 3, 2, or 1 .

The amino acid changes are generally of a minor nature, that is conservative amino acid substitutions or insertions that do not significantly affect the folding and/or activity of the protein; small deletions, typically of one to about 30 amino acids; small amino-terminal or carboxyl-terminal extensions, such as an amino-terminal methionine residue; a small linker peptide of up to about 20-25 residues; or a small extension that facilitates purification by changing net charge or another function, such as a poly-histidine tract, an antigenic epitope or a binding domain.

Examples of conservative substitutions are within the group of basic amino acids (arginine, lysine and histidine), acidic amino acids (glutamic acid and aspartic acid), polar amino acids (glutamine and asparagine), hydrophobic amino acids (leucine, isoleucine and valine), aromatic amino acids (phenylalanine, tryptophan and tyrosine), and small amino acids (glycine, alanine, serine, threonine and methionine). Amino acid substitutions that do not generally alter specific activity are known in the art and are described, for example, by H. Neurath and R.L. Hill, 1979, In, The Proteins, Academic Press, New York. The most commonly occurring exchanges are Ala/Ser, Val/lle, Asp/Glu, Thr/Ser, Ala/Gly, Ala/Thr, Ser/Asn, Ala/Val, Ser/Gly, Tyr/Phe, Ala/Pro, Lys/Arg, Asp/Asn, Leu/lle, LeuA al, Ala/Glu, and Asp/Gly.

Alternatively, the amino acid changes are of such a nature that the physico- chemical properties of the polypeptides are altered. For example, amino acid changes may improve the thermal stability of the methylglyoxal synthase, alter the substrate specificity, change the pH optimum, and the like.

Essential amino acids can be identified according to procedures known in the art, such as site-directed mutagenesis or alanine-scanning mutagenesis (Cunningham and Wells, 1989, Science 244: 1081 -1085). In the latter technique, single alanine mutations are introduced at every residue in the molecule, and the resultant mutant molecules are tested for methylglyoxal synthase activity to identify amino acid residues that are critical to the activity of the molecule. See also, Hilton et al., 1996, J. Biol. Chem. 271 : 4699-4708. The active site of the methylglyoxal synthase or other biological interaction can also be determined by physical analysis of structure, as determined by such techniques as nuclear magnetic resonance, crystallography, electron diffraction, or photoaffinity labeling, in conjunction with mutation of putative contact site amino acids. See, for example, de Vos et al., 1992, Science 255: 306-312; Smith et al., 1992, J. Mol. Biol. 224: 899-904; Wlodaver et al., 1992, FEBS Lett. 309: 59-64. The identities of essential amino acids can also be inferred from analysis of identities with other methylglyoxal synthases that are related to the referenced methylglyoxal synthase.

Single or multiple amino acid substitutions, deletions, and/or insertions can be made and tested using known methods of mutagenesis, recombination, and/or shuffling, followed by a relevant screening procedure, such as those disclosed by Reidhaar-Olson and Sauer, 1988, Science 241 : 53-57; Bowie and Sauer, 1989, Proc. Natl. Acad. Sci. USA 86: 2152- 2156; WO 95/17413; or WO 95/22625. Other methods that can be used include error- prone PCR, phage display (e.g., Lowman et al., 1991 , Biochemistry 30: 10832-10837; U.S. Patent No. 5,223,409; WO 92/06204), and region-directed mutagenesis (Derbyshire et al., 1986, Gene 46: 145; Ner et al., 1988, DNA 7: 127).

Mutagenesis/shuffling methods can be combined with high-throughput, automated screening methods to detect activity of cloned, mutagenized polypeptides expressed by host cells (Ness et al., 1999, Nature Biotechnology 17: 893-896). Mutagenized DNA molecules that encode active methylglyoxal synthases can be recovered from the host cells and rapidly sequenced using standard methods in the art. These methods allow the rapid determination of the importance of individual amino acid residues in a polypeptide.

In some aspects, the methylglyoxal synthase has at least 20%, e.g., at least 40%, at 5 least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% of the methylglyoxal synthase activity of any methylglyoxal synthase described herein (e.g., any methylglyoxal synthase of SEQ ID NO: 1 12, 1 13, 1 14, 1 15, 1 16, 1 17, 1 18, 1 19, 120, 121 , 122, 123, 124, or 125) under the same conditions.

i o In one aspect, the methylglyoxal synthase coding sequence hybridizes under at least low stringency conditions, e.g., medium stringency conditions, medium-high stringency conditions, high stringency conditions, or very high stringency conditions with the full-length complementary strand of the coding sequence from any methylglyoxal synthase described herein (e.g., any methylglyoxal synthase of SEQ ID NO: 1 12, 1 13, 1 14, 1 15, 1 16,

15 1 17, 1 18, 1 19, 120, 121 , 122, 123, 124, or 125) (see, e.g., J. Sambrook, E.F. Fritsch, and T. Maniatus, 1989, supra). In one aspect, the methylglyoxal synthase coding sequence has at least 65%, e.g., at least 70%, at least 75%, at least 80%, at least 85%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with the coding

20 sequence from any methylglyoxal synthase described herein (e.g., any methylglyoxal synthase of SEQ ID NO: 1 12, 1 13, 1 14, 1 15, 1 16, 1 17, 1 18, 1 19, 120, 121 , 122, 123, 124, or 125).

In one aspect, the heterologous polynucleotide encoding the methylglyoxal synthase comprises the coding sequence of any methylglyoxal synthase described herein

25 (e.g., any methylglyoxal synthase of SEQ ID NO: 1 12, 1 13, 1 14, 1 15, 1 16, 1 17, 1 18, 1 19, 120, 121 , 122, 123, 124, or 125). In one aspect, the heterologous polynucleotide encoding the methylglyoxal synthase comprises a subsequence of the coding sequence from any methylglyoxal synthase described herein, wherein the subsequence encodes a polypeptide having methylglyoxal synthase activity. In one aspect, the number of nucleotides residues

30 in the coding subsequence is at least 75%, e.g., at least 80%, 85%, 90%, or 95% of the number of the referenced coding sequence.

The referenced coding sequence of any related aspect described herein can be the native coding sequence (e.g., a sequence readily determined by the skilled artisan using available sequence databases) or a degenerate sequence, such as a codon-optimized

35 coding sequence designed for a Lactobacillus reuteri host cell. For example, the coding sequence for the Escherichia coli methylglyoxal synthase encoding SEQ ID NO: 1 12 can be the native Escherichia coli methylglyoxal synthase coding sequence or a codon- optimized version designed for Lactobacillus reuteri, such as the coding sequence listed in SEQ ID NO: 126. Likewise, the coding sequence for the Bacillus subtilis methylglyoxal synthase encoding SEQ ID NO: 1 13 can be the native respective methylglyoxal synthase coding sequences or codon-optimized version designed for Lactobacillus reuteri, such as the coding sequences listed in SEQ ID NO: 127. Furthermore, the native coding sequence can be modified to a degenerate sequence in order to remove certain unwanted sequence features, as known in the art (e.g., the native coding sequence of the Lactobacillus sakei methylglyoxal synthase encoding SEQ ID NO: 1 16 was modified to SEQ ID NO: 130 in order to remove certain undesired restriction sites).

The methylglyoxal synthase may be a fused polypeptide or cleavable fusion polypeptide in which another polypeptide is fused at the N-terminus or the C-terminus of the methylglyoxal synthase. A fused polypeptide may be produced by fusing a polynucleotide encoding another polypeptide to a polynucleotide encoding the methylglyoxal synthase. Techniques for producing fusion polypeptides are known in the art, and include ligating the coding sequences encoding the polypeptides so that they are in frame and that expression of the fused polypeptide is under control of the same promoter(s) and terminator. Fusion proteins may also be constructed using intein technology in which fusions are created post-translationally (Cooper et al., 1993, EMBO J. 12: 2575-2583; Dawson et ai, 1994, Science 266: 776-779).

The methylglyoxal synthase, and activities thereof, can be detected using methods known in the art. These detection methods may include use of specific antibodies, formation of an enzyme product, or disappearance of an enzyme substrate. See, 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); and Hanai et al., Appl. Environ. Microbiol. 73:7814-7818 (2007)).

Applicants have also surprisingly found that overexpression of triosephosphate isomerase in recombinant Lactobacillus reuteri host cells comprising a methylglyoxal synthase can result in even greater production of n-propanol. Without being bound by theory, the triosephosphate isomerase converts cellular glyceraldehyde 3-phosphate to dihydroxyacetone phosphate, which can then be utilized by methylglyoxal synthase for n- propanol production as describe supra.

Accordingly, in some aspects, the Lactobacillus reuteri host cells comprise both a heterologous polynucleotide encoding methylglyoxal synthase and a heterologous polynucleotide encoding a triosphosphate isomerase. In some embodiments, the recombinant host cells comprising both a heterologous polynucleotide encoding a triosphosphate isomerase and a heterologous polynucleotide encoding methylglyoxal synthase produce a greater amount of n-propanol compared to the cells without the heterologous polynucleotide encoding the triosphosphate isomerase when cultivated under identical conditions. In some embodiments, the host cells produce (or are capable of producing) at least 10% more (e.g., at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 75%, at least 100%, at least 200%, at least 500%, at least 750%, at least 1000%, at least 2000%, at least 5000%, or at least 7500% more) n-propanol compared to the cells without the heterologous polynucleotide encoding the triosphosphate isomerase, when cultivated under identical conditions.

In some aspects, the recombinant host cells lack an endogenous triosphosphate isomerase gene and/or have undetectable triosphosphate isomerase activity without the heterologous polynucleotide encoding the triosphosphate isomerase.

The triosphosphate isomerase can be any triosphosphate isomerase that is suitable for the host cells and their methods of use described herein, such as a naturally occurring triosphosphate isomerase (e.g., an endogenous triosphosphate isomerase or one from another species) or a variant thereof that retains triosphosphate isomerase activity. In one aspect, the triosphosphate isomerase is present in the cytosol of the host cells.

In some aspects, the host cells comprising a heterologous polynucleotide encoding a triosephosphate isomerase have an increased level of triosephosphate isomerase activity compared to the host cells without the heterologous polynucleotide encoding the triosephosphate isomerase, when cultivated under the same conditions. In some aspects, the host cells have an increased level of triosephosphate isomerase activity of at least 5%, e.g., at least 10%, at least 15%, at least 20%, at least 25%, at least 50%, at least 100%, at least 150%, at least 200%, at least 300%, or at 500% compared to the host cells without the heterologous polynucleotide encoding the triosephosphate isomerase, when cultivated under the same conditions.

Exemplary triosephosphate isomerase that can be used with the host cells and methods of use described herein include, but are not limited to, those triosephosphate isomerase shown in Table 2.

Table 2.

Organism Sequence Sequence Code SEQ ID NO Database

Lactobacillus reuteri 167

Lactobacillus plantarum UniProtKB Q88YH4 169

Lactobacillus reuteri UniProtKB Q0GL75 170

Lactobacillus reuteri UniProtKB A4L2T9 171

Lactobacillus sakei UniProtKB Q5NJZ4 172

Saccharomyces cerevisiae UniProtKB P00942 173

Escherichia coli UniProtKB P0A858 174

Bacillus subtilis UniProtKB P27876 175

Corynebacterium glutamicum UniProtKB P 19583 176

Schizosaccharomyces pombe UniProtKB P07669 177

Helicobacter pylori UniProtKB B5Z9W5 178

Additional polynucleotides encoding suitable triosephosphate isomerases may be obtained from microorganisms of any genus, including those readily available within the UniProtKB database (www.uniprot.org). In one aspect, the triosephosphate isomerase is a bacterial, a yeast, or a filamentous fungal triosephosphate isomerase, e.g., obtained from any of the microorganisms described supra under the section on methylglyoxal synthases.

The triosephosphate isomerase coding sequences can also be used to design nucleic acid probes to identify and clone DNA encoding triosephosphate isomerases from strains of different genera or species, as described supra.

The polynucleotides encoding the triosephosphate isomerase may also be identified and obtained from other sources including microorganisms isolated from nature (e.g., soil, composts, water, etc.) or DNA samples obtained directly from natural materials (e.g., soil, composts, water, etc.,) as described supra.

Techniques used to isolate or clone polynucleotides encoding triosephosphate isomerases are described supra.

In one aspect, the triosephosphate isomerase has at least 60%, e.g., at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any triosephosphate isomerase described herein (e.g., any triosephosphate isomerase of SEQ ID NO: 167, 169, 170, 171 , 172, 173, 174, 175, 176, 177, 178, or 201 ). In one aspect, the triosephosphate isomerase sequence differs by no more than ten amino acids, e.g., by no more than five amino acids, by no more than four amino acids, by no more than three amino acids, by no more than two amino acids, or by one amino acid from any triosephosphate isomerase described herein (e.g., any triosephosphate isomerase of SEQ ID NO: 167, 169, 170, 171 , 172, 173, 174,

175, 176, 177, 178, or 201 ). In one aspect, the triosephosphate isomerase comprises or consists of the amino acid sequence of any triosephosphate isomerase described herein

(e.g., any triosephosphate isomerase of SEQ ID NO: 167, 169, 170, 171 , 172, 173, 174, 175, 176, 177, 178, or 201 ), allelic variant, or a fragment thereof having triosephosphate isomerase activity. In one aspect, the triosephosphate isomerase has an amino acid substitution, deletion, and/or insertion of one or more (e.g., two, several) amino acids. In some aspects, the total number of amino acid substitutions, deletions and/or insertions is not more than 10, e.g., not more than 9, 8, 7, 6, 5, 4, 3, 2, or 1 .

In some aspects, the triosephosphate isomerase has at least 20%, e.g., at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% of the triosephosphate isomerase activity of any triosephosphate isomerase described herein (e.g., any triosephosphate isomerase of SEQ ID NO: 167, 169, 170, 171 , 172, 173, 174, 175, 176, 177, 178, or 201 ) under the same conditions.

In one aspect, the triosephosphate isomerase coding sequence hybridizes under at least low stringency conditions, e.g., medium stringency conditions, medium-high stringency conditions, high stringency conditions, or very high stringency conditions with the full-length complementary strand of the coding sequence from any triosephosphate isomerase described herein (e.g., any triosephosphate isomerase of SEQ ID NO: 167, 169, 170, 171 , 172, 173, 174, 175, 176, 177, 178, or 201 , such as the coding sequences shown in SEQ ID NO: 166, 168, 194, 195 or 200) (see, e.g., J. Sambrook, E.F. Fritsch, and T. Maniatus, 1989, supra). In one aspect, the triosephosphate isomerase coding sequence has at least 65%, e.g., at least 70%, at least 75%, at least 80%, at least 85%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with the coding sequence from any triosephosphate isomerase described herein (e.g., any triosephosphate isomerase of SEQ ID NO: 167, 169, 170, 171 , 172, 173, 174, 175, 176, 177, 178, or 201 , such as the coding sequences shown in SEQ ID NO: 166, 168, 194, 195, or 200).

In one aspect, the heterologous polynucleotide encoding the triosephosphate isomerase comprises the coding sequence of any triosephosphate isomerase described herein (e.g., any triosephosphate isomerase of SEQ ID NO: 167, 169, 170, 171 , 172, 173, 174, 175, 176, 177, 178, or 201 , such as the coding sequences shown in SEQ ID NO: 166, 168, 194, 195, or 200). In one aspect, the heterologous polynucleotide encoding the triosephosphate isomerase comprises a subsequence of the coding sequence from any triosephosphate isomerase described herein, wherein the subsequence encodes a polypeptide having triosephosphate isomerase activity. In one aspect, the number of nucleotides residues in the coding subsequence is at least 75%, e.g., at least 80%, 85%, 90%, or 95% of the number of the referenced coding sequence. The triosephosphate isomerases can also include fused polypeptides or cleavable fusion polypeptides, as described supra.

Active Isopropanol Pathway

5 The recombinant Lactobacillus reuteri host cells described herein can be used in the coproduction of n-propanol and isopropanol when further comprising an active isopropanol pathway. Isopropanol pathway genes and corresponding engineered Lactobacillus transformants for fermentation of isopropanol are known in the art (e.g., see PCT/US201 1/58405, the content of which is hereby incorporated in its entirety). One i o exemplary isopropanol pathway used in the coproduction of n-propanol and isopropanol from glucose is depicted in Figure 2, wherein cellular acetyl-CoA is converted to acetoacetyl-CoA by a thiolase; acetoacetyl-CoA is converted to acetoacetate by either a CoA-transferase or through HMG-CoA using an HMG-CoA synthase and an HMG-CoA lyase (e.g., see USSN: 61/727,876, filed November 19, 2013, the content of which is

15 hereby incorporated in its entirety); acetoacetate is converted to acetone by an acetoacetate decarboxylase; and acetone is converted to isopropanol by an isopropanol dehydrogenase. Any suitable isopropanol pathway gene, endogenous or heterologous, encoding a thiolase, CoA-transferase, HMG-CoA synthase, HMG-CoA lyase, acetoacetate decarboxylase, and/or isopropanol dehydrogenase, may be used to produce isopropanol.

20 Thus, the host cells comprising an active isopropanol pathway may comprise thiolase activity, CoA-transferase activity, HMG-CoA synthase activity, HMG-CoA lyase activity, acetoacetate decarboxylase activity and/or isopropanol dehydrogenase activity.

The recombinant Lactobacillus reuteri host cells may comprise any one or combination of a plurality of the heterologous isopropanol pathway genes described. For

25 example, in one aspect, the recombinant host cell comprises a heterologous thiolase gene, a heterologous CoA-transferase gene, a heterologous HMG-CoA synthase gene, a heterologous HMG-CoA lyase gene, a heterologous acetoacetate decarboxylase gene, and/or a heterologous isopropanol dehydrogenase gene described herein. In some aspects, the host cell produces (or is capable of producing) a greater amount of

30 isopropanol compared to the host cell without the heterologous polynucleotides when cultivated under the same conditions. In some of these aspects, the host cell lacks an endogenous thiolase gene, CoA-transferase gene, HMG-CoA synthase gene, HMG-CoA lyase gene, acetoacetate decarboxylase gene, and/or isopropanol dehydrogenase.

In one aspect, the recombinant Lactobacillus reuteri host cell comprises one or

35 more (e.g., two, several) heterologous polynucleotides encoding a thiolase described herein. In one aspect, the host cell comprises one or more heterologous polynucleotides encoding a CoA-transferase described herein. In one aspect, the host cell comprises one or more heterologous polynucleotides encoding a HMG-CoA synthase described herein. In one aspect, the host cell comprises one or more heterologous polynucleotides encoding a HMG-CoA lyase described herein. In one aspect, the host cell comprises one or more heterologous polynucleotides encoding an acetoacetate decarboxylase described herein. In one aspect, the host cell comprises one or more heterologous polynucleotides encoding an isopropanol dehydrogenase described herein.

Any suitable isopropanol pathway gene, endogenous or heterologous, may be used and expressed in sufficient amount to produce an enzyme involved in a selected active isopropanol pathway. 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 selected isopropanol pathway enzymatic activities taught herein is routine and well known in the art for a selected host, in light of the teaching from Applicants' earlier references cited herein (e.g., PCT/US201 1/58405, the content of which is hereby incorporated in its entirety). For example, suitable homologues, orthologs, paralogs and nonorthologous gene displacements of known genes, and the interchange of genetic alterations between organisms can be identified in related or distant host to a selected host.

Sequences for isopropanol pathway genes of interest (either as overexpression candidates or as insertion sites) can typically be obtained using techniques known in the art. Routine experimental design can be employed to test expression of various genes and activity of various enzymes, including genes and enzymes that function in an isopropanol pathway. Experiments may be conducted wherein each enzyme is expressed in the Lactobacillus host cell individually and in blocks of enzymes up to and including all pathway enzymes, to establish which are needed (or desired) for improved isopropanol production. One illustrative experimental design tests expression of each individual enzyme as well as of each unique pair of enzymes, and further can test expression of all required enzymes, or each unique combination of enzymes. A number of approaches can be taken, as will be appreciated.

The recombinant Lactobacillus host cells of the invention can be produced by introducing heterologous polynucleotides encoding one or more of the enzymes participating in an isopropanol pathway, as described below. As one in the art will appreciate, in some instances (e.g., depending on the selection of host) the heterologous expression of every gene shown in the isopropanol pathway may not be required for isopropanol production given that the host cell may have endogenous enzymatic activity from one or more pathway genes. For example, if the recombinant Lactobacillus reuteri host cell is deficient in one or more enzymes of an isopropanol pathway, then heterologous polynucleotides for the deficient enzyme(s) are introduced into the host for subsequent expression. Alternatively, if the host exhibits endogenous expression of some pathway genes, but is deficient in others, then an encoding polynucleotide is needed for the deficient enzyme(s) to achieve isopropanol biosynthesis. Thus, a recombinant host cell of the invention can be produced by introducing heterologous polynucleotides to obtain the enzyme activities of a desired biosynthetic pathway or a desired biosynthetic pathway can be obtained by introducing one or more heterologous polynucleotides that, together with one or more endogenous enzymes, produces a desired product such as isopropanol.

Depending on the isopropanol pathway constituents of the Lactobacillus reuteri host cell, the host cells of the invention will include at least one heterologous polynucleotide encoding a methylglyoxal synthase, and optionally include at least one heterologous polynucleotide encoding an enzyme of an isopropanol pathway gene and up to all encoding heterologous polynucleotides for the isopropanol pathway. For example, isopropanol biosynthesis can be established in a host deficient in an isopropanol pathway enzyme through heterologous expression of the corresponding polynucleotide. In a host deficient in all enzymes of an isopropanol pathway, heterologous expression of all enzymes in the pathway can be included, although it is understood that all enzymes of a pathway can be expressed even if the host contains at least one of the pathway enzymes.

The thiolase, CoA-transferase, HMG-CoA synthase, HMG-CoA lyase, acetoacetate decarboxylase, and isopropanol dehydrogenase, and activities thereof, can be detected using methods known in the art or as described herein. These detection methods may include use of specific antibodies, formation of an enzyme product, or disappearance of an enzyme substrate. See, 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); and Hanai et al., Appl. Environ. Microbiol. 73:7814-7818 (2007)).

Thiolases

In some aspects, the recombinant Lactobacillus reuteri host cells comprise a heterologous polynucleotide that encodes a thiolase. The thiolase can be any thiolase that is suitable for the host cells and their methods of use described herein, such as a naturally occurring thiolase (e.g., an endogenous thiolase or one from another species) or a variant thereof that retains thiolase activity. In one aspect, the thiolase is present in the cytosol of the host cells.

In some aspects, the host cells comprising a heterologous polynucleotide encoding a thiolase have an increased level of thiolase activity compared to the host cells without the heterologous polynucleotide encoding the thiolase, when cultivated under the same conditions. In some aspects, the host cells have an increased level of thiolase activity of at least 5%, e.g., at least 10%, at least 15%, at least 20%, at least 25%, at least 50%, at least 100%, at least 150%, at least 200%, at least 300%, or at 500% compared to the host cells without the heterologous polynucleotide encoding the thiolase, when cultivated under the same conditions.

Exemplary thiolase genes that can be used with the host cells and methods of use described herein include, but are not limited to, the Clostridium acetobutylicum thiolase gene comprising SEQ ID NO: 1 (which encodes the thiolase of SEQ ID NO: 3), the Lactobacillus reuteri thiolase gene encoding the thiolase of SEQ ID NO: 26, the Lactobacillus brevis thiolase gene comprising SEQ ID NO: 41 (which encodes the thiolase of SEQ ID NO: 42), the Propionibacterium freudenreichii thiolase gene comprising SEQ ID NO: 39 (which encodes the thiolase of SEQ ID NO: 40), an E. coli thiolase (NP_416728, Martin et al., Nat. Biotechnology 21 :796-802 (2003)), a S. cere visiae thiolase (NP_015297, Hiser et al., J. Biol. Chem. 269:31383 -31389 (1994)), a C. pasteurianum thiolase (e.g., protein ID ABAI8857.I), a C. beijerinckii thiolase (e.g., protein ID EAP59904.1 or EAP59331 .1 ), a Clostridium perfringens thiolase (e.g., protein ID ABG86544.I, ABG83108.I), a Clostridium diflicile thiolase (e.g., protein ID CAJ67900.1 or ZP _01231975.1 ), a Thermoanaerobacterium thermosaccharolyticum thiolase (e.g., protein ID CAB07500.1 ), a Thermoanaerobacter tengcongensis thiolase (e.g., A.L\.M23825.1 ), a Carboxydothermus hydrogenoformans thiolase (e.g., protein ID ABB13995.I), a Desulfotomaculum reducens Ml-I thiolase (e.g., protein ID EAR45123.1 ), or a Candida tropicalis thiolase (e.g., protein ID BAA02716.1 or BAA02715.1 ).

Additional polynucleotides encoding suitable thiolases may be obtained from microorganisms of any genus, including those readily available within the UniProtKB database (www.uniprot.org). In one aspect, the thiolase is a bacterial, a yeast, or a filamentous fungal thiolase, e.g., obtained from any of the microorganisms described herein, as described supra under the section on methylglyoxal synthases.

In one aspect, the thiolase has at least 60%, e.g., at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to the thiolase encoded by any thiolase gene described herein (e.g., any thiolase of SEQ ID NO: 3, 26, 40 or 42). In one aspect, the thiolase sequence differs by no more than ten amino acids, e.g., by no more than five amino acids, by no more than four amino acids, by no more than three amino acids, by no more than two amino acids, or by one amino acid from the thiolase encoded by any thiolase gene described herein (e.g., any thiolase of SEQ ID NO: 3, 26, 40 or 42). In one aspect, the thiolase comprises or consists of the amino acid sequence of the thiolase encoded by any thiolase gene described herein (e.g., any thiolase of SEQ ID NO: 3, 26, 40 or 42), allelic variant, or a fragment thereof having thiolase activity. In one aspect, the thiolase has an amino acid substitution, deletion, and/or insertion of one or more (e.g., two, several) amino acids. In some aspects, the total number of amino acid substitutions, deletions and/or insertions is not more than 10, e.g., not more than 9, 8, 7, 6, 5, 4, 3, 2, or 1 .

In some aspects, the thiolase has at least 20%, e.g., at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% of thiolase activity of any thiolase described herein (e.g., any thiolase of SEQ ID NO: 3, 26, 40 or 42) under the same conditions.

In one aspect, the thiolase coding sequence hybridizes under at least low stringency conditions, e.g., medium stringency conditions, medium-high stringency conditions, high stringency conditions, or very high stringency conditions with the full-length complementary strand of the coding sequence from any thiolase described herein (e.g., any thiolase of SEQ ID NO: 3, 26, 40 or 42, such as the coding sequences shown in SEQ ID NO: 1 , 2, 25, 39, or 41 ). In one aspect, the thiolase coding sequence has at least 65%, e.g., at least 70%, at least 75%, at least 80%, at least 85%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with the coding sequence from any thiolase described herein (e.g., any thiolase of SEQ ID NO: 3, 26, 40 or 42, such as the coding sequences shown in SEQ ID NO: 1 , 2, 25, 39, or 41 ).

In one aspect, the heterologous polynucleotide encoding the thiolase comprises the coding sequence of any thiolase described herein (e.g., any thiolase of SEQ I D NO: 3, 26, 40 or 42, such as the coding sequences shown in SEQ ID NO: 1 , 2, 25, 39, or 41 ). In one aspect, the heterologous polynucleotide encoding the thiolase comprises a subsequence of any thiolase described herein, wherein the subsequence encodes a fragment having thiolase activity. In one aspect, the number of nucleotides residues in the subsequence is at least 75%, e.g., at least 80%, 85%, 90%, or 95% of the number of the referenced sequence. The thiolases can also include fused polypeptides or cleavable fusion polypeptides, as described supra.

CoA-Transferases

In some aspects, the recombinant Lactobacillus reuteri host cells comprise a heterologous polynucleotide that encodes a CoA-transferase. The CoA-transferase can be any CoA-transferase that is suitable for the host cells and their methods of use described herein, such as a naturally occurring CoA-transferase (e.g., an endogenous CoA- transferase or one from another species) or a variant thereof that retains CoA-transferase activity. In some aspects, the CoA-transferase is an acetoacetyl-CoA:acetate/butyrate CoA transferase. In some aspects, the CoA-transferase is an acetoacetyl-CoA hydrolase. In some aspects, the CoA-transferase is a succinyl-CoA:acetoacetate transferase. In some aspects, the CoA-transferase is present in the cytosol of the host cells. In some aspects, the CoA-transferase is a protein complex comprising a first CoA-transferase subunit and the second CoA-transferase subunit wherein the subunits comprise different amino acid sequences.

In some aspects, the host cells comprising a heterologous polynucleotide encoding a CoA-transferase have an increased level of CoA-transferase activity compared to the host cells without the heterologous polynucleotide encoding the CoA-transferase, when cultivated under the same conditions. In some aspects, the host cells have an increased level of CoA-transferase activity of at least 5%, e.g., at least 10%, at least 15%, at least 20%, at least 25%, at least 50%, at least 100%, at least 150%, at least 200%, at least 300%, or at 500% compared to the host cells without the heterologous polynucleotide encoding the CoA-transferase, when cultivated under the same conditions.

Exemplary succinyl-CoA:acetoacetate transferase genes that can be used with the host cells and methods of use described herein include, but are not limited to, a Bacillus subtilis succinyl-CoA:acetoacetate transferase gene comprising SEQ ID NO: 4 and SEQ ID NO: 7 (which encodes a protein complex comprising subunits of SEQ ID NO: 6 and SEQ ID NO: 9, respectively), a Bacillus mojavensis succinyl-CoA:acetoacetate transferase gene comprising SEQ ID NO: 10 and SEQ ID NO: 13 (which encodes a protein complex comprising subunits of SEQ ID NO: 12 and SEQ ID NO: 15, respectively), a Helicobacter pylori succinyl-CoA:acetoacetate transferase (YP_627417, YP_627418, Corthesy-Theulaz, et al ., J Biol Chem 272:25659-25667 (1997); see also SEQ ID NO: 71/72 and SEQ ID NO: 74/75, which encode a protein complex comprising subunits of SEQ ID NO: 73 and SEQ ID NO: 76, respectively)), an Alicyclobacillus succinyl-CoA:acetoacetate transferase gene comprising SEQ ID NO: 67 and SEQ ID NO: 69 (which encode a protein complex comprising subunits of SEQ ID NO: 68 and SEQ ID NO: 70, respectively), and a Homo sapiens succinyl-CoA:acetoacetate transferase (NP_000427, NP071403, Fukao, T., et al., Genomics 68:144-151 (2000); Tanaka, H., et al., Mol Hum Reprod 8:16-23 (2002)).

Exemplary acetoacetyl-CoA:acetate/butyrate CoA transferase genes that can be used with the host cells and methods of use described herein include, but are not limited to, an E. coli acetoacetyl-CoA:acetate CoA transferase (comprising SEQ ID NO: 27 and SEQ ID NO: 29, which encodes a protein complex comprising subunits of SEQ ID NO: 28 and SEQ ID NO: 30, respectively; NP 416726.1 , NP_416725.1 ; Hanai et al., Appl Environ Microbiol 73:7814-7818 (2007)), a Clostridium acetobutylicum acetoacetyl-CoA:acetate CoA transferase (comprising SEQ ID NO: 31 and SEQ I D NO: 33, which encodes a protein complex comprising subunits of SEQ ID NO: 32 and SEQ ID NO: 34, respectively NPJ 49326.1 , NPJ 49327.1 ; Jojima et al., Appl Microbiol Biotechnol 77:1219-1224 (2008)), a Clostridium saccharoperbutylacetonicum acetoacetyl-CoA:acetate CoA transferase (AAP42564.1 , AAP42565.1 ; Kosaka et al., Biosci. Biotechnol Biochem. 71 :58- 68 (2007)), and an Eremococcus coleocola butyrate-acetoacetate CoA transferase (comprising SEQ ID NO: 61/62 together with SEQ ID NO: 64/65, which encodes a protein complex comprising subunits of SEQ ID NO: 63 and SEQ ID NO: 66, respectively; as described in USSN 61/706,661 , the content of which is incorporated herein).

Exemplary acetoacetyl-CoA hydrolase genes that can be used with the host cells and methods of use described herein include, but are not limited to, acyl-CoA hydrolases, 3-hydroxyisobutyryl-CoA hydrolases, acetyl-CoA hydrolases, and dicarboxylic acid thioesterases, such as a Rattus norvegicus 3-hydroxyisobutyryl-CoA hydrolase (Q5XIE6.2; Shimomura et al., J Biol. Chem. 269 :14248-14253 (1994)), a Homo sapiens 3- hydroxyisobutyryl-CoA hydrolase (Q6NVY1 .2; Shimomura et al., supra), a Rattus norvegicus acetyl-CoA hydrolase (NP 570103.1 ; Robinson et al., Res. Commun. 71 :959- 965 (1976)), a Saccharomyces cerevisiae acetyl-CoA hydrolase (NP_009538; Buu et al., J. Biol. Chem. 278: 17203-17209 (2003)), a Homo sapiens dicarboxylic acid thioesterase (CAA15502; Westin et al., J Biol. Chem. 280:38125-38132 (2005)), and an Escherichia coli dicarboxylic acid thioesterase (Naggert et al., J Biol. Chem. 266: 1 1044-1 1050 (1991 )).

The CoA-transferase coding sequences can also be used to design nucleic acid probes to identify and clone DNA encoding CoA-transferases from strains of different genera or species, as described supra.

The polynucleotides encoding CoA-transferases may also be identified and obtained from other sources including microorganisms isolated from nature (e.g., soil, composts, water, etc.) or DNA samples obtained directly from natural materials (e.g., soil, composts, water, etc.) as described supra.

Techniques used to isolate or clone polynucleotides encoding CoA-transferases are described supra.

In one aspect, the CoA-transferase is a protein complex wherein one or more subunits have at least 60%, e.g., at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to the CoA-transferase subunits encoded by any CoA-transferase subunit gene described herein (e.g., any CoA-transferase subunit of the complexes of SEQ ID NOs: 6+9, 12+15, 28+30, 32+34, or 63+66). In one aspect, the sequence of the CoA-transferase subunits differ by no more than ten amino acids, e.g., by no more than five amino acids, by no more than four amino acids, by no more than three amino acids, by no more than two amino acids, or by one amino acid from the CoA-transferase subunits encoded by any CoA- transferase subunit gene described herein (e.g., any CoA-transferase subunit of the complexes of SEQ I D NOs: 6+9, 12+15, 28+30, 32+34, or 63+66). In one aspect, the CoA- transferase is a protein complex comprising or consisting of the amino acid sequences of any CoA-transferase protein complex described herein (e.g., any CoA-transferase complex of SEQ ID NOs: 6+9, 12+15, 28+30, 32+34, or 63+66). In one aspect, the sequence of the CoA-transferase subunit has an amino acid substitution, deletion, and/or insertion of one or more (e.g., two, several) amino acids. In some aspects, the total number of amino acid substitutions, deletions and/or insertions is not more than 10, e.g., not more than 9, 8, 7, 6, 5, 4, 3, 2, or 1 .

In some aspects, the CoA-transferase protein complex has at least 20%, e.g., at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% of the CoA- transferase activity of any CoA-transferase protein complex described herein (e.g., any CoA-transferase complex of SEQ ID NOs: 6+9, 12+15, 28+30, 32+34, or 63+66) under the same conditions.

In one aspect, the CoA-transferase subunit coding sequence hybridizes under at least low stringency conditions, e.g., medium stringency conditions, medium-high stringency conditions, high stringency conditions, or very high stringency conditions with the full-length complementary strand of the coding sequence from any CoA-transferase subunit described herein (e.g., a subunit coding sequence of SEQ ID NO: 4, 5, 7, 8, 10, 1 1 , 13, 14, 27, 29, 31 , 33, 61 , 62, 64, or 65). In one aspect, the CoA-transferase subunit coding sequence has at least 65%, e.g., at least 70%, at least 75%, at least 80%, at least 85%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with the coding sequence from any CoA-transferase subunit described herein (e.g., a subunit coding sequence of SEQ ID NO: 4, 5, 7, 8, 10, 1 1 , 13, 14, 27, 29, 31 , 33, 61 , 62, 5 64, or 65), or the mature polypeptide coding sequence thereof.

In one aspect, the heterologous polynucleotide encoding the CoA-transferase subunit comprises the coding sequence of any CoA-transferase subunit described herein (e.g., a subunit coding sequence of SEQ ID NO: 4, 5, 7, 8, 10, 1 1 , 13, 14, 27, 29, 31 , 33, 61 , 62, 64, or 65). In one aspect, the heterologous polynucleotide encoding the CoA- i o transferase subunit comprises a subsequence of the coding sequence from any CoA- transferase subunit described herein. In one aspect, the number of nucleotides residues in the subsequence is at least 75%, e.g., at least 80%, 85%, 90%, or 95% of the number of the referenced sequence.

The CoA-transferases (and subunits thereof) can also include fused polypeptides or

15 cleavable fusion polypeptides, as described supra.

HMG-CoA synthases

In some aspects, the recombinant Lactobacillus reuteri host cells comprise a heterologous polynucleotide that encodes an HMG-CoA synthase. The HMG-CoA synthase can be any HMG-CoA synthase that is suitable for the host cells and their methods of use 20 described herein, such as a naturally occurring HMG-CoA synthase (e.g., an endogenous HMG-CoA synthase or one from another species) or a variant thereof that retains HMG- CoA synthase activity. In one aspect, the HMG-CoA synthase is present in the cytosol of the host cells.

In some aspects, the host cells comprising a heterologous polynucleotide encoding 25 an HMG-CoA synthase have an increased level of HMG-CoA synthase activity compared to the host cells without the heterologous polynucleotide encoding the HMG-CoA synthase, when cultivated under the same conditions. In some aspects, the host cells have an increased level of HMG-CoA synthase activity of at least 5%, e.g., at least 10%, at least 15%, at least 20%, at least 25%, at least 50%, at least 100%, at least 150%, at least 200%, 30 at least 300%, or at 500% compared to the host cells without the heterologous polynucleotide encoding the HMG-CoA synthase, when cultivated under the same conditions.

Exemplary HMG-CoA synthases that can be used with the host cells and methods of use described herein include, but are not limited to, those HMG-CoA synthases shown in 35 Table 3. Table 3.

Additional polynucleotides encoding suitable HMG-CoA synthases may be obtained from microorganisms of any genus, including those readily available within the UniProtKB database (www.uniprot.org). In one aspect, the HMG-CoA synthase is a bacterial, a yeast, or a filamentous fungal HMG-CoA synthase, e.g., obtained from any of the microorganisms described supra under the section on methylglyoxal synthases.

The HMG-CoA synthase coding sequences can also be used to design nucleic acid probes to identify and clone DNA encoding HMG-CoA synthases from strains of different genera or species, as described supra.

The polynucleotides encoding HMG-CoA synthase may also be identified and obtained from other sources including microorganisms isolated from nature (e.g., soil, composts, water, etc.) or DNA samples obtained directly from natural materials (e.g., soil, composts, water, etc.) as described supra.

Techniques used to isolate or clone polynucleotides encoding HMG-CoA synthases are described supra.

In one aspect, the HMG-CoA synthase has at least 60%, e.g., at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any HMG-CoA synthase described herein (e.g., any HMG-CoA synthase of SEQ ID NO: 77, 78, 79, 80, 81 , 82, 83, 84, 85, 86, 87, 88, 89, 90, 91 , 92, 93, 94. 95, or 96). In one aspect, the HMG-CoA synthase sequence differs by no more than ten amino acids, e.g., by no more than five amino acids, by no more than four amino acids, by no more than three amino acids, by no more than two amino acids, or by one amino acid from any HMG-CoA synthase described herein (e.g., any HMG-CoA 5 synthase of SEQ ID NO: 77, 78, 79, 80, 81 , 82, 83, 84, 85, 86, 87, 88, 89, 90, 91 , 92, 93, 94. 95, or 96). In one aspect, the HMG-CoA synthase comprises or consists of the amino acid sequence of any HMG-CoA synthase described herein (e.g., any HMG-CoA synthase of SEQ ID NO: 77, 78, 79, 80, 81 , 82, 83, 84, 85, 86, 87, 88, 89, 90, 91 , 92, 93, 94. 95, or 96), allelic variant, or a fragment thereof having HMG-CoA synthase activity. In one aspect, i o the HMG-CoA synthase has an amino acid substitution, deletion, and/or insertion of one or more (e.g., two, several) amino acids. In some aspects, the total number of amino acid substitutions, deletions and/or insertions is not more than 10, e.g., not more than 9, 8, 7, 6, 5, 4, 3, 2, or 1 .

In some aspects, the HMG-CoA synthase has at least 20%, e.g., at least 40%, at 15 least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% of the HMG-CoA synthase activity of any HMG-CoA synthase described herein (e.g., any HMG-CoA synthase of SEQ ID NO: 77, 78, 79, 80, 81 , 82, 83, 84, 85, 86, 87, 88, 89, 90, 91 , 92, 93, 94. 95, or 96) under the same conditions.

20 In one aspect, the HMG-CoA synthase coding sequence hybridizes under at least low stringency conditions, e.g., medium stringency conditions, medium-high stringency conditions, high stringency conditions, or very high stringency conditions with the full-length complementary strand of the coding sequence from any HMG-CoA synthase described herein (e.g., any HMG-CoA synthase of SEQ ID NO: 77, 78, 79, 80, 81 , 82, 83, 84, 85, 86,

25 87, 88, 89, 90, 91 , 92, 93, 94. 95, or 96) (see, e.g., J. Sambrook, E.F. Fritsch, and T.

Maniatus, 1989, supra). In one aspect, the HMG-CoA synthase coding sequence has at least 65%, e.g., at least 70%, at least 75%, at least 80%, at least 85%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with the coding

30 sequence from any HMG-CoA synthase described herein (e.g., any HMG-CoA synthase of SEQ ID NO: 77, 78, 79, 80, 81 , 82, 83, 84, 85, 86, 87, 88, 89, 90, 91 , 92, 93, 94. 95, or 96).

In one aspect, the heterologous polynucleotide encoding the HMG-CoA synthase comprises the coding sequence of any HMG-CoA synthase described herein (e.g., any HMG-CoA synthase of SEQ ID NO: 77, 78, 79, 80, 81 , 82, 83, 84, 85, 86, 87, 88, 89, 90,

35 91 , 92, 93, 94. 95, or 96). In one aspect, the heterologous polynucleotide encoding the HMG-CoA synthase comprises a subsequence of the coding sequence from any HMG-CoA synthase described herein, wherein the subsequence encodes a polypeptide having HMG- CoA synthase activity. In one aspect, the number of nucleotides residues in the coding subsequence is at least 75%, e.g., at least 80%, 85%, 90%, or 95% of the number of the referenced coding sequence.

The HMG-CoA synthases can also include fused polypeptides or cleavable fusion polypeptides, as described supra.

HMG-CoA lyases

In some aspects, the recombinant Lactobacillus reuteri host cells comprise a heterologous polynucleotide that encodes an HMG-CoA lyase. The HMG-CoA lyase can be any HMG-CoA lyase that is suitable for the host cells and their methods of use described herein, such as a naturally occurring HMG-CoA lyase (e.g., an endogenous HMG-CoA lyase or one from another species) or a variant thereof that retains HMG-CoA lyase activity. In one aspect, the HMG-CoA lyase is present in the cytosol of the host cells.

In some aspects, the host cells comprising a heterologous polynucleotide encoding an HMG-CoA lyase have an increased level of HMG-CoA lyase activity compared to the host cells without the heterologous polynucleotide encoding the HMG-CoA lyase, when cultivated under the same conditions. In some aspects, the host cells have an increased level of HMG-CoA lyase activity of at least 5%, e.g., at least 10%, at least 15%, at least 20%, at least 25%, at least 50%, at least 100%, at least 150%, at least 200%, at least 300%, or at 500% compared to the host cells without the heterologous polynucleotide encoding the HMG-CoA lyase, when cultivated under the same conditions.

Exemplary HMG-CoA lyases that can be used with the host cells and methods of use described herein include, but are not limited to, those HMG-CoA lyase shown in Table

Table 4.

Gallus gallus UniProtKB P35915 108

Danio rerio UniProtKB A8WG57 109

Macaca fascicularis UniProtKB Q8HXZ6 1 10

Pongo abelii UniProtKB Q5R9E1 1 1 1

Additional polynucleotides encoding suitable HMG-CoA lyases may be obtained from microorganisms of any genus, including those readily available within the UniProtKB database (www.uniprot.org). In one aspect, the HMG-CoA lyase is a bacterial, a yeast, or a filamentous fungal HMG-CoA synthase, e.g., obtained from any of the microorganisms described herein, as described supra under the section on methylglyoxal synthases.

The HMG-CoA lyase coding sequences can also be used to design nucleic acid probes to identify and clone DNA encoding HMG-CoA lyases from strains of different genera or species, as described supra.

The polynucleotides encoding HMG-CoA lyases may also be identified and obtained from other sources including microorganisms isolated from nature (e.g., soil, composts, water, etc.) or DNA samples obtained directly from natural materials (e.g., soil, composts, water, etc.) as described supra.

Techniques used to isolate or clone polynucleotides encoding HMG-CoA lyases are described supra.

In one aspect, the HMG-CoA lyase has at least 60%, e.g., at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any HMG-CoA lyase described herein (e.g., any HMG- CoA lyase of SEQ ID NO: 97, 98, 99, 100, 101 , 102, 103, 104, 105, 106, 107, 108, 109, 1 10, or 1 1 1 ). In one aspect, the HMG-CoA lyase sequence differs by no more than ten amino acids, e.g., by no more than five amino acids, by no more than four amino acids, by no more than three amino acids, by no more than two amino acids, or by one amino acid from any HMG-CoA lyase described herein (e.g., any HMG-CoA lyase of SEQ ID NO: 97, 98, 99, 100, 101 , 102, 103, 104, 105, 106, 107, 108, 109, 1 10, or 1 1 1 ). In one aspect, the HMG-CoA lyase comprises or consists of the amino acid sequence of any HMG-CoA lyase described herein (e.g., any HMG-CoA lyase of SEQ ID NO: 97, 98, 99, 100, 101 , 102, 103, 104, 105, 106, 107, 108, 109, 1 10, or 1 1 1 ), allelic variant, or a fragment thereof having HMG-CoA lyase activity. In one aspect, the HMG-CoA lyase has an amino acid substitution, deletion, and/or insertion of one or more (e.g., two, several) amino acids. In some aspects, the total number of amino acid substitutions, deletions and/or insertions is not more than

10, e.g., not more than 9, 8, 7, 6, 5, 4, 3, 2, or 1 .

In some aspects, the HMG-CoA lyase has at least 20%, e.g., at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% of the HMG-CoA lyase activity of any HMG- CoA lyase described herein (e.g., any HMG-CoA lyase of SEQ ID NO: 97, 98, 99, 100, 101 , 102, 103, 104, 105, 106, 107, 108, 109, 1 10, or 1 1 1 ) under the same conditions.

In one aspect, the HMG-CoA lyase coding sequence hybridizes under at least low stringency conditions, e.g., medium stringency conditions, medium-high stringency conditions, high stringency conditions, or very high stringency conditions with the full-length complementary strand of the coding sequence from any HMG-CoA lyase described herein (e.g., any HMG-CoA lyase of SEQ ID NO: 97, 98, 99, 100, 101 , 102, 103, 104, 105, 106, 107, 108, 109, 1 10, or 1 1 1 ). In one aspect, the HMG-CoA lyase coding sequence has at least 65%, e.g., at least 70%, at least 75%, at least 80%, at least 85%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with the coding sequence from any HMG-CoA lyase described herein (e.g., any HMG-CoA lyase of SEQ ID NO: 97, 98, 99, 100, 101 , 102, 103, 104, 105, 106, 107, 108, 109, 1 10, or 1 1 1 ).

In one aspect, the heterologous polynucleotide encoding the HMG-CoA lyase comprises the coding sequence of any HMG-CoA lyase described herein (e.g., any HMG- CoA lyase of SEQ ID NO: 97, 98, 99, 100, 101 , 102, 103, 104, 105, 106, 107, 108, 109, 1 10, or 1 1 1 ). In one aspect, the heterologous polynucleotide encoding the HMG-CoA lyase comprises a subsequence of the coding sequence from any HMG-CoA lyase described herein, wherein the subsequence encodes a fragment having HMG-CoA lyase activity. In one aspect, the number of nucleotides residues in the subsequence is at least 75%, e.g., at least 80%, 85%, 90%, or 95% of the number of the referenced coding sequence.

The HMG-CoA lyases can also include fused polypeptides or cleavable fusion polypeptides, as described supra.

Acetoacetate decarboxylases

In some aspects, the recombinant Lactobacillus reuteri host cells comprise a heterologous gene that encodes an acetoacetate decarboxylase. The acetoacetate decarboxylase can be any acetoacetate decarboxylase that is suitable for the host cells and their methods of use described herein, such as a naturally occurring acetoacetate decarboxylase (e.g., an endogenous acetoacetate decarboxylase or one from another species) or a variant thereof that retains acetoacetate decarboxylase activity. In one aspect, the acetoacetate decarboxylase is present in the cytosol of the host cells.

In some aspects, the host cells comprising a heterologous polynucleotide encoding an acetoacetate decarboxylase have an increased level of acetoacetate decarboxylase activity compared to the host cells without the heterologous polynucleotide encoding the acetoacetate decarboxylase, when cultivated under the same conditions. In some aspects, the host cells have an increased level of acetoacetate decarboxylase activity of at least 5%, e.g., at least 10%, at least 15%, at least 20%, at least 25%, at least 50%, at least 100%, at least 150%, at least 200%, at least 300%, or at 500% compared to the host cells without the heterologous polynucleotide encoding the acetoacetate decarboxylase, when cultivated under the same conditions.

Exemplary acetoacetate decarboxylase genes that can be used with the host cells and methods of use described herein include, but are not limited to, a Clostridium beijerinckii acetoacetate decarboxylase of SEQ ID NO: 16 (which encodes the acetoacetate decarboxylase of SEQ ID NO: 18), a Lactobacillus salvarius acetoacetate decarboxylase gene comprising SEQ ID NO: 43 (which encodes the acetoacetate decarboxylase of SEQ ID NO: 44), a Lactobacillus plantarum acetoacetate decarboxylase gene comprising SEQ ID NO: 45 (which encodes the acetoacetate decarboxylase of SEQ ID NO: 46), a C. acetobutylicum acetoacetate decarboxylase gene (NP_149328.1 , which encodes the acetoacetate decarboxylase of SEQ ID NO: 36; see Petersen and Bennett, Appl. Environ. Microbiol 56:3491 -3498 (1990)) and a Clostridium saccharoperbutylacetonicum acetoacetate decarboxylase (AAP42566.1 , Kosaka, et al., Biosci. Biotechnol Biochem. 71 :58-68 (2007)).

Additional polynucleotides encoding suitable acetoacetate decarboxylases may be obtained from microorganisms of any genus, including those readily available within the UniProtKB database (www.uniprot.org). In one aspect, the acetoacetate decarboxylase is a bacterial, a yeast, or a filamentous fungal acetoacetate decarboxylase, e.g., obtained from any of the microorganisms described herein, as described supra under the section on methylglyoxal synthases.

The acetoacetate decarboxylase coding sequences can also be used to design nucleic acid probes to identify and clone DNA encoding acetoacetate decarboxylases from strains of different genera or species, as described supra.

The polynucleotides encoding acetoacetate decarboxylases may also be identified and obtained from other sources including microorganisms isolated from nature (e.g., soil, composts, water, etc.) or DNA samples obtained directly from natural materials (e.g., soil, composts, water, etc.) as described supra.

Techniques used to isolate or clone polynucleotides encoding acetoacetate decarboxylases are described supra. In one aspect, the acetoacetate decarboxylase has at least 60%, e.g., at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to the acetoacetate decarboxylase encoded by any acetoacetate decarboxylase gene described herein (e.g., any acetoacetate decarboxylase of SEQ ID NO: 18, 36, 44, or 46). In one aspect, the acetoacetate decarboxylase sequence differs by no more than ten amino acids, e.g., by no more than five amino acids, by no more than four amino acids, by no more than three amino acids, by no more than two amino acids, or by one amino acid from the acetoacetate decarboxylase encoded by any acetoacetate decarboxylase gene described herein (e.g., any acetoacetate decarboxylase of SEQ ID NO: 18, 36, 44, or 46). In one aspect, the acetoacetate decarboxylase comprises or consists of the amino acid sequence of the acetoacetate decarboxylase encoded by any acetoacetate decarboxylase gene described herein (e.g., any acetoacetate decarboxylase of SEQ ID NO: 18, 36, 44, or 46), allelic variant, or a fragment thereof having acetoacetate decarboxylase activity. In one aspect, the acetoacetate decarboxylase has an amino acid substitution, deletion, and/or insertion of one or more (e.g., two, several) amino acids. In some aspects, the total number of amino acid substitutions, deletions and/or insertions is not more than 10, e.g., not more than 9, 8, 7, 6, 5, 4, 3, 2, or 1 .

In one aspect, the acetoacetate decarboxylase coding sequence hybridizes under at least low stringency conditions, e.g., medium stringency conditions, medium-high stringency conditions, high stringency conditions, or very high stringency conditions with the full-length complementary strand of the coding sequence from any acetoacetate decarboxylase described herein (e.g., any acetoacetate decarboxylase of SEQ ID NO: 18, 36, 44, or 46; such as any coding sequence of SEQ ID NO: 16, 17, 35, 43, or 45). In one aspect, the acetoacetate decarboxylase coding sequence has at least 65%, e.g., at least 70%, at least 75%, at least 80%, at least 85%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with the coding sequence from any acetoacetate decarboxylase described herein (e.g., any acetoacetate decarboxylase of SEQ ID NO: 18, 36, 44, or 46; such as any coding sequence of SEQ ID NO: 16, 17, 35, 43, or 45).

In one aspect, the heterologous polynucleotide encoding the acetoacetate decarboxylase comprises the coding sequence of any acetoacetate decarboxylase gene described herein (e.g., any acetoacetate decarboxylase of SEQ ID NO: 18, 36, 44, or 46; such as any coding sequence of SEQ ID NO: 16, 17, 35, 43, or 45). In one aspect, the heterologous polynucleotide encoding the acetoacetate decarboxylase comprises a subsequence of any acetoacetate decarboxylase gene described herein, wherein the subsequence encodes a fragment having acetoacetate decarboxylase activity. In one aspect, the number of nucleotides residues in the subsequence is at least 75%, e.g., at least 80%, 85%, 90%, or 95% of the number of the referenced sequence.

The acetoacetate decarboxylases can also include fused polypeptides or cleavable fusion polypeptides, as described supra.

Isopropanol dehydrogenases

In some aspects, the recombinant Lactobacillus reuteri host cells comprise a heterologous polynucleotide that encodes an isopropanol dehydrogenase. The isopropanol dehydrogenase can be any isopropanol dehydrogenase that is suitable for the host cells and their methods of use described herein, such as a naturally occurring isopropanol dehydrogenase (e.g., an endogenous isopropanol dehydrogenase or one from another species) or a variant thereof that retains isopropanol dehydrogenase activity. In one aspect, the isopropanol dehydrogenase is present in the cytosol of the host cells.

In some aspects, the host cells comprising a heterologous polynucleotide encoding an isopropanol dehydrogenase have an increased level of isopropanol dehydrogenase activity compared to the host cells without the heterologous polynucleotide encoding the isopropanol dehydrogenase, when cultivated under the same conditions. In some aspects, the host cells have an increased level of isopropanol dehydrogenase activity of at least 5%, e.g., at least 10%, at least 15%, at least 20%, at least 25%, at least 50%, at least 100%, at least 150%, at least 200%, at least 300%, or at 500% compared to the host cells without the heterologous polynucleotide encoding the isopropanol dehydrogenase, when cultivated under the same conditions.

Exemplary isopropanol dehydrogenase genes that can be used with the host cells and methods of use described herein include, but are not limited to, a Clostridium beijerinckii isopropanol dehydrogenase of SEQ ID NO: 19 (which encodes the isopropanol dehydrogenase of SEQ ID NO: 21 ), a Thermoanaerobacter ethanolicus isopropanol dehydrogenase gene comprising SEQ ID NO: 22 (which encodes the isopropanol dehydrogenase of SEQ ID NO: 24), a Lactobacillus fermentum isopropanol dehydrogenase gene comprising SEQ ID NO: 47 (which encodes the isopropanol dehydrogenase of SEQ ID NO: 48), a Lactobacillus antri isopropanol dehydrogenase gene comprising SEQ ID NO: 37 (which encodes the isopropanol dehydrogenase of SEQ ID NO: 38), a Thermoanaerobacter brockii isopropanol dehydrogenase (P14941 .1 , Hanai et al., Appl. Environ. Microbiol. 73:7814-7818 (2007); Peretz et al., Anaerobe 3:259-270 (1997)), a Ralstonia eutropha n-propanol dehydrogenase (formerly Alcaligenes eutrophus) (YP_299391 .1 , Steinbuchel and Schlegel et al., Eur. J. Biochem. 141 :555-564 (1984)), a Burkholderia sp. AIU 652 isopropanol dehydrogenase, and a Phytomonas species isopropanol dehydrogenase (AAP39869.1 , Uttaro and Opperdoes et al., Mol. Biochem. Parasitol. 85:213-219 (1997)).

Additional polynucleotides encoding suitable isopropanol dehydrogenases may be obtained from microorganisms of any genus, including those readily available within the UniProtKB database (www.uniprot.org). In one aspect, the isopropanol dehydrogenase is a bacterial, a yeast, or a filamentous fungal isopropanol dehydrogenase, e.g., obtained from any of the microorganisms described herein, as described supra under the section on methylglyoxal synthases.

The isopropanol dehydrogenase coding sequences can also be used to design nucleic acid probes to identify and clone DNA encoding isopropanol dehydrogenases from strains of different genera or species, as described supra.

The polynucleotides encoding isopropanol dehydrogenases may also be identified and obtained from other sources including microorganisms isolated from nature (e.g., soil, composts, water, etc.) or DNA samples obtained directly from natural materials (e.g., soil, composts, water, etc.) as described supra.

Techniques used to isolate or clone polynucleotides encoding isopropanol dehydrogenases are described supra.

In one aspect, the isopropanol dehydrogenase has at least 60%, e.g., at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to the isopropanol dehydrogenase encoded by any isopropanol dehydrogenase gene described herein (e.g., any isopropanol dehydrogenase of SEQ ID NO: 21 , 24, 38, or 48). In one aspect, the isopropanol dehydrogenase sequence differs by no more than ten amino acids, e.g., by no more than five amino acids, by no more than four amino acids, by no more than three amino acids, by no more than two amino acids, or by one amino acid from the isopropanol dehydrogenase encoded by any isopropanol dehydrogenase gene described herein (e.g., any isopropanol dehydrogenase of SEQ I D NO: 21 , 24, 38, or 48). In one aspect, the isopropanol dehydrogenase comprises or consists of the amino acid sequence of the isopropanol dehydrogenase encoded by any isopropanol dehydrogenase gene described herein (e.g., any isopropanol dehydrogenase of SEQ ID NO: 21 , 24, 38, or 48), allelic variant, or a fragment thereof having isopropanol dehydrogenase activity. In one aspect, the isopropanol dehydrogenase has an amino acid substitution, deletion, and/or insertion of one or more (e.g., two, several) amino acids. In some aspects, the total number of amino acid substitutions, deletions and/or insertions is not more than 10, e.g., not more than 9, 8, 7, 6, 5, 4, 3, 2, or 1 .

In some aspects, the isopropanol dehydrogenase has at least 20%, e.g., at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% of the isopropanol dehydrogenase activity of any isopropanol dehydrogenase described herein (e.g., any isopropanol dehydrogenase of SEQ ID NO: 21 , 24, 38, or 48) under the same conditions.

In one aspect, the isopropanol dehydrogenase coding sequence hybridizes under at least low stringency conditions, e.g., medium stringency conditions, medium-high stringency conditions, high stringency conditions, or very high stringency conditions with the full-length complementary strand of the coding sequence from any isopropanol dehydrogenase described herein (e.g., any isopropanol dehydrogenase of SEQ I D NO: 21 , 24, 38, or 48; such as the coding sequence of SEQ ID NO: 19, 20, 22, 23, 37, or 47). In one aspect, the isopropanol dehydrogenase coding sequence has at least 65%, e.g., at least 70%, at least 75%, at least 80%, at least 85%, at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with the coding sequence from any isopropanol dehydrogenase gene described herein (e.g., any isopropanol dehydrogenase of SEQ ID NO: 21 , 24, 38, or 48; such as the coding sequence of SEQ I D NO: 19, 20, 22, 23, 37, or 47).

In one aspect, the heterologous polynucleotide encoding the isopropanol dehydrogenase comprises the coding sequence of any isopropanol dehydrogenase gene described herein (e.g., any isopropanol dehydrogenase of SEQ ID NO: 21 , 24, 38, or 48; such as the coding sequence of SEQ ID NO: 19, 20, 22, 23, 37, or 47). In one aspect, the heterologous polynucleotide encoding the isopropanol dehydrogenase gene comprises a subsequence of the coding sequence from any isopropanol dehydrogenase described herein. In one aspect, the number of nucleotides residues in the subsequence is at least 75%, e.g., at least 80%, 85%, 90%, or 95% of the number of the referenced sequence.

The isopropanol dehydrogenases can also include fused polypeptides or cleavable fusion polypeptides, as described supra.

Hosts, Expression Vectors and Nucleic Acid Constructs

In some aspects, the recombinant Lactobacillus reuteri host cell comprises a heterologous polynucleotide encoding a methylglyoxal synthase, and optionally comprises one or more (e.g., two, several) heterologous polynucleotides of an active isopropanol pathway (e.g., a heterologous polynucleotide encoding a thiolase, a heterologous polynucleotide encoding a CoA-transferase, a heterologous polynucleotide encoding an HMG-CoA synthase, a heterologous polynucleotide encoding a HMG-CoA lyase, a heterologous polynucleotide encoding an acetoacetate decarboxylase, and/or a heterologous polynucleotide encoding isopropanol dehydrogenase), wherein the host cell secretes (and/or is capable of secreting) an increased level of n-propanol compared to the host cell without the heterologous polynucleotide encoding a methylglyoxal synthase when cultivated under the same conditions. Examples of suitable cultivation conditions are described below and will be readily apparent to one of skill in the art based on the teachings herein.

In some aspects, the recombinant Lactobacillus reuteri host cell produces (and/or is capable of producing) n-propanol at a yield of at least than 10%, e.g., at least than 20%, at least than 30%, at least than 40%, at least than 50%, at least than 60%, at least than 70%, at least than 80%, or at least than 90%, of theoretical.

In any of these aspects, the recombinant Lactobacillus reuteri host is capable of an n-propanol volumetric productivity greater than about 0.1 g/L per hour, e.g., greater than about 0.2 g/L per hour, 0.5 g/L per hour, 0.6 g/L per hour, 0.7 g/L per hour, 0.8 g/L per hour, 0.9 g/L per hour, 1 .0 g/L per hour, 1 .1 g/L per hour, 1 .2 g/L per hour, 1.3 g/L per hour, 1 .5 g/L per hour, 1 .75 g/L per hour, 2.0 g/L per hour, 2.25 g/L per hour, 2.5 g/L per hour, or 3.0 g/L per hour; or between about 0.1 g/L per hour and about 2.0 g/L per hour, e.g., between about 0.3 g/L per hour and about 1 .7 g/L per hour, about 0.5 g/L per hour and about 1 .5 g/L per hour, about 0.7 g/L per hour and about 1 .3 g/L per hour, about 0.8 g/L per hour and about 1.2 g/L per hour, or about 0.9 g/L per hour and about 1 .1 g/L per hour.

The recombinant Lactobacillus reuteri host cells may be cultivated in a nutrient medium suitable for production of one or more of the polypeptides described herein and capable of the recombinant production of n-propanol using methods well known in the art. For example, the cell may be cultivated by shake flask cultivation, and small-scale or large- scale fermentation (including continuous, batch, fed-batch, or solid state fermentations) in laboratory or industrial fermentors performed in a suitable medium and under conditions allowing the desired polypeptide to be expressed and/or isolated. The cultivation takes place in a suitable nutrient medium comprising carbon and nitrogen sources and inorganic salts, as described herein, using procedures known in the art. Suitable media are available from commercial suppliers, may be prepared according to published compositions (e.g., in catalogues of the American Type Culture Collection), or may be prepared from commercially available ingredients.

The recombinant Lactobacillus host cells described herein also can be subjected to adaptive evolution to further augment n-propanol biosynthesis, including under conditions approaching theoretical maximum growth.

The recombinant Lactobacillus reuteri host cells described herein may utilize expression vectors comprising the coding sequence of one or more (e.g., two, several) heterologous genes described herein linked to one or more control sequences that direct expression in a suitable host cell under conditions compatible with the control sequence(s). Such expression vectors may be used in any of the host cells and methods describe herein. The polynucleotides described herein may be manipulated in a variety of ways to provide for expression of a desired polypeptide. Manipulation of the polynucleotide prior to its insertion into a vector may be desirable or necessary depending on the expression vector. The techniques for modifying polynucleotides utilizing recombinant DNA methods are well known in the art.

A construct or vector (or multiple constructs or vectors) comprising the heterologous polynucleotide encoding a methylglyoxal synthase and/or the one or more (e.g., two, several) heterologous isopropanol pathway genes may be introduced into a Lactobacillus cell so that the construct or vector is maintained as a chromosomal integrant or as a self- replicating extra-chromosomal vector as described herein.

The various nucleotide and control sequences may be joined together to produce a recombinant expression vector that may include one or more (e.g., two, several) convenient restriction sites to allow for insertion or substitution of the polynucleotide at such sites. Alternatively, the polynucleotide(s) may be expressed by inserting the polynucleotide(s) or a nucleic acid construct comprising the sequence into an appropriate vector for expression. In creating the expression vector, the coding sequence is located in the vector so that the coding sequence is operably linked with the appropriate control sequences for expression.

The recombinant expression vector may be any vector (e.g., a plasmid or virus) that can be conveniently subjected to recombinant DNA procedures and can bring about expression of the polynucleotide. The choice of the vector will typically depend on the compatibility of the vector with the host cell into which the vector is to be introduced. The vector may be a linear or closed circular plasmid.

In one aspect, each heterologous polynucleotide is contained on an independent vector. In one aspect, at least two of the heterologous polynucleotides are contained on a single vector. In one aspect, at least three of the heterologous polynucleotides are contained on a single vector. In one aspect, at least four of the heterologous polynucleotides are contained on a single vector. In one aspect, all the heterologous polynucleotides are contained on a single vector. Polynucleotides encoding heteromeric subunits of a protein complex (e.g., a CoA-transferase) may be contained in a single heterologous polynucleotide on a single vector or alternatively contained in separate heterologous polynucleotides on separate vectors.

The vector may be an autonomously replicating vector, i.e., a vector that exists as an extrachromosomal entity, the replication of which is independent of chromosomal replication, e.g., a plasmid, an extrachromosomal element, a minichromosome, or an artificial chromosome. The vector may contain any means for assuring self-replication. Alternatively, the vector may be one that, when introduced into the host cell, is integrated into the genome and replicated together with the chromosome(s) into which it has been integrated. Furthermore, a single vector or plasmid or two or more vectors or plasmids that together contain the total DNA to be introduced into the genome of the host cell, or a transposon, may be used.

The expression vector may contain any suitable promoter sequence that is recognized by a host cell for expression of the methylglyoxal synthase gene, or any isopropanol pathway gene described herein. The promoter sequence contains transcriptional control sequences that mediate the expression of the polypeptide. The promoter may be any polynucleotide that shows transcriptional activity in the host cell of choice including mutant, truncated, and hybrid promoters, and may be obtained from genes encoding extracellular or intracellular polypeptides either homologous or heterologous to the host cell.

Each gene coding sequence described herein may be operably linked to a promoter that is foreign to the gene. For example, in one aspect, a methylglyoxal synthase coding sequence may be operably linked to promoter foreign to the polynucleotide.

Each heterologous polynucleotide described herein may be operably linked to a promoter that is foreign to the polynucleotide. For example, in one aspect, the heterologous polynucleotide encoding the methylglyoxal synthase is operably linked to a promoter foreign to the polynucleotide. In another aspect, the heterologous polynucleotide encoding a polypeptide of an isopropanol pathway described herein (e.g., a thiolase, a CoA- transferase, an HMG-CoA synthase, an HMG-CoA lyase, an acetoacetate decarboxylase, and/or an isopropanol dehydrogenase) is operably linked to promoter foreign to the polynucleotide. The promoters may be identical to or share a high degree of sequence identity (e.g., at least about 80%, at least about 85%, at least about 90%, at least about 95%, or at least about 99%) with a selected native promoter. As described supra, polynucleotides encoding heteromeric subunits of a protein complex may be contained in a single heterologous polynucleotide (e.g., a single plasmid), or alternatively contained in separate heterologous polynucleotides (e.g., on separate plasmids). For example, in one aspect, the first heterologous polynucleotide encoding a first subunit, and the second heterologous polynucleotide encoding a second subunit are contained in a single heterologous polynucleotide operably linked to a promoter that is foreign to both the heterologous polynucleotide encoding the first subunit and the heterologous polynucleotide encoding the second subunit. In one aspect, the first heterologous polynucleotide encoding a first subunit, and the second heterologous polynucleotide encoding a second subunit are each contained in separate unlinked heterologous polynucleotides, wherein the heterologous polynucleotide encoding the first subunit is operably linked to a foreign promoter, and the heterologous polynucleotide encoding the second subunit is operably linked to a foreign promoter. The promoters in the foregoing may be the same or different.

Examples of suitable promoters for directing the transcription of the nucleic acid constructs in a bacterial host cell are the promoters obtained from the Bacillus amyloliquefaciens alpha-amylase gene (amyQ), Bacillus licheniformis alpha-amylase gene (amyL), Bacillus licheniformis penicillinase gene (penP), Bacillus stearothermophilus maltogenic amylase gene (amyM), Bacillus subtilis levansucrase gene (sacB), Bacillus subtilis xylA and xylB genes, E. coli lac operon, E. coli trc promoter (Egon et al., 1988, Gene 69: 301 -315), Streptomyces coelicolor agarase gene {dagA), and prokaryotic beta- lactamase gene (Villa-Kamaroff et al., 1978, Proc. Natl. Acad. Sci. USA 75: 3727-3731 ), as well as the tac promoter (DeBoer et al., 1983, Proc. Natl. Acad. Sci. USA 80: 21 -25). Further promoters are described in "Useful proteins from recombinant bacteria" in Gilbert et al., 1980, Scientific American, 242: 74-94; and in Sambrook et al., 1989, supra.

The control sequence may also be a suitable transcription terminator sequence, which is recognized by a host cell to terminate transcription. The terminator sequence is operably linked to the 3'-terminus of the polynucleotide encoding the polypeptide. Any terminator that is functional in the host cell of choice may be used. The terminator may be identical to or share a high degree of sequence identity (e.g., at least about 80%, at least about 85%, at least about 90%, at least about 95%, or at least about 99%) with the selected native terminator.

The control sequence may also be a suitable leader sequence, when transcribed is a nontranslated region of an mRNA that is important for translation by the host cell. The leader sequence is operably linked to the 5'-terminus of the polynucleotide encoding the polypeptide. Any leader sequence that is functional in the host cell of choice may be used.

The control sequence may also be a polyadenylation sequence; a sequence operably linked to the 3'-terminus of the polynucleotide and, when transcribed, is recognized by the host cell as a signal to add polyadenosine residues to transcribed mRNA. Any polyadenylation sequence that is functional in the host cell of choice may be used.

It may also be desirable to add regulatory sequences that allow the regulation of the expression of the polypeptide relative to the growth of the host cell. Examples of regulatory systems are those that cause the expression of the gene to be turned on or off in response to a chemical or physical stimulus, including the presence of a regulatory compound. Regulatory systems in prokaryotic systems include the lac, tac, and trp operator systems. Other examples of regulatory sequences are those that allow for gene amplification.

The vectors may contain one or more (e.g., two, several) selectable markers that permit easy selection of transformed, transfected, transduced, or the like cells. A selectable marker is a gene the product of which provides for biocide or viral resistance, resistance to heavy metals, prototrophy to auxotrophs, and the like. Examples of bacterial selectable markers are the dal genes from Bacillus subtilis or Bacillus licheniformis, or markers that confer antibiotic resistance such as ampicillin, chloramphenicol, kanamycin, or tetracycline resistance.

The vectors may contain one or more (e.g., two, several) elements that permit integration of the vector into the host cell's genome or autonomous replication of the vector in the cell independent of the genome.

Any of the heterologous polynucleotides described herein (e.g., a methylglyoxal synthase coding sequence) may be chromosomally integrated into the Lactobacillus reuteri genome. For genome integration, a vector may rely on the polynucleotide's sequence encoding the polypeptide or any other element of the vector for integration into the genome by homologous or non-homologous recombination. Alternatively, the vector may contain additional polynucleotides for directing integration by homologous recombination into the genome of the host cell at a precise location(s) in the chromosome(s). To increase the likelihood of integration at a precise location, the integrational elements should contain a sufficient number of nucleic acids, such as 100 to 10,000 base pairs, 400 to 10,000 base pairs, and 800 to 10,000 base pairs, which have a high degree of sequence identity to the corresponding target sequence to enhance the probability of homologous recombination. The integrational elements may be any sequence that is homologous with the target sequence in the genome of the host cell. Furthermore, the integrational elements may be non-encoding or encoding polynucleotides. On the other hand, the vector may be integrated into the genome of the host cell by non-homologous recombination.

For autonomous replication, the vector may further comprise an origin of replication enabling the vector to replicate autonomously in the host cell in question. The origin of replication may be any plasmid replicator mediating autonomous replication that functions in a cell. The term "origin of replication" or "plasmid replicator" means a polynucleotide that enables a plasmid or vector to replicate in vivo. Examples of bacterial origins of replication are the origins of replication of plasmids pBR322, pUC19, pACYC177, and pACYC184 permitting replication in E. coli, and pUB1 10, pE194, pTA1060, and ρΑΜβΙ permitting replication in Bacillus.

More than one copy of a polynucleotide described herein may be inserted into a host cell to increase production of a polypeptide. An increase in the copy number of the polynucleotide can be obtained by integrating at least one additional copy of the sequence into the host cell genome or by including an amplifiable selectable marker gene with the polynucleotide where cells containing amplified copies of the selectable marker gene, and thereby additional copies of the polynucleotide, can be selected for by cultivating the cells in the presence of the appropriate selectable agent.

The procedures used to ligate the elements described above to construct the recombinant expression vectors described herein are well known to one skilled in the art (see, e.g., Sambrook et al., 1989, supra).

The introduction of a construct or vector containing one or more heterologous polynucleotides into a Lactobacillus cell may, for instance, be effected by protoplast transformation (see, e.g., Chang and Cohen, 1979, Mol. Gen. Genet. 168: 1 1 1 -1 15), by using competent cells (see, e.g., Young and Spizizen, 1961 , J. Bacteriol. 81 : 823-829, or Dubnau and Davidoff-Abelson, 1971 , J. Mol. Biol. 56: 209-221 ), by electroporation (see, e.g., Shigekawa and Dower, 1988, Biotechniques 6: 742-751 ), or by conjugation (see, e.g., Koehler and Thorne, 1987, J. Bacteriol. 169: 5271 -5278).

Gene Disruptions

The recombinant Lactobacillus reuteri host cells may also comprise one or more

(e.g., two, several) gene disruptions, e.g., to divert sugar metabolism from undesired products to n-propanol and/or isopropanol.

In one aspect, the recombinant Lactobacillus reuteri host cells comprise a disrupted endogenous adhE gene which encodes a bifunctional alcohol/acetaldehyde dehydrogenase. Disruption to adhE may reduce consumption of NADH used for ethanol production, thereby providing more NADH for use in the methylglyoxal pathway of Figures 1 and 2. In some embodiments, the Lactobacillus host cells produce a greater amount of n- propanol compared to the cell without the adhE disruption when cultivated under identical conditions. In some embodiments, the endogenous adhE gene is inactivated. In one embodiment, the endogenous adhE gene has the coding sequence shown in SEQ ID NO: 131 , which encodes a polypeptide having the amino acid sequence of SEQ ID NO: 132.

In another aspect, the recombinant Lactobacillus reuteri host cells comprise a disrupted endogenous pduP gene which encodes a propionaldehyde dehydrogenase. Disruption to pduP may reduce depletion of propanal generated in the methylglyoxal pathway of Figures 1 and 2. In some of these embodiments, the Lactobacillus host cells produce a greater amount of n-propanol compared to the cell without the disruption when cultivated under identical conditions. In some embodiments, the endogenous pduP gene is inactivated. In one embodiment, the disrupted pduP gene has the coding sequence shown in SEQ ID NO: 133, which encodes a polypeptide having the amino acid sequence of SEQ ID NO: 134.

In another aspect, the recombinant Lactobacillus reuteri host cells comprise a disrupted endogenous acetate kinase gene, which may decrease the amount of cellular acetate kinase and increase the amount of available acetyl-CoA for conversion to isopropanol (as described in USSN 61/653,908, the content of which is hereby incorporated by reference). In some embodiments, the Lactobacillus host cells produce a greater amount of isopropanol compared to the cell without the acetate kinase disruption when cultivated under identical conditions. In some embodiments, the endogenous acetate kinase gene is inactivated. In one embodiment, the endogenous acetate kinase gene has the coding sequence shown in SEQ ID NO: 59, which encodes a polypeptide having the amino acid sequence of SEQ ID NO: 60.

Modeling can also be used to design gene disruptions 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 n-propanol and isopropanol. One exemplary 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).

The recombinant Lactobacillus reuteri host cells comprising a disrupted endogenous gene may be constructed using methods well known in the art, including those methods described herein. A portion of the gene can be disrupted such as the coding region or a control sequence required for expression of the coding region. Such a control sequence of the gene may be a promoter sequence or a functional part thereof, i.e., a part that is sufficient for affecting expression of the gene. For example, a promoter sequence may be inactivated resulting in no expression or a weaker promoter may be substituted for the native promoter sequence to reduce expression of the coding sequence. Other control sequences for possible modification include, but are not limited to, a leader, propeptide sequence, signal sequence, transcription terminator, and transcriptional activator.

The recombinant Lactobacillus reuteri host cells comprising a disrupted endogenous gene may be constructed by gene deletion techniques to eliminate or reduce expression of the gene. Gene deletion techniques enable the partial or complete removal of the gene thereby eliminating their expression. In such methods, deletion of the gene is accomplished by homologous recombination using a plasmid that has been constructed to contiguously contain the 5' and 3' regions flanking the gene.

The recombinant Lactobacillus reuteri host cells comprising a disrupted endogenous gene may also be constructed by introducing, substituting, and/or removing one or more (several) nucleotides in the gene or a control sequence thereof required for the transcription or translation thereof. For example, nucleotides may be inserted or removed for the introduction of a stop codon, the removal of the start codon, or a frame-shift of the open reading frame. Such a modification may be accomplished by site-directed mutagenesis or PCR generated mutagenesis in accordance with methods known in the art. See, for example, Botstein and Shortle, Science 1985, 229, 4719; Lo et al., Proc. Natl. Acad. Sci. U.S.A. 1985, 81, 2285; Higuchi et al., Nucleic Acids Res 1988, 16, 7351 ; Shimada, Meth. Mol. Biol. 1996, 57, 157; Ho et al., Gene 1989, 77, 61 ; Horton et al., Gene 1989, 77, 61 ; and Sarkar and Sommer, BioTechniques 1990, 8, 404.

The recombinant Lactobacillus reuteri host cells comprising a disrupted endogenous gene may also be constructed by gene disruption techniques by inserting into the gene a disruptive nucleic acid construct comprising a nucleic acid fragment homologous to the gene that will create a duplication of the region of homology and incorporate construct DNA between the duplicated regions. Such a gene disruption can eliminate gene expression if the inserted construct separates the promoter of the gene from the coding region or interrupts the coding sequence such that a non-functional gene product results. A disrupting construct may be simply a selectable marker gene accompanied by 5' and 3' regions homologous to the gene. The selectable marker enables identification of transformants containing the disrupted gene.

The recombinant Lactobacillus reuteri host cells comprising a disrupted endogenous gene may also be constructed by the process of gene conversion (see, for example, Iglesias and Trautner, Molecular General Genetics 1983, 189, 73-76). For example, in the gene conversion method, a nucleotide sequence corresponding to the gene is mutagenized in vitro to produce a defective nucleotide sequence, which is then transformed into the Lactobacillus strain to produce a defective gene. By homologous recombination, the defective nucleotide sequence replaces the endogenous gene. It may be desirable that the defective nucleotide sequence also comprises a marker for selection of transformants containing the defective gene.

The recombinant Lactobacillus reuteri host cells comprising a disrupted endogenous gene may be further constructed by random or specific mutagenesis using methods well known in the art, including, but not limited to, chemical mutagenesis (see, for example, Hopwood, The Isolation of Mutants in Methods in Microbiology (J.R. Norris and D.W. Ribbons, eds.) pp. 363-433, Academic Press, New York, 1970). Modification of the gene may be performed by subjecting the parent strain to mutagenesis and screening for mutant strains in which expression of the gene has been reduced or inactivated. The mutagenesis, which may be specific or random, may be performed, for example, by use of a suitable physical or chemical mutagenizing agent, use of a suitable oligonucleotide, or subjecting the DNA sequence to PCR generated mutagenesis. Furthermore, the mutagenesis may be performed by use of any combination of these mutagenizing methods.

Examples of a physical or chemical mutagenizing agent suitable for the present purpose include ultraviolet (UV) irradiation, hydroxylamine, N-methyl-N'-nitro-N- nitrosoguanidine (MNNG), N-methyl-N'-nitrosogaunidine (NTG) O-methyl hydroxylamine, nitrous acid, ethyl methane sulphonate (EMS), sodium bisulphite, formic acid, and nucleotide analogues. When such agents are used, the mutagenesis is typically performed by incubating the parent strain to be mutagenized in the presence of the mutagenizing agent of choice under suitable conditions, and selecting for mutants exhibiting reduced or no expression of the gene.

A nucleotide sequence homologous or complementary to a gene described herein may be used from other microbial sources to disrupt the corresponding gene in the recombinant Lactobacillus reuteri host cell.

In one aspect, the modification of a gene in the recombinant Lactobacillus reuteri host cell is unmarked with a selectable marker. Removal of the selectable marker gene may be accomplished by culturing the mutants on a counter-selection medium. Where the selectable marker gene contains repeats flanking its 5' and 3' ends, the repeats will facilitate the looping out of the selectable marker gene by homologous recombination when the mutant strain is submitted to counter-selection. The selectable marker gene may also be removed by homologous recombination by introducing into the mutant strain a nucleic acid fragment comprising 5' and 3' regions of the defective gene, but lacking the selectable marker gene, followed by selecting on the counter-selection medium. By homologous recombination, the defective gene containing the selectable marker gene is replaced with the nucleic acid fragment lacking the selectable marker gene. Other methods known in the art may also be used.

Methods of producing n-propanol and coproduction of n-propanol/isopropanol

The recombinant Lactobacillus reuteri host cells described herein may be used for the production of n-propanol and the coproduction of n-propanol + isopropanol. In one aspect is a method of producing n-propanol, comprising: (a) cultivating any one of the recombinant Lactobacillus reuteri host cells comprising a heterologous polynucleotide encoding a methylglyoxal synthase described herein in a medium under suitable conditions to produce the n-propanol; and (b) recovering the n-propanol. In another aspect is a method of producing n-propanol and isopropanol, comprising: (a) cultivating any one of the recombinant Lactobacillus reuteri host cells comprising a heterologous polynucleotide encoding a methylglyoxal synthase and an active isopropanol pathway described herein in a medium under suitable conditions to produce n-propanol and isopropanol; and (b) recovering the n-propanol and isopropanol.

The recombinant Lactobacillus reuteri host cells may be cultivated in a nutrient medium suitable for isopropanol production using methods well known in the art. For example, the cell may be cultivated by shake flask cultivation, and small-scale or large- scale fermentation (including continuous, batch, fed-batch, or solid state fermentations) in laboratory or industrial fermentors performed in a suitable fermentation medium and under conditions allowing isopropanol production.

The recombinant Lactobacillus reuteri host cells may produce n-propanol or coproduce n-propanol + isopropanol in a fermentable medium comprising any one or more (e.g., two, several) sugars, such as glucose, fructose, sucrose, cellobiose, xylose, xylulose, arabinose, mannose, galactose, and/or soluble oligosaccharides. In some instances, the fermentable medium is derived from a natural source, such as sugar cane, starch, or cellulose, and may be the result of pretreating the source by enzymatic hydrolysis (saccharification). In some aspects, the fermentable medium comprises sugar cane juice. Suitable media are available from commercial suppliers, may be prepared according to published compositions (e.g., in catalogues of the American Type Culture Collection), or may be prepared from commercially available ingredients.

In one aspect, the recombinant Lactobacillus reuteri host cells are cultivated in the presence of fructose, such as at least 0.1 %, 0.25%, 0.5%, 0.75%, 1 %, 2%, 3%, 4%, 5% or 10% fructose. In one aspect, the host cells are cultivated in the presence of 1 ,2- propanediol, such as at least 0.1 %, 0.25%, 0.5%, 0.75%, 1 %, 2%, 3%, 4%, 5% or 10% 1 ,2- propanediol.

In addition to the appropriate carbon sources from one or more (e.g., two, several) sugar(s), the fermentable medium may contain other nutrients or stimulators known to those skilled in the art, such as macronutrients (e.g., nitrogen sources) and micronutrients (e.g., vitamins, mineral salts, and metallic cofactors). In some aspects, the carbon source can be preferentially supplied with at least one nitrogen source, such as yeast extract, N₂, peptone (e.g., Bacto™ Peptone), or soytone (e.g., Bacto™ Soytone). Nonlimiting examples of vitamins include multivitamins, biotin, pantothenate, nicotinic acid, meso-inositol, thiamine, pyridoxine, para-aminobenzoic acid, folic acid, riboflavin, and Vitamins A, B, C, D, and E. Examples of mineral salts and metallic cofactors include, but are not limited to Na, P, K, Mg, S, Ca, Fe, Zn, Mn, Co, and Cu.

Suitable conditions used for the methods of propanol production may be determined by one skilled in the art in light of the teachings herein. In some aspects of the methods, the host cells are cultivated for about 12 hours to about 216 hours, such as about 24 hours to about 144 hours, or about 36 hours to about 96 hours. The temperature is typically between about 26°C to about 60°C, e.g., about 34°C to about 50°C, and at a pH of about 3.0 to about 8.0, such as about 3.0 to about 7.0, about 3.0 to about 6.0, about 3.0 to about 5.0, about 3.5 to about 4.5, about 4.0 to about 8.0, about 4.0 to about 7.0, about 4.0 to about 6.0, about 4.0 to about 5.0, about 5.0 to about 8.0, about 5.0 to about 7.0, or about 5.0 to about 6.0 or less than about 6.0, less than about 5.5, less than about 5.0, less than about 4.5, less than about 4.0, less than about 3.5, less than about 3.0, or less than about 2.5. In some aspects of the methods, the resulting intracellular pH of the host cell is about 2.0 to about 8.0, such as about 2.0 to about 7.0, about 2.0 to about 6.0, about 2.0 to about 5.0, about 1 .5 to about 4.5, about 3.0 to about 8.0, such as about 3.0 to about 7.0, about 3.0 to about 6.0, about 3.0 to about 5.0, about 3.5 to about 4.5, about 4.0 to about 8.0, about 4.0 to about 7.0, about 4.0 to about 6.0, about 4.0 to about 5.0, about 5.0 to about 8.0, about 5.0 to about 7.0, or about 5.0 to about 6.0, or less than about 6.0, less than about 5.5, less than about 5.0, less than about 4.5, less than about 4.0, less than about 3.5, less than about 3.0, or less than about 2.5. Cultivation may be performed under anaerobic, microaerobic, or aerobic conditions, as appropriate. In some aspects, the cultivation is performed under anaerobic conditions.

Cultivation may be performed under anaerobic, substantially anaerobic (microaerobic), or aerobic conditions, as appropriate. Briefly, anaerobic refers to an environment devoid of oxygen, substantially anaerobic (microaerobic) refers to an environment in which the concentration of oxygen is less than air, and aerobic refers to an environment wherein the oxygen concentration is approximately equal to or greater than that of the air. Substantially anaerobic conditions include, for example, a culture, batch fermentation or continuous fermentation such that the dissolved oxygen concentration in the medium remains less than 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/C0₂ mixture or other suitable non-oxygen gas or gases. In some embodiments, the cultivation is performed under anaerobic conditions or substantially anaerobic conditions.

The methods of described herein can employ any suitable fermentation operation mode. For example, a batch mode fermentation may be used with a close system where culture media and host microorganism, set at the beginning of fermentation, have no additional input except for the reagents certain reagents, e.g., for pH control, foam control or others required for process sustenance. The process described herein can also be employed in Fed-batch or continuous mode.

The methods described herein may be practiced in several bioreactor configurations, such as stirred tank, bubble column, airlift reactor and others known to those skilled in the art. The methods may be performed in free cell culture or in immobilized cell culture as appropriate. Any material support for immobilized cell culture may be used, such as alginates, fibrous bed, or argyle materials such as chrysotile, montmorillonite KSF and montmorillonite K-10.

In one aspect of the methods, the n-propanol, isopropanol, or combined n-propanol + isopropanol is produced at a titer greater than about 1 g/L, e.g., greater than about 2 g/L, 5 g/L, 10 g/L, 25 g/L, 50 g/L, 75 g/L, 100 g/L, 125 g/L, 150 g/L, 160 g/L, 170 g/L, 180 g/L, 190 g/L, 200 g/L, 210 g/L, 225 g/L, 250 g/L, 275 g/L, 300 g/L, 325 g/L, 350 g/L, 400 g/L, or 500g/L; or between about 10 g/L and about 500 g/L, e.g., between about 50 g/L and about 350 g/L, about 100 g/L and about 300 g/L, about 150 g/L and about 250 g/L, about 175 g/L and about 225 g/L, or about 190 g/L and about 210 g/L. In one aspect of the methods, the n-propanol and/or isopropanol is produced at a titer greater than about 0.01 gram per gram of carbohydrate, e.g., greater than about 0.02, 0.05, 0.75, 0.1 , 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, or 1.0 gram per gram of carbohydrate.

In one aspect of the methods, the amount of produced n-propanol is at least 5%, e.g., at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 50%, or at least 100% greater compared to cultivating the host cell without the heterologous polynucleotides encoding the methylglyoxal synthase when cultivated under the same conditions.

The n-propanol and isopropanol can be optionally recovered from the fermentation medium using any procedure known in the art including, but not limited to, chromatography (e.g., size exclusion chromatography, adsorption chromatography, ion exchange chromatography), electrophoretic procedures, differential solubility, osmosis, distillation, extraction (e.g., liquid-liquid extraction), pervaporation, extractive filtration, membrane filtration, membrane separation, reverse, or ultrafiltration. In one aspect, the isopropanol is separated from other fermented material and purified by conventional methods of distillation. Accordingly, in one aspect, the method further comprises purifying the recovered isopropanol by distillation.

The recombinant n-propanol and isopropanol may also be purified by the chemical conversion of impurities (contaminants) to products more easily removed from isopropanol by the procedures described above (e.g., chromatography, electrophoretic procedures, differential solubility, distillation, or extraction) and/or by direct chemical conversion of impurities to isopropanol. For example, in one aspect, the method further comprises purifying the n-propanol by converting propanal contaminant to n-propanol, or purifying isopropanol by converting acetone contaminant to isopropanol. Conversion of propanal to n-propanol and acetone to isopropanol may be accomplished using any suitable reducing agent known in the art (e.g., lithium aluminium hydride (LiAIH₄), a sodium species (such as sodium amalgam or sodium borohydride (NaBH₄)), tin species (such as tin(ll) chloride), hydrazine, zinc-mercury amalgam (Zn(Hg)), diisobutylaluminum hydride (DIBAH), oxalic acid (C₂H₂0₄), formic acid (HCOOH), Ascorbic acid, iron species (such as iron(ll) sulfate), and the like).

In some aspects of the methods, the recombinant propanol preparation before and/or after being optionally purified is substantially pure. With respect to the methods of producing n-propanol and coproducing n-propanol + isopropanol, "substantially pure" intends a recovered preparation that contains no more than 15% impurity, wherein impurity intends compounds other than propanol but does not include either propanol isomer. This, a substantially pure preparation of isopropanol may contain n-propanol in excess of 15%. In one variation, a substantially pure preparation is provided wherein the preparation contains no more than 25% impurity, or no more than 20% impurity, or no more than 10% impurity, or no more than 5% impurity, or no more than 3% impurity, or no more than 1 % impurity, or no more than 0.5% impurity.

The n-propanol and isopropanol produced by any of the methods described herein may be converted to propylene. Propylene can be produced by the chemical dehydration of isopropanol using acidic catalysts known in the art, such as acidic alumina, zeolites, and other metallic oxides; acidic organic-sulfonic acid resins; mineral acids such as phosphoric and sulfuric acids; and Lewis acids such as boron trifluoride and aluminum compounds (March, Jerry. Advanced Organic Chemistry. New York: John Wiley and Sons, 1992). Suitable temperatures for dehydration of isopropanol to propylene typically range from about 180°C to about 600°C, e.g., 300°C to about 500°C, or 350°C to about 450°C.

The dehydration reaction of n-propanol and/or isopropanol is typically conduced in an adiabatic or isothermal reactor, which can also be a fixed or a fluidized bed reactor; and can be optimized using residence time ranging from about 0.1 to about 60 seconds, e.g., from about 1 to about 30 seconds. Non-converted alcohol can be recycled to the dehydration reactor.

In one aspect is a method of producing propylene, comprising: (a) cultivating any one of the recombinant Lactobacillus reuteri host cells comprising a heterologous polynucleotide encoding a methylglyoxal synthase described herein in a medium under suitable conditions to produce the n-propanol; (b) recovering the n-propanol; (c) dehydrating the n-propanol under suitable conditions to produce propylene; and (d) recovering the propylene.

In another aspect is a method of producing propylene, comprising: (a) cultivating any one of the recombinant Lactobacillus reuteri host cell comprising a heterologous polynucleotide encoding a methylglyoxal synthase and an active isopropanol pathway described herein in a medium under suitable conditions to produce n-propanol and isopropanol; (b) recovering the n-propanol and isopropanol; (c) dehydrating the n-propanol and isopropanol under suitable conditions to produce propylene; and (d) recovering the propylene.

Contaminants that may be generated during dehydration may be removed through purification using techniques known in the art. For example, propylene can be washed with water or a caustic solution to remove acidic compounds like carbon dioxide and/or fed into beds to absorb polar compounds like water or for the removal of, e.g., carbon monoxide. Alternatively, a distillation column can be used to separate higher hydrocarbons such as propane, butane, butylene and higher compounds. The separation of propylene from contaminants like ethylene may be carried out by methods known in the art, such as cryogenic distillation.

Suitable assays to test for the production of n-propanol, isopropanol and propylene for the methods of production and host cells described herein can be performed using methods known in the art. For example, final n-propanol and isopropanol product, as well as intermediates (e.g., acetone, propanal) 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 n-propanol and isopropanol in the fermentation broth can also be tested with the culture supernatant. Byproducts and residual sugar in the fermentation medium (e.g., 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 using other suitable assay and detection methods well known in the art.

The propylene produced from n-propanol and isopropanol may be further converted to polypropylene or polypropylene copolymers by polymerization processes known in the art. Suitable temperatures typically range from about 105°C to about 300°C for bulk polymerization, or from about 50°C to about 100°C for polymerization in suspension. Alternatively, polypropylene can be produced in a gas phase reactor in the presence of a polymerization catalyst such as Ziegler-Natta or metalocene catalysts with temperatures ranging from about 60°C to about 80°C.

The following examples are provided by way of illustration and are not intended to be limiting of the invention.

Examples

Chemicals used as buffers and substrates were commercial products of at least reagent grade.

Base Host Strains

Lactobacillus reuteri SJ10655 (Q4ZXV)

Strain Lactobacillus reuteri DSM20016 was obtained from a public strain collection. This strain was subcultured in MRS medium, and an aliquot frozen as SJ10468. SJ10468 was inoculated into MRS medium, propagated without shaking for one day at 37°C, and spread on MRS agar plates to obtain single colonies. After two days of growth at 37°C, a single colony was reisolated on a MRS agar plate, the plate incubated at 37°C for three days, and the cell growth on the plate was scraped off and stored in the strain collection as SJ 10655 (alternative name: 04ZXV).

The same cell growth was used to inoculate a 10 ml MRS culture, which was incubated without shaking at 37°C for 3 days, whereafter cells were harvested by centrifugation and genomic DNA was prepared using a QIAamp DNA Blood Kit (Qiagen, Hilden, Germany) and sent for genome sequencing.

The genome sequence revealed that the isolate SJ 10655 (04ZXV) has a genome essentially identical to that of JCM1 1 12, rather than to that of the closely related strain DSM20016. JCM1 1 12 and DSM20016 are derived from the same original isolate, L. reuteri F275 (Morita et al. DNA research, 2008, 15, 151 -161 .) Lactobacillus reuteri SJ11044

Strain Lactobacillus reuteri SJ1 1044 was derived from strain SJ 10655 (04ZXV) as described in U.S. Provisional Application No. 61/653,908.

Lactobacillus reuteri SJ11294 (SJ11400)

L. reuteri strain SJ1 1294 (SJ 1 1400) is a modified version of 04ZXV (supra) which has improved transformation efficiency from a disrupted gene encoding a specificity subunit (LAR_0818) of a type I restriction modification system, as described in U.S. Provisional Application No. 61/648,958.

Lactobacillus reuteri SJ11360

L. reuteri strain SJ1 1360 is strain SJ 1 1294 (supra) transformed with the empty vector pSJ10600 (Figure 3), as described in U.S. Provisional Application No. 61/720,832.

Lactobacillus reuteri TRGU975 and SJ11538

L. reuteri strains TRGU975 and SJ1 1538 are modified versions of L. reuteri strains 04ZXV and SJ1 1400, respectively, having a disrupted acetate kinase gene (SEQ ID NO: 59), as described in U.S. Provisional Application No. 61/653,908.

Bacillus subtilis DN 1885 (see Diderichsen et al. J. Bacteriol. 1990, 172, 4315-4321 ).

Escherichia coli SJ2 (see Diderichsen et al. J. Bacteriol. 1990, 172, 4315-4321 ).

Escherichia coli MG1655 (see Blattner et al. Science 1997, 277, 1453-1462). Escherichia coli TG1

TG1 is a commonly used cloning strain and was obtained from a commercial supplier; it has the following genotype: F'[traD36 laclq A(lacZ) M15 proA+B+] glnV (supE) thi-1 A(mcrB-hsdSM)5 (rK- mK- McrB-) thi A(lac-proAB).

Media

LB plates were composed of 37 g LB agar (Sigma cat no. L3027) and double distilled water to 1 L.

LBPGS plates were composed of 37 g LB agar (Sigma cat no. L3027), 0.5% starch (Merck cat. no. 101252), 0.01 M K₂P0₄, 0.4% glucose, and double distilled water to 1 L.

TY bouillon medium was composed of 20 g tryptone (Difco cat no. 21 1699), 5 g yeast extract (Difco cat no. 212750), 7^*10^"3 g ferrochloride, 1 ^*10^"3 g manganese(ll)-chloride, 1 .5^*10^"3 g magnesium sulfate, and double distilled water to 1 L.

Minimal medium (MM) was composed of 20 g glucose, 1 .1 g KH₂P0₄, 8.9 g K₂HP0₄; 1 .0 g (NH₄)₂S0₄; 0.5 g Na-citrate; 5.0 g MgS0₄-7H₂0; 4.8 mg MnS0₄-H₂0; 2 mg thiamine; 0.4 mg/L biotin; 0.135 g FeCI₃-6H₂0; 10 mg ZnCI₂-4H₂0; 10 mg CaCI₂-6H₂0; 10 mg Na₂Mo0₄-2H₂0; 9.5 mg CuS0₄-5H₂0; 2.5 mg H₃B0₃; and double distilled water to 1 L, pH adjusted to 7 with HCI.

MRS medium was obtained from Difco™, as either Difco™ Lactobacilli MRS Agar or Difco™ Lactobacilli MRS Broth, having the following compositions— Difco™ Lactobacilli MRS Agar: Proteose Peptone No. 3 (10.0 g), Beef Extract (10.0 g), Yeast Extract (5.0 g), Dextrose (20.0 g), Polysorbate 80 (1 .0 g), Ammonium Citrate (2.0 g), Sodium Acetate (5.0 g), Magnesium Sulfate (0.1 g), Manganese Sulfate (0.05 g), Dipotassium Phosphate (2.0 g), Agar (15.0 g) and water to 1 L. Difco™ Lactobacilli MRS Broth: Consists of the same ingredients without the agar.

LC (Lactobacillus Carrying) medium (LCM) was composed of Trypticase (10 g), Tryptose (3 g), Yeast extract (5 g), KH₂P0₄ (3 g), Tween 80 (1 ml), sodium-acetate (1 g), ammonium citrate (1 .5 g), Cystein-HCI (0.2 g), MgS0₄.7H₂0 (12 mg), FeS0₄.7H₂0 (0.68 mg), MnS0₄.2H₂0 (25 mg), and double distilled water to 1 L, pH adjusted to 7.0. Stearile glucose is added after autoclaving, to 1 % (5 ml of a 20 % glucose stock solution/100 ml medium).

Transformation protocols.

Lactobacillus strains

Unless noted otherwise, plasmid DNA constructed in E. coli was purified from 2 ml of an overnight culture grown in TY medium, and supplemented with appropriate antibiotics using a QIAprep Spin Miniprep Kit (Qiagen, Hilden, Germany) as described by the manufacturer. The plasmid DNA was recovered in a volume of 50 microliters, and one microliter of this plasmid preparation was used for electroporation of Lactobacillus.

Plasmid DNA was transformed into Lactobacillus strains by electroporation. The L. reuteri strains were prepared for electroporation as follows: The strain was inoculated from a frozen stock culture into LCM medium, and incubated without shaking at 37°C overnight. A 5 ml aliquot was transferred into 500 ml LCM medium and incubated at 37°C without shaking until OD₆oo reached approximately 0.8. The cells were harvested by centrifugation as above, resuspended and washed 2 times in 50 ml of ion-exchanged sterile water at room temperature, and harvested by centrifugation. The cells were finally gently resuspended in 2.5 ml of 30% PEG1500, and 50 microliter aliquots were quickly frozen in an alcohol/dry ice bath, and stored at -80°C until use. Variations to the electroporation procedures below are described in the respective examples.

Electroporation procedure A: The frozen cells were thawed on ice, and 2 microliter of a DNA suspension in TE buffer was added. 40 microliters of the mixture was transferred to an ice-cold 2 mm electroporation cuvette, kept on ice for 1 -3 minutes, and electroporation carried out in a BioRad Gene Pulser™ with a setting of 1 .5 kV; 25 microFarad; 400 Ohms. 500 microliter of LCM was added, and the mixture incubated without shaking for 2 hours at 37°C before plating. Cells were plated on either LCM agar plates (LCM medium solidified with % agar) or MRS agar plates, supplemented with the required antibiotics, and incubated in an anaerobic chamber (Oxoid; equipped with Anaerogen sachet).

Electroporation procedure B: The frozen cells were thawed on ice, and 1 microliter of a DNA suspension in TE buffer was added. 40 microliters of the mixture was transferred to an ice-cold 1 mm electroporation cuvette, kept on ice for 1 -3 minutes, and electroporation carried out in a BioRad Gene Pulser™ with a setting of 1 .2 kV; 25 microFarad; 400 Ohms. 500 microliter of LCM was added, and the mixture incubated without shaking for 4 hours at 37°C before plating on MRS agar plates, supplemented with the required antibiotics, and incubation in an anaerobic chamber.

E. coli strains

Transformation of E. coli was conducted by electroporation using either a BioRad Gene Pulser™ (BioRad, Hercules, CA, USA) as described by the manufacturer, or by using chemically competent cells prepared following ordinary textbook procedures commonly known in the art. Example 1 : Preparation of an adhE-lpduP- double knockout Lactobacillus reuteri strain (TRGU1013).

MRS medium containing 5 μg ml erythromycin was inoculated with a L. reuteri MM4 strain harboring pJP042 (Pijkeren and Britton Nuc. Acids Res. 2012, 1-13; Figure 9) and incubated overnight at 37°C. The strain was subcultured in 10 ml MRS containing 5 μg ml erythromycin to OD₆oo=0.1 . The culture was incubated at 37°C for approximately 4 hours to OD₆oo=0.8 and centrifuged at 8000 x g for 5 minutes. The supernatant was discarded and the cells were resuspended in 10 ml SET buffer (0.1 M NaCI, 1 mM EDTA, 10 mM Tris-CI). The suspension was centrifuged at 8000 x g for 5 minutes and the supernatant was discarded. The cells were then resuspended in 1 ml lysis buffer (6.7 % saccharose, 50 mM Tris-CI pH 8, 0.1 mM EDTA). Lysozyme was added to 10 mg/ml and the mixture was incubated at 37 °C for 1 hour. The lysate was then centrifuged at 8000 x g for 5 minutes. The plasmid pJP042 DNA was isolated from the supernatant using a PureYield™ MiniPrep kit (Promega, USA) following the directions of the manufacturer.

JCM1 1 12 cells were made competent from an overnight culture in MRS containing 5 μg/ml erythromycin by subculturing in 40 ml MRS containing 5 μg/ml erythromycin to OD=0.1 and harvested at OD=0.8. The cells were kept on ice and washed carefully twice with 40 ml ice cold Wash Buffer (0.5M sucrose, 10% (V/V) glycerol), and resuspended in 400 μΙ Wash Buffer.

5 μΙ of isolated pJP042 (supra) was added to 100 μΙ freshly prepared competent cells (supra) and electroporated in a BioRad Gene Pulser™ with a setting of 2.5 kV, 25 microFarad and 400 Ohms. To this was added 1 ml MRS medium and the cells incubated at 37 °C for 3 hours. The electroporated cells were incubated anaerobically overnight at 37°C on MRS agar plates (MRS medium containing 15 g/l Bacto Agar) containing 5 μg/ml erythromycin for selection of pJP042 transformants. Erythromycin resistant colonies were checked for presence of pJP042 with colony PCR using primers flanking the recT1. Out of 1 1 transformants, 2 were isolated and confirmed to harbor pJP042. One of these strains was stored as TRGU768 in 10% glycerol at -80 °C.

To disrupt the adhE gene (LAR_0310) by recombineering, four oligonucleotides below were designed using PyRec 3.1 (obtained from Robert Britton, Microbial Genomics Laboratory, Michigan State University, Ml, USA).

o310: 5'-CAAGA AACAA GTTGA AAAGA AAGAA TTAAC TGCTG AAGAA AAGCT TTAAA ACGCC CAAAA GCTAG TTGAC GATTT AATGA CTAAG AGTCA-3' (SEQ ID NO: 153)

o310_fwd: 5'-AGGGT GTTGG AGTAA TGCGG T-3' (SEQ ID NO: 154)

o310_mama: 5'-AGAAA GAATT AACTG CTGAA GAAAA GCTTT-3' (SEQ ID NO: 155) o310_rev: 5'-TGAAT GATAG TGATT ATGAC GTTAA AGATC-3' (SEQ ID NO: 156)

The four oligonucleotides were designed to construct and screen for mutants with an in-frame stop codon and a Hind\\\ restriction site. Sequence o310 was used for the recombineering and incorporation of the nucleotides GCTTT which in the complementary direction implements a stop codon in the reading frame and thus results in disruption of gene translation. Oligonucleotides o310_fwd, o310_mama, and o310_rev were used in a PCR screen of all colonies screened. A 578 bp amplicon indicates that the mutations had been incorporated, whereas a single 1031 bp amplicon indicates that the o310_mama primer did not anneal due to the mismatch between oligo and the wild type sequence.

An overnight culture of TRGU768 was subcultured in 40 ml MRS medium containing 5 μg ml erythromycin to OD₆oo 0.1 . After approximately 2 hours incubation at 37°C, OD₆oo reached approximately 0.55 and recT1 expression was induced by addition of induction peptide (8 μΙ; 50 Mg/ml) MAGNSSNFIHKIKQIFTHR (SEQ ID NO: 161 ). The incubation at 37°C was prolonged for 20 minutes. Competent cells were then prepared by centrifugation and washing of the cells twice in 40 ml ice-cold Wash-Buffer (0.5M sucrose, 10% (v/v) glycerol). Finally the cells were resuspended in 800 ul Wash Buffer. 100 μΙ of the resuspended cells was used for each transformation. The cells were then transformed by electroporation with 5 μΙ o819 (20 μ9/μΙ) as described in procedure A above. After 2 hours incubation in 1 ml MRS medium at 37°C, the cells were incubated anaerobically overnight on MRS agar plates.

After overnight incubation, 192 individual colonies were analyzed by PCR using o310_fwd, o310_mama, and o310_rev. After overnight incubation, 192 colonies were screened with colony PCR using o310_fwd, o310_mama, and o310_rev. No colonies resulted in two bands of correct sizes as estimated by agarose gel electrophoresis. Prolonged anaerobic incubation of the transformation plates at 37°C, resulted in appearance of additional 7 colonies. These were tested with PCR as above and 1 colony, referred to as a mixed genotype mutant, resulted in two bands of correct sizes as estimated by agarose gel electrophoresis. This colony was streak purified on MRS agar plates containing 2% fructose and the following day, 94 colonies and the original mixed genotype mutant as control were checked with PCR with the primers o310_fwd, o310_mama, and o310_rev. The PCR reactions were analyzed on a 96 well gel and 18 colonies as well as the original isolated transformant resulted in two bands of correct sizes as estimated by agarose gel electrophoresis. Each of the 96 colony-PCR reactions were digested with Hind\\\ by addition of 2 μΙ Hind\\\ + 3 μΙ H₂0. The digested PCR reactions were analyzed on agarose gel electrophoresis and 16 colonies were shown to be pure genotypes by complete digestion of the 1031 bp band. All colonies were incubated overnight at 37°C in MRS and in MRS+10 microgram/ml erythromycin as well as on MRS agar plates +/- 10 microgram/ml erythromycin. All cultivations contained 2% fructose. This resulted for all strains in growth in media without erythromycin and one strain was able to grow in the presence of erythromycin. The erythromycin resistance indicated presence of plasmid pJP042. This strain was designated TRGU980 and stored in 10% glycerol at -80°C.

Strain TRGU980 contains a disruption to the coding sequence of the adhE gene

(SEQ ID NO: 131 ) which encodes the bifunctional alcohol/acetaldehyde dehydrogenase of SEQ ID NO: 132.

To disrupt the pduP gene (LAR_1623) by recombineering, four oligonucleotides below were designed using PyRec 3.1 (obtained from Robert Britton, Microbial Genomics Laboratory, Michigan State University, Ml, USA).

01623: 5'-GTAGT AGCTG CAACG TTTGC ACTTG AAGAG CTGGC ATTAT CCTAA GCTTC GGCAA GAATT TTGCG TACAG CACTT TCAAT ATCAT TAATC-3' (SEQ ID NO: 157)

o1623_fwd: 5'-ACAAC TAAAT TATGA AGGCC TGTTG C-3' (SEQ ID NO: 158)

o1623_mama: 5'-CGCAA AATTC TTGCC GAAGA ACTA-3' (SEQ ID NO: 159)

o1623_rev: 5'-ATAAT GCTTC TAAAA ATCTA TTTGA TCGGC-3' (SEQ ID NO: 160)

The four oligonucleotides were designed to construct and screen for mutants with an in-frame stop codon and a Hind\\\ restriction site. Sequence o1623 was used for the recombineering and incorporation of the nucleotides CTAAG which in the complementary direction implements a stop codon in the reading frame and thus results in disruption of gene translation. Oligonucleotides o1623_fwd, o1623_mama, and o1623_rev were used in a PCR screen of all colonies screened. A 566 bp amplicon indicates that the mutations had been incorporated, whereas a single 1026 bp amplicon indicates that the o1623_mama primer did not anneal due to the mismatch between oligo and the wild type sequence.

An overnight culture of TRGU980 was subcultured in 40 ml MRS medium containing

5 μg/ml erythromycin and 2% fructose to OD₆oo 0.1 . After approximately 3 hours incubation at 37°C, OD₆oo reached 0.53 and recT1 expression was induced by addition of induction peptide (8 μΙ; 50 microgram/ml) MAGNSSNFIHKIKQIFTHR (SEQ ID NO: 161 ). The incubation at 37°C was prolonged for 20 minutes. Competent cells were then prepared by centrifugation and washing of the cells twice in 40 ml ice-cold Wash-Buffer (0.5M sucrose, 10% (v/v) glycerol). Finally the cells were resuspended in 800 ul Wash Buffer. The cells were then transformed by electroporation with 5 μΙ o819 (20 g/μΙ) as described in procedure A above. After 2 hours incubation in 1 ml MRS medium containing 2% fructose at 37°C, the cells were incubated anaerobically overnight on MRS agar plates containing 2% fructose.

After overnight incubation, 96 colonies were screened with colony PCR using o1623_fwd, o1623_mama, and o1623_rev. Four colonies, referred to as a mixed genotype mutants, resulted in two bands of approximately correct sizes as estimated by agarose gel electrophoresis. 7 ul of each PCR reaction was then digested by addition of 4 ul H₂0, 0.5 ul NEB 4 buffer, and 0.5 ul Hind\\\ and incubation overnight at 37°C. After digest, all four colonies still resulted in both the 1026 bp and 566 bp bands when analyzed on agarose gel electrophoresis, indicating that all four colonies contained cells of both wild type and mutant genotypes. Three out of the four mixed genotype strains were then restreaked onto MRS agar plates containing 2% fructose. After overnight incubation at 37°C, 48 colonies of each were tested by colony PCR using primers o1623_fwd and o1623_rev. Half of each PCR reaction was then digested with Hind\\\ and agarose gel electrophoresis analysis showed that one of the 3 mixed genotype strains resulted in 5 isolates with pure genotypes. The other 2 resulted in no restriction of the 1026 bp band. DNA sequencing of all 5 pure genotype isolates verified that the correct mutations had been incorporated in LAR_1623. Four of the pure genotype mutants were incubated overnight in MRS medium containing 2% fructose at 30°C. The cultures were diluted 1 e5 fold and 100 microliters were plated onto MRS agar medium plates containing 2% fructose. After overnight anaerobic incubation at 37°C, 50 colonies were streaked out on MRS agar medium plates containing 2% fructose and +/- 10 microgram/ml erythromycin. 1 colony was picked randomly due to its sensitivity to 10 microgram/ml erythromycin and stored as TRGU1013 in MRS containing 2% fructose and 10% glycerol at -80°C.

Strain TRGU1013 contains a disruption to the coding sequence of the adhE gene (SEQ ID NO: 131 ) which encodes the bifunctional alcohol/acetaldehyde dehydrogenase of SEQ ID NO: 132, and a disruption to the coding sequence of the pduP gene (SEQ I D NO: 133) which encodes the propionaldehyde dehydrogenase of SEQ ID NO: 134.

Example 2: Conversion of lactaldehyde, hydroxyacetone, and methylglyoxal to n- propanol in Lactobacillus reuteri.

In order to determine the capacity of Lactobacillus reuteri to convert the intermediates methylglyoxal, hydroxyacetone and lactaldehyde into n-propanol (Figures 1 and 2), these precursor compounds were added to growing cultures of a selection of Lactobacillus reuteri strains and analyzed for product formation. Wildtype Lactobacillus reuteri was found to have the capacity to convert all these metabolic intermediates into 1 ,2- propanediol and n-propanol as shown in the following experiments.

Lactaldehyde addition (lactaldehyde -> 1 ,2-propanediol -> propanal -> n-propanol)

In order to determine the capacity of Lactobacillus reuteri to convert lactaldehyde to 1 ,2-propanediol and subsequently to n-propanol an experiment was performed in which lactaldehyde was added to the growth medium.

Lactobacillus reuteri strains SJ 1 1294 and TRGU1013 were inoculated from glycerol stocks in 10ml of MRS medium containing 10 microgram/ml of erythromycin and incubated under anaerobic conditions overnight at 37°C. The following day, 2 x 0.95 ml of each culture was harvested. The cells then were washed with fresh MRS-medium (pH adjusted to 4.0 with lactic acid) and subsequently inoculated in 2 x 0.95 ml of fresh MRS-medium (pH adjusted to 4.0 with lactic acid). The cultures were incubated for 2 hours at 37°C, followed by addition of lactaldehyde to a final concentration of either 0% (v/v) or 0.1221 % (v/v). The cultures were incubated overnight at 37°C. The following day samples of supernatant from the cultures were analyzed for n-propanol and 1 ,2-propanediol by GC- FI D. Samples were diluted 1 + 1 with 0.05% tetrahydrofuran in methanol and analyzed using the GC parameters are listed in Table 5. Results are shown in Table 6.

Table 5.

Table 6.

n-propanol 1 ,2-propanediol C-recovery^*

Strain Medium (fl/L) (fl/L) (lactaldehyde -> n-propanol)

MRS+0.1221 %

SJ 1 1294 (v/v) lactaldehyde 0.72 <0.1 >0.70

SJ 1 1294 MRS 0.03 <0.1 —

MRS+0.1221 %

TRGU1013 (v/v) lactaldehyde 0.71 <0.1 >0.69

TRGU1013 MRS 0.03 <0.1 —

^*C-recovery values were correcl ed for background level of n-propanol production

(Correlation factors: Lactaldehyde -> n-propanol: 1.232612)

Lactobacillus reuteri was found to convert lactaldehyde to n-propanol by at least 70% with both the adhElpduP mutant Lb. reuteri TRGU1013 and strain Lb. reuteri SJ1 1294. Since lactaldehyde was not analyzed in the supernatant, the total C-recovery may be even higher. Thus, if lactaldehyde can successfully be produced in strain Lb. reuteri SJ1 1294 by conversion of methylglyoxal to lactaldehyde, the produced lactaldehyde will eventually be converted to n-propanol.

Hydroxyacetone addition (hydroxyacetone -> 1 ,2-propanediol -> propanal -> n-propanol)

In order to determine the capacity of Lactobacillus reuteri to convert hydroxyacetone to 1 ,2-propanediol and subsequently to n-propanol an experiment was performed in which hydroxyacetone was added to the growth medium.

Lactobacillus reuteri strains SJ1 1294 and TRGU1013 were inoculated from glycerol stocks in 2 ml MRS medium containing +/-1 % (v/v) hydroxyacetone and incubated under anaerobic conditions overnight at 37°C. After two days, samples of supernatant were analyzed for n-propanol, 1 ,2-propanediol and hydroxyacetone using the parameters described supra. Results are shown in Table 7.

Table 7.

*C-recovery values were corrected for background signal.

(Correlation factors: Hydroxyacetone -> n-propanol: 1 .232612, Hydroxyacetone -> 1 ,2- propanediol: 0.973589) Lactobacillus reuteri was found to convert hydroxyacetone to n-propanol with both the adhElpduP mutant Lb. reuteri TRGU1013 and strain Lb. reuteri SJ1 1294. Strain Lb. reuteri TRGU1013, in which adhE and pduP were disrupted, showed the highest conversion of hydroxyacetone to n-propanol, possibly due to the greater availability of NAD(P)H for metabolism of hydroxyacetone to 1 ,2-propanediol and subsequently to n- propanol.

Methylglyoxal addition (methylglyoxal -> hvdroxyacetone/lactaldehvde -> 1 ,2-propanediol -> propanal -> n-propanol)

In order to determine the capacity of Lactobacillus reuteri to convert methylglyoxal to n-propanol using the methylglyoxal pathway shown in Figures 1 & 2, an experiment was performed in which methylglyoxal was added to the growth medium.

Lactobacillus reuteri strain SJ1 1400 was inoculated from glycerol stock in MRS medium and incubated under anaerobic conditions overnight at 37°C. The following day, 2 x 1 ml of the culture were harvested and washed with fresh MRS medium, followed by removal of the supernatant. The resulting cell pellets were then inoculated in 2 x 2 ml of fresh MRS medium +/- 0.2% (v/v) methylglyoxal and incubated overnight. The following day, samples of the supernatant were analyzed for n-propanol and 1 ,2-propanediol using the parameters described supra. Results are shown in Table 8.

Table 8.

As shown in Table 7, Lactobacillus reuteri was found to convert methylglyoxal to n- propanol without the use of any additional heterologous n-propanol pathway genes.

Example 3: Construction of plasmid constructs for expressing a methylglyoxal synthase gene.

Construction of pVS2-based vectors pSJ 10600 and pSJ 10603 for constitutive expression

A set of constitutive expression vectors were constructed based on the plasmid pVS2 (von Wright et al., Appl. Environ. Microbiol. 1987, 53, 1584-1588) and promoters described by Rud et al. (Rud et al. Microbiology 2006, 152, 101 1 -1019). A DNA fragment containing the P1 1 promoter with a selection of flanking restriction sites, and another fragment containing P27 with a selection of flanking restriction sites, was chemically synthesized by Geneart AG (Regenburg, Germany).

The DNA fragment containing P1 1 with flanking restriction sites, and the DNA fragment containing P27 with flanking restriction sites are shown in SEQ ID NOs: 49 and 50, respectively. Both DNA fragments were obtained in the form of DNA preparations, where the fragments had been inserted into the standard Geneart vector, pMA. The vector containing P1 1 was transformed into £ coli SJ2 cells, and a transformant kept as SJ 10560, containing plasmid pSJ10560. The vector containing P27 was transformed into £ coli SJ2 cells, and a transformant kept as SJ10561 , containing plasmid pSJ10561 .

The promoter-containing fragments, in the form of 176 bp Hindi 11 fragments, were excised from the Geneart vectors and ligated to Hindlll-digested pUC19. The P1 1 - containing fragment was excised from the vector prepared from SJ 10560, ligated to pUC19, and correct transformants of £ coli SJ2 were kept as SJ10585 and SJ10586, containing pSJ10585 and pSJ10586, respectively. The P27 containing fragment was excised from the vector prepared from SJ10561 , ligated to pUC19, and correct transformants of £. coli SJ2 were kept as SJ 10587 and SJ 10588, containing pSJ 10587 and pSJ 10588, respectively.

Plasmid pVS2 was obtained in Lactobacillus plantarum NC8, a strain kept as

SJ10491 , extracted from this strain by standard plasmid preparation procedures known in the art, and transformed into £ coli MG1655 selecting erythromycin resistance (200 microgram/ml) on LB agar plates at 37°C. Two such transformants were kept as SJ 10583 and SJ10584.

To insert P1 1 into pVS2, the P1 1 -containing 176 bp Hindi 11 fragment was excised and purified by agarose gel electrophoresis from pSJ10585, and ligated to Hindlll-digested pVS2, which had been prepared from SJ 10583. The ligation mixture was transformed by electroporation into £ coli MG1655, selecting erythromycin resistance (200 microgram/ml) on LB agar plates, and two transformants, which both harbor plasmids with the promoter insert in one particular of the two possible orientations, were kept as SJ 10600 and SJ10601 , containing pSJ 10600 (Figure 3) and pSJ10601 .

Another transformant, having the promoter insert in the other of the two possible orientations, was kept as SJ10602, containing pSJ10602. The plasmid preparation from SJ 10602 appeared to contain less DNA than the comparable preparations from SJ 10600 and SJ10601 , and, upon further work, pSJ10602 appeared to be rather unstable, with deletion derivatives dominating in the plasmid population.

To insert P27 into pVS2, the P27-containing 176 bp Hind 111 fragment was excised and purified by agarose gel electrophoresis from pSJ10588, and ligated to Hindi ll-digested pVS2, which had been prepared from SJ 10583. The ligation mixture was transformed by electroporation into E. coli MG1655, selecting erythromycin resistance (200 microgram/ml) on LB agar plates, and two transformants, which both harbor plasmids with the promoter insert in one particular of the two possible orientations, were kept as SJ 10603 and SJ10604, containing pSJ 10603 (Figure 4) and pSJ10604.

Another transformant, having the promoter insert in the other of the two possible orientations, was kept as SJ10605, containing pSJ10605. The promoter orientation in this plasmid is the same as in pSJ10602, described above. The plasmid preparation from SJ 10605 appeared to contain less DNA than the comparable preparations from SJ 10603 and SJ10604, and, upon further work, pSJ10605 appeared to be rather unstable, with deletion derivatives dominating in the plasmid population.

Construction of plasmids pBK476 and pBKQ478 expressing a synthetic version of the methylglyoxal synthase coding sequence (rngsA Ec) from Escherichia coli K12 substr. MG1655

Plasmid pBKQ464 (Figure 10) was obtained from Geneart AG (Regenburg, Germany) containing a codon-optimized coding sequence (mgsA_Ec; SEQ ID NO: 126) encoding the Escherichia coli methylglyoxal synthase of SEQ ID NO: 1 12. The mgsA_Ec sequence was obtained from pBKQ464 by digestion with Nco\+Xba\, resulting in a 461 bp fragment. Plasmid vector pSJ 10600 and pSJ 10603 (supra) were digested with Nco\+Xba\ and treated with calf intestine alkaline phosphatase resulting in a 5.1 kb fragment. The fragment containing mgsA_Ec was then ligated to pSJ 10600 and pSJ 10603 and introduced to E. coli TG1 by electroporation as described above. The resulting plasmids pBKQ476 (Figure 12) and pBKQ478 (Figures 13) contained the mgsA_Ec sequence expressed from the synthetic promoter P1 1 or P27, respectively. The plasmid constructions were then introduced to Lb. reuteri SJ1 1400 by electroporation Procedure B as described above, plated on MRS medium containing 10 microgram/ml of erythromycin and incubated anaerobically at 37°C. The resulting strains were designated Lb. reuteri BKQ488 (contains pBKQ476) and Lb. reuteri BKQ490 (contains pBKQ478).

Construction of plasmids pBKQ557 and pBKQ559 expressing a synthetic version of the methylglyoxal synthase coding sequence (rngsA Bs) from Bacillus subtilis (strain 168)

Plasmid pBKQ466 (Figure 1 1 ) was obtained from Geneart AG (Regenburg, Germany) containing a codon-optimized coding sequence (mgsA_ Bs; SEQ ID NO: 127) encoding the Bacillus subtilis methylglyoxal synthase of SEQ ID NO: 1 13. The mgsA_Bs sequence was obtained from pBKQ466 by digestion with BspH\+Xba\, resulting in a 416 bp fragment. Plasmid vector pSJ 10600 and pSJ 10603 (supra) were digested with Nco\+Xba\ 5 and treated with calf intestine alkaline phosphatase resulting in a 5.1 kb fragment. The fragment containing mgsA_Bs was then ligated to pSJ 10600 and pSJ 10603 and introduced to E. coli TG1 by electroporation as described above. The resulting plasmids pBKQ557 (Figure 15) and pBKQ559 (Figure 16) contained the mgsA_Bs sequence expressed from the synthetic promoter P1 1 or P27, respectively. The plasmid constructions were then i o introduced to Lb. reuteri SJ1 1400 by electroporation Procedure B as described above, plated on MRS medium containing 10 microgram/ml of erythromycin and incubated anaerobically at 37°C. The resulting strains were designated Lb. reuteri BKQ577 (contains pBKQ557) and Lb. reuteri BKQ579 (contains pBKQ559).

Construction of plasmid pBKQ568 expressing a synthetic version of the methylglyoxal

15 synthase coding sequence (mgsA Lsak) from Lactobacillus sakei subsp. sakei 23K

Plasmid pBKQ535 (Figure 14) was obtained from Geneart AG (Regenburg, Germany) containing a coding sequence (mgsA_Lsak; SEQ ID NO: 130) encoding the Lactobacillus sakei methylglyoxal synthase of SEQ ID NO: 1 16. The mgsA_Lsak sequence was obtained from pBKQ535 by PCR amplification with primer pr059 (5'-GTGAG GGTAC

20 CCGGG TCTAG ATCAA GCATC GCGAT CCCTG C-3'; SEQ ID NO: 135) and primer pr060 (5'-GATCA TACGG TTTGC CATCA TG-3'; SEQ ID NO: 136) using Phusion DNA polymerase. The PCR fragment was purified with Qiagen PCR purification kit and then digested with SspHI, resulting in a 436 bp fragment. Plasmid vector pSJ 10603 (supra) was digested with Nco\ and treated with calf intestine alkaline phosphatase resulting in a 5.1 kb

25 fragment. The fragment containing mgsA_Lsak was then ligated to pSJ 10603 and introduced to E. coli TG1 by electroporation as described above. The resulting plasmid pBKQ568 (Figure 17) contained the mgsA_Lsak sequence expressed from the synthetic promoter P27. The plasmid construction was then introduced to Lb. reuteri SJ1 1400 by electroporation Procedure B as described above, plated on MRS medium containing 10

30 microgram/ml of erythromycin, and incubated anaerobically at 37°C. The resulting strain was designated Lb. reuteri BKQ586 (contains pBKQ568).

Construction of plasmid pBKQ621 expressing a synthetic version of the methylglyoxal synthase coding sequence (mgsA Blic) from Bacillus licheniformis

PCR amplification with primer set pr065 (5'-GTGCT TCATG AAAAT CGCCT TAATT GCCC-3'; SEQ ID NO: 137) and pr066 (5'-GTGAG GGTAC CCGGG TCTAG ATCAA GCATC GCGAT CCCTG C-3'; SEQ ID NO: 138) together with B. licheniformis chromosomal template DNA (e.g., see WO2005/069762 for genomic DNA isolation methods) and Phusion DNA polymerase, was used to amplify the mgsA_Blic coding 5 sequence (SEQ ID NO: 129) which encodes the Bacillus licheniformis methylglyoxal synthase of SEQ I D NO: 1 15. The PCR fragment was purified with Qiagen PCR purification kit and then digested with Bsp \+Kpn\, resulting in a 430 bp fragment. Plasmid vector pSJ 10603 (supra) was digested with Nco\+Kpn\ and treated with calf intestine alkaline phosphatase resulting in a 5.1 kb fragment. The fragment containing the mgsA_Blic i o sequence was then ligated to pSJ10603 and introduced to E. coli TG1 by electroporation as described above. The resulting plasmid pBKQ574 (Figure 18) contained the mgsA_Blic sequence expressed from the synthetic promoter P27. The plasmid construction was then introduced to Lb. reuteri SJ1 1400 by electroporation Procedure B as described above , plated on MRS medium containing 10 microgram/ml of erythromycin, and incubated

15 anaerobically at 37°C. The resulting strain was designated Lb. reuteri BKQ621 (contains pBKQ574).

Example 4: Production of n-propanol from Lactobacillus reuteri strains expressing a Bacillus subtilis methylglyoxal synthase gene sequence.

20 The ability of Lactobacillus reuteri to produce recombinant n-propanol with heterologous expression of a single methylglyoxal synthase gene sequence was analyzed using strains Lb. reuteri BKQ577 and Lb. reuteri pBKQ579 (supra), each containing the plasmids pBKQ557 and pBKQ559, respectively, which encode for the Bacillus subtilis methylglyoxal synthase of SEQ ID NO: 1 13. The Lb. reuteri strains were inoculated in MRS

25 medium supplemented with 10 microgram/ml of erythromycin and incubated anaerobically at 37°C for two days. Samples of supernatant from the cultures were then analyzed for n- propanol and 1 ,2-propanediol production using the parameters described supra. Results are shown in Table 9.

Strains BKQ577 and BKQ579 (containing a polynucleotide encoding a Bacillus subtilis methylglyoxal synthase) produced of up to 130 mg/L of n-propanol, compared to 30 mg/L of n-propanol in the Lb. reuteri SJ1 1360 (lacking a heterologous polynucleotide encoding a methylglyoxal synthase). Production of up to 0.49 g/L of 1 ,2-propanediol was also observed, which was significantly above the control level. As it is known that produced 1 ,2-propanediol may eventually be converted to n-propanol given sufficient time, the final levels of n-propanol would likely have been even higher had the cultures been incubated for a longer period of time.

A detailed study of Lb. reuteri BKQ577 and Lb. reuteri BKQ579 in pH controlled fermentor (pH 5.5) using sugar cane juice based medium consisting of the following components: Sugar cane juice diluted to BRIX 10; 1 mL/L pluronic; 10 g/L yeast extract (Bacto 212720); 1 g/L Tween 80; 25 mg/L MnS0₄,H₂0; 650 mg/L phytic acid; 1 g/L L- serine; 1 g/L L-threonine; 2 g/L KH₂P0₄; 400 mg/L uridine; 1 g/L glutamic acid monosodium salt; 10 microgram/ml of erythromycin. Uridine and glutamic acid monosodium salt was added after autoclavation. The fermentation temperature was 37°C and pH was kept constant at pH 5.5. Nitrogen was used to keep the culture anaerobic.

After 46 hours of cultivation, samples of supernatant were analyzed for n-propanol production. A production of 1 .8 g/L of n-propanol was observed for BKq579 and no 1 ,2- propanediol was detected. Strain BKq577 grown under identical conditions produced approximately 40 mg/L n-propanol and 1 .6 g/L 1 ,2-propanediol. Corresponding fermentation profiles for n-propanol production with BKq579 and BKq577 are shown in Figures 5 and 6, respectively. Typical titers for the parent strain (lacking a heterologous polynucleotide encoding a methylglyoxal synthase) grown under similar conditions are about 50 mg/L n-propanol and undetectable amounts of 1 ,2-propanediol. These data confirmed that expression of only a methylglyoxal synthase gene enables the Lactobacillus reuteri strain to produce n-propanol from sugar.

Example 5: Comparison of n-propanol production from Lactobacillus reuteri strains expressing different methylglyoxal synthase gene sequences.

The ability of Lactobacillus reuteri to produce recombinant n-propanol with heterologous expression of using different methylglyoxal synthase gene sequences was analyzed. Strains BKQ488 and BKQ490 (each containing a polynucleotide encoding the Escherichia coli methylglyoxal synthase of SEQ ID NO: 1 12, as described supra), BKQ577 and BKQ579 (each containing a heterologous polynucleotide encoding the Bacillus subtilis methylglyoxal synthase of SEQ ID NO: 1 13, as described supra), and BKQ586 (containing a heterologous polynucleotide encoding the Lactobacillus sakei methylglyoxal synthase of SEQ ID NO: 1 16, as described supra) were subjected to anaerobic growth at 37°C in pH adjusted (pH 5.5) fermentor using sugar cane juice supplemented with yeast extract and 10 5 microgram/ml of erythromycin. Samples of supernatant were analyzed for n-propanol production after growth for up to three days using the parameters described supra. Results are shown in Table 10.

Table 10.

i o Tested Lb. reuteri strains containing a polynucleotide encoding a methylglyoxal synthase were found to produce significantly higher n-propanol compared to the SJ1 1400 control strain (lacking a heterologous polynucleotide encoding a methylglyoxal synthase). Based on the amount of n-propanol produced, the most promising mgsA candidate was from Lb. sakei, where the highest concentration of n-propanol was measured to 1 .9 g/L.

15

Example 6: Increased methylglyoxal synthase activity in recombinant Lactobacillus reuteri strains expressing methylglyoxal synthase gene sequences.

Lactobacillus reuteri strains containing a heterologous polynucleotide encoding a methylglyoxal synthase were tested for methylglyoxal synthase activity. Strains BKQ490,

20 BKQ579, BKQ586, and BKQ621 {supra) were inoculated in 10 ml MRS medium supplemented with 10 microgram/ml of erythromycin and incubated anaerobically overnight at 37°C. The following day, the Lb. reuteri strains were inoculated in pre-heated (37°C) 40 ml MRS medium supplemented with 10 microgram/ml of erythromycin to OD600=0.1 . The cultures were incubated at 37°C until OD600=0.5-0.7. The cell cultures were then

25 harvested by centrifugation and lysates were prepared as follows:

A volume of 4 ml of each culture were harvested by centrifugation and washed with 1 ml H₂0 by resuspension followed by centrifugation to obtain cell pellets. The cells were then resuspended in 50 microliter of 0.1 M TRIS (pH 7.5) + 2mM DTT and added to Eppendorf tubes containing glass beads. The cells were crushed in Fastprep 5 x 45 sec

30 and placed on ice after each interval of 45 sec to avoid heating. To the samples was added 450 microliter of 0.1 M TRIS (pH 7.5) + 2mM DTT. The supernatant was subsequently purified from cell debris and glass beads by centrifugation and removal of the supernatant to new Eppendorf tubes. The samples were then stored at -20°C until use.

The lysates were analyzed for methylglyoxal synthase activity. 50 μΙ diluted cell lysate (5x, 25x and 100x diluted with 0.1 M imidazole, pH 7.0) was mixed with 50 μΙ 40 mM dihydroxyacetone phosphate (Sigma 51269), 50 μΙ 16 mM glutathione (Sigma G4251 ) and 50 μΙ Glyoxalase I from Saccharomyces cerevisiae (Sigma G4252, 1000x diluted) in the well of a UV transparent microtiter plate. Methylglyoxal synthase activity in the lysates reacts with dihydroxyacetone phosphate to produce methylglyoxal, which is then reacted with excess of reduced glutathione and glyoxalase I to form S-lactoylglutathione. The formation of S-lactoylglutathione is followed spectrophotometrically by measuring absorbance every 20 seconds for 15 min at 240 nm. Blank samples with 50 μΙ Milli Q water added instead of dihydroxyacetone phosphate were seen not to give increase in absorbance at 240 nm. Activities were calculated from slopes in the linear range of the assay multiplied by dilution factor of cell lysate. Results are shown in Table 1 1 .

Table 1 1 .

Lysates from all tested strains comprising a heterologous polynucleotide encoding a methylglyoxal synthase showed high methylglyoxal synthase activity. Corresponding methylglyoxal synthase activity for control strain (lacking a heterologous polynucleotide encoding a methylglyoxal synthase) was negligible. The experiment supports the finding that expression of mgsA alone in Lactobacillus reuteri strains enables conversion of sugar into n-propanol.

Example 7: Production of n-propanol from an adhE-lpduP- double knockout Lactobacillus reuteri strain containing a methylglyoxal synthase gene sequence (BKQ627).

Plasmids pBKQ557 and pBKQ559 (each containing heterologous polynucleotide encoding the B. subtilis methylglyoxal synthase of SEQ ID NO: 1 13 expressed from a synthetic promoter, as described supra) were introduced to Lb. reuteri TRGU1013 (supra) by electroporation Procedure B as described above, resulting in Lb. reuteri strains BKQ627 and BKQ629, respectively. The knockout of adhE, which codes for a bifunctional alcohol dehydrogenase, reduces the consumption of NAD(P)H for ethanol production, which may then be available for the methylglyoxal pathway. Indeed, when grown in MRS medium 5 supplemented with 0.5% of fructose, production of up to 220 mg/L of n-propanol was measured in the adhE/pduP knockout mutant containing mgsA from B. subtilis using the parameters described supra (see Table 12).

i o Strain Lb. reuteri BKQ627 was then studied in more detail in a pH adjusted fermenter (pH 5.5) using sugar cane juice based medium using the same fermentation conditions as described in Example 3. After 42 hours of fermentation the n-propanol titer reached 2.3 g/L, which was an improvement by almost 22% when compared to the n- propanol from Lb. reuteri BKQ577 (supra) which contains the same heterologous

15 polynucleotide encoding the B. subtilis methylglyoxal synthase.

In another fermentation performed with sugar cane juice based medium as described above, strain Lb. reuteri BKQ629 was grown as fed-batch with continuously addition of sugar cane juice, while maintaining pH at 5.5, resulting in a significant increase in n-propanol equivalents (0.48 g/L of n-propanol and 4.08 g/L of 1 ,2-propanediol) after 2

20 days of fermentation. This would correspond to a total titer of n-propanol of 3.7 g/L, once all the 1 ,2-propanediol is converted to n-propanol. The result of the fed-batch fermentation suggests that Lb. reuteri prefers to use the available NAD(P)H for conversion of methylglyoxal to hydroxyacetone or lactaldehyde and subsequently to 1 ,2-propanediol, while further conversion to n-propanol either requires additional NAD(P)H or increased

25 activities of the downstream enzymatic steps.

Example 8: Production of n-propanol from a Lactobacillus reuteri strain containing a chromosomally integrated methylglyoxal synthase gene sequence.

Construction of pSJ1 1298 (Temperature-sensitive vector)

30 A temperature-sensitive vector useful for integration into and subsequent excision from the Lactobacillus chromosome was based on pG+Host4 (Aggligene, France; see Biswas et al. J. Bacteriol. 1993, 175, 3628-3635). A plasmid replication origin functioning in E. coli was obtained by PCR amplification, using plasmid pUC19 (Yanisch-Perron et al. Gene 1985, 33, 103-1 19) as template, and primers 689229 and 689230 shown below. Primer 689229:

5'-GACTA AGCTT GGGCC CTCCT CGCTC ACTGA CTCGC T-3' (SEQ ID NO: 139) Primer 689230:

5'-GACTG AATTC GGGCC CTCAT GACCA AAATC CCTTA ACG-3' (SEQ ID NO: 140)

The approximately 0.8 kb DNA fragment obtained by the PCR amplification was digested with EcoRI + Hindi 11 , and purified by agarose gel electrophoresis.

Plasmid pG+Host4 was digested with EcoRI + Hind\\\, and the digested vector treated with alkaline phosphatase. Fragment and vector DNA was mixed, ligated, and the ligation mixture transformed into E. coli SJ2 competent cells, selecting erythromycin resistance (200 microgram/ml). A transformant, containing a plasmid with the pUC19 origin of replication inserted into the pG+Host4 backbone, was kept as SJ1 1298 (SJ2/pSJ1 1298).

Plasmid pSJ1 1298 was introduced into the SJ1 1294 mutant Lactobacillus strain {supra) according to procedure B, described above, and a transformant obtained (SJ1 1487) was propagated at either 30°C or 37°C in MRS medium supplemented with erythromycin (10 microgram/ml). After overnight incubation, 400 μΙ_ of culture was inoculated into 1 .5 mL MRS with 10 microgram/ml erythromycin. After growth for 3 hours at either 30°C or 37°C, the cells were washed in 0.8 mL STE-buffer (containing, per liter, 26.8 ml 25% sucrose, 50 ml 1 M Tris (pH 7.5), and 2 ml 0.5 M EDTA), and plasmid DNA extracted using a Qiagen spin kit (Qiagen, Hilden, Germany).

Substantially more plasmid DNA was obtained from the 30°C culture, as compared to the 37°C culture, confirming the temperature-sensitive nature of the replication of this plasmid in L. reuteri.

Construction of LAR 1344 disruption vector pSJ 1 1679 (and pSJ 1 1680).

As a tool to disrupt/delete the L. reuteri gene encoding the type I restriction modification system specificity subunit LAR_1344 (SEQ ID NO: 141 ) via homologous recombination, a pSJ1 1298-derived vector was constructed which contained a chromosomal fragment extending from upstream and just into the LAR_1344 coding sequence (5'_LAR_1344; SEQ ID NO: 142), followed by a chromosomal fragment extending from the end of the LAR_1344 coding sequence and downstream (3'_LAR_1344; SEQ ID NO: 143).

The two fragments were obtained by PCR amplification using chromosomal DNA from SJ10655 as template, and primers 697369 + 697370 for the 5' fragment, and primers 697371 + 697372 for the 3' fragment.

Primer 697369:

5'-GACTG AATTC CTTCA AGATA AGAAG AAA-3' (SEQ ID NO: 144)

5 Primer 697370:

5'-GACTG GATCC GTAGT TAAAT ATTCA TCTTT GG-3' (SEQ ID NO: 145)

Primer 697371 :

5'-GACTG GATCC AGCTT AATGC AAGAA TATTT TGGG-3' (SEQ ID NO: 146)

Primer 697372:

0 5'-GACTC GGCCG TCTGA ACTTA TGTGG ATGAA-3' (SEQ ID NO: 147)

The approximately 0.99 kb 5' fragment obtained using primers 697369 + 697370 was digested with EcoRI + BamHI, and purified by agarose gel electrophoresis. The approximately 1 .0 kb 3' fragment obtained using primers 697371 + 697372 was digested with BamHI + Eagl, and purified by agarose gel electrophoresis.

5 The cloning vector pSJ1 1298 was digested with EcoRI + Eag\, treated with alkaline phosphatase, mixed with the purified 5'- and 3'-fragments of LAR_1344, ligated, and the ligation mixture transformed into E. coli SJ2 chemically competent cells, selecting erythromycin resistance (200 microgram/ml). Two transformants, harboring a plasmid deemed correct by restriction analysis and DNA sequencing, were kept as SJ1 1679 o (S J2/pS J 1 1679) and S J 1 1680 (S J2/pS J 1 1680).

Construction of an improved LAR 1344 disruption vector pSJ 1 1698 (and pSJ 1 1699).

To make a more versatile disruption/integration vector, a multiple cloning site was inserted into the BamHI site separating the LAR_1344 5' and 3' fragments.

The multiple cloning site was excised as a BamHI-Bcll fragment of 75 bp from5 plasmid pDN3000 (Diderichsen et al., J. Bacteriol. 1990, 172, 4315-4321 ), prepared from a dam- E. coli host strain, and purified by agarose gel electrophoresis. This fragment was ligated to BamHI-digested, alkaline phosphatase treated pSJ1 1679 DNA, purified by agarose gel electrophoresis, and the ligation mixture transformed into E. coli SJ2 chemically competent cells. Two transformants, deemed to contain correct plasmids by o restriction analysis, were kept as SJ 1 1698 (S J2/pS J 1 1698) and SJ 1 1699 (S J2/pS J 1 1699).

Insertion of antibiotic resistance markers into LAR 1344 disruption vectors.

To further improve the disruption vectors, an antibiotic resistance marker was inserted next to the multiple cloning site between the 5' and 3' chromosomal fragments.

The antibiotic resistance marker was further flanked by recognition sites (res) for the site-5 specific recombination enzyme (resolvase) from plasmid ρΑΜβΙ , thus allowing the eventual deletion of the marker by site-specific recombination mediated by the resolvase (see WO 96/23073).

To insert a chloramphenicol resistance gene flanked by resolvase sites, the appropriate 1 .2 kb fragment was prepared from pSJ3372 by digestion with Bcll-BamHI and purified by agarose gel electrophoresis (prepared from a dam- E. coli host; see WO 96/23073, Figure 9 and examples). The fragment was mixed and ligated with a BamHI- digested, alkaline phosphatase treated and agarose gel purified pSJ1 1698 DNA. The ligation mixture was transformed into E. coli SJ2 by electroporation as described above, selecting chloramphenicol resistance (10 microgram/ml), with colonies checked for both chloramphenicol and erythromycin resistance by replica plating. Two resulting transformants in which plasmids were deemed correct by restriction analysis and DNA sequencing, were kept as SJ 1 1728 (SJ2/pSJ1 1728) and SJ 1 1729 (SJ2/pSJ1 1729).

Construction of Lactobacillus reuteri host strain SJ 1 1774

Plasmid pSJ1 1729 (supra) was introduced into strain SJ 1 1400 (supra) by electroporation using protocol B described above, selecting erythromycin resistance (10 microgram/ml) on MRS agar plates incubated anaerobically at 37°C. Two of the 1 1 transformants obtained were propagated in MRS medium with 10 microgram/ml erythromycin at 30°C for 4 days, whereafter a 100 microliter aliquot was transferred to 1 .8 ml MRS medium with 6 microgram/ml chloramphenicol, incubated at 45°C overnight, and subsequently plated for single colonies on MRS with 6 microgram/ml chloramphenicol.

Two such colonies were inoculated into MRS medium and incubated overnight at 30°C, whereafter an aliquot was used to inoculate new MRS medium cultures that were again incubated overnight at 30°C. These cultures were subsequently plated for single colonies on MRS agar plates with 6 microgram/ml chloramphenicol, plates incubated overnight at 45°C, replica plated to MRS agar plates with either 6 microgram/ml chloramphenicol or 10 microgram/ml erythromycin, replica plates incubated overnight at 37°C, and erythromycin sensitive, chloramphenicol resistant strains isolated. PCR amplification confirmed absence of the ermR gene, and that the res-cat-res segment of plasmid pSJ1 1729 had been inserted into the L. reuteri SJ1 1400 chromosome, replacing the LAR_1344 gene originally present at that chromosomal location. One such strain was kept at SJ 1 1774.

Construction of integration vectors carrying methylglyoxal synthase genes

Plasmid pBKQ476 (containing a synthetic version of the E. coli methylglyoxal synthase coding sequence; see above) was digested with Nhe\ and Xba\, and the 0.78 kb Nhe\-Xba\ fragment purified by agarose gel electrophoresis. Plasmid pSJ1 1698 (supra) was digested with Nhel, treated with alkaline phosphatase, and the 6.59 kb fragment purified by agarose gel electrophoresis. The two fragments were mixed, ligated, and transformed into £ coli TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml). Three of the resulting transformants were deemed correct by restriction analysis, and kept as SJ 1 1815 (TG1/pSJ1 1815), SJ1 1816 (TG1/pSJ1 1816), and SJ1 1817 (TG1/pSJ1 1817).

Plasmid pBKQ557 (containing a synthetic version of the B. subtilis methylglyoxal synthase coding sequence; see above) was digested with Nhel and Xbal, and the 0.74 kb Nhel-Xbal fragment purified by agarose gel electrophoresis. Plasmid pSJ1 1698 (supra) was digested with Nhel, treated with alkaline phosphatase, and the 6.59 kb fragment purified by agarose gel electrophoresis. The two fragments were mixed, ligated, and transformed into £. coli TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml). Two of the resulting transformants were deemed correct by restriction analysis, and kept as SJ1 1818 (TG1/pSJ 1 1818) and SJ1 1819 (TG1/pSJ1 1819).

Chromosomal integration of methylglyoxal synthase genes and n-propanol production

The five plasmids pSJ1 1815 to pSJ1 1819 above were transformed into L. reuteri

SJ1 1774, using transformation protocol B described above, plated on MRS agar plates with erythromycin (10 microgram/ml), and incubated for 3 days at 30°C.

Four transformants with each plasmid were inoculated into liquid MRS with erythromycin (10 microgram/ml), and incubated overnight at 30°C, whereafter supernatants were harvested and analyzed for isopropanol, acetone, n-propanol, and ethanol. Results are shown in Table 13.

Table 13.

pSJ1 1818 - 4 0.03 0.02 0.09 3.19 pSJ1 1819 - 1 0.03 0.03 0.09 3.10

pSJ1 1819 - 2 0.03 0.02 0.09 3.21

pSJ1 1819 - 3 0.03 0.02 0.09 3.19

pSJ1 1819 - 4 0.03 0.02 0.09 3.22

After overnight incubation, all transformants produced n-propanol at a level above that seen for wildtype L. reuteri (0.03 g/L or below in similar experiments). Strains containing a polynucleotide encoding a B. subtilis methylglyoxal synthase (pSJ 1 1818 and pSJ1 1819) produced more n-propanol than the strains containing a polynucleotide encoding an £. co// methylglyoxal synthase (pSJ1 1815 to pSJ1 1817).

In order to have the chromosomally located cat gene of the host strain exchanged with the methylglyoxal synthase gene of the transformed plasmid, transformants were plated on MRS agar plates with 10 microgram/ml erythromycin, and incubated at 45°C for 2 days. Single colonies were subsequently inoculated in liquid MRS medium, propagated for 4 days at 30°C, and aliquots used to inoculate new MRS cultures which were propagated at 30°C overnight, and subsequently plated for single colonies on MRS agar plates. These were replica plated onto MRS, MRS with chloramphenicol (6 microgram/ml), and MRS with erythromycin (10 microgram/ml).

Colonies sensitive to both erythromycin and chloramphenicol were reisolated on MRS agar plates, and colonies appearing were tested again on plates with erythromycin and chloramphenicol.

A number of such sensitive colonies, as well as SJ1 1400, BKQ488 (containing pBKQ476 encoding an E. coli methylglyoxal synthase), and BKQ577 (containing pBKQ557 encoding a B. subtilis methylglyoxal synthase), were inoculated into MRS medium (for the BKQ strains erythromycin (10 microgram/ml) was added), and incubated at 37°C for 3 days, whereafter supernatants were harvested and analyzed for isopropanol, acetone, n- propanol, and ethanol.

Strain SJ1 1400 (negative control) produced n-propanol at 0.004%, whereas strains

BKQ488 and BKQ557 produced n-propanol at 0.013% and 0.019%, respectively. All strains (a total of 25) derived from integration and excision using either pSJ1 1815, pSJ1 1816, or pSJ1 1817 (containing a polynucleotide encoding an E. coli methylglyoxal synthase) produced n-propanol at the control strain level (0.004%). All strains (a total of 10) derived from integration and excision of pSJ1 1818 (containing a polynucleotide encoding a B. subtilis methylglyoxal synthase) produced n-propanol at between 0.006% and 0.01 1 %.

Selected strains were further analyzed by PCR amplification and DNA sequence analysis of the amplified DNA. Three further reisolated colonies, all derived from the same original integration of the B. subtilis mgsA gene using pSJ1 1818 (confirmed by PCR and sequencing), were kept as SJ 1 1875, SJ 1 1876, and SJ 1 1877. Two further reisolated colonies, both derived from the same original integration of the £ coli mgsA gene using 5 pSJ1 1816 (confirmed by PCR and sequencing), were kept as SJ1 1878 and SJ 1 1879. Two other reisolated colonies, both derived from another integration of the £ coli mgsA gene using pSJ1 1816 (confirmed by PCR and sequencing), were kept as SJ1 1880 and SJ1 1881 . Finally, two reisolated colonies both derived from an integration of the £. coli mgsA gene using pSJ1 1817 (confirmed by PCR and sequencing), were kept as SJ1 1882 i o and SJ1 1883. These nine saved strains were compared in a new growth experiment, inoculated in duplicate into MRS medium and incubated at 37°C for 3 days, and supernatants analyzed as described above. All cultures containing £ coli mgsA constructs produced n-propanol at 0.004%, whereas all cultures containing B. subtilis mgsA constructs produced n-propanol at 0.005-0.006%.

15

Example 9: Peptide-inducible pSIP expression vectors.

The peptide-inducible expression vectors pSIP409, pSIP410, and pSIP41 1 (S0rvig, et al. Microbiology 2005, 151 , 2439-2449.) were received from Lars Axelsson, Nofima Mat AS, Norway. pSIP409 and pSIP410 were transformed into £ coli SJ2 by electroporation,

20 selecting erythromycin resistance (150 microgram/ml) on LB agar plates at 37°C. Two transformants containing pSIP409 were kept as SJ10517 and SJ10518, and two transformants containing pSIP410 were kept as SJ 10519 and SJ10520.

pSIP41 1 was transformed into naturally competent Bacillus subtilis DN 1885 cells, essentially as described (Yasbin et al. J Bacteriol 1975, 121 , 296-304), selecting for

25 erythromycin resistance (10 microgram/ml) on LBPGS plates at 37°C. Two such transformants were kept as SJ 10513 and SJ 10514.

pSIP41 1 was in addition transformed into £ coli MG1655 by electroporation, selecting erythromycin resistance (200 microgram/ml) on LB agar plates at 37°C, and two transformants kept as SJ 10542 and SJ 10543.

30 For use in induction of gene expression from these vectors in Lactobacillus, the inducing peptide, here named M-19-R and having the following amino acid sequence: "Met- Ala-Gly-Asn-Ser-Ser-Asn-Phe-lle-His-Lys-lle-Lys-Gln-lle-Phe-Thr-His-Arg", was obtained from "Polypeptide Laboratories France, 7 rue de Boulogne, 67100 Strasbourg, France". Example 10: Cloning of isopropanol pathway genes.

Cloning of a Clostridium acetobutylicum thiolase gene and construction of vector pSJ10705 The 1 176 bp coding sequence (without stop codon) of a thiolase gene identified in Clostridium acetobutylicum was designed for optimized expression in the three organisms 5 Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ10705. The DNA fragment containing the codon optimized coding sequence was designed with the sequence 5'-AAGCTTTC-3' immediately prior to the start codon (to add a Hindi 11 site and convert the start region to a Ncol-compatible BspHI site), and the sequence 5'-TAGTC TAGAC TCGAG GAATT CGGTA CC-3' (SEQ ID NO: 53) i o immediately downstream (to add a stop codon, and restriction sites Xbal-Xhol-EcoRI-Kpnl).

The resulting sequence was then submitted to and synthesized by Geneart AG (Regenburg, Germany) and delivered in the pMA backbone vector containing the β- lactamase encoding gene blaTEM-1 . The DNA preparation delivered from Geneart was transformed into E. coli SJ2 by electroporation, selecting ampicillin resistance (200

15 microgram/ml) and two transformants kept, as SJ 10705 (SJ2/pSJ 10705) and SJ 10706 (SJ2/pSJ 10706).

The wild-type nucleotide sequence (WT), codon-optimized nucleotide sequence (CO), and deduced amino acid sequence of the C. acetobutylicum thiolase gene are SEQ ID NOs: 1 , 2, and 3, respectively. The coding sequence is 1 179 bp including the stop codon 20 and the encoded predicted protein is 392 amino acids. Using the SignalP program (Nielsen et al., 1997, Protein Engineering 10: 1 -6), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 392 amino acids with a predicted molecular mass of 41 .4 kDa and an isoelectric pH of 7.08.

Cloning of a Lactobacillus reuteri thiolase gene and construction of vector pSJ 10694 25 The 1 176 bp thiolase coding sequence (without stop codon) from Lactobacillus reuteri was amplified from chromosomal DNA of SJ10468 (supra) using primers 671826 and 671827 shown below.

Primer 671826:

5'-AGTCA AGCTT CCATG GAGAA GGTTT ACATT GTTGC-3' (SEQ ID NO: 51 ) 30 Primer 671827:

5'-ATGCG GTACC GAATT CCTCG AGTCT AGACT AAATT TTCTT AAGCA GAACC G-3' (SEQ ID NO: 52)

The PCR reaction was programmed for 94°C for 2 minutes; and then 19 cycles each at 95°C for 30 seconds, 59°C for 1 minute, and 72°C for 2 minute; then one cycle at 72°C for 5 minutes. A PCR amplified fragment of approximately 1 .2 kb was digested with Ncol + EcoRI, purified by agarose gel electrophoresis, and then ligated to the agarose gel electrophoresis purified EcoRI-Ncol vector fragment of plasmid pSIP409. The ligation mixture was transformed into E. coli SJ2, selecting ampicillin resistance (200 microgram/ml), and a transformant, deemed correct by restriction digest and DNA sequencing, was kept as SJ10694 (SJ2/pSJ 10694).

The codon-optimized nucleotide sequence (CO), and deduced amino acid sequence of the L. reuteri thiolase gene are SEQ ID NOs: 25 and 26, respectively. The coding sequence is 1 179 bp including the stop codon and the encoded predicted protein is 392 amino acids. Using the SignalP program (Nielsen et al., 1997, Protein Engineering 10: 1 -6), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 392 amino acids with a predicted molecular mass of 41 .0 kDa and an isoelectric pH of 5.4.

Cloning of a Propionibacterium freudenreichii thiolase gene and construction of vector PSJ10676

The 1 152 bp coding sequence (without stop codon) of a thiolase gene identified in Propionibacterium freudenreichii was optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ10676. The DNA fragment containing the codon optimized CDS was designed with the sequence 5'-AAGCTTTC-3' immediately prior to the start codon (to add a Hind 111 site and convert the start region to a Ncol-compatible BspHI site), and the sequence 5'-TAGTC TAGAC TCGAG GAATT CGGTA CC-3' (SEQ ID NO: 53) immediately downstream (to add a stop codon, and restriction sites Xbal-Xhol-EcoRI-Kpnl).

The resulting sequence was then submitted to and synthesized by Geneart AG (Regenburg, Germany) and delivered in the pMA backbone vector containing the β- lactamase encoding gene blaTEM-1 . The DNA preparation delivered from Geneart was transformed into E. coli SJ2 by electroporation, selecting ampicillin resistance (200 microgram/ml) and two transformants kept, as SJ 10676 (SJ2/pSJ 10676) and SJ 10677 (SJ2/pSJ 10677).

The codon-optimized nucleotide sequence (CO), and deduced amino acid sequence of the P. freudenreichii thiolase gene are SEQ ID NOs: 39 and 40, respectively. The coding sequence is 1 155 bp including the stop codon and the encoded predicted protein is 384 amino acids. Using the SignalP program (Nielsen et al., 1997, Protein Engineering 10: 1 -6), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 384 amino acids with a predicted molecular mass of 39.8 kDa and an isoelectric pH of 6.1 .

Cloning of a Lactobacillus brevis thiolase gene and construction of vector pSJ 10699

The 1 167 bp coding sequence (without stop codon) of a thiolase gene identified in Lactobacillus brevis was optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ 10699. The DNA fragment containing the codon optimized CDS was designed with the sequence 5'- AAGCTTCC-3' immediately prior to the start codon (to add a Hindi 11 site and convert the start region to a Ncol site), and the sequence 5'-TAGTC TAGAC TCGAG GAATT CGGTA CC-3' (SEQ ID NO: 53) immediately downstream (to add a stop codon, and restriction sites Xbal-Xhol-EcoRI-Kpnl).

The resulting sequence was then submitted to and synthesized by Geneart AG (Regenburg, Germany) and delivered in the pMA backbone vector containing the β- lactamase encoding gene blaTEM-1 . The DNA preparation delivered from Geneart was transformed into E. coli SJ2 by electroporation, selecting ampicillin resistance (200 microgram/ml) and two transformants kept, as SJ 10699 (SJ2/pSJ 10699) and SJ 10700 (SJ2/pSJ 10700).

The codon-optimized nucleotide sequence (CO), and deduced amino acid sequence of the L. brevis thiolase gene are SEQ ID NOs: 41 and 42, respectively. The coding sequence is 1 170 bp including the stop codon and the encoded predicted protein is 389 amino acids. Using the SignalP program (Nielsen et al., 1997, Protein Engineering 10: 1 -6), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 389 amino acids with a predicted molecular mass of 40.4 kDa and an isoelectric pH of 6.5.

Cloning of B. subtilis succinyl-CoA:acetoacetate transferase genes and construction of vectors pSJ 10695 and pSJ 10697

The 699 bp coding sequence (without stop codon) of the scoA subunit of the B. subtilis succinyl-CoA:acetoacetate transferase and the 648 bp coding sequence of the scoB subunit of the B. subtilis succinyl-CoA:acetoacetate transferase were designed for optimized expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ 10695 and pSJ 10697, respectively.

The DNA fragment containing the codon-optimized scoA coding sequence was designed with the sequence 5'-AAGCT TCTCG AGACT ATTAC AAGGA GATTT TAGCC-3' (SEQ ID NO: 54) immediately prior to the start codon (to add a Hindll l site, a Lactobacillus RBS, and to have the start codon within a Ncol site), and an EcoRI restriction site immediately downstream. The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ 10695 (SJ2/pSJ 10695) and SJ 10696 (SJ2/pSJ 10696).

The wild-type nucleotide sequence (WT), codon-optimized nucleotide sequence

(CO), and deduced amino acid sequence of the B. subtilis scoA subunit of the succinyl- CoA:acetoacetate transferase are SEQ ID NOs: 4, 5, and 6, respectively. The coding sequence is 702 bp including the stop codon and the encoded predicted protein is 233 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 233 amino acids with a predicted molecular mass of 25.1 kDa and an isoelectric pH of 6.50.

The DNA fragment containing the codon optimized scoB coding sequence was designed with the sequence 5'-GAATT CACTA TTACA AGGAG ATTTT AGTC-3' (SEQ ID NO: 55) immediately prior to the start codon (to add a EcoRI site, a Lactobacillus RBS, and to have the start codon within a Ncol-compatible BspHI site), and Eagl and Kpnl restriction sites immediately downstream. The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ 10697 (SJ2/pSJ 10697) and SJ 10698 (SJ2/pSJ 10698).

The wild-type nucleotide sequence (WT), codon-optimized nucleotide sequence (CO), and deduced amino acid sequence of the B. subtilis scoB subunit of the succinyl- CoA:acetoacetate transferase are SEQ ID NOs: 7, 8, and 9, respectively. The coding sequence is 651 bp including the stop codon and the encoded predicted protein is 216 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 216 amino acids with a predicted molecular mass of 23.4 kDa and an isoelectric pH of 5.07.

Cloning of B. mojavensis succinyl-CoA:acetoacetate transferase genes and construction of vectors pSJ10721 and pSJ10723

The 71 1 bp coding sequence (without stop codon) of the scoA subunit of the B. mojavensis succinyl-CoA:acetoacetate transferase and the 654 bp coding sequence (without stop codon) of the scoB subunit of the B. mojavensis succinyl-CoA:acetoacetate transferase were designed for optimized expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ10721 and pSJ 10723, respectively.

The DNA fragment containing the codon-optimized scoA coding sequence was designed with the sequence 5'-AAGCT TCTCG AGACT ATTAC AAGGA GATTT TAGCC-3' (SEQ ID NO: 54) immediately prior to the start codon (to add a Hindll l site, a Lactobacillus RBS, and to have the start codon within a Ncol site), and an EcoRI restriction site immediately downstream. The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ10721 (SJ2/pSJ 10721 ) and SJ10722 (SJ2/pSJ 10722).

The wild-type nucleotide sequence (WT), codon-optimized nucleotide sequence (CO), and deduced amino acid sequence of the B. mojavensis scoA subunit of the succinyl- CoA:acetoacetate transferase are SEQ ID NOs: 10, 1 1 , and 12, respectively. The coding sequence is 714 bp including the stop codon and the encoded predicted protein is 237 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 237 amino acids with a predicted molecular mass of 25.5 kDa and an isoelectric pH of 5.82.

The DNA fragment containing the codon optimized scoB nucleotide coding sequence was designed with the sequence 5'-GAATT CACTA TTACA AGGAG ATTTT AGTC-3' (SEQ ID NO: 55) immediately prior to the start codon (to add a EcoRI site, a Lactobacillus RBS, and to have the start codon within a Ncol-compatible BspHI site), and Eagl and Kpnl restriction sites immediately downstream. The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ 10723 (SJ2/pSJ 10723) and SJ 10724 (SJ2/pSJ 10724).

The wild-type nucleotide sequence (WT), codon-optimized nucleotide sequence

(CO), and deduced amino acid sequence of the B. mojavensis scoB subunit of the succinyl- CoA:acetoacetate transferase are SEQ ID NOs: 13, 14, and 15, respectively. The coding sequence is 657 bp including the stop codon and the encoded predicted protein is 218 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 218 amino acids with a predicted molecular mass of 23.7 kDa and an isoelectric pH of 5.40.

Cloning of E. coli acetoacetyl-CoA transferase genes and construction of vectors pSJ 10715 and PSJ10717

The 648 bp coding sequence (without stop codon) of the atoA subunit (uniprot:P76459) of the E. coli acetyl-CoA transferase and the 660 bp coding sequence (without stop codon) of the atoD subunit (uniprot:P76458) of the E. coli acetyl-CoA transferase were optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ10715 and pSJ 10717, respectively.

The DNA fragment containing the codon-optimized atoA subunit nucleotide coding sequence was designed with the sequence 5'-AAGCT TCTCG AGACT ATTAC AAGGA GATTT TAGCC-3' (SEQ ID NO: 54) immediately prior to the start codon (to add Hindi 11 and Xhol sites, a Lactobacillus RBS, and to have the start codon within a Ncol site), and an EcoRI restriction site immediately downstream. The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ10715 (SJ2/pSJ10715) and SJ10716 (SJ2/pSJ10716).

The codon-optimized nucleotide sequence (CO) and deduced amino acid sequence of the E. coli atoA subunit of the acetoacetyl-CoA transferase are SEQ ID NOs: 27 and 28, respectively. The coding sequence is 651 bp including the stop codon and the encoded predicted protein is 216 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 216 amino acids with a predicted molecular mass of 23.0 kDa and an isoelectric pH of 5.9.

The DNA fragment containing the codon optimized atoD nucleotide coding sequence was designed with the sequence 5'-GAATT CACTA TTACA AGGAG ATTTT AGTC-3' (SEQ ID NO: 55) immediately prior to the start codon (to add a EcoRI site, a Lactobacillus RBS, and to have the start codon within a Ncol-compatible BspHI site), and Eagl and Kpnl restriction sites immediately downstream. The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ10717 (SJ2/pSJ 10717) and SJ10718 (SJ2/pSJ10718).

The codon-optimized nucleotide sequence (CO) and deduced amino acid sequence of the E. coli atoD subunit of the acetoacetyl-CoA transferase are SEQ ID NOs: 29 and 30, respectively. The coding sequence is 663 bp including the stop codon and the encoded predicted protein is 220 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 220 amino acids with a predicted molecular mass of 23.5 kDa and an isoelectric pH of 4.9.

Cloning of Clostridium acetobutylicum acetoacetyl-CoA transferase genes and construction of vectors pSJ10727 and pSJ10731

The 654 bp coding sequence (without stop codon) of the ctfA subunit

(uniprot:P33752) of the C. acetobutylicum acetyl-CoA transferase and the 663 bp coding sequence (without stop codon) of the ctfB subunit (uniprot:P23673) of the C. acetobutylicum acetyl-CoA transferase were optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ10727 and pSJ10731 , respectively. The DNA fragment containing the codon optimized ctfA subunit coding sequence was designed with the sequence 5'-AAGCT TCTCG AGACT ATTAC AAGGA GATTT TAGTC-3' (SEQ ID NO: 56) immediately prior to the start codon (to add Hindi 11 and Xhol sites, a Lactobacillus RBS, and to have the start codon within a Ncol-compatible BspHI site), and an EcoRI restriction site immediately downstream. The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ 10727 (SJ2/pSJ 10727) and SJ 10728 (SJ2/pSJ 10728).

The codon-optimized nucleotide sequence (CO) and deduced amino acid sequence of the C. acetobutylicum ctfA subunit of the acetoacetyl-CoA transferase are SEQ ID NOs: 31 and 32, respectively. The coding sequence is 657 bp including the stop codon and the encoded predicted protein is 218 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 218 amino acids with a predicted molecular mass of 23.6 kDa and an isoelectric pH of 9.3.

The DNA fragment containing the codon optimized ctfB subunit coding sequence was designed with the sequence 5'-GAATT CACTA TTACA AGGAG ATTTT AGTC-3' (SEQ ID NO: 55) immediately prior to the start codon (to add a EcoRI site, a Lactobacillus RBS, and to have the start codon within a Ncol-compatible BspHI site), and Eagl and Kpnl restriction sites immediately downstream. The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ10731 (SJ2/pSJ 10731 ) and SJ 10732 (SJ2/pSJ 10732).

The codon-optimized nucleotide sequence (CO) and deduced amino acid sequence of the C. acetobutylicum ctfB subunit of the acetoacetyl-CoA transferase are SEQ ID NOs: 33 and 34, respectively. The coding sequence is 666 bp including the stop codon and the encoded predicted protein is 221 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 221 amino acids with a predicted molecular mass of 23.6 kDa and an isoelectric pH of 8.5.

Cloning of Eremococcus coleocola butyrate-acetoacetate CoA transferase genes and construction of vectors pSJ 1 1474 and pSJ 1 1476

The 651 bp coding sequence (without stop codon) of the subunit A (UniProt:E4KQS7) of the Eremococcus coleocola putative butyrate-acetoacetate CoA transferase and the 639 bp coding sequence (without stop codon) of the subunit (UniProt:E4KQS6) of the Eremococcus coleocola putative butyrate-acetoacetate CoA transferase were optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri.

Two different codon-optimized nucleotide sequences (CO) designated D1396V and D1396W (SEQ ID NOs: 61 and 62, respectively) encode the subunit A (SEQ ID NO: 63). The coding sequence is 654 bp including the stop codon and the encoded predicted protein is 217 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 217 amino acids with a predicted molecular mass of 24 kDa and an isoelectric pH of 4.8.

Two different codon-optimized nucleotide sequences (CO) designated D1396X and D1396Y (SEQ ID NOs: 64 and 65, respectively) encode the subunit B (SEQ ID NO: 66). The coding sequence is 642 bp including the stop codon and the encoded predicted protein is 213 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 213 amino acids with a predicted molecular mass of 23 kDa and an isoelectric pH of 4.2.

Two different combinations of subunit A and subunit B encoding genes were designed. In the first combination, the DNA sequence 5'-GGATC CCTCG AGACT ATTAC AAGGA GATTT TAGTC-3' (SEQ ID NO: 148) was inserted in front of the first subunit gene (D1396V), followed by the sequence 5'-TAGAA TTCAC TATTA CAAGG AGATT TTAGC C- 3' (SEQ ID NO: 149), the second subunit gene (D1396X), and the sequence 5'-TAGCG GCCGG GTACC-3' (SEQ ID NO: 150).

In the second combination, the DNA sequence 5'-GGATC CCTCG AGACT ATTAC AAGGA GATTT TAGTC-3' (SEQ ID NO: 148) was inserted in front of the first subunit gene (D1396W), followed by the sequence 5'-TAGAA TTCAC TATTA CAAGG AGATT TTAGC C-3' (SEQ ID NO: 149), the second subunit gene (D1396Y), and the sequence 5'-TAGCG GCCGG GTACC-3' (SEQ ID NO: 150).

The designed constructs were obtained from Geneart AG and transformed as described above, resulting in SJ1 1474 (SJ2/pSJ1 1474) and SJ1 1475 (SJ2/pSJ1 1475), both harboring the second combination (D1396W + D1396Y) of genes above, and SJ 1 1476 (SJ2/pSJ 1 1476) and SJ 1 1477 (SJ2/pSJ1 1477), both harboring the first combination (D1396V + D1396X) of genes above.

Cloning of Alicyclobacillus sp-18711 putative succinyl-CoA:acetoacetate transferase genes and construction of vector pSJ1 1472

The 696 bp coding sequence (without stop codon) of the ScoA subunit of the Alicyclobacillus succinyl-CoA:acetoacetate transferase and the 657 bp coding sequence (without stop codon) of the ScoB subunit of the Alicyclobacillus succinyl-CoA:acetoacetate transferase were optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri.

A codon-optimized nucleotide sequence (CO) designated D1396R (SEQ ID NO: 67) encodes the ScoA subunit (SEQ ID NO: 68). The coding sequence is 699 bp including the stop codon and the encoded predicted protein is 232 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 232 amino acids with a predicted molecular mass of 25 kDa and an isoelectric pH of 5.2.

A codon-optimized nucleotide sequence (CO) designated D1396T (SEQ ID NO: 69) encodes the ScoB subunit (SEQ ID NO: 70). The coding sequence is 660 bp including the stop codon and the encoded predicted protein is 219 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 219 amino acids with a predicted molecular mass of 23 kDa and an isoelectric pH of 4.7.

To combine the ScoA and ScoB encoding genes, the DNA sequence 5'-GGATC CCTCG AGACT ATTAC AAGGA GATTT TAGTC-3' (SEQ ID NO: 148) was inserted in front of the first subunit gene (D1396R), followed by the sequence 5'-TAGAA TTCAC TATTA CAAGG AGATT TTAGC C-3' (SEQ ID NO: 149), the second subunit gene (D1396T), and the sequence 5'-TAGCG GCCGG GTACC-3' (SEQ ID NO: 150).

The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ 1 1472 (SJ2/pSJ1 1472) and SJ 1 1473 (SJ2/pSJ 1 1473).

Cloning of Helicobacter pylori succinyl-CoA:acetoacetate transferase genes and construction of vectors pSJ 1 1466 and pSJ 1 1468

The 696 bp coding sequence (without stop codon) of the ScoA subunit

(UniProt:032639) of the H. pylori succinyl-CoA:acetoacetate transferase and the 621 bp coding sequence (without stop codon) of the ScoB subunit (UniProt P56007) of the H. pylori succinyl-CoA:acetoacetate transferase were optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri.

Two different codon-optimized nucleotide sequences (CO) designated D1396J and

D1396K (SEQ ID NOs: 71 and 72, respectively) encode the ScoA subunit (SEQ ID NO: 73). The coding sequence is 699 bp including the stop codon and the encoded predicted protein is 232 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 232 amino acids with a predicted molecular mass of 25 kDa and an isoelectric pH of 5.9.

Two different codon-optimized nucleotide sequences (CO) designated D1396M and D1396N (SEQ ID NOs: 74 and 75, respectively) encode the ScoB subunit (SEQ ID NO: 76). The coding sequence is 624 bp including the stop codon and the encoded predicted protein is 207 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 207 amino acids with a predicted molecular mass of 22 kDa and an isoelectric pH of 5.3.

Two different combinations of ScoA and ScoB encoding genes were designed. In the first combination, the DNA sequence 5'-GGATC CCTCG AGACT ATTAC AAGGA GATTT TAGTC-3' (SEQ ID NO: 148) was inserted in front of the first subunit gene (D1396J), followed by the sequence 5'-TAGAA TTCAC TATTA CAAGG AGATT TTAGC C- 3' (SEQ ID NO: 149), the second subunit gene (D1396M), and the sequence 5'-TAGCG GCCGG GTACC-3' (SEQ ID NO: 150).

In the second combination, the DNA sequence 5'-GGATC CCTCG AGACT ATTAC

AAGGA GATTT TAGTC-3' (SEQ ID NO: 148) was inserted in front of the first subunit gene (D1396K), followed by the sequence 5'-TAGAA TTCAC TATTA CAAGG AGATT TTAGC C- 3' (SEQ ID NO: 149), the second subunit gene (D1396N), and the sequence 5'-TAGCG GCCGG GTACC-3' (SEQ ID NO: 150).

The designed constructs were obtained from Geneart AG and transformed as described above, resulting in SJ1 1466 (SJ2/pSJ1 1466) and SJ1 1467 (SJ2/pSJ1 1467), both harboring the first combination (D1396J + D1396M) of genes above, and SJ1 1468 (SJ2/pSJ 1 1468) and SJ1 1469 (SJ2/pSJ1 1469), both harboring the second combination (D1396K + D1396N) of genes above.

Cloning of a Clostridium acetobutylicum acetoacetate decarboxylase gene and construction of vector pSJ1071 1

The 777 bp coding sequence (without stop codon) of the acetoacetate decarboxylase (uniprot:P23670) from C. acetobutylicum was optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ1071 1 .

The DNA fragment containing the codon-optimized acetoacetate decarboxylase coding sequence (adc) was designed with the sequence 5'-AAGCT TCGGC CGACT ATTAC AAGGA GATTT TAGCC-3' (SEQ ID NO: 57) immediately prior to the start codon (to add Hindi 11 and Eagl sites and a Lactobacillus RBS), and a Kpnl restriction site immediately downstream. The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ1071 1 (SJ2/pSJ1071 1 ) and SJ10712 (SJ2/pSJ10712).

The codon-optimized nucleotide sequence (CO) and deduced amino acid sequence of the C. acetobutylicum acetoacetate decarboxylase gene are SEQ ID NOs: 35 and 36, 5 respectively. The coding sequence is 780 bp including the stop codon and the encoded predicted protein is 259 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 259 amino acids with a predicted molecular mass of 27.5 kDa and an isoelectric pH of 6.2.

i o Cloning of a Clostridium beijerinckii acetoacetate decarboxylase gene and construction of vector PSJ 10713

The 738 bp coding sequence (without stop codon) of the acetoacetate decarboxylase (uniprot:Q716S5) from C. beijerinckii was optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and

15 synthetically constructed into pSJ10713.

The DNA fragment containing the codon optimized acetoacetate decarboxylase coding sequence (adc Cb) was designed with the sequence 5'-AAGCT TCGGC CGACT ATTAC AAGGA GATTT TAGCC-3' (SEQ ID NO: 57) immediately prior to the start codon (to add Hindi 11 and Eagl sites and a Lactobacillus RBS), and a Kpnl restriction site

20 immediately downstream. The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ10713 (SJ2/pSJ10713) and SJ10714 (SJ2/pSJ10714).

The wild-type nucleotide sequence (WT), codon-optimized nucleotide sequence (CO), and deduced amino acid sequence of the C. beijerinckii acetoacetate decarboxylase 25 gene is SEQ ID NO: 16, 17, and 18, respectively. The coding sequence is 741 bp including the stop codon and the encoded predicted protein is 246 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 246 amino acids with a predicted molecular mass of 27.5 kDa and an isoelectric pH of 6.18.

30 Cloning of a Lactobacillus salivarius acetoacetate decarboxylase gene and construction of vector pSJ 10707

The 831 bp CDS (without stop codon) of the acetoacetate decarboxylase (SWISSPROT:Q1 WVG5) from L. salivarius was optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ 10707.

The DNA fragment containing the codon optimized acetoacetate decarboxylase CDS (adc Ls) was designed with the sequence 5'-AAGCT TCGGC CGACT ATTAC AAGGA GATTT TAGAC-3' (SEQ ID NO: 57) immediately prior to the start codon (to add Hindi II and Eagl sites and a Lactobacillus RBS), and a Kpnl restriction site immediately downstream. The constructs were obtained from Geneart AG and transformed as previously described, resulting in SJ10707 (SJ2/pSJ 10707) and SJ 10708 (SJ2/pSJ 10708).

The codon-optimized nucleotide sequence (CO) and deduced amino acid sequence of the L. salivarius acetoacetate decarboxylase gene is SEQ ID NO: 43 and 44, respectively. The coding sequence is 834 bp including the stop codon and the encoded predicted protein is 277 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 277 amino acids with a predicted molecular mass of 30.9 kDa and an isoelectric pH of 4.6.

Cloning of a Lactobacillus plantarum acetoacetate decarboxylase gene and construction of vector pSJ 10701

The 843 bp CDS (without stop codon) of the acetoacetate decarboxylase (SWISSPROT:Q890G0) from L. plantarum was optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ10701 .

The DNA fragment containing the codon optimized acetoacetate decarboxylase CDS (adc Lp) was designed with the sequence 5'-AAGCT TCGGC CGACT ATTAC AAGGA GATTT TAGCC-3' (SEQ ID NO: 57) immediately prior to the start codon (to add Hindi 11 and Eagl sites and a Lactobacillus RBS), and a Kpnl restriction site immediately downstream. The constructs were obtained from Geneart AG and transformed as previously described, resulting in SJ10701 (SJ2/pSJ 10701 ) and SJ 10702 (SJ2/pSJ 10702).

The codon-optimized nucleotide sequence (CO) and deduced amino acid sequence of the L. plantarum acetoacetate decarboxylase gene is SEQ ID NO: 45 and 46, respectively. The coding sequence is 846 bp including the stop codon and the encoded predicted protein is 281 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 281 amino acids with a predicted molecular mass of 30.8 kDa and an isoelectric pH of 4.7.

Cloning of a Thermoanaerobacter ethanolicus isopropanol dehydrogenase gene and construction of vector pSJ10719

The 1056 bp coding sequence (without stop codon) of the isopropanol dehydrogenase (uniprot:Q2MJT8) from T. ethanolicus was optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ10719.

The DNA fragment containing the codon optimized isopropanol dehydrogenase coding sequence (adh Te) was designed with the sequence 5'-GGTAC CACTA TTACA AGGAG ATTTT AGTC-3' (SEQ ID NO: 58) immediately prior to the start codon (to add a Kpnl site, a Lactobacillus RBS, and to have the start codon within a Ncol-compatible BspHI site), and Xmal and Hindi 11 restriction sites immediately downstream. The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ10719 (SJ2/pSJ10719) and SJ10720 (SJ2/pSJ 10720).

The wild-type nucleotide sequence (WT), codon-optimized nucleotide sequence (CO), and deduced amino acid sequence of the T. ethanolicus isopropanol dehydrogenase gene is SEQ ID NO: 22, 23, and 24, respectively. The coding sequence is 1059 bp including the stop codon and the encoded predicted protein is 352 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 352 amino acids with a predicted molecular mass of 37.7 kDa and an isoelectric pH of 6.23.

Cloning of a Clostridium beijerinckii isopropanol dehydrogenase gene and construction of vector pSJ 10725

The 1053 bp coding sequence (without stop codon) of the isopropanol dehydrogenase (uniprot:P25984) from C. beijerinckii was optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ 10725.

The DNA fragment containing the codon optimized isopropanol dehydrogenase coding sequence (adh Cb) was designed with the sequence 5'-GGTAC CACTA TTACA AGGAG ATTTT AGTC-3' (SEQ ID NO: 58) immediately prior to the start codon (to add a Kpnl site, a Lactobacillus RBS, and to have the start codon within a Ncol-compatible BspHI site), and Xmal and Hindi 11 restriction sites immediately downstream. The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ 10725 (SJ2/pSJ 10725) and SJ 10726 (SJ2/pSJ 10726).

The wild-type nucleotide sequence (WT), codon-optimized nucleotide sequence (CO), and deduced amino acid sequence of the C. beijerinckii isopropanol dehydrogenase gene is SEQ ID NO: 19, 20, and 21 , respectively. The coding sequence is 1056 bp including the stop codon and the encoded predicted protein is 351 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 351 amino acids with a predicted molecular mass of 37.8 kDa and an isoelectric pH of 6.64.

Cloning of a Lactobacillus antri isopropanol dehydrogenase gene and construction of vector pSJ 10709

The 1068 bp coding sequence (without stop codon) of the isopropanol dehydrogenase (SWISSPROT:C8P9V7) from L. antri was optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ 10709.

The DNA fragment containing the codon-optimized isopropanol dehydrogenase coding sequence (sadh La) was designed with the sequence 5'-GGTAC CACTA TTACA AGGAG ATTTT AGTC-3' (SEQ ID NO: 58) immediately prior to the start codon (to add a Kpnl site and a Lactobacillus RBS), and Xmal and Hindi 11 restriction sites immediately downstream. The designed construct was obtained from Geneart AG and transformed as described above, resulting in SJ10709 (SJ2/pSJ 10709) and SJ10710 (SJ2/pSJ 10710).

The codon-optimized nucleotide sequence (CO) and deduced amino acid sequence of the L. antri isopropanol dehydrogenase gene is SEQ ID NO: 37 and 38, respectively. The coding sequence is 1071 bp including the stop codon and the encoded predicted protein is 356 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 356 amino acids with a predicted molecular mass of 38.0 kDa and an isoelectric pH of 4.9.

Cloning of a Lactobacillus fermentum isopropanol dehydrogenase gene and construction of vector pSJ 10703

The 1068 bp CDS (without stop codon) of the isopropanol dehydrogenase (SWISSPROT:B2GDH6) from L. fermentum was optimized for expression in the three organisms Escherichia coli, Lactobacillus plantarum, and Lactobacillus reuteri and synthetically constructed into pSJ 10703.

The DNA fragment containing the codon optimized isopropanol dehydrogenase

CDS (sadh Lf) was designed with the sequence 5'-GGTAC CACTA TTACA AGGAG ATTTT AGTC-3' (SEQ ID NO: 58) immediately prior to the start codon (to add a Kpnl site and a Lactobacillus RBS), and Xmal and Hindi 11 restriction sites immediately downstream. The constructs were obtained from Geneart AG and transformed as previously described, resulting in SJ 10703 (SJ2/pSJ 10703) and SJ 10704 (SJ2/pSJ 10704).

The codon-optimized nucleotide sequence (CO) and deduced amino acid sequence of the L. fermentum isopropanol dehydrogenase gene is SEQ ID NO: 47 and 48, respectively. The coding sequence is 1071 bp including the stop codon and the encoded 5 predicted protein is 356 amino acids. Using the SignalP program (Nielsen et al., supra), no signal peptide in the sequence was predicted. Based on this program, the predicted mature protein contains 356 amino acids with a predicted molecular mass of 37.9 kDa and an isoelectric pH of 5.2. i o Example 11 : Isopropanol pathway constructs.

Construction of pSJ10843 containing a C. beijerinckii acetoacetate decarboxylase gene and a C. beijerinckii alcohol dehydrogenase gene

Plasmids pSJ10725 and pSJ10713 were digested individually with Kpnl+AlwNI. Plasmid pSJ 10725 was further digested with Pvul to reduce the size of unwanted

15 fragments. The resulting 1689 bp fragment of pSJ 10725 and the 2557 bp fragment of pSJ10713 were each purified using gel electrophoresis and subsequently ligated as outlined herein. An aliquot of the ligation mixture was used for transformation of E. coli SJ2 chemically competent cells, and transformants selected on LB plates with 200 microgram/ml ampicillin. Four colonies, picked among more than 100 transformants, were

20 all deemed to contain the desired recombinant plasmid by restriction analysis using Hindi 11 , and two of these were kept, resulting in SJ 10843 (SJ2/pSJ 10843) and SJ 10844 (SJ2/pSJ 10844).

Construction of pSJ10841 containing a C. acetobutylicum acetoacetate decarboxylase gene and a C. beijerinckii alcohol dehydrogenase gene

25 Plasmids pSJ10725 and pSJ1071 1 were digested individually with Kpnl+AlwNI; in addition, pSJ 10725 was digested with Pvul to reduce the size of unwanted fragments. The resulting 1689 bp fragment of pSJ 10725 and the 2596 bp fragment of pSJ 1071 1 were each purified using gel electrophoresis and subsequently ligated as outlined herein. An aliquot of the ligation mixture was used for transformation of E. coli SJ2 chemically competent cells,

30 and transformants selected on LB plates with 200 microgram/ml ampicillin. 4 colonies, picked among more than 100 transformants, were all deemed to contain the desired recombinant plasmid by restriction analysis using Bsgl, and two of these were kept, resulting in SJ10841 (SJ2/pSJ10841 ) and SJ10842 (SJ2/pSJ 10842).

Construction of pSJ 10748 containing a B. subtilis succinyl-CoA:acetoacetate transferase genes

Plasmids pSJ10697 and pSJ10695 were each digested with EcoRI and Kpnl. The resulting 690 bp fragment of pSJ 10697 and the 3106 bp fragment of pSJ 10695 were each purified using gel electrophoresis and subsequently ligated as outlined herein.

5 An aliquot of the ligation mixture was used for transformation of £ coli SJ2 by electroporation, and transformants selected on LB plates with 200 microgram/ml ampicillin. 3 colonies, picked among more than 50 transformants, were all deemed to contain the desired recombinant plasmid by restriction analysis using Pvul, and two of these were kept, resulting in SJ 10748 (SJ2/pSJ 10748) and SJ 10749 (SJ2/pSJ 10749).

i o Construction of pSJ 10777 containing a B. mojavensis succinyl-CoA:acetoacetate transferase genes

Plasmids pSJ10723 and pSJ10721 were each digested with EcoRI + Kpnl. The resulting 696 bp fragment of pSJ 10723 and the 31 18 bp fragment of pSJ 10721 were each purified using gel electrophoresis and subsequently ligated as outlined herein.

15 An aliquot of the ligation mixture was used for transformation of £ coli SJ2 chemically competent cells, and transformants selected on LB plates with 200 microgram/ml ampicillin. 4 colonies, picked among more than 500 transformants, were analyzed and one, deemed to contain the desired recombinant plasmid by restriction analysis using Pvul, was kept, resulting in SJ10777 (SJ2/pSJ 10777).

20 Construction of pSJ 10750 containing a E. coli acetoacetyl-CoA transferase genes

Plasmids pSJ10717 and pSJ10715 were each digested with EcoRI + Kpnl. The resulting 702 bp fragment of pSJ 10717 and the 3051 bp fragment of pSJ10715 were each purified using gel electrophoresis and subsequently ligated as outlined herein.

An aliquot of the ligation mixture was used for transformation of £ coli SJ2 by 25 electroporation, and transformants selected on LB plates with 200 microgram/ml ampicillin.

3 colonies, picked among more than 50 transformants, were all deemed to contain the desired recombinant plasmid by restriction analysis using ApaLI, and two of these were kept, resulting in SJ10750 (SJ2/pSJ 10750) and SJ10751 (SJ2/pSJ10751 ).

Construction of pSJ10752 containing a Clostridium acetobutylicum acetoacetyl-CoA 30 transferase genes

Plasmids pSJ10731 and pSJ10727 were each digested with EcoRI + Kpnl. The resulting 705 bp fragment of pSJ 10731 and the 3061 bp fragment of pSJ10727 were each purified using gel electrophoresis and subsequently ligated as outlined herein.

An aliquot of the ligation mixture was used for transformation of £. coli SJ2 by electroporation, and transformants selected on LB plates with 200 microgram/ml ampicillin. 3 colonies, picked among more than 50 transformants, were all deemed to contain the desired recombinant plasmid by restriction analysis using Pvul, and two of these were kept, resulting in SJ 10752 (SJ2/pSJ 10752) and SJ 10753 (SJ2/pSJ 10753).

Construction of expression vector pSJ10798 containing a Clostridium acetobutylicum thiolase gene

Plasmid pSJ10705 was digested with BspHI and EcoRI, whereas pSJ10600 was digested with Ncol and EcoRI. The resulting 1 193 bp fragment of pSJ10705 and the 5147 bp fragment of pSJ 10600 were each purified using gel electrophoresis and subsequently ligated as outlined herein.

An aliquot of the ligation mixture was used for transformation of E. coli TG1 by electroporation, and transformants selected on LB plates with 200 microgram/ml erythromycin. 3 of 4 colonies analyzed were deemed to contain the desired recombinant plasmid by restriction analysis using Nsil as well as DNA sequencing, and two of these were kept, resulting in SJ 10798 (TG1/pSJ10798) and SJ10799 (TG1/pSJ10799).

Construction of expression vector pSJ 10796 containing a L. reuteri thiolase gene

Plasmid pSJ10694 was digested with Ncol and EcoRI, and the resulting 1 .19 kb fragment purified using gel electrophoresis. Plasmid pSJ10600 was digested with Ncol and EcoRI, and the 5.2 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Nsil, and two of these, further verified by DNA sequencing, were kept, resulting in SJ10796 (TG1/pSJ 10796) and SJ 10797 (TG1/pSJ10797).

Construction of expression vector pSJ 10802 containing a Lactobacillus reuteri thiolase gene

Plasmid pSJ10694 was digested with Ncol and EcoRI, and the resulting 1 .19 kb fragment purified using gel electrophoresis. Plasmid pSJ10603 was digested with Ncol and EcoRI, and the 5.2 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Nsil, and two of these, further verified by DNA sequencing, were kept, resulting in SJ10802 (TG1/pSJ10802) and SJ10803 (TG1/pSJ10803). Construction of expression vector pSJ 10795 containing a Propionibacterium freudenreichii thiolase gene

Plasmid pSJ10676 was digested with BspHI and EcoRI, and the resulting 1 .17 kb fragment purified using gel electrophoresis. Plasmid pSJ 10600 was digested with Ncol and EcoRI, and the 5.2 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Nsil, and one of these, further verified by DNA sequencing, was kept, resulting in SJ10795 (TG1/pSJ10795).

Construction of expression vector pSJ 10743 containing a Lactobacillus brevis thiolase gene

Plasmid pSJ10699 was digested with Ncol and EcoRI, and the resulting 1 .18 kb fragment purified using gel electrophoresis. Plasmid pSJ10600 was digested with Ncol and EcoRI, and the 5.2 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into MG1655 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. 16 of the resulting colonies were analyzed and two, deemed to contain the desired recombinant plasmid by restriction analysis using Clal and further verified by DNA sequencing, were kept, resulting in SJ10743 (TG1/pSJ10743) and SJ10757 (TG1/pSJ10757).

Construction of expression vector pSJ10886 containing a Bacillus subtilis succinyl- CoA:acetoacetate transferase genes

Plasmid pSJ10748 was digested with Ncol and Kpnl, and the resulting 1 .4 kb fragment purified using gel electrophoresis. Plasmid pSJ10600 was digested with Ncol and Kpnl, and the resulting 5.1 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Hind III, and two of these, further verified by DNA sequencing, were kept, resulting in SJ10886 (TG1/pSJ10886) and SJ10887 (TG1/pSJ10887).

Construction of expression vector pSJ10888 containing E. coli acetoacetyl-CoA transferase genes

Plasmid pSJ10750 was digested with Ncol and Kpnl, and the resulting 1 .35 kb fragment purified using gel electrophoresis. Plasmid pSJ10600 was digested with Ncol and Kpnl, and the resulting 5.1 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Hind III, and two of these, further verified by DNA sequencing, were kept, resulting in SJ10888 (TG1/pSJ10888) and SJ10889 (TG1/pSJ10889).

Construction of expression vector pSJ10756 containing a C. beijerinckii acetoacetate decarboxylase gene

Plasmid pSJ10713 was digested with Eagl and Kpnl, and the resulting 0.77 kb fragment purified using gel electrophoresis. Plasmid pSJ10600 was digested with Eagl and Kpnl, and the resulting 5.1 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and one, deemed to contain the desired recombinant plasmid by restriction analysis using Clal and verified by DNA sequencing, was kept as SJ10756 (TG1/pSJ10756).

Construction of expression vector pSJ 10754 containing a C. acetobutylicum acetoacetate decarboxylase gene

Plasmid pSJ1071 1 was digested with Eagl and Kpnl, and the resulting 0.81 kb fragment purified using gel electrophoresis. Plasmid pSJ 10600 was digested with Eagl and Kpnl, and the resulting 5.1 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into MG1655 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed, three deemed to contain the desired recombinant plasmid by restriction analysis using Clal and two, verified by DNA sequencing, were kept as SJ10754 (MG1655/pSJ 10754) and SJ10755 (MG1655/pSJ 10755).

Construction of expression vector pSJ 10780 containing a L. salvarius acetoacetate decarboxylase gene

Plasmid pSJ10707 was digested with Pcil and Kpnl, and the resulting 0.84 kb fragment purified using gel electrophoresis. Plasmid pSJ10600 was digested with Ncol and Kpnl, and the resulting 5.1 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into MG1655 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed, all deemed to contain the desired recombinant plasmid by restriction analysis using Clal and two, verified by DNA 5 sequencing, were kept as SJ10780 (MG1655/pSJ 10780) and SJ10781 (MG1655/pSJ 10781 ).

Construction of expression vector pSJ 10778 containing a L. plantarum acetoacetate decarboxylase gene

Plasmid pSJ10701 was digested with Ncol and Kpnl, and the resulting 0.85 kb i o fragment purified using gel electrophoresis. Plasmid pSJ10600 was digested with Ncol and Kpnl, and the resulting 5.1 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into MG1655 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed, all deemed to contain the 15 desired recombinant plasmid by restriction analysis using Clal and two, verified by DNA sequencing, were kept as SJ10778 (MG1655/pSJ 10778) and SJ10779 (MG1655/pSJ 10779).

Construction of expression vector pSJ 10768 containing a Lactobacillus antri isopropanol dehydrogenase gene

20 Plasmid pSJ10709 was digested with Kpnl and Xmal, and the resulting 1 .1 kb fragment purified using gel electrophoresis. Plasmid pSJ10600 was digested with Xmal and Kpnl, and the resulting 5.1 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates

25 at 37°C. Four of the resulting colonies were analyzed and two deemed to contain the desired recombinant plasmid by restriction analysis using Clal and verified by DNA sequencing, were kept as SJ 10768 (TG1/pSJ10768) and SJ 10769 (TG1/pSJ 10769).

Construction of expression vectors pSJ 10745, pSJ 10763, pSJ 10764, and pSJ 10767, containing a Thermoanaerobacter ethanolicus isopropanol dehydrogenase gene

30 Plasmid pSJ10719 was digested with BspHI and Xmal, and the resulting 1 .06 kb fragment purified using gel electrophoresis. Plasmid pSJ10600 was digested with Ncol and Xmal, and the resulting 5.1 kb fragment purified using gel electrophoresis. The purified fragments were mixed and ligated. The ligation mixture was transformed into MG1655 electrocompetent cells, and one of the resulting colonies, deemed to contain the desired recombinant plasmid by restriction analysis using Clal and verified by DNA sequencing, was kept as SJ10745 (MG1655/pSJ10745). The ligation mixture was also transformed into electrocompetent E. coli JM103, where two of four colonies were deemed to contain the desired plasmid by restriction analysis using Clal, and these kept as SJ10763 5 (JM103/pSJ10763) and SJ10764 (JM103/pSJ10764).

Finally, the ligation mixture was transformed into electrocompetent TG1 , where three of four colonies were deemed to contain the desired plasmid by restriction analysis using Clal, and one, SJ10767 (JM103/pSJ10767), was verified by DNA sequencing.

Construction of expression vector pSJ 10782 containing a Clostridium beijerinckii i o isopropanol dehydrogenase gene

Plasmid pSJ10725 was digested with BspHI and Xmal, and the resulting 1 .06 kb fragment purified using gel electrophoresis. Plasmid pSJ10600 was digested with Ncol and Xmal, and the resulting 5.1 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into MG1655

15 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and two, deemed to contain the desired recombinant plasmid by restriction analysis using Clal and verified by DNA sequencing, were kept as SJ10782 (TG1/pSJ10782) and SJ10783 (TG1/pSJ10783).

Construction of expression vector pSJ 10762 containing a Lactobacillus fermentum

20 isopropanol dehydrogenase gene

Plasmid pSJ10703 was digested with BspHI and Xmal, and the resulting 1 .1 kb fragment purified using gel electrophoresis. Plasmid pSJ10600 was digested with Xmal and Ncol, and the resulting 5.1 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into JM103 as well as

25 TG1 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Transformants were analyzed and two (one from each host strain), deemed to contain the desired recombinant plasmid by restriction analysis using Clal and verified by DNA sequencing, were kept as SJ10762 (JM103/pSJ10762) and SJ10765 (TG1/pSJ10765). Transformant SJ10766 (JM103/pSJ10766) was also verified to contain

30 the Lactobacillus fermentum isopropanol dehydrogenase gene.

Construction of expression vector pSJ 10954 containing a C. acetobutylicum thiolase gene, B. mojavensis succinyl-CoA:acetoacetate transferase genes (both subunits), a C. beijerinckii acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene Plasmid pSJ10798 was digested with Xhol and Xmal, and the resulting 6.3 kb fragment purified using gel electrophoresis. Plasmid pSJ10777 was digested with Xhol and Eagl, and the resulting 1 .43 kb fragment purified using gel electrophoresis. Plasmid pSJ10843 was digested with Eagl and Xmal, and the resulting 1 .85 kb fragment purified using gel electrophoresis. The three purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Xbal, and two of these were kept, resulting in SJ10954 (TG1/pSJ 10954) and SJ10955 (TG1/pSJ10955).

Construction of expression vector pSJ 10956 containing a C. acetobutylicum thiolase gene, B. mojavensis succinyl-CoA:acetoacetate transferase genes (both subunits), a C. acetobutylicum acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

Plasmid pSJ10798 was digested with Xhol and Xmal, and the resulting 6.3 kb fragment purified using gel electrophoresis. Plasmid pSJ10777 was digested with Xhol and Eagl, and the resulting 1 .43 kb fragment purified using gel electrophoresis. Plasmid pSJ10841 was digested with Eagl and Xmal, and the resulting 1 .89 kb fragment purified using gel electrophoresis. The three purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Xbal, and two of these were kept, resulting in SJ10956 (TG1/pSJ 10956) and SJ10957 (TG1/pSJ10957).

From an independent construction process (digestion, fragment purification, ligation, transformation by electroporation) one transformant, deemed to contain the desired recombinant plasmid by restriction analysis using Xbal, was kept as SJ 10926 (TG1 PSJ10926).

Construction of expression vector pSJ 10942 containing a C. acetobutylicum thiolase gene, B. subtilis succinyl-CoA:acetoacetate transferase genes (both subunits), a C. beijerinckii acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

Plasmid pSJ10798 was digested with Xhol and Xmal, and the resulting 6.3 kb fragment purified using gel electrophoresis. Plasmid pSJ10748 was digested with Xhol and

Eagl, and the resulting 1 .43 kb fragment purified using gel electrophoresis. Plasmid pSJ10843 was digested with Eagl and Xmal, and the resulting 1 .85 kb fragment purified using gel electrophoresis. The three purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Xbal, and two of these were kept, resulting in SJ10942 (TG1/pSJ 10942) and SJ10943 (TG1/pSJ10943).

Construction of expression vector pSJ 10944 containing a C. acetobutylicum thiolase gene, B. subtilis succinyl-CoA:acetoacetate transferase genes (both subunits), a C. acetobutylicum acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

Plasmid pSJ10798 was digested with Xhol and Xmal, and the resulting 6.3 kb fragment purified using gel electrophoresis. Plasmid pSJ10748 was digested with Xhol and Eagl, and the resulting 1 .43 kb fragment purified using gel electrophoresis. Plasmid pSJ10841 was digested with Eagl and Xmal, and the resulting 1 .89 kb fragment purified using gel electrophoresis. The three purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Xbal, and two of these were kept, resulting in SJ10944 (TG1/pSJ 10944) and SJ10945 (TG1/pSJ10945).

Construction of expression vector pSJ 10946 containing a C. acetobutylicum thiolase gene, an E. coli acetoacetyl-CoA transferase genes (both subunits), a C. beijerinckii acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

Plasmid pSJ10798 was digested with Xhol and Xmal, and the resulting 6.3 kb fragment purified using gel electrophoresis. Plasmid pSJ10750 was digested with Xhol and Eagl, and the resulting 1 .37 kb fragment purified using gel electrophoresis. Plasmid pSJ10843 was digested with Eagl and Xmal, and the resulting 1 .85 kb fragment purified using gel electrophoresis. The three purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Xbal, and two of these were kept, resulting in SJ10946 (TG1/pSJ 10946) and SJ10947 (TG1/pSJ10947).

Construction of expression vector pSJ 10948 containing a C. acetobutylicum thiolase gene, E. coli acetoacetyl-CoA transferase genes (both subunits), a C. acetobutylicum acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

Plasmid pSJ10798 was digested with Xhol and Xmal, and the resulting 6.3 kb fragment purified using gel electrophoresis. Plasmid pSJ10750 was digested with Xhol and 5 Eagl, and the resulting 1 .37 kb fragment purified using gel electrophoresis. Plasmid pSJ10841 was digested with Eagl and Xmal, and the resulting 1 .89 kb fragment purified using gel electrophoresis. The three purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and i o deemed to contain the desired recombinant plasmid by restriction analysis using Xbal, and two of these were kept, resulting in SJ10948 (TG1/pSJ 10948) and SJ10949 (TG1/pSJ10949).

Construction of expression vector pSJ 10950 containing a C. acetobutylicum thiolase gene, C. acetobutylicum acetoacetyl-CoA transferase genes (both subunits), a C. beijerinckii

15 acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

Plasmid pSJ10798 was digested with Xhol and Xmal, and the resulting 6.3 kb fragment purified using gel electrophoresis. Plasmid pSJ10752 was digested with Xhol and Eagl, and the resulting 1 .38 kb fragment purified using gel electrophoresis. Plasmid pSJ10843 was digested with Eagl and Xmal, and the resulting 1 .85 kb fragment purified

20 using gel electrophoresis. The three purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Xbal, and two of these were kept, resulting in SJ10950 (TG1/pSJ 10950) and SJ10951

25 (TG1/pSJ10951 ).

Construction of expression vector pSJ 10952 containing a C. acetobutylicum thiolase gene, C. acetobutylicum acetoacetyl-CoA transferase genes (both subunits), a C. acetobutylicum acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

Plasmid pSJ10798 was digested with Xhol and Xmal, and the resulting 6.3 kb 30 fragment purified using gel electrophoresis. Plasmid pSJ10752 was digested with Xhol and Eagl, and the resulting 1 .38 kb fragment purified using gel electrophoresis. Plasmid pSJ10841 was digested with Eagl and Xmal, and the resulting 1 .89 kb fragment purified using gel electrophoresis. The three purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Xbal, and two of these were kept, resulting in SJ10952 (TG1/pSJ 10952) and SJ10953 (TG1/pSJ10953).

Construction of expression vector pSJ 10790 containing a C. acetobutylicum thiolase gene, B. mojavensis succinyl-CoA:acetoacetate transferase genes (both subunits), a C. beijerinckii acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene under control of the P1 1 promoter

Plasmid pTRGU00178 (see US Provisional Patent Application No. 61/408,138, filed October 29, 2010) was digested with Ncol and BamHI, and the resulting 1 .2 kb fragment purified using gel electrophoresis. pTRGU00178 was also digested with BamHI and Sail, and the resulting 2.1 kb fragment purified using gel electrophoresis. pSIP409 was digested with Ncol and Xhol, and the resulting 5.7 kb fragment purified using gel electrophoresis. The three purified fragments were mixed, ligated, and the ligation mixture transformed into SJ2 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Two transformants, deemed to contain the desired recombinant plasmid by restriction analysis using EcoRI, Bglll, and Hindlll, were kept as SJ10562 (SJ2/pSJ 10562) and SJ 10563 (SJ2/pSJ 10563).

Plasmid pSJ10562 was digested with Xbal and Notl, and the resulting 7.57 kb fragment purified using gel electrophoresis. Plasmid pTRGU00200 (supra) was digested with Xbal and Notl, and the resulting 2.52 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into MG1655 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Two transformants, deemed to contain the desired recombinant plasmid by restriction analysis using Notl + Xbal, were kept as SJ10593 (MG1655/pSJ 10593) and SJ10594 (MG1655/pSJ 10594).

Plasmid pTRGU00200 was digested with EcoRI and BamHI, and the resulting 1 .2 kb fragment purified using gel electrophoresis. pSJ 10600 was digested with EcoRI and BamHI, and the resulting 5.2 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into MG1655 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Two transformants, deemed to contain the desired recombinant plasmid by restriction analysis using EcoRI + BamHI, were kept as SJ10690 (MG1655/pSJ 10690) and SJ10691 (MG1655/pSJ10691 ).

Plasmid pSJ10593 was digested with BamHI and Xbal, and the resulting 3.25 kb fragment purified using gel electrophoresis. pSJ 10690 was digested with BamHI and Xbal, and the resulting 6.3 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Two transformants, deemed to contain the desired recombinant plasmid by restriction analysis using Nsil, were kept as SJ10790 (TG1/pSJ10790) and SJ10791 (TG1/pSJ10791 ).

Construction of pSJ10792 containing a C. acetobutylicum thiolase gene, B. mojavensis succinyl-CoA:acetoacetate transferase genes (both subunits), a C. beijerinckii acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene under control of the P27 promoter

Plasmid pTRGU00200 was digested with EcoRI and BamHI, and the resulting 1 .2 kb fragment purified using gel electrophoresis. pSJ 10603 was digested with EcoRI and BamHI, and the resulting 5.2 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into MG1655 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Two transformants, deemed to contain the desired recombinant plasmid by restriction analysis using EcoRI + BamHI, were kept as SJ10692 (MG1655/pSJ 10692) and SJ10693 (MG1655/pSJ 10693).

Plasmid pSJ10593 was digested with BamHI and Xbal, and the resulting 3.25 kb fragment purified using gel electrophoresis. pSJ 10692 was digested with BamHI and Xbal, and the resulting 6.3 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Two transformants, deemed to contain the desired recombinant plasmid by restriction analysis using Nsil, were kept as SJ10792 (TG1/pSJ 10792) and SJ 10793 (TG1/pSJ10793).

Construction of expression vector pSJ1 1208 containing a L. reuteri thiolase gene, B. mojavensis succinyl-CoA:acetoacetate transferase genes (both subunits), a C. beijerinckii acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

Plasmid pSJ10796 was digested with Xhol and Xmal, and the resulting 6.3 kb fragment purified using gel electrophoresis. Plasmid pSJ10954 was digested with Xhol and

Xmal, and the resulting 3.28 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at

37°C. Three of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Xbal, and two of these were kept, resulting in SJ1 1208 (TG1/pSJ1 1208) and SJ1 1209 (TG1/pSJ1 1209).

Construction of expression vector pSJ1 1204 containing a L. reuteri thiolase gene, B. subtilis succinyl-CoA:acetoacetate transferase genes (both subunits), a C. beijerinckii acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene 5 Plasmid pSJ10796 was digested with Xhol and Xmal, and the resulting 6.3 kb fragment purified using gel electrophoresis. Plasmid pSJ10942 was digested with Xhol and Xmal, and the resulting 3.26 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at i o 37°C. Four of the resulting colonies were analyzed and deemed to contain the desired recombinant plasmid by restriction analysis using Xbal, and two of these were kept, resulting in SJ1 1204 (TG1/pSJ1 1204) and SJ1 1205 (TG1/pSJ1 1205).

Construction of expression vector pSJ1 1339 containing a L. reuteri thiolase gene, B. subtilis succinyl-CoA:acetoacetate transferase genes (both subunits), a C. beijerinckii

15 acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

Plasmid pSJ10802 was digested with Nhel and Xhol, and the resulting 1 .5 kb fragment purified using gel electrophoresis. Plasmid pSJ 10802 was separately digested with Xmal and Nhel, and the resulting 4.8 kb fragment purified using gel electrophoresis. Plasmid pSJ10942 was digested with Xhol and Xmal, and the resulting 3.26 kb fragment

20 purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 electrocompetent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed, two were deemed to contain the desired recombinant plasmid by restriction analysis using EcoRI + Nhel, and these were kept as SJ1 1339 (TG1/pSJ1 1339) and

25 SJ1 1340 (TG1/pSJ1 1340).

Construction of expression vector pSJ1 1230 containing a L. reuteri thiolase gene, E. coli acetoacetyl-CoA transferase genes (both subunits), a C. beijerinckii acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

Plasmid pSJ10796 was digested with Xhol and Xmal, and the resulting 6.3 kb 30 fragment purified using gel electrophoresis. Plasmid pSJ10946 was digested with Xhol and Xmal, and the resulting 3.23 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Seven of the resulting colonies were analyzed and 5 deemed to contain the desired recombinant plasmid by restriction analysis using Xbal, and two of these were kept, resulting in SJ1 1230 (TG1/pSJ1 1230) and SJ1 1231 (TG1/pSJ 1 1231 ).

Construction of expression vector pSJ1 1206 containing a L. reuteri thiolase gene, C. acetobutylicum acetoacetyl-CoA transferase genes (both subunits), a C. beijerinckii acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

Plasmid pSJ10796 was digested with Xhol and Xmal, and the resulting 6.3 kb fragment purified using gel electrophoresis. Plasmid pSJ 10951 was digested with Xhol and Xmal, and the resulting 3.23 kb fragment purified using gel electrophoresis. The purified fragments were mixed, ligated, and the ligation mixture transformed into TG1 chemically competent cells, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the resulting colonies were analyzed and two, deemed to contain the desired recombinant plasmid by restriction analysis using Xbal, were kept as SJ1 1206 (TG1/pSJ1 1206) and SJ1 1207 (TG1/pSJ1 1207).

Construction of expression vectors pSJ1 1492 and pSJ1 1533 containing a L. reuteri thiolase gene, Eremococcus coleocola butyrate-acetoacetate CoA transferase genes (both subunits), a C. beijerinckii acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

Each of the two different combinations of codon-optimized Eremococcus coleocola butyrate-acetoacetate CoA transferase genes were excised from plasmids pSJ 1 1474 and pSJ1 1476, respectively, by digestion with Xhol and Eagl, and the resulting 1 .35 kb fragments purified using gel electrophoresis. Plasmid pSJ1 1231 was digested with Xhol and Eagl, and the resulting 8.15 kb fragment purified using gel electrophoresis. Each of the Eremococcus coleocola butyrate-acetoacetate CoA transferase gene fragments was individually ligated to the pSJ1 1231 fragment, and the ligation mixture transformed into £. coli TG1 by electroporation, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C. Four of the colonies resulting from ligation with the pSJ1 1474 fragment were analyzed, three were deemed to contain the desired recombinant plasmid by restriction analysis using Ncol, and two were kept as SJ1 1492 (TG1/pSJ1 1492) and SJ1 1493 (TG1/pSJ1 1493). 16 of the colonies resulting from ligation with the pSJ1 1476 fragment were analyzed, four deemed to contain the desired recombinant plasmid by restriction analysis using Ncol, and two were kept as SJ1 1533 (TG1/pSJ1 1533) and SJ1 1539 (TG1/pSJ1 1539).

Construction of expression vector pSJ1 1490 containing a L. reuteri thiolase gene, Alicyclobacillus succinyl-CoA:acetoacetate transferase genes (both subunits), a C. beijerinckii acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

The combination of codon-optimized Alicyclobacillus succinyl-CoA:acetoacetate transferase genes was excised from plasmid pSJ1 1472, by digestion with Xhol and EagI, 5 and the resulting 1 .35 kb fragment purified using gel electrophoresis. Plasmid pSJ1 1231 was digested with Xhol and EagI, and the resulting 8.15 kb fragment purified using gel electrophoresis. The Alicyclobacillus succinyl-CoA:acetoacetate transferase gene fragment was ligated to the pSJ1 1231 fragment, and the ligation mixture transformed into E. coli TG1 by electroporation, selecting erythromycin resistance (200 microgram/ml) on LB plates i o at 37°C. Four of the resulting colonies were analyzed, two were deemed to contain the desired recombinant plasmid by restriction analysis using Bglll, and these were kept as SJ1 1490 (TG1/pSJ1 1490) and SJ1 1491 (TG1/pSJ1 1491 ).

Construction of expression vectors pSJ1 1540 and pSJ1 1513 containing a L. reuteri thiolase gene, Helicobacter pylori succinyl-CoA:acetoacetate transferase genes (both

15 subunits), a C. beijerinckii acetoacetate decarboxylase gene, and a C. beijerinckii alcohol dehydrogenase gene

Each of the two different combinations of codon-optimized Helicobacter pylori succinyl-CoA:acetoacetate transferase genes were excised from plasmids pSJ 1 1466 and pSJ1 1468, respectively, by digestion with Xhol and EagI, and the resulting 1 .35 kb

20 fragments purified using gel electrophoresis. Plasmid pSJ1 1231 was digested with Xhol and EagI, and the resulting 8.15 kb fragment purified using gel electrophoresis. Each of the Helicobacter pylori succinyl-CoA:acetoacetate transferase gene fragments was individually ligated to the pSJ1 1231 fragment, and the ligation mixture transformed into E. coli TG1 by electroporation, selecting erythromycin resistance (200 microgram/ml) on LB plates at 37°C.

25 Seven of the colonies resulting from ligation with the pSJ1 1466 fragment were analyzed, five were deemed to contain the desired recombinant plasmid by restriction analysis using BsaHI, and two were kept as SJ1 1540 (TG1/pSJ1 1540) and SJ1 1541 (TG1/pSJ1 1541 ). 4 of the colonies resulting from ligation with the pSJ1 1468 fragment were analyzed, three deemed to contain the desired recombinant plasmid by restriction analysis using Ndel and

30 Bglll, and two were kept as SJ1 1513 (TG1/pSJ1 1513) and SJ1 1514 (TG1/pSJ1 1514).

Example 12: Co-production of n-propanol and isopropanol from a Lactobacillus reuteri strain containing a methylglyoxal synthase gene sequence and an active isopropanol pathway.

35 A plasmid containing a heterologous polynucleotide encoding the B. subtilis methylglyoxal synthase of SEQ ID NO: 1 13 and an active isopropanol pathway genes was constructed as follows. The B. subtilis methylglyoxal synthase coding sequence from pBKQ559 was obtained by PCR amplification using primers pr068 (5'-CTGCG GTCGA CGGGG TTTAG TTGTT GACAG GG-3'; SEQ ID NO: 151 ) and 677372 (5'-GAGCA CACGG TTTAA CGAC-3'; SEQ ID NO: 152) using Phusion DNA polymerase, which resulted in a 645 bp PCR fragment. The PCR fragment was purified with Qiagen PCR purification kit and then digested with Sal\+Xba\, resulting in a 532 bp fragment containing the P27 promoter upstream to B. subtilis methylglyoxal synthase coding sequence. Plasmid pSJ 1 1533 {supra; containing heterologous polynucleotides encoding a thiolase, a butyrate- acetoacetate CoA transferase, an acetoacetate decarboxylase, and a C. beijerinckii alcohol dehydrogenase in operon structure expressed from the synthetic P1 1 promoter) was digested with Nhe\+Xma\+PshA\ of which a fragment of 4.7 kb contained the P1 1 promoter upstream to the isopropanol pathway genes. Plasmid vector pSIP41 1 (supra) was digested with Sal\+Xma\, resulting in a plasmid vector fragment of 3.1 kb. A triple fragment cloning was performed by ligation of 1 ) the PCR product containing P27-S. subtilis methylglyoxal synthase coding sequence, 2) the P1 1 -isopropanol pathway fragment, and 3) the plasmid vector part from pSIP41 1 . The ligation mixture was introduced to E. coli TG1 by electroporation. The resulting plasmid, pBKQ643 (Figure 19), contained the mgsA_Bs gene expressed from the P27 promoter and the isopropanol pathway genes expressed from the P1 1 promoter. The plasmid construction was then introduced to Lb. reuteri TRGU1013 by standard electroporation plated on MRS+10microgram/ml of erythromycin and incubated anaerobically at 37°C. The resulting strain was designated Lb. reuteri BKQ676 (contains pBKQ643).

Lactobacillus reuteri strain BKQ676 was subjected to batch fermentation of sugar cane juice supplemented with yeast extract in pH adjusted fermenter. A preculture was made by using 100 μΙ_ of a glycerol (20% w/w) preserved preparation of BKq676 to inoculate 40 mL of a sugar cane juice based medium (BRIX 5) containing 10 g/L yeast extract, 1 g/L Tween80, 50 mg/L MnS0₄-H₂0 and 10 mg/L erythromycin. The culture was incubated in a closed tube at 37°C. After approximately 24 hours, the preculture was used as seed to a 3 liter tank containing 2 liter of fermentation medium. The medium was prepared by adding 1 mL/L pluronic, 10 g/L yeast extract (Bacto 212720), 1 g/L Tween 80, 25 mg/L MnS0₄-H₂0, 650 mg/L phytic acid, 1 g/L L-serine, 1 g/L l-threonine, 2 g/L KH₂P0₄ to sugar cane juice that had been adjusted with water to BRIX 10. This mixture was autoclaved at 121 -123 C for 30 minutes. After reaching room temperature, the following components were added by sterile filtration: 400 mg/L uridine and 1 g/L L-glutamic acid dissolved in 10 mL water; and 10 mg/L erythromycin (added as a 5 mg/mL solution in ethanol). The temperature was held constant at 37°C and pH was held constant at pH 5.5 by addition of aqueous ammonia (NH3, aq.). The culture was kept anaerobic by constantly adding nitrogen (N2) to the tank. The agitation rate (to ensure mixing) was set at 400 RPM. 5 After 49 hours the fermentation resulted in combined production of 0.81 g/L of n- propanol, 0.86 g/L of isopropanol and 0.63 g/L of acetone. Assuming that all acetone can be converted into isopropanol the corresponding propanol equivalent titer would reach 2.3 g/L. i o Example 13: Co-production of n-propanol and isopropanol from a Lactobacillus reuteri strain containing a chromosomally integrated methylglyoxal synthase gene sequence and an active isopropanol pathway.

Chromosomal integration of a B. subtilis methylglyoxal synthase gene sequence (construction of TRGU1215)

15 Bacillus subtilis subsp. spizizenii was inoculated overnight in TY bouillon at 37°C.

The following day, the culture was harvested and genomic DNA was isolated using QiaAMP^(R) DNA Blood mini kit (Qiagen). The Bacillus subtilis subsp. spizizenii methylglyoxal synthase coding sequence (SEQ ID NO: 128) encoding the methylglyoxal synthase of SEQ ID NO: 1 14 was amplified with PCR on the isolated genomic DNA as template using

20 forward primer P643 (5'-CATGg gatcc ACCTG CCCCG TTAGt tttcg gcagt ttgtt gacag tgcct ccgac tcatg gtaaa atggg aatat agcgt actta gctgg ccagc atata tgtat tctat aaaat actat tacaa ggaga tttta gccat gaaaa ttgct ttgat cgcgc atga-3'; SEQ ID NO: 179) and reverse primer P644 (5'-CATGg gatcc CTAAC GGGGC AGGTT TATAC ATTCG GCTCT TCTCC CCGAA- 3'; SEQ ID NO: 180). Primers P643 and P644 were designed to incorporate BamH\ sites

25 flanking the amplified gene and in addition P643 incorporated a synthetic promoter SP2 (indicated with lowercase letters), which had previously showed high expression levels of GusA in Lactobacillus reuteri. Ten PCR reactions were set up using Phusion High-Fidelity DNA Polymerase (New England Biolabs) with the following components for each reaction: 28.5 microliter H₂0, 10 microliter HF buffer, 8 microliter dNTP (5mM), 1 microliter P643 (20

30 micromolar), 1 microliter P644 (20 micromolar), 1 microliter genomic DNA diluted 10 fold, and 0.5 microliter Phusion polymerase.

The resulting 576 bp PCR product was purified from solution and 24 ul was digested with BamHI overnight at 37°C using 3 ul BamHI (20000 U/microliter), 3 microliter NEB 3 buffer and 0.3 microliter BSA. The digested PCR product was then purified using agarose

35 gel electrophoresis by excising the band of correct size as estimated by the applied DNA size marker. The resulting DNA concentration after purification was measured on a Nanodrop Model ND-1000 (Nanodrop Technologies, DE, USA) to 18 ng/ul.

Four ml of an overnight 10 ml culture of E. coli TG1 SJ1 1646 (comprising the LAR_0818 disruption vector pSJ1 1646; see U.S. Provisional Application No. 61/648,958) was harvested and the plasmid pSJ 1 1646 was isolated using a Qiagen miniprep kit. Twenty microliters of the miniprep was digested with 4 ul BamHI (20000 U/microliter), 0.3 microliter BSA, 3 microliter NEB 3 buffer, and 3 microliter H₂0. The two restriction reactions were incubated overnight at 37°C, then heat inactivated at 65°C for 30 min and finally 1 microliter calf intestine protease (CIP) was added and incubated at 37 °C for 1 hr in order to dephosphorylate the vector prior to agarose gel electrophoresis purification and ligation. The concentration of the vector was measured after purification to 12 ng/microliter. The ligation was set up using 4 microliter pSJ1 1646 (BamHI + CIP) and 4 microliter PCR product (BamHI) together with 1 microliter 10x ligation buffer and 1 ul T4 DNA ligase. A negative control was included with 4 microliter pSJ1 1646 (BamHI + CIP) and 4 microliter H₂0 together with 1 microliter 10x ligation buffer and 1 microliter T4 DNA ligase. The ligation was incubated overnight at room temperature and 5 microliter was withdrawn and used for chemical transformation of E. coli TG1 as described above. The transformed cells were plated after 2 hours incubation at 37°C with vigorous shaking onto LB agar plates containing 200 microgram/ml erythromycin. The following day six transformants with pSJ1 1646/PCR and two transformants for pSJ1 1646/H₂0 were inoculated for minipreps and the isolated plasmids were subjected to restriction analysis with BamHI. The expected restriction pattern on agarose gel electrophoresis consisted of a 5722 and a 562 bp band. One colony had this restriction profile and the plasmid was submitted for sequencing. The sequence showed that the gene had been cloned in the counter clockwise orientation (see pTRGU 1200; Figure 24) and the strain designated TRGU1200 was stored at -80°C in 10% glycerol.

To chromosomally integrate the Bacillus subtilis subsp. spizizenii methylglyoxal synthase coding sequence (SEQ ID NO: 128) from pTRGU 1200 on Lactobacillus reuteri SJ1 1400 {supra), strain SJ1 1400 was transformed with pTRGU 1200 via electroporation by procedure B (except that 3 microliter plasmids was used). Five transformants were obtained after 2 days incubation at 30°C, and were inoculated in MRS containing 10 microgram/ml erythromycin. After overnight growth the cultures were streak purified on MRS agar plates containing 10 microgram/ml erythromycin and incubated at 45°C to select for cells in which the first homologous recombination event had occurred as this temperature is non- permissive of plasmid replication. After overnight incubation at 45°C, numerous colonies had appeared and 40 of them were inoculated in MRS at 30°C for 1 day to increase the chance for the second homologous recombination event to occur. The cells then were plated onto MRS agar plates and after 2 days incubation at 45°C, Plenty of colonies were replicated on to MRS agar plates and MRS agar plates containing 10 microgram/ml 5 erythromycin to test for loss of the erythromycin resistance gene. In addition, during replication the colonies were subjected to colony PCR using primers BKL700148 (5'- GCGAT GGTTA AACAA CAAAA TG-3'; SEQ ID NO: 181 ) and BKL700149 (5'-CCACA ATAAA TCACC TCTTT CTG-3'; SEQ ID NO: 182).

When the methylglyoxal synthase gene sequence replaces LAR_0818 the amplicon i o is expected to be 1677 bp, whereas when the mutant reverts to the wild type during the second homologous recombination event, the amplicon is expected be approximately 1 100 bp. A wild type SJ1 1400 colony was included in the colony PCR analysis. Several of the mutants had reverted to the wild type, while several colonies also resulted in two amplicons of exactly 1 100 bp and 1677 bp. All 40 colonies were analyzed with colony PCR using

15 primers P643 and P644 specific for the methylglyoxal synthase gene sequence. All colonies that had resulted in only the wild type 1 100 bp band with primer set BKL700148/BKL700149 did not result in amplification product with P643/P644. However, all colonies resulting in two bands with primer set BKL700148/BKL700149 also resulted in amplification products with methylglyoxal synthase gene sequence primers. Thus, colonies

20 consisted of cells of wild type and mutant genotypes. Several of the mixed genotype colonies were streak purified and the incubated overnight at 37°C on MRS agar plates. 96 colonies of one of these streak purified colonies were then subjected to colony PCR analyses with the primer set BKL700148/BKL700149. One of the colonies resulted in a strong 1672 bp amplifcation product that was subsequently analyzed by restriction analysis

25 using BamHI to produce three fragments of approximately 500 bp (commensurate with the expected 631 bp, 562 bp, and 484 bp fragments). The resulting strain designated TRGU1215 was stored at -80 °C in 10% glycerol. The genotype was confirmed by sequencing of the BKL700148/BKL700149 amplification product.

Expression of methylglyoxal synthase of SEQ ID NO: 1 14 was confirmed by

30 incubating TRGU1215 overnight in 10 ml MRS. Lysate then was prepared from the isolated cells and a sample submitted for MS analysis. A significant hit of the methylglyoxal synthase protein was detected and estimated to approximately 0.007% of all intracellular proteins. Disruption of the acetate kinase gene ackA1 and transformation of plasmid pSJ 1 1533 into TRGU1215 (construction of BKq776)

Competent cells of TRGU1215 {supra) was prepared by standard methods as described above and transformed with pSJ1 1503 (Figure 25; see U.S. Provisional Application No. 61/653,908) by electroporation procedure B. Nine transformants were obtained after 2 days on MRS agar plates containing 10 microgram/ml erythromycin at 30°C. All transformants were inoculated in MRS containing 10 microgram/ml erythromycin and incubated overnight at 30°C. Weak growth was detected in all cultures. One transformant was selected based on a positive PCR analysis as verified by a 2400 bp PCR amplification product using primers BKL692431 (5'-TTTGA ATTAA TGGAG GCTCG T-3'; SEQ ID NO: 183) and BKL309520 (5'-GCCAG TCATT AGGCC TATC-3'; SEQ ID NO: 184) The selected transformant was propagated in MRS containing 6 microgram/ml chloramphenicol at 30°C for several days, and then plated onto MRS agar plates containing 6 microgram/ml chloramphenicol. 40 colonies were subjected to PCR analyses using primers BKL692431 {supra) and BKL692432 (5'-CGCGG TAACA TTAAT ATCAT GA-3'; SEQ ID NO: 185) flanking the homologous regions flanking Res-cat-Res. 15 microliter of each PCR reaction was digested with BamHI using 1 microliter BamHI, 0.2 microliter BSA, 2 microliter NEB 3 buffer and 3.8 microliter H₂0. One colony resulted in a vague PCR amplification product which was digested by BamHI, confirming that the colony harbored the correct ackA1 negative genotype. The inability of this strain to grow on MRS agar plate containing 6 microgram/ml chloramphenicol and 10 microgram/ml erythromycin but clear growth on MRS agar plates with 6 microgram/ml chloramphenicol showed that it had lost the erythromycin resistance gene. The resulting strain was designated TRGU1276 stored at -80°C in 10% glycerol.

Strain TRGU1276 was then transformed with pSJ1 1533 as described above to produce the desired strain designated BKq776 which contains a chromosomally integrated B. subtilis subsp. spizizenii methylglyoxal synthase gene sequence and an active isopropanol pathway.

Fermentation of Lactobacillus reuteri strain BKq776 containing a chromosomally integrated B. subtilis methylglyoxal synthase gene sequence and an active isopropanol pathway

A preculture was made by using 100 μΙ_ of a glycerol (20% w/w) preserved preparation of BKq776 to inoculated 40 mL of a sugar cane juice based medium (BRIX 5) containing 10 g/L yeast extract, 1 g/L Tween80, 50 mg/L MnS04-H20 and 10 mg/L erythromycin. The culture was incubated in a closed tube at 37°C. After approximately 48 hours, the preculture was used as seed to a 3 liter tank containing 2 liter of fermentation medium. The medium was prepared by adding 1 mL/L pluronic, 20 g/L yeast extract (Bacto 212720), 1 g/L Tween 80, 25 mg/L MnS0₄-H₂0, 650 mg/L phytic acid, 1 g/L L-serine, 1 g/L L-threonine, 2 g/L KH₂P0₄ to sugar cane juice that had been adjusted with water to BRIX 5. This mixture was autoclaved at 121 -123°C for 30 minutes. After reaching room temperature, the following components were added by sterile filtration: 400 mg/L uridine and 1 g/L L-glutamic acid dissolved in 10 mL water; and 10 mg/L erythromycin (added as a 5 mg/mL solution in ethanol).

In a first fermentation, the temperature was held constant at 37°C and pH was held constant at pH 5.5 by addition of aqueous ammonia (NH₃, aq.) ant the culture was kept anaerobic by constantly adding nitrogen (N₂) to the tank. The agitation rate (to ensure mixing) was set at 400 RPM. The feed was started 6 hours after inoculation and the feed rate was held constant at 0.1 g/min for the remaining part of the fermentation. The feed consisted of sugar cane syrup (BRIX 64) and pluronic 1 g/L. Samples were taken throughout the fermentation, resulting in significant coproduction of n-propanol and isopropanol (reaching >7 g/L total propanol), as shown in Figure 7.

A second fermentation was conducted as described above, except that the temperature was held constant at 37°C for the first 16 hours of the fermentation, ramped down in a linear fashion to 34°C between 16 and 19 hours, and then held constant at 34°C for the remaining part of the fermentation. Samples were taken throughout the fermentation, resulting in significant coproduction of n-propanol and isopropanol (reaching >8 g/L total propanol), as shown in Figure 8.

Example 14: Construction of plasmid constructs for expressing a triosephosphate isomerase gene.

Construction of plasmid pBKQ729 expressing tpiA LP 0791 from Lb. plantarum WCFS1

PCR amplification with primer set pr078 (5'-GTTGC TCATG AGGAC ACCTA TTATT GCCGG TAAC-3'; SEQ ID NO: 162) and pr079 (5'-GGCTT TCTAG AGTTT GTCAA TCAAC TTGCA ATAAC-3'; SEQ ID NO: 163) using chromosomal DNA from Lactobacillus plantarum WCFS1 as template was performed to amplify a coding sequence (tpiA_LP_0791; SEQ ID NO: 168) encoding the Lactobacillus plantarum triosephosphate isomerase of SEQ ID NO: 169 using Phusion DNA polymerase. The PCR fragment was purified with Qiagen PCR purification kit and then digested with BspH\+Xba\, resulting in a 0.8 kb fragment. Plasmid vector pSJ 10603 (supra) was digested with Nco\+Xba\ and treated with calf intestine alkaline phosphatase resulting in a 5.1 kb fragment. The fragment containing tpiA_LP_0791 was then ligated to pSJ10603 and introduced to E. coli TG1 by electroporation. The resulting plasmid pBKQ729 (Figure 20) contained tpiA_LP_0791 expressed from the synthetic promoter P27. Plasmid pBKQ729 was then introduced to Lb. reuteri TRGU1215 {supra) by electroporation Procedure B as described above, plated on MRS+10 microgram/ml of erythromycin and incubated anaerobically at 37°C. The resulting strain was designated Lb. reuteri BKQ760.

Construction of plasmid pBKQ731 expressing tpiA LAR 0383 from Lb. reuteri JCM1 1 12

PCR amplification with primer set pr080 (5'-GTTGC TCATG AGAGT ACCGA TTATT GCTG-3'; SEQ ID NO: 164) and pr081 (5'-GGCTT TCTAG ATTAA AAACA AG CAT TTAAC CGC-3'; SEQ ID NO: 165) using chromosomal DNA of Lactobacillus reuteri JCM1 1 12 as template was performed to amplify a coding sequence (tpiA_LAR_0383; SEQ ID NO: 166) encoding the Lactobacillus reuteri triosephosphate isomerase of SEQ ID NO: 167 using phusion DNA polymerase. The PCR fragment was purified with Qiagen PCR purification kit and then digested with BspH\+Xba\, resulting in a 0.8 kb fragment. Plasmid vector pSJ 10603 (supra) was digested with Nco\+Xba\ and treated with calf intestine alkaline phosphatase resulting in a 5.1 kb fragment. The fragment containing tpiA_LAR_0383 was then ligated to pSJ10603 and introduced to £ coli TG1 by electroporation. The resulting plasmid pBKQ731 (Figure 21 ) contained tpiA_LAR_0383 expressed from the synthetic promoter P27. Plasmid pBKQ731 was then introduced to Lb. reuteri TRGU1215 {supra) by electroporation Procedure B as described above, plated on MRS+10microgram/ml of erythromycin and incubated anaerobically at 37°C. The resulting strain was designated Lb. reuteri BKQ764.

Construction of plasmid pTOID55 expressing tpiA b3919 from E. coli K-12 MG1655

PCR amplification with primer set ptoid227 (5'-GCAGC ATCAT GAGAC ATCCT TTAGT GATGG GTAAC-3'; SEQ ID NO: 192) and ptoid228 (5'-GCAGC AAGAT CTTTA AGCCT GTTTA GCCGC T-3'; SEQ ID NO: 193) using chromosomal DNA from £ coli K-12 MG1655 as template was performed to amplify a coding sequence (tpiA_b3919; SEQ ID NO: 194) encoding the £. coli triosephosphate isomerase of SEQ ID NO: 174 using Phusion DNA polymerase. The PCR fragment was purified with Qiagen PCR purification kit and then digested with BspH\+Bgl\\, resulting in a 0.8 kb fragment. Plasmid vector pSJ10603 (supra) was digested with Nco\+BamH\ and treated with calf intestine alkaline phosphatase resulting in a 5.1 kb fragment. The fragment containing tpiA_b3919 was then ligated to pSJ10603 and introduced to £ coli TG1 by heat shock. The resulting plasmid pTOID55 (Figure 28) contained tpiA_b3919 expressed from the synthetic promoter P27. Plasmid pTOID55 was then introduced to Lb. reuteri TRGU1215 (supra) by electroporation Procedure B as described above, plated on MRS+10 microgram/ml of erythromycin and incubated anaerobically at 37°C. The resulting strain was designated Lb. reuteri TOID69. Construction of plasmid pTOID81 expressing TPI1 YDR050C from S. cerevisiae S288c

PCR amplification with primer set ptoid252 (5'-GCAGC ACCAT GGCTA GAACT 5 TTCTT TGTCG-3'; SEQ ID NO: 196) and ptoid253 (5'-GCAGC AGGAT CCTTA GTTTC TAGAG TTGAT GATAT CAACA-3'; SEQ ID NO: 197) using chromosomal DNA from S. cerevisiae S288c as template was performed to amplify a coding sequence (TPI1_YDR050C; SEQ ID NO: 195) encoding the S. cerevisiae triosephosphate isomerase of SEQ ID NO: 173 using Phusion DNA polymerase. The PCR fragment was purified with i o Qiagen PCR purification kit and then digested with Nco\+BamH\, resulting in a 0.8 kb fragment. Plasmid vector pSJ10603 (supra) was digested with Nco\+BamH\ and treated with calf intestine alkaline phosphatase resulting in a 5.1 kb fragment. The fragment containing TPI1_ was then ligated to pSJ10603 and introduced to E. coli TG1 by heat shock. The resulting plasmid pTOID81 (Figure 29) contained TPI1_YDR050C expressed

15 from the synthetic promoter P27. Plasmid pTOID81 was then introduced to Lb. reuteri TRGU1215 {supra) by electroporation Procedure B as described above, plated on MRS+10 microgram/ml of erythromycin and incubated anaerobically at 37°C. The resulting strain was designated Lb. reuteri TOID95.

Construction of plasmid pTOID79 expressing a L. fermentum triose phosphate isomerase

20 gene (to/ Lreu) from L. reuteri CCUG 38012

PCR amplification with primer set ptoid250 (5'-GCAGC ACCAT GGCTC GCAAA CCATT TGTGG TC-3'; SEQ ID NO: 198) and ptoid251 (5'-GCAGC AGGAT CCCTA TTTAG CACCC TTATA AAGTT CG-3'; SEQ ID NO: 199) using chromosomal DNA from L. fermentum (AKA, L. reuteri CCUG 38012) as template was performed to amplify a coding

25 sequence (SEQ ID NO: 200) encoding the L. fermentum triosephosphate isomerase of SEQ ID NO: 201 using Phusion DNA polymerase. The PCR fragment was purified with Qiagen PCR purification kit and then digested with Nco\+BamY\\, resulting in a 0.8 kb fragment. Plasmid vector pSJ10603 (supra) was digested with Nco\+BamH\ and treated with calf intestine alkaline phosphatase resulting in a 5.1 kb fragment. The fragment

30 containing TPI coding sequence was then ligated to pSJ 10603 and introduced to E. coli

TG1 by heat shock. The resulting plasmid pTOID79 (Figure 30) contained the TPI coding sequence expressed from the synthetic promoter P27. Plasmid pTOID79 was then introduced to Lb. reuteri TRGU 1215 (supra) by electroporation Procedure B as described above, plated on MRS+10 microgram/ml of erythromycin and incubated anaerobically at

35 37°C. The resulting strain was designated Lb. reuteri TO I D94. Example 15: Production of n-propanol from Lactobacillus reuteri strains containing a chromosomally integrated methylglyoxal synthase gene sequence and expressing a triosephosphate isomerase gene sequence.

Strains Lb. reuteri BKQ760, Lb. reuteri BKQ764 and Lb. reuteri TOID69 {supra), containing a chromosomally integrated B. subtilis methylglyoxal synthase gene sequence and either tpiA_LP_0791 , tpiA_LAR_0383 or tpiA_b3919, respectively, were grown anaerobic at 37°C in MRS supplemented with 10 microgram/ml of erythromycin. Samples of supernatant were analyzed for product formation after two days as described supra, which resulted in production of up to 1 g/L of n-propanol+0.4 g/L of 1 ,2-propanediol or 1 .3 g/L of n-propanol (Table 14). The production of ethanol was found to decrease as the production of n-propanol increased, which is in good agreement with the fact that NAD(P)H is used in the production of both ethanol and n-propanol.

Strains Lb. reuteri BKQ760, Lb. reuteri BKQ764, Lb. reuteri TOID69, Lb. reuteri TOID94 and Lb. reuteri TOID95 (supra), containing a chromosomally integrated B. subtilis methylglyoxal synthase gene sequence and either tpiA_LP_0791 , tpiA_LAR_0383, tpiA_b3919, tpi_Lreu or TPI1_YDR050C, respectively, were grown anaerobic at 37°C in MRS supplemented with 10 microgram/ml of erythromycin. Samples of supernatant were analyzed for product formation after two days as described supra, which resulted in production of up to 1 g/L of n-propanol+0.4 g/L of 1 ,2-propanediol or 1 .3 g/L of n-propanol (Table 14). The production of ethanol was found to decrease as the production of n- propanol increased, which is in good agreement with the fact that NAD(P)H is used in the production of both ethanol and n-propanol.

Table 14.

Example 16: Increased production of n-propanol from a Lactobacillus reuteri strain containing a chromosomally integrated methylglyoxal synthase gene sequence and a chromosomally integrated triosephosphate isomerase gene sequence.

Chromosomal resolvation of Res-cat-Res in TRGU1276, isolation of chloramphenicol sensitive strain TRGU 1283

5 Strain TRGU1276 (supra) was transformed with pSJ3008 (see U.S. Provisional

Application No. 61/653,908) using electroporation by procedure B. Transformants were isolated and cultivated overnight in MRS containing 10 microgram/ml erythromycin. The cells were then plated on MRS agar plates and colonies were isolated and replicated on to MRS agar plates containing either 6 microgram/ml chloramphenicol or 10 microgram/ml i o erythromycin. Colonies were isolated that both had resolved the cat gene as well as lost the pSJ3008 plasmid by their inability to grow on neither chloramphenicol nor erythromycin. One such colony, designated TRGU1283, was inoculated in MRS overnight at 37°C and stored at -80°C in 10% glycerol.

Construction of plasmid pBKQ726

15 PCR amplification with primer set pr080 (5'-GTTGC TCATG AGAGT ACCGA TTATT

GCTG-3'; SEQ ID NO: 164) and pr081 (5'-GGCTT TCTAG ATTAA AAACA AG CAT TTAAC CGC-3'; SEQ ID NO: 165) using chromosomal DNA of Lactobacillus reuteri JCM1 1 12 as template was performed to amplify a coding sequence (tpiA_LAR_0383; SEQ ID NO: 166) encoding the Lactobacillus reuteri triosephosphate isomerase of SEQ ID NO: 167 using

20 phusion DNA polymerase. The PCR fragment was purified with Qiagen PCR purification kit and then digested with BspH\+Xba\, resulting in a 0.8 kb fragment. Plasmid vector pSJ 10603 (supra) was digested with Nco\+Xba\ and treated with calf intestine alkaline phosphatase resulting in a 5.1 kb fragment. The fragment containing tpiA_LAR_0383 was then ligated to pSJ10600 and introduced to E. coli TG1 by electroporation. The resulting

25 plasmid pBKQ726 (Figure 26) contained tpiA_LAR_0383 expressed from the synthetic promoter P1 1 .

Construction of plasmid pTRGU1279

The triosephosphate isomerase gene sequence (including the promoter region) was excised from pBKQ726 by Nhel and Xbal restriction. The reaction contained 10 microliter

30 pBKQ726, 1 ul Xbal, 1 microliter Nhel, 2 microliter NEB 4 buffer and 6 microliter H₂0. A band of 1238 bp corresponding to tpiA was isolated after agarose gel electrophoresis. Ten microliter plasmid pSJ1 1728 (supra) was digested with 1 microliter Nhel, 2 microliter NEB 4 buffer and 7 microliter H₂0. The reaction was incubated overnight at 37°C. Subsequently, the vector was dephosphorylated by adding 1 microliter CI P and incubating 1 hour at 37°C.

35 A band of correct size corresponding to the linearized vector was purified from agarose gel electrophoresis. After purification of both vector and insert with Qiagen gel purification kit, the concentrations were measured to 36 ng/microliter for pSJ1 1728 and 9 ng/microliter for the sequence containing tpiA. A ligation was set up containing 3 microliter pSJ1 1728 (Nhel CIP), 1 .5 microliter of the tpiA-insert (Nhel Xbal), 0.5 microliter T4 DNA ligase, 1 microliter ligation buffer, and 5 microliter H₂0. A negative control was included with 5 microliter H₂0 instead of tpiA-insert. After 2 hours incubation at room temperature, E. coli TG1 cells were transformed by chemical transformation as described above using 5 microliter from each reaction. The following day, no E. coli TG1 transformants were obtained on the negative control whereas 6 transformants were obtained on the ligation reaction. Cultures were inoculated in order to make minipreps. The following day, the cultures were weak, but minipreps and subsequent restriction profile analyses with Xbal+Agel and Xbal+Nhel clearly showed that several plasmids were correctly cloned and that all plasmids had the triosephosphate isomerase gene sequence cloned counter clockwise. One such plasmid was designated pTRGU 1279 (Figure 27).

Chromosomal integration of tpiA in Lactobacillus reuteri TRGU1283, construction of strain TRGU1321

Stain TRGU1283 was transformed with pTRGU 1279 (supra) by electroporation using procedure B described above. Four transformants were grown overnight at 30°C, plated on MRS agar plates containing 10 microgram/ml eryhtomycin and incubated at 45°C overnight. 24 colonies were analyzed by PCR using primers P697/P698 and P699/P700 to check whether the integration events had occurred over the 5' homologous region or the 3' homologous region.

P697: 5'-TACTGATAAAATTTGTGATCCAGCT-3' (SEQ ID NO: 186)

P698: 5'-AGGAATTGTCAGATAGGCCTAATG-3' (SEQ ID NO: 187)

P699: 5'-CTTATG CATTTTC CAATTACCAG C-3 ' (SEQ ID NO: 188)

P700: 5'-TCGATCATTTGGCTTATCTAAAGA-3' (SEQ ID NO: 189)

Two colonies were selected for further incubation in MRS containing 6 microgram/ml chloramphenicol at 30°C as the PCR analyses showed that these colonies contained cells with mixed genotypes. The colonies were inoculated in MRS containing 6 microgram/ml chloramphenicol and incubated at 45°C for about 2 days. The cells were then plated on to MRS containing 6 microgram/ml chloramphenicol. After overnight incubation at 45°C, replicate 50 colonies onto MRS containing either 6 microgram/ml chloramphenicol or 10 microgram/ml erythromycin. After two days incubation at 45°C, one colony did not grow on MRS agar plates containing 10 microgram/ml erythromycin and was subjected to PCR analyses according to Table15 below. Table 15.

Primer set P701 (5'-TCTTA CCGTA TCTTG GTATT AACAA-3'; SEQ ID NO: 190) and P702 (5'-ATCTC TCCTC TGTCA TTTAT TTGGA-3'; SEQ ID NO: 191 ) was used to check that the plasmid had not recombined over the homologous LAR_0383 region. Agarose gel electrophoresis showed the correct pattern for the erythromycin sensitive mutant, indicating that the correct mutant had been generated. This strain, designated TRGU1321 , was stored at -80°C in 10% glycerol.

Fermentation of Lactobacillus reuteri strains TRgu1283 and TRgu1321

Lactobacillus reuteri strains TRgu1283 (containing a chromosomally integrated B. subtilis subsp. spizizenii methylglyoxal synthase gene sequence) and TRgu1321 (containing a chromosomally integrated B. subtilis subsp. spizizenii methylglyoxal synthase gene sequence and a chromosomally integrated Lb. reuteri triosephosphate isomerase gene sequence) were cultivated as follows: A preculture was made by using 100 μΙ_ of a glycerol (20% w/w) preserved preparation of the strain to inoculated 40 mL of a sugar cane juice based medium (BRIX 5) containing 10 g/L yeast extract, 1 g/L Tween80, 50 mg/L MnS0₄-H₂0 and 10 mg/L erythromycin. The culture was incubated in a closed tube at 37 C. After approximately 48 hours, the preculture was used as seed to a 3 liter tank containing 2 liter of fermentation medium. The medium was prepared by adding 1 mL/L pluronic, 20 g/L yeast extract (Bacto 212720), 1 g/L Tween 80, 25 mg/L MnS0₄-H₂0, 650 mg/L phytic acid, 1 g/L L-serine, 1 g/L L-threonine, 2 g/L KH₂P0₄ to sugar cane juice that had been adjusted with water to BRIX 10. This mixture was autoclaved at 121 -123°C for 30 minutes. After reaching room temperature, the following components were added by sterile filtration: 400 mg/L uridine and 1 g/L L-glutamic acid dissolved in 10 mL water; and 10 mg/L erythromycin (added as a 5 mg/mL solution in ethanol). The temperature was held constant at 37°C and pH was held constant at pH 5.5 by addition of aqueous ammonia (NH₃, aq.). The culture was kept anaerobic by constantly adding nitrogen (N₂) to the tank. The agitation rate (to ensure mixing) was set at 400 RPM. The feed was started 6 hours after inoculation and the feed rate was held constant at 0.1 g/min for the remaining part of the fermentation. The feed consisted of sugar cane syrup (BRIX 64) and pluronic 1 g/L. Samples were taken throughout the fermentation showed a significant increase in n-propanol production for TRgu1321 (containing a chromosomally integrated B. subtilis methylglyoxal synthase gene sequence and a chromosomally integrated Lb. reuteri triosephosphate isomerase gene sequence) compared to TRgu1283 (containing only a chromosomally integrated B. subtilis methylglyoxal synthase gene sequence) as shown in Figure 22. 1 ,2-propanediol was not detected in either of these fermentations.

Example 17: Co-production of n-propanol and isopropanol from a Lactobacillus reuteri strain containing a chromosomally integrated methylglyoxal synthase gene sequence, a chromosomally integrated triosephosphate isomerase gene sequence, and an active isopropanol pathway.

Lactobacillus reuteri strain TRgu1321 (supra) was transformed with pSJ 1 1533 using procedure B as described above to produce the desired strain designated TRgu1378 which contains a chromosomally integrated B. subtilis subsp. spizizenii methylglyoxal synthase gene sequence, a chromosomally integrated Lb. reuteri triosephosphate isomerase gene sequence, and an active isopropanol pathway.

Strain 1378 was cultivated as follows: A preculture was prepared by using 100 μΙ_ of a glycerol (20% w/w) preserved preparation of TRgu1378 to inoculated 40 mL of a sugar cane juice based medium (BRIX 5) containing 10 g/L yeast extract, 1 g/L Tween80, 50 mg/L MnS0₄-H₂0 and 10 mg/L erythromycin. The culture was incubated in a closed tube at 37 °C. After 1 day, the preculture was used as seed to a 3 liter tank containing 2 liter of fermentation medium. The medium was prepared by adding 1 mL/L pluronic, 20 g/L yeast extract (Bacto 212720), 1 g/L Tween 80, 25 mg/L MnS0₄-H₂0, 650 mg/L phytic acid, 1 g/L L-serine, 1 g/L l-threonine, 2 g/L KH₂P0₄ to sugar cane juice that had been adjusted with water to BRIX 5. This mixture was autoclaved at 121 -123°C for 30 minutes. After reaching room temperature, the following components were added by sterile filtration: 400 mg/L uridine and 1 g/L L-glutamic acid dissolved in 10 mL water; and 10 mg/L erythromycin (added as a 5 mg/mL solution in ethanol).

The temperature was held constant at 37°C and pH was held constant at pH 5.5 by addition of aqueous ammonia (NH₃, aq.). The culture was kept anaerobic by constantly adding nitrogen (N₂) to the tank. The agitation rate (to ensure mixing) was set at 400 RPM. The feed was started 6 hours after inoculation and the feed rate was held constant at 0.1 g/min for the remaining part of the fermentation. The feed consisted of sugar cane syrup (BRIX 64) and pluronic 1 g/L. Samples were taken throughout the fermentation, resulting in significant coproduction of n-propanol and isopropanol, as shown in Figure 23.

Although the foregoing has been described in some detail by way of illustration and example for the purposes of clarity of understanding, it is apparent to those skilled in the art that any equivalent aspect or modification may be practiced. Therefore, the description and examples should not be construed as limiting the scope of the invention.

In some aspects, the invention may be described by the following numbered paragraphs: [1 ] A recombinant Lactobacillus reuteri host cell comprising a heterologous polynucleotide encoding a methylglyoxal synthase, wherein the cell is capable of producing n-propanol.

[2] The recombinant host cell of paragraph [1 ], wherein the host cell is capable of producing a greater amount of n-propanol compared to the cell without the heterologous polynucleotide encoding the methylglyoxal synthase, when cultivated under identical conditions.

[3] The recombinant host cell of paragraph [1 ] or [2], wherein the host cell is capable of producing at least 10% more (e.g., at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 75%, at least 100%, or at least 200% more) n-propanol compared to the cell without the heterologous polynucleotide encoding the methylglyoxal synthase, when cultivated under identical conditions.

[4] The recombinant host cell of any one of paragraphs [1 ]-[3], wherein the host cell is capable of producing a greater amount of n-propanol when consisting only of the heterologous polynucleotide encoding the methylglyoxal synthase, compared to the cell without the heterologous polynucleotide encoding the methylglyoxal synthase.

[5] The recombinant host cell of any one of paragraphs [1 ]-[4], wherein the methylglyoxal synthase has at least 60%, e.g., at least 65%, at least 70%, at least 75%, at least 80% at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ I D NOs: 1 12, 1 13, 1 14, 1 15, 1 16, 1 17, 1 18, 1 19, 120, 121 , 122, 123, 124, or 125.

[6] The recombinant host cell of any one of paragraphs [1 ]-[4], wherein the methylglyoxal synthase comprises or consists of the amino acid sequence of any one of SEQ ID NOs: 1 12, 1 13, 1 14, 1 15, 1 16, 1 17, 1 18, 1 19, 120, 121 , 122, 123, 124, or 125. [7] The recombinant host cell of any one of paragraphs [1 ]-[4], wherein the methylglyoxal synthase comprises or consists of the amino acid sequence of SEQ ID NO: 1 13 or 1 16.

[8] The recombinant host cell of any one of paragraphs [1 ]-[7], wherein the heterologous polynucleotide encoding the methylglyoxal synthase is operably linked to a foreign promoter.

[9] The recombinant host cell of any one of paragraphs [1 ]-[8], wherein the heterologous polynucleotide encoding a methylglyoxal synthase is chromosomally integrated. [10] The recombinant host cell of any one of paragraphs [1 ]-[9], wherein the cell comprises a heterologous polynucleotide encoding a triosephosphate isomerase.

[1 1 ] The recombinant host cell of paragraph [10], wherein the host cell is capable of producing a greater amount of n-propanol compared to the cell without the heterologous polynucleotide encoding the triosephosphate isomerase, when cultivated under identical conditions.

[12] The recombinant host cell of paragraph [10] or [1 1 ], wherein the host cell is capable of producing at least 10% more (e.g., at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 75%, at least 100%, or at least 200% more) n-propanol compared to the cell without the heterologous polynucleotide encoding the triosephosphate isomerase, when cultivated under identical conditions.

[13] The recombinant host cell of any one of paragraphs [10]-[12], wherein the triosephosphate isomerase has at least 60%, e.g., at least 65%, at least 70%, at least 75%, at least 80% at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ I D NOs: 167, 169, 170, 171 , 172, 173, 174, 175, 176, 177, 178, or 201 .

[14] The recombinant host cell of any one of paragraphs [10]-[12], wherein the triosephosphate isomerase comprises or consists of the amino acid sequence of any one of SEQ ID NOs: 167, 169, 170, 171 , 172, 173, 174, 175, 176, 177, 178, or 201 . [15] The recombinant host cell of any one of paragraphs [10]-[14], wherein the heterologous polynucleotide encoding the triosephosphate isomerase is operably linked to a foreign promoter.

[16] The recombinant host cell of any one of paragraphs [10]-[15], wherein the heterologous polynucleotide encoding the triosephosphate isomerase is chromosomally integrated.

[17] The recombinant host cell of any one of paragraphs [1 ]-[16], wherein the cell comprises a disruption to an endogenous adhE gene.

[18] The recombinant host cell of paragraph [17], wherein the cell comprises a disruption to an endogenous adhE gene that encodes for a polypeptide having an amino acid sequence comprising or consisting of SEQ ID NO: 132. [19] The recombinant host cell of paragraph [17] or [18], wherein the coding sequence of the adhE gene comprises or consists of SEQ ID NO: 131 .

[20] The recombinant host cell of any one of paragraphs [17]-[19], wherein the disruption occurs in the coding sequence of the adhE gene.

[21 ] The recombinant host cell of any one of paragraphs [17]-[19], wherein the disruption occurs in a promoter sequence of the adhE gene.

[22] The recombinant host cell of any one of paragraphs [17]-[21 ], wherein the endogenous adhE gene is inactivated. [23] The recombinant host cell of any one of paragraphs [17]-[22], wherein the cell produces a greater amount of n-propanol compared to the cell without the disruption to an endogenous adhE gene when cultivated under identical conditions. [24] The recombinant host cell of any one of paragraphs [17]-[23], wherein the cell is capable of producing at least 10% more (e.g., at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 75%, at least 100%, or at least 200% more) n-propanol compared to the cell without the disruption to an endogenous adhE gene, when cultivated under identical conditions.

[25] The recombinant host cell of any one of paragraphs [1 ]-[24], wherein the cell comprises a disruption to an endogenous pduP gene.

[26] The recombinant host cell of paragraph [25], wherein the cell comprises a disruption to an endogenous pduP gene that encodes for a polypeptide having an amino acid sequence comprising or consisting of SEQ ID NO: 134.

[27] The recombinant host cell of paragraph [25] or [26], wherein the coding sequence of the pduP gene comprises or consists of SEQ ID NO: 133.

[28] The recombinant host cell of any one of paragraphs [25]-[27], wherein the disruption occurs in the coding sequence of the pduP gene.

[29] The recombinant host cell of any one of paragraphs [25]-[27], wherein the disruption occurs in a promoter sequence of the pduP gene.

[30] The recombinant host cell of any one of paragraphs [25]-[29], wherein the endogenous pduP gene is inactivated. [31 ] The recombinant host cell of any one of paragraphs [25]-[30], wherein the cell produces a greater amount of n-propanol compared to the cell without the disruption to an endogenous pduP gene when cultivated under identical conditions.

[32] The recombinant host cell of any one of paragraphs [25]-[31 ], wherein the cell is capable of producing at least 10% more (e.g., at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 75%, at least 100%, or at least 200% more) n-propanol compared to the cell without the disruption to an endogenous pduP gene, when cultivated under identical conditions.

[33] The recombinant host cell of any one of paragraphs [1 ]-[32], comprising an active isopropanol pathway.

[34] The recombinant host cell of paragraph [1 ]-[33], wherein the cell comprises one or more heterologous polynucleotides selected from:

a heterologous polynucleotide encoding a thiolase;

a heterologous polynucleotide encoding a CoA-transferase;

a heterologous polynucleotide encoding an HMG-CoA synthase;

a heterologous polynucleotide encoding an HMG-CoA lyase;

a heterologous polynucleotide encoding an acetoacetate decarboxylase; and a heterologous polynucleotide encoding an isopropanol dehydrogenase.

[35] A composition comprising the recombinant host cell of any one of paragraphs [1 ]-[34].

[36] The composition of paragraph [35], wherein the composition comprises a fermentable medium.

[37] The composition of paragraph [36], wherein the fermentable medium comprises glucose and fructose.

[38] The composition of paragraph [36], wherein the fermentable medium comprises sugarcane juice (e.g., non-sterilized sugarcane juice).

[39] The composition of any of paragraphs [35]-[38], further comprising n-propanol and/or isopropanol. [40] The composition of paragraph [39], wherein the n-propanol and/or isopropanol is at a titer greater than about 0.01 g/L, e.g., greater than about 0.02 g/L, 0.05 g/L, 0.075 g/L, 0.1 g/L, 0.5 g/L, 1 g/L, 2 g/L, 5 g/L, 10 g/L, 15 g/L, 20 g/L, 25 g/L, 30 g/L, 35 g/L, 40 g/L, 45 g/L, 50 g/L, 55 g/L, 60 g/L, 65 g/L, 70 g/L, 75 g/L, 80 g/L, 85 g/L, 90 g/L, 95 g/L, 100 g/L, 125 g/L, 150 g/L, 200 g/L, or 250 g/L.

[41 ] A method of producing n-propanol, comprising:

(a) cultivating the recombinant Lactobacillus reuteri host cell of any one of paragraphs [1 ]-[34] in a medium under suitable conditions to produce n-propanol; and (b) recovering the n-propanol.

[42] The method of paragraph [41 ], wherein the medium is a fermentable medium.

[43] The method of paragraph [42], wherein the fermentable medium comprises glucose and fructose.

[44] The method of paragraph [42], wherein the fermentable medium comprises sugarcane juice (e.g., non-sterilized sugarcane juice).

[45] The method of any of paragraphs [41 ]-[44], wherein the produced n-propanol is at a titer greater than about 0.01 g/L, e.g., greater than about 0.02 g/L, 0.05 g/L, 0.075 g/L, 0.1 g/L, 0.5 g/L, 1 g/L, 2 g/L, 5 g/L, 10 g/L, 15 g/L, 20 g/L, 25 g/L, 30 g/L, 35 g/L, 40 g/L, 45 g/L, 50 g/L, 55 g/L, 60 g/L, 65 g/L, 70 g/L, 75 g/L, 80 g/L, 85 g/L, 90 g/L, 95 g/L, 100 g/L, 125 g/L, 150 g/L, 200 g/L, or 250 g/L.

[46] The method of any of paragraphs [41 ]-[45], further comprising purifying the recovered n-propanol by distillation.

[47] The method of any of paragraphs [41 ]-[46], further comprising purifying the recovered n-propanol by converting propionaldehyde contaminant to n-propanol in the presence of a reducing agent. [48] The method of any of paragraphs [41 ]-[47], wherein the resulting n-propanol is substantially pure.

[49] A method of producing propylene, comprising:

(a) cultivating the recombinant host cell of any of paragraphs [1 ]-[34] in a medium under suitable conditions to produce n-propanol;

(b) recovering the n-propanol;

(c) dehydrating the n-propanol under suitable conditions to produce propylene; and

(d) recovering the propylene.

[50] The method of paragraph [49], wherein the medium is a fermentable medium.

[51 ] The method of paragraph [50], wherein the fermentable medium comprises glucose and fructose.

[52] The method of paragraph [50], wherein the fermentable medium comprises sugarcane juice (e.g., non-sterilized sugarcane juice).

[53] The method of any of paragraphs [49]-[52], wherein the produced n-propanol is at a titer greater than about 0.01 g/L, e.g., greater than about 0.02 g/L, 0.05 g/L, 0.075 g/L, 0.1 g/L, 0.5 g/L, 1 g/L, 2 g/L, 5 g/L, 10 g/L, 15 g/L, 20 g/L, 25 g/L, 30 g/L, 35 g/L, 40 g/L, 45 g/L, 50 g/L, 55 g/L, 60 g/L, 65 g/L, 70 g/L, 75 g/L, 80 g/L, 85 g/L, 90 g/L, 95 g/L, 100 g/L, 125 g/L, 150 g/L, 200 g/L, or 250 g/L.

[54] The method of any one of paragraphs [49]-[53], wherein dehydrating the n-propanol comprises treating the n-propanol with an acid catalyst. [55] A method of producing n-propanol and isopropanol, comprising:

(a) cultivating the recombinant Lactobacillus reuteri host cell of paragraph [33] or [34] in a medium under suitable conditions to produce n-propanol and isopropanol; and

(b) recovering the n-propanol and isopropanol. [56] The method of paragraph [55], wherein the medium is a fermentable medium.

[57] The method of paragraph [56], wherein the fermentable medium comprises glucose and fructose. [58] The method of paragraph [56], wherein the fermentable medium comprises sugarcane juice (e.g., non-sterilized sugarcane juice).

[59] The method of any of paragraphs [55]-[58], wherein the produced n-propanol and/or isopropanol is at a titer greater than about 0.01 g/L, e.g., greater than about 0.02 g/L, 0.05 g/L, 0.075 g/L, 0.1 g/L, 0.5 g/L, 1 g/L, 2 g/L, 5 g/L, 10 g/L, 15 g/L, 20 g/L, 25 g/L, 30 g/L, 35 g/L, 40 g/L, 45 g/L, 50 g/L, 55 g/L, 60 g/L, 65 g/L, 70 g/L, 75 g/L, 80 g/L, 85 g/L, 90 g/L, 95 g/L, 100 g/L, 125 g/L, 150 g/L, 200 g/L, or 250 g/L.

[60] The method of any of paragraphs [55]-[59], further comprising purifying the recovered n-propanol and isopropanol by distillation.

[61 ] The method of any of paragraphs [55]-[60], further comprising purifying the recovered n-propanol by converting propionaldehyde contaminant to n-propanol and/or converting acetone contaminant to isopropanol in the presence of a reducing agent.

[62] The method of any of paragraphs [55]-[61 ], wherein the resulting n-propanol and isopropanol is substantially pure.

5

[63] A method of producing propylene, comprising:

(a) cultivating the recombinant host cell of paragraph [33] or [34] in a medium under suitable conditions to produce n-propanol and isopropanol;

(b) recovering the n-propanol and isopropanol;

i o (c) dehydrating the n-propanol and isopropanol under suitable conditions to produce propylene; and

(d) recovering the propylene.

[64] The method of paragraph [63], wherein the medium is a fermentable medium.

15

[65] The method of paragraph [64], wherein the fermentable medium comprises glucose and fructose.

[66] The method of paragraph [64], wherein the fermentable medium comprises sugarcane 20 juice (e.g., non-sterilized sugarcane juice).

[67] The method of any of paragraphs [63]-[66], wherein the produced n-propanol and/or isopropanol is at a titer greater than about 0.01 g/L, e.g., greater than about 0.02 g/L, 0.05 g/L, 0.075 g/L, 0.1 g/L, 0.5 g/L, 1 g/L, 2 g/L, 5 g/L, 10 g/L, 15 g/L, 20 g/L, 25 g/L, 30 g/L, 35 25 g/L, 40 g/L, 45 g/L, 50 g/L, 55 g/L, 60 g/L, 65 g/L, 70 g/L, 75 g/L, 80 g/L, 85 g/L, 90 g/L, 95 g/L, 100 g/L, 125 g/L, 150 g/L, 200 g/L, or 250 g/L.

[68] The method of any one of paragraphs [63]-[67], wherein dehydrating the n-propanol and isopropanol comprises treating the n-propanol and isopropanol with an acid catalyst.

30

Claims

Claims What is claimed is:

1 . A recombinant Lactobacillus reuteri host cell comprising a heterologous polynucleotide encoding a methylglyoxal synthase, wherein the host cell produces a greater amount of n-propanol compared to the cell without the heterologous polynucleotide encoding the methylglyoxal synthase, when cultivated under identical conditions.

2. The recombinant host cell of claim 1 , wherein the host cell produces at least 10% more (e.g., at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least

40%, at least 45%, at least 50%, at least 75%, at least 100%, or at least 200% more) n- propanol compared to the cell without the heterologous polynucleotide encoding the methylglyoxal synthase, when cultivated under identical conditions.

3. The recombinant host cell of claim 1 or 2, wherein the methylglyoxal synthase has at least 60%, e.g., at least 65%, at least 70%, at least 75%, at least 80% at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ ID NOs: 1 12, 1 13, 1 14, 1 15, 1 16, 1 17, 1 18, 1 19, 120, 121 , 122, 123, 124, or 125.

4. The recombinant host cell of any one of claims 1 -3, wherein the methylglyoxal synthase comprises or consists of the amino acid sequence of any one of SEQ ID NOs: 1 12, 1 13, 1 14, 1 15, 1 16, 1 17, 1 18, 1 19, 120, 121 , 122, 123, 124, or 125.

5. The recombinant host cell of any one of claims 1 -4, wherein the heterologous polynucleotide encoding the methylglyoxal synthase is operably linked to a foreign promoter.

6. The recombinant host cell of any one of claims 1 -5, wherein the heterologous polynucleotide encoding a methylglyoxal synthase is chromosomally integrated.

7. The recombinant host cell of any one of claims 1 -6, wherein the cell comprises a heterologous polynucleotide encoding a triosephosphate isomerase.

8. The recombinant host cell of claim 7, wherein the triosephosphate isomerase has at least 60%, e.g., at least 65%, at least 70%, at least 75%, at least 80% at least 85%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity to any one of SEQ ID NOs: 167, 169, 170, 171 , 172, 173, 174, 175, 176, 177, or 178.

9. The recombinant host cell of claim 7 or 8, wherein the triosephosphate isomerase comprises or consists of the amino acid sequence of any one of SEQ ID NOs: 167, 169, 170, 171 , 172, 173, 174, 175, 176, 177, or 178.

10. The recombinant host cell of any one of claims 1 -9, wherein the cell comprises a disruption to an endogenous adhE gene.

1 1 . The recombinant host cell of claim 10, wherein the cell comprises a disruption to an endogenous adhE gene that encodes for a polypeptide having an amino acid sequence comprising or consisting of SEQ ID NO: 132.

12. The recombinant host cell of claim 10 or 1 1 , wherein the endogenous adhE gene is inactivated.

13. The recombinant host cell of any one of claims 1 -12, wherein the cell comprises a disruption to an endogenous pduP gene.

14. The recombinant host cell of claim 13, wherein the cell comprises a disruption to an endogenous pduP gene that encodes for a polypeptide having an amino acid sequence comprising or consisting of SEQ ID NO: 134.

15. The recombinant host cell of claim 13 or 14, wherein the endogenous pduP gene is inactivated.

16. The recombinant host cell of any one of claims 1 -15, comprising an active isopropanol pathway.

17. The recombinant host cell of claim 1 -16, wherein the cell comprises one or more heterologous polynucleotides selected from:

a heterologous polynucleotide encoding a thiolase;

a heterologous polynucleotide encoding a CoA-transferase; a heterologous polynucleotide encoding an HMG-CoA synthase;

a heterologous polynucleotide encoding an HMG-CoA lyase;

a heterologous polynucleotide encoding an acetoacetate decarboxylase; and a heterologous polynucleotide encoding an isopropanol dehydrogenase.

18. A method of producing n-propanol, comprising:

(a) cultivating the recombinant Lactobacillus reuteri host cell of any one of claims 1 -17 in a medium under suitable conditions to produce n-propanol; and

(b) recovering the n-propanol.

19. A method of producing propylene, comprising:

(a) cultivating the recombinant host cell of any of claims 1 -17 in a medium under suitable conditions to produce n-propanol;

(b) recovering the n-propanol;

(c) dehydrating the n-propanol under suitable conditions to produce propylene; and

(d) recovering the propylene.

20. A method of producing n-propanol and isopropanol, comprising:

(a) cultivating the recombinant Lactobacillus reuteri host cell of claim 16 or 17 in a medium under suitable conditions to produce n-propanol and isopropanol; and

(b) recovering the n-propanol and isopropanol.

21 . A method of producing propylene, comprising:

(a) cultivating the recombinant host cell of claim 16 or 17 in a medium under suitable conditions to produce n-propanol and isopropanol;

(b) recovering the n-propanol and isopropanol;

(c) dehydrating the n-propanol and isopropanol under suitable conditions to produce propylene; and

(d) recovering the propylene.