IMPROVED PROPIONIBACTERIUM STRAINS FOR THE PRODUCTION OF

PROPIONIC ACID

CROSS REFERENCE TO RELATED APPLICATIONS

This application claims priority to U. S. Provisional Patent Application No. 62/234,900, filed on September 30, 2015, which is incorporated by reference in its entirety.

INCORPORATION BY REFERENCE OF MATERIAL SUBMITTED ELECTRONICALLY

Incorporated by reference in its entirety is a computer-readable nucleotide/amino acid sequence listing submitted concurrently herewith and identified as follows: One 76 kilobyte ASCn (Text) file named "258544_SeqListing.txt," created on September 26, 2016.

BACKGROUND

Propionic acid (PA) is widely used in the food industry as a food preservative. PA has also found use as a precursor for the synthesis of polymers, including but not limited to polypropylene and vinyl propionate. In addition propanol and other valuable chemicals can be derived from PA. PA has traditionally been derived from fossil fuels until recently when mounting environmental concerns have shifted end users' interest to a sustainable alternative. This search for sustainable alternatives has revived bacterial fermentation as an alternative for the production of C3 chemicals.

Propionibacterium sp are pleomorphic rods, gram-positive bacteria that naturally produce PA as their main fermentation product through the Wood-Werckman cycle. Natively, PA is produced along with other organic acids (lactate, succinate, and acetate) resulting in low productivities and modest yields which translate in costly downstream processes. Until recently, metabolic engineering in P. acidipropionici had proven to be challenging mainly due to the seven Clustered Regularly Interspaced Short Palindromic Repeats (CRISPRs) which provide resistance against conjugative plasmids and bacteriophages. As disclosed herein genome shuffling (GS) can be used to improve the growth rate and PA production of P.

acidipropionici.

Originally described in the mid 70' s as protoplast fusion, GS has been extensively used in industry. Examples include increased production of tylosin in Streptomyces fradiae (Zhang et al., Nature, 2002, 415, 644-646), ethanol in Saccharomyces cerevisiae (L. Hou, Appl.

Biochem. Biotechnol., 2010, 160, 1084-1093), vitamin B 12 in P. shermanii (Zhang et al., J. Biotechnol., 2010, 148, 139-143), lactic acid in Lactobacillus (Patnaik et al., Nat. Biotechnol., 2002, 20, 707-12), 1,3 - propanediol in Clostridium diolis (Otte et al., Appl. Environ.

Microbiol., 2009, 75, 7610-7616), and PA in P. acidipropionici (Guan et al., "Genome- Shuffling Improves Acid Tolerance..." in Advances in Chemistry Research, Vol 15, Chapter 8 (2012) Nova Science Publishers, Inc., pp 143- 152) amongst others.

To obtain Propionibacterium strains with higher propionic acid yields and lower byproducts, a genome shuffling (GS) protocol has been used to transfer genetic material between two strains of Propionibacterium resulting in novel strains with the potential for improved propionic acid production. GS combines the advantages of multi-parental crossing facilitated by DNA exchange, where the donor provides a small amount of DNA material leaving the rest of the recipient intact. The end result of GS is genetically unique strains with potentially novel pathways and regulatory mechanisms.

Previous attempts to use GS for enhancing PA production have not delivered strains with the desired phenotypes. This is due in part to the lack of genomic diversity in the known strains used in such procedures. Thus, previous efforts have failed to produce

Propionibacterium strains that are capable of producing propionic acid in excess of 0.54 g/g using glucose or sucrose fermentations, that also retain multiple byproducts in appreciable quantities (Stowers et al., J. Ind. Microbiol. Biotechnol., 2014, 41, 837-852).

Accordingly, the success of using GS for enhancing PA production is predicated on the ability to select two strains with desirable phenotypes that if combined could result in an advantaged propionic acid production strain. Applicants have selected strains of

Propionibacterium that they have identified as having high potential for propionic acid production. These selected strains have now been used to produce novel strains of

Propionibacterium that have improved growth rates, enhanced propionic acid production (e.g., exceeding 0.54 g/g) and optionally, a reduced production of undesired byproducts such as acetic acid and succinic acid.

SUMMARY

In accordance with one embodiment compositions and methods are provided for the biosynthetic production of propionic acid. In accordance with one embodiment, a new strain of Propionibacterium is provided that has an improved ability to utilize glucose or sucrose as a carbon and energy source for enhanced yields of propionic acid relative to native Propionibacterium and other known derivative strains. In one embodiment a new

Propionibacterium strain is provided having enhanced propionic acid production. In one embodiment the new strain is produced by the transfer of genetic material between a first and second strain of Propionibacterium, optionally by genome shuffling, wherein the first and second strains have different genomes. The strains generated after genomic shuffling are subsequently screened for strains that have an improved ability to utilize glucose/sucrose as a carbon and energy source for enhanced yields of propionic acid relative to the parent strains. In one embodiment the Propionibacterium strains used for genomic shuffling are selected from strains: P. acidipropionici ATCC 55737, P. acidipropionici ATCC 4875, P. acidipropionici ATCC 4965, P. intermedium ATCC 14072, and P. jensenii ATCC 9617. In one embodiment the new Propionibacterium strain having enhanced propionic acid production is a P.

acidipropionici strain.

In accordance with one embodiment a P. acidipropionici strain is produced by exchanging genetic material between P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737, optionally through genomic shuffling, wherein the growth rate of the resulting strains and the production of propionic acid is enhanced in the new strain relative to parental strains P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737 and other known P. acidipropionici strains. In another embodiment a P. acidipropionici strain is produced by exchanging genetic material between P. acidipropionici ATCC 4875 and P. acidipropionici strain F3E8, deposited with the American Type Culture Collection (ATCC), on June 25, 2015 under ATCC Accession No. PTA- 122267, wherein the new generated strain has an increase in total yield (g/g) of propionic acid relative to that produced by P. acidipropionici ATCC 55737 when the two strains are cultured under identical conditions.

In accordance with one embodiment a method of producing a new strain of

Propionibacterium is provided. The method comprises the steps of producing a library of Propionibacterium strains by conducting separate genomic shuffling reactions between various combinations of two different known strains; and combining the resulting strains generated from each reaction to form a library of Propionibacterium strains having enhanced genomic diversity relative to the parental strains.

In one embodiment a library of Propionibacterium strains is created using the strains: P. acidipropionici ATCC 4875, P. acidipropionici ATCC 4965, P. intermedium ATCC 14072, and P. jensenii ATCC 9617. In one embodiment, to obtain a library of diverse strains, P. acidipropionici ATCC 4875 was used to perform genomic shuffling with the other three Propionibacterium strains in three separate genomic shuffling reactions. In one embodiment 1, 2 or 3 rounds of genome shuffling are performed with each set of strains. Strains generated from each of the three sets of genomic shuffling reactions are combined to produce a library of Propionibacterium strains.

This library of strains can then be use to conduct a second round of genomic shuffling between the library of strains and selected Propionibacterium strains, including for example, P. acidipropionici ATCC 4875 or P. acidipropionici strain F3E8, deposited with the American Type Culture Collection (ATCC), on June 25, 2015 under ATCC Accession No. PTA-122267, to produce a second set of GS generated strains. The amount of propionic acid produced in the second set of GS generated strains is then measured to identify those new strains having improved yields of propionic acid relative to P. acidipropionici ATCC 4875 or P.

acidipropionici ATCC 55737. In one embodiment the second set of GS generated strains is cultured prior to the step of measuring propionic acid production to identify those strains that have enhanced growth rates relative to P. acidipropionici ATCC 4875 or P. acidipropionici ATCC 55737. Members of the second set of GS generated strains having enhanced growth rate relative to the parental strains are selected and only the selected strains are analyzed to identify those new strains having improved yields of propionic acid relative to P. acidipropionici ATCC 4875 or P. acidipropionici ATCC 55737.

In accordance with one embodiment, an isolated Propionibacterium strain is provided wherein said strain has a maximum growth rate of greater than 0.18/hr, and optionally at least 0.24/hr as calculated for the exponential phase of cells cultured at 32 °C, pH 6.5 in PAM media. In one embodiment a new strain of Propionibacterium is provided having a maximum growth rate at least 1.5 times that of P. acidipropionici ATCC strain 4875 when the two strains are grown under identical conditions, including for example anaerobically at 32 °C, pH 6.5 in media suitable for growth, including for example PAM media. In one embodiment a P.

acidipropionici strain is provided having a growth rate of about 0.24/hr to about 0.26/hr, as calculated for the exponential phase of cells, including for example cells cultured at 32 °C, pH 6.5 in PAM media.

The novel Propionibacterium strains of the present disclosure also have improved yields of propionic acid relative to P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737. In one embodiment a novel Propionibacterium strain of the present disclosure has an increase of at least 10%, and optionally a 10 to 15%, a 15 to 20%, or a 20 to 25% or a 30 to 40% increase in total yield (g/g) of propionic acid relative to that produced by P. acidipropionici ATCC strain 4875 or ATCC strain 55737 when the two strains are cultured under identical conditions (e.g., anaerobically at 32 °C, pH 6.5 in PAM media). In one embodiment the novel Propionibacterium strain produces a propionic acid yield of about 0.49 to about 0.8 g/g, about 0.5 to about 0.7 g/g, or about 0.5 to about 0.66, or about 0.66 g/g. In one embodiment the novel P. acidipropionici strain produces 0.77 to about 0.92 g/L/hr or about 0.84 g/L/hr or about 0.95 g/L/hr of propionic acid under optimal culture conditions, including for example cultured anaerobically at 32 °C, pH 6.5 in PAM media. In one embodiment the novel P. acidipropionici strain produces about 0.8 to about 1.1 g/L/hr or about 0.95 g/L/hr of propionic acid under optimal culture conditions, including for example cultured anaerobically at 32 °C, pH 6.5 in PAM media.

In one embodiment an isolated Propionibacterium strain is provided wherein said strain produces a higher yield of propionic acid than P. acidipropionici strain ATCC 55737. More particularly, the isolated strain is further characterized as comprising a modified gene, relative to P. acidipropionici strain ATCC 55737, selected from the group consisting of the ABC polar amino acid transporter gene, the Cytochrome C biogenesis gene or the ABC multiple sugar transporter gene. In one embodiment the isolated Propionibacterium strain encodes an altered gene product for the ABC polar amino acid transporter protein and/or the Cytochrome C biogenesis protein. In one embodiment the isolated strain comprises modified genes for each of the ABC polar amino acid transporter gene, the Cytochrome C biogenesis gene and the ABC multiple sugar transporter gene.

In one embodiment an isolated or purified Propionibacterium strain is provided wherein the strain has improved growth rates (e.g., greater than about 0.24/hr) and/or improved propionic acid yields (e.g., about 0.55 g/g or greater) and comprises

(i) a modified large ribosomal RNA gene (SEQ ID NO 15) comprising a nucleotide substitution at position 1441 relative to the same gene in P. acidipropionici ATCC 55737, and positions 1425, 1270, and 1271 relative to the same gene in P. acidipropionici ATCC 4875;and/or

(ii) a modified long chain acyl-CoA synthetase gene (SEQ ID NO: 16) comprising a deletion at position 2269 and nonsense codon after position 1500 relative to the same gene in P. acidipropionici ATCC 55737, and positions 264, 1650, and 2295 relative to the same gene in P. acidipropionici ATCC 4875. The amino acid changes in the encoding protein are at positions 500 and 757 referencing P. acidipropionici ATCC 55737 and positions 88 and 757 relative to the same protein in P. acidipropionici ATCC 4875 (see SEQ ID NO: 16; the encoded truncated peptide is provided as SEQ ID NO: 17); and/or

(iii) a modified cation diffusion facilitator gene (SEQ ID NO: 13) comprising a nucleotide substitution at position 953 relative to the same gene in P. acidipropionici ATCC 55737 and positions 18, 19, 56, 211, 334, 340, 406, 451, 484, 574, and 953 relative to the same gene in P. acidipropionici ATCC 4875. The amino acid changes of the encoding protein (SEQ ID NO: 14) comprises an amino acid substitution at position 318 relative to the same proteins in P. acidipropionici ATCC 55737 and positions 7, 60, 71, 86, 112, 114, 136, 151, 162, 192, and 318 relative to the same protein in P. acidipropionici ATCC 4875; and/or

(iv) an extra copy of the whole ribosomal RNA gene (SEQ ID NO: 18) relative to both parental strains (P. acidipropionici ATCC 55737 and P. acidipropionici ATCC 4875); and/or (v) an extra copy of the arginine deiminase regulon (SEQ ID NO: 21) relative to both parental strains (P. acidipropionici ATCC 55737 and P. acidipropionici ATCC 4875). In one embodiment the extra copy comprises a mutation at position 37 relative to both parental strains (P. acidipropionici ATCC 55737 and P. acidipropionici ATCC 4875).

In accordance with one embodiment the novel strain, P. acidipropionici F3E8 is provided, as deposited with the American Type Culture Collection (ATCC), 10801 University Boulevard, Manassas, VA, 20110-2209 USA, on June 25, 2015 under ATCC Accession No. PTA-122267. The present disclosure also encompasses variants of said deposited strain, wherein the variant has been subjected to further genetic manipulations. Such variants include derivatives of strain F3E8 generated by subjecting the F3E8 strain to genomic shuffling with different strains of P. acidipropionici. In accordance with one embodiment the novel strain, P. acidipropionici WGS 7 is provided, as deposited with the American Type Culture Collection (ATCC), 10801 University Boulevard, Manassas, VA, 20110-2209 USA, on September 2, 2016 under ATCC Accession No. . Strain, P. acidipropionici F3E8 was generated using glucose as the carbon and energy source whereas WGS 7 was generated using sucrose as the carbon and energy source.

In accordance with one embodiment, a method of producing a Propionibacterium strain having improved yields of propionic acid production is provided wherein the method comprises subjecting a strain selected from the group consisting of P. acidipropionici F3E8, P.

acidipropionici WGS 7, P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737 to genomic shuffling with a different strain selected from the group consisting of P.

acidipropionici F3E8, P. acidipropionici WGS 7, P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737, and recovering the resulting modified strains that have enhanced growth rates or improved propionic acid yields relative to the parental strains. In one embodiment a method of producing a novel P. acidipropionici strain having improved yields of propionic acid production is provided wherein the method comprises subjecting a first and second strain of P. acidipropionici to genome shuffling. More particularly, in one embodiment the first strain is selected from the group consisting of P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737 and the second strain is P. acidipropionici strain F3E8, deposited with the American Type Culture Collection (ATCC), on June 25, 2015 under ATCC Accession No. PTA-122267. The new strains produced by subjecting the first and second strains of P. acidipropionici to genome shuffling are then cultured to identify strains that have enhanced growth rates or propionic acid production relative to P. acidipropionici ATCC 4875 or P.

acidipropionici ATCC 55737. In one embodiment the method comprises the steps of first selecting the resulting new strains having enhance growth rate relative to P. acidipropionici ATCC 4875 or P. acidipropionici ATCC 55737; and then measuring propionic acid production in the selected strains to identify those new strains having improved yields of propionic acid relative to P. acidipropionici ATCC 4875 or P. acidipropionici ATCC 55737.

In one embodiment the modified strains have one, two, three, four, or five modifications in a cellular component and/or the gene encoding the cell component, wherein the cell component in one embodiment is selected from the ABC polar amino acid transporter, the Cytochrome C biogenesis, the ABC multiple sugar transporter, the large subunit of ribosomal RNA, the long chain acyl-CoA synthetase, the cation diffusion facilitator, the whole ribosomal RNA operon, and the arginine deminnase regulon. In one embodiment the modified strains have one, two, or three modifications in a cellular component and/or the gene encoding the cell component, wherein the cell component in one embodiment is selected from the large subunit of ribosomal RNA, the long chain acyl-CoA synthetase, and the cation diffusion facilitator.

In one embodiment a method is provided for producing propionic acid. In one embodiment the method comprises the steps of culturing a P. acidipropionici strain under conditions suitable for growth of the strain and recovering the propionic acid produced by said strain. In one embodiment the P. acidipropionici strain has a maximum growth rate of at least 0.24/hr calculated for the exponential phase of cells cultured at 32 °C, pH 6.5 in PAM media. In a further embodiment the P. acidipropionici strain produces a propionic acid yield of about 0.49 to about 0.8 g/g, about 7.0 to about 8.0 g/g, about 0.5 to about 0.7 g/g, or about 0.5 to about 0.66, or about 0.66 g/g. In one embodiment the P. acidipropionici strain used to produce propionic acid is strain F3E8, deposited with the American Type Culture Collection (ATCC), 10801 University Boulevard, Manassas, VA, 20110-2209 USA, on June 25, 2015, and assigned ATCC Accession No. PTA- 122267. The present disclosure also encompasses novel P.

acidipropionici strains produced by the transfer of genetic material between P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737, including strains produced during genome shuffling of P. acidipropionici strain F3E8 with either P. acidipropionici ATCC 4875 or P. acidipropionici ATCC 55737 and/or genome shuffling of P. acidipropionici strain F3E8 and a different P. acidipropionici strain. In one embodiment the P. acidipropionici strain used to produce propionic acid is strain WGS 7, as deposited with the American Type Culture

Collection (ATCC), 10801 University Boulevard, Manassas, VA, 20110-2209 USA, on September 2, 2016 under ATCC Accession No. .

BRIEF DESCRIPTION OF THE DRAWINGS

Figs. 1 A -1C are graphs demonstrating the fermentation profile in 2 L bioreactors for P. acidipropionici ATCC 4875 (Fig. 1 A), P. acidipropionici ATCC 55737 (Fig. IB), and P. acidipropionici F3E8 strain (Fig. 1C). Plots display the average of two biological replicates.

Optical Density measured at 600 nm: Black line and♦; Propionic Acid: Black line and■;

Glucose: Black line and ·; Succinic Acid: Grey line and□; Acetic Acid: Grey line and 0; Pyruvate: Grey line and o.

Fig. 2 is a graph demonstrating the fermentation growth profile of P. acidipropionici

F3E8 in serum bottles (three biological replicates).

Fig. 3 A & 3B are graphs demonstrating the kinetics of PA inhibition at different pHs

(o: pH 6.5. D: pH 5.5. Δ: ρΗ 4.5). Fig. 3A is a graph of the inhibition of P. acidipropionici ATCC 4875, whereas Fig. 3B provides data for the inhibition of P. acidipropionici F3E8.

Figs. 4A-4D show fermentation profiles in 2 L bioreactors and their external metabolite specific rates. Fig. 4A: P. acidipropionici ATCC 55737, Fig. 4B: P. acidipropionici WGS 7 strain obtained from GS. Optical Density: solid diamonds; Propionic Acid: solid squares;

Glucose: solid circles; Succinic Acid: open squares; Acetic Acid: open diamonds; Pyruvate: open circles. Fig. 4C is a bar graph showing the specific consumption rate of sucrose (qs), the specific production rate of PA (qp), and the specific growth rate (μ). Fig. 4D is a bar graph demonstrating the specific consumption rates of the free amino acids presented in the PAM media, shaded bars: P. acidipropionici ATCC 55737. Open bars: P. acidipropionici WGS 7. The data represent the average of two biological replicates for each strain. The specific rates were calculated in the middle exponential phase (15 to 25 hours).

Fig. 5 shows the effects of exogenous addition of 50 mM of lactate (LAC), 10 mM of fumarate (FUM), 10 mM of arginine (Arg), 10 mM of lysine (Lys) or 10 mM of proline (Pro) in P. acidipropionici ATCC 55737 (shaded bars) and P. acidipropionici WGS 7 (open bars).

Fermentations were performed by duplicate serum bottle fermentations containing CDM media.

DETAILED DESCRIPTION DEFINITIONS

In describing and claiming the invention, the following terminology will be used in accordance with the definitions set forth below.

The term "about" as used herein means greater or lesser than the value or range of values stated by 10 percent but is not intended to designate any value or range of values to only this broader definition. Each value or range of values preceded by the term "about" is also intended to encompass the embodiment of the stated absolute value or range of values.

As used herein, the term "purified" and like terms define the isolation of bacteria, or a compound in a form that is substantially free of contaminants normally associated with the bacteria, or compound in a native or natural environment. As used herein, the term "purified" does not require absolute purity; rather, it is intended as a relative definition. The term

"purified bacteria" is used herein to describe a bacterial population which has been separated from other bacteria and is present as a homogenous population.

The term "isolated" requires that the referenced material be removed from its original environment (e.g., the natural environment if it is naturally occurring).

As used herein an "altered gene product" defines an RNA or amino acid sequence encoded by a modified gene, wherein the RNA or amino acid sequence has a different primary sequence relative to the corresponding gene product produced by the wild type gene/reference gene.

As used herein a 'copy number variation' defines a different number of a specific gene relative to the number of the same gene in the wild type strains. The variation can be a duplication or deletion of a sequence encoding for RNA or an amino acid sequence. The extra copy can be an exact sequence or a modified one relative to the same sequence of the strain. As used herein an "amino acid modification" encompasses (i) a substitution of an amino acid with a different amino acid, (ii) an addition/insertion of an amino acid, or (iii) a deletion of one or more amino acids.

As used herein an amino acid "substitution" refers to the replacement of one amino acid residue by a different amino acid residue.

As used herein, the term "conservative amino acid substitution" is defined herein as exchanges within one of the following five groups:

I. Small aliphatic, nonpolar or slightly polar residues:

Ala, Ser, Thr, Pro, Gly;

II. Polar, negatively charged residues and their amides:

Asp, Asn, Glu, Gin;

III. Polar, positively charged residues:

His, Arg, Lys;

IV. Large, aliphatic, nonpolar residues:

Met, Leu, lie, Val, Cys

V. Large, aromatic residues:

Phe, Tyr, Trp.

A "nonconservative amino acid substitution" is defined as an exchange of a first amino acid with a second amino acid wherein the second amino acid is selected from one of the above five groups outside of the group of the first amino acid.

A "promoter" is a DNA sequence that directs the transcription of a polynucleotide. Typically a promoter is located in the 5' region of a polynucleotide to be transcribed, proximal to the transcriptional start site of such polynucleotide.

A "gene" as used herein describes nucleic acid molecules comprising an open reading frame encoding a gene product (i.e., an RNA transcript and/or polypeptide) as well as the associated regulatory regions controlling the expression of the gene product and non-coding transcribed regions (e.g., introns and 5' and 3' untranslated regions).

A "modified gene" as used herein is a gene that has been altered in some manner relative to the corresponding gene found in a reference strain, (e.g. P. acidipropionici ATCC strain 4875 or 55737). Alterations may include duplication or deletion of the entire gene, duplication or deletion of portions of the gene, or insertions or substitutions of nucleotides in the gene sequence. The term "identity" as used herein defines the similarity between two or more sequences. Identity is measured by dividing the number of identical residues by the total number of residues and multiplying the product by 100 to achieve a percentage. Thus, two copies of exactly the same sequence have 100% identity, whereas two sequences that have amino acid deletions, additions, or substitutions relative to one another have a lower degree of identity. Those skilled in the art will recognize that several computer programs, such as those that employ algorithms such as BLAST (Basic Local Alignment Search Tool, Altschul et al. (1993) J. Mol. Biol. 215:403-410) are available for determining sequence identity.

A "culture medium" refers to any liquid, semi-solid or solid media that can be used to support the growth of a microorganism used in the methods of the invention. In some embodiments, the microorganism is a bacterium, e.g., P. acidipropionici. Media for growing microorganisms are well known, see, e.g., Sambrook et al. and Current Protocols in Molecular Biology, F. M. Ausubel et al., eds., Current Protocols, a joint venture between Greene

Publishing Associates, Inc. and John Wiley & Sons, Inc., (1998 Supplement) (Ausubel).

As used herein the term "genome shuffling" or "GS" is a process of transferring genetic material between two bacterial strains to create a new recombined strain comprising elements from the two initial strains.

As used herein the term "growth rate" absent any further qualification, means the maximum specific growth rate ^max) calculated based on the change in cell density in exponential phase as measured by the change in optical density based on light absorption at 600 nm. Growth rate is expressed in the units of per time or 1/time using the formula N = Νο β ^{μ ί} wherein "N" is the cell density measured by the OD600, "No" is the original or previous cell density, "t" is time, "e" is a constant (Euler number, e = 2.71828), and μ is the growth rate.

As used herein the term "yield" defines the amount of product of interest produced divided by the amount of total substrate consumed (g/g).

As used herein the term "production rate of propionic acid" represents the total grams of propionic acid produced per liter per hour.

As used herein the term "titer" represents the total grams of product produced per liter of fermentation broth.

The term "fermentation" refers to a metabolic process performed by an organism that converts one substrate to another in which the cell is able to obtain cellular energy, such as when an organism utilizes glucose and converts it to propionic acid. EMBODIMENTS

In accordance with one embodiment compositions and methods are provided for the biosynthetic production of propionic acid. In accordance with one embodiment, a new strain of Propionibacterium is provided that has an improved ability to utilize glucose as a carbon and energy source for enhanced yields of propionic acid relative to native Propionibacterium and other known derivative strains. In accordance with one embodiment, the novel

Propionibacterium strain is a P. acidipropionici strain.

Novel strains having the requisite improved growth rates and/or propionic acid yields can be produced by conducting genomic shuffling between any suitable Propionibacterium strains. In one embodiment, strains with enhanced yields of propionic acid are produced by exchanging genetic material between P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737, optionally through genomic shuffling. The resulting strains can then be further manipulated by subjecting the resulting strains, including for example any of P. acidipropionici strains F3E8, F3G8, F3C8, F3H8, F3E9, F3F8, F3F6, F3D9, F3C1, F3B9, WGS.l, WGS.2, WGS.3, WGS.4, WGS.5, WGS.6, WGS 7, WGS.8, WGS.9, WGS.10, WGS.ll, WGS.12, and WGS.13 disclosed herein, to further genomic shuffling with different strains of P.

acidipropionici. In accordance with one embodiment, a new P. acidipropionici strain is produced by exchanging genetic material between P. acidipropionici strain F3E8 and a different P. acidipropionici strain. In one embodiment a new P. acidipropionici strain is produced by exchanging genetic material, optionally through genomic shuffling, from a first strain selected from the group consisting of P. acidipropionici ATCC 4875 and P.

acidipropionici ATCC 55737 and a second strain, wherein the second strain is selected from the group consisting of P. acidipropionici strain F3E8, deposited with the American Type Culture Collection (ATCC), on June 25, 2015 under ATCC Accession No. PTA- 122267 or P.

acidipropionici strain WGS 7, deposited with the American Type Culture Collection (ATCC), on September 2, 2016 under ATCC Accession No. , wherein the new strain has an increase in total yield (g/g) of propionic acid relative to that produced by P. acidipropionici ATCC 55737 when the two strains are cultured under identical conditions. In one embodiment a new P. acidipropionici strain is produced by exchanging genetic material, optionally through genomic shuffling, between P. acidipropionici strain F3E8 and P. acidipropionici strain WGS 7.

In one embodiment the resulting strain will have a maximum growth rate of at least 0.24/hr as calculated for the exponential phase of cells cultured under optimal conditions, including for example anaerobically at 32 °C, pH 6.5 in PAM media, and/or a propionic acid yield of about 0.55 g/g or higher. In one embodiment the new strain will produce propionic acid at a yield of about 0.49 to about 0.8 g/g, a about 0.5 to about 0.7 g/g, or about 0.5 to about 0.66, or about 0.66 g/g.

In accordance with one embodiment, the new strain is generated by conducting genomic shuffling between two parental P. acidipropionici strains. Briefly, the two parental strains are separately cultured and then collected from the culture by centrifugation and subjected to enzymatic treatment (lysozyme) to generate protoplasts. Equal numbers of protoplasts from each population are mixed together in protoplast formation buffer. Polyethylene glycol and CaCl ₂ are added to the suspension to induce fusion of the protoplasts, and the fused protoplasts are subsequently centrifuged, washed and resuspended in regeneration medium. Cells are then plated on growth media and cells exhibiting growth rates of at least 0.19/hr are selected for further analysis.

In accordance with one embodiment a method of producing a novel Propionibacterium strain having enhanced growth rates or propionic acid production relative to P. acidipropionici ATCC 4875 or P. acidipropionici ATCC 55737 is provided. The method comprises the steps of exchanging genetic material, optionally through genomic shuffling, between a first and second set of Propionibacterium strains. More particularly, the first set of Propionibacterium strains is selected from the group consisting of P. acidipropionici ATCC 4875, P.

acidipropionici ATCC 55737 or a strain generated from genomic shuffling between P.

acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737, P. acidipropionici ATCC 4965, P. intermedium ATCC 14072, or P. jensenii ATCC 9617. The second set of

Propionibacterium strains is P. acidipropionici strain F3E8, deposited with the American Type Culture Collection (ATCC), on June 25, 2015 under ATCC Accession No. PTA-122267. The strains generated by the genomic shuffling between the first and second sets of

Propionibacterium strains are then cultured to identify a novel Propionibacterium strain having enhanced growth rates or propionic acid production relative to P. acidipropionici ATCC 4875 or P. acidipropionici ATCC 55737. In one embodiment, the strains generated by the genomic shuffling between the first and second sets of Propionibacterium strains are first cultured to identify those strains exhibiting an enhance growth rate relative to P. acidipropionici ATCC 4875 or P. acidipropionici ATCC 55737. Those strains that exhibit a relative enhanced growth rate are then selected and the selected strains are then separately cultured to identify which of the selected strains have improved yields of propionic acid relative to P. acidipropionici ATCC 4875 or P. acidipropionici ATCC 55737. In accordance with one embodiment the first set of strains is generated from genomic shuffling between P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737. In one embodiment the first set of strains is generated from genomic shuffling between P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 14072. In one embodiment the first set of strains is generated from genomic shuffling between P. acidipropionici ATCC 4875 and P.jensenii ATCC 9617. In one embodiment the first set of strains represents a library of strains produced by the combination of two or more strains generated by separate genomic shuffling reactions. In one embodiment, the first set of strains in the disclosed method of producing a novel Propionibacterium strain having enhanced growth rates or propionic acid production relative to P. acidipropionici ATCC 4875 or P.

acidipropionici ATCC 55737 comprises a mixture of strains from 2, 3 or 4 of the following separate genomic shuffling reactions:

i) genomic shuffling between P. acidipropionici ATCC 4875 and P.

acidipropionici ATCC 55737;

ii) genomic shuffling between P. acidipropionici ATCC 4875 and P.

acidipropionici ATCC 4965;

iii) genomic shuffling between P. acidipropionici ATCC 4875 and P. intermedium ATCC 14072; and

iv) genomic shuffling between P. acidipropionici ATCC 4875 and P. jensenii ATCC 9617. In one embodiment, the first set of strains represents a combination of strains generated from i) and ii); or ii) and iii); or iii) and iv); or ii) and iv) or ii), iii) and iv); or i), ii), iii) and iv).

In accordance with one embodiment a method of producing a novel P. acidipropionici strain having improved yields of propionic acid production, said method comprises the steps of subjecting a first and second strain of P. acidipropionici to genome shuffling, wherein said first strain is selected from the group consisting of P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737 and said second strain is P. acidipropionici strain F3E8, deposited with the American Type Culture Collection (ATCC), on June 25, 2015 under ATCC Accession No. PTA-122267, to produce new strains;

optionally culturing said new strains to identify strains that have enhanced growth rates relative to P. acidipropionici ATCC 4875 or P. acidipropionici ATCC 55737 and selecting the resulting new strains having enhance growth rate; and measuring propionic acid production in said new strains to identify strains having improved yields of propionic acid relative to P. acidipropionici ATCC 4875 or P.

acidipropionici ATCC 55737.

The present disclosure also encompasses any of the novel P. acidipropionici strains produced in accordance with the methods disclosed herein.

In accordance with one embodiment an isolated or purified novel Propionibacterium strain is provided wherein the novel strain exhibits a faster growth rate and/or enhanced yields of propionic acid relative to P. acidipropionici strain F3E8, P. acidipropionici ATCC 4875 and/or P. acidipropionici ATCC 55737. In one embodiment the novel strains of the present disclosure have a maximum growth rate at least 1.5 times that of P. acidipropionici ATCC strain 4875 and/or an increase of 15 to 20 % in total yield (g/g) relative to that of P.

acidipropionici ATCC strain 4875, when the two strains are grown under identical conditions, including for example anaerobically at 32 °C, pH 6.5 in PAM media. In one embodiment the Propionibacterium strain of the present disclosure has a growth rate of about 0.24/hr to about 0.26/hr, as calculated for the exponential phase of cells cultured at 32 °C, pH 6.5 in PAM media. In one embodiment the Propionibacterium strain is a P. acidipropionici strain.

In one embodiment a P. acidipropionici strain of the present disclosure has an increase of at least 10%, and optionally a 10 to 15%, 15 to 20%, or a 20 to 25% in yield (g/g) of propionic acid relative to that produced by P. acidipropionici ATCC strain 4875 when the two strains are cultured under identical conditions (e.g., anaerobically at 32 °C, pH 6.5 in PAM media). In one embodiment the novel P. acidipropionici strain produces propionic acid with a yield of about 0.49 to about 0.7 g/g, about 0.5 to about 0.7 g/g, or about 0.5 to about 0.66, or about 0.66 g/g. In one embodiment the novel P. acidipropionici strain produces at least about 0.77 to about 0.92 g/L/hr or about 0.84 g/L/hr of propionic acid under optimal culture conditions, including for example anaerobically at 32 °C, pH 6.5 in PAM media. In one embodiment a P. acidipropionici strain of the present disclosure has an increase of at least 10%, and optionally a 10 to 15%, 15 to 20%, or a 20 to 25% in yield (g/g) of propionic acid relative to that produced by P. acidipropionici strain ATCC 55737 when the two strains are cultured under identical conditions (e.g., anaerobically at 32 °C, pH 6.5 in PAM media).

In one embodiment the P. acidipropionici strain is an isolated or purified P.

acidipropionici strain wherein said strain has an increase of 15 to 20 % in yield (g/g) of propionic acid relative to that produced by P. acidipropionici ATCC strain 4875, and optionally has a maximum growth rate at least 1.5 times that of P. acidipropionici ATCC strain 4875 when both strains are cultured under identical conditions, optionally anaerobically at 32 °C, pH 6.5 in an acceptable media such as PAM. In one embodiment the P. acidipropionici strain is an isolated or purified P. acidipropionici strain wherein the strain has a propionic acid yield of about 0.49 to about 0.7 g/g, about 0.6 to about 0.8 g/g, about 0.5 to about 0.7 g/g, or about 0.5 to about 0.66, or about 0.66 g/g, or about 0.75 g/g, and optionally has a maximum growth rate at least 1.5 times that of P. acidipropionici ATCC strain 4875 when both strains are cultured under identical conditions, optionally anaerobically at 32 °C, pH 6.5 in an acceptable media such as PAM. In one embodiment a new P. acidipropionici strain is provided where the strain is an isolated or purified P. acidipropionici strain having a propionic acid yield that is 20, 30 or 40% higher than the propionic acid yield of P. acidipropionici strain ATCC 55737 when both strains are grown under identical conditions, optionally, anaerobically at 32 °C, pH 6.5 in an acceptable media such as PAM. In one embodiment the P. acidipropionici strain is an isolated or purified P. acidipropionici strain wherein the strain has a propionic acid yield of about 0.49 to about 0.8 g/g, about 0.5 to about 0.7 g/g, or about 0.5 to about 0.66, or about 0.55 or 0.66 g/g, and optionally has a maximum growth rate of about 0.24/hr to about 0.26/hr, as calculated for the exponential phase of cells cultured at 32 °C, pH 6.5 in PAM media. In one embodiment the cultured P. acidipropionici strain is an isolated or purified P. acidipropionici strain wherein the strain produces about 0.84 g/L/hr to about 0.95 g/L/hr of propionic acid, and optionally has a maximum growth rate of at least 0.24/hr OD600 as calculated for the exponential phase of cells cultured at 32 °C, pH 6.5 in an acceptable media such as PAM. In one embodiment the cultured P. acidipropionici strain is an isolated or purified P. acidipropionici strain wherein the strain produces at least about 0.77 to about 0.95 g/L/hr or about 0.84 g/L/hr of propionic acid under optimal culture conditions and optionally has a growth rate of about 0.24/hr to about 0.26/hr, as calculated for the exponential phase of cells cultured at 32 °C, pH 6.5 in PAM media.

In accordance with one embodiment, any of the above described Propionibacterium strains are further characterized by comprising one or more modified genes relative to P.

acidipropionici ATCC 4875 and ATCC 55737 as per Tables 1 and 7. In one embodiment the Propionibacterium strain comprises a modified gene, relative to strain P. acidipropionici ATCC 4875 and ATCC 55737, encoding for the large subunit ribosome, the long chain acyl-CoA synthetase, or the cation diffusion facilitator (Table 1). In one embodiment the

Propionibacterium strain comprises a modified gene, relative to strain P. acidipropionici strain ATCC 55737, selected from the group consisting of the ABC polar amino acid transporter gene, the Cytochrome C biogenesis gene and the ABC multiple sugar transporter gene (Table 7). In one embodiment the modified gene comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 modifications relative to the corresponding gene found in strain P. acidipropionici ATCC 55737 or ATCC 4875 , wherein each modification represents a single nucleotide insertion, deletion or substitution. In one embodiment a

Propionibacterium strain is provided that comprises modified genes, relative to P.

acidipropionici ATCC 55737 or ATCC 4875, for each of the large subunit ribosome, the long chain acyl-CoA synthetase, and the cation diffusion facilitator.

In accordance with one embodiment an isolated Propionibacterium strain is provided, optionally a P. acidipropionici strain, wherein said strain has a maximum growth rate of at least 0.24/hr as calculated for the exponential phase of cells cultured anaerobically at 32 °C, pH 6.5 in PAM media, and said strain comprises a modified gene, relative to strain P. acidipropionici ATCC 55737 or ATCC 4875, that encodes an altered gene product for the ABC polar amino acid transporter, the Cytochrome C biogenesis protein, the large subunit ribosomal RNA, the long chain acyl-CoA synthetase, or the cation diffusion facilitator.

In accordance with one embodiment an isolated Propionibacterium strain is provided, optionally a P. acidipropionici strain, having an improved propionic acid yield relative to P. acidipropionici ATCC strain 4875 and ATCC 55737, wherein the strain comprises a modified gene, relative to strain P. acidipropionici ATCC 4875 and ATCC 55737 for the ABC polar amino acid transporter gene, the Cytochrome C biogenesis gene, the ABC multiple sugar transporter gene, the large subunit ribosome, the long chain acyl-CoA synthetase, or the cation diffusion facilitator . In one embodiment the Propionibacterium strain comprising the modified large subunit ribosome, the long chain acyl-CoA synthetase, or the cation diffusion facilitator has an increase of at least 15 to 20 % in total yield (g/g) of propionic acid relative to that produced by P. acidipropionici ATCC 4875 when the two strains are cultured under identical conditions, and in one embodiment producing a propionic acid yield of at least 0.55 or 0.66 g/g. In one embodiment the strain comprises a modified large subunit ribosome, long chain acyl- CoA synthetase, and cation diffusion facilitator genes. In one embodiment the modified gene encodes an altered gene product.

In accordance with one embodiment of the present disclosure isolated

Propionibacterium strains were initially selected that produced enhanced yields of propionic acid. After analyzing ten selected strains, three variant genes were conserved in all ten selected strains (Table 1). The variations were: i) in the cation diffusion facilitator,

ii) in the large subunit ribosomal RNA (23S),

iii) in the long chain fatty acyl-CoA synthetase.

The positions of the variants are specified in Table 1.

Table 1: Positions of the conserved gene variants found in the 10 new strains taking as reference P. acidipropionici ATCC 4875 or P. acidipropionici ATCC 55737. All 10 strains had the conserved mutations in their genome.

*New strains: the 10 strains resulted from genome shuffling

ATCC4875: P. acidipropionici ATCC 4875

ATCC55737: P. acidipropionici ATCC 55737

SNP: Single nucleotide polymorphism

INDEL: Insertion or deletion

X: Indicates the amino acid cannot be defined Legend of amino acids

SLC: Single-letter data-base codes; A: Acidic; B:Basic; Ali: Aliphatic; Ami: Amine; Aro: Aromatic; N:Nonpolar; PU:Polar uncharged; NEG: negativel charged; POS: Positively charged.

In one embodiment an isolated Propionibacterium strain is provided, optionally a P. acidipropionici strain, that comprises a modified large subunit ribosome gene that encodes a large subunit ribosomal RNA (SEQ ID NO: 15) wherein the gene is altered in its primary sequence relative to the same gene in P. acidipropionici ATCC 55737 or ATCC 4875. In one embodiment the large subunit ribosomal gene of the presently disclosed Propionibacterium strain is altered by 1, 2, or 3 nucleotide substitutions. In accordance with one embodiment the large subunit ribosomal (23S) gene of the present Propionibacterium strain differs from the large subunit ribosomal gene of P. acidipropionici ATCC 4875 by 1, 2, or 3 nucleic acid substitutions at positions 1270, 1271 and 1425, and more particularly in one embodiment the nucleotide substitutions are G1270A, T1271C and G1425T. In one embodiment the large subunit ribosomal (23S) gene comprises a sequence having 99% sequence identity relative to the same gene in P. acidipropionici ATCC 55737 with a substitution at position 1425, and more particularly in one embodiment the nucleotide substitution is A1441G.

In one embodiment an isolated Propionibacterium strain is provided, optionally a P. acidipropionici strain, wherein the strain comprises a modified long chain acyl-CoA synthetase gene relative to P. acidipropionici ATCC 55737 or ATCC 4875. In one embodiment the modification to the long chain acyl-CoA synthetase gene prevents the expression of a functional long chain acyl-CoA synthetase. In one embodiment the gene comprises a nucleotide insertion or deletion that causes a frameshift mutation. In one embodiment the insertion or deletion introduces a stop codon into the reading frame of the transcribed mRNA resulting in the production of a truncated peptide. In one embodiment the gene encoding the long chain acyl-CoA synthetase comprises a nonsense codon after nucleotide position 1500 and deletion or insertion at position 2269 relative to the same gene in P. acidipropionici ATCC 55737 and positions 264, 1271, and 2295 to ATCC 4875. In one embodiment the encoded long chain acyl-CoA synthetase is truncated near amino acid position 500 of the native protein and position 757 relative to P. acidipropionici ATCC 55737 and amino acid substitution at position 88 and non-sense amino acid substitution at position 757 relative to same amino acid sequence in P. acidipropioici ATCC 4875 In one embodiment the modified long chain acyl-CoA synthetase gene comprises the sequence of SEQ ID NO: 16 or a sequence that encodes the truncated peptide of SEQ ID NO: 17.

In one embodiment an isolated Propionibacterium strain is provided, optionally a P. acidipropionici strain, wherein the strain comprises a modified cation diffusion facilitator gene. In accordance with one embodiment the cation diffusion facilitator gene (SEQ ID NO: 13) of the present Propionibacterium strain differs from the cation diffusion facilitator gene in P. acidipropionici ATCC 4875 by 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or more nucleic acid substitutions at positions 19,178, 256, 211, 334, 340, 406, 451, 484, 574, and 953. The particular substitutions are G19A, T178G, G211A, A256G, G334A, C340G, A406C, C451A, T484C, A574G, and G953A. In one embodiment the modified cation diffusion facilitator gene comprises a sequence having 99.5% or 99% sequence identity relative to the same gene in P.

acidipropionici ATCC 55737 and optionally having an adenosine at position 953. In one embodiment the modified gene encodes an altered gene product, wherein the encoded polypeptide comprises one or more amino acid substitutions relative to the cation diffusion facilitator protein encoded by P. acidipropionici ATCC 4875. In accordance with one embodiment the cation diffusion facilitator protein (SEQ ID: 14) of the present Propionibacterium strain differs from the cation diffusion facilitator protein encoded in P. acidipropionici ATCC 4875 by 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or more amino acid substitutions at positions 7, 60, 71, 86, 112, 114, 136, 151, 162, 192, and 318 with amino acid residues I, A, N, V, I, A, H, T, P, A, and Y respectively. In one embodiment the altered cation diffusion facilitator protein differs from the cation diffusion facilitator protein encoded by P. acidipropionici ATCC 55737 by 1, 2, or 3 non-conservative amino acid substitutions. In one embodiment the altered gene product of the cation diffusion facilitator of the present strain comprises an amino acid substitution at amino acid position 318 relative to the cation diffusion facilitator protein encoded by P. acidipropionici ATCC 4875, and more particularly the amino acid substitution is a non-conservative amino acid substitution. In one embodiment the isolated Propionibacterium strain of the present disclosure encodes a cation diffusion facilitator protein wherein the native C at amino acid position 318 is substituted with F, Y, and W. In one embodiment the isolated Propionibacterium strain of the present disclosure encodes a cation diffusion facilitator protein wherein the native C at amino acid position 318 is substituted with Y. In one embodiment the modified cation diffusion facilitator protein gene comprises the sequence of SEQ ID NO: 13 or a gene that encodes a peptide of SEQ ID NO: 14. SEQ ID NO: 13 differs in sequence from the cation diffusion facilitator protein gene of P. acidipropionici ATCC 4875, at 11 different nucleotides (at positions 19, 178, 211, 256, 334, 340, 406, 451, 484, 574, and 953). In one embodiment the encoded cation diffusion facilitator protein comprises the sequence of SEQ ID NO: 14, or a polypeptide that differs from the same protein sequence in P. acidipropionici ATCC 4875 or ATCC 55737 by 1, 2, 3, or more amino acid substitutions at positions referred in Table 1.

In one embodiment the new strain has an extra copy of the whole ribosomal RNA and an extra copy of the arginine deiminase regulon (ArgR). The extra copy of the ArgR has a mutation in position 37 of the gene (DEL of GC). The mutation is present in one of the two active domains of the protein. The mutation causes a frame- shift in the gene. This gene has been associated with the regulation of the arginine deiminase pathway (involved in acid tolerance mechanism). The mutation reported here has a positive effect contributing to the improvement of the new strain P. acidipropionici F3E8.

In one embodiment an isolated Propionibacterium strain is provided, optionally a P. acidipropionici strain, wherein the strain comprises

(i) a modified large ribosomal RNA comprising a nucleotide substitution of the large ribosomal gene at position 1441 relative to the same gene in P. acidipropionici ATCC 55737 and position 1270, 1271, and 1425 relative to the same gene in P. acidipropionici ATCC 4875; and/or

(ii) a modified long chain acyl-CoA synthetase gene comprising a nonsense codon after nucleotide position 1500 and position 2269 relative to the same gene in P. acidipropionici ATCC 55737 and positions 264, 1650, and 2295 relative to the same gene in P. acidipropionici ATCC 4875; and/or

(iii) a modified cation diffusion facilitator gene encoding a gene product comprising an amino acid substitution at amino acid position 318 relative to the encoded protein in P. acidipropionici ATCC 55737 and positions relative to P. acidipropionici ATCC 4875 according to Table 1 ; and/or

(iv) a copy number variation (e.g., an extra copy) of the whole RNA operon (SEQ ID NO: 18) relative to P. acidipropionici ATCC 4875 or ATCC 55737.

(v) a copy number variation (e.g., an extra copy) of the arginine deiminase regulon (SEQ ID NO: 19) with a deletion at position 37 (SEQ ID NO: 21) relative to P. acidipropionici ATCC 4875 or ATCC 55737. In one embodiment an isolated Propionibacterium strain is provided, optionally a P. acidipropionici strain, comprising a gene sequence modified relative to a gene sequence selected from the group consisting of SEQ ID NO: 1, SEQ ID NO: 3 and SEQ ID NO: 4, or a sequence that shares 85%, 90%, 95% or 99% sequence identity with SEQ ID NO: 1, SEQ ID NO: 3 and SEQ ID NO: 4 but less than 100% sequence identity.

In one embodiment an isolated Propionibacterium strain is provided, optionally a P. acidipropionici strain, wherein the strain comprises

(i) a modified large ribosomal RNA comprising the sequence of SEQ ID NO: 15, or a sequence having 95% or 99% sequence identity with same sequences in P. acidipropionici ATCC 4875 or ATCC 55737; and/or

(ii) a modified long chain acyl-CoA synthetase comprising the sequence of SEQ ID NO: 17, or a polypeptide that differs from SEQ ID NO: 17 by 1, 2 or 3 amino acid substitutions; and/or

(iii) a modified cation diffusion facilitator comprising the sequence of SEQ ID NO: 14, or a polypeptide that differs from SEQ ID NO: 14 by 1, 2 or 3 amino acid substitutions; and/or

(iv) a copy number variation (e.g., an extra copy) of the whole RNA operon (SEQ ID NO: 17) relative to P. acidipropionici ATCC 4875 or ATCC 55737; and/or

(v) a copy number variation (e.g., an extra) copy of the arginine deiminase regulon (SEQ ID NO: 20) with a deletion at position 37 relative to P.

acidipropionici ATCC 4875 or ATCC 55737 that causes a frame shift. The deletion is located in one of the two domains.

In one embodiment, an isolated Propionibacterium strain is provided, optionally a P. acidipropionici strain, comprising a large subunit ribosome gene sequence of SEQ ID NO: 15, a long chain acyl-CoA synthetase gene sequence of SEQ ID NO: 16, and a cation diffusion facilitator gene sequence of SEQ ID NO: 13. In one embodiment, an isolated P. acidipropionici strain is provided comprising gene sequences encoding a large subunit ribosome RNA comprising SEQ ID NO: 15, a long chain acyl-CoA synthetase comprising SEQ ID NO: 17, and a cation diffusion facilitator comprising SEQ ID NO: 2.

In one embodiment an isolated Propionibacterium strain is provided, optionally a P. acidipropionici strain, wherein said strain comprises one or more duplication of proteins responsible for regulation of the arginine deiminase pathway, optionally in combination with one or more modified genes, relative to strain P. acidipropionici ATCC 4875, encoding for the large subunit ribosome, the long chain acyl-CoA synthetase, or the cation diffusion facilitator as disclosed above. In one embodiment the strains include any of the variations listed in Table 1. These include for example modifications to transcriptional regulators, transport, and genes linked to acid tolerance mechanisms. For example, additional Propionibacterium strains of the present disclosure in some embodiments include mutations in the transcriptional regulator MerR gene, the transcriptional regulator of the DeoR family, and/or the sigma 54 specific transcriptional regulator of the Fis family. Examples of mutations related to transport include a mutation in the Na+/H+ antiporter, the ABC transporter binding protein, the ABC-type nitrate/sulfonate/bicarbonate transport system, the arsenic efflux pump protein and the oligopeptide transport ATP-binding protein. In some embodiments, the Propionibacterium strains of the present disclosure include mutations in genes related to acid tolerance mechanisms such as a mutation in the malto-oligotreahalose trehalohydrolase and mutations in phosphogluconate dehydrogenase decarboxylating gene. Other mutations included dihydrolipoamide succinyltransferase component E2 of 2-oxoglutarate dehydrogenase complex/2- oxoglutarate dehydrogenase El component and D-3-phosphoglycerate dehydrogenase and mutations in the catalase/peroxidase gene.

In accordance with one embodiment a novel P. acidipropionici strain is provided selected from the group consisting of strains F3E8, F3G8, F3C8, F3H8, F3E9, F3F8, F3F6, F3D9, F3C1, F3B9, WGS.l, WGS.2, WGS.3, WGS .4, WGS.5, WGS.6, WGS 7, WGS.8, WGS.9, WGS.IO, WGS.ll, WGS.12, and WGS.13. In one embodiment the novel P. acidipropionici strain is selected from the group consisting of strains F3E8, F3G8, F3C8, F3H8, F3E9, F3F8, F3F6, F3D9, F3C1, and F3B9. In one embodiment the novel P. acidipropionici strain is selected from the group consisting of strains F3E8 and F3F6. In accordance with one embodiment a novel P. acidipropionici strain is provided selected from the group consisting of strains WGS.l, WGS.2, WGS.3, WGS.4, WGS.5, WGS.6, WGS 7, WGS.8, WGS.9, WGS.IO, WGS.ll, WGS.12, and WGS.13.

In accordance with one embodiment the novel strain is P. acidipropionici strain F3E8, deposited with the American Type Culture Collection (ATCC), 10801 University Boulevard, Manassas, VA, 20110-2209 USA on June 25, 2015, and assigned ATCC Accession No.122267. In accordance with one embodiment the novel strain is P. acidipropionici strain WGS 7, deposited with the American Type

Culture Collection (ATCC), on under ATCC Accession No. PTA- Access to these deposits will be available during the pendency of the application to the

Commissioner of Patents and Trademarks and persons determined by the

Commissioner to be entitled thereto upon request. The deposits will be maintained in the ATCC Depository, which is a public depository, for a period of 30 years, or 5 years after the most recent request, or for the enforceable life of the patent, whichever is longer, and will be replaced if it becomes nonviable during that period. Applicant does not waive any infringement of their rights granted under this patent.

The present disclosure also encompasses derivatives of P. acidipropionici strains F3E8, F3C8, F3H8, F3E9, F3F8, F3F6, F3D9, F3C1, F3B9, WGS.l, WGS.2, WGS.3, WGS .4, WGS.5, WGS.6, WGS 7, WGS.8, WGS.9, WGS.10, WGS.11, WGS.12, and WGS.13, formed by genomic shuffling with other known P.

acidipropionici strains or wherein one or more genes of the parent strains have been modified to produce the derivative. In one embodiment the strains F3E8, F3C8, F3H8, F3E9, F3F8, F3F6, F3D9, F3C1 and F3B9 are subjected to one or more rounds of genetic shuffling with a different strain of P. acidipropionici. In one embodiment a derivative is formed by conducting genomic shuffling of one strain selected from the group consisting of strains F3E8, F3C8, F3H8, F3E9, F3F8, F3F6, F3D9, F3C1,

F3B9, WGS.l, WGS.2, WGS.3, WGS.4, WGS.5, WGS.6, WGS 7, WGS.8, WGS.9, WGS.10, WGS.l 1, WGS.12, and WGS.13 and a second strain selected from either P. acidipropionici ATCC 4875 or P. acidipropionici ATCC 55737. In another embodiment a derivative is formed by conducting genomic shuffling between one strain selected from the group consisting of strains F3E8, F3C8, F3H8, F3E9, F3F8, F3F6, F3D9, F3C1, F3B9, WGS.l, WGS.2, WGS.3, WGS .4, WGS.5, WGS.6, WGS 7, WGS.8, WGS.9, WGS.10, WGS.l 1, WGS.12, and WGS.13 and a different strain selected from the group consisting of strains F3E8, F3C8, F3H8, F3E9, F3F8, F3F6, F3D9, F3C1, F3B9, WGS.l, WGS.2, WGS.3, WGS .4, WGS.5, WGS.6, WGS 7, WGS.8, WGS.9, WGS.10, WGS.l 1, WGS.12, and WGS.13. In one embodiment the derivative is formed by conducting genomic shuffling between one strain selected from the group consisting of strains F3E8, F3C8, F3H8, F3E9, F3F8, F3F6, F3D9, F3C1 and F3B9 and a second different P. acidipropionici strain (i.e., not ATCC 4875, ATCC 55737, F3E8, F3C8, F3H8, F3E9, F3F8, F3F6, F3D9, F3C1 or F3B9). In one embodiment the derivative is formed by conducting genomic shuffling between strain F3E8 and a second different P. acidipropionici strain.

In accordance with one embodiment strains F3E8, F3G8, F3C8, F3H8, F3E9, F3F8, F3F6, F3D9, F3C1 and F3B9 are further modified to reduce the production of byproducts such as acetate and succinate. The production of these byproducts not only lowers the yield of the main fermentation product but also causes difficulty in product purification.

In accordance with one embodiment, a method for the commercial production of propionic acid or one of its derivatives such as propanol or propylene is provided comprising batch fermentation utilizing the novel Propionibacterium acidipropionici disclosed herein. In accordance with one embodiment, the primary substrates for the fermentation are sucrose and/or glucose. In one embodiment the primary nitrogen source will be gaseous ammonia that is supplemented with an amino acids/protein cocktail from plant derived flour (soy flour, cotton seed flour, corn steep flour, etc.). However, other nitrogen sources are known to those skilled in the art and are suitable for use in the disclosed methods.

In one embodiment the method for producing propionic acid or one of its derivatives such as propanol or propylene is provided. The method comprises the steps of culturing a P. acidipropionici strain as disclosed herein under conditions suitable for growth of the strain and recovering the propionic acid produced by said strain. In one embodiment the cultured P. acidipropionici strain is an isolated or purified P. acidipropionici strain wherein said strain has a maximum growth rate of at least 0.24/hr as calculated for the exponential phase of cells cultured at 32 °C, pH 6.5 in an acceptable media such as PAM. In one embodiment the cultured P.

acidipropionici strain is an isolated or purified P. acidipropionici strain wherein the strain has a maximum growth rate at least 1.5 times that of P. acidipropionici ATCC strain 4875. In one embodiment the cultured P. acidipropionici strain is an isolated or purified P. acidipropionici strain wherein said strain has an increase of 15 to 20 % in total yield (g/g) of propionic acid relative to that produced by P. acidipropionici ATCC strain 4875, and optionally has a maximum growth rate at least 1.5 times that of P. acidipropionici ATCC strain 4875 when both strains are cultured under identical anaerobic conditions, optionally at 32 °C, pH 6.5 in an acceptable media such as PAM. In one embodiment the cultured P. acidipropionici strain is an isolated or purified P. acidipropionici strain wherein the strain has a propionic acid yield of about 0.55 or 0.66 g/g, and optionally has a maximum growth rate at least 1.5 times that of P. acidipropionici ATCC strain 4875 when both strains are cultured under identical anaerobic conditions, optionally at 32 °C, pH 6.5 in an acceptable media such as PAM. In one embodiment the cultured P. acidipropionici strain is an isolated or purified P. acidipropionici strain wherein the strain produces about 0.84 g/L/hr of propionic acid, and optionally has a maximum growth rate of at least 0.24/hr as calculated for the exponential phase of cells cultured at 32 °C, pH 6.5 in an acceptable media such as PAM. '

In one embodiment the cultured Propionibacterium strain, optionally a P. acidipropionici strain, comprises

(i) a modified large ribosomal RNA comprising a nucleotide substitution of the large ribosomal gene at positions according to Table 1 ;

(ii) a modified long chain acyl-CoA synthetase gene comprising a nonsense codon according to Table 1 ; and

(iii) a modified cation diffusion facilitator gene encoding a gene product comprising an amino acid substitution at amino acid position 318 relative to the same gene in P. acidipropionici ATCC 55737 and positions as shown in Table 1 relative to P. acidipropionici ATCC 4875. In one embodiment the cultured

Propionibacterium strain, optionally a P. acidipropionici strain, comprises a modified gene sequence relative to a gene sequence selected from the group consisting of SEQ ID NO: 1, SEQ ID NO: 3 and SEQ ID NO: 4, or a sequence that shares 85%, 90%, 95% or 99% sequence identity with SEQ ID NO: 1, SEQ ID NO: 3 and SEQ ID NO: 4 but less than 100% sequence identity. In one embodiment the cultured P.

acidipropionici strain comprises modified gene sequences relative to each of the gene sequences SEQ ID NO: 1, SEQ ID NO: 3 and SEQ ID NO: 4, or a sequence that shares 85%, 90%, 95% or 99% sequence identity with SEQ ID NO: 1, SEQ ID NO: 3 and SEQ ID NO: 4, but less than 100% sequence identity;

(iv) a copy number variation (e.g., an extra copy) of the whole RNA operon (SEQ ID NO: 18) relative to P. acidipropionici ATCC 4875 or ATCC 55737; and/or

(v) a copy number variation (e.g., an extra copy) of the arginine deiminase regulon (SEQ ID NO: 20) with a deletion at position 37 relative to P. acidipropionici ATCC 4875 or ATCC 55737 that causes a frame shift. The deletion is located in one of the two domains. In one embodiment the cultured P. acidipropionici strain is an isolated or purified P. acidipropionici strain selected from the group consisting of strains F3E8, F3C8, F3H8, F3E9, F3F8, F3F6, F3D9, F3C1, F3B9, WGS.l, WGS.2, WGS.3, WGS.4, WGS.5, WGS.6, WGS 7, WGS.8, WGS.9, WGS.10, WGS.l 1, WGS.12, and WGS.13 or from strains F3E8, F3G8, F3C8, F3H8, F3E9, F3F6, F3D, and F3C1, or from strains WGS 7, F3E8 and F3F6. In one embodiment the cultured P.

acidipropionici strain is isolated or purified P. acidipropionici strain F3E8, deposited under ATCC Accession No. 122267. In accordance with one embodiment the novel strain is P. acidipropionici strain WGS 7, deposited with the American Type Culture Collection (ATCC), on September 2, 2016 under ATCC Accession No. .

EXAMPLE 1

Use of Genome Shuffling (GS) to Prepare New Strains of Propionibacterium acidipropionici

Propionibacterium sp are pleomorphic rods, gram-positive bacteria that naturally produce PA as their main fermentation product through the Wood- Werckman cycle. Natively, PA is produced along with other organic acids (lactate, succinate, and acetate) resulting in low productivities and modest yields which translate in costly downstream processes. Until recently, metabolic engineering in P. acidipropionici had proven challenging mainly due to the seven Clustered Regularly Interspaced Short Palindromic Repeats (CRISPRs) which provide resistance against conjugative plasmids and bacteriophages. Therefore genome shuffling (GS) was used to improve the growth rate and PA production of P. acidipropionici, obtaining a strain that can achieve improved yields and higher growth rates.

In addition, to understand genomic changes leading to the improved phenotype, next generation sequencing (NGS) was used to characterize genotypic changes. Using ClonalFrame (Didelot and D. Falush, Genetics, 2007, 175, 1251-66), changes in the genome are shown to correspond to regions of high recombination probability, thus, reinforcing the mechanism of gene conversion during GS. GS in bacteria has mainly been linked to gene conversion (asymmetric contributions) where the donor provide a small amount of DNA material leaving the rest of the recipient intact. MATERIAL AND METHODS

Bacteria. P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737 were selected from a collection of 17 strains(Stowers et al., J. Ind. Microbiol.

Biotechnol., 2014, 41, 837-852) for their fast growth and ability to achieve high PA yields. The strains were kept at - 80 °C using glycerol (20%) as cryoprotector.

Media. The culture media (PAM) for pre-inoculum preparation consisted of yeast extract (10 g/L), trypticase soy (5 g/L), K ₂ HP0 ₄ (0.05 g/L), MnS0 ₄ (0.05 g/L), and glucose (40 g/L). Agar (15 g/L) was added only to prepare PAM plates. Glucose (100 g/L) was used as the carbon source. Media components and the carbon source were sterilized separately for 20 min at 121 °C.

Protoplast formation buffer (PFB). This buffer consisted of sodium succinate (40.5 g/L), sucrose (42.75 g/L), and MgCl ₂ (1.9 g/L) dissolved in one litre of Tris-HCl 0.05 mol L at pH 7.1.

Regeneration buffer (RB). This buffer consisted of yeast extract (10 g/L), trypticase soy (5 g/L), KH ₂ P0 ₄ (1.5 g/L), K ₂ HP0 ₄ (2.5 g/L), and Bovine Serum Albumin (BSA) (5 g/L) adjusted to pH 7.

Analytical methods. The optical density of the culture was measured at 600 nm using a Biochrom Libra S12 UV/Vis Spectrophotometer. Organic acids, carbohydrates, and alcohol were quantified by ion-exclusion chromatography using an Agilent 1200 HPLC system and an Agilent Hiplex H column (300 x 7.7 mm, PLl 170- 6830) with a guard column (SecurityGuard Carbo-H, Phenomenex PN: AJO-4490). Sugars and alcohols were monitored using a refractive index detector (Agilent RID, G1362A) set on positive polarity and optical unit temperature of 40 °C, while organic acids were monitored at 210 nm (Agilent MWD, G1365B). 30 uL of sample was injected onto the column using an auto-sampler (Agilent HiP-ALS, G1367B) and column temperature kept at 65 °C using a thermostatted column compartment (Agilent TCC, G1316A). Analytes were eluted isocratically with 4 mM H ₂ S0 ₄ at 0.6 mL/min for 26 min. Chromatograms were integrated using ChemStation (Rev B.03.02[341]).

Inoculum preparation. Under sterile conditions, bacteria activation was carried out in a 1.5 mL Eppendorf tube with 1 mL of PAM media inoculated with 0.8% (v/v) of a glycerol stock. This culture was allowed to grow for 24 hours at 32 °C. The cultures were transferred to a 15 mL Falcon tube containing 14 mL of PAM media and allowed to grow for an additional 24 hours. 5% (v/v) of this culture was used to inoculate 250 mL serum bottles containing 100 mL of PAM media and allowed to grow for an additional 24 hours. Cells from the serum bottles in mid- exponential phase were used to inoculate the fermenters at an initial OD600nmof 0.3.

Protoplast preparation and regeneration. Protoplasts were prepared as described in Guan et al. (Guan et al., J. Biotechnol., 2013, 167, 56-63) with minor modifications. Cells were grown for 24 hours in PAM media supplemented with 40 g/L of glucose and 1 % of glycine. Cells were then conditioned in PAM media containing 1% of glycine and 120 g/L of glucose for an extra 24 hours. After at least ten generations, cells were washed two times using PBS and fixed to an OD600nm = 0.2 in a lysozyme solution containing 15 mg/mL - 600,000 U/mL - lysozyme solution in PFB. Cell walls were digested for two hours at 120 rpm and 40 °C.

Protoplasts were detected using a light microscope using the lOOx oil immersion objective and counted using a haemocytometer. When appropriate, protoplasts were regenerated in RB (pH = 7 for 48 h at 32 °C).

Genome shuffling (GS). The protocol from Guan et al (Guan et al., J.

Biotechnol., 2013, 167, 56-63) was used with some modifications. Protoplasts were treated with UV light for 0.5 min or heated at 60 °C for 2 h. Cells were mixed, centrifuged and re-suspended in 500 of PFB. Then 500 uL of PEG 6000 (80%) with 20 mmol/L CaCl ₂ was added. Fusion conditions were conducted at pH 7.4, 32 °C for 30 min. After fusion, 5 mL of PFB was added, and the suspension was centrifuged at 2500 rpm for 5 min. Protoplasts were washed two times with 5 mL of PFB and re-suspended in 1 mL of RB.

Screening. On a first stage, PAM media plates were used to screen for strains with improved growth rate. PAM plates were incubated in an anaerobic chamber at 32 °C and monitored for 24 hours every 6 hours. The first colonies on the plate were labeled and selected for the next round of GS. After three rounds of GS, isolated colonies were randomly selected for screening in 96-well plates containing 100 μL of PAM media. Growth was monitored using a micro-plate reader Omega FLUOstar® adapted to maintain anaerobic conditions through a constant injection of nitrogen. The best performing strains were scaled up to 250 mL serum bottles with a working volume of 100 mL under N ₂ . Serum bottles were incubated using an orbital shaker incubator (Infors HT multitron standard) with an agitation rate of 100 rpm and a working temperature of 32 °C for 96 hours. Instrumented fermenters. Fermentations were performed using 2 L

Applikon fermenters with a working volume of 1 L. Fermenters were equipped with probes and controllers for pH, dissolved oxygen, temperature, and agitation. The agitation rate was controlled with two Rushton impellers at 300 rpm. The pH was controlled at 6.5 using 10 M NaOH. The temperature of the culture was maintained at 32 °C using an electric jacket. Prior to inoculation, the fermenters were sparged with N ₂ for at least 15 minutes. A constant N ₂ flow was kept for the entire fermentation at a flow rate of 0.3 L/min.

Fermentation calculations. Maximum specific growth rate ^ _m ax) was calculated in exponential phase. For consistency, volumetric productivity (Pv) was calculated for the same time interval (ranging from 22.5 to 52 hours). Yield (Yps) was calculated using the total PA produced over the total substrate consumed.

Finally, ratios PA: Acetic Acid and PA:Succinic Acid were calculated using total organic acid production.

DNA-sequencing and de novo assembly. Genomic DNA of the different

Propionibacterium strains were extracted using PureLink® genomic DNA mini kit (Invitrogen Cat. No. K1820-01) and quantified using Nanodrop™ 1000 (Thermo Scientific) and Qubit® dsDNA BR assay kit (Life Technologies Cat. No. Q32850). Quality of the DNA was determined by running a 1% agarose gel with DNA gel stain SYBR safe (Life Technologies Cat. No. S33102). The gel was visualized in a

ChemiDoc MP system (Bio-Rad). The Illumina platform was used to sequence the genomes of 11 strains (n=l parental strain P. acidipropionici ATCC 55737 and n=10 strains from GS). Sequencing was performed using TrueSeq® Illimina 300 PE.

Library were prepared using Illumina TrueSeq® DNA HT sample preparation kit (illumina Cat. No. FC- 121-2003). The assembly of the reads was performed using the SPAdes genome assembly algorithm (Bankevich et al., Comput. Biol., 2012, 19, 455- 477. To close the genomes, two strains (the parental strain P. acidipropionici ATCC 55737 and the recombinant P. acidipropionici F3E8) were sequenced using PacBio RS II technology. The PacBio® library preparation was performed using the protocol for 20 Kb selected with the BluePippin™ system. The sequencing chemistry was the most recent released P6-C4 by PacBio and loaded by magnetic beads. The genome assembly was performed with the SMRT® portal.

Bioinformatics tools. RAST server was used to annotate the different assembled genomes (Aziz et al., BMC Genomics, 2008, 9, 75). Mauve software was used to perform a multiple whole genome alignments. Mauve was also used to move the contigs of the strains from GS taking as a reference the closed genome P.

acidipropionici F3E8 (Darling et al., Genome Res., 2004, 14, 1394^03). GRIL was used to find rearrangements in the Mauve alignment (Darling et al., Bioinformatics, 2003, 20, 122-124). Genome-to-genome distance calculator (GGDC 2.0) was used to calculate the genome distance between two genomes (Auch et al., Genomic Sci., 2010, 2, 142-148). Bowtie2 and SAMtools or SMRT® portal were used to align the reads and call the variants of the sequenced genomes. SnpEff was used to annotate the variations and SnpSift to filter them (Cingolani et al., Front. Genet., 2012, 3, 35. IGV viewer was used to visualizing the variations. ClonalFrame, was used to determine the number of homologous recombination events in the recombinants. Bio- Python programming language was used to develop a script to perform a Blast gene- gene differential and score system comparison between two genomes (Altschul et al., J. Mol. Biol., 1990, 215, 403-10).

RESULTS

Screening of superior strains.

To increase PA production using GS, P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737 were selected as the initial strains. After GS, colonies were selected using high-throughput micro kinetics to select strains with higher growth rate. The growth rate of the new colonies was compared to the growth rate of the parental strains (grown on the same plate). From each plate, the 10 fastest growing colonies were grown in serum bottles. One of the fastest growing strains (referred hereafter as F3E8) out of the lot of 10 strains was characterized on bioreactors (Fig. 1C). The new strain displayed a growth rate of 0.26 h ^"1 compared to the growth rate of the parental strains of 0.15 h ^"1 and 0.18 h ^"1 (Table 2) which is in accordance with the growth rate observed in serum bottles (Table 3). Table 2 Kinetic parameters of the 2 L instrumented fermentations to test the stability of the parental strains Propionibacterium acidipropionici ATCC 4875 and

Propionibacterium acidipropionici ATCC 55737, and the new strain from GS Propionibacterium acidipropionici F3E8. PA:SA is the PA titer divided by the SA titer at the end of the fermentation. PA: AA is the PA titer divided by the AA titer at the end of the fermentation.

PA:SA

Strain Yps (g/g) PA:AA (g/g) *Pv (g/Lh) μ (1/η)

(g/g)

P. acidipropionici ATCC

0.45±0.01 3.87±0.76 4.7710.89 0.6110.01 0.15+0.001 4875

P. acidipropionici ATCC

0.43±0.02 3.44±0.0.3 7.8310.1 0.6410.01 0.18+0.02 55737

P. acidipropionici F3E8 0.55±0.02 3.51+0.39 5.7611.33 0.8410.02 0.2610.01

*Range of time: 22.5 - 52 h

Table 3 Genomic and kinetic features of the propionibacteria strains

No. of Maximum

Genome size

Strain Project Contigs %GC CDS Specific growth

(Mb)

• rate (h ^"1 )

P. acidipropionici ATCC

3.66 1 68.8 3362 0.15+0.004

4875 ¹

P. acidipropionici ATCC

3.71

55737* • 1 68.7 3406 0.18+0.003

P. acidipropionici F3E8 ³ 3.63 • 1 68.7 3350 0.26+0.001

P. acidipropionici

3.58 357 68.6 3301 0.24+0.007 F3G8 ³

P. acidipropionici F3C8 ³ 3.61 333 68.6 3347 0.25+0.007

P. acidipropionici

3.57 C ⁾ 381 68.7 3296 0.25+0.014

F3H8 ³

P. acidipropionici F3E9 ³ 3.61 e 218 68.6 3327 0.25+0.007

P. acidipropionici F3F8 ³ 3.61 C) 255 68.7 3329 0.24+0.007

P. acidipropionici F3F6 ³ 3.59 € 212 68.7 3312 0.26+0.014

P. acidipropionici

3.58 € 180 68.7 3302 0.25+0.007 F3D9 ³

P. acidipropionici F3C1 ³ 3.59 400 68.7 3305 0.25+0.007

P. acidipropionici F3B9 ³ 3.62 € 447 68.7 3325 0.24+0.003

1. Genome sequence downloaded from the NCBI web page (NC_019395.1). 2. Strain in- house sequenced, assembled, and annotated. 3. Recombinants obtained from genome shuffling between P. acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737. The specific growth rate data are the average of two biological replicates in serum bottles. · Genome closed C> Genome not closed. At the end of the fermentation, F3E8 had produced 40 g/L of PA compared to the parental strains which produced 30 g/L (72 hours of fermentation) (Fig. 1A-1C) resulting in a global improvement of 25% in PA production. In terms of volumetric productivity, the new strain displayed an improvement of 27% relative to the best parental strain. In terms of yield (Yps) the new strain had a yield of 0.55 g/g compared to the 0.44 g/g yield for the best parent representing an increase of 20% in yield.

Discussion

Genome shuffling has been extensively used in industry to increase bacterial phenotypes. As reported herein, two distinct strains that displayed strong genomic differences were used as parental strains for genome shuffling (GS). Comparative analysis of strains produced by GS showed that recombination only happens between highly conserved genes of these two strains. After three rounds of GS using P.

acidipropionici ATCC 4875 and P. acidipropionici ATCC 55737, 10 mutants with increased growth rate and PA production were obtained.

EXAMPLE 2

Serum bottle P. acidipropionici F3E8 Fermentations

Material and Methods

Bacteria. P. acidipropionici F3E8 obtained from genome shuffling disclosed in Example 1 was used to perform the kinetic evaluations in serum bottles. The strain was taken from glycerol (20%, v/v) stock kept at -80 °C.

Media. The culture media (PAM) for pre-inoculum preparation consisted of yeast extract (10 g/L), trypticase soy (5 g/L), K ₂ HP0 ₄ (0.05 g/L), MnS0 ₄ (0.05 g/L), and glucose (40 g/L). Media components and the carbon source were sterilized separately for 20 min at 121 °C in an autoclave.

Serum bottles. 250 mL serum bottles with 100 mL of working volume with PAM media were used. The carbon source and media were sterilized inside the serum bottles separately and mixed under sterile conditions. After autoclaving, the head space of the serum bottles was washed with sterile nitrogen. The serum bottles were incubated in an orbital incubator at 32 °C and 90 rpm for 96 h.

Inoculum preparation. 1.5 mL Eppendorf tubes containing 1 mL of PAM media were inoculated with 0.8% (v/v) of a F3E8 glycerol stock under sterile conditions to perform the first step of the cultivation. This culture was allowed to grow for 24 hours at 32 °C. The cultures were then transferred to a 15 mL Falcon tube containing 14 mL of PAM media and allowed to grow for an additional 24 hours. 5% (v/v) of this culture was used to inoculate 250 mL serum bottles containing 100 mL of PAM media and allowed to grow for an additional 24 hours. Cells from the serum bottles in mid-exponential phase were used to inoculate other serum bottles to an OD600nm of 0.3.

Analytical methods. The optical density of the culture was measured at 600 nm using a Biochrom Libra S12 UV/Vis Spectrophotometer. Organic acids and carbohydrates were quantified by ion-exclusion chromatography using an Agilent 1200 HPLC system and an Agilent Hiplex H column (300 x 7.7 mm, PL1170-6830) with a guard column (SecurityGuard Carbo-H, Phenomenex PN: AJO-4490).

Results. Fig. 2 shows the growth profile of P. acidipropionici F3E8 and Table 4 shows the corresponding fermentation performance parameters. As can be seen, the maximum specific growth rate was around 0.26 h ^"1 and the final PA production was around of 10.35 g/L. The PA conversion (Yps yield) was around 0.66 g/g. As byproducts, there were detected around 0.23 g/L of succinic acid (SA) and 1.29 g/L of acetic acid (AA).

Table 4 Parameters of the serum bottle kinetics to test growth rate and production of the mutant from genome shuffling P. acidipropionici F3E8.

Strain PA (g/L) SA (g) AA (g) Yps(g/g) μ (1/ ^η )

P. acidipropionici F3E8 10.35±0.070 0.23±0.005 1.29+0.044 0.66±0.0023 0.26±0.005

Discussion. The results indicate that the efficiency of P. acidipropionici F3E8 to convert PA from glucose is above 0.6 g/g.

EXAMPLE 4

pH and propionic acid tolerance in P. acidipropionici ATCC 4875 and P.

acidipropionici F3E8 To compare acid tolerance between P. acidipropionici ATCC 4875 and P. acidipropionici F3E8, the cells were grown in a 96-well plates at different pH values (6.5, 5.5, and 4.5) and different propionic acid (PA) concentrations (100, 50, 25, 12.5, 6.25, 3.125, 1.56, 0.78, and 0). The growth was constantly monitored in a micro-plate reader for three days.

Material and Methods

Bacteria. P. acidipropionici ATCC 4875 and P. acidipropionici F3E8 were used. The strains were kept at - 80 °C using glycerol (20%, v/v) as cryoprotector.

Media. The culture media (PAM) for pre-inoculum preparation consisted of yeast extract (10 g/L), trypticase soy (5 g/L), K ₂ HP0 ₄ (0.05 g/L), MnS0 ₄ (0.05 g/L), and glucose (40 g/L). Media components and the carbon source were sterilized separately for 20 min at 121 °C in an autoclave. The pH was adjusted using 5 M HC1 and sodioum propionate was used to prepare a stock of 100 g/L of PA.

Inoculum preparation. 1.5 mL Eppendorf tubes containing 1 mL of PAM media were inoculated with 0.8% (v/v) of an F3E8 glycerol stock under sterile conditions to perform the first step of the cultivation. This culture was allowed to grow for 24 hours at 32 °C. The cultures were transferred to a 15 mL Falcon tube containing 14 mL of PAM media and allowed to grow for an additional 24 hours. Cells in mid-exponential phase were used to inoculate the micro-plates to an

OD600nm from 0.2 to 0.3.

Micro kinetics. 96 well-plates were used to test the growth rate of the strain at different acidic conditions. The working volume of each well was 100 uL. The kinetics were studied in duplicate. Three pHs (6.5, 5.5, and 4.5) were tested at different PA acid concentrations (100, 50, 25, 12.5, 6.25, 3.125, 1.56, 0.78, and 0). First, respective rows were filled with 100 uL of PAM at the different pHs. The PA concentrations were established by double dilution with the stock PA solution at 100 g/L. The micro-plates were incubated in a micro-plate reader OMEGA Fluostar adapted to maintain anaerobic conditions. The micro-plate was incubated at 32 °C for 72 h. Finally, the maximum specific growth rates were calculated using GrowthRates program (Hall et al., Mol. Biol. Evol., 2014, 31, 232-238).

Results. Figs. 3A-3B show the specific growth rate profile at different acidic conditions for the strains. As can be seen in the charts, the growth rate was inhibited in both strains as the PA concentration was increased. Nevertheless, the inhibition pattern was more severe for the strain P. acidipropionici ATCC 4875 (Fig. 3A) than for the mutant strain P. acidipropionici F3E8 (Fig. 3B). The minimum inhibitory condition (MIC) for P. acidipropionici F3E8 was pH 4.5 and PA 25 g/L meanwhile the MIC for P. acidipropionici ATCC 4875 was pH 6.5 and 50 g/L.

EXAMPLE 5

Propionibacteria sequencing, de novo assembly, and annotation.

The genome of P. acidipropionici ATCC 4875 has previously been sequenced and annotated (Parizzi et al., BMC Genomics, 2012, 13, 562). P. acidipropionici ATCC 55737 and 10 GS strains were sequenced using TruSeq Illimina 300 base pair paired end sequencing. To close the genome, the recombinant P. acidipropionici F3E8 and the parental P. acidipropionici ATCC 55737 were also sequenced using PacBio RS II. RAST and the SEED viewer were used for genome annotation (Aziz et al., BMC Genomics, 2008, 9, 75; Overbeek et al., Nucleic Acids Res., 2005, 33, 5691-5702).

P. acidipropionici ATCC 55737 has a genome of 3.71 Mb, a GC content of 68.7%, 3406 CDS (42% in a subsystem), and 65 RNAs. The new strain from GS named P. acidipropionici F3E8 has a size of 3.63 Mb, 68.7% of GC content, 3350 CDS (43% in a subsystem), and 73 RNAs. The other nine strains from GS have similar genome features as the strain P. acidipropionici F3E8 (Table 3). As can be seen, the genome size of the strains from GS is similar to the parental strain P.

acidipropionici ATCC 4875. Surprisingly, the number of RNAs in the recombinants is 11% higher than the parentals strains. Genome association between P. acidipropionici ATCC 4875 and P.

acidipropionici ATCC 55737.

In order to elucidate the relationship between phenotype and genotype, a systematic genomic comparison between P. acidipropionici ATCC 4875 and P.

acdipropionici ATCC 55737 was performed. These two strains display distinct specific growth rates. First, a gene-gene comparison to determine gene presence, absence, and similarity was performed. This comparison suggests that P.

acidipropionici ATCC 4875 has 345 unique genes and P. acidipropionici ATCC 55737 has 423 unique genes (E. value <0.0001). All the genes involved in PA production were conserved within 98 %. (E-value <0.0001). PA is produced through the Wood-Werckman cycle which involves ten genes: methylmalonyl-CoA carboxyl- transferase (two subunits), malate dehydro- genase, fumarate hydratase, succinate dehydrogenase (two subunits), propionyl-Co A: succinate CoA-transferase, methyl- malonyl-CoA epimerase and methylmalonyl-CoA mutase (two subunits).

Major changes between the two strains were observed for the subsystem "prophages". Prophages are mobile elements which help bacteria cope with adverse environmental conditions such as sub-lethal concentration of antibiotics or withstanding osmotic, oxidative, and acid stresses. An increased number of prophages has previously been reported to have an effect on growth and biofilm formation. P.

acidipropionici ATCC 55737 presents 30 prophages-associated proteins with a total size of 33,000 bp, whereas P. acidipropionici ATCC 4875 has only 17 prophages- associated proteins with a total size of 20,143 bp; Out of those, 11 are shared between the two strains. Genome comparisons

In an effort to compare the new strains obtained through GS, we aligned the genomes of the new strains using Mauve; P. acidipropionici ATCC 4875 was selected as the reference. The analyses showed 38 LCBs, of which four are inversions and 11 are rearrangements. The alignment shows increased similarity to the parental P.

acidipropionici ATCC 55737 compared to P. acidipropionici ATCC 4875. To calculate similarities, the genomic distance was used to calculate differences between the recombinant strains and the parents using GGDC 2.0. This comparison suggest that the recombinants have a genomic distance of 0.1090 ± 0.0013 using as a reference P. acidipropionici ATCC 4875 and 0.0156 ± 0.0019 using P.

acidipropionici ATCC 55737 as a reference.

Variant analysis

Bowtie2 and SAMtools, or SMRT® were used to align reads and to call for variants between the new strains and the parents. SNPeff was used to annotate variations. The variant analyses were performed against P. acidipropionici ATCC 55737. Significant variations are presented in Table 1. Many of the mutations, while not in the same genes, were part of the same functional group of genes which included most notably transcriptional regulators, transport, and genes linked to acid tolerance mechanisms. For example, the strain P. acidipropionici F3B9 had a mutation in the transcriptional regulator MerR gene, whereas the strain P. acidipropionici F3D9 had two mutations in regulatory genes: one in the transcriptional regulator of the DeoR family, and the other in the sigma 54 specific transcriptional regulators of the Fis family. Several mutations related to transport were also found. For instance, the recombinant P. acidipropionici F3G9 had a mutation in the Na ⁺ /H ⁺ antiporter, whereas the strain P. acidipropionici F3E8 had a mutation in one ABC transporter binding protein and the strain P. acidipropionici F3G8 had one mutation in the ABC- type nitrate/sulfonate/bicarbonate transport system and another one in the arsenic efflux pump protein; finally P. acidipropionici F3C8 had a mutation in the

oligopeptide transport ATP-binding protein. Mutations in genes related to acid tolerance mechanisms included a mutation in the malto-oligotreahalose

trehalohydrolase in the strain P. acidipropionici F3B9 and two mutations in phophogluconate dehydrogenase decarboxylating gene in the P. acidipropionici F3F6. Other mutations included dihydrolipoamide succinyltransferase component E2 of 2- oxoglutarate dehydrogenase complex/2-oxoglutarate dehydrogenase El component and D-3-phosphoglycerate dehydrogenase and mutations in the catalase/peroxidase gene. Out of all the individual variations, only three were conserved in all 10 strains.

The first SNP was identified in the large subunit ribosomal RNA, the second one in the cation diffusion facilitator protein and the last one in the long chain fatty acyl- CoA synthetase. Having a closed genome using PacBio sequencing also allowed identifying for repeats elements in strain F3E8. An extra copy of the whole ribosomal RNA operon and one extra copy of the repressor of the arginine regulon were found in F3E8 compared to the parental strain.

Analyses of recombination mechanism leading to genomic variations

ClonalFrame is a Bayesian phylogenetic method which performs inference under evolutionary models accounting for the effect of homologous recombination (X. Didelot and D. Falush, Genetics, 2007, 175, 1251-66). To determine regions in the genome of high probability of homologous recombination ClonalFrame was used. First, variable regions between the recombinants and the parents were removed from individual Mauve alignments using the stripSubsetLCBs. This script is distributed by Mauve as a minor program to complement the principal one (Mauve). The core regions were submitted to 10,000 ClonalFrame iterations in which the first half were discarded as a burn-in (data used at the beginning of probabilistic determinations to establish stationary iterations). We detected 305 ± 42 recombination events across the recombinant genomes which corresponded to regions where variants were found. In addition, we observed that the variations in the recombinants are located in very well conserved genes between the parental strains suggesting that recombination events are responsible for the variants found in the new strains. Discussion

To understand the relationship between genotype and phenotype of the new strains generated by genome shuffling, DNA sequencing was performed for the 10 strains with the largest improvements in growth rate and PA production.

A comparison between the two parental strains revealed the presence of 30 prophages as the most significant difference between the two strains. Other studies have previously reported that the presence of cryptic prophages in bacteria can provide, among other important phenotypes, an increase in growth rate. The GS strains had 29 genes associated to prophages with 47% of them shared with the parental strain P. acidipropionici ATCC 4875. To explain the increase in growth relative to the parents, variants and multi copy variation analyses were performed. Three mutations were shared in all strains.

The first mutation is in the large subunit ribosomal RNA. The mutated ribosomal RNA was aligned against all the strains used in this study. Similar mutations have been previously reported in other GS studies. Resistance to avilamycin in Streptomyces viridochmogenes was linked to a mutation in the ribosome protein S12. The second shared mutation was found in the long chain acyl- CoA synthetase. This enzyme is a member of the ligase family that activates the breakdown of complex fatty acids. This enzyme plays a role in the physiological regulation of various cellular functions via the production of long chain acyl-CoA esters, which affect protein transport, enzyme activation, protein acylation, cell signalling, and transcriptional regulation. We speculate that the enzyme has a role in the effective use of the CoA in the Wood-Werkman cycle. The third shared mutation was in the cation diffusion facilitator. This protein is considered to be the efflux pumps that remove ions from cells using a proton antiport to drive substrate translocation across the membrane. This enzyme is likely playing an importance role in cell homeostasis. In addition to shared mutations, we found changes in genes associated with regulation, acid tolerance, and transport. For example, the histone HI -like protein HC2 has mutations in five locations in five recombinants. Similar results have been reported previously using GS in S. fradiae, where mutations in a regulatory gene were found in three mutants. During that investigation, authors also observed conservation of a mutation of a relevant gene to achieve the improved phenotype.

Pacbio sequencing allowed for closing the genome of the F3E8 strain, which in turn allowed for the detection of gene duplications which may be contributing to the improved phenotype. An extra copy of a ribosomal RNA operon was found. Changes in ribosomal RNA have been associated previously with carboxylic acids tolerance through a signal recognition particle modification. The presence of new ribosomal machinery in the improved cell would allow for a new translational program which could result in a new gene expression program. This would be analogous to strains obtained using global transcriptional machinery engineering or ribosome engineering where whole gene expression alteration results in improved phenotypes. Similarly, an extra copy of a gene encoding for a protein responsible for regulation of the arginine deiminase pathway was also found. This pathway has been linked to acid tolerance in P. acidipropionici. The arginine deiminase pathway transports arginine inside the cell, transforms arginine into citrulline and ammonia to yield ornithine and carbamyl phosphate. Finally, the carbamate kinase reversibly transforms the carbamyl phosphate and ADP into ATP, C0 ₂ , and ammonia. This system is not only useful to produce energy but also produces ammonia to alkalinize the media. Using the assembled recombinant and parents genomes, we detected around 300 recombination events across the recombinant genomes which strongly correlate with the position of the variants identified in the new strains. Variations were found in well conserved genes between the two parental strains. No variants were found in unique genes of the parental strains providing further evidence of the mechanism of recombination in GS. EXAMPLE 6

Use of Subsequent Rounds of Genome Shuffling to Further Improve Propionibacterium acidipropionici strains

In order to further enhance growth rate, pH tolerance, propionic acid productivity and yield, a further round of genome shuffling was conducted using the highest performing strain, F3E8, from the original round of genome shuffling. Four additional Propionibacterium strains were included in the genome shuffling to ensure genetic diversity: P. acidipropionici ATCC 4875, P. acidipropionici ATCC 4965, P. intermedium ATCC 14072, P.jensenii ATCC 9617.

MATERIAL AND METHODS

Bacteria. To access genomic diversity, four propionibacteria wild type strains and one mutant were used for genome shuffling. From a collection of 17 wild type strains (Stowers et al., J. Ind. Microbiol. Biotechnol., 2014, 41, 837-852), the following four strains were selected: P. acidipropionici ATCC 4875, P.

acidipropionici ATCC 4965, P. intermedium ATCC 14072, P. jensenii ATCC 9617. We also used the derivative strain, P. acidipropionici F3E8, described in the present disclosure. The strains were kept at -80 °C using glycerol (20%) as cryoprotector.

Media. The culture media for pre-inoculum preparation was PAM as described in Example 2. Agar (15 g/L) was added to prepare PAM plates. Sucrose (80 g/L) was added to perform fermentations in instrumented fermenters. Media components and the carbon source (i.e. sucrose) were sterilized separately for 20 min at 121 °C. Chemical defined medium (CDM) was used to test effects of exogenous metabolite additions. The base CDM media consisted in (mg/L): sucrose (20000), FeS0 ₄ .7H ₂ 0 (10), Fe(N0 ₃ ) ₂ .9H ₂ 0 (1), K ₂ HP0 ₄ (100), KH ₂ P0 ₄ (500), MgS0 ₄ .7H ₂ 0 (500), MnS0 ₄ (10), CaCl ₂ .6H ₂ 0 (10), NaH ₂ PO ₄ .H ₂ 0 (1597.5), CoCl ₂ .6H ₂ 0 (10),

Na ₂ HP0 ₄ (3675), arginine (200), asparagine (2000), cysteine (200), glutamine (200), histidine (200), leucine (200), methionine (200), phenylalanine (200), proline (200), serine (200), tryptophan (200), tyrosine (200), biotin (0.2), riboflavin (2), thiamine hydrochloride (1), vitamin B12 (0.2), and pantothenic acid (2).

pH gradient plate. The PA/pH gradient plate was prepared using PAM agar supplemented with 5 g/L of PA salt at either low pH (3) or near to neutral pH (6.5). To create the gradient, a square plate (100 x 100 mm) was raised 0.5 cm on the side and the agar enriched with PA at pH 6.5 was poured into the plate. Once this layer was solidified, the plate was placed horizontally and the second layer of agar supplemented with PA at pH 3 was poured. Finally, the plate was left overnight at room temperature to allow the formation of the PA/pH gradient. The pH gradient was confirmed using pH indicator strips.

Measurement of internal pH (pHi). The protocol to measure the internal pH was adapted from ( Guan et al., J. Biotechnol., 2013, 167, 56-63). The fluorescence method with 2', 7'-bis-(2-carboxyethyl)-5(and 6)-carboxyfluorescein acetoxymethyl ester (BCECF AM) was used to determine the pHi. Cells (OD600=2 for a working volume of 2 mL) were centrifuged (12,000 rpm, 1 min) and washed with 50 mM HEPES-K buffer (pH 8). The pellet was re-suspended in 2 mL of the same HEPES-K buffer and incubated with 1 uL of 1 uM BCECF AM for 20 min at 32 °C. After, the cells were washed with 50 mM potassium phosphate buffer (pH 7). The pellet was re- suspended in the HEPES-K buffer, and half of the suspension was filtered.

Fluorescence intensities were measured with a fluorescence spectrophotometer with an excitation spectrum of both 490 nm (pH sensitive) and 440 nm (pH insensitive). The emission was at 535 nm. The ratio of the emission intensity of both the suspension (S) and filtrate (F) at 490 and 440 was determined: R=(S ₄ 9o - S ₄₄ o)/(F ₄ 9 ₀ - F ₄₄ o). This ratio and a calibration curve were used to calculate the pHi of all subsequent samples.

The calibration curve was determined for each strain as follows. Valinomycin and nigericin (Sigma) were added to each strain to a final concentration of 50 uM to maintain equilibration of pHi with extracellular pH (pHex). The cultures were then incubated at 32 °C for 20 min. Cells (A600 for a working volume of 2 mL) were centrifuged, washed and re-suspended in 2 mL of a buffer at pH 4, 5, 6, 7, or 8 (50 mM citrate buffer, pH 4 and 5; 50 mM phosphate buffer, pH 6, 7 and 8). After, 1 uL of 1 uM BCECF AM was added and incubated for 20 min at 32 °C. Once incubated, the cells were washed and re-suspended using the respective buffer. Finally, half of the suspension was filtered and both the fluorescence determination and ratio calculation were performed as described above.

Analytical methods. Optical density of the culture was measured at 600 nm using a Biochrom Libra S12 UV/Vis Spectrophotometer. Organic acids,

carbohydrates, and alcohols were quantified by ion-exclusion chromatography using an Agilent 1200 HPLC system and an Agilent Hiplex H column (300 x 7.7 mm, PLl 170-6830) with a guard column (SecurityGuard Carbo-H, Phenomenex PN: AJO- 4490). Sugars and alcohols were monitored using a refractive index detector (Agilent RID, G1362A) set on positive polarity and optical unit temperature of 40 °C, while organic acids were monitored at 210 nm (Agilent MWD, G1365B). 30 uL of sample was injected into the column using an auto-sampler (Agilent HiP-ALS, G1367B) while the column temperature kept at 65 °C using a thermostatted column

compartment (Agilent TCC, G1316A). Analytes were eluted isocratically with 4 mM H ₂ S0 ₄ at 0.6 mL/min for 26 min. Chromatograms were integrated using ChemStation (Rev B.03.02[341]).

Amino acids were quantified as described by (Chacko et al., Mol. Microbiol., 2014, 93(4), 797-813). In brief, samples were diluted 1:1 with internal standards and derivatized amino acids were analysed by RP-HPLC. Derivatization was performed in a high-performance autosampler (Agilent HiP-ALS SL, G1367C). 0.5 uL of sample containing 250 uM of internal standards, sarcosine and 2-aminobutanoic acid, was added into 2.5 uL of borate buffer (0.4 N, pH 10.2, Agilent PN: 5061-3339), mixed and incubated for 20 s at 4 °C. 1 uL of OP A reagent (10 mg o-pthalaldehyde/mL in 3- mercaptopropionic acid, Agilent PN: 5061-3335) was then added to initially derivatize primary amino acids. The reaction was mixed and incubated for 20 s at 4 °C. Then 0.4 uL of FMOC reagent (2.5 mg 9-fluorenylmethyl chloroformate /mL in acetonitrile, Agilent PN:5061-3337) was added, mixed and incubated for 20 s at 4 °C to derivatize other amino acids. 45.6 uL of buffer A (40 mM Na ₂ HP0 ₄ , 0.02% NaN ₃ , pH 7.8) was added to lower the pH of the reaction prior to injecting the 50 uL reaction mixture onto an Agilent Zorbax Extend C-18 column (3.5 um, 4.6 x 150 mm, Agilent PN: 763953-902) with a guard column (SecurityGuard Gemini CI 8, Phenomenex PN: AJO-7597). The column temperature was kept at 37 °C in a thermostatted column compartment (Agilent TCC, G1316B). Chromatography was performed using an Agilent 1200-SL HPLC system, equipped with an active seal wash and a degasser (Agilent Degasser, G1379B). The HPLC gradient was 2-45% B from 0-18 min, 50- 60% B from 18.1-20 min, 100% B from 20.1-24 min, and 2% B from 24.1-27 min - using a binary pump (Agilent Bin Pump SL, G1312B). Buffer B was 45%

acetonitrile, 45% methanol, and 10% water. Flow rate was 2 mL/min. Derivatised amino acids were monitored using a fluorescence detector (Agilent FLD, G1321A). OPA-derivatised amino acids were detected at 340 _ex and 450 _em nm from 1-18 min, and FMOC-derivatised amino acids at 266 _ex and 305 _em nm from 18-27 min. Quantifications were based on standard curves derived from serial dilutions of amino acid standard (Sigma, AAS18-10ML) and amino acid supplement (Agilent, 5062- 2478) kits. Chromatograms were integrated using ChemStation (Rev B.03.02[341]).

Intracellular metabolites extraction. Cells sampled at mid-exponential phase of instrumented fermenters were used for intracellular metabolomics analyses.

Metabolite extraction was performed with 50% acetonitrile (ACN). Briefly, a cell volume corresponding to 1-20 optical density units (ODs) were harvested and centrifuged at 20,172 G for 2 min at room temperature. The supernatant was then discarded, and the pellet resuspended in 50% ACN. This solution was vortexed for 10 seconds every 2 min for three times and centrifuged for 3 min at 4 °C at 20,172 G. After, the supernatant was placed into a tube and frozen at -80 °C before being freeze- dried. Finally, the powder was resuspended in 0.5 mL of MilliQ water. Intracellular metabolites of the central carbon metabolism were analyzed by LC-MS and intracellular amino acids by HPLC (method described above). Metabolite

concentrations were standardized using dry cell weight values. The factor to convert OD600 to dry cell weight (g/L) was 0.25 for P. acidipropionici ATCC 55737 and 0.29 for P. acidipropionici WGS 7.

Inoculum preparation. Under sterile conditions, the bacteria activation was carried out in a 1.5 mL Eppendorf tube with 1 mL of PAM media inoculated with 0.8% (v/v) of a glycerol stock. This culture was allowed to grow for 24 hours at 32 °C. The cultures were transferred into a 15 mL Falcon tube containing 14 mL of PAM media and allowed to grow for an additional 24 hours at 32 °C. 5% (vol/vol) of this culture was used to inoculate 250 mL serum bottles containing 100 mL of PAM media and allowed to grow for an additional 24 hours. Cells from the serum bottles in mid-exponential phase were used to inoculate the fermenters at an initial optical density of 0.3 measured at 600 nm.

Protoplast formation, fusion, and regeneration. Protoplast formation buffer (PFB) was made from (g/L): sodium succinate (40.5), sucrose (42.75), and MgCl ₂ (1.9). PFB was dissolved in one litre of Tris-HCl 0.05 mol/L at pH 7.1 ( Guan et al., "Genome-Shuffling Improves Acid Tolerance... " in Advances in Chemistry Research, Vol 15, Chapter 8 (2012) Nova Science Publishers, Inc., pp 143- 152). Regeneration buffer (RB) was made from (g/L): yeast extract (10), trypticase soy (5), KH ₂ P0 ₄ (1.5), K ₂ HP0 ₄ (2.5), and BSA (5). pH was adjusted to 7 ( Guan et al., "Genome- Shuffling Improves Acid Tolerance..." in Advances in Chemistry Research, Vol 15, Chapter 8 (2012) Nova Science Publishers, Inc., pp 143- 152). Protoplasts were prepared as described in ( Guan et al., "Genome-Shuffling Improves Acid

Tolerance..." in Advances in Chemistry Research, Vol 15, Chapter 8 (2012) Nova Science Publishers, Inc., pp 143- 152) with some modifications as subsequently described. Cells were grown for 24 hours in PAM media supplemented with 40 g/L of sucrose and 1 % of glycine. Cells were then conditioned in PAM media containing 1% of glycine and 120 g/L of sucrose for an extra 24 hours. After at least ten generations, cells were washed twice using PBS and fixed to an OD600nm = 0.2 in a lysozyme solution containing 15 mg/mL (600,000 U/mL) in PFB. Cell walls were digested in a 125 mL flask for two hours in a shaker incubator at 120 rpm and 40 °C. Protoplasts were detected in a light microscope using the lOOx oil immersion objective and counted using a haemocytometer. When appropriate, protoplasts were regenerated in RB (pH = 7 for 48 h at 32 °C). For the protoplast fusion, the protocol from Guan et al., "Genome-Shuffling Improves Acid Tolerance... " in Advances in Chemistry Research, Vol 15, Chapter 8 (2012) Nova Science Publishers, Inc., pp 143- 152) was used with minor modifications. Protoplasts were treated with UV light for 1 min or heated at 60 °C for 2 h. Cells were mixed, centrifuged and re-suspended in 500 μL of PFB. Then 500 of PEG 6000 (80%) with 20 mmol/L CaCl ₂ was added. Fusion conditions were pH 7.1, time 30 min, and temperature 32 °C. After fusion, 5 mL of PFB was added, and the sample was centrifuged at 3500 rpm for 5 min. Protoplasts were washed two times with 5 mL of PFB and re-suspended in 1 mL of RB.

Obtaining recombinants with high PA yields. Strain diversity was created using the wild type strains: P. acidipropionici ATCC 4875, P. acidipropionici ATCC 4965, P. intermedium ATCC 14072, and P.jensenii ATCC 9617. To obtain this library, P. acidipropionici ATCC 4875 was used to perform GS with the other three wild type Propionibacterium strains separately. In total, three rounds of genome shuffling were performed with each set of strains. Cells for subsequent GS rounds were selected from the acidic side of pH/PA gradient plates. Next, another three rounds of GS were performed with the obtained library of Propionibacterium strains, the parental strains (P. acidipropionici ATCC 4875, P. acidipropionici ATCC 4965, P. intermedium ATCC 14072, P.jensenii ATCC 9617) and the previously created mutant P. acidipropionici F3E8. Finally, the new recombinants were isolated by serial dilutions in PAM media agar plates. Individual recombinants were randomly selected and screened in a 96 well plate containing 100 μΐ., of PAM media at pH 5 and 25 g/L of PA. Growth was monitored using a micro-plate reader (FLUOStar Omega, BMG Labtech, Mornington, Victoria, Australia) adapted to maintain anaerobic conditions through a constant injection of nitrogen.

The selection strategy was based on an acid tolerance improvement which was determined by an acidic ratio comparison between the new strains and the wild type - the individual ratios were calculated dividing the specific growth rate under acidic conditions over specific growth rate under non-acidic conditions. The best performing strains were scaled up to 250 mL serum bottles with a working volume of 100 mL. Serum bottles were incubated using an orbital shaker incubator (Multitron, Infors-HT, Bottmingen, Switzerland) at an agitation rate of 100 rpm (2.5 cm orbit) and a working temperature at 32 °C for 96 hours.

Instrumented fermenters. Fermentations were performed using 2 L

Applikon fermenters with a working volume of 1 L. Fermenters were equipped with probes and controllers for pH, dissolved oxygen, temperature, and agitation. The agitation rate was controlled with two Rushton impellers at 300 rpm. The pH was controlled at 6.5 using 10 M NaOH. The temperature of the culture was maintained at 32 °C using an electric jacket. Prior to inoculation, the fermenters were sparged with N ₂ for at least 15 minutes. A constant N ₂ flow was kept for the duration of the fermentation at a flow rate of 0.3 L/min. Samples for metabolomics analyses were taken in the middle exponential phase.

Growth characterization. Specific growth rate (μ) was calculated at the middle exponential phase using the logarithm method. Specific growth rate calculations of kinetics in 96 well plates were performed using the program

GrowthRates ( Hall et al., Mol. Biol. Evol., 2014, 31, 232-238. For consistency, volumetric productivity (Pv) was calculated for the same time interval (ranging from 15 to 30 hours). Yield (Yps) was calculated using the total PA produced over the total substrate consumed. PA: Acetic Acid and PA: Succinic Acid ratios were calculated using total organic acid production. The specific consumption rate of sucrose (qs) and the specific production rate of PA (qp) were calculated at the middle exponential phase multiplying specific growth rate by the linear correlations of sugar or PA with biomass. RESULTS

Obtaining strains with high PA yields

Propionibacterium strains were generated and screened using the Genome Shuffling methodology described in the Materials and Methods. After the screening, 13 strains with an apparent improvement in the acid tolerance were obtained. To confirm the results, select strains were screened in serum bottle fermentations. As can be seen in Table 5, the new strains showed diversity in growth rate and PA yield. The recombinants presented an increase in the average final PA titer of 38%, an average improvement in the PA yield of 41% and a higher average growth density of 27%. Also, the average final pH was 0.05 - 0.15 pH units lower than what was observed during studies with the wild type strain. One of the strains, WGS 7, achieved a propionic acid yield of 0.75 +/- 0.05 g/g, the highest reported to date. To validate the results of serum bottles fermentations, we selected the recombinant with the highest yield (P. acidipropionici WGS 7) for further study in 2L fermenters.

Table 5. Parameters of the fermentations in serum bottles for strains generated using Genome Shuffling.

Final PA PA:AA Yps

Strain Final pH

OD600nm (g L) (g g) (g g)

P. acidipropionici WGS.l ¹ 19.13±0.44 3.82±0.01 8.39±0.26 5.58±0.25 0.59+0.01

P. acidipropionici WGS.2 ¹ 18.07±0.92 3.77±0.01 9.64±0.12 6.85±0.02 0.63+0.02

P. acidipropionici WGS.3 ¹ 19.65±0.21 3.81±0.01 8.67+0.05 6.25±0.21 0.58+0.02

P. acidipropionici WGS.4 ¹ 19.78+0.99 3.82+0.00 8.29+0.17 4.88+0.14 0.64+0.02

P. acidipropionici WGS.5 ¹ 19.67+0.18 3.82+0.02 9.37+0.88 6.30+0.91 0.70+0.10

P. acidipropionici WGS.6 ¹ 19.03±0.52 3.82±0.01 8.80±1.08 5.35+1.04 0.65+0.03

P. acidipropionici WGS 7 ¹ 19.06+0.20 3.81+0.00 8.67+0.27 5.66+0.56 0.75+0.05

P. acidipropionici WGS.8 ¹ 20.27±0.78 3.83±0.01 8.82±0.02 5.96±0.56 0.60+0.09

P. acidipropionici WGS.9 ¹ 19.47±0.27 3.83±0.01 8.57±0.11 6.01±0.06 0.55+0.00

P. acidipropionici WGS.10 ¹ 20.46±0.37 3.85±0.01 8.20±0.05 5.27±0.07 0.53+0.02

P. acidipropionici WGS.l l ¹ 19.69±0.58 3.83±0.01 8.81±0.98 5.89±0.46 0.57+0.08

P. acidipropionici WGS.12 ¹ 19.29±1.23 3.83±0.00 8.64+0.49 6.30±1.21 0.57+0.05

P. acidipropionici WGS.13 ¹ 18.14+0.40 3.87+0.01 8.13+0.26 6.43+0.96 0.53+0.02

P. acidipropionici F3E8 ² 16.04+0.20 3.92+0.01 6.95+0.18 5.26+0.29 0.53+0.02

1. New strains from Genome Shuffling. 2.

Reference strain. PA: Propionic acid. PA:AA: Propionic acid/ Acetic acid ratio. Yps: propionic acid conversion per sugar consumed on a gram basis. The data represent the average of two biological replicates. The fermentation time was 96 h.

Table 6. Kinetic parameters of 2 L fermentations of the parental strains P. acidipropionici ATCC 55737, P. acidipropionici F3E8, and the strain from GS P. acidipropionici WGS 7.

Strain Yps (g/g) PA:SA (g/g) PA:AA (g/g) *P _v {g/L.h) ΔρΗ Final PA (g/L)

P. acidipropionici

0.45±0.028 6.28±0.940 2.95±0.35 0.530±0.212 0.43±0.035 26.28±1.880

ATCC 55737

P^aadipropionia ₀ .62±0.014 6.19±0.233 5.45±0.410 0.955±0.148 1.27±0.099 44.21±0.933

*Range of time: 15 - 30 h

ΔρΗ: pHi - pHext (pHi: internal pH; pHext: external pH)

As can be seen in Table 6 and Fig. 4A-4D, the new strain (P. acidipropionici WGS 7) presented a remarkable improvement in its ability to produce PA. The yield was increased by 37%, the PA: AA ratio around 85%, and the volumetric productivity around 80%. In addition, the internal ΔρΗ (pHi - pHext)in the middle exponential phase of the fermentations; as can be seen in Table 6, was found to be ΔρΗ (pHi - pHext) 1.27 in WGS 7 and 0.43 in the wild type (ATCC 55737). It can also be seen from Fig. 4D that the amino acid consumption rates were notably different between WGS 7 and ATCC 55737.

EXAMPLE 7 Genomic and Metabolomic Characterization of the WGS 7 Mutant Obtained from Genome Shuffling

In order to better understand the causes for improved performance of the WGS 7 strain, a series of genomic, metabolomics and transcriptiomic analysis were performed to characterize the strain relative to parental strains. MATERIAL AND METHODS

RNA extraction, sequencing, and analyses

Cells sampled at mid-exponential phase from instrumented fermenters were used for RNA extraction. 50 ODs were harvested and centrifuged at 4000g for 10 min at room temperature. The supernatant was removed, and 5 mL of RNAlater ^® reagent was added to the pellet. After 8-24 h of incubation at 4 °C, the RNAlater ^® was removed by centrifugation and the pellet stored at -80 °C for further use. RNeasy ^® Mini Kit (Qiagen) was used to extract the RNA and the RNA Clean and Concentrator- 25 Kit (Zymo) to clean it. Next, the RNA was enriched depleting the ribosomal RNA with the Ribbo-Zero Magnetic Kit (Illumina). The samples were cleaned and concentrated with the RNA Clean and Concentrator-5 Kit (Zymo). The quality of the RNA was evaluated by a Bioanalyzer (Agilent 2100). Finally, the samples were sequenced using the illumina platform 100 bp PE. Tophat, Cufflinks, and CuffDiff were used to align the RNAseq reads against the reference genome P. acidipropionici ATCC 55737 (Flores et al., Genome Announc, 2016, 4 (3), 00248-16), normalize and annotate the transcripts, and evaluate differential expression, respectively (Trapnell et al., Nat. Protoc, 2012, 7(3), 562-78). DNA-sequencing, de novo assembly and annotation.

Genomic DNA of the different Propionibacterium strains were extracted using PureLink genomic DNA mini kit (Invitrogen Cat. No. Kl 820-01) and quantified using Nanodrop 1000 (Thermo Scientific) and Qubit dsDNA BR assay kit (Life Technologies Cat. No. Q32850). The quality of the DNA was determined by running a 1% agarose gel with DNA gel stain SYBR™ safe (Life Technologies Cat. No. S33102). The gel was visualized in a ChemiDoc MP system (Bio-Rad). PacBio was used to obtain the complete genome of new recombinant strain. The PacBio library preparation was performed using the protocol for 20 Kb selected with the BluePippin system. The library was prepared using the P6-C4 chemistry from Pacbio and was loaded using magnetic beads. The genome assembly was performed with the SMRT portal. This portal was also used to align the reads and call the variants of the sequenced genomes. Finally, the RAST and SEED viewer servers were used to respectively annotate and visualize the assembled genome ( Aziz et al., BMC

Genomics, 2008, 9, 75; et al., Methods Mol. Biol., 2013, 985, 17-45).

RESULTS

P. acidipropionici WGS 7 sequencing, de-novo assembly, annotation, and variant detection

The WGS 7 genome was annotated using the RAST server ( Aziz et al., BMC

Genomics, 2008, 9, 75). The genome was found to have a size of 3.61 Mb with 3333 CDS and 65 RNAs. After mapping the WGS 7 genome against the wild type strain P. acidipropionici ATCC 55737, 17 SNPs and 7 indels were identified as shown in Table 7. The annotation of these variants rendered 10 hypothetical proteins and 14 proteins inside of a subsystem (Table 7). Three mutations that seem likely to impact propionic acid production are SNP C2293187T in the ABC polar amino acid transporter gene, SNP C1487806T in the Cytochrome C biogenesis gene and SNP G1917729A in the promoter of the ABC multiple sugar transporter gene. Table 7. Variants and copy number variation found in strain P. acidipropionici WGS 7 relative to P. acidipropionici ATCC 55737.

Genome

Type Reference Alternate Gene function Remark

coordinate*

294317 SNP CC TT ATP-dependent DNA helicase SC05184 -

371247 INDEL A delA Hypothetical protein -

567004 INDEL G delG 2-dehydropantoate 2-reductase -

1105158 INDEL C delC Hypothetical protein -

1344547 SNP G A Cobalt-zinc-cadmium resistance protein -

1487806 SNP C T Cytochrome c-type biogenesis protein CcdA (DsbD analog) -

1605833 INDEL G delG Putative integral membrane protein Intergenic

1685121 SNP C T Hypothetical protein -

1799226 INDEL - insG Conserved membrane protein, putative permease -

1913841 SNP G T Hypothetical protein -

1917729 SNP G A Multiple sugar ABC transporter, substrate-binding protein Promoter

2190342 SNP C T Myo-inositol 2-dehydrogenase -

2293187 SNP C T Amino acid ABC transporter, ATP-binding protein -

2312506 INDEL - insGCAC Hypothetical protein Intergenic

2360492 SNP C T UDP-glucose 4-epimerase -

2440725 SNP G C Hypothetical protein -

2440740 SNP G A Hypothetical protein -

2440971 SNP A C Hypothetical protein -

2505914 INDEL G delG Ribose ABC transport system, ATP-binding protein RbsA -

3092919 SNP C G Chromosome segregation ATPases -

3167868 SNP C G Hypothetical protein -

3184231 SNP T G Hypothetical protein Intergenic

3279129 SNP G A Hypothetical protein -

3335969 SNP A G LSU rRNA -

SNP: single nucleotide polymorphism; INDEL: insertion or deletion; del: deletion; ins: insertion; * Genome coordinate in P. acidipropionici ATCC 55737.

Extracellular metabolites

Extracellular metabolites were measured across 2 L fermentations with the WGS 7 and ATCC 55737 (Fig. 4A & 4B). Specific growth rate, the specific consumption rate of sucrose, the specific consumption rate of free amino acids, and specific production rate of PA were calculated at the mid-exponential phase. As can be seen in Fig. 4C, WGS 7 demonstrated a 96% increase in the specific consumption rate of sucrose and a 216% increase in the specific production rate of PA relative to ATCC 55737. Interestingly, the WGS 7 depleted the sucrose in the culture media, whereas ATCC 55737 consumed only 69% of the total sugar. This observation supports the hypothesis that the SNP G1917729A, which was identified in the promoter of the putative ABC multiple sugar transporter gene, upregulated its expression and improved sucrose consumption. Changes in the specific consumption rates of the free amino acids were also detected. As can be seen in Fig. 4D, WGS 7 had an increase in the specific consumption rate of the following amino acids: serine, arginine, tyrosine, valine, methionine, tryptophan, proline, phenylalanine, isoleucine, leucine, and lysine.

Differential RNAseq analyses

Differential expression between wild type strain P. acidipropionici ATCC 55737 and shuffled strain P. acidipropionici WGS 7 was evaluated. The analysis identified 2406 transcribed genes, of which 76 were expressed significantly (q < 0.05) different- 13 downregulated and 63 upregulated.

Of the genes that were significantly differentially expressed, the new strain presented upregulation of eight ABC sugar transporters. Among these transporters, the ABC sugar transporter XLOC_000592 had the G1917729A mutation, which is located in the promoter of the gene. The shuffled strain also presented significant upregulation of three ABC amino acid transporters and three ABC oligopeptides transporters. Inside these category of genes, the ABC amino acid transporter

XLOC_000834 presented the mutation C2293187T (see Table 7). Regarding electron transport genes, upregulation of the nitric oxide reductase XLOC_002350

(cytochrome c) was identified, which is likely related to the SNP C1487806T in the cytochrome c biogenesis gene. Interestingly, the methylglyoxal pathway was also found overexpressed in the new strain. This pathway aids to degrade the

methylglyoxal compound through the formation of L-lactate and D-lactate, which are finally converted to pyruvate (Weber et al., Microbiology+, 2005, 151(3), 707-716). In addition, significant upregulation of three genes involved in the pentose and glucoronate interconversions were identified: altronate dehydratase (uxaA), altronate oxidoreductase (uxaB), and unronate isomerase (uxaC). The oxidative reactions of the pentose phosphate pathway were found to be overexpressed. The reactions associated with acetate interconversion were also more highly expressed in WGS 7 than in ATCC 5573. The intermediary compound of acetate dissimilation or assimilation, namely acetyl-CoA, is used to enter the TCA cycle through the citrate synthase (gltA) reaction. The transcription of this gene was also found overexpressed in WGS 7. This suggests that TCA cycle was more active in the shuffled strain than in WGS 7.

Experiments on CDM to test individual effects of metabolites

Based on the genomic variant and metabolomic profiles analyses, experiments were conducted to test the effect of individual metabolites in the production of PA. The experiments were performed in serum bottles using CDM. As can be seen in Fig. 5, P. acidipropionici WGS 7 increased PA production by approximately 112%, 50%, 38%, 38%, and 25% when lactate, fumarate, proline, lysine, or arginine were supplemented to the CDM, respectively. On the contrary, the wild type strain P. acidipropionici ATCC 55737 increased the PA production by 66% only when lactate was added to the CDM media. This data is consistent with the genome sequencing data, which revealed a SNP (C2293187T) in a gene (XLOC_000834) annotated as a putative ABC transporter of polar amino acids. This seems to have resulted in increased consumption of several polar amino acids. RNAseq data confirms this assertion, as indicated by the upregulation of XLOC_000834. Further, it can be seen in Fig. 5 that the feeding of certain amino acids can substantially improve the PA titer of strain WGS 7, far more than the improvement observed with ATCC 55737.

Adding serine at 10 mM improved WGS 7 titer by 75% and adding alanine at 10 mM improved titer by 25%.

As mentioned previously, the genomics analysis of WGS 7 revealed SNP C1487806T located within the cytochrome c biogenesis gene. Correspondingly, the RNAseq analysis showed significant upregulation of the nitric oxide reductase - XLOC_000683 - (cytochrome c) (q<0.05, log2 1.30 fold), suggesting that the C1487806T mutation in the cytochrome c biogenesis is likely playing an important role in the generation of the c-type cytochromes. This mutation could also be indirectly aiding to neutralize acidic environments inside the cells, as the final nitrate degradation product is ammonia. In addition, it is reported that lactate requires an electron acceptor to be oxidized and produce pyruvate through the reaction lactate dehydrogenase. Some authors suggest that cytochrome c can act as natural electron transport in the lactate-pyruvate reaction (Ogata et al., J. Biochem., 1981, 89(5), 1423-1431; Yoshimura et al., Biochim Biophys Acta, 1977, 492, 331-339). Thus, the upregulated XLOC_000683 gene could be improving the electron transport in the lactate-pyruvate reaction. To test this hypothesis, CDM serum bottle fermentations enriched with 50 mM of exogenous lactate were conducted. Interestingly, as can be seen in Fig. 5, WGS 7 produced 63% more PA than ATCC 55737 by the addition of lactate. This suggests that the mutation C1487806T is having an important role in the electron transport chain in P. acidipropionici WGS 7. An electron transport system in P. acidipropionici is also responsible for the reduction of fumarate in the Wood Werkman Cycle ( Parizzi et al., BMC Genomics, 2012, 13, 562). Flavin, in the form of flavin adenine dinucleotide (FAD), has been identified in the fumarate reductase enzyme (Ingledew et al., Microbiol. Rev., 1984, 48(3), 222-71). Regarding this system, we found significant upregulation of the flavin reductase gene

(XLOC_001914) in our transcriptome data (q<0.05, log2 1.31). This suggests that the step fumarate-succinate is contributing to the increase in PA production, as it has been associated with ATP generation (Ingledew et al., Microbiol. Rev., 1984, 48(3), 222-71). To evaluate our hypothesis, we added exogenous fumarate (10 mM) on CDM media and grew both strains in serum bottles. As can be seen in Fig. 5, the new strain had a PA titer that was two-fold higher than the wild type strain. The fact that the latter strain did not present any benefit for the exogenous addition of fumarate strengthens our hypothesis of a probable improvement in the electron transport chain in the fumarate-succinate reaction (Fig. 5). It seems likely that modifications in the electron transport chain are also affecting the metabolite pools in the TCA cycle. The new strain only presented an increase of 1.13 folds of citrate and 1.65 folds of succinate; intracellular malate did not present any change. Contrary to our results, Guan et al. ( Guan et al., Metabolomics, 2015, 11(5), 1106-1116) found more than two levels of magnitude of intracellular citrate, succinate, fumarate, or malate in their mutant strain P. acidipropionici WSH1105 than in wild type P. acidipropionici CGMCC 1.2230. In that study, to test the individual effect of fumarate, 0-50 mM of this metabolite was added to buffered PAM media; the addition of 30 mM of fumarate increased PA production 10.52% ( Guan et al., Metabolomics, 2015, 11(5), 1106— 1116).