3-EPIMERASE

Abstract

A protein comprising a polypeptide sequence having at least 70% sequence identity to SEQ ID NO: 6, SEQ ID NO: 2 or SEQ ID NO: 4. The protein has ketose 3-epimerase activity.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is a Continuation of U.S. application Ser. No. 15/639,214, filed Jun. 30, 2017, which is a continuation of U.S. application Ser. No. 14/431,549, filed Mar. 26, 2015, now U.S. Pat. No. 9,725,707, issued Aug. 8, 2017, which is the U.S. National Phase of International Application No. PCT/GB2013/052531, filed Sep. 27, 2013, which claims priority from U.S. Provisional Application No. 61/706,338, filed Sep. 27, 2012, and British Application No. 1220554.8, filed Nov. 15, 2012. The disclosures of each of these applications are incorporated herein by reference in their entirety for all purposes.

FIELD OF THE INVENTION

The present invention relates to a protein having ketose 3-epimerase activity and a nucleic acid molecule encoding said protein. The invention also relates to a vector and a host cell comprising the nucleic acid molecule. The invention also relates to a method of synthesising allulose using the protein and to allulose produced in such a manner.

BACKGROUND OF THE INVENTION

Allulose is a “zero-calorie” sweetener and has sweetness suggested to be similar to dextrose. It also has bulking and browning properties similar to those of other sugars. The primary target market for allulose is food and beverage manufacturers that currently use dextrose, fructose or HFCS in their products and that are looking to significantly reduce calories without significantly altering other properties imparted by the sugar component, for example, bulking, browning, texture and sweetness.

Allulose is not Generally Regarded As Safe (GRAS) in the United States but there is currently a GRAS notice pending (GRN400). Allulose is present in processed cane and beet molasses, steam treated coffee, wheat plant products and high fructose corn syrup. The typical total daily intake of allulose has been estimated to be greater than 0.2 grams per day. D-allulose is the C-3 epimer of D-fructose, and the structural difference between allulose and fructose results in allulose not being metabolized by the human body and thus having zero calories. Therefore, allulose is thought to be a promising candidate as a sweet bulking agent as it has no calories and is reported to be sweet while maintaining similar properties to typical monosaccharides.

Ketose-3-epimerases can interconvert fructose and allulose. U.S. Pat. No. 8,030,035 and PCT publication no. WO2011/040708 disclose that D-psicose (an alternative name for allulose) can be produced by reacting a protein derived from Agrobacterium tumefaciens, and having psicose 3-epimerase activity, with D-fructose.

US patent publication no. 2011/0275138 discloses a ketose 3-epimerase derived from a microorganism of the Rhizobium genus. This protein shows a high specificity to D- or L-ketopentose and D- or L-ketohexose, and especially to D-fructose and D-psicose. This document also discloses a process for producing ketoses by using the protein.

Korean patent no. 100832339 discloses a Sinorhizobium YB-58 strain which is capable of converting fructose into psicose (i.e. allulose), and a method of producing psicose using a fungus body of the Sinorhizobium YB-58 strain.

Korean patent application no. 1020090098938 discloses a method of producing psicose using E. coli wherein the E. coli expresses a polynucleotide encoding a psicose 3-epimerase.

The present invention seeks to provide an improvement in the production of allulose over existing technology. The present invention seeks to provide a ketose-3-epimerase with higher rates of conversion and volumetric productivity in a whole cell system than previously reported.

SUMMARY OF THE INVENTION

The present invention arises from the identification and characterisation of three ketose-3-epimerase enzymes, exemplary amino acid sequences of which are shown in SEQ. ID NOS. 2, 4 and 6. The ketose-3-epimerases may be used to convert fructose to allulose. These proteins had previously been identified as hypothetical proteins or as having tagatose epimerase activity. However, the present inventors have now surprisingly found that these enzymes have psicose-3-epimerase activity.

According to a first aspect of the present invention there is provided a protein comprising a polypeptide sequence having at least 70% sequence identity to SEQ ID NO: 2, SEQ ID NO: 4 or SEQ ID NO: 6, wherein the protein has ketose 3-epimerase activity.

Conveniently, the polypeptide sequence has at least 80%, 90%, 95% or 99% sequence identity, or has 100% sequence identity, to SEQ ID NO: 2, SEQ ID NO: 4 or SEQ ID NO: 6.

Advantageously, the polypeptide sequence comprises the sequence of SEQ ID NO: 13.

Preferably, the protein is immobilized on a solid substrate.

According to a second aspect of the present invention there is provided the use of a protein according to the first aspect of the invention for synthesizing allulose.

According to a third aspect of the present invention there is provided a nucleic acid molecule comprising a polynucleotide sequence encoding a protein according to the first aspect of the invention.

Advantageously, the nucleic acid molecule comprises a polynucleotide sequence which:

• • i) has at least 70%, 80%, 90%, 95% or 99% sequence identity, or has 100% sequence identity, to SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3; or
  • ii) hybridizes under highly stringent conditions to a polynucleotide having a sequence complementary to the sequence set forth in SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3.

According to a fourth aspect of the present invention there is provided a vector comprising a nucleic acid molecule according to the third aspect of the invention.

According to a fifth aspect of the present invention there is provided a host cell comprising a recombinant nucleic acid molecule according to the third aspect of the invention.

Conveniently, the host cell is a yeast, bacterium or other microorganism, or is a mammalian, plant or other cell culture.

Preferably, the host cell is E. coli.

According to a sixth aspect of the present invention there is provided allulose produced by a protein according to the first aspect of the invention.

According to a seventh aspect of the present invention there is provided a method of producing allulose comprising:

• • i) providing a protein according to the first aspect of the invention; and
  • ii) contacting the protein with a fructose substrate under conditions such that the fructose substrate is converted into allulose. A method of producing allulose is also provided by the invention, the method comprising contacting a protein according to the first aspect of the invention with a fructose substrate under conditions such that the fructose substrate is converted to allulose.

Advantageously, the protein is present in a host cell.

Alternatively, the protein is in isolated form.

Conveniently, the conditions comprise maintaining the protein and the fructose substrate at a temperature between 25° C. and 75° C., preferably between 50° C. and 60° C., more preferably between 52° C. and 55° C., more preferably 55° C.

Preferably, the conditions comprise maintaining the protein and the fructose substrate between pH 4 and pH10.

Advantageously, the conditions comprise maintaining the fructose substrate concentration between 75% and 95% (WN).

According to an eighth aspect of the present invention there is provided a nucleic acid molecule comprising a polynucleotide sequence which:

• • i) has at least 70% sequence identity to SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3; or
  • ii) hybridizes under highly stringent conditions to a polynucleotide having a sequence complementary to the sequence set forth in SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3.

The nucleic acid molecule encodes a polypeptide having ketose 3-epimerase activity. The nucleic acid molecule may be in isolated form, according to one aspect of the invention.

According to a ninth aspect of the present invention there is provided a host cell comprising a recombinant nucleic acid molecule comprising a polynucleotide sequence encoding a polypeptide having ketose 3-epimerase activity, wherein the polynucleotide sequence:

• • i) has at least 70% sequence identity to SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3; or
  • ii) hybridizes under highly stringent conditions to a polynucleotide having a sequence complementary to the sequence set forth in SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3.

According to a tenth aspect of the present invention there is provided a vector comprising a nucleic acid molecule comprising a polynucleotide encoding a polypeptide having ketose 3-epimerase activity, wherein the polynucleotide sequence:

• • i) has at least 70% sequence identity to SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3; or
  • ii) hybridizes under highly stringent conditions to a polynucleotide having a sequence complementary to the sequence set forth in SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3.

According to an eleventh aspect of the present invention there is provided a method of producing allulose comprising the steps of:

• • i) providing a vector comprising a nucleic acid molecule having a polynucleotide sequence encoding a protein having ketose 3-epimerase activity wherein the polynucleotide sequence: a) has at least 70% sequence identity to SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3; or b) hybridizes under highly stringent conditions to a polynucleotide having a sequence complementary to the sequence set forth in SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3;
  • ii) synthesising the protein having ketose 3-epimerase activity encoded by the polynucleotide sequence;
  • iii) contacting fructose with the protein having ketose 3-epimerase activity and maintaining the fructose and protein under conditions to permit the conversion of fructose to allulose; and
  • iv) at least partially purifying the allulose produced in step iii).

The terms “polypeptide”, “peptide” and “protein” are used interchangeably herein to refer to a polymer of amino acid residues. The terms also apply to amino acid polymers in which one or more amino acid residues is a modified residue, or a non-naturally occurring residue, such as an artificial chemical mimetic of a corresponding naturally occurring amino acid, as well as to naturally occurring amino acid polymers. The polypeptide may or may not be “isolated”, that is to say removed from the components which exist around it when naturally occurring.

The term “amino acid” as used herein refers to naturally occurring and synthetic amino acids, as well as amino acid analogues and amino acid mimetics that have a function that is similar to naturally occurring amino acids. Naturally occurring amino acids are those encoded by the genetic code, as well as those modified after translation in cells (e.g. hydroxyproline, gamma-carboxyglutamate, and O-phosphoserine). The phrase “amino acid analogue” refers to compounds that have the same basic chemical structure (an alpha carbon bound to a hydrogen, a carboxy group, an amino group, and an R group) as a naturally occurring amino acid but have a modified R group or modified backbones (e.g. homoserine, norleucine, methionine sulfoxide, methionine methyl sulphonium). The phrase “amino acid mimetic” refers to chemical compounds that have different structures from, but similar functions to, naturally occurring amino acids. It is to be appreciated that, owing to the degeneracy of the genetic code, nucleic acid molecules encoding a particular polypeptide may have a range of polynucleotide sequences. For example, the codons GCA, GCC, GCG and GCT all encode the amino acid alanine.

The percentage “identity” between two sequences may be determined using the BLASTP algorithm version 2.2.2 (Altschul, Stephen F., Thomas L. Madden, Alejandro A. Schïffer, Jinghui Zhang, Zheng Zhang, Webb Miller, and David J. Lipman (1997), “Gapped BLAST and PSI-BLAST: a new generation of protein database search programs”, Nucleic Acids Res. 25:3389-3402) using default parameters. In particular, the BLAST algorithm can be accessed on the internet using the URL http://www.ncbi.nlm.nih.gov/blast/.

The term “ketose-3-epimerase activity” as used herein means that an enzyme is capable of catalyzing the inversion of stereochemistry of ketoses, in particular the conversion of fructose to allulose. For example, in one embodiment, “ketose-3-epimerase activity” is defined as being the capacity of an enzyme to increase the rate of interconversion of fructose to allulose by at least 10 micromol/min per mg of added enzyme (0.1 U/mg) over a reaction mixture under the same conditions in the absence of the enzyme. In alternative embodiments an increase in rate of interconversion of fructose to allulose of at least 0.05 U/mg or 0.2 U/mg is is considered to be “ketose-3-epimerase activity”. A suitable assay for determining the activity of an enzyme in converting D-fructose into allulose is as follows. A reaction mixture comprising 1 ml D-fructose (50 g/L), Tris-HCL buffer (50 mM, pH 8.0), and 0.5 μM enzyme is incubated at 55° C. for 2 minutes.

The reaction is stopped after 10 minutes by boiling. The amount of D-allulose produced is determined by the HPLC method. One unit of enzyme activity is defined as the amount of enzyme catalysing the formation of 1 μmol of D-allulose/min at pH 8.0 and 55° C. (J. Agric. Food Chem. 2011, 59, 7785-7792).

The terms “gene”, “polynucleotides”, and “nucleic acid molecules” are used interchangeably herein to refer to a polymer of multiple nucleotides. The nucleic acid molecules may comprise naturally occurring nucleic acids (i.e. DNA or RNA) or may comprise artificial nucleic acids such as peptide nucleic acids, morpholin and locked nucleic acids as well as glycol nucleic acids and threose nucleic acids.

The term “nucleotide” as used herein refers to naturally occurring nucleotides and synthetic nucleotide analogues that are recognised by cellular enzymes.

The term “vector” as used herein refers to any natural or artificial construct containing a nucleic acid molecule in which the nucleic acid molecule can be subject to cellular transcription and/or translation enzymes. Exemplary vectors include: a plasmid, a virus (including bacteriophage), a cosmid, an artificial chromosome or a transposable element.

The term “host cell” as used herein refers to any biological cell which can be cultured in medium and used for the expression of a recombinant gene. Such host cells may be eukaryotic or prokaryotic and may be a microorganism such as a bacterial cell, or may be a cell from a cell line (such as an immortal mammalian cell line).

The term “highly stringent conditions” as used herein when referring to hybridization conditions means: at least about 6×SSC and 1% SDS at 65° C., with a first wash for 10 minutes at about 42° C. with about 20% (v/v) formamide in 0.1×SSC, and with a subsequent wash with 0.2×SSC and 0.1% SDS at 65° C. It is known in the art that hybridization techniques using a known nucleic acid as a probe under highly stringent conditions, such as those set forth in the specification, will identify structurally similar nucleic acids.

The term “allulose” as used herein refers to a monosaccharide sugar of the structure shown in Formula I. It is also known as “D-Psicose”.

The term “fructose” as used herein refers to a monosaccharide sugar having the structure shown in Formula II. Examples of fructose substrate include but are not limited to crystalline fructose and crystalline fructose greens. As used herein, “crystalline fructose greens” refers to a process stream created during fructose crystallization from the non-crystallizing portion of the crystallization mother liquor.

The term “recombinant” as used herein refers to a nucleic acid molecule or a polypeptide which is located in a non-naturally occurring context and which has been produced by artificial intervention. For example, a first polypeptide isolated from other polypeptides or linked by a peptide bond to a second polypeptide sequence having a different amino acid sequence from any polypeptide with which the first polypeptide is associated in nature is a recombinant polypeptide.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 shows the amino acid sequence of the ketose-3-epimerase from Clostridium scindens in accordance with one embodiment of the present invention (SEQ ID NO: 2).

FIG. 2 shows the amino acid sequence of the ketose-3-epimerase from Clostridium hylemonae in accordance with another embodiment of the present invention (SEQ ID NO: 4).

FIG. 3 shows the amino acid sequence of the ketose-3-epimerase from Desmospora sp. in accordance with a further embodiment of the present invention (SEQ ID NO: 6).

FIG. 4 shows the amino acid sequence of a previously known xylose isomerase from Clostridium cellulolyticum (SEQ ID NO: 8).

FIG. 5 shows a sequence comparison between the three ketose-3-epimerases shown in FIGS. 1 to 3 and three previously known ketose-3-epimerases. Completely conserved residues are highlighted.

FIG. 6 is a graph showing the rate of conversion of fructose to allulose by E. coli transformed to express an enzyme in accordance with an embodiment of the present invention and a control.

FIG. 7 shows the optimised gene sequence (SEQ ID NO: 7) encoding the amino acid sequence shown in FIG. 4 and a comparison of the optimised sequence with the original sequence.

FIG. 8 shows the optimised gene sequence (SEQ ID NO: 5) encoding the amino acid sequence shown in FIG. 3, and a comparison of the optimised sequence with the original sequence.

FIG. 9 shows the optimised gene sequence (SEQ ID NO: 1) encoding the amino acid sequence shown in FIG. 1, and a comparison of the optimised sequence with the original sequence.

FIG. 10 shows the optimised gene sequence (SEQ ID NO: 3) encoding the amino acid sequence shown in FIG. 2, and a comparison of the optimised sequence with the original sequence.

FIG. 11 is a graph showing the preparation of conversion of fructose substrate to allulose by E. coli transformed to express an enzyme in accordance with one embodiment of the present invention (the ketose-3-epimerase from Desmospora sp) at 18 L scale.

FIG. 12 is a graph showing the rate of allulose conversion by enzymes according to embodiments of the invention (CH P3E, CS P3E and DS P3E) and a known ketose-3-epimerase (CC P3E).

FIG. 13 is a graph showing the rate of allulose conversion by DS P3E in a 30 ml fixed bed reactor packed with A568 resin.

FIG. 14 is a graph showing the rate of allulose conversion by DS P3E in a 300 ml fixed bed reactor packed with A568 resin.

FIG. 15 shows the amino acid sequence of an artificial variant of the naturally occurring ketose-3-epimerase from Desmospora sp. (SEQ ID NO. 13).

BRIEF DESCRIPTION OF THE SEQUENCE LISTING

SEQ ID NO: 1 shows a gene sequence (optimised for expression in E. coli) encoding a ketose-3-epimerase from Clostridium scindens.

SEQ ID NO: 2 shows the amino acid sequence of the ketose-3-epimerase encoded by the gene sequence of SEQ ID NO. 1.

SEQ ID NO: 3 shows a gene sequence (optimised for expression in E. coli) encoding a ketose-3-epimerase from Clostridium hylemonae.

SEQ ID NO: 4 shows the amino acid sequence of the ketose-3-epimerase encoded by the gene sequence of SEQ ID NO. 3.

SEQ ID NO: 5 shows a gene sequence (optimised for expression in E. coli) encoding a ketose-3-epimerase from Desmospora sp. 8437.

SEQ ID NO: 6 shows the amino acid sequence of the ketose-3-epimerase encoded by the gene sequence of SEQ ID NO. 5.

SEQ ID NO: 7 shows a gene sequence (optimised for expression in E. coli) encoding a ketose-3-epimerase from Clostridium cellulolyticum.

SEQ ID NO: 8 shows the amino acid sequence of the ketose-3-epimerase encoded by the gene sequence of SEQ ID NO. 7.

SEQ ID NO: 9 shows the naturally occurring gene sequence encoding the ketose-3-epimerase from Clostridium scindens.

SEQ ID NO: 10 shows the naturally occurring gene sequence encoding the ketose-3-epimerase from Clostridium hylemonae.

SEQ ID NO: 11 shows the naturally occurring gene sequence encoding the ketose-3-epimerase from Desmospora sp. 8437.

SEQ ID NO: 12 shows the naturally occurring gene sequence encoding the ketose-3-epimerase from Clostridium cellulolyticum.

SEQ ID NO: 13 shows the amino acid sequence of an artificial variant of the ketose-3-epimerase of Desmospora sp. 8437.

DETAILED DESCRIPTION

The present invention relates, in general terms, to a protein comprising a polypeptide having an amino acid sequence shown in one of SEQ. ID NO. 2, 4 or 6. The source organisms of the polypeptides of SEQ. ID NOS. 2, 4 and 6 are shown in Table 1.

TABLE 1
Source Organism                 SEQ. ID NO.
Clostridium scindens ATCC 35704 2
Clostridium hylemonae DSM 15053 4
Desmospora sp. 8437             6

However, in alternative embodiments, the polypeptide sequence is not identical to that shown in SEQ. ID NOs. 2, 4 or 6 but has at least 70% sequence identity thereto. It is preferred that the polypeptide sequence has at least 80%, 90%, 95% or 99% sequence identity, or 100% sequence identity, to SEQ. ID NO. 2, 4 or 6.

For example, in one embodiment the polypeptide sequence comprises the sequence of SEQ. ID NO. 13 which has 89% sequence identity to SEQ. ID NO. 6. The polypeptide sequence has ketose 3-epimerase activity.

Thus in some embodiments, one or more amino acids of the peptides are omitted or are substituted for a different amino acid, preferably a similar amino acid. A similar amino acid is one which has a side chain moiety with related properties and the naturally occurring amino acids may be categorized into the following groups. The group having basic side chains: lysine, arginine, histidine. The group having acidic side chains: aspartic acid and glutamic acid. The group having uncharged polar side chains: aspargine, glutamine, serine, threonine and tyrosine. The group having non-polar side chains: glycine, alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, tryptophan and cysteine. Therefore it is preferred to substitute amino acids within these groups.

It is generally preferred that the polypeptide conforms with the chemistry of naturally occurring polypeptides (although it may be synthesized in vitro) but in some alternative embodiments the polypeptide is a peptidomimetic, that is to say a modification of a polypeptide in a manner that will not naturally occur. Such peptidomimetics include the replacement of naturally occurring amino acids with synthetic amino acids and/or a modification of the polypeptide backbone. For example in some embodiments, the peptide bonds are replaced with a reverse peptide bond to generate a retro-inverso peptidomimetic (see Méziére et al J Immunol. 1997 Oct. 1; 159(7):3230-7, which is incorporated herein by reference.) Alternatively, the amino acids are linked by a covalent bond other than a peptide bond but which maintains the spacing and orientation of the amino acid residues forming the polymer chain.

All such modified and unmodified polypeptides of the invention have ketose-3-epimerase polymerase activity. That is to say, the protein, when purified or expressed in a host cell, has the capacity to catalyze the conversion of fructose to allulose. Suitable conditions for testing the presence of ketose-3-epimerase activity are shown in Example 1.

The polypeptide of the invention may be contained within a whole cell or may be an isolated protein, a partially purified protein or an immobilized protein. Purification of the protein may be by standard methods such as cell disruption and filtration. Other standard methods are known to those skilled in the art.

In embodiments of the present invention, there is provided a nucleic acid molecule which comprises a polynucleotide sequence encoding a protein having an amino acid sequence with at least 70% sequence identity to SEQ. ID NO. 2, 4 or 6, where the protein has ketose-3-epimerase activity. For example, in one embodiment the nucleic acid molecule comprises a sequence encoding the polypeptide sequence of SEQ. IN NO. 13.

In addition to the sequence specifically encoding the protein of the invention, the nucleic acid molecule may contain other sequences such as primer sites, transcription factor binding sites, vector insertion sites and sequences which resist nucleolytic degradation (e.g. polyadenosine tails). The nucleic acid molecule may be DNA or RNA and may include synthetic nucleotides, provided that the polynucleotide is still capable of being translated in order to synthesize a protein of the invention.

As described above, the amino acid sequence of the protein of the present invention may differ from the specific sequences disclosed herein. In preferred embodiments, the nucleic acid molecule comprises a polynucleotide having the sequence of SEQ. ID NO. 1, 3 or 5, which has been optimised by expression in E. coli host cells. In alternative embodiments, the polynucleotide sequence has at least 70% sequence identity to any one of SEQ. ID NO. 1, 3 or 5 and encodes a protein which has ketose-3-epimerase activity. It is preferred that the polynucleotide sequence has at least 80%, 90%, 95% or 99% sequence identity, or 100% sequence identity, to one of SEQ. ID NO. 1, 3 or 5. In alternative embodiments the nucleic acid molecule comprises a polynucleotide sequence which hybridizes under highly stringent conditions to a polynucleotide having a sequence complementary to the sequence set forth in SEQ ID NO: 1, 3 or 5 and which encodes a protein which has ketose-3-epimerase activity. In some embodiments, there is provided a nucleic acid molecule comprising a polynucleotide having the sequence of SEQ. ID NO. 9, 10 or 11, which are the naturally occurring sequences of the enzymes.

In some embodiments, the nucleic acid molecule forms part of a vector such as a plasmid. In addition to the nucleic acid sequence described above, the plasmid comprises other elements such as a prokaryotic origin of replication (for example, the E. coli OR1 origin of replication) an autonomous replication sequence, a centromere sequence; a promoter sequence, upstream of the nucleic acid sequence, a terminator sequence located downstream of the nucleic acid sequence, an antibiotic resistance gene and/or a secretion signal sequence. A vector comprising an autonomous replication sequence is also a yeast artificial chromosome.

In some alternative embodiments, the vector is a virus, such as a bacteriophage and comprises, in addition to the nucleic acid sequence of the invention, nucleic acid sequences for replication of the bacteriophage, such as structural proteins, promoters, transcription activators and the like.

The nucleic acid molecule of the invention may be used to transfect or transform host cells in order to synthesize the protein of the invention. Suitable host cells include prokaryotic cells such as E. coli and eukaryotic cells such as yeast cells, or mammalian or plant cell lines. Host cells are transfected or transformed using techniques known in the art such as electroporation; calcium phosphate base methods; a biolistic technique or by use of a viral vector.

After transfection, the nucleic acid molecule of the invention is transcribed as necessary and translated. In some embodiments, the synthesized protein is allowed to remain in the host cell and cultures of the recombinant host cell are subsequently used. In other embodiments, the synthesized protein is extracted from the host cell, either by virtue of its being secreted from the cell due to, for example, the presence of secretion signal in the vector, or by lysis of the host cell and purification of the protein therefrom.

The protein of the present invention is used to catalyze the conversion of fructose to allulose. In some embodiments, the protein is present in host cells and is mixed, to form a conversion mixture, with a fructose substrate, such as borate buffered fructose substrate, at a concentration from 1 to 1000 g/L under suitable conditions, such as incubation at a temperature from 25° C. to 75° C., pH from 4 to 10. The conversion mixture may also comprise a solvent and optionally additional co-solvents (in addition to water) for example ethanol, toluene and methanol. The fructose substrate may also contain other sugars such as glucose or sucrose. The protein catalyzes a conversion of the fructose substrate to allulose. In practice, not all fructose in the conversion mixture is converted to allulose so there is typically a subsequent step of extracting and purifying the allulose through evaporation and crystallisation. Residual fructose in the mixture may be removed by yeast fermentation.

In alternative embodiments, the protein of the present invention is provided in purified form and mixed with a fructose substrate together with suitable solvent for an entirely in vitro conversion. In one embodiment the conditions are pH 4-10, a temperature between 30° C. and 70° C. and a fructose concentration of 10-95% w/v, with water as the solvent. Alternative concentration ranges for fructose include but are not limited to 20-95%, 30-95%, 40-95%, 50-95%, 60-95%, 70-95%, 75-95%. It is particularly preferred that the fructose substrate is provided at a concentration between 70 and 95%. In other preferred embodiments, the fructose concentration is 75-95%.

The conversion reaction of ketoses is usually carried out using a substrate concentration of 1-60% (w/v), preferably about 5-50%. It is a particular advantage of the present invention that the protein can be used in ketose conversion reactions under usual operating conditions but with higher fructose concentrations than have previously been used. Thus, a greater volumetric productivity is possible with the protein of the present invention.

In some embodiments, the protein of the present invention is immobilised on a solid substrate. This provides the advantage that the enzyme has a longer usage life, can be packed in a smaller fixed bed reactor, and has greater tolerance to contaminants and to fluctuations in the conditions of the process. Exemplary solid substrates include ion exchange resins and polymer encapsulations. In some embodiments, the protein of the present invention is immobilised on Duolite A568 resin. In some embodiments, the protein of the invention is immobilised on a substrate by weakly basic ion exchange (i.e. electrostatic interaction based on the charge of the protein and the charge of a substrate such as a resin). In other embodiments the protein is immobilised by non-specific binding to porous regions of a substrate such as a resin.

In another embodiment the invention relates to a method of producing allulose. The method comprises the following steps.

• • 1) Providing a vector comprising a nucleic acid molecule having at least 70% sequence identity to SEQ ID NO:1, SEQ ID NO:3 or SEQ ID NO: 5.
  • 2) Transforming competent host cells with said vector.
  • 3) Optionally, culturing the transformed host cells.
  • 4) Mixing the transformed cells with a fructose substrate and maintaining under conditions to permit conversion of fructose to allulose.
  • 5) Purifying the allulose produced using standard methods in the art, such as evaporation and crystallization.

In alternative embodiments, step 4) is omitted. Instead, the protein encoded by said nucleic acid molecule is isolated from the transformed host cell and optionally immobilized on a substrate. The protein is then mixed with the fructose substrate and maintained under conditions to permit conversion of fructose to allulose. Step 5) is then performed. In other embodiments, step 2) is omitted and the protein is instead synthesised by in vitro translation. Subsequently, the protein is isolated and mixed with the fructose substrate.

Allulose produced by the method of the present invention may be used in a product for human and/or animal consumption. In some embodiments, the product may be a food product, a beverage product, a pharmaceutical product, a nutritional product, a sports product, or a cosmetic product.

For example, when the product is a food product, the food product can be selected from the group consisting of a confectionary product, a dessert product, a cereal product, baked goods, frozen dairy products, meats, dairy products, condiments, snack bars, soups, dressings, mixes, prepared foods, baby foods, diet preparations, syrups, food coatings, dried fruit, sauces, gravies, and jams/jellies. In some embodiments, the food product may comprise allulose produced by the method of the present invention as a coating or frosting formed on the surface of the product.

Alternatively, when the product is a beverage product, the beverage product can be selected from the group consisting of a carbonated beverage, a non-carbonated beverage, fruit-flavoured beverage, fruit-juice, tea, milk, coffee, and the like.

EXAMPLES

Example 1

In this example, host cells were transformed to express one of three putative ketose-3-epimerase enzymes. Transformed host cells were tested for ketose-3-epimerase activity by incubating them with a fructose substrate.

Materials

Borate Buffer 1M pH8:

• • i) 62 g boric acid dissolved in 11 DI water
  • ii) adjust to pH8 with 10M NaOH
  • iii) store in 1 L bottle in 4° C. refrigerator

Borate Buffered Fructose Substrate:

• • i) 970 g liquid fructose (77% DS) in 50 ml Borate buffer pH8
  • ii) water up to 1 L final volume
  • iii) adjust pH to 8 with 5M NaOH

Expression Medium LB—4×2.8 L baffled shake flasks:

• • i) 10 g tryptone, 7 g NaCl and 10 g yeast in 1 L DI water
  • ii) autoclave

Methods

Three putative ketose-3-epimerase gene sequences and one control sequence were selected to be synthetically constructed by Genscript USA, Inc. The putative ketose-3-epimerase sequences encoded:

• • i) hypothetical protein CLOSCI_02526 from Clostridium scindens ATCC 35704 (accession ZP_02432281) (SEQ. ID NO. 2)
  • ii) hypothetical protein CLOHYLEM_05645 from Clostridium hylemonae DSM 15053 (accession ZP_03778576.1) (SEQ. ID NO. 4)
  • iii) D-tagatose 3-epimerase from Desmospora sp. 8437 (accession ZP_08466075) (SEQ. ID NO. 6).

The control sequence encoded a xylose isomerase protein from Clostridium cellulolyticum H10 (accession YP_002505284) (SEQ. ID NO. 8).

The genes were synthetically constructed with sequences optimized for expression in E. coli (see FIGS. 7 to 10) and each of the resulting four genes was cloned into an expression vector, pET15b. Other combinations of microorgansims and expression vectors known to one skilled in the art are expected to perform equally well.

Competent cells used for the transformation were prepared by inoculating 3 ml Lysogeny Broth (LB) with E. coli BL21 (DE3) and allowing the bacteria to propagate overnight at 37° C. 300 ml LB was inoculated with this 3 ml culture and the cells were grown at 37° C. with shaking to 0.7-1.0 OD (600). Optical densities (OD) were measured in a 1 cm cell at 600 nm wavelength on a typical spectrophotometer. The cells were chilled on ice for 10 mins and then spun down at 7500×g at 4° C. for 15 minutes. The media was poured off and the cells resuspended in 300 ml cold water. The spin was repeated and the cell resuspended in 150 ml cold water. The spin was repeated again and the cells were suspended in about 2 ml cold sterile 10% glycerol. The cells were spun down as previously and were suspended in about 2 ml cold sterile 10% glycerol. The suspension was divided into 100 μl aliquots in sterile eppendorf tubes and stored at −80° C.

The expression vectors provided by Genscript were subsequently used to transform competent E. coli BL21 (DE3) by electroporation and positive transformants were selected on ampicillin containing LB agar. 1 L LB was poured into each of four 2.8 L baffled flasks and was autoclaved. Once cool, 1 ml of 100 mg/I ampicillin was added to each flask aseptically and each flask was inoculated with 2-3 ml of the overnight culture of competent cells prepared above (1 flask per expression strain). The cells were allowed to grow for about 3 hours at 37° C. with 200 rpm shaking in order to achieve an OD of 0.8-1.5. 1 ml of a freshly prepared 1M isopropyl β-D-1-thiogalactopyranoside solution was added to each flask, the temperature was reduced to room temperature (i.e. 25-30° C.) and induction was allowed to proceed for about 5 hours. The cells were spun down at about 5000×g for 30 minutes at 4° C. and the supernatant decanted. The cell pellet was transferred to a weighed 50 ml centrifuge tube and the cell mass was recorded. The cells were resuspended in a few ml of sterile glycerol (10% w/w) and were frozen at −80° C.

The conversion activity of the cells was checked by mixing the whole cells into a borate buffered fructose substrate and analysing by HPLC using DP 1-4 method with a Ca²⁺ column. Four flasks containing 250 ml of borate buffered fructose substrate were warmed to 55° C. and the frozen cells were thawed at room temperature. The cells were pelleted at 6500×g and resuspended in DI water. 2 g (wet weight) of cells were mixed in borate buffered fructose substrate and were incubated at 55° C. with 90 rpm mixing in a 1 L baffled flask. Samples were taken at 0, 1, 2 and 5 hours, and were submitted for HPLC analysis.

HPLC analysis consisted of injection of 20 μL of a sample to be analysed at 0.1% (WN) into a chromatographic system consisting of a water mobile phase with a flow rate between 0.1 and 1.5 mL/min and a stationary phase consisting of a resin of particle size between 1 and 10 μm in the Ca²⁺ form maintained at 80° C. Peaks were detected and quantitated by a refractive index detector and qualitatively assigned based on retention time of known standards.

Results and Discussion

Three protein sequences were identified to be tested as ketose-3-epimerase proteins. The sequences of these proteins are given in FIGS. 1 to 3 (SEQ. ID NOS 2, 4 and 6). The xylose isomerase from Clostridium cellulolyticum H10, used as a control, has previously been suggested to produce allulose from fructose. Its amino acid sequence is shown in FIG. 4 (SEQ. ID NO. 8).

The amino acid sequences of these proteins were aligned with those of other known ketose-3-epimerases and the aligned sequences are shown in FIG. 5. Completely conserved residues are highlighted. There are very few conserved residues between these sequences, with fewer than 65% of the residues being conserved from one sequence to the next.

The degree of sequence identity between each of the sequences of SEQ. ID NOS. 2, 4, and 6 with each of Accession Nos NP_535228 and BAA24429 and SEQ. ID NO 8 was determined. The results are shown in Table 2. Between 40 and 63% sequence identity for the known ketose-3-epimerases was observed for each of the selected protein sequences. There was no overall strong homology based on the sequence alignment of all of the sequences. The selected proteins had genes optimized for expression by E. coli synthetically constructed and cloned into commercial expression vector pET15b by Genscript. Transformation of E. coli BL21 (DE3) was successful for each construct and frozen stocks of each strain were saved along with the expression vectors. Protein expression was carried out on the 1 L scale and whole cells were harvested as the crude catalyst. Conversion activity was checked and FIG. 6 shows the % DSB allulose produced during the experiment by the four different strains tested.

	TABLE 2
	% Identical Protein Sequence
	SEQ	A. tumefaciens	P. cichorii	C. cellulolyticum
	ID	(accession	(accession	(accession
Organism	NO.	NP_535228)	BAA24429)	NC_011898)
C. scindens	2	59	43	43
C. hylemonae	4	60	41	63
Desmospora	6	50	43	51
sp.

Expression of the three putative ketose-3-epimerases was successfully carried out and all three could successfully convert fructose into allulose, confirming that each protein is, indeed, a ketose-3-epimerase.

The most active protein was DS P3E (D-tagatose 3-epimerase from Desmospora sp. 8437, SEQ. ID NO. 6) which was capable of converting 30% of a 750 g/L fructose solution in just 2 hours utilizing 8 g of wet cell weight per litre for a volumetric productivity of 112 g/L/hour.

Example 2

In this example, the current best conditions for cell growth and conversion were carried out on an 18 L scale to determine scaleability and to produce allulose for further sensory and clinical investigation. Following the conversion, an initial clean-up step to remove fructose was carried out. This example was for the purpose of identifying the scaleability of this process, any unforeseen problems with scale-up and the amount of allulose that could reasonably be produced in a laboratory.

Materials

Isopropylthiogalactopyranoside (IPTG)

Filter sterilized aqueous ampicillin solution 100 mg/ml

Crystalline fructose greens

Liquid fructose (77% DS)

Growth media:

• • i) 25 g NaCl, 25 g Staleydex® 333, 6 g glycerol, 50 g tryptone (Difco), 60 g yeast extract (Difco), 8 g potassium phosphate dibasic and 8 g potassium phosphate monobasic in 6 l DI water
  • ii) adjust pH to ˜7.8 with Tris base (solid)
  • iii) autoclave 1 L per flask in 6×2.8 L baffled flasks with foil on top

Tris buffer 1 M pH8:

• • iv) 121 g in 1 L DI water
  • iv) adjust pH to 8 with HCl
  • iv) store in 1 L bottle in 4° C. refrigerator

Method

To propagate the cells, six overnight cultures of 5 ml LB medium supplemented with 100 μg/ml ampicillin were started. The cultures were inoculated with the E. coli production strain (BL21-DE3 pET15b-DS-P3E expressing the protein of SEQ. ID NO. 6) and allowed to grow overnight (˜16 hours) at 37° C. 6 L of growth media was prepared and autoclaved as described above, 5 ml of the overnight culture was added to each flask and this was shaken at 190 rpm at 37° C. for 4 hours. 1 mM IPTG was added to each flask by preparing a fresh 1M solution and adding 1 ml per litre to the flasks. The temperature was reduced to 25° C. with continued shaking for 14-16 hours.

In order to harvest the cells, the cultures were centrifuged at 6000 rpm for 20 minutes using floor centrifuge in Mod 322 and 1 L bottles (filled with not more than 800 ml of media). The media was decanted into a kill bucket, to which 1% by volume bleach was added, and was allowed to sit for 30 minutes. The centrifuge tubes were weighed and 3 ml DI water per gram of cells was added to the tubes. The cells were re-suspended using a spatula and vortex genie until a uniform cell slurry was obtained. The suspension was transferred to 40 ml centrifuge tubes and re-pelleted at 6500×g. The wash was decanted into the kill bucket and the cells were re-suspended in the same volume of water.

The propagation and harvesting of cells was repeated with a second batch of cells.

The crystalline fructose greens conversion substrate was prepared by warming a 5 gallon (18.9 L) bucket of crystalline fructose greens to room temperature and adding 16,506 g crystalline fructose greens to a sanitised 5 gallon (18.9 L) plastic bucket with an 18 L calibration mark. 900 ml of 1M Tris pH 8.0 prepared as above was added to the bucket, followed by water up to the 18 L calibration mark, and was mixed using an overhead mixer until homogeneous. The mixture and the unused crystalline fructose greens were returned to a cold room for storage.

The liquid fructose conversion substrate was prepared by combining 17,460 g liquid fructose (77% DS) and 500 ml of 1M Tris pH 8.0 (prepared as above) in a sanitised 5 gallon (18.9 L) plastic bucket with an 18 L calibration mark. Water was added up to 18 L calibration mark and an overhead mixer was used to mix until homogeneous.

For the whole cell conversion, 18 L of prepared crystalline fructose greens conversion substrate was heated to 55° C. in a water bath and was gently mixed with an overhead mixer at about 150 rpm. The re-suspended cell paste obtained from the cell harvesting was added to a total of 100 g wet weight of cells. After 5 hours a sample was removed and was submitted HPLC analysis. The reaction was stopped by refrigerating the entire bucket at 4° C. A sample was submitted for microbial analysis for E. coli, coliforms and TPC.

The whole cell conversion process was repeated with 18 L of prepared liquid fructose conversion substrate. 120 g of cells wet weight was used and samples were taken at 2 and 4 hours for HPLC Analysis.

Yeast fermentation was used to remove fructose from the crystalline fructose greens conversion substrate. The crystalline fructose greens conversion product was diluted with 2 volumes of water for a final concentration of ˜250 g/L of combined allulose and fructose in a total volume of 54 L. The 54 L of diluted mixture was split between four sanitised 5 gallon (18.9 L) buckets, with approximately 13 L per bucket. Two of the buckets were stored in a refrigerator. The remaining two buckets were set up with vigorous agitation from overhead mixers and aeration from 9 L/min air pumps with diffusers for approximately 0.3 VVM air flow. 120 g dry active baker's yeast (Fleishman's brand) was added to each bucket, and these were mixed and aerated for 2 days (˜36 hours) with occasional sampling for DP1-4 allulose analysis. The buckets were transferred to the cold room overnight to allow the yeast to settle. The supernatant was then transferred to two new clean sanitised buckets and the remaining yeast fraction was transferred to the two refrigerated crystalline fructose greens-containing buckets prepared above. The agitation and mixing process was repeated, followed by the removal of yeast. Following the yeast fermentation step, about 45 L of supernatant was obtained and sterile filtered into 3 clean sanitsed buckets, which were stored at 4° C. for further processing.

Results and Discussion

Approximately 220 g of BL21 (DE3) pET-15b-DS P3E cells were obtained from 12 L of culture and split into two 18 L bioconversions as described above. Thus, the total whole cell biocatalyst concentration was 5.6 g/L for the crystalline fructose greens conversion and 6.7 g/L for the liquid fructose conversion.

FIG. 11 shows that both conversions rapidly reached ˜25%, calculated as allulose as a percentage of allulose+fructose. This is slightly lower but very near the conversion level previously achieved using this cell type and similar conversion media on the small scale. Conversion had already reached 22% after just 2 hours with the liquid fructose substrate.

In each 18 L conversion, approximately 3.3 kg of allulose was produced. There did not appear to be a significant difference between the two substrates.

The scale up from 250 ml did not produce any unforeseen issues and proceeded as expected.

The microbial testing resulted in no live E. coli with a negative result and <3 coliforms per gram, and a total plate count of two. Therefore, a temperature of 55° C. combined with a high percentage of DS syrup was sufficient to kill the whole cell biocatalyst.

The bioconversion of fructose to allulose using the newly identified enzyme DS P3E was successfully scaled up to 18 L.

Example 3

An E. coli strain containing the newly identified DS P3E protein (SEQ ID NO: 6) was produced by two 10 L fermentations in a fermentation lab using a pH control feed batch culture method with glucose yeast extract media. The fermentations proceeded as expected.

During the fermentation batch growth and fed batch phases the cells grew exponentially with a doubling time of approximately 1 hr. Glucose concentration dropped from about 9 g/L to <1 g/L in about 5.5 hours (OD ˜28). During the Induction phase for enzyme production, the OD continued to rise to about 130 and then was not observed to change significantly. Harvesting of the fermentation by centrifugation resulted in 4.5 kg (10 lbs) of wet cell paste or approximately 1.1 kg (2.5 lbs) dry cell weight.

Fructose substrate (836 kg DS (dry solids) basis) was diluted to 69% DS (920 grams/L) with RO water and heated to 52° C. and pH adjusted to 7.8. Low agitation (˜50 rpms) was utilized to promote mixing throughout the reaction and the entire batch of 4.5 kg (wet paste) of expressed whole cells from above was added to the reaction and a time 0 sample was taken. This provided a 0.48 g/L biocatalyst load which is similar to the previously tested lab scale conversions, however, the substrate concentration was higher at 920 g/L. Samples were taken at 4 and 16 hrs and analyzed by HPLC.

No loss of DS was observed and no bioproducts were produced during the reaction. The reaction proceeded nearly to the equilibrium value of −30% allulose at the end of the 16 hr reaction. At 4 hrs the reaction had already proceeded to 18% conversion. The volumetric conversion rate previously obtained using 0.5 g/L biocatalyst with 750 g/L substrate (Examples 1 and 2) was 46 g/L*hr or per unit biocatalyst 92 g/L*hr/gram biocatalyst. Here, using a higher substrate concentration and slightly lower temperature (52° C. vs 55° C.), the volumetric conversion rate was 41 g/L*hr or 85 g/L*hr/gram biocatalyst (calculated using 4 hr data point). This demonstrates the remarkable flexibility of the epimerase reaction. When the reaction was completed at 16 hrs, 230 kg of allulose were present in the 28:72 mixture of allulose:fructose.

Example 4

Conversion of fructose by the four different enzymes (SEQ ID No 2, 4, 6 and 8) was compared on Tris buffered fructose substrate at 750 g/L. Cells were induced at 16° C. instead of 25-30° C. and the rate of conversion was slower than in previous experiments. To 200 mL of the substrate, 2 g wet weight of resuspended cells was added in 500 mL baffled flasks and incubated at 55° C. with 90 rpm shaking. Samples were taken at 2 hrs and 3.5 hrs for HPLC analysis. The results are shown in FIG. 12 in which CC P3E corresponds to SEQ ID NO. 8 and CH P3E, CS P3E and DS P3E correspond to SEQ ID NOs. 4, 2 and 6, respectively. In this experiment, all 4 strains expressing one of the proteins set forth in SEQ ID 2, 4, 6 or 8 appear to have approximately the same level of activity converting approximately 5% of the substrate to allulose in 3.5 hrs.

Example 5

In this example, the first trial of allulose production using immobilised enzymes was carried out. This example was for the purpose of improving enzyme utilisation.

Materials

Lyophilized enzyme powder as prepared from Codexis, Lot D13007 or D13008

• • Desmospora sp. Psicose 3-epimerase

Duolite A568 (Dow)

Amberlite XAD2 (Sigma)

Tris buffer 1M:

• • i) Prepared in water by dissolving at a concentration of 1M
  • ii) Adjusted pH to 8.0 with HCl
  • iii) Diluted to 100 mM before use

Crystalline Fructose Greens, 80% dry solids with a composition of:

• • i) 90% DSB fructose
  • ii) 7% DSB dextrose
  • iii) 3% DP2+
  • iv) Other monosaccherides

MnCl₂ (Sigma)

Method

1) Small Scale Immobilisation

In order to test efficiency of immobilisation, a small fixed bed reaction was carried out in jacketed columns of approximately 30 ml volume (11 mm×300 mm column dimensions). Both XAD2 and A568 resin was washed several times with water to remove fines (i.e. fine resin particles (i.e. broken/fractured particles) that are a by-product of resin manufacturing) and any residuals from manufacturing. 2 g of lyophilised enzyme (i.e. epimerase) was dissolved in approximately 50 ml of water and split into two aliquots. The pH was measured and determined to be 6.5. Approximately 30 ml of each resin was incubated with one aliquot of epimerase solution for approximately one hour at room temperature and light agitation. The resins were then packed into the jacketed column and peristaltic pumps were used to recycle the epimerase solution through the fixed bed for an additional two hours. The columns were then washed with 10 bed volumes of Tris buffer 100 mM pH 8.0. The effluent at this point looked clear and free of protein as measured spectrophotometrically at A280. Crystalline fructose greens feed was prepared by diluting crystalline fructose greens down to 60% DS with RO water, followed by adjusting the pH to 8.0 and then addition of 28 ppm MnCl₂ and 10 mM Tris buffer pH 8.0.

Feed was then pumped through the 30 ml fixed bed reactors at a rate of 8 bed volumes per hour (BV/h) with the jacketed columns heated with a recirculating water bath to a temperature of 57° C. The reactor effluent was collected and analysed by FT-IR to provide a relative concentration of allulose and fructose. This process was continued for a total of 5 days. The feed rate was adjusted from 8 bed volumes per hour down to 6 bed volumes per hour over the course of the test production run in order to maintain conversion rate.

2. Scale-Up 300 ml Fixed Bed Reactor

In order to test scaled-up efficiency of immobilisation, a larger fixed bed reactor was created in jacketed columns of approximately 300 ml volume (25 mm×600 mm column dimensions). Both XAD2 and A568 resin was washed several times with water to remove fines and any residuals from manufacturing. 10 g of lyophilised epimerase was dissolved in approximately 100 ml of water. Approximately 300 ml of A568 resin was packed into the 300 ml jacketed column and peristaltic pumps were used to recycle the epimerase solution through the fixed bed for approximately two hours at room temperature. The columns were then washed with 5 bed volumes of Tris buffer 100 mM, pH 8.0, at room temperature. The effluent at this point looked clear and free of protein as measured spectrophotometrically at A280. Crystalline fructose greens feed was prepared by diluting crystalline fructose greens down to 60% DS with RO water, followed by adjusting the pH to 8.0 and then addition of 28 ppm MnCl₂ and 10 mM Tris buffer pH 8.0. Feed was then pumped through the 30 ml fixed bed reactors at a rate of 8 bed volumes per hour (BV/h) with the jacketed columns heated with a recirculating water bath to a temperature of 57° C. The reactor effluent was collected and analysed by FT-IR to provide a relative concentration of allulose and fructose. This process was continued for a total of 4 days. The feed rate was adjusted from eight bed volumes per hour down to two bed volumes per hour over the course of the test production run. Additionally, the column was allowed to sit at room temperature for two weeks and then restarted to determine the epimerase stability during column storage.

Results and Discussion

The 30 ml column with XAD2 displayed no significant conversion for any of the samples checked and, therefore, no further analysis was performed. However, significant conversion was observed with Dowex A568. FIG. 13 shows the time course of reaction for the 30 ml fixed bed reactor with A568 resin. More than 4 kg of allulose was produced over the course of 120 hour fixed bed conversion containing just 1 g of epimerase. The percentage conversion gradually decreased over the course of the 120 hours and the flow rate through the column was reduced at 72 hours to compensate. Near equilibrium concentrations of allulose were produced during the reaction.

FIG. 14 displays the course of the 300 ml reaction with A568 resin. Flow rate was decreased at 24 hours due to limitations in the amount of feed available (861 of feed were used in 72 hours). More than 20 kg of allulose was produced in 72 hours from 10 g of epimerase. The conversion rate was still high at the end of 72 hours, although some decline in performance was observed.

Epimerase stability in solution has previously been determined in flask reactions. More than 90% of activity is lost within 8 hours at 53° C. In this example, the conversion was carried out at 57° C. A higher temperature is advantageous in terms of reaction rate, equilibrium ratio, and microbiological stability. In this example, significant epimerase activity remained even after 120 hours.

In the large scale reaction, which was feed limited, 20 kg of allulose was produced using 10 g of epimerase, resulting in a net epimerase dosing rate of 0.05% (m/m). In the small scale reaction, 4.8 kg of allulose was produced from 1 g of epimerase, resulting in a net epimerase dosing rate of 0.02% (m/m). In standard fructose production, immobilised glucoisomerase is used at a rate of 0.01-0.005% (m/), although this is over the course of 6-12 months of operation.

Claims

1. A protein comprising a polypeptide sequence having at least 70% sequence identity to SEQ ID NO: 6, SEQ ID NO: 2 or SEQ ID NO: 4, wherein the protein has ketose 3-epimerase activity.

2. A protein according to claim 1, wherein the polypeptide sequence has at least 90% sequence identity to SEQ ID NO: 6, SEQ ID NO: 2 or SEQ ID NO: 4.

3. A protein according to claim 1, wherein the polypeptide sequence comprises the sequence of SEQ ID NO: 13.

4. A protein according to claim 1, wherein the protein is immobilized on a solid substrate.

5. A nucleic acid molecule comprising a polynucleotide sequence encoding a protein according to claim 1.

6. A nucleic acid molecule according to claim 5, comprising a polynucleotide sequence which:

i) has at least 70% sequence identity to SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3; or

ii) hybridizes under highly stringent conditions to a polynucleotide having a sequence complementary to the sequence set forth in SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3.

7. A vector comprising a nucleic acid molecule according to claim 5.

8. A host cell comprising a recombinant nucleic acid molecule according to claim 5.

9. A host cell according to claim 8, wherein the host cell is a yeast, bacterium or other microorganism, or is a mammalian, plant or other cell culture.

10. A host cell according to claim 9, wherein the host cell is E. coli.

11. Allulose produced by a protein according to claim 1.

12. A method of producing allulose comprising contacting a protein according to claim 1 with a fructose substrate under conditions such that the fructose substrate is converted into allulose.

13. A method according to claim 12, wherein the protein is present in a host cell.

14. A method according to claim 12, wherein the protein is in isolated form.

15. A method according to claim 12, wherein the conditions comprise maintaining the protein and the fructose substrate at a temperature between 25° C. and 75° C.

16. A method according to claim 12, wherein the conditions comprise maintaining the protein and the fructose substrate between pH 4 and pH10.

17. A method according to claim 12, wherein the conditions comprise maintaining the fructose substrate concentration between 75% and 95% (WN).

18. A nucleic acid molecule comprising a polynucleotide sequence which:

i) has at least 70% sequence identity to SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3; or

ii) hybridizes under highly stringent conditions to a polynucleotide having a sequence complementary to the sequence set forth in SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3.

19. A host cell comprising a recombinant nucleic acid molecule comprising a polynucleotide sequence encoding a polypeptide having ketose 3-epimerase activity, wherein the polynucleotide sequence:

i) has at least 70% sequence identity to SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3; or

ii) hybridizes under highly stringent conditions to a polynucleotide having a sequence complementary to the sequence set forth in SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3.

20. A vector comprising a nucleic acid molecule comprising a polynucleotide encoding a polypeptide having ketose 3-epimerase activity, wherein the polynucleotide sequence:

i) has at least 70% sequence identity to SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3; or

ii) hybridizes under highly stringent conditions to a polynucleotide having a sequence complementary to the sequence set forth in SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3.

21. A method of producing allulose comprising the steps of:

i) providing a vector comprising a nucleic acid molecule having a polynucleotide sequence encoding a protein having ketose 3-epimerase activity wherein the polynucleotide sequence: a) has at least 70% sequence identity to SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3; or b) hybridizes under highly stringent conditions to a polynucleotide having a sequence complementary to the sequence set forth in SEQ ID NO: 5, SEQ ID NO: 1 or SEQ ID NO: 3;

ii) synthesising the protein having ketose 3-epimerase activity encoded by the polynucleotide sequence;

iii) contacting fructose with the protein having ketose 3-epimerase activity and maintaining the fructose and protein under conditions to permit the conversion of fructose to allulose; and

iv) at least partially purifying the allulose produced in step iii).