EP4689097A1

EP4689097A1 - Method for the preparation of lactulose

Info

Publication number: EP4689097A1
Application number: EP24713492.7A
Authority: EP
Inventors: Ana TOPLAK; Hein Jakob Wijma; Walter Cabri; Timo Nuijens; Antonio Ricci; Dirk Barend Janssen
Original assignee: Fresenius Kabi Ipsum SRL
Current assignee: Fresenius Kabi Ipsum SRL
Priority date: 2023-04-06
Filing date: 2024-03-27
Publication date: 2026-02-11
Also published as: WO2024208687A1

Abstract

The present invention relates to a method for the preparation of lactulose by enzymatic conversion of lactose. In particular, it relates to a method which utilizes a cellobiose 2-epimerase mutant variant for the enzymatic conversion of lactose into lactulose. It further relates to such cellobiose 2-epimerase variant itself.

Description

Description _{Title of Invention:} METHOD FOR THE PREPARATION OF LACTULOSE Technical Field [0001] The present invention relates to a method for the preparation of lactulose by enzymatic catalysis. In particular, it relates to a method which utilizes a cellobiose 2-epimerase variant for the enzymatic conversion of lactose into lactulose. It further relates to a cellobiose 2-epimerase variant as such, and to a composition comprising lactulose made by such a process. Background Art [0002] Lactulose (CAS No.63-42-3) is a simple disaccharide consisting of one galactose and one fructose sugar unit joined by a β-1,4-glycosidic bond. Lactulose is of interest to the pharmaceutical, nutraceutical and food industry. As a non- absorbable sugar, lactulose can be used in high doses as oral laxative for the treatment of chronic constipation or as detoxifying agent for the treatment of hepatic encephalopathy. In small amounts it is used as a prebiotic nutrient. The recent rise in lactulose demand is mainly due to its non-toxicity and to various nutraceutical applications either as a sole ingredient or in a mixture. Lactulose is present on the market in two forms: crystalline lactulose or lactulose solution in various concentrations. The substance lactulose for both products is currently produced by chemical isomerization of lactose. Lactose is a disaccharide present in milk, which is consisting of a D-Galactose and a D-Glucose, which are linked via a β-1,4-glycosidic bond. Lactulose may be obtained by treating lactose with alkaline agents such as calcium, sodium or potassium hydroxide, triethylamine or magnesium oxide and sodium hydrosulfite or heterogenous catalysts, zeolites, at high temperatures. Lactulose may also be obtained by treating lactose with agents such as borate or aluminate in combination with an alkaline catalyst or sodium aluminate, also at high temperatures. Both reactions involve a ring opening of the glucose, for an illustration see Figure 1. The high temperatures are essential to make the process economically viable a) due to the increasing solubility of lactose with increasing temperatures, b) due to the required energy for the ring opening and c) due to a certain amount of protection from microbial contamination provided by the high temperature. The alkaline catalysis greatly suffers from low yields, maximum yields of 20-30% have been reported. High temperatures (70˚-100˚C) need to be applied to make the process commercially interesting. At the same time, the high temperatures promote degradation of lactulose and formation of acidic byproducts that further inactivate the alkaline reagents and stop the isomerization process. [0003] Downstream processing of each chemical isomerization process is a complex, multi-step procedure that in case of alkaline catalysis involves additional lactose recycling steps. [0004] An enzymatic process offers an alternative approach for lactulose synthesis and its up-scaled production. Three possible enzymatic strategies have been described so far: redox isomerization by pyranose dehydrogenase/aldose reductase, transgalactosylation by glycosyl hydrolases (GH, EC 3.2.1.X) and isomerization mediated by cellobiose 2-epimerase (CE-enzyme, EC 5.1.3.11). Out of these three enzymatic strategies, only CE-enzyme mediated isomerization is a single step process utilizing a single enzyme. CE-enzymes also catalyze the competing conversion of lactose into epilactose. Epilactose (CAS No.20869-27-6) is a C2 epimer of lactose, which has the same molecular formula and weight (C12H22O11 and 342.3 g/mol, respectively) as lactose. Up to today over 20 thermophilic and mesophilic CE-enzymes have been identified in bacterial microorganisms. Many naturally occurring CE-enzymes catalyze both, lactose epimerization and isomerization, forming epilactose and lactulose products. Some of the CE-enzymes are used industrially in the production of epilactose. In particular, thermophilic CE-enzymes, like Cs-CE isolated from Caldicellulosiruptor saccharolyticum, have been reported to have high lactose isomerization activity, at high temperatures. Cs-CE (PDB: 4Z4J) has been extensively studied as a lactulose-producing catalyst. During the reaction catalysed by Cs-CE, the catalysis of the undesired epimerization reaction into epilactose occurs at the beginning of the reaction, the isomerization reaction wherein lactulose is formed occurs later in the reaction (reaction equilibrium)^. It was reported that the Cs-CE catalysed conversion of lactose yields lactulose and epilactose at 58 wt% and 15 wt% respectively (Kim et al. (2012) Bioresource Technol., 104:668-672), while another study reported that the reaction reaches steady-state at a conversion to 55 wt% lactulose, 14.5 wt% epilactose and 30.5 wt% lactose (Kuschel et al. (2016), J. Mol. Cat. B., 133:80-87). Although lactulose yield could be improved by addition of complexing salts, like borate, the amount of epilactose by-product could not be reduced. [0005] For use of lactulose as pharmaceutical agent in the final liquid lactulose product only a maximum of 8 % w/w epilactose with respect to the amount of lactulose is allowed (USP Monographs, USP40, 2017, p4763). With the (desired) isomerization rate being ten times lower than the (undesired) epimerization rate, the applicability of CE-enzymes in the industrial lactulose production is limited. Many attempts have been made to increase the isomerization activity of CE-enzymes and to engineer the selectivity especially of the Cs-CE enzyme, without success. Only one Cs-CE variant (referred to as 5xCs-CE, (PDB:4Z4L)), obtained by mutagenesis, was reported to be 3-fold faster than the wild-type Cs-CE and to produce 76 wt% lactulose, allegedly without producing any detectable amounts of epilactose (Shen et al., Food Chem., 2016; 207:60-67). [0006] However, to our knowledge this result could not be reproduced so far. It seems that under conditions applicable to an industrial process, such as high volumes, high concentrations and high temperatures, applied over an extended period of time, the 5xCs-CE variant is not selective in lactose transformation and cannot reach 76 wt% conversion to lactulose. Instead, the enzyme reached the same equilibrium as the wild-type Cs-CE, albeit faster. The increase in the reaction rate of 5xCs-CE enzyme could be confirmed, but the determined epilactose levels were in the same range as those produced during the isomerization reaction of the wild- type Cs-CE. The reaction equilibrium was reached at 57 wt% lactulose, 15 wt% epilactose and 28 wt% lactose, when performed at 50˚C. The final ratio of epilactose to lactulose was 0.26, which will be referred to as 26% of w/w with respect to lactulose. At 80˚C the values differed slightly : 57 wt% lactulose, 16 wt% epilactose, 26 wt% lactose, 1wt% other. The final ratio of epilactose to lactulose was 0.29, thereby much higher than the USP required maximum ratio of not more than of 0.08, i.e.8% w/w with respect to the amount of lactulose. [0007] There remains a need in the art to provide a more efficient and selective method for enzymatically synthesizing lactulose from lactose, especially a method which is suitable for industrial application. [0008] The enzymes and methods according to the invention solve or at least lessen some or all of the above-mentioned problems. Summary of the invention [0009] It has been surprisingly found that an efficient conversion of lactose into lactulose can be achieved by enzymatic catalysis employing cellobiose 2- epimerase variants characterized by at least one mutated histidine residue located in the enzyme’s catalytic site. [0010] Accordingly, the present invention provides a cellobiose 2-epimerase variant of a wild-type cellobiose 2-epimerase, wherein the wild-type cellobiose 2-epimerase is selected from the group consisting of Cs-CE (SEQ ID NO: 1), 5xCs-CE (SEQ ID NO: 2), Rm-CE (SEQ ID NO: 3), Ts-CE (SEQ ID NO: 4), Co-CE (SEQ ID NO: 5), Dt-CE (SEQ ID NO: 7), Ta-CE (SEQ ID NO: 8), St-CE (SEQ ID NO: 9), Bt-CE (SEQ ID NO: 10), Ec-CE (SEQ ID NO: 11), Bf-CE (SEQ ID NO: 12), Cl-CE (SEQ ID NO: 13), Dg-CE (SEQ ID NO: 14), Fj-CE (SEQ ID NO: 15), Ph-CE (SEQ ID NO: 16), Df-CE (SEQ ID NO: 17), Ha-CE (SEQ ID NO: 18), Sd-CE (SEQ ID NO: 19), Sl-CE (SEQ ID NO: 20) and Tt-CE (SEQ ID NO: 21), characterized in that at least one histidine residue selected from the group consisting of histidine residues, His 1, His 2 and His 3, located in the catalytic site of the cellobiose 2-epimerase variant is mutated. His 1, His 2 and His 3 are histidine residues that are conserved sites across the wild-types of at least all the CE enzymes listed above, and which are described as the “triplet histidine center”. [0011] The specific amino acid positions of these conserved histidine residues differ in the different cellobiose 2-epimerase amino acid sequences. But all three are located in the catalytic site of the cellobiose 2-epimerase. They are positioned in a rather conserved distance from each other. Wild-type cellobiose 2-epimerases that are considered suitable for the purpose of generating more efficiently converting variants of these enzymes by introducing a mutation are selected from the group consisting of Cs-CE (SEQ ID NO: 1), 5xCs-CE (SEQ ID NO: 2), Rm-CE (SEQ ID NO: 3), Ts-CE (SEQ ID NO: 4), Co-CE (SEQ ID NO: 5), Dt-CE (SEQ ID NO: 7), Ta-CE (SEQ ID NO: 8), St-CE (SEQ ID NO: 9), Bt-CE (SEQ ID NO: 10), Ec-CE (SEQ ID NO: 11), Bf-CE (SEQ ID NO: 12), Cl-CE (SEQ ID NO: 13), Dg-CE (SEQ ID NO: 14), Fj-CE (SEQ ID NO: 15), Ph-CE (SEQ ID NO: 16), Df-CE (SEQ ID NO: 17), Ha-CE (SEQ ID NO: 18), Sd-CE (SEQ ID NO: 19), Sl-CE (SEQ ID NO: 20) and Tt-CE (SEQ ID NO: 21). [0012] It is known at which positions in the primary structure of these enzymes the three conserved histidine residues are located. These positions are provided in Table 1 below. Accordingly, a preferred embodiment of the invention is a cellobiose 2-epimerase variant, wherein at least one of the three histidine residues positioned as defined in Table 1 is mutated. It is even more preferred that the mutation is at His 2, the second conserved histidine, which is located at the position according to the Table 2 below. [0013] The invention also provides a process for the preparation of a composition comprising lactulose, wherein the process comprises a step of contacting a solution comprising lactose with a cellobiose 2-epimerase variant according to the invention. This process is advantageous compared to the processes disclosed in the art because it results in a lactulose composition with an amount of epilactose below 14% w/w with respect to the amount of lactulose in the composition. [0014] Accordingly, the invention also provides for a composition, preferably an aqueous solution, comprising lactulose obtained by the process described above, that is characterized by a low amount of epilactose. Preferably these compositions have an epilactose amount of below 14% w/w, more preferably below 13%, 12% or 10% w/w, even more preferably below 8% w/w and most preferably below 6% w/w with respect to lactulose. [0015] Brief Description of Drawings [0016] Figure 1 is a scheme of the proposed CE enzyme catalysed mechanism of the conversion of lactose to lactulose and epilactose starting with a ring opening, followed by epimerization or isomerization via the “cis-enediol intermediate” followed by the final ring closure. The epimerization reaction yields epilactose and isomerization reaction yields lactulose. [0017] Figure 2 is a graphic illustration of the catalytic site of CE enzymes. The CE enzymes which were laid on top of each other are PDB:3WKH (Rm-CE); PDB:4Z4J (Cs-CE); PDB:5ZIG (St-CE); PDB:4Z4L (5xCs-CE) and PDB: 5ZHB (Bt-CE). Displayed are the amino acid structures that form the catalytic site. The first circled group of histidine residues comprises the His1 residues (H200 for PDB:3WKH; H188 for PDB:4Z4J; H188 for PDB: 4Z4L; H196 for PDB:5ZIG; H188 for PDB:5ZHB). The second circled group of histidine residues comprises the His2 residues: H259 for PDB:3WKH; H247 for PDB:4Z4J; H247 for PDB: 4Z4L; H253 for PDB:5ZIG; H247 for PDB:5ZHB). The third circled group of histidine residues comprises the His3 residues (H390 for PDB:3WKH; H377 for PDB:4Z4J; H385 for PDB: 4Z4L; H385 for PDB:5ZIG; H376 for PDB:5ZHB). [0018] Other residues shown are: PDB:3WKH (Rm-CE): Y124, F128, N195, R66, E203, Y389,W322, E326, E262; PDB:4Z4J (Cs-CE): Y114, F118, N184, R56, E191, Y376, W308, E312, E250; PDB:5ZIG (St-CE): Y120, F124, N192, R62, E199, Y384, W316, E320, E256; PDB:4Z4L (5xCs-CE): Y114, F118, N184, R56, E191, Y376, W308, E312, E250; PDB:5ZHB (Bt-CE): Y112, F116, N184, R54, E191, Y375, W307, E311, E250. [0019] Figure 3 is a graphic representation of lactose conversion over an extended period of time of up to 72 hrs, catalyzed by 5xCs-CE. The experimental conditions are described in Example 2 (50˚C). Full squares represent amounts of lactulose, full diamonds represent the amounts of epilactose, full triangles represent the amount of lactose, full circles represent other sugars. [0020] Figure 4 is a graphic representation of lactose conversion over an extended period of time of up to 72 hrs, catalyzed by 5xCs-CE H247L. The experimental conditions are described in Example 2 (50˚C). Full squares represent amounts of lactulose, full diamonds represent the amounts of epilactose, full triangles represent the amount of lactose, full circles represent other sugars. [0021] Figure 5 is a graphic representation of lactose conversion over an extended period of time of up to 64 hrs, catalyzed by 5xCs-CE H247D. The experimental conditions are described in Example 2 (50˚C). Full squares represent amounts of lactulose, full diamonds represent the amounts of epilactose, full triangles represent the amount of lactose, full circles represent other sugars. [0022] Figure 6 is a graphic representation of lactose conversion over an extended period of time of 120 hours, catalyzed by 5xCs-CE wild-type. The experimental conditions are described in Example 3. Full squares represent amounts of lactulose, full diamonds represent the amounts of epilactose, full triangles represent the amount of lactose, full circles represent other sugars. The steady-state is reached already after 8 h with epilactose amounts of 15 wt% epilactose. [0023] Figure 7 is a graphic representation of lactose conversion for 120 hours, catalyzed by Cs-CE wild-type. The experimental conditions are described in Example 3. Full squares represent amounts of lactulose, full diamonds represent the amounts of epilactose, full triangles represent the amount of lactose, full circles represent other sugars. The steady-state is reached after 24 h with epilactose amounts of 15 wt% epilactose. [0024] Figure 8 is a graphic representation of lactose conversion over 24 hrs, catalyzed by 5xCs-CE H247D. The experimental conditions are described in Example 2. The reaction was performed at a temperature of 80°C. Full squares represent amounts of lactulose, full diamonds represent the amounts of epilactose, full triangles represent the amount of lactose, full circles represent other sugars. [0025] Figure 9 is a graphic representation of lactose conversion over 24 hrs, catalyzed by wild-type 5xCs-CE H247H. The experimental conditions are described in Example 2. The reaction was performed at a temperature of 80°C. Full squares represent amounts of lactulose, full diamonds represent the amounts of epilactose, full triangles represent the amount of lactose, full circles represent other sugars. Description of the Invention [0026] Definitions [0027] The term “neutraceutical industry” refers to the industry that is concerned with the making, marketing and selling of nutraceuticals. Nutraceuticals are products derived from food sources that are purported to provide extra health benefits, in addition to the basic nutritional value found in foods. Products that are not classified as a pharmaceutical product, but rather a nutritional product, that claim physiological benefits. [0028] The term “catalytic site” refers to the site in an enzyme‘s tertiary structure, that interacts with the substrate(s) whose conversion or reaction the enzyme catalyzes. It is the portion of the enzyme at which the actual reaction proceeds, it consists of several residues in a spatial arrangement that permits interaction with the substrate to catalyze its reaction. CE enzymes only provide one catalytic site. A schematic representation thereof is given in Figure 2. [0029] The term “conserved” when used herein in combination with an amino acid residue or a short sequence thereof refers to an amino acid or sequence that is identical at the same position (in the primary or secondary or tertiary structure of the enzyme) across several different enzymes from different species. Within a sequence, amino acids that are important for folding, structural stability, catalytic activity or that form a binding site may be more highly conserved than other amino acids. [0030] The primary structure is comprised of a linear chain of amino acids. The secondary structure contains regions of amino acid chains that are stabilized by hydrogen bonds from the polypeptide backbone. These hydrogen bonds create alpha-helix and/or beta-pleated sheets of the secondary structure. The overall three-dimensional structure of a polypeptide is called its tertiary structure. [0031] The term wt% when used throughout the application refers to the weight percentage based on dried matter only. If a composition, such as an aqueous solution is described to have 23 wt% lactulose that is by reference to a total of all other dry matter in the composition. [0032] The term “% w/w X with respect to Y” is used to describe a value that describes the ratio of two products X and Y to each other based on their dry mass times 100. 50% w/w epilactose with respect to lactulose refers to a ratio of amount of epilactose divided by amount of lactulose of 0.5. For example, 3 mol of lactose could be converted in 1 mol of epilactose and 2 mol of lactulose, resulting in a 50% w/w epilactose with respect to lactulose. In this case this is equivalent to the ratio of the substances’ dry mass ratio, because their molecular weights are identical. [0033] The term "mutated" or “mutation” as used herein regarding proteins or enzymes - means that at least one amino acid in the wild-type or naturally occurring protein sequence has been replaced with a different amino acid, via mutagenesis of nucleic acids encoding these amino acids. Mutagenesis is a well-known method in the art, and includes, for example, site-directed mutagenesis by means of PCR or via oligonucleotide-mediated mutagenesis. The term "mutated" or “mutation” as used herein regarding genes means that at least one nucleotide in the nucleic acid sequence of that gene, has been replaced with a different nucleotide, via mutagenesis, resulting in the transcription and translation of a protein sequence with a qualitatively of quantitatively altered function or resulting in the knock-out of that gene. [0034] The term “variant of an enzyme” is used to describe an enzyme that has an amino acid sequence which is not 100% identical to the amino acid sequence of the wild-type enzyme. Enzymes may vary in their sequences to some extent without significant changes in their activity. It is known in the field which regions of the amino acid sequence are prone to cause reduced or increased specificity or efficiency if changed. If used without a sequence identity specification (such as for example “wherein the variant has a sequence identity of at least 80%”), the term is referring to variants the sequence of which differs from the wildtype sequence only insignificantly, without effecting its activity. [0035] In the present specification, a shorthand for denoting amino acid substitutions employs the single letter amino acid code of the amino acid that is substituted, followed by the number designating where in the protein amino acid sequence the substitution is made. This number is the amino acid position of the wild-type amino acid sequence. Thus, for the mutated amino acid sequence it is the amino acid position corresponding to the position with that number in the wild-type enzyme. Due to one or more other mutations at a lower position (additions, insertions, deletions, etc.) the actual position does not need to be the same. The skilled person will be able to determine the corresponding positions using a generally known alignment technique, such as NEEDLE. The number is followed by the single letter code of the amino acid that replaces the wild-type amino acid therein. For example, H221C denotes the substitution of histidine (H) at the position corresponding to position 221 to cysteine (C). X is used to indicate any other proteinogenic amino acid than the amino acid to be substituted. For example, H221X denotes the substitution of histidine at the position corresponding to position 221 to any other proteinogenic amino acid. [0036] The CE enzymes are a subset of enzymes which are part of the N-acyl-D- glucosamine-2-epimerase superfamily (AGE, EC 5.1.3.8). It is reported that certain amino acids are strictly conserved across this family of CE enzymes. [0037] In the following the enzyme names used throughout the application are referenced to their protein ID# in the UniProt database (https://www.uniprot.org/) and/or the RSCB PDB Protein Data Bank (https://www.rcsb.org/) Cs-CE (SEQ ID NO: 1) - UniProt:A4XGA6; PDB: 4Z4J 5xCs-CE (SEQ ID NO: 2) - UniProt:A4XGA6; PDB: 4Z4L Rm-CE (SEQ ID NO: 3) - UniProt:F8WRK9; PDB:3WKF Ts-CE (SEQ ID NO: 4) - UniProt:I3VXT6 Co-CE (SEQ ID NO: 5) – UniProt:D9TH31 Ra-CE (SEQ ID NO: 6) – UniProt:PODKY5; PDB:3VW5 Dt-CE (SEQ ID NO: 7) – UniProt:B8DZK4 Ta-CE (SEQ ID NO: 8) – UniProt:W9E788 St-CE (SEQ ID NO: 9) – UniProt:E0RU15; PDB:5ZIG Bt-CE (SEQ ID NO: 10) – UniProt:A0A0D0GHZ5; PDB:5ZHB Ec-CE (SEQ ID NO: 11) - UniProt:B3XZI5 Bf-CE (SEQ ID NO: 12) - UniProt:Q5LH66 Cl-CE (SEQ ID NO: 13) - UniProt:F2JR71 Dg-CE (SEQ ID NO: 14) - UniProt:F5IT64 Fj-CE (SEQ ID NO: 15) - UniProt:A5FA14 Ph-CE (SEQ ID NO: 16) - UniProt:C6XVU9 Df-CE (SEQ ID NO: 17) - UniProt:C6VW66 Ha-CE (SEQ ID NO: 18) - UniProt:A9AYF3 Sd-CE (SEQ ID NO: 19) - UniProt:Q21NF7 Sl-CE (SEQ ID NO: 20) - UniProt:D2QDA5 Tt-CE (SEQ ID NO: 21) - UniProt:C5BUF0. [0038] It is proposed that the mechanism of isomerization and epimerization by CE enzymes proceeds via the formation of a cis-enediol intermediate of the lactose (see Figure 1). It is further proposed that this reaction is catalysed at an active site present in the CE enzymes. The active site of the CE enzymes was described as an α6/ α6- barrel scaffold composed of 12 α-helices. The active site comprises the catalytic site and a binding site, where the substrate of the reaction is temporarily bound. Feng proposed that three conserved histidine residues are located in the catalytic site of the CE enzymes, that is catalyzing the conversion (Feng, Yinghui et al. (2020), Acta Crystallographica 2020, 76.Jg., No 11, p.1104–1113). Without being bound to any of these hypotheses, throughout the application the term catalytic site will be used to describe the site that comprises the three histidine residues, His 1, His 2 and His 3, also referred to as “triplet histidine center”. [0039] Even though not in all CE enzymes these three histidine residues are located at the same position of the amino acid strain, i.e. not at the same position of the peptide’s primary structure, all three histidine residues are located at conserved positions in the tertiary structure of the enzyme. This is nicely illustrated in Figure 2c) of Feng et al. (ibidem). To differentiate the positions of the histidine residues from one another these shall be referred to as first, second and third, that is His 1, His 2 and His 3, according to their occurrence in the primary structure (amino acid sequence). In Cs-CE the first histidine residue, His 1, is His188, the second histidine residue, His 2, is His247, the third histidine residue, His 3, is His377. All three are located within the catalytic site of the enzyme. Additional histidine residues which may be present in the amino acid chain, but which are not located in the catalytic site are disregarded for this numbering. [0040] The three histidine residues are completely conserved over the whole family of cellobiose 2-epimerases (CE enzymes). For the CE enzymes, these three histidine residues are even considered essential for the catalytic activity of the enzyme. This was concluded by two groups of scientists, both stating that when any one of these three conserved histidine residues in the CE enzyme Ra-CE which was found in Ruminococcus albus, was mutated the catalytic activity of the enzymes was lost completely. In Ra-CE the three histidine residues are His184, His243 and His374, according to the numbering of Ra-CE in PDB:3VW (Ito et al., Biotechnol. Lett., 2009, 31:1065-1071 for mutating His243 and His374; Fujiwara et al., FEBS Lett., 2013, 587:840-846 for mutating His184). Relying on these studies it had to be expected that by mutating any of these three histidine residues, His 1, His 2 or His 3 in a CE enzyme the catalytic activity would be abolished. [0041] The inventors found that mutating at least one of the three histidine residues in the catalytic site of an CE enzyme does not result in a complete loss of activity but results instead in a new selective isomerization enzyme with virtually no epimerase activity. In particular, the mutation of His 2 in the catalytic site of CE enzymes results in a significant change of their catalytic activity, towards a reduction of the isomerization reaction, and thereby to a significantly reduced amount of epilactose as by-product. The mutation of His 2 in the Cs-CE (His247), as well as the same mutation in the 5xCs-CE and these mutations’ effects are disclosed herein. These mutations surprisingly result in a selective isomerization enzyme which can perform the conversion of lactose into lactulose with high efficiency and without any significant formation of epilactose. [0042] It has been surprisingly found that an efficient conversion of lactose into lactulose can be achieved by enzymatic catalysis employing cellobiose 2- epimerase variants characterized by at least one mutated histidine residue located in the enzyme’s catalytic site. [0043] To create more selective enzymes for industrially suitable lactulose production, the 5xCs-CE enzyme was selected as starting enzyme and processed to insert further mutations. Site saturation libraries were produced for different selected sites at first. Those enzymes which carried a mutation at one of the three conserved histidine sites, which had been reported as being essential for the catalytic activity, were initially selected to work as negative controls. When it turned out that surprisingly these “negative controls” remained active, a site saturation library for the selected site was produced. Surprisingly, the variants which carried a mutation at the His 2 position, remained active in the lactose conversion reaction and also showed superior selectivity towards the isomerization reaction resulting in higher conversion rates to lactulose. Most importantly, the amount of epilactose detected was significantly lower than in the unmutated wild-type enzymes, even below 8% w/w with respect to lactulose. This is an important achievement because such a lactulose solution meets the limits of the allowed epilactose content in a lactulose solution according to the US pharmacopoeia without having to apply an additional purification treatment to remove the epilactose content. [0044] Accordingly, the present invention provides a cellobiose 2-epimerase variant of a wild-type cellobiose 2-epimerase, characterized in that at least one histidine residue selected from the group consisting of histidine residues, His 1, His 2 and His 3, located in the catalytic site of the cellobiose 2-epimerase variant is mutated. The wild-type cellobiose 2-epimerase is selected from the group consisting of Cs- CE (SEQ ID NO: 1), 5xCs-CE (SEQ ID NO: 2), Rm-CE (SEQ ID NO: 3), Ts-CE (SEQ ID NO: 4), Co-CE (SEQ ID NO: 5), Dt-CE (SEQ ID NO: 7), Ta-CE (SEQ ID NO: 8), St-CE (SEQ ID NO: 9), Bt-CE (SEQ ID NO: 10), Ec-CE (SEQ ID NO: 11), Bf-CE (SEQ ID NO: 12), Cl-CE (SEQ ID NO: 13), Dg-CE (SEQ ID NO: 14), Fj-CE (SEQ ID NO: 15), Ph-CE (SEQ ID NO: 16), Df-CE (SEQ ID NO: 17), Ha-CE (SEQ ID NO: 18), Sd-CE (SEQ ID NO: 19), Sl-CE (SEQ ID NO: 20) and Tt-CE (SEQ ID NO: 21). The specific positions of these conserved histidine residues differ in the different cellobiose 2-epimerases. It is a preferred embodiment wherein the histidine residue His 1 is located in the amino acid sequence starting with position 184 and ending with 208, His 2 is located in the amino acid sequence starting with position 239 and ending with 273, and His 3 is located in the amino acid sequence starting with position 372 and ending with position 405 wherein the amino acid sequences are those of the different cellobiose 2-epimerases. The distance between His 1 and His 2 varies slightly across the different species of CE enzymes, but it is commonly between 50 and 70 amino acids, in the CE enzymes disclosed herein the distance is not more than 65 amino acids and not less than 54 amino acids. The observed distance between His 2 and His 3 across the different CE enzymes is between 129 and 136 amino acids. The exact position in the primary sequence is of less relevance than the fact that the histidine is located in the catalytic site in the tertiary structure. [0045] In a preferred embodiment of the invention the at least one mutated histidine residue is the second of the three histidine residues located in the catalytic site, His 2, located at a position in between His 1 and His 3 in the amino acid sequence of the cellobiose 2-epimerase. Accordingly, it is preferred that the mutated histidine residue is located in the amino acid sequence starting with position 239 and ending with position 273. [0046] It is known at which positions in the primary structure of the CE enzymes the three conserved histidine residues are located. These positions are provided in Table 1, below. Accordingly, a preferred embodiment of the invention is a cellobiose 2-epimerase variant, wherein at least one of the three histidine residues located at the positions as defined in Table 1 is mutated. [0047] Table 1: Conserved histidine positions located in the catalytic site of different CE enzymes: CE enzyme (wild-type) amino acid position of His 1, His 2 and His 3 Cs-CE 188, 247 and 377 5xCs-CE 188, 247 and 377 Rm-CE 200, 259 and 390 Ts-CE 186, 245 and 375 Co-CE 188, 247 and 377 Dt-CE 188, 247 and 377 Ta-CE 206, 265 and 395 St-CE 196, 253 and 385 Bt-CE 188, 247 and 376 Ec-CE 184, 259 and 389 Bf-CE 190, 249 and 381 Cl-CE 190, 249 and 380 Dg-CE 185, 239 and 372 Fj-CE 192, 251 and 382 Ph-CE 189, 248 and 378 Df-CE 187, 246 and 377 Ha-CE 202, 261 and 393 Sd-CE 199, 258 and 390 Sl-CE 190, 249 and 385 Tt-CE 208, 273 and 405 [0048] It is even more preferred that the mutation is located at the second histidine residue, His 2, which is located at the position relative to the corresponding wild- type cellobiose 2-epimerase amino acid sequence as indicated in the following table (Table 2): [0049] Table 2: Conserved His 2 positions located in the catalytic site of different CE enzymes CE enzyme amino acid CE enzyme amino acid (wild-type) position of His 2 (wild-type) position of His 2 Cs-CE 247 Bf-CE 249 5xCs-CE 247 Cl-CE 249 Rm-CE 259 Dg-CE 239 Ts-CE 245 Fj-CE 251 Co-CE 247 Ph-CE 248 Dt-CE 247 Df-CE 246 Ta-CE 265 Ha-CE 261 St-CE 253 Sd-CE 258 Bt-CE 247 Sl-CE 249 Ec-CE 259 Tt-CE 273 [0050] In a preferred embodiment according to the invention the cellobiose 2- epimerase variant, is a variant of a wild-type cellobiose 2-epimerase selected from the group consisting of Cs-CE (SEQ ID NO: 1) and 5xCs-CE (SEQ ID NO: 2). The term wild-type cellobiose 2-epimerase refers to the enzymes with the sequences reported at the RCSB protein databank. Cs-CE is registered as PDB: 4Z4J and 5xCs-CE is registered as PDB:4Z4L. Therein none of the three conserved histidine residues is mutated. These correspond also to the sequences given in the sequence protocol. It is especially preferred that the CE variant according to the invention is a variant of the cellobiose 2-epimerase wild-type that is selected from the group consisting of Cs-CE (SEQ ID NO: 1), and 5xCs-CE (SEQ ID NO: 2), and the at least one mutated histidine residue of this CE variant located at positions 188, 247 and 377 of the corresponding wild-type cellobiose 2-epimerase. [0051] In a more preferred embodiment, the cellobiose 2-epimerase variant is a variant of the cellobiose 2-epimerase wild-type that is selected from the group consisting of Cs-CE (SEQ ID NO: 1) and 5xCs-CE (SEQ ID NO: 2), and the at least one mutated histidine residue of said variant is histidine 247. [0052] In an even more preferred embodiment this cellobiose 2-epimerase variant, described above has a histidine mutation that is selected from the group consisting of H247E, H247D, H247L, H247F and H247A. [0053] The most preferred cellobiose 2-epimerase variant is a variant of the cellobiose 2-epimerase wild-type 5xCs-CE with a histidine mutation that is selected from the group consisting of H247E, H247D, H247L, H247F and H247A [0054] All tested variants according to the invention showed reduced epimerization activity resulting in lower epilactose amounts formed compared to the so-called wild-type, i.e. the enzyme that is not mutated in any of the three histidine residues His 1 to His 3, and thereby resulted in a significantly reduced amount of epilactose as by-product. When the mutants of the site-saturation library at the His 2 position of 5xCs-CE, which is His247, were tested, all tested variants of 5xCs-CE (H247X) showed reduced epilactose forming activity compared to the wild-type 5xCs-CE (H247H). The variants which are best suited for an industrial lactulose production are the ones presenting with one of the preferred mutations H247E, H247D, H247L, H247F or H247A, which also showed a sufficiently high isomerization activity. These enzymes are best suited for the industrial lactulose production because they have a reduced epilactose forming activity and also a high isomerization activity. The advantageous properties remained when tested under conditions, such as high temperature, but also at high concentrations of the enzymes used. [0055] The cellobiose 2-epimerase variant according to the invention may differ in its amino acid sequence from the wild-type amino acid sequence also in a limited number of other positions. But the variant’s amino acid sequence according to the invention has at least 80%, preferably at least 85%, more preferably at least 90% or even more preferably at least 95% or even 99% sequence identity to the respective cellobiose 2-epimerase wild-type enzyme’s amino acid sequence. [0056] For the purpose of this invention, it is defined here that in order to determine the percent identity of two amino acid sequences, the complete sequences are aligned for optimal comparison purposes. To optimize the alignment between the two sequences gaps may be introduced in any of the two sequences that are compared. Such alignment is carried out over the full length of the sequences being compared. Alternatively, the alignment may be carried out over a shorter length, for example over about 20, about 50, about 100 or more nucleic acids or amino acids. The percentage identity is the percentage of identical matches between the two sequences over the reported aligned region. [0057] A comparison of sequences and determination of percent identity between two sequences can be accomplished using a mathematical algorithm. The skilled person will know several different computer programs are available to align two sequences and determine the homology between two sequences (Kruskal, J. B. (1983) An overview of sequence comparison In D. Sankoff and J. B. Kruskal, (ed.), Time warps, string edits and macromolecules: the theory and practice of sequence comparison, pp.1-44 Addison Wesley). The percent identity between two amino acid sequences can be determined using the Needleman and Wunsch algorithm for the alignment of two sequences. (Needleman, S. B. and Wunsch, C. D. (1970) J. Mol. Biol. 48, pp 443-453). The Needleman-Wunsch algorithm has been implemented in the computer program NEEDLE. For the purpose of this invention the NEEDLE program from the EMBOSS package was used (version 2.8.0 or higher, EMBOSS: The European Molecular Biology Open Software Suite (2000) Rice,P. Longden,I. and Bleasby,A. Trends in Genetics 16, (6) pp 276—277, http://emboss.bioinformatics.nl/). For protein sequences, EBLOSUM62 is used for the substitution matrix. Other matrices can be specified. The optional parameters used for alignment of amino acid sequences are a gap-open penalty of 10 and a gap extension penalty of 0.5. The skilled person will appreciate that all these different parameters will yield slightly different results but that the overall percentage identity of two sequences is not significantly altered when using different algorithms. [0058] The sequence identity between the two aligned sequences is calculated as follows: the number of corresponding positions in the alignment showing an identical amino acid in both sequences divided by the total length of the alignment after subtraction of the total number of gaps in the alignment. The identity defined as herein can be obtained from NEEDLE by using the NOBRIEF option and is labelled in the output of the program as “longest-identity”. For purposes of the invention the level of identity (homology) between two sequences is calculated according to the definition of “longest-identity” as can be carried out by using the program NEEDLE. [0059] The enzyme’s sequence can further be used as a “query sequence” to perform a search against sequence databases, for example to identify other family members or related sequences. Such searches can be performed using the BLAST programs. Software for performing BLAST analyses is publicly available through the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov). BLASTP is used for amino acid sequences. The BLAST program uses as defaults: -Cost to open gap: default = 11 for proteins -Cost to extend gap: default = 1 for proteins -Expect value: default = 10 -Wordsize: default = 28 for megablast/ 3 for proteins [0060] Furthermore, the degree of local identity between the amino acid sequence query and the retrieved homologous sequences is determined by the BLAST program. However only those sequence segments are compared that give a match above a certain threshold. Accordingly, the program calculates the identity only for these matching segments. Therefore, the identity calculated in this way is referred to as local identity. [0061] Another embodiment of the invention is a process for the preparation of a composition comprising lactulose, wherein the process comprises a step of contacting a solution comprising lactose with a cellobiose 2-epimerase variant according to the invention. [0062] Preferably the process comprises the steps of: a) providing an aqueous solution comprising lactose b) contacting said solution with a cellobiose 2-epimerase variant, and c) collecting a composition comprising the lactulose formed during step b); and wherein the cellobiose 2-epimerase variant is a variant according to the embodiments described above. Preferably the aqueous solution of step b) is free of other disaccharides. [0063] It is preferred that step b) of the process described above is performed at a pH in the range from 6 to 9, preferably from 7 to 8, and most preferably at pH 7.5. [0064] This process comprises the enzymatically catalysed conversion reaction, which takes place at certain temperatures. In principle the temperature during the enzymatic conversion reaction is not critical, as long as a temperature is chosen at which the selected CE enzyme shows sufficient activity and stability. The optimal temperature for enzymatic transformations by Cs-CE reported in the literature is close to 80°C, but these enzymes are known to work also at lower temperatures. A suitable temperature can be routinely determined. Generally, the temperature may be in a range from 15°C to 100°C. Preferably, the temperature is in the range from 40° to 90°C, depending on the nature of the enzymes. In particular, if mesophilic enzymes are used, it is preferred that step b) of the process described above is performed at a temperature range of 15°C to 60°C, most preferably at 25°C. If thermophilic enzymes are used, it is preferred that step b) of the process described above is performed at a temperature range of 70° to 90°C, most preferably at 80°C. Optimal temperature conditions can easily be identified for a specific CE enzyme by a person skilled in the art through routine experimentation based on common general knowledge and the information disclosed herein. In general, for a process that is used in industrial conditions, at large scale, the preferred temperature is 80°C. [0065] At a higher concentration range, such as for example, 10-20 mg/ml the mutated CE enzymes described herein showed superior transformation rates, with lesser epilactose formation than the 5xCs-CE (see Example 2, Figures 8 and 9). Accordingly, it is a preferred embodiment, wherein the process for the preparation of a composition comprising lactulose as described above is characterized by a step b) wherein the cellobiose 2-epimerase variant as described above is present at a concentration of 10-20 mg/ml, preferably of 14 mg/ml. [0066] The process according to the invention is characterized in resulting in lactulose comprising compositions with low levels of epilactose impurities. Therefore, a preferred process according to the invention is a process wherein the lactulose comprising composition collected in step c) has an epilactose content of below 14% w/w with respect to lactulose, preferably of below 13%, 12% or 10% w/w with respect to lactulose, more preferably below 8% w/w with respect to lactulose and most preferably below 6% w/w with respect to lactulose in the composition. [0067] A further embodiment of the invention is an aqueous liquid composition comprising lactulose obtained by the process described above, wherein the composition has an epilactose to lactulose ratio of below 0.14, preferably below 0.13, 0.12, or 0.10, more preferably below 0.8. In other words, an aqueous liquid composition comprising lactulose obtained by the process described above is a further embodiment of the invention, wherein the composition has an epilactose amount of below 14% w/w, preferably of below 13%, 12% or 10% w/w, more preferably of below 8% w/w. All % w/w values are given with respect to lactulose, unless indicated otherwise. [0068] Sequences Cs-CE (SEQ ID NO: 1) - UniProt:A4XGA6; PDB: 4Z4J MDITRFKEDLKAHLEEKIIPFWQSLKDDEFGGYYGYMDFNLNIDRKAQKGCILNSRILWF FSACYNVLKSEKCKEMAFHAFEFLKNKFWDKEYEGLFWSVSHKGVPVDVTKHVYVQAFGI YGLSEYYEASGDEEALHMAKRLFEILETKCKRENGYTEQFERNWQEKENRFLSENGVIAS KTMNTHLHVLESYTNLYRLLKLDDVYEALEWIVRLFVDKIYKKGTGHFKVFCDDNWNELI KAVSYGHDIEASWLLDQAAKYLKDEKLKEEVEKLALEVAQITLKEAFDGQSLINEMIEDR IDRSKIWWVEAETVVGFFNAYQKTKEEKYLDAAIKTWEFIKEHLVDRRKNSEWLWKVNED LEAVNMPIVEQWKCPYHNGRMCLEIIKRVD 5xCs-CE (SEQ ID NO: 2) - UniProt:A4XGA6; PDB: 4Z4L MDITMFKEDLKSHLEEKIIPFWQSLKDDEFGGYYGYMDFNLNIDRKAQKGCVLNSRILWF FSACYNVLKSEKCKEMAFHAFEFLKNKFWDKEYEGLFWSVSHKGVPVDVTKHVYVQAFGI YGLSEYYEASGDEEALHMAKRLFEILETKCKRENGYTEQFERNWQEKENRFLSENGVIAS KTMNTHLHVLESYTNLYRLLKLDDVYEALEWIVRLFVDKIYKKGTGHFKVLCDDNWNELI KAVSYGHDIEASWLLDQAAKYLKDEKLKEEVEKLALEVAQITLKEAFDGQSLINEMIEDR IDRSKIWWVEAETVVGFFNAYQKTKEEIYLDAAIKTWEFIKEHLVDRRKNSEWLWKVNED LEAVNMPIVEQWKCPYHNGRMCLEIIKRVD Rm-CE (SEQ ID NO: 3) - UniProt:F8WRK9; PDB:3WKF MSTETIPDVRRLRALQAEVHEELTENILKFWATRTHDPVHGGFVGRVGPDGRPHPEAPRG AILNARILWTFAAAYRQLGTPLYREMAERAYRYFVRHFVDAEHGGVYWMVAADGRPLDTR KHVYAQSFAIYALSEWHRATGGEAALALARSIYDLIETHCADRVHGGYVEACDRAWRPLE DARLSAKDAPEPRSMNTHLHVLEAYANLYRVWPETELAARLQALIELFLRAIYHPATGHL ILFFDERWRPRSRAVSFGHDIEASWLLLEAVDVLGQATLRPRVQQASLHLARATLAEGRA PDGSLYYEIGEQGHLDTDRHWWPQAEALVGFLNAYQESGEVLFYEAAEDVWRYIRERQRD TRGGEWFARVRDDGAPYPDDKVDFWKGPYHNGRACLEAIQRLRHLLEHVRSR Ts-CE (SEQ ID NO: 4) - UniProt:I3VXT6 MEKIVHEMNKELKDRIFPFWSKLKDEENGGFYGYVDYDLNIDNKALKGSILNSRILWFFS AFYCLDKEKKALELAHHAYKFLRDNILDKENKGLYWMVDYKGEPVDTRKHTYSQAFGIYG LAQYYKATGNEEALNIAIDLFNTIERNCRDENGYLEEFDRNWNLKENYELSEHGVISSRT MNTHLHILEAYTLLYDVWKNSNLKKSIIYLLNLFKDKIYSEDGKYLKVFFDDNWNTTIDI KSYGHDIEASWLLDKAVDVLCDEEIKNTTKKYTMEIAENILNNVYSPDGMVNETVEGRTD LSRVWWVQAESVVGFFNAYQKTNDVRFIEASKNIWEYIKRYAIDKRDGGEWYWKVDENGK PFEMPIVEPWKCPYHNGRMCIEIIERVKNSEV Co-CE (SEQ ID NO: 5) – UniProt:D9TH31 MDITSFKKELKSHLEEKIIPFWQSLKDDEFGGYYGYMDFNLNINKKAQKGCILNSRILWF FSACYNVLKSEKCKELAFHAFEFLKNKFWDKDYEGLFWSVSLEGLPVDVTKHVYVQAFGI YGLSEYYEASGDKEALFLARKLFEILETRCKRENGYTEQFERNWQEKENRFLSENGVIAS KTMNTHLHVLESYTNLYKVLKLDDVYEALEWLVRLFVEKIYKKGTGHFKVFCDDNWNELI KAVSYGHDIEASWLLDETAKYLRDEKLKEEVEKLTLEVAQVTLQEAFDGKSLINEKVEDR VDRSKIWWVEAETVVGFFNAYQKTKEEKYLDAAIKTWEFIKEYLVDKRKNSEWLWKVDEN LNPVQMPIVEPWKCPYHNGRMCLEIIKRVG Ra-CE (SEQ ID NO: 6) – UniProt:PODKY5; PDB:3VW5 MMISEIRQELTDHIIPFWNKLRDDENGGFYGYLSYGLGLDKKADKGVILHSRILWFYSNA YMTLGGDELLDNAKHAYEFIKNNCIDYEYGGVYWMMDFEGKPADTMKHTYNIAFAIYALS SYYRASGDKEALALAYRPFEDIEKNTLYEYGYREAFDRQWRLVDNEALSENGLKADKTMN AILHLIEAYTELYKADGNEKVADRLKFQLGQMRDIVYTPDTNALKVFFDTAFNLVGDIHS YGHDIEATWLMDRACDVLGDEDLKKQFAEMDLKISHNIQDIALEDGALNNERDKNEIDKT RVWWVQAEAVVGFINAYQHSGDEKFLESAKSVWENIKEYIIDKREGGEWYSEVTFDHTPH DYKETVGPWKCPYHNGRMCMEVITRGVDI Dt-CE (SEQ ID NO: 7) – UniProt:B8DZK4 MDLKVLKSEIFEHLNNKIIPFWEELKDENNGGYISYVGFDLKPDPYAPKGLVLTSRILWF FSRLYNQLRKEEFINFADHSYKFLIKSFLDKENKGFYWMVDYKGEPIDKRKHLYGQAFVL YGLSEYYKATQKKESLDLALEIYKIIEEVCKNDVGYKEEFDEKWNPKENIIVSEYGIICE RSMNTLLHILEAYTNLFTATYDQSIKKKIEDLIILFKEKIYDSKTNHLYVFFDKKMNPII DAISYGHDIEATWLIDEALRYIDNNKLIKEMSEINLKIAEKVLEEAFESGSLLNERVRGI VDKNRIWWVQAEALVGFLNAYQKSKLDKFLKAVFELWEFIKDFLVDKRAQGEWFWKLDEN YIPSPMPEVDLWKCPYHNGRMCLEVIKRI Ta-CE (SEQ ID NO: 8) – UniProt:W9E788 MVNEGLKFSLTLILGEGTCQMEKIVHEMNKELKDRIFPFWSKLKDEENGGFYGYVDYDLN IDNKALKGSILNSRILWFFSAFYCLDKEKKALELAHHAYKFLRDNILDKENKGLYWMVDY KGEPVDTRKHTYSQAFGIYGLAQYYKATGNEEALNIAIDLFNTIERNCRDENGYLEEFDR NWNLKENYELSEHGVISSRTMNTHLHILEAYTLLYDVWKNSNLKKSIIYLLNLFKDKIYS EDGKYLKVFFDDNWNTTIDIKSYGHDIEASWLLDKAVDVLCDEEIKNTTKKYTMEIAENI LNNVYSPDGMVNETVEGRTDLSRVWWVQAESVVGFFNAYQKTNDVRFIEASKNIWEYIKR YAIDKRDGGEWYWKVDENGKPFEMPIVEPWKCPYHNGRMCIEIIERVKNSEV St-CE (SEQ ID NO: 9) – UniProt:E0RU15; PDB:5ZIG MPLPTTLRPTLRQIRSELAAQLFDHILPFWLGQQDPIHGGFYGSITTGPDPTAPKGLVMT ARHLWTFSQAFLSRPNPAYLEAAGNAYRFLTHALYDATHRGFFWSVHPDGTPLSRVKKLY GNAFAVYALAAYHTASGDREALTLAWETFDLLEDRGRDRRHGGYYEAFTEDWSTPLPEPL GEGETPAPKTMNTHLHILEAYSTLFRTTKEPRVREAMEHLILIFRTHIAPSSHLGLYFAE DWAPMGGGISFGHDIEATWLLTESVELLYGDPLPEWFLSWIRPVMEETARALDTHGGSLP NEQREDGSVDRARVWWVQAEAFVGFLNAYSLFEEPRYLDHACTVWRFIMDHLVDREGGEW FWAVTPEGSPLAGYEKGGMWKASYHNSRACLEGMRRIDTILEEERR Bt-CE (SEQ ID NO: 10) – UniProt:A0A0D0GHZ5; PDB:5ZHB MNTFVNEFRNELETHILPFWAKLKDDENGGYYGLVDYDLHVHKDAGKGGIATCRQLWAFS AAYRVLKKEAYLQQANHAYRFLTEYVFDHQYKGLYWMVDYKGNPSDDRKHVYAQAFGVYA LTEYYRVTQNQEALDYAKQLYKLIETVGFNEETNAYKEEFNRKWEEQSNEMLSENGVIAD ITMNTHLHVLEAYTNLYRVWEDEQLKGRIANLIDLFYEKVFDKQSKFLQVFFNNHWESII DLKSYGHDIEASWLIDDALKVTGNNDRKYTQMVIDIAYNIEKKGVLKDGSLAYENENGKI DYTRVWWVQVEAMVGFYNAYEKTKDEKFLKAVERIWDYVKTYMIDSREGGEWYWSVEADG QPTKREIAGPWKCPYHNARFCLEFIERVGK Ec-CE (SEQ ID NO: 11) - UniProt:B3XZI5 MKNEVVYKQLTEKILPFWNAMRDDENGGFYGYMSEDLHIDDHADKGCILNSRILWFYSTA YMYLQDEKLLDNAKHAFEFLKTYCFDPMCGGIFWSVRYNGKPADTTKHTYNQAFAIYALS AYYEATGSIEAIAIAEIIYEKIEDTMRDTKGYLEAFTRDFRPADNDKLSENGVMAERTMN TLLHIIEAYSALVHALRKKVADPAKGDVRDELFMNVVENKLAAALELMRDKFYNSDRHRL DVFFDKEYESLIDLTSYGHDIEASWLLEWAAGILDDEEITESLHPISSDLVEKVYKEAFD GHSIVNECEDGDVNTDRIWWVEAESVLGFLKAFEREGKEEYRKAAHEILAFILDKQVDKR EGSEWFEMLKEDGTPCHKPMVREWKCPYHNGRMCLEILKSGIEIG Bf-CE (SEQ ID NO: 12) - UniProt:Q5LH66 MDEILKQEMQKELTTRILPYWMERMVDQENGGFYGRITGQEELMPRADKGAILNARILWT YSAAYRLLGREEYKEMANRAKRYLIDHFYDSEFGGVYWSLNYRGEPLDTKKQIYAIGFAI YGLSEFHRATGDPEALMYAVRLFNDIESHSFDGLKNGYCEALTREWNEIADMRLSEKDAN ERKTMNTHLHILEPYTNLYRVWKDARLERQLYNLIGLFTEKILDKDTSHLQLFFDNDWQS KYPVVSYGHDIEASWLLHEAARVLGDAGLIAEIEPVVKKIAAAASEGLTSDGGMIYEKNL TTGHIDGDYHWWVQAETVVGYYNLFRYFGDRGALQHSIDCWEFIKRHLTDDVHGEWFWSL RADGSLNRDDDKAGFWKCPYHNGRMCIELLGE Cl-CE (SEQ ID NO: 13) - UniProt:F2JR71 MFVAEVKEHLVNGIIPFWKKLRDNTYGGYYGYMNYDLELDQTAIKGCILNSRIMWFFSNA YLTLKDESLLEEATHAFEFMKAHCIDYKYGGVYWMLDYKGNPVEDMKHTYNQAFAIYALS SYYVASGNSEALDLAMGLFHKIESTCKDEYGYLEAFDRTWEPIDNHKLSDNKHMESEGKV AEKTMNTILHVLEAYTELYRVGKDVEVGRCLEKLLQMTELKVYNKEKRQLEVFFDTQLRS IADMHSYGHDIEAAWLLDRAALVLGNQDIIDRTKAYTVPIAYKVKEVAFEKGALNNERFN NDIDKTRIWWVQAESVVGFINAYEKTQDVAFLEVAKEIWQYIKTYFIDSRAHSEWYWQVD ENGRPNKSYPIVEPWKCPYHNGRMCLEVIRRDINV Dg-CE (SEQ ID NO: 14) - UniProt:F5IT64 MDFKKLEKQYKSELLDNVLPFWLKNSQDKEYGGYFSCLDRDGTVFDTDKFIWLQGREVWM FSMLYNNVEKRQDWLDCAVQGGEFLKKYGHDGNYNWYFSLDREGNPLIEPYNIFSYTFAT MAFGQLNKATGNEEYAEIARKTFDIILSKVDNPKGKWNKAYPGTRNLKNFALPMILCNLA LEIEHLLPEAFLNKVMDDCIHEVMEVFLRPELGGLIVENVTVNGELSDTFDGRLMNPGHA IEAMWFIMDLGHRLNRPELINKAVETTLKMIDYGWDKEYGGIFYYMDRLGRPPQQLEWDQ KLWWVHIETLIALIKGYYLTGNQECAEWFESIHNYTWQHFKDNEYPEWYGYLNRQGDVLL SLKGGKWKGCFHIPRGLYQIYKTISENQNDL Fj-CE (SEQ ID NO: 15) - UniProt:A5FA14 MPANLKQLKSELTAELDSILNYWSKHTLDNQNEGFVGQIDFNDHIVANAEKGSVLNSRIL WTFSSSYQITKKENHKEIAKRAFEYLSKYFYDPEFNGLFWSINADKTPKDTKNQIYALAF AIYGLSEYYVISQDQKALETAVNLYKSIQKYSYDPVNKGYLEAFTRNWQPIEDLRLSDKD ANEKKTMNTHLHIIEAYANLFKVWKDETLKKDSIELLETIEKHFINTETGHLRLFFDENW IEKPDVISYGHDIEAAWLLLQCAEVLEEENLIANYKKHAVQIAEVTKEGFDTDGGLWYEF EPKENELIAEKHWWVQAEALIGFYNAYQLTGNEEYLKIVFENWNFIKKHILDHKNGEWFW GIHKDYSLIQKDKAGFWKCPYHNSRACLELINRIKT Ph-CE (SEQ ID NO: 16) - UniProt:C6XVU9 MSEILIQEFEKELAGMLNYWATYTVDKVHTGFYGQVDNDNNAIEKADKGAVLNSRILWFF SAAYNYKSWPESLHLARRSYDYIRDHFVDKQFGGVYWSVDFQGKPADTKKQMYALAFALY GITEFYKASREEEALALAKILYADIEKHSFDPINNGYFEAFSCQWSELTDQRLSDKDANE KKTMNTHLHVLEAYTSLYTVWPDEGLGRQIRNLLGVFTDKIIDRDTHHLMLFFDESWHSK SRAISFGHDIEASWLLLEAAESLGDENLIWQFKDVAVKMAMASIQGLDENGGLNYEFEPS NWSREKHWWVQAEAMVGFFNAFQLTKEQTYYDKFLKCWEFTKAHIINTQKGEWFWGVNED LSLMPEQYKVGLWKCPYHNGRACLEMIRRLGVNFDFS Df-CE (SEQ ID NO: 17) - UniProt:C6VW66 MAAFTPEAFKKEFIAILDYWEKYGLDSEKGGFYGRVNYENQPVKDAAKSVVLTGRILWTF SLAHRLLKEAKYLTLADRAYQQLARHFFDPEHGGVYWSVNADGSPLETKKQIYGNAFAMY GLAEYYRVTHFKPALEKAQSLFEIIEKHAFDPVNGGYREAFARDWSATDDYILSKSPWNK SMNTHLHLVEAYTNLYSVWPDARLKKQTAGMLDTIITRIVNPKTETMQLFFDEQWKAKDN IVSYGHDIEASWLLFETAEILHDEKLIGRMKQKSIAMAATAAKGLGADGALNYEYDPETK HTQTDRSWWVVAEQLVGFYNAYQLTKDAQFKTKAEKSWDYIVNEFIDHERGEWFGTVKED GTPVKGDKINFWKCPYHNARACAEMWRRTGKA Ha-CE (SEQ ID NO: 18) - UniProt:A9AYF3 MANLVDGGFDHGLREMWQAQFRDEVLGNILPFWANHTLDHEHGGFYGGLTNDLTIHNEFP RSAVLCGRILWTFACAYRMLGDPRDLAVAEYAYAYLKQAFWDQTYGGLYWSIDANGQPLA DHKQTYAQSFAIYGLAEYVRATGDQSALELAQTLFHLIENHAFDAVYGGYIEGCDRVWQP LGDSRLSKLEPEARKTMNTMLHMMEAYANLLRVWDVADVRQQLASLIEACCEHIIDPVQG RFHLFFDDQWNHHEHGISYGHDIEGSWLLMEAAHVLGDEHLIAKAETLAIGMADAVYRNG RHADGSIIHERAPDGSINLERHWWPQAEAVVGFYNAYQATGKPEFAQAAYDSWNFIQRYF IDHDHGDWFKILDAQNQPLGAIPKVGPWECPYHHARVCFEMIERLAEHNVSVQMRG Sd-CE (SEQ ID NO: 19) - UniProt:Q21NF7 SEQ ID NO: 19 MGVPLKKNKLNYSDLNNAFESELHAIADWWVDNTVDEIHGGFVGEMSVDNKVNATADKGI ILNTRILWFFSEAALFTGNKRYAQMAERAYYYIVDHFLDTEHGGVFWALDYRGNVVNNRK QVYAQAFAIYGLSAYYRLTKQADVLAQAHAIFELTEANAQDENHQGYLEAFSNTWGRLDD MRLSEKDLNSPKSMNTHLHILEAYTALHIAAPSPKVSAAIKHCLELFDRYIINKANWHLR MFQDMDWEDVSASYSYGHDIECSWLMWEAVESLESEELHAHYKPAVIALAETCLEQGIGR YNEVLDAYNFETQTLVDERVWWVQAEALVGFLNAYHLAGEQRYLDAALNVWEFINSYQKD SVGGEWHWLSRLDLPHTGDCKMGPWKAPYHNGRAMMESIKLVQRIFGNNE Sl-CE (SEQ ID NO: 20) - UniProt:D2QDA5 SEQ ID NO: 20 MDLKQLRAELRQEHKDILAYWTRYAPDPVHGGFYGRVNYQNQPDPNAEKGIVLNARILWT FSAALRETQHEEYRSIADQAFQYIKTYFVDPEYGGIYWSVDATGKPKNALKHLYGQAFTL YGLSEYVRATKSEPALALAKEVFQNMVKHAYDPKKGGFVEALARDWSPATDYIISKYDNQ EIKTMNTHLHILESFTCLYHVWPDKSVAEQMRGMLHMFQIHIIDPTTYRMNLFMDMNWKV RRTAVSYGHDIEASWLLPEAADLLAAKNPADKPLQKAFQHIGVKMARAASTGYDPTDGGM NYELEPDGHLNKERSWWVLAEAMVGFMNAYQLTHEKPFLDKSIKSWEFTKKYLLDRQNGE WYMGVTGDHTIVGKDKISMWKCPYHNSRACLEMLERLDHLHG Tt-CE (SEQ ID NO: 21) - UniProt:C5BUF0. SEQ ID NO: 21 METETDTREQLILLQNNCANELAHLVSWWASNVVGDGNFSGEINSQGLPQHQAGKGGIYG SRLLWFFSEVAILGRDSLAPASGPTSGLASAKVSAQALYQYFSHHFIDPEYGGVVWELHA NGELRDGRKQIYAQAFAIYCLSAYYRLTSDEHALAQALTIYELVEEFALDSAHGGYFEAF SREWRELDDIRLSEKDLASPKTMNTHLHLLEAYTGLLRALRQAESPAVQQVAASVQQLLE LYVSRVYNSSSRHVHMFMNTHWEDESLAFSYGHDIESSWLLWEAADVLGDDALKQSYRHD VIALAETCLAQARATDGSLYDVFDKQRQLNIPERVWWVQAEAMVGFFNAWQLTGQQKYLD AVFSLWRYLEAEFITPGSEWPWLARSDQGAGYRAYLAGFWKGPYHNGRALMELISRINSV LNVDC

[0069] Examples [0070] The following examples provide detailed experimental conditions for the method of the present invention and are intended to be illustrative and not limiting of all possible embodiments of the same. [0071] Unless otherwise noted, all materials, solvents and reagents were obtained from commercial suppliers, of the best grade, and used without further purification. Example 1 - Enzyme production For each enzyme variant, DNA was obtained by gene synthesis and the sequence was codon optimized for E.coli expression. The backbone was a pET21a+ vector (Novagen) with an antibiotic resistance for kanamycin. The respective vectors carrying desired DNA were transformed into cloning strain E.coli TOP10 for long term storage as glycerol cryo-stock, and into expression strain E.coli BL21(DE3) following the manufacturer instructions. CE enzyme production protocol is performed at 100 ml scale using Overnight Express™ Instant TB Medium. The cultures were incubated in shaking incubator at 37°C and 200 rpm for 24h and harvested by centrifugation. The cell pellets are washed and resuspended in the buffer (10 mM PIPES, 300 mM NaCl, pH 7.5) to 30%w/v and sonicated according to the manufacturer instructions (Sonics Vibra cell). The obtained total cell lysate (TCL) is processed using a heat-shock purification to obtain pure enzyme fraction and simultaneously remove temperature- sensitive proteins of E.coli. In more details, TCL is incubated for 1h at temperature 10 degrees lower then reported CE enzyme melting temperature (Tm), followed by a high- speed centrifugation step to remove the cell debris. The clarified supernatant represents the heated cell-free extract (CFE delta). To obtain very pure enzyme, heated CFE is applied onto His-tag affinity resin (Talon, Takara) and eluted with imidazole. In the final step, imidazole was removed by ultrafiltration (Vivaspin 15R, Sartorius) and the enzyme was stored in the 10 mM PIPES, 100 mM NaCl pH 7.5 buffer. Enzyme concentration was determined by measuring the absorbance at 280 nm using E1% method (factor E is calculated from the enzyme sequence by ProtParam program) and purity was estimated by SDS-PAGE densitometric analysis. Example 2 - Enzyme screening Enzymes were tested under laboratory conditions (in screening assays) where side- reactions are minimized at 50˚C, using 50 g/L lactose in 50 mM PIPES buffer pH507.5, 1.5 mg/ml enzyme in 10 mM PIPES, 0.1M NaCl, pH 7.5 in 1.2 ml total volume. Under these assay conditions several enzymes were tested: With 5xCs-CE (H247H) the reaction is very fast in reaching >55 wt% conversion to lactulose, but the amount of epilactose remains high (15.8 wt%), resulting in 28.9% w/w epilactose with respect to lactulose. This result is illustrated in Figure 3. With 5xCs-CE H247L the reaction is slow but results in low epilactose amounts (<3 wt%), of 5.7% w/w epilactose with respect to the amount of lactulose. This result is illustrated in Figure 4. With 5xCs-CE H247D the reaction is slower but results in epilactose amounts below 7 wt% and 12.9 w/w % of epilactose with respect to lactulose. This result is illustrated in Figure 5. We also tested the enzymes under industrially relevant conditions at 80˚C, using 400 g/L lactose, 50 mM PIPES buffer pH807.5, 20 mg enzyme in 10 mM PIPES, 0.1M NaCl, pH 7.5 in 1.5 ml total volume. With 5xCs-CE H247D the reaction reaches steady state after 5h with 2 wt% epilactose and 3.2% w/w with respect to lactulose. This result is illustrated in Figure 8. With wild-type 5xCs-CE (H247H) the reaction reaches steady state after 2h with 15 wt% epilactose and 25% w/w with respect to lactulose. This result is illustrated in Figure 9. Accordingly, 5xCs-CE H247L as well as 5xCs-CE H247D are well suited to generate a lactulose solution within USP specifications. As negative control reaction, vials were analysed without any enzyme added. Each reaction was monitored in time and allowed to reach equilibrium. The time samples were taken as described: 40 µl reaction sample was quenched with 760 µl mobile phase (composition: ACN:MeOH:H2O (64:16:20)) containing 10mM urea as internal standard. Additionally, pH was measured at 0min, 4h, 24h, 48h, and 72h to monitor the drop in pH. The reaction samples were analyzed using a HPLC-CAD detector with a Shodex column, assay temperature at 40°C, eluent ACN:MeOH:H2O 75:15:8, and a flow rate of 1 mL/min. The output data from the CAD detector is Area % which in this case is interchangeable with lactulose amount in wt% because the CAD signal is independent of the chemical structure of the substrate, and because the isomers have the same molecular weight (MW). Results of screening assay: Table 3: Overview of activity of different 5xCs-CE H247X variants, herein referred to as 5xCasa-CE variants: Screening assay: 50˚C using 50 g/L lactose in 50 mM PIPES buffer pH507.5, 1.5 mg enzyme in 10 mM PIPES, 0.1M NaCl, pH 7.5 in 1.2 ml total volume after 72 h. Enzyme Lactulose content a_{fter 72h} ^Selectivity % EPILACTOSE (w/w vs wt% LACTULOSE) after 72 h Rx1a 5xCasa-CE-H247A 30% 6.9 5xCasa-CE-H247C 23% 8.9 5xCasa-CE-H247D 52% 12.9 5xCasa-CE-H247E 52%(24h) 6.7 5xCasa-CE-H247F 32% 7.1 5xCasa-CE-H247G 29% 2.7 5xCasa-CE-H247H (_{positive control)} ^{55% 27.5} 5xCasa-CE-H247I 34% 11.5 5xCasa-CE-H247L 42% 5.7 negative control (no e_nzyme) ^{5% 15} 5xCasa-CE-H247M 27% 6.9 Rx1b 5xCasa-CE-H247N 56% 27 5xCasa-CE-H247P 22% 16 5xCasa-CE-H247Q 22% 24 5xCasa-CE-H247R 22% 36 5xCasa-CE-H247S 25% 6.8 5xCasa-CE-H247T 21% 7.9 5xCasa-CE-H247V 25% 9 5xCasa-CE-H247H (_{positive control)} ^{55% 28.9} negative control (no e^{nzyme) 5% NA} Almost all variants 5xCs-CE H247X (X= all canonical amino acids) showed reduced epimerization activity resulting in lower amounts of epilactose formed compared to CE enzymes which are not mutated at any of the three histidine residues. The most suitable enzymes for the lactulose production that have reduced epimerization activity but high isomerization activity are the variants 5xCs-CE H247E, 5xCs-CE H247D, 5xCs-CE H247L, 5xCs-CE H247F and 5xCs-CE H247A. Example 3 – Testing of Cs-CE wild-type and 5xCs-CE wild-type Screening assay: 50˚C using 50 g/L lactose in 50 mM PIPES buffer pH507.5, 1.5 mg enzyme in 10 mM PIPES, 0.1M NaCl, pH 7.5 in 1.5 ml total volume for 120 hours. The results show that for 5xCs-CE a reaction equilibrium is reached after 8 h and the reaction results in 57 wt% lactulose, 15 wt% epilactose and 28 wt% lactose. The ratio of the amount epilactose versus the amount of lactulose is 0.26, also referred to as 26% w/w epilactose with respect to lactulose. The results are illustrated in Figure 6. Figure 7 shows the results for Cs-CE, which do not differ very much, but show that a full conversion (max lactulose) for a wild-type Cs-CE only occurs after 24 h, whereas the full conversion of lactose to lactulose was reached already after 4-8h when 5xCs- CE was used.

Claims

Claims [Claim 1] A cellobiose 2-epimerase variant of a wild-type cellobiose 2- epimerase, wherein the wild-type cellobiose 2-epimerase is selected from the group consisting of Cs-CE (SEQ ID NO: 1), 5xCs-CE (SEQ ID NO: 2), Rm- CE (SEQ ID NO: 3), Ts-CE (SEQ ID NO: 4), Co-CE (SEQ ID NO: 5), Dt-CE (SEQ ID NO: 7), Ta-CE (SEQ ID NO: 8), St-CE (SEQ ID NO: 9), Bt-CE (SEQ ID NO: 10), Ec-CE (SEQ ID NO: 11), Bf-CE (SEQ ID NO: 12), Cl-CE (SEQ ID NO: 13), Dg-CE (SEQ ID NO: 14), Fj-CE (SEQ ID NO: 15), Ph-CE (SEQ ID NO: 16), Df-CE (SEQ ID NO: 17), Ha-CE (SEQ ID NO: 18), Sd-CE (SEQ ID NO: 19), Sl-CE (SEQ ID NO: 20) and Tt-CE (SEQ ID NO: 21), characterized in that at least one histidine residue selected from the group consisting of histidine residues, His 1, His 2 and His 3, located in the catalytic site of the cellobiose 2-epimerase variant, is mutated. [Claim 2] The cellobiose 2-epimerase variant, according to claim 1 wherein the at least one mutated histidine residue is the second of the three histidine residues located in the catalytic site, His 2, located in between His 1 and His 3 in the amino acid sequence of the cellobiose 2-epimerase. [Claim 3] The cellobiose 2-epimerase variant, according to any of the preceding claims, wherein the mutation is located at the second histidine residue, His 2, which is located at the position relative to the corresponding wild-type cellobiose 2-epimerase amino acid sequence as indicated in the following table:

CE enzyme amino acid CE enzyme amino acid (wild-type) position of His 2 (wild-type) position of His 2 Cs-CE 247 Bf-CE 249 5xCs-CE 247 Cl-CE 249 Rm-CE 259 Dg-CE 239 Ts-CE 245 Fj-CE 251 Co-CE 247 Ph-CE 248 Dt-CE 247 Df-CE 246 Ta-CE 265 Ha-CE 261 St-CE 253 Sd-CE 258 Bt-CE 247 Sl-CE 249 Ec-CE 259 Tt-CE 273 [Claim 4] The cellobiose 2-epimerase variant, according to any of the preceding claims, wherein it is a variant of a cellobiose 2-epimerase wild- type selected from the group consisting of Cs-CE (SEQ ID NO: 1) and 5xCs-CE (SEQ ID NO: 2). [Claim 5] The cellobiose 2-epimerase variant, according to claim 1, wherein it is a variant of a cellobiose 2-epimerase wild-type selected from the group consisting of Cs-CE (SEQ ID NO: 1) and 5xCs-CE (SEQ ID NO: 2), and wherein the at least one mutated histidine residue of said variant is located at positions 188, 247 or 377 of the corresponding wild-type cellobiose 2-epimerase. [Claim 6] The cellobiose 2-epimerase variant according to any of the preceding claims wherein it is a variant of a cellobiose 2-epimerase wild- type selected from the group consisting of Cs-CE (SEQ ID NO: 1) and 5xCs-CE (SEQ ID NO: 2), and wherein the at least one mutated histidine residue of said variant is histidine 247. [Claim 7] The cellobiose 2-epimerase variant according to claim 6 wherein the histidine mutation is selected from the group consisting of H247E, H247D, H247L, H247F und H247A. [Claim 8] A cellobiose 2-epimerase variant, according to any of the preceding claims wherein said variant’s amino acid sequence has at least 80%, preferably at least 85%, more preferably at least 90% or even more preferably at least 95% sequence identity to the respective cellobiose 2- epimerase wild-type enzyme’s amino acid sequence. [Claim 9] A process for the preparation of a composition comprising lactulose, wherein the process comprises a step of contacting a solution comprising lactose with a cellobiose 2-epimerase variant according to any of claims 1 to 8. [Claim 10] The process according to claim 9 comprising the steps of: a) providing an aqueous solution comprising lactose b) contacting said solution with a cellobiose 2-epimerase variant, and c) collecting a composition comprising the lactulose formed during step b); wherein said cellobiose 2-epimerase variant is defined according to any one of the claims 1 to 8. [Claim 11] The process according to claim 10, wherein the step b) of the process is performed at a pH in the range from 6 to 9, preferably from 7 to 8, and most preferably at pH 7.5. [Claim 12] The process according to claim 10 or 11, wherein the step b) of the process is performed at a temperature range of 15°C to 60°C, most preferably at 25°C, wherein the cellobiose 2-epimerase variant is a mesophilic enzyme. [Claim 13] The process according to claim 10 or 11, wherein the step b) of the process is performed at a temperature range of 70° to 90°C, most preferably at 80°C, wherein the cellobiose 2-epimerase variant is a thermophilic enzyme. [Claim 14] The process according to any of claims 10 to 13, wherein the cellobiose 2-epimerase variant in step b) is present at a concentration of 10- 20 mg/ml, preferably of 14 mg/ml. [Claim 15] The process according to any of claims 10-14, wherein the lactulose comprising composition collected in step c) has an epilactose content of below 14% w/w with respect to lactulose, preferably of below 13%, 12% or 10% w/w with respect to lactulose, more preferably below 8% w/w with respect to lactulose and most preferably below 6% w/w with respect to lactulose in the composition. [Claim 16] An aqueous liquid composition comprising lactulose obtained by the process according to any of claims 9-15, wherein the composition obtained has an epilactose content of below 14% w/w with respect to lactulose, preferably of below 13%, 12% or 10% w/w with respect to lactulose, more preferably below 8% w/w with respect to lactulose and most preferably below 6% w/w with respect to lactulose.