WO2024003317A1 - Method for generating epilactose preparation - Google Patents

Abstract

The present invention provides a method for increasing the ratio of epilactose:lactose and/or epilactose:lactulose in a mixture containing epilactose with other disaccharides by using a microorganism which preferentially consumes lactose and/or lactulose as a substrate in preference to epilactose. Also provided is a method for producing a preparation comprising epilactose in which a starting material containing lactose is contacted with an enzyme capable of converting lactose to epilactose to provide a reaction mixture comprising lactose, epilactose and optionally lactulose, and performing the above method to increase the ratio of epilactose:lactose and/or epilactose:lactulose. Kits for this purpose and methods of identifying microorganisms useful for this purpose are also provided.

Description

Method for generating epilactose preparation

The present invention relates to a method for increasing the ratio of epilactose: lactose and/or epilactose: lactulose in a mixture containing epilactose with other disaccharides by using a microorganism which preferentially uses lactose/lactulose as a fermentation substrate instead of epilactose. This method may be used to provide enriched or purified preparations of epilactose. Kits for this purpose and methods of identifying microorganisms useful for this purpose are also provided.

Lactose is the most abundant by-product of the dairy industry, particularly in the form of whey, and is increasingly recognized as an important feedstock to produce value-added compounds. Millions of tons of whey are produced annually, but direct utilization is difficult due to its low solubility and sweetening power. However, conversion of lactose to useful derivatives with applications in the food and pharmaceutical fields is of interest.

Lactulose (4-O-p-D-galactopyranosyl-D-fructose) and epilactose (4-O-p- galactopyranosyl-D-mannose) are valuable prebiotics that can be generated from lactose with cellobiose 2-epimerases (CEases). Epilactose has the potential to replace or be a cost- effective alternative to well-known health promoting fructooligosaccharides (FOS). FOS are well established in the food ingredient market as a sweetener and a prebiotic, with a global market size of around 500 million euro/year. Like FOS, epilactose has several documented prebiotic effects and has been shown to promote the growth of probiotic bifidobacteria and lactic acid bacteria in the gastro-intestinal tract (Watanabe et al., 2008, J. Diary Sci. , 91, pp4518-4526). However, compared to FOS, epilactose also promotes the growth of various probiotic bacteria in the upper intestine and may thus provide added health benefits and/or act as a supplement to FOS. Epilactose has also been found to have numerous beneficial health properties (Xiao et al., 2019, Appl. Microbiol., Biol., 102, pp3683-3691).

Epilactose is currently not available as a commercial product, apart from Sigma Aldrich (Merck - 5 million Euro/kg) and is typically used in milligram amounts in microbiology research. Large scale production of epilactose has not been performed.

Chemical epilactose production from lactose is challenging and epilactose is generally obtained enzymatically via epimerization of lactose by epimerases (Ito et al., 2008, Appl. Microbiol. Biotechnol., 79, pp433-441 ; Chen et al., 2015, J. Molec. Catalysis B: Enzymatic, 116, pp39-44; Taguchi et al., 2008, FEMS Microbiol. Lett., 287, pp34-40). However, epimerases are inherently equilibrium type enzymes and complete enzymatic conversion of lactose to epilactose is not achieved therefore requiring further purification steps to obtain the epilactose at useable purity.

Purification of the epilactose is time-consuming, expensive and uses harmful chemicals. It is normally conducted by using chromatography with varying yield and purity (Saburi et al., 2010, Biosci. Biotech. Bioch., 74, pp1736-1737; Kuschel et al., 2017, Eur. Food Res. Technol., 243, p391-402; Chen et al., 2018, J. Dairy Science, 101, pp1872-1882; Watanabe et al., 2008, supra, EP2395080). There therefore remains a need for a method to obtain epilactose in large quantities and with low levels of other disaccharides.

As an alternative to purification of epilactose from mixtures containing other disaccharides by chromatographic methods, the present inventors have developed a method in which those disaccharides, particularly lactose and lactulose, may be removed from the mixture by performing selective fermentation with a microorganism that preferentially uses disaccharides other than epilactose as its primary carbon source thus increasing the relative levels of epilactose to provide a high yield purified or enriched epilactose product. The methods described herein provide a final product with more than 95% purity.

Conveniently, the method may start with a mixture which contains large amounts of lactose, e.g. by using whey as a starting material and subjecting the whey to enzymatic conversion of lactose to epilactose and lactulose using an epimerase. By subjecting this mixture to fermentation with a microorganism that preferentially uses lactose as its carbon source, lactose (as well as lactulose, depending on the selection of the microorganism) is fermented (thus depleted in the culture) by the microorganism and the epilactose remains in an essentially pure form.

Bacteria such as Lactiplantibacillus plantarum (and Lactiplantbacillus pentosus) have been identified as being able to utilize lactose and lactulose, but not epilactose as a carbon source (see Figure 2). In short, the microorganism will ferment (remove) all carbohydrates in the bioreactor, but have no enzymes to degrade epilactose. When appropriate microorganisms and starting materials are selected the entire content of the fermentor will be food grade and the fermentation product can be used directly as/in a food supplement. Notably lactulose and epilactose are formed by thermic conversion of lactose and are commonly found in minor amounts in ultra-high temperature (UHT) treated products (Schuster-Wolff-Buhring et al., 2010, Int. Dairy J., 20, pp.731-741), which make them acceptable for consumption as they are already consumed in conventional products.

Prior to the invention, microorganisms which preferentially utilized lactose and/or lactulose were not known to exist. Indeed it was assumed that those that consumed lactose would also consume epilactose. However, surprisingly it has been found that microorganisms that are selective in their consumption of these disaccharides do exist and can be readily identified. Several such bacteria have been identified and screening methods to identify further microorganisms useful in methods of the invention are provided.

Chen et al., 2018, supra, used yeast fermentation to consume monosaccharides to purify epilactose, but first used a p-galactosidase to degrade lactose into galactose and glucose for consumption in the fermentation reaction. (Saburi et al., 2010, supra., 74(8), pp1736-1737 and JP2011217701 similarly describe the use of p-galactosidase to degrade lactose.) The present invention provides a single step to remove disaccharides such as lactose and lactulose without the need to convert them first to monosaccharides.

Method of increasing epilactose: lactose and/or epilactoselactulose ratio

Thus, in a first aspect the present invention provides a method of increasing the ratio of epilactoselactose and/or epilactoselactulose in a mixture containing i) epilactose, and ii) lactose and/or lactulose, comprising fermenting said mixture with a microorganism which consumes lactose and/or lactulose as a substrate in preference to epilactose as a substrate during said fermentation.

In some methods of the invention it is likely that lactulose will be present only in small quantities, e.g. as a minor product of an epimerase reaction, compared to lactose which may be present in much larger quantities. Consequently, in a preferred aspect the ratio of epilactoselactose is increased and the microorganism consumes lactose, and optionally lactulose, as a substrate in preference to epilactose as a substrate during said fermentation.

As referred to herein, epilactose, lactose and lactulose are disaccharides having the structures as set out in Figure 1. The “ratio” of epilactoselactose or epilactoselactulose is the mass ratio of the different disaccharides and may be determined by any appropriate mechanism. The g/L concentration of epilactose (for example) may be determined in the mixture by HPLC, for example, e.g. as illustrated in Example 1. However, any other appropriate method for quantification of the amount of these disaccharides in the mixture may be used. Equal amounts of the compared disaccharides in the mixture provides a 1 :1 ratio. By way of example, if the epilactose is present at two times the amount of the lactose this may be presented as a ratio of 2:1 or 1 :0.5. The former is considered preferable and where possible recital of ratios using numbers below 1 are avoided. A higher ratio refers to a ratio in which the amount of the first recited disaccharide is increased, i.e. 5:1 is considered a higher ratio than 2:1.

The “mixture” containing i) epilactose, and ii) lactose and/or lactulose may be obtained from any source, but generally contains epilactose at a concentration of at least 0.1 g/L, preferably at least 0.5, 1 or 2g/L. The lactose may similarly be at a concentration of at least 0.1 g/L, preferably at least 0.5, 1 or 2g/L. Lactulose, where present, may be present at the same levels as lactose, or also at much lower levels, e.g. at least 0.01 g/L, preferably at least 0.05 or 0.1 g/L. Conveniently the mixture is prepared as described hereinafter using an epimerase reaction to produce epilactose from a lactose-rich substrate, but any source of the epilactose may be used as the starting material if lactose/lactulose are present at the levels described such that enrichment relative to those disaccharides is justified.

As described herein the step of “fermenting” the mixture with the microorganism refers to an in vitro method of culture in which microorganisms are grown in a culture medium over a period of time under conditions appropriate to sustain their viability and in which large organic molecules are broken down into simpler molecules by enzymes present in the microorganisms. Such conditions include appropriate pH, temperature, osmolality and gas concentration and using a medium containing essential components necessary for growth. As referred to herein, viability refers to microorganisms that remain alive and able to replicate.

The microorganisms are grown in a “culture medium” to achieve fermentation. This liquid medium contains essential components for growth and typically includes: an energy/carbon source, usually in the form of a saccharide such as glucose, all essential amino acids, vitamins, free fatty acids, inorganic salts, trace elements (usually in the micromolar range), for example, which are necessary for growth and/or survival. The solution may also contain components that enhance growth. The solution is preferably optimized to an appropriate pH and salt concentration suitable for cell survival and proliferation and a buffer may be used to maintain pH (e.g. HEPES).

The culture medium comprises the mixture to be fermented. The mixture to be fermented, depending on its source may provide essentially all the requirements for a culture medium or supplementation may be required to allow growth. When supplementation is required a defined or natural medium or mix thereof may be used. Additions may be made to the medium during culture (= culture feed), to provide components necessary for sustained growth to high cell densities. Appropriate media/culture feed for growth may be selected according to the microorganism to be grown. Depending on the microorganism, suitable media formulations include commercially available media such as MRSLIS (De Man, Rogosa and Sharpe (MRS) medium without glucose and tween-20). MRS is available from ThermoFisher Scientific as Thermo Scientific™ Oxoid™ MRS. Only small amounts of medium may be necessary if the mixture to be used for fermentation provides most of the necessary components for growth. MRSLIS for example provides a nitrogen source. Other food-grade nitrogen sources such as chicken-extract, ammonium sulfate or nitrate may be used instead. In the alternative trace elements may be added directed to the mixture to be used for fermentation. The requirements of the microorganism to be used and the mixture to be fermented should be taken into account in selecting the medium, or additional components, that needs to be added.

In accordance with the invention it is important for lactose and/or lactulose to be consumed during fermentation and thus lactose and/or lactulose forms at least one of the carbon sources to be consumed by the microorganism as the substrate on which the microorganism grows. Preferably it is the primary carbon source (e.g. provides > 50% of the total useable carbon source in the culture medium) and any medium used in the fermentation process is devoid of another carbon source. The primary carbon source (lactose/lactulose) is provided in the mixture comprising also epilactose (though the latter is preferentially not consumed), e.g. in processed whey as discussed hereinafter. To this may be added appropriate media to provide essential elements necessary for the growth of the microorganism.

The fermentation is conveniently performed in a bioreactor which is a vessel suitable for growth of cells. Conventional bioreactors can include fed-batch stirred reactors, batch stirred reactors or continuous flow stirred reactors. The bioreactor is of a suitable size to allow cell growth, preferably to a high density. The bioreactor contains a vessel in which the fermentation is performed. For example, the bioreactor vessel has a capacity of at least 1 litre, preferably at least 100, 1000, 10,000 litres or more. Suitable bioreactors are well known and include bioreactors such as autoclavable glass fermenters (for example from Applikon) or stainless steel fermenters (for example from Biolaffite and Infers). The internal conditions of the bioreactor vessel, including, but not limited to pH and temperature, are typically controlled during the fermentation period.

The microorganism to be used in the method consumes lactose and/or lactulose as a substrate in preference to epilactose as a substrate during said fermentation. By consumes is meant that the lactose/lactulose acts as a substrate/carbon source during fermentation, i.e. is used as an energy source and broken down to smaller molecules. These smaller molecules are monosaccharides which are directly metabolized by the microorganism. (Conversion of lactose/lactulose to epilactose by an enzyme or microorganism constitutes bioconversion and is not considered fermentation.)

The microorganism preferentially consumes lactose and/or lactulose rather than epilactose. This does not mean necessarily that the microorganism is incapable of consuming epilactose (e.g. it may do so when in high concentrations and other substrates are unavailable or in few concentration). Rather, it means that if epilactose and lactose (or lactulose) were present in the mixture at a 1:1 ratio that more of the lactose (or lactulose) would be consumed (at least in the short time until the ratio significantly favoured epilactose, e.g. until a ratio of 5:1 or 10:1 was reached). Preferably lactose/lactulose is consumed at least 2, 3, 5 or 10 times as fast (e.g. up to 20 times as fast) as epilactose over the test period. Consumption is conveniently assessed directly, by monitoring the levels of the different disaccharides over the time of culture/fermentation, or may be inferred from indirect measurements, e.g. by monitoring bacterial growth using the different substrates.

It will be appreciated that substrate consumption and preference is affected by culture conditions. To determine if a microorganism has a preference for consumption of lactose/lactulose rather than epilactose, fermentation in the presence of each of these substrates as the primary (or sole) carbon source may be assessed. Such methods are described in the Examples and methods of screening to identify suitable microorganisms are described in more detail hereinafter. Conveniently, therefore, the microorganism may be grown in MRS medium where the glucose is replaced with lactose, lactulose or epilactose as the carbon source. Consumption may be assessed over a period of 24 hours under appropriate growth conditions for the microorganism. Consumption of the substrate may be determined directly by measuring the amount of the substrate present over the growth period, e.g. by HPLC analysis, see e.g. Jameson et al., 2021, Current Res. Biotech., 3, p57- 64. In the alternative indirect assessment of consumption of the substrate may be performed, for example, by measuring optical density for 24 hours.

Advantageously the method of the invention avoids steps in which lactose or lactulose is removed by conversion to monosaccharides (e.g. using a beta-galactosidase for lactose) and subsequent consumption of those monosaccharides. Thus, in a preferred aspect in methods of the invention, lactose and/or lactulose in the mixture is not subjected to hydrolysation to form monosaccharides and/or fermentation of monosaccharides is not performed, e.g. using yeast. Microorganisms

The method of the invention provides a purified or enriched preparation of epilactose which may be used for its health benefits. Therefore, it is preferable to use a microorganism that is safe for human or animal consumption. Thus, in a preferred aspect the microorganism is a food grade or (generally recognized as safe) GRAS microorganism, preferably a food-derived microorganism. Food grade and GRAS microorganisms and genera and species of such microorganisms are well known. A “food-derived” microorganism is a microorganism that is isolated from a food source, e.g. a fermented food source, particularly a dairy food source, such as milk, cheese, yoghurt or extracts of such products, e.g. whey.

The inventors have identified a number of strains of bacteria that have the desired properties. These include commercially available strains, well known strains and strains from their own collection. Alternative suitable strains may be readily identified. Preferably the microorganism is a bacterium or a yeast. When the microorganism is a bacterium it preferably is from the order Lactobacillales, or from a genus selected from Bacillus or Bifidobacteria. In particular the bacteria from the order Lactobacillales may be from a genus selected from Lactobacillus, Streptococcus, Enterococcus, Lactiplantibacillus, Latilactobacillus and Pediococcus. It should be noted that Lactobacillus has recently been recategorized into 25 genera (including Lactobacillus which remains but is more limited), see www.ncbi.nlm. nih.gov/Taxonomy/Browser/wwwtax.cgi?id=186826. The new nomenclature has been used. When the microorganism is a yeast it may be from the genus Saccharomyces or Kluyveromyces. The yeast may be from the species Saccharomyces fragilis or Kluyveromyces lactis which are known to consume lactose.

The inventors have identified suitable strains from Lacti pl anti bacillus plantarum and Lactiplantibacillus pentosus (as disclosed in the Examples) and bacteria from this species are conveniently used in methods of the invention. Strains of utility have been identified as WCFS1 (from L. plantarum), KW1 and KW2 (from L. pentosus, whose genome sequences have been deposited under BioProject Accession numbers PRJNA850900 and PRJNA850901, respectively). Strain TMW1.25 (from L. plantarum) has also been found to have the required properties (this strain is a well-known strain, see e.g. Reminger et al., 1999, J. Appl. Microbiol., 86(6), pp1053-1058). Conveniently methods of the invention may use the bacterium Lactobacillus plantarum WCFS1 (strain NCIMB 8826), deposited under accession number ATCC BAA-793 or the bacterium Lactobacillus pentosus KW1 or KW2. Culture protocols

To increase the epilactose: lactose or epilactoselactulose ratio during culture the microorganism and/or culture conditions are selected to maximize that increase. This may be reflected in the speed with which lactose/lactulose is consumed relative to the consumption of epilactose. Conveniently a microorganism is selected that has a strong preference for lactose/lactulose relative to epilactose, e.g. as discussed above when present at a 1 :1 ratio it consumes the lactose/lactulose at least 2, 3, 5 or 10 times as fast as the epilactose over the test period. Furthermore, the fermentation conditions may be modified to ensure a high rate of lactose/lactulose consumption. This may mean ensuring that the concentration of epilactose is not allowed to exceed a certain level, i.e. to avoid substrate depletion necessitating use of the less preferred substrate.

To ensure that enrichment of the epilactose is achieved relative to lactose during the claimed method, conveniently when the ratio of epilactose: lactose present in said fermentation is less than 10:1 e.g. less than 5:1 , 3:1 or 1:1 (but preferably more than 1:100, 1 :50, 1:25 or 1 :10, e.g. the epilactose is present at approximately 1-90% relative to the concentration of lactose), said microorganism consumes lactose at least 1.5 times faster, preferably at least 2, 3 or 5 times faster (e.g. up to 10 or 20 times faster), than epilactose during said fermentation. When the epilactose: lactose ratio is less than 10:1 the lactose is present at less than 10% concentration relative to epilactose. At this point the microorganism may start to consume epilactose in preference and is to be avoided. In the Examples, it has been shown that lactose is consumed in preference to epilactose until the epilactose: lactose ratio exceeds around 5:1. Furthermore, rates of up to 5 or more times faster consumption were observed for lactose relative to epilactose. Similarly, to ensure that enrichment of the epilactose is achieved relative to lactulose during the claimed method, conveniently when the ratio of epilactoseJactulose in said fermentation is less than 20:1, e.g. less than 10:1 , 5:1 or 2:1 (but preferably more than 1 :100, 1 :50, 1 :25 or 1 :10, e.g. the epilactose is present at approximately 1-80% relative to the concentration of lactulose), said microorganism consumes lactulose at least 1.5 times faster, preferably at least 2, 3 or 5 times faster, than epilactose during said fermentation. When the epilactoseJactulose ratio is less than 20:1 the lactulose is present at less than 5% concentration relative to epilactose. At this point the microorganism may start to consume epilactose in preference and is to be avoided. Lactulose may be present at much lower concentrations than lactose depending on the starting culture mixture that is used. In such cases, lower consumption of that substrate can be tolerated without significantly affecting the purity of the final product.

At the start of the fermentation process it is likely that the concentration of lactose/lactulose will significantly outweigh the concentration of epilactose and hence the method serves the purpose of removing the lactose/lactulose. Thus, conveniently, the mixture before fermentation comprises a) epilactose: lactose in a ratio of less than 5:1, preferably less than 2:1, 1:1 , 1 :2, 1:5 or 1:10; and/or b) epilactoseJactulose in a ratio of less than 50:1, preferably less than 25:1 or 10:1. In an alternative the mixture before fermentation may contain epilactoseJactulose ratios as set out in a) above.

Conveniently the ratio of epilactose: lactose in the mixture is more than 1 :20, e.g. more than 1 :10. Thus, in preferred aspect the epilactose is present at about 10-80% of the lactose levels. The same ratios may be present for epilactoseJactulose. However, conveniently the ratio of epilactoseJactulose in the mixture is more than 5:1 , e.g. more than 10:1. Thus, in preferred aspect the epilactose is present at about 80-98% of the lactulose levels.

When using epimerase-treated whey, as shown in the examples, an epilactoseJactose ratio of 1 :1.7 was observed in the mixture before fermentation (35% epilactose, 60% lactose,).

The method achieves a reduction in lactose and lactulose. Conveniently, following the fermentation step: a) the epilactoseJactose ratio is at least 2:1, preferably at least 5:1, preferably at least 10:1, 15:1, 20:1 or 40:1 ; and/or b) the epilactoseJactulose ratio is at Ieast 2:1 , preferably at least 5:1, preferably at least 10:1, 15:1, 20:1 , 100:1 or 1000:1 ; and/or c) the yield of epilactose is at least 25, 30, 40 or 50%, preferably at least 60, 70, 80 or 90% relative to the amount of epilactose at the start of the fermentation step.

It has been shown in the experiments that epilactose may be consumed in the final stages of fermentation as the available lactose and lactulose decreases. To avoid loss of yield it is preferable to end the fermentation at this point. Therefore, in some cases it may be preferable to cease fermentation when the epilactoseJactose ratio is higher than 5:1 , 10:1, 20:1 or 40:1, though this may depend on the conditions and microorganism used. In the Examples an epilactoseJactose ratio of over 10:1 was achieved. Similar comments apply to lactulose and fermentation may be ended when the epilactoseJactulose ratio is as disclosed above for the epilactoseJactose ratio or conveniently may be more than 20:1 , 100:1 or 1000:1.

The yield of the epilactose is determined relative to the amount of epilactose at the start of the fermentation step based on mass. In the Examples yields of over 50% were achieved. However, lower yields may be accepted to improve purity. The purity that may be achieved is as described hereinafter.

Fermentation conditions are adjusted as appropriate for the microorganism, media and other culture conditions to be used. By way of example, the fermentation may be carried out: a) at between 20 and 40°C; b) until the epilactoseJactose ratio is higher than 10:1 , preferably higher than 20:1 or 40:1; c) until the epilactoselactulose ratio is higher than 100:1, preferably higher than 1000:1 (or has a ratio as disclosed in b); and/or d) for at least 8-24 hours.

The relative conditions are selected to optimize the purity and/or yield of epilactose remaining after the fermentation reaction. Thus, the temperature may be selected to reflect the microorganism’s optimum temperature (i.e. the temperature at which growth is fastest, e.g. exponential growth) or the temperature at which it exhibits the highest preference for disaccharides other than epilactose as substrate. Some bacteria have a temperature optimum around 20°C and thus low temperatures may be appropriate. When using L. plantarum or L. pentosus a temperature of around 35-40°C may be used. Conveniently a temperature of ± 2 °C of the microorganism’s temperature optimum for growth is used. Appropriate pH for growth may be anywhere between pH 5 and 7, for example. This pH is appropriate for L. plantarum or L. pentosus and some other microorganisms. The appropriate pH is of course dependent on the microorganism to be used and may be selected accordingly. The timing of the culture may be short or long but is established based on the yield and purity of epilactose. Different types of culture methods may be used. Whilst conveniently a single culture without intervention once commenced may be used, batch, fed-batch, repeated fed-batch or continuous cultures may also be used.

A “fed-batch” culture refers to a method of culturing cells/microorganisms in which additional components are added to the culture after the start of the culture process, particularly components that have been depleted. In the methods of the invention, this could be the addition of more of the mixture containing the epilactose, lactose and/or lactulose, or other components essential for growth. A fed-batch culture is typically halted at some point and the microorganisms or the medium collected. However, a fed-batch culture may be sustained by harvesting culture volume, e.g. 50-90% at intervals (repeated fed-batch)

In a continuous culture, a steady rate of growth is achieved in a constant volume over an extended period of time. This requires the continued input of required components (as discussed above) and an equal rate of removal of culture medium/microorganisms to sustain a constant environment with a constant volume and number of microorganisms.

As discussed above, in non-continuous methods of culture, the fermentation may be stopped once the preference for disaccharides other than epilactose wanes, e.g. as the lactose and lactulose levels drop. To avoid a loss of yield, the fermentation may be carried out until the ratios indicated above, or lower ratios, are reached. For example, to preserve yield it may be preferable to cease fermentation when the ratio of epilactose: lactose is higher than 5:1, 10:1 , 20:1 or 400:1. Conveniently the epilactose: lactose ratio that is reached is lower than 100:1, 40:1 or 20:1. In the case of lactulose, to preserve yield it may be preferable to cease fermentation when the ratio of epilactoselactulose is as set forth for the epilactoselactose ratio, preferably higher than 20:1 , 100:1 or 1000:1. Conveniently the epilactoselactulose ratio that is reached is lower than 10000:1 or 1000:1.

To determine the point at which fermentation should be terminated, various parameters of the process may be monitored, e.g. cell growth or the consumption of starting materials or production of products. Conveniently test runs are performed to establish appropriate conditions for the reaction and to identify the most appropriate timing for termination of fermentation, preferably specifically based on epilactose yield and purity. The epilactose may subsequently be harvested and optionally further purified and/or processed. In the alternative, a continuous culture may be used in which additional starting materials are added during the culture and a portion of the culture is periodically harvested (e.g. repeated fed-batch).

Preparation of the epilactose-containinq mixture

The mixture to be used for fermentation may be prepared by any appropriate means and is simply required to contain at least epilactose, lactose and/or lactulose. Conveniently it is prepared from a lactose-rich starting material which has been treated with enzymes that are able to convert lactose into epilactose. Such methods inevitably fail to convert all lactose to epilactose and may generate some lactulose and hence the methods of invention which serve to remove lactose/lactulose may be used.

Thus, in a further aspect the present invention provides a method of producing a preparation comprising epilactose comprising the steps of: a) contacting a starting material containing lactose with an enzyme, which is capable of converting lactose to epilactose, under conditions appropriate to generate epilactose from the lactose in the starting material, to provide a reaction mixture comprising lactose, epilactose and optionally lactulose; and b) performing a method as defined hereinbefore to increase the ratio of epilactose:lactose and/or epilactose: lactulose in said reaction mixture to provide said preparation comprising epilactose.

As referred to herein a “contacting” step refers to bringing the starting material and enzyme into contact such that the enzyme is able to perform its reaction. As noted above, conditions appropriate to generate epilactose from the lactose in the starting material should be used and thus appropriate conditions such as temperature, pH, concentrations etc. must be selected in line with the enzyme’s requirements to achieve conversion to epilactose.

The starting material may be any appropriate material which contains sufficiently high levels of lactose that enzymatic conversion to epilactose would produce reasonable amounts of that product. Conveniently the lactose-containing starting material contains at least 50 mg/g (dry weight) lactose, preferably at least 500 mg/g (dry weight). For example, the lactose-containing starting material contains at least 20g/L, preferably at least 40g/L lactose. Suitably whey, or products derived from whey which contain lactose, may be used. Whey is the liquid remaining after milk has been curdled and the solids removed. It is a common byproduct during cheese production. Untreated wet whey contains around 50mg/g (wet weight) lactose and may be used as is. In the alternative lactose-containing material derived from whey may also be used such as other products containing whey or produced from whey. This may include whey protein powder (which may contain residual lactose) or acid whey (which is produced after acid coagulation of milk). Conveniently whey permeate may be used. To prepare the permeate, protein and other solids are removed from whey resulting in a product with a high concentration of lactose, up to 850mg/g (dry weight) lactose.

The enzyme, which is capable of converting lactose to epilactose, is preferably an epimerase, preferably a cellobiose 2-epimerase. The Examples use the cellobiose 2- epimerase from Caldicellulosiruptor bescii DSM 6735 and this is a preferred epimerase (Jameson et al., 2021 , supra). The amino acid sequence for this enzyme is set forth in SEQ ID NO. 1. Preferred enzymes for use are cellobiose-2-epimerases such as the enzyme with the amino acid sequence set forth in SEQ ID NO:1 or having at least 80, 90 or 95 % sequence identity thereto and having cellobiose-2-epimerase activity, in particular being able to convert lactose to epilactose.

Sequence identity may be assessed by any convenient method. However, for determining the degree of sequence identity between sequences, computer programmes that make pairwise or multiple alignments of sequences are useful, for instance EMBOSS Needle or EMBOSS stretcher (both Rice, P. et al., Trends Genet. 16, (6) pp. 276-277, 2000) may be used for pairwise sequence alignments while Clustal Omega (Sievers F et al., Mol. Syst. Biol. 7:539, 2011) or MUSCLE (Edgar, R.C., Nucleic Acids Res. 32(5): 1792-1797, 2004) may be used for multiple sequence alignments, though any other appropriate programme may be used. Whether the alignment is pairwise or multiple, it must be performed globally (i.e. across the entirety of the reference sequence) rather than locally.

Sequence alignments and % identity calculations may be determined using for instance standard Clustal Omega parameters: matrix Gonnet, gap opening penalty 6, gap extension penalty 1. Alternatively, the standard EMBOSS Needle parameters may be used: matrix BLOSUM62, gap opening penalty 10, gap extension penalty 0.5. Any other suitable parameters may alternatively be used.

For the purposes of this application, where there is dispute between sequence identity values obtained by different methods, the value obtained by global pairwise alignment using EMBOSS Needle with default parameters shall be considered valid.

Jameson also describes an epimerase from Roseburia faecis M72 which may be used. A suitable cold-active cellobiose 2-epimerase from Roseburia intestinalis is also described by Chen et al., (2020, J. Dairy Sci., 103(9), pp7730-7741) which was able to convert nearly 30% of lactose to epilactose at 8°C. Suitable enzymes from other sources such as from Thermoanaerobacterium saccharolyticum JQ/SL-YS485 and C. saccharolyticus are also known (described in Jameson, et al., 2021, supra).

These enzymes generally convert lactose to a mixture of epilactose and lactulose. The proportion of these products and the % conversion can be optimized by appropriate control of the reaction, e.g. by selection of the enzyme and reaction conditions, e.g. temperature. Preferably the enzyme and reaction conditions are selected such that conversion of lactose to epilactose is in excess of 25 or 30%.

Conveniently, the conversion of lactose to epilactose is performed at 25°C to 80°C, preferably 40-70°C. However, it will be appreciated that this temperature may depend on the conditions used, particularly the microorganism that is used. As described above the epimerase disclosed by Chen et al., 2020, supra, is effective at 8°C, but is optimally effective at 45°C. The epimerases from Caldicellulosiruptor bescii and Roseburia faecis (Jameson et al., 2021, supra), referred to as CbCEP and RfCEP, have maximal activity at 70°C (at pH 7.5) and 50 °C (at pH 8), respectively. Thus temperatures in the range of e.g. 5-80°C may be used by appropriate selection of the microorganism and reaction conditions. As discussed in Jameson et al., 2021, supra, the temperature used may also affect the conversion efficiency such that more epilactose or lactulose may be produced and thus methods of the invention should also take such variation into account.

The mixture that remains after the enzymatic conversion of lactose to epilactose (and optionally also lactulose, depending on the conditions used), is the so-called reaction mixture. This reaction mixture is used as the mixture containing i) epilactose, and ii) lactose and/or lactulose as used in the methods of the invention to increase the ratio of epilactose: lactose and/or epilactose: lactulose in the mixture. This reaction mixture is thus used in said method as described hereinbefore to increase said ratio to provide the preparation comprising epilactose. The preparation thus obtained may simply be enriched for epilactose (e.g. the ratio of epilactose: lactose is increased) or may be enriched to the extent that it may be considered purified, as discussed hereinafter.

Isolation/separation of produced epilactose

Conveniently the epilactose that is present after said fermentation performed in the method of the invention as described hereinbefore, or the epilactose that is present in the preparation comprising epilactose produced by the method of the invention as described hereinbefore, is harvested, isolated and/or further purified.

To harvest the epilactose or preparation comprising epilactose the culture/fermentation medium is collected (also referred to herein as the fermentation product). The obtained medium containing the epilactose may be retained in the form in which it has been obtained, or may be further treated as described below.

To further harvest the epilactose it may be separated from the liquid in which it was contained, e.g. by concentration and/or drying. In a preferred aspect the epilactose is provided in dried form. Optionally, the epilactose is separated from the culture medium to provide epilactose which has been isolated. The epilactose may be separated and/or purified from the medium by an appropriate method, e.g. filtration, centrifugation, chromatography (e.g. HPLC), precipitation, crystallization etc.

The separated epilactose may be further purified, e.g. using the above described separation means, e.g. in combination, to remove any contaminants. In a preferred aspect, the epilactose (obtained by a method of the invention, which may be in a preparation) is purified to the extent that it has a purity of at least >50%, e.g. >60, 70, 80, 85, 90, or 95% (g/g dry weight).

However, conveniently the purity may be determined by reference to the amount of lactose and/or lactulose that remains (i.e. not taking into account spent media or other remaining impurities). Thus, conveniently, the harvested, isolated or purified epilactose has: a) an epilactose: lactose ratio of at least 2:1, preferably at least 5:1 , preferably at least 10:1, 15:1, 20:1 or 40:1 ; and/or b) an epilactoselactulose ratio of at least 2: 1 , preferably at least 5: 1 , preferably at least 10: 1 , 15:1, 20:1, 100:1 or 1000:1.

In methods of the invention a purity of at least 95%, when assessed by the epilactoselactose ratio, i.e. a ratio of at least 20:1 is achieved, even before further purification is contemplated. In certain embodiments using methods of the invention, lactulose has been found in only negligible quantities at the end of fermentation. Ratios up to 100:1 or up to 1000:1 for both epilactoselactose and epilactoselactulose are considered readily achievable.

The preparation or products containing epilactose which are obtained by the above described methods may be further processed to provide a form suitable for commercial use, e.g. as a prebiotic or for other nutritional or medical uses. Thus, in a further preferred aspect the method additionally comprises the step of processing the obtained (e.g. separated and optionally additionally purified) epilactose (or preparation comprising epilactose) to provide a processed epilactose product. As referred to herein, processing refers to methods of further manipulation of the epilactose (or preparation) beyond simple separation and or isolation, e.g. to provide the product in a different form or to add further components to generate a composition. The resulting products are referred to as processed epilactose products. By way of example, the processing step may be to dry the epilactose (or preparation) and/or to formulate the epilactose (or preparation) into a particular formulation or composition.

When the composition is for medical or nutritional use the composition is appropriately formulated, e.g. to provide a nutritional composition (which may contain other nutritional components) or a pharmaceutical composition, which provides the epilactose in a relevant, pharmaceutically acceptable, carrier and may contain other pharmaceutically relevant components. In the alternative the processing may simply be to alter the formulation of the obtained fermentation product, epilactose or preparation containing epilactose, e.g. pressed into tablets, pellets or particles, e.g. for nutritional purposes.

The invention further extends to a fermentation product, preparation comprising epilactose, or harvested, isolated, purified epilactose or processed epilactose product obtainable by a method as described hereinbefore.

The invention also provides the use of a microorganism to increase the ratio of epilactoselactose and/or epilactoselactulose in a mixture containing i) epilactose, and ii) lactose and/or lactulose, wherein the microorganism is as defined hereinbefore. The terms as referred to here are as defined hereinbefore. Method of screening

As illustrated in the examples, microorganisms that may be used in methods of the invention may be readily identified. Preferably, strains are sought from the above described preferred order, genera or species. Preferred candidates are of food-grade status and grow rapidly as well as having the desired substrate preference.

In this aspect, the invention provides a method of screening to identify a microorganism which consumes lactose and/or lactulose as a substrate in preference to epilactose during fermentation, comprising: i) fermenting said microorganism in a culture medium comprising a) lactose as the carbon source; b) epilactose as the carbon source; and/or c) lactulose as the carbon source, and ii) after said fermentation step identifying a microorganism which consumes lactose and/or lactulose as a substrate during fermentation in preference to epilactose based on the microorganism’s growth and/or consumption of epilactose, and lactose and/or lactulose.

The terms used herein are as described hereinbefore. In addition, reference is made in this method to fermenting said microorganism in a culture medium. The culture medium is as described hereinbefore, insofar as it contains all the essential components for growth. It also contains the carbon source. In methods of the invention in which a lactose/epilactose/lactulose mixture is fermented, the mixture which contains those disaccharides may provide other essential components for growth. However, in the screening method, to allow analysis, it is convenient to add a purified carbon source to the culture medium to assess its consumption. Conveniently, therefore, the microorganism may be grown in MRS medium where the glucose is replaced with lactose, lactulose and/or epilactose as the carbon source.

In one option all of the carbon sources to be tested may be used together. However, this may be more complex to analyse and thus conveniently culture media are provided as follows: a) a culture medium comprising lactose as the carbon source; b) a culture medium comprising epilactose as the carbon source; and/or c) a culture medium lactulose as the carbon source, and each culture medium is tested separately. Where the screening method is to be used to identify a microorganism for use on a mixture with limited amounts of lactulose, it may only be necessary to perform the screen with the media of a) and b) (or a medium containing both epilactose and lactose).

The carbon source is at least predominantly lactose, epilactose and/or lactulose, respectively (i.e. they are the primary carbon source in the medium or in their respective media), to allow assessment of the consumption of those disaccharides separately. However, other carbon sources may be present providing the results remain interpretable, e.g. in which all media used have a comparable level of other carbon sources (besides the carbon source being tested), where present. As noted above, in some screening methods only a single culture medium may be used, in which the culture medium contains epilactose, lactose and/or lactulose. In that case the levels of each of these disaccharides is assessed during fermentation to determine if the microorganism selectively consumes epilactose.

Fermentation is performed on the one or more media and the consumption of epilactose, lactose and/or lactulose compared to identify if the tested microorganism consumes lactose and/or lactulose as a substrate during fermentation in preference to epilactose based on the microorganism’s growth and/or consumption of epilactose, and lactose and/or lactulose. Fermentation is performed as described hereinbefore, but it is expected that only small bioreactors or culture wells are required for these test purposes. As with fermentation methods of the invention, consumption is conveniently assessed directly, by monitoring the levels of the different disaccharides over the time of culture/fermentation or may be inferred from indirect measurements, e.g. by monitoring growth using the different substrates. Ideally the culture is performed for 12-48 hours, e.g. from 18-36 hours, e.g. 24 hours under appropriate growth conditions for the microorganism. The microorganism’s growth may be assessed e.g. by measuring optical density or assessing the mass of cells at the end of culture. Consumption of epilactose, and lactose and/or lactulose may be assessed by appropriate determination of the amount of those disaccharides during culture (or at least at the start and end of culture), e.g. using HPLC, as described hereinbefore. The Examples illustrate appropriate screening methods according to the invention. Kits

The invention also provides a kit comprising: a) an enzyme as described hereinbefore (which is capable of converting lactose to epilactose); and b) a microorganism as described hereinbefore (which consumes lactose and/or lactulose as a substrate in preference to epilactose as a substrate during fermentation).

Such kits may be used for putting the methods of the invention into practice. Conveniently they may contain other components useful for putting the methods of invention into practice, e.g. they may further comprise a suitable buffer or instructions for putting the methods of the invention into effect.

The methods described in the Examples form further preferred aspects of the invention. All combinations of the preferred features described above are contemplated, particularly as described in the Examples. The invention will now be described by way of non-limiting Examples with reference to the drawings in which:

Figure 1 shows the structures of lactulose, epilactose and lactose. Figure 2 shows the growth of Lactiplantibacillus plantarum in MRS medium supplemented with 0.1% glucose, lactose, lactulose or epilactose.

Figure 3 shows growth analysis from a 96-well plate reader of L. plantarum WCFS1 with 10% MRSLIS Epi. The indicated number (g/L) shows the total sugar concentration from Epi in the medium. The graphs are based on data from 3 to 6 replicates.

Figure 4 shows the amount of various sugars remaining at timepoints as shown during fermentation using L. plantarum WCFS1 in a 50 mL culture with 10% MRSLIS and 4.5 g/L Epi at 37 °C without agitation. Representative analysis, data based on technical triplicates.

Figure 5 shows analysis of the 'analytical' 100 mL reactor with L. plantarum with 10% MRSUS and 9 g/L Epi. Lactulose has been omitted from the figure as it was of a negligible amount and was removed from the reaction during the fermentation.

Figure 6 shows the results from the ‘analytical’ 2.0-liter reactor with L. plantarum, 10% MRSUS and 9 g/L Epi. Lactulose has been omitted from the figure as it was of a negligible amount and was removed from the reaction during the fermentation.

Example 1 : Identification of microorganisms which consume lactose and lactulose in preference to epilactose during fermentation and their use to generate purified epilactose product

Various bacteria were screened for their ability to grow in media with different primary carbon sources to identify microorganism with a preference for disaccharides other than epilactose. Use of a selected strain achieved production of an epilactose product with more than 95% purity.

Materials and methods

Media and substrates

The substrates and media tested in growth experiments were MRS (broth from Oxoid) and MRSUS (MRS without glucose and tween-20) as well as whey powder, whey powder permeate (MPP) and whey powder concentrate (WPC80), which differ in their lactose and protein content. Media used in fermentation was MPP treated with a cellobiose 2-epimerase (CbCEP) derived from the bacterium Caldicellulosiruptor bescii (Jameson et al., 2021, Current Res. Biotech., 3, p57-64), either at 30°C or 70°C, henceforth named Epi and Epilactu, respectively. The CbCEP efficiently converts lactose into lactulose and epilactose. The Epi substrate contains less lactulose compared to Epilactu.

Solutions of various concentrations were made by dissolving in dFW. Insoluble components in the whey powder media, Epi and Epilactu were pelleted and removed by centrifugation before being used as substrate for fermentation.

Screening of bacterial strains

For screening, samples were taken from fermented food sources and grown on selective media. Promising strains were cultivated for selection on lactose, lactulose and epilactose and sequenced using 16S-rRNA sequencing. Growth was conducted in media (MRSLIS) supplemented with 0.1% lactose, 0.1 % lactulose or 0.1% epilactose (all sourced from Sigma-Aldrich).

Growth Analysis

For growth analysis, the selected promising strains were cultivated overnight in MRSLIS media supplemented with 1 g/L glucose and incubated at 37°C without shaking. For growth analysis using microtiter plates, the overnight culture was diluted in fresh MRSLIS to appropriate ODeoo, then centrifuged to remove the remaining glucose. The cell pellet was resuspended in fresh MRSLIS, 180 pL of the resuspended bacteria were added to wells containing 20 pL of 10 g/L various carbohydrates (containing glucose, lactose, lactulose or epilactose) to a final concentration of 0.1%, to an OD of ~0.1. The microtiter plate was incubated at 37°C and the growth was measured over time in a MultiSkan FC plate reader. An analog method was used for growth analysis in 50 mL cultures. pH-analysis pH was measured in the medium before and after cultivation using pH strips or pH-probes.

Carbohydrate analysis by HPLC

Quantification of lactose, lactulose and epilactose was performed using HPAEC-PAD on the ICS6000. Samples were taken from microtiter plates or cultures, immediately sterile filtered through 0.22 pm filters to remove bacteria and stop fermentation of the remaining sugars. Filtered samples were diluted with dH2O then loaded into the ICS6000 for analysis.

When analyzing samples, four external standards with predetermined amounts of lactose, lactulose and epilactose were included in the sequence for quantification. In addition, mannose was added to all samples prepared for analysis (including the external standards) as an internal standard to a total concentration of 0.050 g/L, for additional accuracy of quantification.

Results

Screening of bacteria

To screen for bacteria that strictly ferment lactose and lactulose and not epilactose, samples from eight fermented food products known to be inhabited by lactic acid bacteria (LAB), such as two separate sourdough cultures, olives, kombucha with SCOBY and sauerkraut, were taken. Initially, the samples were streaked on agar plates with MRS (Oxoid) or LBS medium for selective growth of LAB. More than thirty bacterial colonies were picked for further analysis. The selected strains were subsequently grown in MRSUS made according to Oxoid recipe ‘CM0361’, but without tween-20 and glucose was replaced with lactose, lactulose or epilactose as the carbon source. Two strains proved potential candidates for the EpiWhey process, as they could grow well using lactose and lactulose, but not epilactose as their carbon source. Both strains were 16S-rRNA sequenced to Lactiplantibacillus pentosus and named KW1 and KW2. In addition to these bacteria screened from food, we analyzed bacteria in our strain collection for similar traits. We identified one strain, L. plantarum WCFS1 which had the desired properties and was included for further analysis. The results of the screen for Lactiplantibacillus plantarum WCFS1 is shown in Figure 2 by way of example.

Growth experiments

Initial growth experiments investigated all three strains’ ability to utilize, or not, glucose, lactose, Epi and Epilactu.

As expected, all three strains grew well in MRS, and in MRSUS where the glucose was replaced with 1% lactose. When whey powder, Epi or Epilactu was used as the only nutrient source, none of the selected strains grew, showing that the whey lacks some essential growth factors. However, supplementing the media with 10% MRSUS restored the growth. The growth analysis also showed that Epi supplemented with 10% MRSUS yielded a better growth rate than using Epilactu. Epi was therefore selected as the medium in all subsequent experiments.

Experiments with various Epi concentrations supplemented with 10% MRSUS were tested in microtiter plates to obtain the highest possible ratio of epilactose/lactose after fermentation. Figure 3 shows the same growth rate in all cultures except for ‘Epi 4.5’ (contains a total sugar content of 4.5 g/L) which reached the stationary phase after 18 hours, likely caused by depletion of available carbohydrates. The stationary phase was reached simultaneously for the other cultures while still having plenty of fermentable carbohydrates left. The reason for this stagnation of growth could be due to a decrease in pH, accumulation of toxic byproducts or a lack of other nutrients.

In addition, an Epi concentration of 4.5 g/l gave the same growth rate as the other cultures for up to 15 hours (Figure 3). Interestingly, when analyzing the lactose/epilactose ratio in the cultures after 24 hours of growth, cultures with 4.5 g/L had an epilactose/lactose fraction of -95%.

The plate reader experiments also revealed that all strains slowly fermented some degree of epilactose, most likely at the end of the fermentation, when the lactose concentration became low. However, all strains preferentially fermented lactose to a high degree compared to epilactose, which leaves epilactose mostly intact. Thus, this indicates that high purity is at the expense of yield to a certain extent.

A 50 mL fermentation was run with L. plantarum WCFS1 with 10% MRSUS and 4.5 g/L Epi as the substrate to better understand the fermentation process and the lactose consumption. Figure 4 shows the results from sampling the 50 mL culture. There is a substantial reduction in lactose as the culture grows, with low consumption of epilactose. However, it also confirms that at low lactose concentration, the bacteria slowly start to consume epilactose.

Conclusion

Screening trials yielded three strains of LAB with the required properties. These strains were characterized and tested for growth on various substrates. Further, several substrate and growth conditions were tested to achieve high concentration and purity of epilactose in the fermentation. For the best performing strain L. plantarum WCFS1 , the most optimal conditions were 10% MRSUS with 4.5 g/L Epi at 37 °C without agitation, which reached a purity of epilactose of >95% at the end of fermentation. (Purity here is based on the measured lactose and epilactose and does not include lactulose, spent media or other remaining impurities. However, based on HPLC results, lactulose is presented only in negligible quantities at the end of fermentation.)

Example 2: Scale-up to produce purified epilactose in bioreactors

Materials and methods

The growth media and strains were as used in Example 1. Bacterial cultures grown overnight were initially used as inoculum for the 100 mL reactors. Eventually the overnight cultures were replaced with frozen bacteria obtained from the 100 mL reactors harvested in the exponential phase. The sampled bacteria were frozen directly as a pellet or in the growth medium at -85°C and either were used successfully as the inoculum. All reactors were run at 37 °C, pH 6.2 with magnetic stirring at 200 rpm. The pH was controlled by automatically adding 1.0 M NaOH or 1.0 M HCI to the growing culture as needed. Samples were taken from the reactors by inserting a needle through a septum on the reactor with 1.0-mL syringes. The samples were placed on ice or sterile filtered immediately to quench the reaction and prepare the samples for HPLC analysis.

2.0-liter reactors were used for cultures of 1.5 liters. They functioned in much the same way as the 100 mL reactors and were used according to the manufacturers' operating procedures. The settings were 37°C, pH 6.2 and 200 rpm stirring.

The largest bioreactor has a capacity of 100 liters, but the working volume of 15 liters was used for the final experiment. This industrial-scale reactor was used with the same setting as the smaller reactors and was operated according to the manufacturer's specifications. Two-phase separator (EasyScalelO Westfalia Separator (Gea)) was used after harvesting from the 100 liter-reactor for separating cells from the liquid. The liquid fraction was subsequently further purified by nanofiltration on a GEA model L filtration unit set up with a TriSep-Nanofiltration membrane.

Results

It was observed from the results of Example 1, that the optimal epilactose/lactose ratio was achieved by fermentation using L. plantarum WCFS1 and 10% MRSUS supplemented with 4.5 g/L Epi as substrate. However, an increased concentration of 9.0 g/L Epi was found to be more successful in the 100 mL reactors (automatic pH-control at ~6.2). Samples were withdrawn for HPLC analysis and ODeoo measurements at various time points during the fermentation (Figure 5). The results from these experiments were comparable to results seen in the 50 mL culture without pH control and stirring presented in Figure 4. Figure 5 shows representative fermentation in a 100 mL reactor. The fermentation clearly shows that lactose is metabolized efficiently until about 12 hours after the start. At around 12 hours, when lactose is close to depletion the fermentation of epilactose (that is fermented at a much slower rate) is more evident.

The 100mL reactor was used as a guide to select harvest times for the 10O-liter reactor and the former was used to prepare cultures as inoculum for 1.5-liter cultures in the 2.0-liter reactors. These were in turn used as inoculum for the final 15-liter fermentation in the 100- liter reactor. The 100 mL- and 2.0 liter reactors had reactors used as inoculum as well as a separate reactor designated as an analytical reactor used for regular OD-measurements and sampling for HPLC analysis into the stationary phase of reaction.

Figure 6 shows the results of analysis from the 2.0-liter analytical reactor. The growth rate was significantly higher in the 2.0-liter reactor compared to the 100 mL reactor (Figure 5). The increased growth rate made it more difficult to time when to harvest the larger reactors, as the increased growth speed also increases the rate of metabolizing available sugars. However, it is evident that lactose was preferentially used as substrate and the ratio of epilactose: lactose increased during fermentation.

The working volume in the 10O-liter reactor was 15 liters in total. Signs of growth, such as oxygen depletion was immediately measured by the reactor instruments which indicated good growth in the 10O-liter reactor. The optimal harvest point for the 10O-liter reactor was difficult to estimate, but we decided to harvest ~7 hours after inoculation at a similar OD as when the 100 mL reactor had the best epilactose/lactose ratio. The fermentation process was quenched when the OD-value reached the same levels as in the analytical 100 mL reactors. The fermentation reaction was stopped by cooling down the culture by adding cold water (4 °C), at the same time as setting cooling on the fermenter. The fermentation liquid was subsequently pumped to a 2-phase separator and cells were separated from the liquid containing the enriched epilactose. The liquid fraction was further pumped over to the nanofiltration unit for concentrating and diafiltration of the product. Diafiltration was continued until reaching a stable conductivity.

The most important results from the analytical reactors in the scale-up experiments are shown below in Table 1. Notably, when scaling up to larger reactors, a much higher OD- value is reached at earlier timepoints.

Table 1 . Selected samples from the scale-up experiment from all reactor sizes. Samples around the timepoint they were used as inoculum and at points of high epilactose purity is displayed.

Reactor 2.0 litres

Asterisks (*) marks point where the culture was used for inoculation of the next reactor.

Analysis of the prepared product harvested from the 100-liter reactor revealed the reaction was stopped too early to yield the same high purity of epilactose as seen for the smaller scale reactors. After further filtration and 2-phase separation, the total yield was estimated to be -52% carbohydrate (the remaining dried material is made up of residual media components). A total of 32 grams of epilactose was recovered in the process.

The product was freeze-dried before storage. This powder was re-dissolved in dFW for reanalysis by HPLC. The results obtained confirmed previous analysis, confirming the ratio of epilactose/lactose to be -54%.

SEQUENCE LISTING

SEQ ID NO: 1 : Amino acid sequence of cellobiose 2-epimerase from Caldicellulosiruptor bescii DSM 6735

MDITKFKEDLKAHLEEKIVPFWQSLKDDEFGGYYGYMDFNLNIHRKAQKGCILNSRILWF

FSACYNVLKNEKCKELAFHAFEFLKNKFWDKEYEGLFWNVSHKGVPVDMTKHVYVQAF

GIYGLSEYYEASGDKEALQMAKKLFEILETKCKRENGYTEQFERNWQEKENRFLSENG

VIASKTMNTHLHVLESYTNLYKVLRTKDVYEALEWIVRLFVDKIYKKGTGHFKVFCDDNW

NELIKAVSYGHDIEASWLLDEAARYLKDEKLKEEVEKLTLEVAQVTLKEAFDGQSLINEM

VEDRVDRSKIWWVEAETVVGFFNAYQKSKEEKFLDAAIKTWKFIEEHLVDKRKNSEWL

WKVSEDLKALDMPIVEPWKCPYHNGRMCLEIIKRVG

Claims

Claims

1. A method of increasing the ratio of epilactose: lactose and/or epilactose:lactulose in a mixture containing i) epilactose, and ii) lactose and/or lactulose, comprising fermenting said mixture with a microorganism which consumes lactose and/or lactulose as a substrate in preference to epilactose as a substrate during said fermentation.

2. The method as claimed in claim 1 wherein the ratio of epilactoseJactose is increased and said microorganism consumes lactose, and optionally lactulose, as a substrate in preference to epilactose as a substrate during said fermentation.

3. The method as claimed in claim 1 or 2, wherein when the ratio of epilactoseJactose present in said fermentation is less than 10:1 , said microorganism consumes lactose at least 1.5 times faster, preferably at least 2, 3 or 5 times faster, than epilactose during said fermentation.

4. The method as claimed in any one of claims 1 to 3, wherein when the ratio of epilactoselactulose in said fermentation is less than 20:1, said microorganism consumes lactulose at least 1.5 times faster, preferably at least 2, 3 or 5 times faster, than epilactose during said fermentation.

5. The method as claimed in any one of claims 1 to 4, wherein the microorganism is a food grade or GRAS microorganism, preferably a food-derived microorganism.

6. The method as claimed in any one of claims 1 to 5, wherein the microorganism is a bacterium or a yeast.

7. The method as claimed in any one of claims 1 to 6, wherein the bacteria is a bacterium from the order Lactobacillales, or a genus selected from Bacillus or Bifidobacteria or the yeast is a yeast from the genus Saccharomyces or Kluyveromyces, wherein preferably the bacterium from the order Lactobacillales is from a genus selected from Lactobacillus, Streptococcus, Enterococcus, Lactiplantibacillus, Latilactobacillus and Pediococcus.

8. The method as claimed in claim 7, wherein the yeast is a yeast from the species Saccharomyces fragilis or Kluyveromyces lactis.

9. The method as claimed in claim 7, wherein the bacteria is a bacterium from the species Lactiplantibacillus plantarum or Lactiplantibacillus pentosus.

10. The method as claimed in claim 9, wherein the bacterium is Lactiplantibacillus plantarum WCFS1 (strain NCIMB 8826), deposited under accession number ATCC BAA- 793.

11. The method as claimed in any one of claims 1 to 10, wherein said mixture before fermentation comprises: a) epilactose: lactose in a ratio of less than 5: 1 , preferably less than 2:1 , 1 :1, 1:2, 1 :5 or 1:10; and/or b) epilactose: lactulose in a ratio of less than 50:1, preferably less than 25:1 or 10:1.

12. The method as claimed in any one of claims 1 to 11 , wherein following said fermentation step: a) said epilactoselactose ratio is at least 2:1, preferably at least 5:1 , preferably at least 10:1 , 15:1 , 20:1 or 40:1 ; and/or b) said epilactose:lactulose ratio is at least 2:1 , preferably at least 5:1, preferably at least 10:1 , 15:1 , 20:1, 100:1 or 1000:1 ; and/or c) the yield of epilactose is at least 50%, preferably at least 60, 70, 80 or 90% relative to the amount of epilactose at the start of the fermentation step.

13. The method as claimed in any one of claims 1 to 12, wherein said fermentation is carried out: a) at between 20 and 40°C; b) until the epilactoselactose ratio is higher than 10:1 , preferably higher than 20:1 or 40:1 ; c) until the epilactose: lactulose ratio is higher than 100:1, preferably higher than 1000:1; and/or d) for at least 8-24 hours.

14. A method of producing a preparation comprising epilactose comprising the steps of: a) contacting a starting material containing lactose with an enzyme, which is capable of converting lactose to epilactose, under conditions appropriate to generate epilactose from the lactose in the starting material, to provide a reaction mixture comprising lactose, epilactose and optionally lactulose; and b) performing a method as defined in any one of claims 1 to 13 to increase the ratio of epilactose: lactose and/or epilactose: lactulose in said reaction mixture to provide said preparation comprising epilactose.

15. The method of claim 14, wherein said conversion of lactose to epilactose is performed at 25°C to 80°C, preferably 40-70°C.

16. The method as claimed in claim 14 or 15, wherein the lactose-containing starting material contains at least 50 mg/g (dry weight) lactose.

17. The method as claimed in any one of claims 1 to 16, wherein the lactose-containing starting material is whey or lactose-containing material derived from whey.

18. The method as claimed in any one of claims 14 to 17, wherein the enzyme is an epimerase, preferably a cellobiose 2-epimerase, preferably a cellobiose 2-epimerase from Caldicellulosiruptor bescii.

19. The method as claimed in any one of claims 1 to 18, wherein the epilactose that is present after said fermentation performed in the method of any one of claims 1 to 13, or the epilactose that is present in the preparation comprising epilactose produced by the method of any one of claims 14 to 18, is harvested, isolated and/or further purified.

20. The method as claimed in claim 19, wherein the harvested, isolated or purified epilactose has: a) an epilactoselactose ratio of at least 2:1 , preferably at least 5:1, preferably at least 10:1, 15:1 , 20:1 or 40:1 ; and/or b) an epilactoselactulose ratio of at least 2:1 , preferably at least 5:1 , preferably at least 10:1 , 15:1, 20:1 , 100:1 or 1000:1.

21. The method as claimed in any one of claims 1 to 20, wherein the epilactose that is present after said fermentation performed in the method of any one of claims 1 to 13, 19 or 20, or the epilactose that is present in the preparation comprising epilactose produced by the method of any one of claims 14 to 18, 19 or 20, has a purity of at least 80, 85, 90 or 95% (g/g dry weight).

22. A fermentation product, preparation comprising epilactose, or harvested, isolated or purified epilactose obtainable by a method as defined in any one of claims 1 to 21.

23. The use of a microorganism to increase the ratio of epilactose: lactose and/or epilactose: lactulose in a mixture containing i) epilactose, and ii) lactose and/or lactulose, wherein said microorganism is as defined in any one of claims 1 or 5-10.

24. A method of screening to identify a microorganism which consumes lactose and/or lactulose as a substrate in preference to epilactose during fermentation, comprising: i) fermenting said microorganism in a culture medium comprising a) lactose as the carbon source; b) epilactose as the carbon source; and/or c) lactulose as the carbon source, and ii) after said fermentation step identifying a microorganism which consumes lactose and/or lactulose as a substrate during fermentation in preference to epilactose based on the microorganism’s growth and/or consumption of epilactose, and lactose and/or lactulose.

25. The method of claim 24 wherein said microorganism is fermented in a) a culture medium comprising lactose as the carbon source; b) a culture medium comprising epilactose as the carbon source; and/or c) a culture medium comprising lactulose as the carbon source.

26. A kit comprising: a) an enzyme as defined in claim 14 or 18; and b) a microorganism as defined in any one of claims 1 or 5-10.

27. The kit as claimed in claim 26, further comprising a buffer.