Dual strain system for producing oligosaccharides

Abstract

The present invention relates to a method for producing one or more oligosaccharides by co-culturing two different oligosaccharide producing strains. The process can be used to produce a balanced mixture of oligosaccharides, with less by-product formation of selected oligosaccharides as well as complex oligosaccharides, especially complex fucosylated and sialylated HMO. Co-culturing is controlled by controlling a carbon source on which each strain can grow.

Description

Technical Field

[0001] The present invention relates to a method for producing one or more oligosaccharides by co - culturing two different oligosaccharide - producing strains. The method can be used to produce balanced oligosaccharide mixtures, selected oligosaccharides with less by - product formation, and complex oligosaccharides, in particular complex fucosylated and sialylated HMOs, as well as neutral oligosaccharides of six or more monosaccharide units. The co - culture is controlled by controlling the carbon sources on which the individual strains can grow. Background Art

[0002] The production of human milk oligosaccharides (HMOs) has been carried out through chemical synthesis pathways, enzymatic pathways, and in - vivo fermentation methods. For industrial applications, the chemical pathway is too complex and costly. Due to the nature of enzymatic reactions, HMOs produced through the enzymatic pathway are usually a mixture of donors, acceptors, and a third oligosaccharide (HMO), as well as by - products released from the donor substrate (leaving group, such as lactose) (see, for example, WO2012 / 156897, WO2012 / 156898, and WO2016 / 063262).

[0003] Currently, for smaller fucosylated, sialylated, and neutral core HMOs, a biological production system using in - vivo fermentation of HMOs is the preferred production method (for details, see Bych et al. 2019, Current Opinion in Biotechnology 56:130–137). In production cells that require multiple glycosidase activities to produce the desired HMO product, by - product HMOs may also be present at the end of fermentation. If a mixture of HMOs is required, usually two desired HMOs are produced in two separate fermentations, then through separate purification processes, and two or more purified HMOs are mixed.

[0004] WO 2015 / 032413 describes a method for producing more complex HMOs of at least four monosaccharide units by adding an exogenous receptor molecule to a culture, enabling the cells to internalize the receptor molecule and thus produce HMOs of at least four monosaccharide units.

[0005] Most fermentation processes use lactose as the initial substrate. WO 2015 / 150328 describes a method in which instead of adding lactose to the culture medium, the cells are engineered to produce lactose, which is then further glycosylated by the cells to produce HMOs; as an alternative, it is mentioned that one type of cell can produce lactose and another type of cell can internalize lactose and produce HMOs. It is not clearly specified whether these cells grow in the same culture medium or whether they grow on different carbon sources.

[0006] WO 2015 / 036138 describes the production of desired oligosaccharides using a first strain and the expression of glycosidases using a second strain to remove by-products produced by the first strain. This setup is described as a continuous fermentation setup with two separate fermenters. Similarly, WO 2022 / 242860 describes the sequential fermentation for the production of oligosaccharides in two separate compartments for culturing two different genetically engineered microbial cells separated by a semi-permeable membrane. Summary of the Invention

[0007] Object of the Invention

[0008] The present disclosure has demonstrated that by controlling the carbon sources on which individual strains can grow, the co-culture of two strains producing two different oligosaccharides can be controlled.

[0009] The ability to co-culture two strains to produce two different oligosaccharides can be applied to a variety of uses

[0010] i) The co-cultured strains can produce a balanced mixture of oligosaccharides produced by each cell, where the amount of each oligosaccharide can be controlled by controlling the amounts of different carbon sources. The advantage of this system is that production capacity can be saved because a mixture of two HMOs can be produced in a single fermentation, rather than producing the oligosaccharides separately and then mixing them after their respective fermentations and purifications.

[0011] ii) The co-cultured strains can produce oligosaccharides (such as HMOs) with reduced oligosaccharide by-products. Among them, the first strain produces a first oligosaccharide as an intermediate oligosaccharide, which is taken up by the second strain, and the second strain uses the internalized intermediate oligosaccharide as a substrate to generate a second oligosaccharide (as Figure 1 shown). The advantage of this system is that by separating two or more glycosyltransferases into two different cells, the formation of oligosaccharide by-products common in single-cell systems can be reduced. The way these glycosyltransferases are separated between the two cells depends on the desired HMO outcome / product. In addition, separating the glycosyltransferase reactions into two different cells allows direct control of their respective rates by adjusting the sugar addition profile, which can be used to balance their respective rates to control the formation of by-products.

[0012] iii) The co-cultured strains can be used to produce complex oligosaccharides by reacting the oligosaccharides produced by each cell with a transglycosidase in the culture medium to form a third complex oligosaccharide (as Figure 2As shown). The application of this dual - strain system is also known as a hybrid dual - strain system. One advantage of this system is that complex oligosaccharides that are usually difficult to export from cells (which can lead to low yields, poor quality of the fermentation broth, and poor fermentation performance) can be produced extracellularly, thus eliminating the export problem and potentially increasing product yields and fermentation performance. Other advantages of this system are that the formation of by - products can be controlled, and lactose can be recovered in situ, resulting in a very low lactose content at the end of fermentation. In addition, compared with in vitro processes (in which the two oligosaccharides used for the transglycosylation reaction are supplied in purified form), this process also has many advantages, such as low raw material costs; eliminating the kinetic equilibrium barrier by recovering lactose, enabling complete conversion of either of the two oligosaccharides produced by the cells, thus increasing the yield of the desired complex oligosaccharide; and being able to control the composition of by - products, which is beneficial for purifying and removing lactose at the end of the process.

[0013] Summary of content

[0014] This application relates to a method for producing one or more oligosaccharides having at least three monosaccharide units, such as: i) a balanced mixture of oligosaccharides; ii) a selected oligosaccharide with reduced by - products; or iii) an oligosaccharide having at least three or four monosaccharide units, such as a complex sialylated and / or fucosylated oligosaccharide.

[0015] This method is based on a surprising discovery: by controlling the carbon sources on which two strains can grow, the two strains can be cultured in a controlled manner in the same culture medium, with each strain producing a different oligosaccharide.

[0016] On the one hand, there is a method for producing one or more oligosaccharides having at least three monosaccharide units, the method comprising the step of co - culturing first and second genetically modified microbial cells in a culture medium, wherein,

[0017] a) The first genetically modified microbial cell is capable of producing a disaccharide or a first oligosaccharide having at least three monosaccharide units, and wherein the genetically modified cell

[0018] i) is capable of growing on a first carbon source and is growth - restricted or non - growing on a second carbon source, and

[0019] ii) contains one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase; and

[0020] iii) contains at least one pathway for producing activated sugar nucleotides from the first carbon source;

[0021] And

[0022] b) The second genetically modified microbial cell is capable of producing a second oligosaccharide having at least three monosaccharide units, and wherein the genetically modified cell

[0023] i) capable of growing on the second carbon source while being growth - restricted or non - growing on the first carbon source; and

[0024] ii) comprising one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase; and

[0025] iii) comprising a biosynthetic pathway for producing activated sugar nucleotides from the second carbon source.

[0026] In some embodiments, the ability to grow on one carbon source and not on a second carbon source is achieved by ensuring that the cell expresses the correct transporter (sugar transport system) for the selected carbon source, while the cell does not express the transporter for the second carbon source or lacks the ability to utilize it for growth after the second carbon source enters the cell, e.g., the kinase required to phosphorylate the imported carbon source. If the selective carbon source growth is not naturally present in the cell, it can be genetically engineered to grow on the desired carbon source and not to grow on the desired carbon source.

[0027] In embodiments, the first and second genetically modified microbial cells are capable of independently producing one or more disaccharides or oligosaccharides selected from: LNB, LacNAc, 2’FL, 3FL, 2’FLacNAc, 2’FLNB, Lewis A, Lewis X, 3’SL, 6’SL, 3’SLacNAc, 3’SLNB, sialyl - Lewis A, sialyl - Lewis X, DFL, LNT - II, LNT, LNnT, LNFP - I, LNFP - III, LNFP - IV, LNFP - V, LNFP - VI, FSL, LST - a, LST - b, LST - c, LST - d, LNDFH - II and LNDFH - III, DSLNT, pLNH, pLNnH, LNH, LNnH, (D)F - LNH - I, (D)F - LNH - II, (D)F - LNH - III, F - para - LNH - I, DF - para - LNH, DF - para - LNnH, TF - LNH, FLSTb, FLSTa, FLST - c, S - LNH, S - LNnH - I, FS - LNH, FS - LNnH - I, DS - F - LNH - II.

[0028] In further embodiments, the first genetically modified cell is capable of producing LacNAc, LNB 2’FL, 3FL, LNT - II, LNT, LNnT, LNFP - I, LST - c or LST - a as the most abundant disaccharide or oligosaccharide (e.g., HMO), and preferably, the disaccharide or oligosaccharide is transported out of the first genetically modified cell, e.g., by a sugar efflux transporter or a major facilitator superfamily transporter.

[0029] In a second aspect, the second genetically modified microbial cell is capable of importing the disaccharide or oligosaccharide produced by the first genetically modified microbial cell. In some embodiments, the import of the disaccharide or oligosaccharide produced by the first genetically modified microbial cell is facilitated by a protein or protein complex selected from Table 1 or Table 2.

[0030] In some embodiments, the second genetically modified microbial cell produces an oligosaccharide having at least three, such as at least four, monosaccharide units. Preferably, the oligosaccharide having at least three monosaccharide units is selected from Lewis A, Lewis X, sialyl-LacNAc. Preferably, the oligosaccharide having at least four monosaccharide units is selected from sialyl-Lewis X, Lewis B, DFL, FSL, LNT, LNnT, LST-a, LST-b, LST-c, LST-d, LNFP-I, LNFP-II, LNFP-III, LNFP-V, LNFP-VI, LNDFH-I, LNDFH-II, LNDFH-III, DSLNT, pLNH, pLNnH, LNH, LNnH, (D)F-LNH-I, (D)F-LNH-II, (D)F-LNH-III, F-para-LNH-I, DF-para-LNH, DF-para-LNnH, TF-LNH, S-LNFP-I (FLSTb), S-LNFP-II (FLSTa), S-LNH, S-LNnH-I, FS-LNH, FS-LNnH-I, DS-F-LNH-II, or a mixture thereof.

[0031] In a third aspect, the one or more oligosaccharides produced are complex oligosaccharides having at least three, such as at least four, monosaccharide units, wherein at least the first or second genetically modified microbial cell produces a donor oligosaccharide and another cell produces a receptor oligosaccharide, and the method further comprises the steps of:

[0032] a) making an enzyme having transglycosidase activity available in the culture medium;

[0033] b) incubating the first disaccharide or first oligosaccharide, the second oligosaccharide produced in co-culture, and the transglycosidase in the same culture medium to form a third oligosaccharide having at least four monosaccharide units in the culture medium.

[0034] In an embodiment, the transglycosidase is selected from alpha-1,2-tranfucosidase, alpha-1,3-transfucosidase, alpha-1,3 / 4-transfucosidase, alpha-2,3-transialidase, and alpha-2,6-transialidase, and the donor oligosaccharide is selected from 2’FL, 3FL, DFL, FSL, LNT, LNnT, sialyl-LacNAc, sialyl-LNB, 3’SL, and 6’SL, and the acceptor oligosaccharide is selected from LacNAc, LNB, 2’FL, 3FL, LNT-II, LNT, LNnT, Para-LNnH, LNFP-I, LNFP-II, LNFP-III, LNFP-IV, LNFP-V, LNFP-VI, 3’SL, 6’SL, LST-a, and LST-c.

[0035] Another aspect is the use of a composition of oligosaccharides (such as HMO) produced by the methods described herein in the production of a nutritional composition. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1:Non-limiting illustration of a dual-strain system with intermediate oligosaccharide uptake to produce target HMOs with fewer by-products. The first strain takes up lactose (Lac) via transporter 1 (TP-1), and the cells of this strain are modified to express at least one glycosyltransferase (GT-1+), e.g., one or two glycosyltransferases, which add one or more additional sugar moieties to lactose (modified lactose) to produce a precursor sugar of at least three monosaccharide units (e.g., four monosaccharide units), e.g., HMO. The precursor sugar is exported from strain 1 to the culture medium by passive diffusion or transporter 2 (TP-2). The precursor sugar is internalized by strain 2 via transporter 3 (TP-3). TP-1 and TP-3 can be the same or closely related proteins, e.g., lacY and LacY variants. The second type of cells is engineered to express one or more glycosyltransferases (GT-2+), e.g., one or two glycosyltransferases, which add one or more additional sugar moieties to the precursor sugar produced by strain 1, thereby generating the desired oligosaccharides, e.g., HMO or at least four monosaccharide units, e.g., five, six, or seven monosaccharides. The desired oligosaccharides are preferably exported from the second strain to the culture medium via transporter 4 (TP-4). However, the oligosaccharides produced by the second strain can also be harvested from the culture medium and biomass. The first and second strains are engineered so that they do not grow on the same carbon source. Preferably, the two strains are inoculated into the bioreactor at the start of fermentation, and their growth can be controlled by supplying the two carbon sources at different rates.

[0037] Figure 2: A non - limiting illustration of a dual - strain co - fermentation - enzymatic process carried out in a fermentative bioreactor. A first oligosaccharide / HMO - producing strain in the bioreactor is supplied with a first carbon source (e.g., glucose, glycerol, sucrose, fructose, galactose, maltose, sorbitol, arabinose, etc.), and an initial amount of lactose (lac) is provided to produce the first oligosaccharide / HMO. The bioreactor also contains a second oligosaccharide / HMO - producing strain, which is supplied with a second carbon source different from the first carbon source (e.g., glucose, glycerol, sucrose, fructose, galactose, maltose, sorbitol, arabinose, etc.) and produces a second oligosaccharide / HMO from lactose. Preferably, the two strains are inoculated into the bioreactor at the start of fermentation, and by supplying the two carbon sources at different rates, their growth and the production rate of their respective oligosaccharides can be controlled. The first and second oligosaccharides / HMOs produced by the cells, once present in sufficient amounts in the culture medium, can serve as donors and acceptors in a transglycosylation reaction. The transglycosylation reaction is catalyzed by a transglycosidase, which is provided to the culture medium, for example, by addition or produced by one of the strains. Once the transglycosidase is present in the ongoing fermentation medium, it catalyzes the transfer of a glycosyl moiety from a donor oligosaccharide / HMO (e.g., sialylated or fucosylated lactose) to an acceptor oligosaccharide / HMO, thereby generating complex HMOs, such as a third sialylated or fucosylated complex HMO, and lactose as a by - product (the leaving group of the enzymatic step). Lactose is sequentially taken up by the first and second strains (in a cycle) to produce more of the first and second oligosaccharides / HMOs, thus shifting the equilibrium towards the formation of the third complex HMO. Near the end of fermentation, feeding lactose to the fermentation broth can be stopped, and the remaining lactose is consumed until fermentation is complete. If the rate of recycling of the leaving group (lactose) is faster than the rate of the enzymatic reaction, the reverse enzymatic reaction (represented by is very low.

[0038] Figure 3 : Co - culture of the 3’SL strain (MF1) and the LNT strain (MF2) in a 2L fermenter, showing the formation of the desired oligosaccharides 3’SL and LNT, as well as the by - products LNT - II and pLNH2, expressed as a percentage of the total HMO (mM)+lactose produced throughout the fermentation process.

[0039] Figure 4: Shows the concentration curves in the dual-strain mixing process, expressed as weight percentages relative to the total weight of the substrate and product (mass fraction %), illustrating the process of synthesizing 6’SL (strain MF5) and LNnT (strain MF6), as well as the decline of lactose as the two HMOs are produced. In the trans-sialylation reaction, 6’SL serves as the donor substrate and LNnT as the acceptor. This reaction is initiated 113 hours after the start of fermentation, when α-2,6-sialyltransferase PITS-197 is added to the fermentation. At this time, LST-c begins to form, and the levels of 6’SL and LNnT decrease, while lactose is produced, stabilizing the production of 6’SL and LNnT. The lactose concentration is represented by a dotted line with circles, the 6’SL concentration by a short dashed line with squares, the LNnT concentration by a long dashed line with triangles, and the LST-c concentration by a solid line with diamonds.

[0040] Figure 5 : Shows the progress curve of the in vitro enzymatic reaction, expressed as a weight percentage relative to the total weight of the substrate (mass fraction %). α-2,6-sialyltransferase (PITS-197) uses 6’SL as the sialic acid donor to catalyze the sialic acid transfer of LNnT (acceptor) to synthesize LST-c. The lactose concentration is represented by a dotted line with circles, the 6’SL concentration by a short dashed line with squares, the LNnT concentration by a long dashed line with triangles, and the LST-c concentration by a solid line with diamonds.

[0041] Figure 6 : Co-culture of 2’FL strain (MF3) and 3FL strain (MF4), where the 3FL strain takes up 2’FL as an intermediate substrate to produce DFL in the MF4 strain.

[0042] Figure 7 : Shows the concentration distribution curve in the single-strain mixing process as a weight percentage relative to the total weight of the substrate and product (mass fraction %). This curve demonstrates the addition of 6’SL to the LNnT strain culture and the process of lactose content decline as LNnT is produced. In the sialic acid transfer reaction initiated by adding α-2,6-sialyltransferase PITS-197, 6’SL serves as the donor substrate and LNnT as the acceptor. At this time, LST-c begins to form, and the levels of 6’SL and LNnT decrease. The lactose concentration is represented by a dotted line with circles, the 6’SL concentration by a short dashed line with squares, the LNnT concentration by a long dashed line with triangles, and the LST-c concentration by a solid line with diamonds.

[0043] Figure 8Shows the changes in the weight percentages of lactose, LNT, 3'SL, and LST-a relative to the total weight of the substrate and product in the dual-strain process. In the sialic acid transfer reaction initiated by adding α-2,3-sialyltransferase TcTS to the fermentation, 3'SL serves as the donor substrate and LNT serves as the acceptor. At this time, LST-a begins to form, and the levels of 3'SL and LNT decrease. The lactose concentration is represented by dotted lines with circles, the 3'SL concentration is represented by short dashed lines with squares, the LNT concentration is represented by long dashed lines with triangles, and the LST-a concentration is represented by solid lines with diamonds.

[0044] Figure 9 : Shows the progress curve of the in vitro enzymatic reaction, expressed as the weight percentage (mass fraction %) relative to the total weight of the substrate. α-2,3-sialyltransferase TcTS uses 3'SL as the fucosyl donor to catalyze the sialic acid transfer of LNT (acceptor) to synthesize LST-a, with an initial molar ratio of LNT to 3'SL of 1:1. The lactose concentration is represented by dotted lines with circles, the 3'SL concentration is represented by short dashed lines with squares, the LNnT concentration is represented by long dashed lines with triangles, and the LST-a concentration is represented by solid lines with diamonds.

[0045] Figure 10 : The 3'SL strain (MF1) and the LNT strain (MF2) are co-cultured at different carbon source ratios, showing the formation of the required oligosaccharides 3'SL and LNT, as well as the by-products LNT-II and pLNH2, expressed as a percentage of the total HMO (mM) + lactose produced during fermentation.

[0046] Figure 11 : Experimental setup for the regeneration and activity assessment of freeze-dried probiotics under acidic conditions of pH 3.0.

[0047] Figure 12 : Shows the colony-forming units (CFU) calculated based on the colonies of Lactobacillus rhamnosus DSM32550 on agar plates. A) The CFU of mixtures B (mixB) and C (mix C) (mixtures containing LST-a and LNT) are counted on agar plates at the dilution gradient E-2. B) The CFU of mixture D (mix D) are counted on agar plates at the dilution gradient E-4.

[0048] Figure 13: Showing the regeneration and viability of freeze-dried Lactobacillus rhamnosus (DSM 32550), incubated at pH 3.0 for 3 hours and inoculated onto plates at two dilution gradients of 1:100 (E-2) and 1:1000 (E-3). A) is the control without HMO; B) is the combination of Lactobacillus rhamnosus (DSM 32550) and an HMO mixture (mix B) containing 55% LST-a and 45% LNT; C) is the combination of Lactobacillus rhamnosus (DSM 32550) and an HMO mixture (mix C) containing 65% LST-a, 55% LNT, and 10% 3'SL; D) is the combination of Lactobacillus rhamnosus (DSM 32550) and an HMO mixture (mix D) containing 25% LNnT, 50% LST-c, and 25% 6'SL.

[0049] Figure 14 : Showing the regeneration and viability of freeze-dried Bifidobacterium longum (DSM 32946), incubated at pH 3.0 for 30 minutes and inoculated onto plates with undiluted A) as the control without HMO; B) as the combination of Bifidobacterium longum (DSM 32946) and an HMO mixture (mix B) containing 55% LST-a and 45% LNT; C) as the combination of Bifidobacterium longum (DSM 32946) and an HMO mixture (mix C) containing 65% LST-a, 55% LNT, and 10% 3'SL; D) as the combination of Bifidobacterium longum (DSM 32946) and an HMO mixture (mix D) containing 25% LNnT, 50% LST-c, and 25% 6'SL. Detailed Description of the Invention

[0050] The present invention is based on a surprising discovery that by controlling the carbon sources on which two strains can grow, the two strains can be cultured in a controlled manner in the same culture medium, with each strain producing a different disaccharide or oligosaccharide. In essence, the first genetically modified cell (the first strain) that produces the first disaccharide or the first oligosaccharide can grow on one carbon source, while its growth is limited or it does not grow on the second carbon source; and the second genetically modified cell (the second strain) that produces the second oligosaccharide can grow on the carbon source on which the first cell has limited growth or does not grow, while its growth is limited or it does not grow on the carbon source of the first genetically modified cell.

[0051] As shown in the examples of the present application, this allows two different strains to grow in the same culture medium (also known as co-culture) without one strain outgrowing the other. Growth on the selected carbon sources also allows for simultaneous inoculation of the strains, while controlling the product formation of the strains by supplementing the carbon sources at the selected ratios and / or at different time points.

[0052] The ability of co - culturing two strains to produce two different oligosaccharides can be applied to various uses

[0053] i) The strains can be used to produce a mixture of oligosaccharides produced separately by each cell, where the amounts of the individual oligosaccharides can be balanced (controlled) by controlling the amounts of different carbon sources. This is the simplest form of the two - strain system, where the two oligosaccharide products are produced independently, and there is no interaction or further processing during the cultivation of the oligosaccharides produced by the cells. One advantage of this system is that production capacity can be saved because a mixture of two HMOs can be produced in a single fermentation, rather than producing the oligosaccharides separately and mixing them after production.

[0054] ii) The first strain can be used to produce a first oligosaccharide as an intermediate oligosaccharide, which is taken up by the second strain, which uses the internalized intermediate oligosaccharide as a substrate to generate a second oligosaccharide (as Figure 1 shown). In this application of the two - strain system, the second strain is dependent on the product produced by the first strain. Thus, the production of the second oligosaccharide is dependent on the production of the first oligosaccharide. One advantage of this system is that the formation of oligosaccharide by - products common in single - cell systems can be reduced by separating two or more glycosyltransferases into two different cells. The way these glycosyltransferases are separated between the two cells depends on the desired HMO outcome / product. In addition, reducing the number of glycosyltransferases in a single cell may also be beneficial for the cell to reduce its metabolic burden when expressing multiple recombinant / heterologous proteins. Furthermore, separating the glycosyltransferase reaction into two different cells allows direct control of their respective rates by regulating the sugar addition curve, which can be used to balance their respective rates to control the formation of by - products.

[0055] iii) The first strain produces a first oligosaccharide and the second strain produces a second oligosaccharide, where in a transglycosylation reaction carried out in the same medium as the growth of the two cells, one oligosaccharide acts as a donor oligosaccharide and the other acts as an acceptor oligosaccharide, resulting in the formation of a third complex oligosaccharide (as Figure 2 shown). This application of the two - strain system is also known as a hybrid two - strain system because it produces the first oligosaccharide and the second oligosaccharide (e.g., HMO) in situ, which react with a transglycosidase in the medium to generate a third oligosaccharide. One advantage of this system is that complex oligosaccharides, which are usually difficult to export from cells, are produced extracellularly, thus eliminating the export problem and potentially increasing product yield and fermentation performance. In addition, since the lactose leaving in the transglycosidase reaction is recycled in situ by genetically modified cells to generate additional first and second oligosaccharides, the lactose content in this system may be very low, which results in a very low lactose content at the end of fermentation and a higher yield of the desired complex oligosaccharide because the kinetic equilibrium barrier for the formation of the third complex oligosaccharide is eliminated by recycling the leaving group.

[0056] In the present invention, a "donor oligosaccharide" should be understood as an oligosaccharide that provides a specific moiety to another compound (preferably an acceptor) in a chemical reaction (such as a nucleophilic substitution reaction or an electrophilic substitution reaction). Similarly, an "acceptor oligosaccharide" should be understood as an oligosaccharide that receives a specific moiety from a donor in a chemical reaction (such as a nucleophilic substitution reaction or an electrophilic substitution reaction) to form a third compound.

[0057] Oligosaccharide

[0058] As used herein, the term "oligosaccharide" refers to a sugar polymer containing at least three monosaccharide units, i.e., oligosaccharides of three, four, five, six, seven, eight or more monosaccharide units. The oligosaccharide can have a linear or branched structure, in which the monosaccharide units are interconnected by glycosidic linkages. In a preferred embodiment, the oligosaccharide contains a lactose, lacto-N-biose (LNB) or N-acetyl lactosamine (LacNAc) residue / moiety at the reducing end, and one or more naturally occurring monosaccharides having 5-9 carbon atoms, which are selected from aldoses (such as glucose, galactose, ribose, arabinose, xylose, etc.), ketoses (such as fructose, sorbose, tagatose, etc.), deoxysugars (such as rhamnose, fucose, etc.), deoxyamino sugars (such as N-acetylglucosamine, N-acetylmannosamine, N-acetylgalactosamine, etc.), uronic acids and ketoaldonic acids (such as sialic acid).

[0059] In an embodiment of the present invention, the oligosaccharide produced by the method described herein can have a lactose (Galβ1-4Glc) moiety or a lacto-N-biose (LNB or Galβ1-3GlcNAc) moiety or an N-acetyl lactosamine (LacNAc or Galβ1-4GlcNAc) moiety at the reducing end.

[0060] In an embodiment, the oligosaccharides produced by the methods described herein comprise at least 3 monosaccharide units and a reducing-end lacto-N-biose (LNB or Galβ1-3GlcNAc) moiety or N-acetyl lactosamine (LacNAc or Galβ1-4GlcNAc) moiety, such as Lewis A (LeA or Galβ1-3[fucα1-4]GlcNAc) or Lewis X (LeX or Galβ1-4[fucα1-3]GlcNAc), or Lewis Y (LeY or Fucα1-2Galβ1-4[Fucα1-3]GlcNAc) or Lewis-B (LeB or Fucα1-2Galβ1-3[Fucα1-4]GlcNAc) or 3'-sialyl-LNB (Neu5Ac-α2-3Galβ1-3-GlcNAc) 3'-sialyl-LacNAc (Neu5Ac-α2-3Galβ1-4-GlcNAc) or 6'-sialyl-LacNAc (6'LN or Neu5Ac-α2-6Galβ1-4-GlcNAc), sialyl-Lewis A (SLeA or Neu5Ac-α2-3Galβ1-3[fucα1-4]GlcNAc) or sialyl-Lewis X (SLeX or Neu5Ac-α2-3Galβ1-4[fucα1-3]GlcNAc).

[0061] In the present invention, complex oligosaccharides refer to oligosaccharides belonging to one of the following three categories: i) oligosaccharides composed of at least four monosaccharide units, wherein at least two monosaccharide units are selected from fucosyl and / or sialyl moieties; ii) oligosaccharides composed of at least five monosaccharide units, preferably having at least one sialyl or fucosyl monosaccharide; and iii) oligosaccharides composed of at least six monosaccharide units, preferably neutral oligosaccharides. In the present disclosure, a subcategory of complex oligosaccharides is highly complex oligosaccharides, wherein at least one monosaccharide unit in the oligosaccharide contains at least three glycosidic linkages to other monosaccharide units.

[0062] HMO

[0063] Preferred oligosaccharides of the present disclosure are human milk oligosaccharides (HMO).

[0064] The term "human milk oligosaccharide" or "HMO" in this text refers to carbohydrates found in human breast milk. HMOs have a core structure, the reducing end of which contains a lactose unit that can be extended by one or more β-N-acetyl lactosamine groups and / or one or more β-lacto-N-biose units, and the core structure can be substituted by α-L-fucosyl (fucosylation) and / or α-N-acetylneuraminyl moieties (sialylation). For example, Xi Chen disclosed HMO structures in Chapter 4, Volume 72 of Advances in Carbohydrate Chemistry and Biochemistry in 2015.

[0065] In the context of this disclosure, lactose is not considered an HMO species but a substrate for the process. It is desirable to minimize the lactose content as much as possible at the end of the production process.

[0066] HMOs are classified into neutral HMOs and acidic HMOs. In this regard, non-acidic (or neutral) HMOs do not contain sialic acid residues, while acidic HMOs have at least one sialic acid residue in their structure. Non-acidic (or neutral) HMOs can be fucosylated or non-fucosylated.

[0067] Examples of such neutral non-fucosylated HMOs (neutral core HMOs) include lacto-N-triose II (LNT-II), lacto-N-tetraose (LNT or Galβ1-3GlcNAcβ1-3Galβ1-4Glc), lacto-N-neotetraose (LNnT or Galβ1-4GlcNAcβ1-3Galβ1-4Glc), lacto-N-neohexose (LNnH or Galβ1-4GlcNAcβ1-3(Galβ1-4GlcNAcβ1-6)Galβ1-4Glc), para-lacto-N-neohexose (pLNnH or Galβ1-4GlcNAcβ1-3Galβ1-4GlcNAcβ1-3Galβ1-4Glc), para-lacto-N-hexose (pLNH or Galβ1-3GlcNAcβ1-3Galβ1-4GlcNAcβ1-3Galβ1-4Glc), and lacto-N-hexose (LNH or Galβ1-3GlcNAcβ1-3(Galβ1-4GlcNAcβ1-6)Galβ1-4Glc).

[0068] Examples of neutral fucosylated HMOs include 2'-fucosyllactose (2'FL or Fucα1-2Galβ1-4Glc), 3-fucosyllactose (3FL or Galβ1-4(Fucα1-3)Glc), difucosyllactose (DFL or LDFT or Fucα1-2Galβ1-4(Fucα1-3)Glc), lacto-N-fucopentaose I (LNFP-I or Fucα1-2Galβ1-3GlcNAcβ1-3Galβ1-4Glc), lacto-N-fucopentaose II (LNFP-II or Galβ1-3[Fuc-α1-4]GlcNAcβ1-3Galβ1-4Glc), lacto-N-fucopentaose III (LNFP-III or Galβ1-4[Fuc-α1-3]GlcNAcβ1-3Galβ1-4Glc), lacto-N-fucopentaose IV (LNFP-IV or Fuc-α1-2Galβ1-4GlcNAcβ1-3Galβ1-4Glc), lacto-N-fucopentaose V (LNFP-V or Galβ1-3GlcNAcβ1-3Galβ1-4[Fuc-α1-3]Glc), lacto-N-fucopentaose VI (LNFP-VI or Galβ1-4GlcNAcβ1-3Galβ1-4[Fucα1-3]Glc), lacto-N-difucosylhexose I (LNDFH-I or Fucα1-2Galβ1-3[Fucα1-4]GlcNAcβ1-3Galβ1-4Glc), lacto-N-difucosylhexose II (LNDFH-II or Galβ1-3[Fuc-(α1-4)]GlcNAcβ1-3Galβ1-4[Fucα1-3]Glc), lacto-N-difucosylhexose III (LNDFH-III or Galβ1-4[Fuc-(α1-3)]GlcNAcβ1-3Galβ1-4[Fucα1-3]Glc), fucosyl-lacto-N-hexose I (FLNH-I or Fuc-α1-2Galβ1-3GlcNAcβ1-3(Galβ1-4GlcNAcβ1-6)Galβ1-4Glc), fucosyl-lacto-N-hexose II (FLNH-II or Galβ1-3(Fuc-α1-4)GlcNAcβ1-3(Galβ1-4GlcNAcβ1-6)Galβ1-4Glc), fucosyl-lacto-N-hexose III (FLNH-III or Galβ1-3GlcNAcβ1-3((Gal(β1-4(Fucα1-3)GlcNAcβ1-6))Galβ1-4Glc), fucosyl-p-lacto-N-hexose I (FpLNH-I or Galβ1-3GlcNAcβ1-3Galβ1-4[Fuc-(α1-3)]GlcNAcβ1-3Galβ1-4Glc),Fucosyl - para - lacto - N - neotetraose II (FpLNnH - II or Galβ1 - 4GlcNAcβ1 - 3Gal(fucα1 - 3)β1 - 4GlcNAcβ1 - 3Galβ1 - 4Glc), difucosyl - lacto - N - hexose I (DF - LNH - I or DF - LNHa or fuc - α1 - 2Galβ1 - 3GlcNAcβ1 - 3((Gal(β1 - 4(Fuc1 - 3)GlcNAcβ1 - 6))Galβ1 - 4Glc), difucosyl - lacto - N - hexose II (DF - LNH - II or DF - LNHb or Galβ1 - 3(fuc - α1 - 4)GlcNAcβ1 - 3,

[0069] ((Gal(β1 - 4(Fuc1 - 3)GlcNAcβ1 - 6))Galβ1 - 4Glc), difucosyl - lacto - N - hexose III (DF - LNH - III or DF - LNHc or fuc - α1 - 2Galβ1 - 3(fuc - α1 - 4)GlcNAcβ1 - 3(Gal(β1 - 4GlcNAcβ1 - 6)Galβ1 - 4Glc), difucosyl - para - lacto - N - hexose (DF - para - LNH or Galβ1 - 3(fuc - α1 - 4)GlcNAcβ1 - 3Galβ1 - 4(fuc - α1 - 3)GlcNAcβ1 - 3Galβ1 - 4Glc), difucosyl - para - lacto - N - neotetraose (DF - para - LNnH or Galβ1 - 4[Fucα1 - 3]GlcNAcβ1 - 3Galβ1 - 4[Fuc - α1 - 3]-GlcNAcβ1 - 3Galβ1 - 4Glc), fucosyl - lacto - N - neotetraose a (FLNnHa), fucosyl - lacto - N - neotetraose b (FLNnHb), difucosyl - lacto - N - neotetraose (DFLNnH) and trifucosyl - lacto - N - hexose (TF - LNH or fuc - α1 - 2Galβ1 - 3(fuc - α1 - 4)GlcNAcβ1 - 3((Gal(β1 - 4(Fuc1 - 3)GlcNAcβ1 - 6))Galβ1 - 4Glc).

[0070] Examples of acidic HMOs include 3'-sialyllactose (3'SL or Neu5Ac-α2-3Galβ1-4-Glc), 6'-sialyllactose (6'SL or Neu5Ac-α2-6Galβ1-4-Glc), 3-fucosyl-3'-sialyllactose (FSL or Neu5Ac-α2-3Galβ1-4(Fucα1-3)Glc), 3'-sialyllacto-N-tetraose a (LSTa or Neu5Ac-α2-3Galβ1-3GlcNAcβ1-3Galβ1-4Glc), fucosyl-LSTa (FLSTa or Neu5Ac-α2-3Galβ1-3(Fucα1-4)GlcNAcβ1-3Galβ1-4Glc), 6'-sialyllacto-N-tetraose b (LSTb or Galβ1-3(Neu5Ac-α2-6)GlcNAcβ1-3Galβ1-4Glc), fucosyl-LSTb (FLSTb or Fucα1-2Galβ1-3(Neu5Ac-α2-6)GlcNAcβ1-3Galβ1-4Glc), 6'-sialyllacto-N-neotetraose (LSTc or Neu5Ac-α2-6Galβ1-4GlcNAcβ1-3Galβ1-4Glc), fucosyl-LSTc (FLSTc or Neu5Ac-α2-6Galβ1-4GlcNAcβ1-3Galβ1-4(Fucα1-3)Glc), 3'-sialyllacto-N-neotetraose (LSTd), fucosyl-LST d (FLSTd), disialyl-lacto-N-tetraose (DSLNT or Neu5Ac-α2-3Galβ1-3(Neu5Ac-α2-6)GlcNAcβ1-3Galβ1-4Glc), sialyl-paralacto-N-neohexose (S-pLNnH or Neu5Ac-α2-3Galβ1-4GlcNAcβ1-3Galβ1-4GlcNAcβ1-3Galβ1-4Glc), sialyl-lacto-N-hexose (SLNH or Neu5Ac-α2-6Galβ1-4GlcNAcβ1-6(Galβ1-3GlcNAcβ1-3)Galβ1-4Glc), fucosyl-sialyl-lacto-N-hexose (FSLNH or Neu5Ac-α2-6Galβ1-4GlcNAcβ1-6(Fucα1-2Galβ1-3GlcNAcβ1-3)Galβ1-4Glc), sialyl-lacto-N-neohexose I (SLNnH-I or Neu5Ac-α2-3Galβ1-4GlcNAcβ1-6(Galβ1-4GlcNAcβ1-3Galβ1-4Glc), fucosyl-sialyl-lacto-N-neohexose I (FSLNnH-I or Neu5Ac-α2-6Galβ1-4GlcNAcβ1-3(Galβ1-4(Fucα1-3)GlcNAcβ1-6)Galβ1-4Glc), sialyl-lacto-N-neohexose II (SLNnH-II or Neu5Ac-α2-6Galβ1-4GlcNAcβ1-3(Galβ1-4GlcNAcβ1-6)Galβ1-4Glc) and disialyl-fucosyl-lacto-N-hexose II (DS-FLNH-II or Neu5Ac-α2-3Galβ1-3(Neu5Ac-α2-6)GlcNAcβ1-3(Galβ1-4(Fucα1-3)GlcNAcβ1-6)Galβ1-4Glc).

[0071] In the context of the present disclosure, a complex HMO is an HMO belonging to one or more of the following three categories: i) an HMO composed of at least four monosaccharide units, wherein at least two monosaccharide units are selected from fucosyl and / or sialic acid moieties; ii) an HMO composed of at least five monosaccharide units, preferably having at least one sialyl or fucosyl monosaccharide, non-limiting examples of which are LNFP-I, LNFP-II, LNFP-V, LST-a, LST-c, and many highly complex HMOs mentioned below; and iii) an HMO composed of at least six monosaccharide units, preferably neutral non-fucosylated oligosaccharides, such as pLNH-I, pLNnH, LNH, and LNnH. In the context of the present disclosure, a subcategory of complex HMOs is highly complex HMOs, wherein at least one monosaccharide unit in the HMO contains at least three glycosidic linkages to other monosaccharide units. Non-limiting examples of highly complex HMOs include LNH, LNnH, LNFP-II, LNFP-III, LST-b, DSLNT, LNDFH-I, LNDFH-II, FLST-a, FLST-b, FpLNnH, FpLNnH-II, F-LNH-I, F-LNH-II, DF-LNH-I, DF-LNH-II, DF-LNH-III, TF-LNH, DFpLNH, DFpLNnH, S-LNFP-I, S-LNH, S-LNnH-I, FS-LNH, FS-LNnH-I, and DS-F-LNH-II.

[0072] Preferably, if produced by fermentation, the complex HMOs of the present invention are not easily exported from the cell cytosol to the supernatant. The complex HMOs produced by the dual-strain mixing method described herein require the action of at least three enzymes. These three enzymes can be, for example, at least two glycosyltransferases present in the cytosol of each genetically engineered cell used in the process, and a transglycosidase present in the fermentation medium. For example, the production of FSL using the dual-strain mixing system described herein requires the presence of an α-1,3-fucosyltransferase in one of the genetically modified cells to form 3FL, and the presence of an α-2,3-sialyltransferase in a second cell to form 3'SL, where both 3FL and 3'SL are exported to the medium, and the presence of an α-2,3-transsialidase in the culture broth to form FSL from the 3FL and 3'SL produced by the cells. Export from the genetically modified cells to the medium may require the presence of a recombinant transporter in the genetically modified cells. Non-limiting examples of suitable transporters can be found, for example, in WO2010 / 142305, WO2021 / 148615, WO2021 / 148614, WO2021 / 148611, WO 2021 / 148610, WO2021 / 148620, and WO2021 / 148618.

[0073] In one method according to the present specification, fucosylated and / or sialylated oligosaccharides having at least four monosaccharide units (such as human milk oligosaccharides (HMOs)) are HMOs having four monosaccharide units, such as DFL or FSL. Preferably, the complex HMO having four monosaccharide units is FSL.

[0074] In one method according to the present specification, fucosylated and / or sialylated oligosaccharides having at least four monosaccharide units (such as human milk oligosaccharides (HMOs)) are oligosaccharides having five monosaccharide units, such as oligosaccharides selected from LNFP-I, LNFP-II, LNFP-III, LNFP-IV, LNFP-V, LNFP-VI, LST-a, LST-b, LST-c, and LST-d. Specifically, fucosylated and / or sialylated HMOs having five monosaccharide units can be selected from HMOs of LNFP-I, LNFP-II, LNFP-III, LNFP-V, LNFP-VI, LST-a, LST-b, and LST-c.

[0075] In one method according to the present specification, fucosylated and / or sialylated oligosaccharides having at least four monosaccharide units, such as human milk oligosaccharides (HMOs), are HMOs having six monosaccharide units. Specifically, fucosylated and / or sialylated HMOs having six monosaccharide units can be selected from DSLNT, LNDFH-I, LNDFH-II, LNDFH-III, FLST-a, FLST-b, and FLST-c.

[0076] On the one hand, human milk oligosaccharides (HMOs) composed of seven or eight monosaccharide units are produced by the method according to the present specification, such as HMOs selected from the following: FLNH-I, FLNH-II, FLNH-III, FpLNH-I, FpLNnH II, DF-LNF-I, DF-LNF-II, DF-LNF-III, DF-para-LNH, DF-para LNnH, FLNnHa, FLNnHb, DFLNnH, TF-LNH, SLNH, FSLNH, SLNnH-I, FSLNnH-I, SLNnH-II, and DS-FLNH-II. The production of these HMOs may require the presence of three or more glycosyltransferase and / or transglycosylase activities.

[0077] Dual-strain system

[0078] The dual-strain oligosaccharide production system disclosed herein is based on culturing two different genetically modified cells in the same culture medium, taking advantage of their ability to grow on different carbon sources. Thus, these cells are cultured in the same container, which does not contain any device for separating the components in the container.

[0079] One aspect of the present disclosure is a method for producing one or more oligosaccharides having at least three monosaccharide units, the method comprising the step of co-culturing first and second genetically modified microbial cells in a culture medium, wherein:

[0080] a) the first genetically modified microbial cell is capable of producing a first disaccharide or a first oligosaccharide of at least three monosaccharide units, and wherein the genetically modified cell

[0081] i) is capable of growing on a first carbon source while being growth-limited or non-growing on a second carbon source, and

[0082] ii) comprises one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase; and

[0083] iii) has at least one pathway for producing nucleotide-activated sugars from the first carbon source; and

[0084] iv) is preferably capable of exporting the first oligosaccharide into the culture medium; and

[0085] b) the second genetically modified microbial cell is capable of producing a second oligosaccharide of at least three monosaccharide units, and wherein the genetically modified cell

[0086] i) is capable of growing on the second carbon source while being growth-limited or non-growing on the first carbon source; and

[0087] ii) comprises one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase; and

[0088] iii) comprises at least one biosynthetic pathway for preparing the activated sugar nucleotide from the second carbon source; and

[0089] iv) is preferably capable of exporting the second oligosaccharide into the culture medium.

[0090] The carbon sources most commonly used in the biological production of oligosaccharides are glucose, glycerol and sucrose, which are preferred carbon sources because they are the most economically attractive. Other alternative carbon sources include fructose, galactose, sorbitol, arabinose and maltose.

[0091] In an embodiment of the present specification, the first and second carbon sources are different and are preferably selected from glucose, glycerol, sucrose, fructose, galactose, sorbitol, arabinose and maltose.

[0092] In order to make the use of fructose, galactose and maltose economically viable, they can be obtained from cheaper sources.

[0093] Fructose can also be obtained from the hydrolysis of sucrose or high fructose syrup (made from glucose). For example, if one cell grows on fructose rather than glucose and another cell grows on glucose rather than fructose, it may be advantageous to use hydrolyzed sucrose, in which case both the first and second genetically modified cells can be fed with hydrolyzed sucrose to produce a 50:50 mixture of fructose and glucose, the ratio of which can be altered by adding high fructose syrup or glucose to the culture medium (e.g., by feeding).

[0094] Hydrolyzed lactose is another option for obtaining glucose and galactose. Thus, if one cell grows on galactose rather than glucose and another cell grows on glucose rather than galactose, both the first and second genetically modified cells can be fed with hydrolyzed lactose to produce a 50:50 mixture of galactose and glucose, the ratio of which can be altered by adding galactose or glucose to the culture medium (e.g., by feeding).

[0095] Glucose syrup is made from starch-rich sources such as (but not limited to) corn, potato, rice, wheat, and barley, and contains glucose, maltose, and longer non-fermentable sugars such as maltotriose. High maltose syrup can also be prepared from starch sources using maltogenic amylase, which produces mainly maltose rather than glucose. Glucose syrup is also known as a carbon source and energy source for bioethanol formation, but can also be used for the bioproduction described herein, where one cell grows on maltose rather than glucose and another cell grows on glucose rather than maltose.

[0096] The inventors of the present application have realized that if the first and second genetically modified strains have different growth capabilities on two selected carbon sources, their growth in the same culture medium can be controlled. Preferably, the first genetically modified cell grows on the first carbon source while being restricted or not growing on the second carbon source. On the other hand, the second genetically modified cell grows on the second carbon source while being restricted or not growing on the first carbon source.

[0097] In this specification, terms such as "culturing" or "fermenting" or "fermentation" are used interchangeably to refer to culturing genetically modified cells (strains) in a bioreactor to produce the first and second oligosaccharides. The culture contains cells and a liquid, also known as the culture broth. The culture medium is the liquid in which the cells can grow. The products in the culture broth or culture medium or the enzymatic reactions occurring in the culture broth or culture medium should be understood to exist / occur extracellularly in the culture medium.

[0098] The term "co-culture" or "co-culturing" as used in the present disclosure refers to the growth of two different genetically modified cells (strains) in the same culture vessel (such as a shake flask, fermenter or bioreactor) such that their products are produced in the same culture medium. In the present disclosure, it is understood that the two different strains are not separated by a semi-permeable membrane. Preferably, the two different strains are grown simultaneously in the culture medium so that cells, components, metabolites and products are freely distributed. For example, the two different strains can be inoculated into the vessel at the start of the culture. This allows the strains to grow simultaneously in the same vessel and produce their products in the same culture medium. However, if it is desired to give the first strain the opportunity to increase its biomass before adding the second strain, the second strain can also be added to the vessel at a later time point. This will still result in the strains growing simultaneously for part of the culture time. If it is known that the initial growth rate of one strain is slower than that of the other strain, or the oligosaccharide yields per mole of carbon source of the two strains are different, or the ratio of the products produced by the two strains is desired to be different, the two different strains can be inoculated (seeded) in different ratios (cells / ml). In the present disclosure, the terms "cell" and "strain" are used interchangeably. In co-culture, the strains do not necessarily need to produce their products simultaneously, as the production of one product may depend on the production of one or more intermediate products.

[0099] Natural microbial fermentation follows four stages, namely the lag phase, the growth phase, the stationary phase, and the death phase. In industrial fed-batch fermentation, cell growth is divided into two stages: In the first stage, cells grow rapidly in the medium, and carbon source acquisition is unrestricted, or carbon source acquisition is restricted after the carbon source is restricted by a rapidly increasing feeding profile. In the second stage, cell growth is more controllable, and industrial products are usually produced in this stage. Carbon (sugar) limitation refers to the stage in fermentation where the growth rate is kinetically controlled by the carbon source (sugar) concentration in the culture broth, and the carbon source concentration is determined by the rate of adding carbon to the fermenter (sugar feeding rate). In the dual-strain method described herein, the cultivation preferably employs fed-batch or continuous fermentation using two different carbon sources. Preferably, the cultivation starts with at least one initial carbon source (batch phase), and when the carbon source is consumed, the carbon source is added to the medium at the required rate throughout the fermentation process (feeding phase). When two different strains are fed with two different carbon sources, the fermentation may essentially comprise two simultaneous batch phases, one for each strain / carbon source. The batch phase (biomass formation) can be carried out using a carbon source different from the carbon source used for producing oligosaccharides in the feeding phase (carbon-limited conditions). This may be particularly useful if the growth rate of one strain on the "production" carbon source is lower than that of the other strain. The batch phase can also be carried out using the same carbon source for culturing both strains, in which case a third carbon source is required because the two different carbon sources used in the feeding phase (production) make it impossible for one of the two strains to grow effectively. The feeding rates of the two carbon sources in the production phase can be predicted based on the knowledge of the growth rates and oligosaccharide product yields of the first and second strains. If a specific oligosaccharide ratio is required, it can be achieved by balancing the ratio of the two carbon sources in the feed according to the oligosaccharide / carbon source yields of the individual strains. As described above, if it is desired for one strain to start growing before the other strain, the batch phases can also be staggered. Industrial fermentation should always be stopped before a large number of cells die.

[0100] In addition to the carbon source for cell growth, cells also require a substrate to form oligosaccharides, which is usually different from the growth carbon source. The most common substrate is lactose, which can be added to the culture or produced by the cells themselves. Other substrates can also be used to produce the first and / or second oligosaccharides. Other substrates are described in the dual-strain mixing system below. If one of the genetically modified strains produces a disaccharide, an additional substrate may not be required because the cells can prepare the disaccharide from the growth carbon source.

[0101] To enable the first and second genetically modified cells to grow on different carbon sources rather than the same carbon source, it may be necessary to select cells with specific growth characteristics or genetically modify the cells to have the required growth characteristics.

[0102] Microorganisms are generally able to utilize a variety of carbon sources to promote their growth. To determine whether a cell can grow on a specific carbon source, for example, the microorganism can be plated on an agar plate containing the selected carbon source and the formation of colonies can be observed.

[0103] Depending on the type of microorganism strain used in this specification, different modifications may be required to ensure that the strain does not grow or has limited growth on the desired carbon source.

[0104] In the co-cultures described herein, growth limitation or reduction of genetically modified cells refers to cells with reduced affinity and uptake rate for a specific carbon source (low-affinity strains), which means that it cannot effectively compete with strains having a higher affinity for the same specific carbon source. This is especially true when growth on a specific carbon source is under carbon-limited conditions, such as in the feed stage of a carbon-limited fed-batch culture or continuous culture, in which case the strain with higher affinity will reduce the residual concentration of the specific carbon source in the medium to such a low level that the strain with lower affinity can hardly grow. In addition, even in the batch stage with an excess of carbon source, the low-affinity strain is at a great disadvantage compared to the high-affinity strain because the maximum growth rate is also affected by the lack of the main carbon source uptake system for the specific carbon source (e.g., deletion of ptsG when the specific carbon source is glucose), as the maximum carbon source uptake rate, and thus the maximum growth rate, will be affected.

[0105] It is known that Gram-negative bacteria have a periplasmic space between the inner cytoplasmic membrane and the outer bacterial membrane. Gram-positive bacteria can also have a periplasmic space, although it is usually much smaller. To prevent or limit the growth of microbial cells on a specific carbon source, one option is to prevent the carbon source from entering the cytosol of the microbial cell. In this way, the carbon source can enter the periplasmic space, but if its entry into the cytosol is blocked, the cell may still not be able to grow on it. Another option is to prevent the cell from further processing the carbon source after it enters the cytosol, so that it cannot enter the energy-producing pathways required for cell growth, such as glycolysis, the pentose phosphate pathway, or the Krebs cycle. This can be achieved by preventing the cell from accessing the enzymes required for phosphorylating the carbon source. This requires eliminating the uptake of the carbon source through the PTS uptake system (if such a system exists) and / or removing the enzymes in the cytosol responsible for phosphorylating the carbon source.

[0106] In some embodiments, the ability to grow on a first carbon source but not on a second carbon source is achieved by ensuring that the cell expresses the correct transporter for the selected carbon source while not expressing an effective transporter for the second carbon source. Whether the required transporter is naturally present in the host cell or not, the cell can be genetically engineered to exhibit the desired carbon source utilization.

[0107] The transport and utilization of different carbon sources (glucose, glycerol, sucrose, galactose, fructose, sorbitol, arabinose, and maltose) will be described below. If alternative carbon sources are needed, those skilled in the art will know how to engineer the cells to adapt to these carbon sources. It should be understood that in the methods described herein, the genetically modified cells can functionally transport / utilize one sugar selected from glucose, glycerol, sucrose, galactose, fructose, sorbitol, arabinose, and maltose, or can reduce or eliminate the transport or utilization enzymes of one or more sugars selected from glucose, glycerol, sucrose, galactose, fructose, sorbitol, arabinose, and maltose, for example, by mutation or deletion of the relevant genes described below.

[0108] In some embodiments, the first genetically modified microbial cell grows on sucrose and does not grow or has limited growth on glucose; the second genetically modified microbial cell grows on glucose and does not grow or has limited growth on sucrose.

[0109] In some embodiments, the first genetically modified microbial cell grows on sucrose and does not grow or has limited growth on glycerol; the second genetically modified microbial cell grows on glycerol and does not grow or has limited growth on sucrose.

[0110] In some embodiments, the first genetically modified microbial cell grows on sucrose and does not grow or has limited growth on maltose, and preferably also does not grow or has limited growth on glucose; the second genetically modified microbial cell grows on maltose and does not grow or has limited growth on sucrose.

[0111] In some embodiments, the first genetically modified microbial cell grows on sucrose and does not grow or has limited growth on galactose; the second genetically modified microbial cell grows on galactose and does not grow or has limited growth on sucrose.

[0112] In some embodiments, the first genetically modified microbial cell grows on sucrose and does not grow or has limited growth on fructose; the second genetically modified microbial cell grows on fructose and does not grow or has limited growth on sucrose.

[0113] In some embodiments, the first genetically modified microbial cell grows on sucrose and does not grow or has limited growth on arabinose, and the second genetically modified microbial cell grows on arabinose and does not grow or has limited growth on sucrose.

[0114] In some embodiments, the first genetically modified microbial cell grows on sucrose and does not grow or has limited growth on sorbitol, and the second genetically modified microbial cell grows on sorbitol and does not grow or has limited growth on sucrose.

[0115] In some embodiments, the first genetically modified microbial cell grows on glucose and does not grow or has limited growth on sucrose, while the second genetically modified microbial cell grows on sucrose and does not grow or has limited growth on glucose.

[0116] In some embodiments, the first genetically modified microbial cell grows on glucose and does not grow or has limited growth on glycerol, while the second genetically modified microbial cell grows on glycerol and does not grow or has limited growth on glucose.

[0117] In some embodiments, the first genetically modified microbial cell grows on glucose and does not grow or has limited growth on galactose, while the second genetically modified microbial cell grows on galactose and does not grow or has limited growth on glucose.

[0118] In some embodiments, the first genetically modified microbial cell grows on glucose and does not grow or has limited growth on fructose, while the second genetically modified microbial cell grows on fructose and does not grow or has limited growth on glucose.

[0119] In some embodiments, the first genetically modified microbial cell grows on glucose and does not grow or has limited growth on arabinose, while the second genetically modified microbial cell grows on arabinose and does not grow or has limited growth on glucose.

[0120] In some embodiments, the first genetically modified microbial cell grows on glucose and does not grow or has limited growth on sorbitol, while the second genetically modified microbial cell grows on sorbitol and does not grow or has limited growth on glucose.

[0121] In some embodiments, the first genetically modified microbial cell grows on glycerol and does not grow or has limited growth on sucrose, while the second genetically modified microbial cell grows on sucrose and does not grow or has limited growth on glycerol.

[0122] In some embodiments, the first genetically modified microbial cell grows on glycerol and does not grow or has limited growth on glucose, while the second genetically modified microbial cell grows on glucose and does not grow or has limited growth on glycerol.

[0123] In some embodiments, the first genetically modified microbial cell grows on glycerol and does not grow or has limited growth on galactose, while the second genetically modified microbial cell grows on galactose and does not grow or has limited growth on glycerol.

[0124] In some embodiments, the first genetically modified microbial cell grows on glycerol and does not grow or has limited growth on fructose, while the second genetically modified microbial cell grows on fructose and does not grow or has limited growth on glycerol.

[0125] In some embodiments, the first genetically modified microbial cell grows on glycerol, does not grow or has limited growth on maltose, and preferably also does not grow or has limited growth on glucose. The second genetically modified microbial cell grows on maltose, does not grow or has limited growth on glycerol.

[0126] In some embodiments, the first genetically modified microbial cell grows on glycerol, does not grow or has limited growth on arabinose. The second genetically modified microbial cell grows on arabinose, does not grow or has limited growth on glycerol.

[0127] In some embodiments, the first genetically modified microbial cell grows on glycerol, does not grow or has limited growth on sorbitol. The second genetically modified microbial cell grows on sorbitol, does not grow or has limited growth on glycerol.

[0128] In some embodiments, the first genetically modified microbial cell grows on galactose, does not grow or has limited growth on maltose, and preferably also does not grow or has limited growth on glucose. The second genetically modified microbial cell grows on maltose, does not grow or has limited growth on galactose.

[0129] In some embodiments, the first genetically modified microbial cell grows on galactose, does not grow or has limited growth on fructose, and preferably also does not grow or has limited growth on glucose. The second genetically modified microbial cell grows on maltose, does not grow or has limited growth on galactose.

[0130] In some embodiments, the first genetically modified microbial cell grows on fructose, does not grow or has limited growth on maltose, and preferably also does not grow or has limited growth on glucose. The second genetically modified microbial cell grows on maltose, does not grow or has limited growth on fructose.

[0131] In some embodiments, one or more oligosaccharides produced by co-culture are a mixture of at least two human milk oligosaccharides (HMOs). Preferably, the at least two oligosaccharides are harvested from the co-culture. The HMOs produced by the first and second genetically modified microbial cells can be independently selected from 2’FL, 3FL, 3’SL, 6’SL, DFL, LNT-II, LNT, LNnT, LNFP-I, LNFP-III, LNFP-V, LNFP-VI, FSL, LST-a, LST-b, LST-c, LST-d, LNDFH-II and LNDFH-III, DSLNT, pLNH, pLNnH, LNH, LNnH, (D)F-LNH-I, (D)F-LNH-II, (D)F-LNH-III, F-para-LNH-I, DF-para-LNH, DF-para-LNnH, TF-LNH, FLSTb, FLSTa, FLST-c, S-LNH, S-LNnH-I, FS-LNH, FS-LNnH-I, DS-F-LNH-II.

[0132] In a preferred embodiment, the HMOs produced by the first and second genetically modified microbial cells can be independently selected from 2’FL, 3FL, 3’SL, 6’SL, DFL, LNT-II, LNT, LNnT, LNFP-I, LNFP-II, LNFP-III, LNFP-V, LNFP-VI, LST-a, LST-c, LNDFH-I, LNDFH-II and LNDFH-III.

[0133] Transport and utilization of glucose

[0134] Glucose is one of the most commonly accepted carbon sources by microorganisms, and microbial cells can uptake glucose through multiple systems and convert it into energy required for growth. Various glucose transport systems have been described in detail. For example, see the description of bacteria in Jaheris et al. 2008 FEMS Microbiol Rev 32:891–907, the description of Escherichia coli in Fuentes et al. 2013 Microbial Cell Factories 12:42, and the description of yeast in Kim et al. 2013 Biochimica et Biophysica Acta 1830:5204–5210.

[0135] In the embodiments described herein, cells grown on glucose have at least one glucose transport system. The glucose transport system can be selected from the systems described by Jaheris et al., Fuents et al., or Kim et al. Specifically, the glucose import system can be selected from the phosphoenolpyruvate:sugar phosphotransferase system (PTS), such as the PTS-dependent glucose (glc) utilization system, the PTS-dependent mannose (man) utilization system, the PTS-dependent maltose (mal) utilization system, the PTS-dependent β-glucoside (bgl) utilization system, or the PTS-dependent N-acetylglucosamine (nag) utilization system. Alternative glucose transport systems include the galactose:H+ symporter GalP, the glucose uptake protein GlcU, the sodium / glucose transporter family (SGLT), or ABC transporters, such as the galactose / glucose ABC transporter (mglABC) system, the trehalose / maltose / sucrose / palatinose (TMSP)-ABC transporter (malEFG) system, and the glucose / mannose ABC transporter (glcEFG) system, or MFS transporter systems, such as the glucose proton symporter (glcP) and the glucose facilitator (glf) or the hexose transporter (HXT).

[0136] In other embodiments, glucose entry into the cytosol of a microorganism can be reduced or blocked by mutating or deleting one or more sequences encoding proteins that affect glucose import ability, as described in the previous paragraph.

[0137] To utilize glucose for energy, once glucose enters the cytosol, the cell phosphorylates it to glucose-6-phosphate, which can enter energy-generating metabolic pathways such as glycolysis and the pentose phosphate pathway. Thus, blocking the formation of glucose-6-phosphate may also help prevent the cell from using glucose as a carbon source for growth.

[0138] In some embodiments, the genetically modified microbial cell is a bacterium that grows poorly or not at all on glucose, wherein the functionality of one or more endogenous proteins involved in glucose import and utilization in the cell can be reduced or eliminated. Preferably, these proteins are selected from the following:

[0139] i) Glucose PTS complex component IICB Glc (ptsG, e.g., Uniprot accession number P69786, or a functional variant thereof);

[0140] ii) β-glucoside PTS complex component IIABC Bgl (bglF, e.g., Uniprot accession number P08722, or a functional variant thereof);

[0141] iii) Mannose PTS complex component - IICD Man (manX, e.g., Uniprot accession number P69797, or a functional variant thereof);

[0142] iv) N - acetylglucosamine PTS complex component IIABC Nag (nagE, e.g., Uniprot accession number P09323, or a functional variant thereof);

[0143] v) Maltose / maltodextrin transport system (malX, e.g., Uniprot accession number P19642, or a functional variant thereof);

[0144] vi) Galactose / glucose high - affinity ABC transporter component (mglC, e.g., Uniprot accession number P23200, or a functional variant thereof);

[0145] vii) Trehalose / maltose / sucrose / trehalulose (TMSP)–ABC transporter (malF, e.g., Uniprot accession number P02916 or a functional variant thereof);

[0146] viii) Trehalose / maltose / sucrose / trehalulose (TMSP)–ABC transporter (malG, e.g., Uniprot accession number P68183, or a functional variant thereof);

[0147] ix) Galactose permease (galP, e.g., Uniprot accession number P0AEP1, or a functional variant thereof);

[0148] x) Glucose - proton symporter (glcP, e.g., Uniprot accession number O07563, or a functional variant thereof);

[0149] xi) Glucose facilitator (glf, e.g., Uniprot accession number P37747 or P21906, or a functional variant thereof);

[0150] xii) Glucose uptake protein (glcU, e.g., Uniprot accession number P40420, or a functional variant thereof);

[0151] xiii) Sodium / glucose transporter family (sglT, e.g., Uniprot accession number P96169, or a functional variant thereof); and

[0152] xiv) Glucokinase (glk, e.g., Uniprot accession number P0A6V8, or a functional variant thereof); and

[0153] xv) Hexose transporter (HXT).

[0154] Among them, the proteins in i)-vi) and xv) are all part of various glucose import complexes, which are usually composed of multiple proteins. The proteins in vii)-xiii) are single protein transporters identified in different bacterial species. The protein in xiv) is an example of a glucose-utilizing enzyme that phosphorylates glucose after it enters the cell. In the context of the present disclosure, it is preferred to reduce or eliminate the activity of membrane-bound transporters. Preferably, the genes to be mutated / deleted in the complex are indicated in italics in parentheses.

[0155] In some embodiments, the genetically modified microbial cell is Escherichia coli, which has reduced (limited) or no growth on glucose, wherein the functionality of one or more endogenous proteins involved in glucose import and utilization in the cell can be reduced or eliminated. Preferably, the protein is selected from the following:

[0156] i) Glucose PTS complex component IICB Glc (ptsG)

[0157] ii) β-Glucoside PTS complex component IIABC Bgl (bglF)

[0158] iii) Mannose PTS complex component - IICD Man (manX)

[0159] iv) N-Acetylglucosamine PTS complex component IIABC Nag (nagE)

[0160] v) Maltose / maltodextrin transport system (malX)

[0161] vi) Galactose / glucose high-affinity ABC transporter component (mglC);

[0162] vii) Galactose permease (galP); and / or

[0163] viii) Glucokinase (glk).

[0164] In the embodiments described herein, at least one genetically modified cell reduces or eliminates the activity of at least one PTS-dependent sugar transport system, and the sugar transport system is selected from the following:

[0165] i) Glucose PTS complex component IICB Glc ;

[0166] ii) β-Glucoside PTS complex component - IIABC Bgl ;

[0167] iii) Mannose PTS complex component - IICD Man ;

[0168] iv) N - acetylglucosamine PTS complex component - IIABC Nag ; and

[0169] v) Maltose / maltodextrin PTS complex - IICB malX

[0170] vi) Sorbitol PTS complex IICB slr

[0171] Preferably, in bacteria that do not grow on glucose or have limited growth on glucose (such as Escherichia coli), at least delete the ptsG gene of the glucose PTS complex component IICB Glc of the ptsG gene.

[0172] Transport and utilization of glycerol

[0173] In the embodiments described herein, cells growing on glycerol have at least one glycerol transport system. This glycerol transport system can be selected from glycerol facilitator (glpF) or glycerol / H + -symporter (stl1).

[0174] In other embodiments, glycerol entry into the cytosol of a microorganism can be reduced or blocked by mutating or deleting one or more sequences encoding proteins that affect glycerol import ability, such as deleting or mutating the nucleic acid sequences encoding glycerol facilitator (glpF) or glycerol / H + -symporter (stl1).

[0175] To utilize glycerol for energy, cells phosphorylate glycerol to glycerol - 3 - phosphate after glycerol enters the cytosol, and this phosphorylation is carried out by glycerol kinase (glpK). Therefore, preventing the formation of glycerol - 3 - phosphate may also help prevent cells from using glycerol as a carbon source for growth. For more information on glycerol utilization in various bacteria, see the review in Lin Ann. Rev. Microbial. 1976 30:535 - 78.

[0176] In some embodiments, the growth of the genetically modified cells on glycerol is reduced or there is no growth, wherein the functionality of one or more endogenous proteins involved in glycerol import and utilization in the cells can be reduced or eliminated. Preferably, the reduction or elimination of the protein activity is achieved by deleting or mutating one or more nucleic acid sequences encoding proteins selected from glpF (e.g., Uniprot accession number, or a functional variant thereof), stl1 (e.g., Uniprot accession number, or a functional variant thereof), and glpK (e.g., Uniprot accession number, or a functional variant thereof).

[0177] If it is desired to reduce or eliminate the growth of bacteria (e.g., Escherichia coli) on glycerol, it is preferred to reduce or eliminate the functionality of the glycerol transporter (also known as glycerol facilitator (glpF)), for example, by mutating or deleting the glpF gene in the cells. In addition, the activity of glycerol kinase can also be reduced or eliminated, for example, by mutating or deleting the glpK gene in the cells.

[0178] In embodiments of the methods described herein, the functionality of proteins involved in glucose and / or glycerol import and utilization is reduced or eliminated by completely or partially inactivating one or more genes selected from ptsG, bglF, manX, nagE, malX, mglC, glk, and glpF.

[0179] Transport and utilization of sucrose

[0180] In the embodiments described herein, the cells growing on sucrose have at least one sucrose transport system. This sucrose transport system can be a PTS-dependent sucrose (suc) utilization system. Alternatively, the cells can grow on sucrose by having an active sucrose invertase or sucrose hydrolase protein in the outer membrane or periplasmic membrane (if present), which are capable of cleaving sucrose into glucose and fructose, and then the cells can take up these glucose and fructose through the relevant fructose and glucose transport systems (see, for example, WO 2013 / 087884).

[0181] In some embodiments, a genetically modified microbial cell capable of growing on sucrose comprises one or more nucleic acid sequences encoding a PTS-dependent sucrose utilization system. For example, the PTS-dependent sucrose utilization system can be encoded by scrY, scrA, scrB, and optionally scrR (see, for example, WO2015 / 197082), wherein the gene scrA encodes the sucrose transporter IIScr enzyme (e.g., SEQ ID NO:98 or ncbi sequence ID: CAA40658.1 or a functional variant thereof), which converts extracellular sucrose into intracellular sucrose-6-phosphate by active transport across the cell membrane and concomitant phosphorylation. The sucrose-specific ScrY porin (e.g., SEQ ID NO:97 encoded by scrY or ncbi sequence ID: CAA40657.1 or a functional variant thereof) facilitates the diffusion of sucrose through the outer membrane. The ScrB invertase (e.g., SEQ ID NO:99 encoded by scrB or ncbi sequence ID: WP_000056853.1 or a functional variant thereof) hydrolyzes the accumulated sucrose-6-phosphate into glucose-6-phosphate and fructose. scrR encodes a LacI family DNA-binding transcriptional regulator (e.g., SEQ ID NO:100 or ncbi sequence ID: WP_000851062.1 or a functional variant thereof).

[0182] The Escherichia coli csc PTS-dependent sucrose system is described in WO2015 / 150328 and is expressed by the cscABKR gene cluster (SEQ ID NO:110), which encodes: a sucrose permease (e.g., cscB with UniProt accession number P30000.1 or a functional variant thereof), a fructokinase (e.g., csck with GenBank accession number EDV65567.1 or a functional variant thereof), a sucrose hydrolase (e.g., cscA with NCBI accession number WP_175214520.1 or a functional variant thereof), and a transcriptional repressor (e.g., cscR with GenBank accession number AJA27326.1 or a functional variant thereof).

[0183] Alternatively, a genetically modified microbial cell capable of growing on sucrose comprises a nucleic acid encoding a sucrase or a sucrose hydrolase that enables the cell to assimilate sucrose. The sucrase can be, for example, a glycoside hydrolase and a sucrose-6-phosphate hydrolase (e.g., SacC_AgaI of GeneBank ID: WP_103853210.1 or SEQ ID NO:111, or a functional variant thereof), or a β-fructofuranosidase (e.g., Bff of GeneBank ID: BAD18121.1 or SEQ ID NO:112, or a functional variant thereof). Since the sucrose hydrolase or sucrase converts sucrose into glucose and fructose in the periplasmic space of the cell, etc., the cell preferably can grow on glucose or fructose, which means that in a two-strain system, the other genetically modified cell should not be able to grow or only be able to grow to a limited extent on fructose and / or glucose.

[0184] In an embodiment where the carbon source is sucrose and its assimilation is achieved by a sucrase or a sucrose hydrolase, the other cell in the two-strain system preferably grows on glycerol or galactose.

[0185] In some embodiments, the growth of the genetically modified cell on sucrose is reduced or it does not grow, wherein the functionality of one or more endogenous proteins involved in sucrose import and utilization in the cell can be reduced or eliminated. Preferably, the reduction or elimination of the protein activity is achieved by deleting or mutating one or more nucleic acid sequences selected from the nucleic acid sequences encoding proteins in the PTS-dependent sucrose utilization system, for example, by mutating or deleting the cscABKR gene cluster (SEQ ID NO:110, or a functional variant thereof) (if present and functional in the cell). In particular, the deletion or mutation of the sucrose permease gene (e.g., Uniprot accession number P30000.1, or a functional variant thereof) or scrY (e.g., Uniprot accession number B1LQA1, or a functional variant thereof) is crucial for rendering the strain unable to grow or having reduced growth on sucrose. Many non-pathogenic Escherichia coli cells have lost the ability to grow on sucrose, so it is usually not necessary to prevent their growth on sucrose by mutation because the cells themselves no longer have this ability.

[0186] Transport and utilization of galactose

[0187] In the embodiments described herein, the cell growing on galactose has at least one galactose transport system. The galactose transport system can be selected from the galactose:H+ symporter GalP, the galactose / glucose ABC transporter (mglABC) system, the PTSLac (lacFE) system, and / or the sodium / glucose transporter family (sglT).

[0188] Galactose that enters the cell via GalP, mglABC, or sglT is converted to galactose-1-phosphate (Gal1P) by the action of galactokinase, which is then metabolized to α-glucose-1-phosphate (G1P) via the Leloir pathway (galMKTE).

[0189] Galactose imported into the cell via the PTSLac (lacFE) system is converted to galactose-6-phosphate (Gal6P) and further metabolized to triose phosphates via the Tag6P pathway (lacABCD).

[0190] Cells growing on galactose preferably also have a functional galactokinase (galK), a functional Leloir pathway, and / or a Tag6P pathway.

[0191] In other embodiments, galactose entry into the microbial cytosol can be reduced or blocked by mutating or deleting one or more sequences encoding proteins that affect galactose import ability, such as deleting or mutating the nucleic acid sequences of the galactose:H+ symporter GalP, the galactose / glucose ABC transporter (mglABC) system, the sodium / glucose transporter family (sglT), or the PTSLac (lacFE) system.

[0192] To utilize galactose for energy, the cell phosphorylates galactose to Gal1P or Gal6P. Thus, mutating, deleting, or blocking the enzymes that convert galactose to Gal1P or Gal6P may also prevent the cell from using galactose as a growth carbon source.

[0193] In some embodiments, the growth of genetically modified cells on galactose is reduced or absent, wherein the functionality of one or more endogenous proteins involved in galactose import and utilization in the cell can be reduced or eliminated. Preferably, the reduction or elimination of the protein activity is achieved by deleting or mutating one or more nucleic acid sequences selected from those encoding galP (e.g., Uniprot accession number P0AEP1, or a functional variant thereof), mglC (e.g., Uniprot accession number P23200, or a functional variant thereof), lacF (e.g., Uniprot accession number P24400, or a functional variant thereof), galK (e.g., Uniprot accession number P0A6T3, or a functional variant thereof), and / or sglT (e.g., Uniprot accession number P96169, or a functional variant thereof).

[0194] Transport and utilization of fructose

[0195] In the embodiments described herein, the cells growing on fructose have at least one fructose transport system. The fructose transport system can be selected from the fructose PTS complex component IIABC Fru , the glucose PTS complex component IICB Glc and the fructose transporter FruP.

[0196] The fructose imported into the cells is converted into fructose-1-phosphate (fru1P) or fructose-6-phosphate (fru6P) by fructokinase.

[0197] In other embodiments, the entry of fructose into the cytosol of the microorganism can be reduced or blocked by mutating or deleting one or more sequences encoding proteins that affect fructose transport ability, such as deleting or mutating the nucleic acid sequences encoding the fructose PTS complex component IIABC Fru , the glucose PTS complex component IICB Glc and the fructose transporter FruP.

[0198] To utilize fructose for energy, the cells phosphorylate fructose to fru1P or fru6P. Therefore, mutating, deleting or blocking the enzymes that convert fructose to fru1P or fru6P may also prevent the cells from using fructose as a growth carbon source.

[0199] In some embodiments, the growth of the genetically modified cells on fructose is reduced or they do not grow, wherein the functionality of one or more endogenous proteins involved in fructose import and utilization in the cells may be reduced or eliminated. Preferably, the reduction or elimination of protein activity is achieved by deleting or mutating one or more nucleic acid sequences selected from those encoding fruA (e.g., Uniprot accession number P20966, or a functional variant thereof), ptsG (e.g., Uniprot accession number P69786, or a functional variant thereof) or FruP (e.g., Uniprot accession number F4TKS5, or a functional variant thereof).

[0200] Transport and utilization of maltose

[0201] In the embodiments described herein, the cells growing on maltose have at least one maltose transport system.

[0202] Such a maltose transport system includes the MalFGK ABC superfamily transport system, which can transport maltose across the cytoplasmic membrane of Escherichia coli. The MalFGK transport system is a heterotetrameric complex composed of the integral membrane proteins MalF and MalG, which bind to two units of the peripheral membrane protein MalK. MalK has ATP-binding properties and can thus provide energy for the maltose permease encoded by malF and malG.

[0203] Alternatively, the maltose transport system can be selected from the maltose / maltodextrin PTS complex - IICB encoded by malX mal (e.g., Uniprot accession number P19642, or a functional variant thereof). The PTS enzyme II protein encoded by malX can recognize glucose and maltose as substrates.

[0204] In embodiments where the maltose / maltodextrin PTS complex is used for a strain to grow on maltose, it means that the second strain does not grow or has limited growth on both glucose and maltose.

[0205] The maltose imported into the cell is converted to glucose by amylomaltase encoded by malQ (e.g., UniProt accession number P15977.2 or a functional variant thereof) or an alternative maltase from other species. The glucose is then phosphorylated as described in the "Glucose transport" section above.

[0206] In other embodiments, the import of maltose into the cytosol of a microorganism can be reduced or blocked by mutating or deleting one or more sequences encoding proteins that affect maltose import ability, such as deleting or mutating the nucleic acid sequences encoding components of the MalFGKABC superfamily transport system or component IICB of the maltose PTS complex mal . In addition, the deletion of glucokinase (glk, e.g., Uniprot accession number P0A6V8, or a functional variant thereof) will prevent the utilization of maltose as a carbon source because glucose needs to be phosphorylated to be converted into energy by the cell.

[0207] In some embodiments, the growth of the genetically modified cell on maltose is reduced or absent, wherein the functionality of one or more endogenous proteins involved in maltose import and utilization in the cell can be reduced or eliminated. Preferably, the reduction or elimination of protein activity is achieved by deleting or mutating one or more nucleic acid sequences selected from those encoding MalF (e.g., UniProt accession number P02916.1 or a functional variant thereof), MalG (e.g., GenBank: AAC77002.1 or a functional variant thereof), MalK (e.g., UniProt accession number P02916.1 or a functional variant thereof), the PTS complex - IICB encoded by malX mal (e.g., Uniprot accession number P19642 or a functional variant thereof) and / or the glucokinase encoded by glk (e.g., Uniprot accession number P0A6V8 or a functional variant thereof).

[0208] Transport and utilization of arabinose

[0209] In the embodiments described herein, the cells growing on arabinose have at least one arabinose transport system.

[0210] The arabinose transport system is the AraFGH arabinose transporter, which is a member of the ATP-binding cassette (ABC) transporter superfamily. AraF is a periplasmic binding protein (e.g., UniProt accession number P02924 or a functional variant thereof), AraH is a membrane component (e.g., UniProt accession number P0AE26 or a functional variant thereof), and AraG is the ATP-binding component of the ABC transporter (e.g., UniProt accession number P0AAF3 or a functional variant thereof).

[0211] Alternatively, the arabinose transport system can be selected from the arabinose-proton symporter AraE (e.g., UniProt accession number P0AE24 or P96710 or a functional variant thereof).

[0212] In some embodiments, the growth of the genetically modified cell on arabinose is reduced or absent, wherein the functionality of one or more endogenous proteins involved in arabinose import and utilization in the cell can be reduced or eliminated. For example, arabinose import into the cytosol of the microorganism can be blocked by mutating or deleting one or more sequences encoding proteins that affect arabinose import ability, such as deleting or mutating the nucleic acid sequences encoding components of the AraFGH ABC superfamily transport system or the maltose arabinose-proton symporter AraE.

[0213] Transport and utilization of sorbitol

[0214] In the embodiments described herein, the cells growing on sorbitol have at least one sorbitol transport system.

[0215] Transport of sorbitol into prokaryotic cells is facilitated by the phosphoenolpyruvate-dependent phosphotransferase system (PTS). The sorbitol-specific enzyme IIB and IIC (EIIBC srl ) components are responsible for binding to sorbitol and initiating its transport into the cell, and the enzymes are encoded by srlA (e.g., Uniprot accession number P56579 or O32333 or a functional variant thereof) and srl E (e.g., Uniprot accession number P56580 or O32332, or a functional variant thereof), respectively. As part of the PTS process, the incoming sorbitol molecule is simultaneously phosphorylated by sorbitol kinase (EIIA srl ) encoded by the gene srlB (e.g., Uniprot accession number P05706 or A5I7D9, or a functional variant thereof).

[0216] In embodiments where the sorbitol PTS complex is used for the growth of a strain on sorbitol, it means that a second strain does not grow or has limited growth on glucose and maltose.

[0217] In some embodiments, the genetically modified cells grow poorly or not at all on sorbitol, and the functionality of one or more endogenous proteins involved in sorbitol import and utilization can be reduced or eliminated in the cells. For example, the import of sorbitol into the cytosol of a microorganism is blocked by mutating or deleting one or more sequences encoding proteins that affect the sorbitol PTS system.

[0218] Dual-strain system with intermediate oligosaccharide uptake

[0219] In this application of the dual-strain system, the second strain is designed to use the product produced by the first strain as an initial substrate for a glycosylation reaction to produce oligosaccharides, thereby generating a second oligosaccharide. The intermediate-dependent dual-strain system utilizes the characteristics of the dual-strain system. Additionally, the second genetically modified cell is capable of importing the disaccharide or oligosaccharide produced by the first genetically modified cell. In this way, the first oligosaccharide serves as an intermediate oligosaccharide and a receptor (as Figure 1 shown) during the production of the second oligosaccharide by the second genetically modified cell.

[0220] A second aspect described herein is a method for producing an oligosaccharide having at least three, such as at least four, monosaccharide units, the method comprising the step of co-culturing first and second genetically modified microbial cells in a culture medium, wherein,

[0221] a) the first genetically modified microbial cell is capable of producing a first disaccharide or a first oligosaccharide having at least three monosaccharide units, and wherein the first genetically modified cell

[0222] i) is capable of growing on a first carbon source while growing poorly or not at all on a second carbon source, and

[0223] ii) comprises one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase; and

[0224] iii) comprises at least one pathway for producing an activated sugar nucleotide from the first carbon source;

[0225] and

[0226] iv) is capable of exporting the first oligosaccharide into the culture medium; and

[0227] b) the second genetically modified microbial cell is capable of internalizing the first oligosaccharide having at least three monosaccharide units, and wherein the second genetically modified cell

[0228] i) is capable of growing on the second carbon source while growing poorly or not at all on the first carbon source; and

[0229] ii) comprises one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase; and

[0230] iii) comprising a biosynthetic pathway for producing an activated sugar nucleotide from the second carbon source;

[0231] wherein the second genetically modified microbial cell produces an oligosaccharide having at least three monosaccharide units, such as at least four monosaccharide units.

[0232] In some embodiments, the first genetically modified microbial cell produces an intermediate disaccharide or oligosaccharide having three, four, or five monosaccharide units. Preferably, the disaccharide or oligosaccharide produced by the first genetically modified cell is selected from LacNAc, LNB, Lewis A, Lewis X, 2’FL, 3FL LNT-II, LNT, LNnT, LNFP-I, LST-c, or LST-a.

[0233] In some embodiments, the first genetically modified cell producing an intermediate disaccharide or oligosaccharide comprises at least one glycosyltransferase selected from the group consisting of: β-1,3-N-acetylglucosaminyltransferase, β-1,3-galactosyltransferase, β-1,4-galactosyltransferase, α-1,2-fucosyltransferase, α-1,3-fucosyltransferase, α-1,3 / 4-fucosyltransferase, and α-2,3-sialyltransferase.

[0234] In some embodiments, the first genetically modified cell producing an intermediate oligosaccharide comprises at least two glycosyltransferases. Preferably, such cells produce LNT or LNnT.

[0235] In some embodiments, the first genetically modified cell producing an intermediate oligosaccharide comprises at least three glycosyltransferases. Preferably, such cells produce LNFP-I, LST-c, or LST-a.

[0236] In a further embodiment, the second genetically modified cell is capable of importing a disaccharide or an oligosaccharide having three, four, or five monosaccharide units produced by the first genetically modified cell. The disaccharides lactose, LacNAc, and LNB can be imported via lactose permease. To be able to import oligosaccharides, the second genetically modified cell can be additionally modified to comprise at least one recombinant nucleic acid sequence and / or cluster of recombinant nucleic acid sequences encoding a transporter protein and / or protein cluster capable of importing the intermediate oligosaccharides produced by the first genetically modified cell.

[0237] The known wild-type lactose permease (LacY) can transport the disaccharide lactose from the outside of the cell into Escherichia coli cells. The wild-type lactose permease can also import 2’FL, LNB, and LAcNAc, so when the first genetically modified cell produces 2’FL, LNB, or LacNAc, it can serve as an import protein in the second genetically modified cell. However, it is preferred to modify the lactose permease to have a higher affinity for 2’FL and potentially 3FL or LNT-II than for lactose.

[0238] Mutant variants of LacY have been described that are capable of transporting the trisaccharide maltotriose (Olsen et al. 1993 J Bacteriol. 175(19):6269-75). In the present disclosure, these mutants are described as potential import proteins for trisaccharides (receptor oligosaccharides / HMO precursor molecules) related to HMO production, such as 2-fucosyllactose (2’FL), 3-fucosyllactose (3FL), and lacto-N-triose (LNT-II).

[0239] The second genetically modified cell according to the invention can comprise a recombinant nucleic acid sequence encoding a transporter protein that can import an intermediate (receptor) oligosaccharide of at least three monosaccharide units produced by the first genetically modified cell into said cell, wherein the transporter protein is a mutant lactose permease (LacY), as shown in Table 1.

[0240] Table 1. List of exemplary mutants of Escherichia coli DH1 K12 lactose permease LacY (SEQ ID NO: 1) that can be used to import 2’FL, 3FL, or LNT-II

[0241]

[0242]

[0243] In a preferred embodiment, the lactose permease variants in Table 1 have a higher affinity for 2’FL, 3FL, and / or LNT-II than for lactose.

[0244] This may be particularly advantageous if the first genetically modified cell needs to internalize lactose to produce 2’FL, 3FL, or LNT-II, otherwise the second genetically modified cell will also take up lactose, and lactose may also serve as a substrate (receptor) for the second genetically modified cell, leading to the formation of unwanted by-products.

[0245] In embodiments where the intermediate oligosaccharides produced by the first genetically modified cell are not 2’FL, LacNAc or LNB, it is preferred that the second strain does not contain any functional endogenous lactose permease. The endogenous lactose permease can be deleted or its function can be eliminated by point mutations (such as stop codons or truncations). The second genetically modified strain lacks functional lactose permease, which will prevent the glycosyltransferases in the second strain from glycosylating lactose unnecessarily, thereby reducing the formation of by-products.

[0246] Generally, by-product oligosaccharides or by-product HMOs are i) oligosaccharide or HMO precursors that are further modified intracellularly to produce the desired oligosaccharide or HMO (product HMO / oligosaccharide); or ii) further modifications to the desired product oligosaccharide. In some embodiments, it may be desirable to produce a large amount of product HMO / oligosaccharide and a small amount of by-product HMO / oligosaccharide. The abundance of the desired oligosaccharide / HMO is, for example, at least 20% of the total oligosaccharide / HMO produced, such as at least 30%, such as at least 50%, such as at least 60%, such as at least 75%, such as at least 80%, such as at least 85%, such as at least 90%. The term "most abundant oligosaccharide / HMO" (in relation to production by genetically modified cells) means that if the cells are fermented alone, it is the most prevalent oligosaccharide / HMO in the culture medium / cells. By-products in HMO production can be, for example, 2’FL and / or 3FL in DFL production, or LNT-II in the formation of LNT and LNnT, in which case not all precursor oligosaccharides are modified in the second glycosylation reaction.

[0247] Another way to prevent the second cell from taking up lactose added to the medium as a substrate for the first cell is to engineer the first cell to produce lactose itself (see, for example, Parschat et al. 2020 ACS Synth. Biol. 9:2784 - 2796). In some embodiments, the first genetically modified microbial cell contains a β-1,4-galactosyltransferase that allows galactosylation of glucose monosaccharides to produce lactose intracellularly, and wherein the glucose kinase activity that converts glucose to glucose-6-phosphate in the cell is reduced or eliminated. Preferably, the lactose-producing cell does not use glucose as a carbon source for growth because this requires the conversion of glucose to glucose-6-phosphate, leaving less free glucose to be converted to lactose. More preferably, the lactose-producing cell uses sucrose or maltose as a carbon source.

[0248] In some embodiments, the first genetically modified cell can produce LNB, LacNAc, Lewis A, Lewis X, 2’FL, 3FL, LNT-II, LNT, LNnT, LNFP-I or LST-a without adding lactose to the medium.

[0249] In some embodiments, when the first genetically modified cell produces an oligosaccharide of four or five monosaccharide units, other transporter molecules may be required in the second genetically modified cell, particularly transporters capable of internalizing LNT, LNnT, LNFP-I, or LST-a.

[0250] The second genetically modified cell according to the invention may comprise a recombinant nucleic acid or a cluster of recombinant nucleic acid sequences encoding a transporter and / or a protein cluster capable of importing an intermediate (acceptor) oligosaccharide of at least three or four monosaccharide units produced by the first genetically modified cell into said cell.

[0251] In Gram-positive (Gram+) bacteria, particularly in the genera Bifidobacterium, Roseburia, and Eubacterium, importer proteins have been identified that are capable of importing oligosaccharides of at least three or four monosaccharide units or more.

[0252] Table 2 lists MFS transporters from Gram-positive bacteria and ABC transporter clusters from Gram-positive bacteria that are capable of importing acceptor oligosaccharides of at least three or four monosaccharide units into the cell. The terms "transporter" and "importer" can be used interchangeably.

[0253] The intermediate oligosaccharide produced by the first genetically engineered cell is preferably a precursor of a more complex HMO and can act as an acceptor oligosaccharide when imported into the second genetically modified cell. Table 2 lists which intermediate oligosaccharide / acceptor oligosaccharide this transporter is expected to import into the cell.

[0254] Table 2: ABC transporters and MFS transporters from Gram-positive bacteria, indicating the precursor oligosaccharides expected to be imported by this transporter. The ABC transporters consist of 3 to 4 genes. For ease of reference, each transporter has been assigned a transporter ID (TPID).

[0255]

[0256]

[0257]

[0258] Typically, the second genetically modified cell lacks the enzymatic activity to degrade the acceptor oligosaccharide of at least three, four, or five monosaccharide units imported into the cell.

[0259] Once the intermediate oligosaccharide produced by the first cell is imported into the second cell, it can serve as an acceptor molecule for further glycosylation by one or more selected glycosyltransferases.

[0260] In an embodiment, the second genetically modified cell comprises at least one glycosyltransferase selected from β-1,3-N-acetylglucosaminyltransferase, β-1,3-galactosyltransferase, β-1,4-galactosyltransferase, α-1,2-fucosyltransferase, α-1,3-fucosyltransferase, α-1,3 / 4-fucosyltransferase, α-1,4-fucosyltransferase, α-2,3-sialyltransferase, and α-2,6-sialyltransferase.

[0261] In an embodiment, one or more oligosaccharides (e.g., HMO) produced by the second genetically modified microbial cell have at least three, e.g., at least four monosaccharide units, and are selected from Lewis A, Lewis X, sialyl-LacNAc, sialyl-LNB, sialyl-Lewis X, sialyl-Lewis A, Lewis B, Lewis Y, DFL, FSL, LNT, LNnT, LST-a, LST-c, LNFP-I, LNFP-II, LNFP-III, LNFP-V, LNFP-VI, LNDFH-I, LNDFH-II, LNDFH-III, LST-a, LST-b, LST-c, LST-d, DSLNT, pLNH, pLNnH, LNH, LNnH, (D)F-LNH-I, (D)F-LNH-II, (D)F-LNH-III, F-para-LNH-I, DF-para-LNH, DF-para-LNnH, TF-LNH, FLSTb, FLSTa, FLST-c, S-LNH, S-LNnH-I, FS-LNH, FS-LNnH-I, and DS-F-LNH-II, or a mixture thereof.

[0262] Preferably, both the first and second genetically engineered cells express a transporter (efflux protein) that can efflux the disaccharide or oligosaccharide produced by the cell. Transporters for oligosaccharide efflux will be described in the corresponding section below. In addition to heterologous transporters, endogenous transporters located in the cytoplasmic membrane and / or outer membrane, such as porins, can also assist in the expected efflux of the precursor oligosaccharide of the first genetically modified cell or the final oligosaccharide produced by the second genetically modified cell.

[0263] One disclosed embodiment is a method for producing LNnT, the method comprising co-culturing:

[0264] a. A first genetically modified microbial cell that is capable of growing on a first carbon source while being growth restricted or not growing on a second carbon source, and that comprises:

[0265] i. A recombinant nucleic acid sequence encoding β-1,3-N-acetylglucosaminyltransferase, and

[0266] ii. Optionally, a recombinant nucleic acid sequence encoding an MFS transporter capable of exporting LNT-II into the extracellular culture medium, and

[0267] b. A second genetically modified microbial cell that is capable of growing on the second carbon source while being growth-limited or non-growing on the first carbon source, and comprising:

[0268] i. A recombinant nucleic acid sequence encoding a protein or protein complex capable of importing the oligosaccharide produced by the first genetically modified microbial cell, the recombinant nucleic acid sequence being recombinant and encoding a transporter capable of importing LNT-II, the transporter being selected from the mutant lacY transporters or potential LNT-II transporters in Table 2;

[0269] ii. A recombinant nucleic acid sequence encoding β-1,4-galactosyltransferase;

[0270] iii. Optionally, a recombinant nucleic acid encoding the MFS transporter vag; and

[0271] c. Harvesting LNnT produced in the co-culture, and

[0272] wherein, compared with the LNnT produced by a single cell, the LNT-II produced by the method is significantly less, and / or the pLNnH by-product is extremely rare or absent.

[0273] In a preferred embodiment, the second genetically modified cell is unable to import lactose, for example, due to its lack of a functional lactose permease.

[0274] One disclosed embodiment is a method for producing LNT, the method comprising co-culturing:

[0275] a. A first genetically modified microbial cell that is capable of growing on a first carbon source while being growth-limited or non-growing on a second carbon source, and comprising:

[0276] i. A recombinant nucleic acid sequence encoding β-1,3-N-acetylglucosaminyltransferase, and

[0277] ii. Optionally, a recombinant nucleic acid sequence encoding an MFS transporter capable of exporting LNT-II into the extracellular culture medium, and

[0278] b. A second genetically modified microbial cell that is capable of growing on the second carbon source while being growth-limited or non-growing on the first carbon source, which comprises:

[0279] i. A recombinant nucleic acid sequence encoding a protein or protein complex capable of importing the oligosaccharides produced by the first genetically modified microbial cell, the recombinant nucleic acid sequence being recombinant and encoding a transporter capable of importing LNT-II, the transporter being selected from the mutant lacY transporters or potential LNT-II transporters in Table 2;

[0280] ii. A recombinant nucleic acid sequence encoding β-1,3-galactosyltransferase;

[0281] iii. Optionally, a recombinant nucleic acid encoding an MFS selected from nec or YberC;

[0282] c. Harvesting the LNT produced in the co-culture,

[0283] wherein, compared with the LNT produced by a single cell, the LNT-II produced by the method is significantly less, and / or the pLNH2 by-product is extremely rare or absent.

[0284] In a preferred embodiment, the second genetically modified cell is unable to import lactose, for example, due to its lack of a functional lactose permease.

[0285] An example of an LNT-II MFS output protein is the putative metabolite transporter YjhB from Escherichia coli.

[0286] One disclosed embodiment is a method for producing LNFP-III, LNFP-VI, and / or LNDFH-III, the method comprising co-culturing

[0287] a. A first genetically modified microbial cell that is capable of growing on a first carbon source while being restricted or not growing on a second carbon source, and comprising:

[0288] i. A recombinant nucleic acid sequence encoding β-1,3-N-acetylglucosaminyltransferase, and

[0289] ii. A recombinant nucleic acid sequence encoding β-1,4-galactosyltransferase, and

[0290] iii. Optionally, a recombinant nucleic acid sequence encoding an MFS transporter capable of exporting LNnT into the extracellular culture medium, and

[0291] b. A second genetically modified microbial cell that is capable of growing on the second carbon source while being restricted or not growing on the first carbon source, which comprises:

[0292] i. A nucleic acid sequence encoding a protein or protein complex capable of importing the oligosaccharides produced by the first genetically modified microbial cell, the nucleic acid sequence being recombinant and encoding a transporter capable of importing LNnT; and

[0293] ii. The recombinant nucleic acid sequence encoding a glycosyltransferase encodes an α-1,3-fucosyltransferase or an α-1,3 / 4-fucosyltransferase; and

[0294] c. Harvesting LNFP-III, LNFP-VI and / or LNDFH-III produced in the co-culture,

[0295] wherein the LNnT, pLNnH and LNT-II by-products and their fucosylated derivatives produced by the method are significantly reduced compared to LNFP-III, LNFP-VI and / or LNDFH-III produced by a single cell.

[0296] In a preferred embodiment, the second genetically modified cell cannot import lactose because, for example, it lacks functional lactose permease.

[0297] Dual-strain mixed system

[0298] Another application of the dual-strain mixed system is very suitable for producing complex oligosaccharides, such as complex fucosylated and / or sialylated oligosaccharides of at least five (e.g., at least six) monosaccharide units, or neutral non-fucosylated oligosaccharides. However, it can also be used to produce shorter oligosaccharides of three or four monosaccharide units.

[0299] Currently, complex oligosaccharides are mainly produced by in vitro enzymatic synthesis. The in vitro enzymatic process relies on the use of donor oligosaccharides (HMO) and acceptor oligosaccharides (HMO), which generate a third oligosaccharide (complex HMO) under the catalysis of an enzyme with transglycosidase activity. However, due to the nature of the enzymatic reaction, the oligosaccharides (HMO) generated by this pathway are always a mixture of the donor, acceptor, and third oligosaccharide (complex HMO), as well as a by-product portion (e.g., lactose) released from the donor substrate (leaving group) due to the enzymatic reaction equilibrium (e.g., see WO2012 / 156897, WO2012 / 156898, and WO2016 / 063262). In addition, this enzymatic process uses separately produced and purified donor and acceptor substrates, which increases the process cost. Moreover, if the by-product released from the donor substrate is lactose, a large amount of purification is required to remove the large amount of unwanted lactose. In addition, due to the kinetic equilibrium in the reaction, traditional enzymatic processes usually cannot remove one of the substrates.

[0300] Currently, for smaller fucosylated, sialylated, and neutral core HMOs, the use of a biological production system of in vivo fermented HMOs is the preferred production method (for details, see Bych et al. 2019, Current Opinion in Biotechnology 56:130–137). However, for more complex HMOs, the fermentation pathway may face challenges in exporting HMOs from the cells to the culture medium, which is crucial for achieving high yields on an industrial scale.

[0301] The two strain hybrid process described herein combines the advantages of in vivo biological production systems and in vitro enzymatic oligosaccharide (e.g., HMO) production systems into a hybrid production system that combines fermentation and enzymatic steps in the same vessel. As Figure 2 shown, two strains can be co-cultured as described in the "Two Strain System" section above, and the oligosaccharides (e.g., HMOs) produced by the two strains can react under the catalysis of transglycosidase in the culture medium to form a third complex oligosaccharide.

[0302] Surprisingly, the enzymatic transglycosylation reaction can be effectively carried out in the culture medium of an ongoing fermentation process to form complex HMOs, as shown in Example 2. In the hybrid process, the reaction conditions are determined by the fermentation conditions (such as temperature, pH, oxygen, carbon dioxide, stirring, etc.), and the reaction environment is significantly more complex. The fermentation broth contains a variety of substrates and metabolites, and may even include proteases released from the cells. In contrast, the traditional in vitro enzymatic process shown in Example 2 only contains two initial substrates (acceptor and donor) and transglycosidase. To our knowledge, this is the first time an enzymatic reaction has been used to synthesize a larger molecule from two smaller molecules simultaneously produced during an ongoing fermentation process.

[0303] The hybrid production system described in this disclosure includes a co-culture fermentation step and an enzymatic step, and these two steps can be carried out in the same vessel as the co-culture.

[0304] In a dual strain mixing process, the first genetically modified cell produces a disaccharide or preferably a first oligosaccharide, which is secreted / output into the fermentation medium. This first oligosaccharide can serve either as a donor oligosaccharide or as a receptor oligosaccharide in subsequent transglycosylation reactions occurring in the medium. The first genetically modified cell is preferably genetically engineered to be able to efficiently produce the first oligosaccharide / HMO and secrete it into the medium. Additionally, the process includes a second genetically modified cell that produces a second oligosaccharide, which is preferably secreted / output into the fermentation medium. This second oligosaccharide can serve either as a donor oligosaccharide or as a receptor oligosaccharide in subsequent transglycosylation reactions occurring in the medium. In embodiments where the complex oligosaccharides produced by the present method are fucosylated and / or sialylated oligosaccharides, the donor oligosaccharide contains a fucosyl or sialic acid residue. In a preferred embodiment, the donor oligosaccharide is fucosyllactose or sialyllactose.

[0305] In embodiments where the complex oligosaccharides produced by the present method are neutral core oligosaccharides (non-fucosylated), the donor oligosaccharide is at least six monosaccharide units, such as hexa-, octa-, deca- or dodeca-oligosaccharides, and the donor oligosaccharide is preferably LNT or LNnT.

[0306] In the production of complex HMOs, the first and second oligosaccharides are preferably HMOs. The genetically modified cells are engineered to be able to efficiently produce oligosaccharides / HMOs when grown in the same vessel and release or secrete them into the medium. Preferably, the co-culture of the cells is controlled by their ability to grow on different carbon sources on which the other cell cannot grow significantly (limitedly). Once the cells have produced a portion of the first and second oligosaccharides / HMOs, they react under the catalysis of transglycosidases in the medium to form a third complex oligosaccharide / HMO, such as sialylated and / or fucosylated and / or hexa- or octa-neutral core oligosaccharide / HMO.

[0307] A third aspect described herein relates to a method for producing an oligosaccharide of at least three monosaccharide units or a complex oligosaccharide of at least four, such as at least five monosaccharide units, from donor and receptor oligosaccharides produced by first and second genetically modified cells, the method comprising the steps of:

[0308] a) co-culturing the first and second genetically modified cells in a medium, wherein

[0309] i) the first genetically modified microbial cell is capable of producing a disaccharide or a first oligosaccharide of at least three monosaccharide units, and wherein the genetically modified cell

[0310] · comprises one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase; and

[0311] · comprises at least one pathway for producing nucleotide-activated sugars from a first carbon source; and

[0312] and

[0313] ·Preferably capable of exporting said first oligosaccharide into the culture medium; and

[0314] ii) said second genetically modified microbial cell is capable of producing a second oligosaccharide of at least three monosaccharide units, and wherein said genetically modified cell

[0315] ·Comprises one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase, and

[0316] ·Comprises a biosynthetic pathway for preparing said activated sugar nucleotide from a second carbon source;

[0317] ·Preferably capable of exporting said second oligosaccharide into the culture medium, and

[0318] b) Making an enzyme with transglycosylase activity available in the culture medium, and

[0319] c) Incubating said first oligosaccharide or disaccharide and said second oligosaccharide together with said transglycosylase in the culture medium, wherein said first oligosaccharide or disaccharide reacts with said second oligosaccharide to form a third oligosaccharide of at least three, such as four, such as five monosaccharide units.

[0320] The embodiments described herein relate to a method for producing sialylated and / or fucosylated oligosaccharides of at least four monosaccharide units from donor oligosaccharides and acceptor oligosaccharides produced from first and second genetically modified cells, said method comprising the steps of:

[0321] a) Co-culturing first and second genetically modified cells in a culture medium, wherein

[0322] i) said first genetically modified microbial cell is capable of producing a disaccharide or a first oligosaccharide of at least three monosaccharide units, and wherein said genetically modified cell

[0323] ·Capable of growing on a first carbon source while being growth-limited or non-growing on a second carbon source, and

[0324] ·Comprises one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase; and

[0325] ·Comprises at least one pathway for producing nucleotide-activated sugar from said first carbon source; and

[0326] ·Preferably capable of exporting said first oligosaccharide into the culture medium; and

[0327] ii) said second genetically modified microbial cell is capable of producing a second oligosaccharide of at least three monosaccharide units, and wherein said genetically modified cell

[0328] · capable of growing on the second carbon source, while having limited growth or no growth on the first carbon source, and

[0329] · comprising one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase, and

[0330] · comprising a biosynthetic pathway for preparing the activated sugar nucleotide from the second carbon source;

[0331] · preferably capable of exporting the second oligosaccharide into the culture medium,

[0332] wherein the first oligosaccharide or the second oligosaccharide is a fucosylated or sialylated donor oligosaccharide, and the other oligosaccharide or disaccharide is a receptor oligosaccharide or receptor disaccharide; and

[0333] b) making an enzyme having transglycosylase activity available in the culture medium, wherein the enzyme having transglycosylase activity is:

[0334] i) a transfucosidase if the donor oligosaccharide is a fucosylated oligosaccharide; or

[0335] ii) a transsialidase if the donor oligosaccharide is a sialylated oligosaccharide; and

[0336] c) incubating the first oligosaccharide or disaccharide and the second oligosaccharide together with the transglycosylase in the culture medium in which the first oligosaccharide and the second oligosaccharide are produced to form a third sialylated and / or fucosylated oligosaccharide having at least three, for example at least four, monosaccharide units.

[0337] The embodiments described herein relate to a method for producing a neutral core oligosaccharide having at least six monosaccharide units from donor oligosaccharides and receptor oligosaccharides produced from first and second genetically modified cells, the method comprising the following steps:

[0338] a) co-culturing first and second genetically modified cells in a culture medium, wherein

[0339] i) the first genetically modified microbial cell is capable of producing a first oligosaccharide having at least four monosaccharide units, such as LNT or LNnT, and wherein the genetically modified cell

[0340] · is capable of growing on the first carbon source, while having limited growth or no growth on the second carbon source, and

[0341] · comprises one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase; and

[0342] · comprises at least one pathway for producing nucleotide-activated sugar from the first carbon source; and

[0343] ·Preferably capable of exporting the first oligosaccharide into the culture medium; and

[0344] ii) The second genetically modified microbial cell is capable of producing a second oligosaccharide different from the first oligosaccharide having at least four monosaccharide units, such as LNT, LNnT, pLNnH or pLNH2, and wherein the genetically modified cell

[0345] ·Capable of growing on the second carbon source while being growth-limited or not growing on the first carbon source, and

[0346] ·Comprising one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase, and

[0347] ·Comprising a biosynthetic pathway for preparing the activated sugar nucleotide from the second carbon source;

[0348] ·Preferably capable of exporting the second oligosaccharide into the culture medium,

[0349] wherein the first oligosaccharide is a donor oligosaccharide and the second oligosaccharide is a receptor oligosaccharide;

[0350] and

[0351] b) Making an enzyme with transglycosidase activity available in the culture medium, wherein the enzyme with transglycosidase activity is:

[0352] i) If the donor oligosaccharide is a fucosylated oligosaccharide, then it is a transfucosidase; or

[0353] ii) If the donor oligosaccharide is a sialylated oligosaccharide, then it is a transsialidase; and

[0354] c) Incubating the first oligosaccharide or disaccharide and the second oligosaccharide together with the transglycosidase in the culture medium in which the first oligosaccharide and the second oligosaccharide are produced to form a third sialylated and / or fucosylated oligosaccharide having at least three, such as at least four, monosaccharide units.

[0355] In the context of a mixed production method, the term "first oligosaccharide" or "first HMO" or "first disaccharide" or simply "disaccharide" refers to an oligosaccharide or disaccharide produced in situ by a first genetically modified cell, which constitutes the first substrate in the enzymatic (transglycosidase) step of the mixed process. In an embodiment of the "first" disaccharide produced by a first genetically modified strain, it preferably acts as a receptor. Additionally, the "first" disaccharide is not lactose and is preferably lacto-N-biose (LNB) or N-acetyl lactosamine (LacNAc). The term "second oligosaccharide" or "second HMO" refers to an oligosaccharide produced in situ by a second genetically modified cell, which constitutes the second substrate in the enzymatic (transglycosidase) step of the mixed process. The first oligosaccharide and the second oligosaccharide are different and are capable of acting as donor and receptor substrates during the transglycosylation process. In embodiments where the oligosaccharides produced in this process are Lewis A- or Lewis X-based oligosaccharides, or complex fucosylated and / or sialylated oligosaccharides, the donor oligosaccharide contains a fucosyl or sialic acid residue.

[0356] When the first and second oligosaccharides react with a transglycosidase, a third oligosaccharide and a by-product (leaving group) are produced. The leaving group is recycled by at least one genetically modified cell, thereby producing more oligosaccharides produced by said cell. The third oligosaccharide is preferably the complex oligosaccharide required for the process, but it can also be an intermediate oligosaccharide for a second transglycosidase reaction, and the fourth oligosaccharide produced by this reaction is the complex oligosaccharide required for the process. If the mixed process comprises two enzymatic steps, it can be a two-step enzymatic process catalyzed by the same transglycosidase or a two-step enzymatic process catalyzed by transglycosidases with different activities, e.g., one is a fucosyltransferase and the other is a sialyltransferase, depending on their selectivity. However, in the case of using two different transglycosidases, it may be necessary to provide another substrate for the second transglycosidase.

[0357] The enzymatic transglycosidase reaction occurs in the culture medium. Therefore, the first and second oligosaccharides produced by the genetically modified cells need to be present in the culture medium before the reaction occurs. Preferably, the first and second oligosaccharides are exported from their respective cells without affecting cell viability. In an alternative embodiment, the first and / or second oligosaccharides can be released (made available) into the culture medium by natural lysis of some cells during fermentation without stopping the growth of the culture.

[0358] In an embodiment, it is desirable for the first genetically modified cell to export the first oligosaccharide it produces into the culture medium, and for the second genetically modified cell to export the second oligosaccharide it produces into the culture medium so that they can easily undergo a transglycosidase reaction in the culture medium.

[0359] In embodiments, oligosaccharides such as HMOs that can be advantageously produced by one of the genetically modified cells and exported into the culture medium and that can act as donor oligosaccharides in a transglycosylation reaction can be selected from Lewis A, Lewis X, 2’FL, 3FL, DFL, sialyl-LacNAc, sialyl-LNB, FSL, LNT, LNnT, LNFP-I, LST-a, 3’SLAcNAc, 3’SLNB, 3’SL, and 6’SL.

[0360] In embodiments, oligosaccharides (such as HMOs) that can be produced by one of the genetically modified cells and that can act as acceptor oligosaccharides in a transglycosylation reaction can be selected from LNB, LacNac, 2’FL, 3FL, 2’FLacNAc, 2’FLNB, Lewis A, Lewis X, LNT-II, LNT, LNnT, Para-LNnH, LNFP-I, LNFP-II, LNFP-III, LNFP-IV, LNFP-V, LNFP-VI, LST-a, LST-c, and LST-d.

[0361] In embodiments, HMOs that can be advantageously produced by one of the genetically modified cells and exported into the culture medium and that can act as acceptor oligosaccharides in a transglycosylation reaction can be selected from LNB, LacNAc, Lewis A, Lewis X, 2’FL, 3FL, LNT-II, LNT, LNnT, and LNFP-I.

[0362] Genetically modified cells are capable of producing a first oligosaccharide from a substrate preferably added to the culture medium and taken up by the cells as an initial substrate for the production of the first oligosaccharide (e.g., HMO). If the reducing end of the third oligosaccharide is N-acetyl lactosamine (LacNAc) lacto-N-diosaccharide (LNB), the initial substrate can be selected from N-acetyl lactosamine (LacNAc) lacto-N-diosaccharide (LNB), which is then modified in the cell in a manner similar to lactose, resulting in, for example, Lewis A, Lewis X, 3'SLacNAc, 3'SLNB, 2'FLacNAc or 2'FLNB. For HMO, lactose is most commonly used as the initial substrate, but LNT-II may also serve as a substrate for the production of LNT or LNnT, or 2'FL or 3FL can be used as a substrate for the production of DFL. In some embodiments, the substrate for the production of the first HMO can be selected from lactose, 2'FL, 3FL or LNT-II. Preferably, the initial substrate is selected from lactose or 2'FL. In a preferred embodiment, the substrate for the production of the first and second oligosaccharides / HMO is lactose. As an alternative to adding the initial substrate for the production of the first and / or second oligosaccharide to the fermentation medium, the genetically modified cells can be further engineered to produce the initial substrate intracellularly (e.g., see WO2015 / 150328).

[0363] To enable in-situ recycling of by-products, it would be highly advantageous if at least one genetically modified cell is capable of internalizing the initial substrate for the production of the first and / or second oligosaccharide.

[0364] In some embodiments, at least one genetically modified cell is capable of internalizing lactose, 2'FL, 3FL and / or LNT-II, depending on which compound is used as the initial substrate for the production of the first or second HMO. This initial substrate internalized by the cell may correspond to a by-product generated by the transglycosylation reaction and thus be recycled. Preferably, at least one genetically modified cell uses lactose as the initial substrate, and if the by-product generated by the transglycosylation reaction is not lactose, lactose is added to the culture during fermentation. In addition, in embodiments where the fermentation is a co-culture and both cells use lactose as the initial substrate, it is also preferred to feed lactose to the culture during fermentation to avoid depletion of the initial substrate.

[0365] In some embodiments, preferably, one of the genetically modified cells is capable of internalizing lactose or 2'FL added to the culture medium, and then at least one cell uses this lactose or 2'FL to produce the first or second oligosaccharide (e.g., HMO). Some microbial cells have an endogenous lactose uptake system, such as lactose permease, which is also capable of importing 2'FL. If a higher lactose uptake capacity is required, lactose permease can also be introduced into the cell by genetic engineering as an additional recombinant copy of a heterologous protein or a native gene.

[0366] In a preferred embodiment, both the first and second genetically modified cells are capable of internalizing lactose as an initial substrate for the production of the first and second oligosaccharides, respectively.

[0367] In an embodiment, the genetically modified cells are cultured starting in the presence of sufficient substrate for the cells to produce the first and second oligosaccharides. In some embodiments, sufficient substrate is present at the start of the culture to produce the desired amounts of the first and second oligosaccharides such that no additional substrate needs to be added to the culture medium after the start of the culture. In other embodiments, during the fermentation process, at least one substrate (such as lactose) for the production of the first and / or second oligosaccharides is continuously added to the culture to ensure that the cells do not deplete the substrate, which is particularly important when both cells use the same substrate. In an embodiment, the initial substrate for the production of the first and / or second oligosaccharides can be independently selected from lactose, LacNAc, LNB, 2’FL, 3FL, and LNT-II. In a preferred embodiment, the substrates for the production of the first and second oligosaccharides are independently selected from lactose or 2’FL. Preferably, the substrates for the production of the first and second oligosaccharides are the same for both strains, and most preferably, the substrate for the production of the first and second oligosaccharides is lactose.

[0368] In an alternative embodiment, at the time of initial carbon source depletion, the substrate (such as lactose) for the production of the first and second oligosaccharides is added to the culture medium, thereby allowing the first and second genetically modified cells to undergo initial growth before starting to produce the first and second oligosaccharides. The substrate (such as lactose) can be added as a single portion, added separately, or added together with one or two carbon sources.

[0369] During the co-culture process described herein, it may be necessary to control the ratio of the first and second oligosaccharides to balance the molar ratio of the first and second oligosaccharides, thereby ensuring optimal formation of the third oligosaccharide. In one embodiment, the molar ratio of the first and second oligosaccharides is 1:1 (equimolar). In some embodiments, it may be advantageous for the second oligosaccharide to be in excess over the first oligosaccharide, such as a ratio of the second oligosaccharide to the first oligosaccharide of 1.5:1 - 10:1, such as 1.5:1 - 5:1, such as 2:1, 3:1, 4:1, 5:1, 6:1, 7:1, 8:1, 9:1, 10:1. For example, in a dual-strain mixing method for forming complex oligosaccharides, the affinity constants (Km) of the enzyme for the donor and acceptor may be different. To maximize the efficiency of the enzymatic reaction, it may thus be advantageous to provide the donor and acceptor in a ratio that reflects Km. The ratio of the first and second oligosaccharides can be controlled by adjusting the feed rates of the first and second carbon sources. In particular, towards the end of fermentation, it may be necessary to reduce the production of one of the oligosaccharides to ensure a lower content of by-products in the final product of the mixing process. This can be achieved by reducing the carbon source of the strain producing that oligosaccharide, or, if the substrate for producing that oligosaccharide is different from the substrate for producing the other oligosaccharide, by reducing the substrate for producing that oligosaccharide.

[0370] To reduce the level of by-products (such as lactose) generated as leaving groups in the transglycosylation reaction (enzymatic reaction) in the dual-strain mixing process, it is preferably to stop feeding the cells with the substrates (such as lactose) for producing the first and second oligosaccharides in advance at the end of the reaction. When the transglycosidase reaction generates lactose and the third complex oligosaccharide, at least one genetically modified cell is capable of internalizing lactose, thereby removing the lactose generated in the enzymatic reaction and converting it into additional first and / or second oligosaccharides. In addition to removing unwanted lactose from the culture medium, this also promotes the balance towards the formation of additional third oligosaccharide / complex HMO. Therefore, compared with traditional enzymatic processes, the yield of the third oligosaccharide / complex HMO can be higher than that of the first and / or second oligosaccharide or both. If lactose feeding is stopped to achieve complete conversion of the acceptor oligosaccharide / HMO or to achieve a higher molar ratio of the third complex HMO to the acceptor oligosaccharide / HMO, it is preferably to inactivate the transglycosidase while reducing lactose feeding to prevent the accumulation of the leaving group (such as lactose) of the enzymatic step due to the side hydrolysis activity of the enzyme, triggering a reverse reaction and maintaining the desired product composition. This can be achieved by changing the pH value, temperature, or adding a protease.

[0371] In other embodiments, the by-products of the transglycosidase reaction can be oligosaccharides, such as oligosaccharides having three monosaccharide units, such as HMO. In such cases, it is preferred to engineer a genetically modified cell to be able to uptake the by-product oligosaccharide and use it as a substrate to generate the first or second oligosaccharide that enters the transglycosylation process in the culture medium. For example, one of the genetically modified cells produces DFL, which acts as a fucosyl donor in the fucosylation reaction, and thereby generates the by-product 2’FL (leaving group). Subsequently, 2’FL is taken up by the cell that produces DFL, and this cell is able to use 2’FL as a substrate to produce DFL. In such cases, another strain is likely to require a different substrate, such as lactose, to produce another oligosaccharide for the transglycosidase reaction.

[0372] In some embodiments, at the end of the process, the weight % of the third oligosaccharide / complex HMO exceeds the weight % of the donor oligosaccharide. In some embodiments, the ratio of the third oligosaccharide to the first and / or second oligosaccharide is higher than 1.5:1, such as higher than 2:1, such as 5:1. Higher than a certain ratio means that the first number indicated in the ratio can be the indicated number or greater than the indicated number.

[0373] In the dual-strain mixing process described herein, the transglycosidase that mediates the glycosyl transfer of the acceptor oligosaccharide and the donor oligosaccharide can be obtained in the fermentation medium.

[0374] In some embodiments, the transglycosidase is expressed by a recombinant nucleic acid in one of the genetically modified cells that produce the first or second oligosaccharide. Alternatively, it can be expressed in a third strain that grows on a third carbon source or one of the carbon sources of the first or second strain, as it may not require balancing the expression of the enzyme. The enzyme can enter the culture medium through the natural lysis of some cells during the fermentation process without stopping the growth of the culture. It is advantageous to export the enzyme into the culture medium. For example, an appropriate signal peptide can be used to facilitate the export of the transglycosidase.

[0375] For expression in E. coli, the signal peptide can be selected, for example, from one of the following well-known signal peptides.

[0376] Table 34 lists the signal peptides suitable for expressing heterologous transglycosidase in E. coli.

[0377]

[0378]

[0379] In other embodiments, during the cultivation of genetically modified cells, the transglycosidase is added exogenously to the culture medium. When adding the enzyme exogenously, it is preferably aseptically filtered before addition to avoid contamination of the culture. The transglycosidase is added to the mixing process with sufficient activity to mediate the transglycosylation of the acceptor oligosaccharide and the donor oligosaccharide. If the activity of the enzyme decreases during fermentation, it may be advantageous to add the enzyme when the genetically modified cells produce sufficient substrate so that it does not limit the rate of the process. The enzyme can also be added multiple times during the cultivation process.

[0380] In some embodiments, at the time point when the genetically modified cells have converted at least 50% of the initial lactose (such as at least 75% of the initial lactose, at least 85% of the initial lactose, at least 90% of the initial lactose, and 95% to 100% of the initial lactose) into the first and second oligosaccharides, the transglycosidase is added to the culture medium. Before initiating the transglycosylation reaction, it is preferable to first form a sufficient amount of the first and second oligosaccharides to ensure that the substrate of the enzymatic reaction does not become a rate-limiting factor and that the unconverted lactose does not inhibit the reaction.

[0381] At the end of fermentation, it is necessary to inactivate the transglycosidase to avoid the shift of the transglycosidase reaction equilibrium after the formation of the first oligosaccharide stops due to the cessation of the carbon source supply (the carbon source provides energy and carbon for the genetically modified cells to continue cycling lactose). Preferably, inactivation is carried out before or immediately after harvesting the cells. Non-limiting examples of transglycosidase inactivation can be selected from: i) heating the fermentation broth to a temperature that denatures the enzyme; ii) adding a protease to the culture broth at the end of fermentation to hydrolyze the enzyme; or iii) changing the pH value of the culture broth so that it is outside the activity range of the enzyme or denatures the enzyme. If heat inactivation is used, it is preferably to heat the culture broth to at least 60°C, such as at least 70°C, such as at least 80°C, such as at least 90°C, such as at least 95°C, and for at least 5 minutes, such as at least 10 minutes, such as at least 15 minutes. If protease is used to inactivate the enzyme, it is preferably to add the protease with sufficient activity at least 10 minutes before cell harvesting, such as at least 20 minutes, such as at least 30 minutes, to hydrolyze all the enzymes. If pH value change is used to inactivate the enzyme, it is preferably to lower the pH value below 5, preferably between 3 and 5, such as between 3.5 and 4.5.

[0382] Transglycosidase

[0383] Glycoside hydrolases are essentially carbohydrate-processing enzymes. In addition to hydrolytic activity, some of them also exhibit high transglycosylation activity, also known as transglycosidase, which is capable of catalyzing the transfer of glycosyl groups between different glycosides and / or oligosaccharides.

[0384] In the dual-strain mixing method described herein, it is more advantageous if the hydrolytic activity of the transglycosidase is as low as possible. For example, the hydrolytic activity of sialidase results in the hydrolysis of donor oligosaccharides. For example, in the case of 3'SL, the hydrolysis reaction produces lactose and sialic acid, as well as the hydrolysis of a third oligosaccharide to form a receptor and sialic acid, for example, in the case of FSL, to form 3FL and sialic acid. For example, the hydrolytic activity of fucosidase results in the hydrolysis of donor oligosaccharides. In the case of 2'FL or 3FL, the hydrolysis reaction produces lactose and fucose, and the hydrolysis of a third oligosaccharide forms a receptor and fucose, for example, in the case of LNFP-III, to produce LNnT and fucose. Generally, when the receptor substrate is in sufficient supply relative to the donor substrate, the hydrolytic activity of the enzyme is inhibited. However, since in the dual-strain mixing process, by-products of hydrolysis (such as lactose) are recycled to generate the first and / or second oligosaccharides in genetically modified cells, the impact on the formation of the third HMO product is very low because the transglycosidase activity subsequently regenerates the third oligosaccharide. The hydrolytic and fucosyltransferase activities of fucosidase can be measured, for example, according to the method of Zeuner et al. (2018, Enzyme and Microbial Technology 115:37–44). A similar assay method is also applicable to sialidase, simply by substituting 3' SL or 6' SL for 3FL. However, in terms of the functionality of the dual-strain mixing process described herein, it is preferable to compare the hydrolytic activity of potential transglycosidases in the actual process and then evaluate the amount of fucose or sialic acid produced by each enzyme. In the mixing method described herein, a fucosidase that produces as little fucose as possible is required. Similarly, in the mixing method described herein, a sialidase that produces as little sialic acid as possible is required. A trans-lacto-N-biosidase (LnbX) that produces as little lacto-N-biose as possible is required. For example, the levels of fucose, sialic acid, and lacto-N-biose can be measured by HPLC or other methods known to those skilled in the art.

[0385] In the dual-strain mixing method described herein, the amount of transglycosidase added to the dual-strain culture is sufficient to mediate the transglycosylation of the glycosyl group of the receptor oligosaccharide with the donor oligosaccharide. In the dual-strain mixing method, if the transferred glycosyl moiety is a fucosyl or sialyl group, the enzymes with transglycosidase activity are fucosidase and sialidase, respectively.

[0386] In the dual-strain mixing method, if the transferred glycosyl moiety is galactose or N-acetylglucosamine (GlcNAc), the enzymes with transglycosidase activity are trans-β-galactosidase, trans-lacto-N-biosidase, or β-N-acetylglucosaminidase, respectively.

[0387] In an embodiment, the transglycosidase is selected from α-1,2-fucosyltransferase, α-1,3-fucosyltransferase, α-1,3 / 4-fucosyltransferase, α-2,3-sialyltransferase, α-2,6-sialyltransferase, trans-lacto-N-biosidase, β-N-acetylglucosaminidase, and trans-β-galactosidase. It is advantageous if the fucosyltransferase can use fucosyllactose (e.g., 2’FL, 3FL, or DFL) as a fucosyl donor and a second oligosaccharide as an acceptor. Similarly, it is advantageous if the sialyltransferase can use sialyllactose (3’SL or 6’SL) as a sialic acid donor and a second oligosaccharide as an acceptor. It is advantageous if β-1,3-N-acetylglucosaminidase can use LNT-II as a GlcNAc donor and a second oligosaccharide as an acceptor. It is advantageous if β-1,3-galactosidase can use LNT as a galactose donor and a second oligosaccharide as an acceptor. It is advantageous if trans-lacto-N-biosidase can use LNT, LNFP-I, and LST-a as donors and a second oligosaccharide as an acceptor.

[0388] In an embodiment where the complex oligosaccharide having at least four or five monosaccharide units produced by the dual-strain mixing method described herein is an HMO, the transglycosidase has substrate specificity for an oligosaccharide receptor, which is preferably an HMO containing at least three monosaccharide units, such as four, five, six, or seven monosaccharide units. In some embodiments, the complex oligosaccharide has at least five monosaccharide units and is a neutral non-fucosylated complex oligosaccharide, such as a neutral core HMO.

[0389] In an embodiment where the sialylated and / or fucosylated oligosaccharide having at least four monosaccharide units produced by the dual-strain mixing method described herein is an HMO, the fucosyltransferase or sialyltransferase has substrate specificity for an oligosaccharide receptor, which is preferably an HMO containing at least three monosaccharide units, such as four, five, six, or seven monosaccharide units.

[0390] In an embodiment, the fucosyltransferase and / or sialyltransferase has substrate specificity for at least one receptor disaccharide or oligosaccharide selected from LNB, LAcNAc, 2’FL, 3FL, Lewix A, Lewis X, 2’FLacNAc, 2’FLNB, LNT, LNnT, LNH, LNnH, para-LNH, para-LNnH, LNFP-I, LNFP-II, LNFP-III, LNFP-V, LNFP-VI, LSTa, and LSTc.

[0391] In a further embodiment, the fucosyltransferase and / or sialyltransferase has substrate specificity for at least one HMO receptor oligosaccharide selected from 2’FL, 3FL, LNT, LNnT, LNH, LNnH, para-LNH, para-LNnH, LNFP-I, LNFP-II, LNFP-III, LNFP-V, LNFP-VI, LSTa and LSTc.

[0392] One embodiment of the present application is a method for producing FSL, comprising the steps of:

[0393] a) co-culturing a first genetically modified cell that grows on the first carbon source and produces 3FL and a second genetically modified cell that grows on the second carbon source and produces 3’SL in a culture medium provided with the first and second carbon sources; and

[0394] b) making an enzyme having sialyltransferase activity available in the culture medium; and

[0395] c) incubating 3FL, 3’SL and sialyltransferase in the culture medium to form FSL and lactose; and

[0396] wherein lactose is recycled by the first and second genetically modified cells to produce more 3FL and 3’SL.

[0397] Another embodiment of the present application is a method for producing LNDFH-I, comprising the steps of:

[0398] a) co-culturing a first genetically modified cell that grows on the first carbon source and produces 3FL and a second genetically modified cell that grows on the second carbon source and produces LNFP-I in a culture medium provided with the first and second carbon sources; and

[0399] b) making an enzyme having fucosyltransferase activity available in the culture medium; and

[0400] c) incubating 3FL, LNFP-I and fucosyltransferase in the culture medium to form LNDFH-I and lactose; and

[0401] wherein lactose is recycled by the first and second genetically modified cells to produce more 3FL and LNFP-I.

[0402] Another embodiment of the present application is a method for producing LST-c, comprising the steps of:

[0403] a) Co-culturing a first genetically modified cell that grows on the first carbon source and produces 6’SL and a second genetically modified cell that grows on the second carbon source and produces LNnT in a culture medium provided with the first and second carbon sources;

[0404] b) Making an enzyme with sialidase activity available in the culture medium;

[0405] c) Incubating 6’SL, LNnT, and sialidase in the culture medium to form LST-c and lactose, and

[0406] wherein the lactose is recycled by the first and second genetically modified cells to produce more 6’SL and LNnT.

[0407] Another embodiment of the present application is a method for producing LNFP-III, comprising the following steps:

[0408] a) Co-culturing a first genetically modified cell that grows on the first carbon source and produces 3FL and a second genetically modified cell that grows on the second carbon source and produces LNnT in a culture medium provided with the first and second carbon sources; and

[0409] b) Making an enzyme with fucosidase activity available in the culture medium; and

[0410] c) Incubating 3FL, LNnT, and fucosidase in the culture medium to form LNFP-III and lactose, and

[0411] wherein the lactose is recycled by the first and second genetically modified cells to produce more 3FL and LNnT.

[0412] Another embodiment of the present application is a method for producing para-LNH, comprising the following steps:

[0413] a) Co-culturing a first genetically modified cell that grows on the first carbon source and produces LNT and a second genetically modified cell that grows on the second carbon source and produces LNnT in a culture medium provided with the first and second carbon sources; and

[0414] b) Making an enzyme with β-galactosyl-N-diasidase activity available in the culture medium; and

[0415] c) Incubating LNT, LNnT, and β-galactosyl-N-diasidase in the culture medium to form para-LNH and lactose, and

[0416] wherein the lactose is recycled by the first and second genetically modified cells to produce more LNT and LNnT.

[0417] To expand on the above specific embodiments, a method for producing specific complex oligosaccharides is described in detail. Table 3 below lists a non-limiting list of sialylated and / or fucosylated oligosaccharides with at least four monosaccharide units, which may be obtained by using different transglycosidase activities, where fucosyllactose or sialyllactose is used as the donor oligosaccharide and a second oligosaccharide is used as the acceptor oligosaccharide.

[0418] Table 3: Non-limiting examples of complex oligosaccharides obtainable using a dual-strain mixed process

[0419]

[0420]

[0421] Taking the trans-lacto-N-diosidase from Bifidobacterium longum subsp. longum JCM1217 (LnbX, Sakamura et al. J. Biol. Chem. 288, 25194 (2013), GenBank nr. DAA64542) and its truncated functional analogs as examples, they can be used to prepare oligosaccharides containing linear lacto-N-biose. In one embodiment, a variant having 70% identity with the amino acid sequence at positions 45 to 625 in GenBank accession number DAA64542 is used, and mutations occur at at least one or more of the amino acid positions 410, 416, 439, and 442, with the amino acid numbering based on GenBank accession number DAA64542, for transferring the lacto-N-biose moiety from the donor oligosaccharide to the acceptor oligosaccharide. Some advantageous variants are described in PA202201151, where it is shown that they play a role in the in vitro generation of pLNH from LNT and LNnT.

[0422] In one embodiment, the trans-lacto-N-diosidase comprises or consists of the amino acid sequence of SEQ ID NO: 122 or a functional variant thereof, particularly where Gly (G) at position 410 is replaced by Trp, Tyr, Phe, or His, preferably Trp; and / or Asp (D) at position 416 is replaced by Asn or Gln, preferably Asn; and / or Met (M) at position 439 is replaced by Leu, Val, or Ile, preferably Leu; and / or Asn (N) at position 442 is replaced by Trp, Tyr, Phe, or His, preferably Trp.

[0423] Non-limiting examples of relevant trans-sialidases and trans-fucosidases are listed in Tables 4 and 5 below.

[0424] Trans-sialidase

[0425] Enzymes having trans-sialidase activity and suitable for preparing sialylated oligosaccharides by the dual-strain mixed process described herein may be selected from sialidases and trans-sialidases.

[0426] Both sialidase or neuraminidase (EC 3.2.1.18) and trans-sialidase (EC 2.4.1.-) belong to the GH33 family defined by the CAZY nomenclature (http: / / www.cazy.org), and refer to enzymes that can hydrolyze the α-linkage of terminal sialic acid (exo-α-sialidase) in various sialic acid glycoconjugates. These sialic acids are bound to galactose or glucose via α-2,3 or α-2,6 linkages. These enzymes are particularly present in various viral families and bacteria, as well as in protozoa, some invertebrates, and mammals. Although sialidase has hydrolytic activity, due to its α-2,3 and / or α-2,6 selective trans-sialidase activity, it can act as a catalyst for the trans-sialylation reaction.

[0427] To improve the trans-sialidase activity of sialidase, it can be modified by various engineering techniques. Preferably, under the conditions of the mixed dual-strain method described herein, the formation of sialic acid is low. Preferably, the amount of sialic acid is less than 5% of the total molar percentage of the donor oligosaccharide and the third oligosaccharide, more preferably less than 3% of the total molar percentage of the donor oligosaccharide and the third oligosaccharide. WO2012 / 007588 describes a series of suitable trans-sialidases.

[0428] Table 4: Suitable trans-sialidases

[0429]

[0430] In an embodiment, the trans-sialidase is selected from the suitable trans-sialidases in Table 4 or a functional homolog thereof, and the amino acid sequence of the functional homolog has at least 70% identity with the sequence of a single trans-sialidase in Table 4, such as at least 80%, at least 85%, at least 90%, at least 95%, or even 97%, 98%, or 99% identity.

[0431] In one embodiment, the fucosidase comprises or consists of the amino acid sequence of SEQ ID NO: 60, 91, 87, 88, or 113.

[0432] In the embodiment of adding the trans-sialidase to the fermentation broth, it is aseptically filtered before being introduced into the dual-strain mixing process. A trans-sialidase with an activity sufficient to mediate the trans-sialylation of the acceptor oligosaccharide and the donor oligosaccharide is added to the dual-strain mixing process.

[0433] In alternative embodiments, at least one genetically modified cell is further modified by introducing a heterologous nucleic acid encoding a trans-sialidase. Preferably, the trans-sialidase is secreted / output by the further genetically modified cell into the culture medium. The heterologous nucleic acid encoding the trans-sialidase can be expressed from an inducible promoter such that the expression of the trans-sialidase is delayed relative to the formation of the oligosaccharides produced by the same cell. The advantage of delaying the expression of the trans-sialidase is that the first and / or second oligosaccharides do not become rate-limiting factors in the enzymatic step in the dual-strain mixing process.

[0434] Fucosyltransferase

[0435] Enzymes having fucosyltransferase activity and suitable for the preparation of fucosylated oligosaccharides by the dual-strain mixing process described herein may be selected from fucosidases and fucosyltransferases.

[0436] α-L-fucosidases are classified according to EC 3.2.1.38 and EC 3.2.1.51 and belong to glycoside hydrolase families 29 and 95 (GH29 and GH95) as defined by the CAZY nomenclature (http: / / www.cazy.org). The GH29 family has a broad substrate specificity, while the GH95 family has a strict substrate specificity for α-1,2-linked fucosyl residues. The GH29 family appears to be divided into two subfamilies. One subfamily generally has a strict specificity for α-1,3- and α-1,4-fucosidic linkages. Members of the other subfamily have a broader specificity covering two or three α-fucosyl linkages. α-L-fucosidases generally hydrolyze the terminal fucosyl residues of glycans. Due to their fucosyltransferase activity, these enzymes can also act as catalysts for fucosylation reactions and can thus be used in the mixing methods described herein.

[0437] To improve the activity of fucosidase, fucosidase can be modified by various engineering techniques. WO2016 / 063261 and Zeuner et al. (2018 Enzyme and Microbial Technology 115:37–44) describe mutants of α-1-3 / 4 fucosidase from Bifidobacterium longum subsp. infantis ATCC 15697 (NCBI accession number WP_012578573) or mutants of fucosidase from Bifidobacterium bifidum JCM1254 (GenBank BAH80310.1), which have increased fucosidase activity and decreased hydrolase activity. Preferably, under the conditions of the mixed dual-strain method described herein, the formation of fucose is low. Preferably, the amount of fucose is less than 5% of the total molar percentage of the donor oligosaccharide and the third oligosaccharide, more preferably less than 3% of the total molar percentage of the donor oligosaccharide and the third oligosaccharide.

[0438] Table 5: Suitable fucosidases

[0439]

[0440]

[0441] In an embodiment, the fucosidase is selected from the suitable fucosidases in Table 5 or a functional homolog thereof, and the amino acid sequence of the functional homolog has at least 70% identity with the sequence of a single fucosidase in Table 5, such as at least 80%, at least 85%, at least 90%, at least 95% or even 97%, 98% or 99% identity.

[0442] In an embodiment, the fucosidase is derived from Bifidobacterium bifidum or Bifidobacterium longum.

[0443] In one embodiment, the fucosidase comprises or consists of the amino acid sequence of SEQ ID NO: 66, 77, 86 or 123.

[0444] In an embodiment, the fucosidase is added to the mixed dual-strain method and aseptically filtered before being introduced into the dual-strain mixing process. The fucosidase is added to the dual-strain mixing process, and its activity is sufficient to mediate the fucosylation of the receptor oligosaccharide with the donor oligosaccharide.

[0445] In alternative embodiments, at least one genetically modified cell is further modified by introducing a heterologous nucleic acid encoding a fucosidase. Preferably, the fucosidase is secreted / output by the further genetically modified cell into the culture medium. The heterologous nucleic acid encoding the fucosidase can be expressed from an inducible promoter such that the expression of the fucosidase is delayed compared to the formation of the oligosaccharides produced by the same cell. The advantage of delaying the expression of the fucosidase is that the first and second oligosaccharides do not become rate-limiting factors in the enzymatic step in the dual-strain mixing process.

[0446] Genetically modified cell

[0447] As used herein, the terms "genetically modified cell" and "genetically engineered cell" are used interchangeably. The "genetically modified cell" as used herein refers to a host cell whose genetic material has been altered by human intervention using genetic engineering techniques, such techniques including but not limited to transformation or transfection, such as using heterologous polynucleotide sequences, Crisper / Cas editing, and / or random mutagenesis. In one embodiment, the genetically engineered cell has been transformed or transfected with a recombinant nucleic acid sequence.

[0448] The genetic modification can for example be selected from the introduction of glycosyltransferases, transglycosidases, and / or metabolic pathway engineering, as well as the introduction of transporters (including the importers and exporters described in the present application), and those skilled in the art will know how to combine all these techniques into a genetically modified cell capable of producing the desired oligosaccharides / HMOs.

[0449] The genetically engineered cell is preferably a microbial cell, such as a prokaryotic cell or a eukaryotic cell. Suitable microbial cells that can serve as host cells include bacterial cells, archaeal cells, algal cells, and fungal cells.

[0450] The genetically engineered cell can be for example a bacterial cell or a yeast cell. In a preferred embodiment, the genetically engineered cell is a bacterial cell.

[0451] Host cell

[0452] In principle, there is no restriction on the bacterial host cell; it can be a eubacterium (Gram-positive or Gram-negative) or an archaeon, as long as it allows the insertion of the gene of interest by genetic manipulation and is capable of large-scale cultivation. Preferably, the host cell has characteristics that allow high cell density cultivation. Non-limiting examples of bacterial host cells suitable for the recombinant industrial production of the HMOs of the present invention can be Erwinia herbicola (Pantoea agglomerans), Citrobacter freundii, Campylobacter sp, Corynebacterium sp., Pantoea citrea, Pectobacterium carotovorum, or Xanthomonas campestris. Bacteria of the genus Bacillus can also be used, including Bacillus subtilis, Bacillus licheniformis, Bacillus coagulans, Bacillus thermophilus, Bacillus laterosporus, Bacillus megaterium, Bacillus mycoides, Bacillus pumilus, Bacillus lentus, Bacillus cereus, and Bacillus circulans.Similarly, the methods of the present invention can be used to engineer bacteria of the genera Lactobacillus and Lactococcus, including but not limited to Lactobacillus acidophilus, Lactobacillus salivarius, Lactobacillus plantarum, Lactobacillus helveticus, Lactobacillus delbrueckii, Lactobacillus rhamnosus, Lactobacillus bulgaricus, Lactobacillus crispatus, Lactobacillus gasseri, Lactobacillus casei, Lactobacillus reuteri, Lactobacillus jensenii, and Lactococcus lactis. Corynebacterium glutamicum, Gluconobacter oxydans, Streptococcus thermophiles, and Proprionibacterium freudenreichii are also bacterial species suitable for the invention described herein.The present invention also includes engineered strains according to what is described herein, which strains are from the genus Enterococcus (e.g., Enterococcus faecium and Enterococcus thermophiles), the genus Bifidobacterium (e.g., Bifidobacterium longum, Bifidobacterium infantis, and Bifidobacterium bifidum), the genus Sporolactobacillus spp., the genus Micromomospora spp., the genus Micrococcus spp., the genus Rhodococcus spp., and the genus Pseudomonas (e.g., Pseudomonas fluorescens and Pseudomonas aeruginosa).

[0453] Non-limiting examples of fungal host cells suitable for the recombinant industrial production of heterologous products include: for example, yeast cells of the genus Komagataella, Kluyveromyces, Yarrowia, Pichia, Saccaromyces, Schizosaccharomyces, or Hansenula, or filamentous fungi from the genus Aspargillus, Fusarium, or Thricoderma. More specifically, yeast cell species include Komagataella phaffii, Kluyveromyces lactis, Yarrowia lipolytica, Pichia pastoris, and Saccharomyces cerevisiae, or filamentous fungi such as Aspergillus niger, Aspergillus nidulans, Aspergillus oryzae, Fusarium solani, Fusarium graminearum, and Trichoderma reesei.

[0454] In one or more exemplary embodiments, the genetically engineered cells are selected from Escherichia coli, Corynebacterium glutamicum, Gluconobacter oxydans, Lactococcus lactis, Bacillus subtilis, Streptomyces lividans, Pichia pastoris, and Saccharomyces cerevisiae.

[0455] In one or more exemplary embodiments, the genetically engineered cells are Saccharomyces cerevisiae or Pichia pastoris.

[0456] In one or more exemplary embodiments, the genetically engineered cells are selected from Escherichia coli, Corynebacterium glutamicum, Gluconobacter oxydans, Lactococcus lactis, Bacillus subtilis, and Streptomyces lividans.

[0457] In one or more exemplary embodiments, the genetically engineered cells are Gram-positive bacteria. In a further embodiment, the Gram-positive bacteria are selected from Bacillus subtilis, Corynebacterium glutamicum, Lactococcus lactis, and Streptomyces lividans.

[0458] In one or more exemplary embodiments, the genetically engineered cells are Bacillus subtilis.

[0459] In one or more exemplary embodiments, the genetically engineered cells are Corynebacterium glutamicum.

[0460] In one or more exemplary embodiments, the genetically engineered cells are Gram-negative bacteria. In a further embodiment, the Gram-negative bacteria are selected from Escherichia coli and Gluconobacter oxydans.

[0461] In an embodiment, the first and second genetically modified microbial cells are both selected from yeast. In one embodiment, one strain is selected from Yarrowia lipolytica, Pichia pastoris, and Saccharomyces cerevisiae, and the other strain is selected from Yarrowia lipolytica, Pichia pastoris, and Saccharomyces cerevisiae. The yeast strains are preferably of the same species.

[0462] In an embodiment, the first genetically modified microbial cell is selected from bacteria, and the second genetically modified microbial cell is selected from yeast. In one embodiment, one strain is selected from Bacillus subtilis, Corynebacterium glutamicum, Lactococcus lactis, and Streptomyces lividans, and the other strain is selected from Yarrowia lipolytica, Pichia pastoris, and Saccharomyces cerevisiae.

[0463] In some embodiments, the first and second genetically modified microbial cells are selected from bacterial species. In one embodiment, one strain is selected from Bacillus subtilis, Corynebacterium glutamicum, Lactococcus lactis, and Streptomyces lividans, and the other strain is selected from Bacillus subtilis, Corynebacterium glutamicum, Lactococcus lactis, and Streptomyces lividans. In one embodiment, one strain is an Escherichia coli strain and the other strain is a Bacillus subtilis or Corynebacterium glutamicum strain. The bacterial strains are preferably of the same species.

[0464] The first and second genetically modified microbial cells are preferably selected from the same species.

[0465] In one or more exemplary embodiments, the genetically modified cell is Escherichia coli. In one embodiment, both the first and second genetically modified microbial cells are Escherichia coli.

[0466] In one or more exemplary embodiments, the invention relates to a genetically modified cell, wherein the cell is derived from Escherichia coli strain K-12 or strain DE3.

[0467] Glycosyltransferase

[0468] The genetically modified cell according to the invention comprises at least one recombinant nucleic acid sequence encoding at least one glycosyltransferase which is capable of transferring a sugar residue from a sugar donor to a receptor oligosaccharide (substrate) to synthesize an oligosaccharide product, such as a human milk oligosaccharide product. The nucleic acid sequence encoding one or more expressed glycosyltransferases may be integrated into the genome of the genetically engineered cell (by chromosomal integration), or it may be comprised in a plasmid and expressed in plasmid form, as described in the present disclosure.

[0469] The genetically modified cell according to the invention may comprise at least two recombinant nucleic acid sequences encoding two different glycosyltransferases which are capable of transferring a sugar residue from a sugar donor to a receptor oligosaccharide.

[0470] The one or more glycosyltransferases are preferably selected from enzymes having α-1,2-fucosyltransferase, α-1,3-fucosyltransferase, α-1,3 / 4-fucosyltransferase, α-1,4-fucosyltransferase, α-2,3-sialyltransferase, α-2,6-sialyltransferase, β-1,3-N-acetylglucosaminyltransferase, β-1,6-N-acetylglucosaminyltransferase, β-1,3-galactosyltransferase and β-1,4-galactosyltransferase activity, as described in more detail below.

[0471] β-1,3-N-acetylglucosaminyltransferase

[0472] β-1,3-N-acetylglucosaminyltransferase refers to any protein that can transfer N-acetylglucosamine from UDP-N-acetylglucosamine to lactose or other acceptor molecules via a β-1,3-linkage. Preferably, the β-1,3-N-acetylglucosaminyltransferase used herein is not from a species of genetically engineered cells, i.e., the gene encoding β-1,3-galactosyltransferase is of heterologous origin. Table 6 lists non-limiting examples of β-1,3-N-acetylglucosaminyltransferase. β-1,3-N-acetylglucosaminyltransferase variants may also be useful, preferably such variants have at least 80%, such as at least 85%, such as at least 90%, such as at least 95% identity with one of the β-1,3-N-acetylglucosaminyltransferases in Table 6.

[0473] Table 6. List of β-1,3-N-acetylglucosaminyltransferases

[0474]

[0475] In an embodiment, the glycosyltransferase encoded by the genetically engineered cell is the β-1,3-N-acetylglucosaminyltransferase in Table 6. Preferably, the glycosyltransferase in the genetically engineered cell is the β-1,3-N-acetylglucosaminyltransferase from Neisseria meningitidis, such as the β-1,3-N-acetylglucosaminyltransferase of SEQ ID NO:95 or its functional variant.

[0476] β-1,3-galactosyltransferase

[0477] β-1,3-galactosyltransferase refers to any protein that can transfer galactose from UDP-galactose to the N-acetylglucosaminyl moiety of an acceptor molecule via a β-1,3-linkage. Preferably, the β-1,3-galactosyltransferase used herein is not from a species of genetically engineered cells, i.e., the gene encoding the β-1,3-galactosyltransferase is of heterologous origin. Table 7 lists non-limiting examples of β-1,3-galactosyltransferase. β-1,3-galactosyltransferase variants may also be useful, preferably such variants have at least 80%, such as at least 85%, such as at least 90%, such as at least 95% identity with one of the β-1,3-galactosyltransferases in Table 7.

[0478] Table 7. List of β-1,3-glycosyltransferases

[0479]

[0480] In an embodiment, at least one glycosyltransferase encoded by the genetically engineered cell is a β-1,3-N-acetylglucosaminyltransferase and a β-1,3-galactosyltransferase. Preferably, the glycosyltransferases in the genetically engineered cell are selected from the β-1,3-N-acetylglucosaminyltransferases in Table 4 and the β-1,3-galactosyltransferases selected from Table 5. More preferably, the 1,3-N-acetylglucosaminyltransferase is from the genus Neisseria, and the β-1,3-galactosyltransferase is from Helicobacter pylori, such as the β-1,3-N-acetylglucosaminyltransferase with GenBank reference number WP_002248149.1 and the β-1,3-galactosyltransferase with GenBank reference number WP_111735921.1 or SEQ ID NO:96, or a functional variant thereof.

[0481] β-1,4-galactosyltransferase

[0482] β-1,4-galactosyltransferase is any protein having the ability to transfer the galactose of UDP-galactose to the N-acetylglucosamine moiety. Preferably, the β-1,4-galactosyltransferase used herein does not originate from the species of the genetically engineered cell, i.e., the gene encoding the β-1,4-galactosyltransferase is of heterologous origin. Table 8 gives non-limiting examples of β-1,4-galactosyltransferases. β-1,4-galactosyltransferase variants may also be useful, and preferably such variants have at least 80%, such as at least 85%, such as at least 90%, such as at least 95% identity with one of the β-1,4-galactosyltransferases in Table 8.

[0483] Table 8. List of β-1,4-glycosyltransferases

[0484] Protein Name GenBank ID Source GalT WP_001262061.1 Helicobacter pylori LgtB AAF42257.1 Neisseria meningitidis MC58

[0485] In an embodiment, at least one glycosyltransferase encoded by the genetically engineered cell is a β-1,3-N-acetylglucosaminyltransferase and a β-1,4-galactosyltransferase. Preferably, the glycosyltransferases in the genetically engineered cell are selected from the β-1,3-N-acetylglucosaminyltransferases in Table 4 and the β-1,4-galactosyltransferases selected from Table 6. More preferably, the 1,3-N-acetylglucosaminyltransferase is from the genus Neisseria, and the β-1,4-galactosyltransferase is from Helicobacter pylori, such as the β-1,3-N-acetylglucosaminyltransferase with GenBank reference number WP_002248149.1 and the β-1,4-galactosyltransferase with GenBank reference number WP_001262061.1 or SEQ ID NO:105, or a functional variant thereof.

[0486] α-1,2-fucosyltransferase

[0487] An α-1,2-fucosyltransferase is a protein having the ability to catalyze the transfer of fucose from a donor substrate (such as GDP-fucose) to a receptor molecule via an α-1,2-linkage bond. Preferably, the α-1,2-fucosyltransferase used herein does not originate from the species of genetically engineered cells, that is, the gene encoding the α-1,2-fucosyltransferase is of heterologous origin. Table 9 gives non-limiting examples of α-1,2-fucosyltransferases. α-1,2-fucosyltransferase variants may also be useful. Preferably, such variants have at least 80%, such as at least 85%, such as at least 90%, such as at least 95% identity with one of the α-1,2-fucosyltransferases in Table 9.

[0488] Table 9. List of α-1,2-fucosyltransferases

[0489]

[0490] In an embodiment, the glycosyltransferase encoded by the genetically engineered cell is the α-1,2-fucosyltransferase in Table 7. Preferably, the glycosyltransferase in the α-1,2-fucosyltransferase is from Helicobacter pylori, such as the α-1,2-fucosyltransferase with GenBank accession number WP_080473865.1 or SEQ ID NO: 108, or a functional variant thereof.

[0491] α-1,3-fucosyltransferase

[0492] An α-1,3-fucosyltransferase refers to a glycosyltransferase that catalyzes the transfer of fucose from a donor substrate (such as GDP-fucose) to a receptor molecule via an α-1,3-linkage bond. Preferably, the α-1,3-fucosyltransferase used herein does not originate from the species of genetically engineered cells, that is, the gene encoding the α-1,3-fucosyltransferase is of heterologous origin. Table 10 gives non-limiting examples of α-1,3-fucosyltransferases. α-1,3-fucosyltransferase variants may also be useful. Preferably, such variants have at least 80%, such as at least 85%, such as at least 90%, such as at least 95% identity with one of the α-1,3-fucosyltransferases in Table 10.

[0493] Table 10. List of α-1,3-fucosyltransferases

[0494]

[0495] In an embodiment, the glycosyltransferase encoded by the genetically engineered cell is the α-1,3-fucosyltransferase in Table 8. Preferably, the glycosyltransferase in the genetically engineered cell is the α-1,3-fucosyltransferase FutA from Helicobacter pylori, such as the α-1,3-fucosyltransferase of SEQ ID NO:89 or a functional variant thereof.

[0496] α-1,3 / 4-fucosyltransferase

[0497] An α-1,3 / 4-fucosyltransferase refers to a glycosyltransferase that catalyzes the transfer of fucose from a donor substrate (such as GDP-fucose) to a receptor molecule via an α-1,3- or α-1,4-linkage. Preferably, the α-1,3 / 4-fucosyltransferase used herein is not from the species of the genetically engineered cell, i.e., the gene encoding the α-1,3 / 4-fucosyltransferase is of heterologous origin. Table 11 lists non-limiting examples of α-1,3 / 4-fucosyltransferases. Variants of the α-1,3 / 4-fucosyltransferase may also be useful. Preferably, such variants have at least 80%, such as at least 85%, such as at least 90%, such as at least 95% identity with one of the α-1,3 / 4-fucosyltransferases in Table 11.

[0498] Table 11. List of α-1,3 / 4-fucosyltransferases

[0499] Protein Name GenBankID Source FucTIII AY450598.1 Helicobacter pylori strain DSM 6709 FutA WP_000487428.1 Helicobacter pylori ATCC 26695

[0500] In an embodiment, the glycosyltransferase encoded by the genetically engineered cell is the α-1,3 / 4-fucosyltransferase in Table 11. Preferably, the glycosyltransferase in the genetically engineered cell is the α-1,3 / 4-fucosyltransferase FutA from Helicobacter pylori, such as the α-1,3 / 4-fucosyltransferase of SEQ ID NO:89, or a functional variant thereof.

[0501] α-2,3-sialyltransferase

[0502] An α-2,3-sialyltransferase refers to a glycosyltransferase that catalyzes the transfer of sialic acid from a donor substrate (such as CMP-N-acetylneuraminic acid) to a receptor molecule via an α-2,3-linkage. Preferably, the α-2,3-sialyltransferase used herein is not from the species of the genetically engineered cell, i.e., the gene encoding the 2,3-sialyltransferase is of heterologous origin. Table 12 gives non-limiting examples of α-2,3-sialyltransferases. Variants of the α-2,3-sialyltransferase may also be useful. Preferably, such variants have at least 80%, such as at least 85%, such as at least 90%, such as at least 95% identity with one of the α-2,3-sialyltransferases in Table 12.

[0503] Table 12. List of α-2,3-sialyltransferases

[0504]

[0505] In an embodiment, the glycosyltransferase encoded by the genetically engineered cell is an α-2,3-sialyltransferase from Table 10. Preferably, the glycosyltransferase in the genetically engineered cell is an α-2,3-sialyltransferase from Campylobacter lari, Neisseria meningitidis, or Pasteurella oralis, such as an α-2,3-sialyltransferase having GenBank accession number EGK8106227.1, AAC44541.1, WP_101774487.1, or SEQ ID NO:91, or a functional variant thereof.

[0506] α-2,6-sialyltransferase

[0507] An α-2,6-sialyltransferase refers to a glycosyltransferase that catalyzes the transfer of sialic acid from a donor substrate (such as CMP-N-acetylneuraminic acid) to a receptor molecule via an α-2,6-linkage. Preferably, the α-2,6-sialyltransferase used herein is not derived from the species of the genetically engineered cell, i.e., the gene encoding the 2,6-sialyltransferase is of heterologous origin. Table 13 gives non-limiting examples of α-2,6-sialyltransferases. α-2,6-sialyltransferase variants may also be useful, preferably such variants are at least 80%, such as at least 85%, such as at least 90%, such as at least 95% identical to one of the α-2,6-sialyltransferases in Table 13.

[0508] Table 13. List of α-2,6-sialyltransferases

[0509]

[0510] In an embodiment, the glycosyltransferase encoded by the genetically engineered cell is an α-2,6-sialyltransferase from Table 11. The glycosyltransferase in the genetically engineered cell is an α-2,6-sialyltransferase from the genus Photobacterium, such as an α-2,6-sialyltransferase having GenBank accession number AB500947.1 or BAF92026.1.

[0511] Nucleotide-activated sugar pathway

[0512] In the genetically engineered cells used in the methods described herein, glycosyltransferase-mediated glycosylation reactions occur intracellularly, where activated sugar nucleotides serve as glycosyl donors. Activated sugar nucleotides typically have a phosphorylated sugar residue linked to a nucleoside. Specific glycosyltransferases accept only specific sugar nucleotides. Thus, the following activated sugar nucleotides are preferably involved in glycosyltransfer: glucose-UDP-GlcNAc, UDP-galactose, UDP-glucose, UDP-N-acetylglucosamine, UDP-N-acetylgalactosamine (GlcNAc), and CMP-N-acetylneuraminic acid. The genetically modified cells of the present invention may comprise one or more pathways for producing nucleotide-activated sugars selected from glucose-UDP-GlcNAc, GDP-fucose, UDP-galactose, UDP-glucose, UDP-N-acetylglucosamine, UDP-N-acetylgalactosamine, and CMP-N-acetylneuraminic acid (CMP-Neu5Ac).

[0513] In some embodiments, the genetically modified cells are capable of producing one or more of the above-activated sugar nucleotides via a de novo pathway. In this regard, the activated sugar nucleotides are generated by the cell from simple carbon sources (such as glycerol, sucrose, maltose, or glucose) in a stepwise reaction sequence under the action of enzymes involved in the corresponding sugar nucleotide de novo biosynthetic pathway (for a review of monosaccharide metabolism, see, for example, H.H. Freeze and A.D. Elbein: Chapter 4: Glycosylation precursors, in: Essentials of Glycobiology, 2nd Edition (Eds. A. Varki et al.), Cold Spring Harbour Laboratory Press (2009)).

[0514] Enzymes involved in the de novo biosynthetic pathway of activated sugar nucleotides may be naturally present in the cell or may be introduced into the cell by genetic or recombinant DNA techniques, all of which are within the common knowledge of those skilled in the art.

[0515] In an embodiment, the pathway for producing nucleotide-activated sugars is the de novo GDP-fucose pathway (gmd, wcaG, manB, manC, and manA) and / or the sialic acid sugar nucleotide pathway (neuB, neuC, and neuA) as described below.

[0516] In another embodiment, the genetically modified cells can utilize salvaged monosaccharides for sugar nucleotide synthesis. In the salvage pathway, monosaccharides from degraded oligosaccharides are phosphorylated by kinases and converted to nucleotide sugars by pyrophosphorylases. The enzymes involved in this process can be heterologous enzymes or can be native enzymes of the host cell.

[0517] Colanic acid gene cluster

[0518] For the production of fucosylated oligosaccharides / HMOs, the colanic acid gene cluster is crucial for ensuring the presence of sufficient GDP-fucose. In Escherichia coli, GDP-fucose is an intermediate in the production of the exopolysaccharide colanic acid, a major oligosaccharide of the bacterial cell wall. In the present invention, the colanic acid gene cluster (from Escherichia coli, SEQ ID NO: 109) encodes most of the enzymes (gmd, wcaG, wcaH, wcaI, manB, manC) involved in the de novo synthesis of GDP-fucose, and one or more genes downstream of GDP-L-fucose (such as wcaJ) can be deleted to prevent the conversion of GDP-fucose to colanic acid.

[0519] The de novo GDP-fucose pathway genes responsible for forming GDP-fucose include or consist of the following genes: i) manA, which encodes the protein mannose-6-phosphate isomerase (EC 5.3.1.8, UniProt accession number P00946), which facilitates the interconversion of fructose-6-phosphate (F6P) and mannose-6-phosphate; ii) manB, which encodes the protein phosphomannomutase (EC 5.4.2.8, UniProt accession number P24175), which participates in the biosynthesis of GDP-mannose by catalyzing the conversion of mannose-6-phosphate to mannose-1-phosphate; iii) manC encodes the protein mannose-1-phosphate guanylyltransferase (EC: 2.7.7.13, UniProt accession number P24174), which participates in the biosynthesis of GDP-mannose by synthesizing GDP-mannose from GTP and α-D-mannose-1-phosphate; iv) gmd encodes the protein GDP-mannose-4,6-dehydratase (UniProt accession number P0AC88), which catalyzes the conversion of GDP-mannose to GDP-4-dehydro-6-deoxy-D-mannose; v) wcaG (fcl), which encodes the protein GDP-L-fucose synthase (EC 1.1.1.271, UniProt accession number P32055), which catalyzes the NADP-dependent two-step conversion of GDP-4-dehydro-6-deoxy-D-mannose into GDP-fucose.

[0520] Therefore, preferably, when a genetically engineered cell produces one or more fucosylated heterologous products, the entire colanic acid gene cluster and / or one or more genes of the de novo GDP-fucose pathway selected from manA, manB, manC, gmd, and wcaG are overexpressed.

[0521] In one or more exemplary embodiments, the colanic acid gene cluster responsible for GDP-fucose formation can be expressed from its native genomic locus. The expression can be actively regulated to increase GDP-fucose formation. The expression can be regulated by exchanging the native promoter with a target promoter, and / or by expressing the gene cluster from other loci outside the native locus, or by episomally expressing the colanic acid gene cluster or its specific genes to increase the copy number of the colanic acid genes encoding the proteins.

[0522] For the purposes of the present disclosure, the term "native locus" in relation to the colanic acid gene cluster refers to the original and natural position of the gene cluster in the genome of a genetically engineered cell.

[0523] Sialic acid sugar nucleotide synthesis pathway

[0524] If a genetically modified cell is capable of producing sialylated oligosaccharides / HMOs, then the genetically modified cell should possess sialic acid sugar nucleotide synthesis capabilities, i.e., the genetically modified cell contains a biosynthetic pathway for generating sialic acid sugar nucleotides, such as using CMP-N-acetylneuraminic acid as the glycosyl donor for sialyltransferases. For example, the genetically modified cell provides an exogenous UDP-GlcNAc 2-epimerase (such as neuC of Campylobacter jejuni (GenBank AAK91727.1) or an equivalent (such as (GenBank CAR04561.1)), Neu5Ac synthase (such as neuB of Campylobacter jejuni (GenBank AAK91726.1) or an equivalent, (such as sialic acid synthase of Flavobacterium limnosediminis, GenBank WP_023580510.1)) and / or CMP-Neu5Ac synthase (such as neuA of Campylobacter jejuni (GenBank AAK91728.1) or an equivalent, (such as CMP-sialic acid synthase of Vibrio brasiliensis, GenBank WP_006881452.1)). SEQ ID NO:94 disclosed herein is an example of the neuBCA gene cluster from Campylobacter jejuni, and other functional variants are also suitable for preparing sialic acid sugar nucleotides in genetically modified cells.

[0525] In addition, the genetically modified cells preferably have a defective sialic acid catabolic pathway. The "sialic acid catabolic pathway" refers to a series of reactions that are typically controlled and catalyzed by enzymes, leading to the degradation of sialic acid. The exemplary sialic acid catabolic pathway described below is the Escherichia coli pathway. In this pathway, sialic acid (Neu5Ac; N-acetylneuraminic acid) is degraded by the enzymes NanA (N-acetylneuraminic acid lyase), NanK (N-acetylmannosamine kinase), and NanE (N-acetylmannosamine-6-phosphate epimerase), which are all encoded by the nanATEK-yhcH operon and are repressed by NanR (http: / / ecocyc.org / ECOLI). The sialic acid catabolic pathway in the Escherichia coli host is inactivated by introducing one or more mutations in the endogenous nanA (N-acetylneuraminic acid lyase) (e.g., GenBank accession number D00067.1 (GL216588)) and / or nanK (N-acetylmannosamine kinase) gene (e.g., GenBank accession number (amino acid) BAE77265.1 (GL85676015)) and / or nanE (N-acetylmannosamine-6-phosphate epimerase, GI: 947745, incorporated herein by reference). Optionally, the nanT (N-acetylneuraminic acid transporter) gene is also inactivated or mutated. Inactivation means that the coding sequence has been altered such that the resulting gene product is inactivated, or the activity of the encoded gene product is less than 100%, such as 90%, 80%, 70%, 60%, 50%, 40%, 30%, or 20% of its naturally occurring endogenous gene product. Thus, in the present invention, the nanA, nanK, nanE, and / or nanT genes are preferably inactivated.

[0526] Lactose importer

[0527] In embodiments where the genetically modified cells use lactose or LacNAc, LNB, 2’FL, or 3FL as the initial substrate for oligosaccharide formation, the cells are preferably capable of importing the substrate into the cell, unless the initial substrate is produced in situ by the cell itself.

[0528] In embodiments of the dual-strain mixing method, at least one genetically modified cell is capable of importing into the cell a by-product generated as a leaving group during the transglycosylation process (enzymatic process) of the mixing process.

[0529] In embodiments of the dual-strain mixing method, the genetically modified cells comprise a by-product importer. Preferably, the by-product importer can import one or more of the following by-products: lactose, 2’FL, and / or 3FL.

[0530] Most lactose importers are capable of importing lactose, LNB, LacNAc, and 2’FL simultaneously. In embodiments where the initial substrate for oligosaccharide production or a by-product in a dual-strain mixing process is lactose, LNB, LacNAc, or 2’FL, the genetically modified cells have a functional lactose importer or a 2’FL importer. Lactose importers are well known in a variety of species including bacteria and yeast.

[0531] The lactose importer can be, for example, lactose permease. The lactose permease can be an endogenous lactose permease naturally expressed by the cell used to produce the first oligosaccharide.

[0532] In one or more embodiments, one or both genetically modified cells contain one or more overexpressed lactose permease genes.

[0533] One or both genetically engineered cells can contain at least one, for example, at least two, three, or four nucleic acid sequences encoding lactose permease.

[0534] In one or more further exemplary embodiments, the one or more lactose permeases are encoded by heterologous and / or recombinant nucleic acid sequences. A native lactose permease can be genetically engineered, for example, to be under the control of a promoter stronger than the native promoter, thereby creating a recombinant lactose permease gene that overexpresses the native lactose permease protein.

[0535] In one or more preferred exemplary embodiments, the nucleic acid sequence encoding the one or more lactose permeases is a native gene of the genetically engineered cell.

[0536] In Escherichia coli, lactose permease is encoded by the lacY gene in the lactose operon. In an exemplary embodiment, the lactose permease in the genetically modified cell is LacY from Escherichia coli. Preferably, the lactose permease comprises or consists of the amino acid sequence of SEQ ID NO:1, or a lactose permease having an amino acid sequence that has at least 80%, for example at least 90%, for example at least 95%, for example at least 99% or 100% identity to SEQ ID NO:1.

[0537] Transporters for oligosaccharide export

[0538] As described above, preferably, the first and second genetically modified cells grown in co-culture are capable of exporting the first and second oligosaccharide products, respectively. This is particularly preferred for a dual-strain mixing process, where the first and second oligosaccharides undergo an enzymatic reaction in the culture broth. Similarly, in a dual-strain system, when the first oligosaccharide is an intermediate oligosaccharide imported by the second genetically modified cell, it is desirable to export the oligosaccharide produced by the first genetically modified cell into the culture medium for use by the second genetically modified cell.

[0539] In embodiments of the present disclosure, the genetically modified cell preferably comprises at least one nucleic acid sequence encoding one or more transporters capable of exporting the first and / or second oligosaccharide from the cell into the culture medium. The genetically modified cell of the present disclosure preferably expresses a sugar efflux transporter (SET) or a heterologous major facilitator superfamily (MFS) transporter.

[0540] A sugar efflux transporter (SET) capable of exporting certain HMOs is described in WO2010142305. In particular, SETA with UniProt accession number P31675 or a functional variant thereof may contribute to the export of 2’FL and 3FL.

[0541] Transporters of the major facilitator superfamily (MFS) facilitate the transport of molecules (such as, but not limited to, oligosaccharides) across the cell membrane.

[0542] The term "MFS transporter" herein refers to a protein that facilitates the passage of oligosaccharides (preferably HMOs) through or across the cell membrane, from the cell cytosol to the cell periplasm and / or the culture medium. Preferably, the MFS transporter transports HMOs / oligosaccharides synthesized by the genetically modified cells described herein. Additionally, or alternatively, the MFS transporter may also facilitate the efflux of molecules that are not HMOs or oligosaccharides (such as lactose, glucose, cell metabolites, and / or toxins). In a preferred embodiment, the MFS transporter is capable of exporting 2’FL, 3FL, 3’SL, 6’SL, LNT-II, LNT, LNnT, and / or LNFP-I from the cell cytosol into the cell culture medium.

[0543] In the context of the present invention, lactose permease is not considered a heterologous MFS transporter.

[0544] In one or more exemplary embodiments, the MFS transporter is selected from Bad, Nec, YjhB, YberC, Fred, Vag, and Marc.

[0545] Thus, in one or more exemplary embodiments, the genetically modified cell of the present disclosure expresses a heterologous MFS transporter selected from Vag, Nec, Fred, Marc, YberC, Bad, and a functional homolog of any one of Vag, Nec, Fred, Marc, YberC, or Bad, and having an amino acid sequence that is 80% identical to said protein.

[0546] Bad

[0547] The MFS transporter referred to herein as "Bad protein" or "Bad transporter" or "Bad" has an amino acid sequence corresponding to GenBank accession number WP_017489914.1.

[0548] In one or more embodiments of the present invention, the genetically engineered cell expresses a heterologous MFS transporter bad or a functional homolog thereof having an amino acid sequence that is at least 80%, such as at least 90%, such as at least 95%, such as at least 99% or 100% identical to GenBank accession number WP_017489914.1.

[0549] Nec

[0550] The MFS transporter interchangeably referred to herein as "Nec protein" or "Nec transporter" or "Nec" has an amino acid sequence corresponding to SEQ ID NO: 107 or GenBank accession number WP_092672081.1.

[0551] In one or more embodiments of the present invention, the genetically engineered cell expresses a heterologous MFS transporter Nec or a functional homolog thereof having an amino acid sequence that is at least 80%, such as at least 90%, such as at least 95%, such as at least 99% or 100% identical to the amino acid sequence of SEQ ID NO: 107 or GenBank accession number WP_092672081.1.

[0552] Nec is particularly suitable for transporting 2’FL, DFL, LNT, 3’SL, 6’SL, and LNFP-I.

[0553] YjhB

[0554] The MFS transporter interchangeably referred to herein as "YhjB protein" or "YjhB transporter" or "YjhB" has an amino acid sequence corresponding to UniProt accession number P39352.

[0555] In one or more embodiments of the present invention, the genetically engineered cell expresses a heterologous MFS transporter YjhB or a functional homolog thereof having an amino acid sequence that is at least 80%, such as at least 90%, such as at least 95%, such as at least 99% or 100% identical to the amino acid sequence of UniProt accession number P39352.

[0556] YjhB is particularly useful in the output of LNT-II.

[0557] YberC

[0558] The MFS transporter, which may be interchangeably referred to herein as "YberC protein" or "YberC transporter" or "YberC", has an amino acid sequence corresponding to SEQ ID NO: 103 or GenBank accession number EEQ08298.1.

[0559] In one or more embodiments of the invention, the genetically engineered cell expresses a heterologous MFS transporter YberC or a functional homolog thereof, the amino acid sequence of which has at least 80%, such as at least 90%, such as at least 95%, such as at least 99% or 100% identity to the amino acid sequence of SEQ ID NO: 103 or GenBank accession number EEQ08298.1.

[0560] YberC is particularly useful in transporting LNT.

[0561] Fred

[0562] The MFS transporter, which may be interchangeably referred to herein as "Fred protein" or "Fred transporter" or "Fred", has an amino acid sequence corresponding to GenBank accession number WP_087817556.1.

[0563] In one or more exemplary embodiments, the MFS transporter expressed according to the present disclosure is Fred. Thus, in one or more exemplary embodiments, the genetically engineered cells of the present disclosure express a heterologous MFS transporter Fred.

[0564] In one or more embodiments of the invention, the genetically engineered cell expresses a heterologous MFS transporter fred or a functional homolog thereof, the amino acid sequence of which has at least 80%, such as at least 90%, such as at least 95%, such as at least 99% or 100% identity to the amino acid sequence of GenBank accession number WP_087817556.1.

[0565] Fred is particularly useful in transporting 3'SL and 6'SL.

[0566] Vag

[0567] The MFS transporter, which may be interchangeably referred to herein as "Vag protein" or "Vag transporter" or "Vag", has an amino acid sequence corresponding to SEQ ID NO: 106 or GenBank accession number WP_048785139.1.

[0568] In one or more embodiments of the present invention, the genetically engineered cell expresses a heterologous MFS transporter vag or a functional homolog thereof, the amino acid sequence of which has at least 80%, such as at least 90%, such as at least 95%, such as at least 99% or 100% identity to the amino acid sequence of SEQ ID NO: 106 or GenBank accession number WP_048785139.1.

[0569] Vag is particularly useful for transporting LNnT.

[0570] Marc

[0571] The MFS transporter, which may be interchangeably referred to herein as "Marc protein" or "Marc transporter" or "Marc", has an amino acid sequence corresponding to SEQ ID NO: 90 or GenBank accession number WP_060448169.1.

[0572] In one or more embodiments of the present invention, the genetically engineered cell expresses a heterologous MFS transporter marc or a functional homolog thereof, the amino acid sequence of which has at least 80%, such as at least 90%, such as at least 95%, such as at least 99% or 100% identity to the amino acid sequence of SEQ ID NO: 90 or GenBank accession number WP_060448169.1.

[0573] Marc is particularly useful for transporting 3FL.

[0574] Edic1

[0575] The MFS transporter, which may be interchangeably referred to herein as "Edic1 protein" or "Edic1 transporter" or "Edic1", is derived from Edwardsiella ictalurid and has an amino acid sequence corresponding to SEQ ID NO: 60 or GenBank accession number WP_015873007.1.

[0576] In one or more embodiments of the present invention, the genetically engineered cell expresses a heterologous MFS transporter marc or a functional homolog thereof, the amino acid sequence of which has at least 80%, such as at least 90%, such as at least 95%, such as at least 99% or 100% identity to the amino acid sequence of SEQ ID NO: 60 or GenBank accession number WP_015873007.1.

[0577] Edic1 is particularly suitable for transporting LNT or LNnT.

[0578] Thus, in one or more exemplary embodiments, the genetically engineered cells of the present disclosure express a heterologous MFS transporter, which is Vag, Nec, Fred, Marc, Edic1, YjhB, YberC or Bad.

[0579] In one or more exemplary embodiments, the genetically engineered cells of the present disclosure express a functional homolog of Vag, Nec, Fred, Marc, Edic1, YjhB, YberC and / or Bad, the amino acid sequence of which has at least 70%, 80%, 85%, 90%, 95% or at least 99% identity with the GenBank numbers of the above-mentioned Vag, Nec, Fred, Marc, Edic1, YberC and / or Bad.

[0580] In a preferred embodiment of the present invention, the expressed MFS transporter is Nec.

[0581] In a particularly preferred embodiment, the expressed MFS transporter is YberC.

[0582] In a particularly preferred embodiment, the expressed MFS transporter is Marc.

[0583] In a particularly preferred embodiment, the expressed MFS transporter is Edic1.

[0584] In a particularly preferred embodiment, the expressed MFS transporter is YjhB.

[0585] In a particularly preferred embodiment, the expressed MFS transporter is Vag.

[0586] In a particularly preferred embodiment, the expressed MFS transporter is Fred.

[0587] Sequence identity

[0588] The term "sequence identity" as used herein describes the relatedness between two amino acid sequences or between two nucleotide sequences based on their paired comparison, the two amino acid sequences being a candidate sequence (e.g., a sequence of the present disclosure) and a reference sequence (e.g., a prior art sequence). For the purposes of the present disclosure, the sequence identity between two amino acid sequences is determined using the Needleman-Wunsch algorithm (Needleman and Wunsch, 1970, J. Mol. Biol. 48:443-453), which is implemented in the Needle program of the EMBOSS package (EMBOSS: The European Molecular Biology Open Software Suite, Rice et al., 2000, Trends Genet. 16:276-277), preferably version 5.0.0 or higher (available at https: / / www.ebi.ac.uk / Tools / psa / emboss_needle / ). The parameters used are a gap open penalty of 10, a gap extension penalty of 0.5, and the EBLOSUM62 (the EMBOSS version of BLOSUM62) substitution matrix. The output of Needle labeled "identity" (obtained using the -nobrief option) is used as the percent identity. Typically sequence identity can be calculated as follows: (identical residues x 100) / (length of the alignment region).

[0589] For the purposes of the present disclosure, the sequence identity between two nucleotide sequences is determined using the Needleman-Wunsch algorithm (Needleman and Wunsch, 1970, ibid.) implemented in the Needle program of the EMBOSS package (EMBOSS: The European Molecular Biology Open Software Suite, Rice et al., 2000, Trends Genet. 16:276-277), preferably version 5.0.0 or higher. The parameters used are a gap open penalty of 10, a gap extension penalty of 0.5, and the DNAFULL (the EMBOSS version of NCBI NUC4.4) substitution matrix. The output of Needle labeled "identity" (obtained using the -nobrief option) is used as the percent identity. Typically sequence identity can be calculated as follows: (identical deoxyribonucleotides x 100) / (length of the alignment region).

[0590] Functional homolog

[0591] Functional homologs or functional variants of the protein / nucleic acid sequences described herein refer to protein / nucleic acid sequences in which the genetic code or amino acid sequence has been altered but still retains its original function. Functional variants can be obtained by mutagenesis or can be natural variants from the same or other species. Compared to the function of the protein / nucleic acid sequence, functional homologs should have at least 50%, such as at least 60%, 70%, 80%, 90% or 100% of the remaining function / activity. In embodiments of the present invention, the functional variant has at least 80% identity, such as at least 85% identity, such as at least 90% identity, such as at least 95% identity with the protein / nucleic acid sequence shown in a given protein, nucleic acid or gene.

[0592] Functional variants or homologs may also exist across species, i.e., different species (such as but not limited to Escherichia coli, Bacillus, Corynebacterium, Lactobacillus, Saccharomyces) may have proteins with similar functions, such as various sugar transport systems, including the phosphoenolpyruvate:sugar phosphotransferase system (PTS). In the context of the present disclosure, such systems that exist across multiple species are considered functional variants. For convenience, only one NCBI or UniProt reference is provided to characterize such proteins, but it should be understood that proteins from other species with equivalent functions should also be considered functional variants.

[0593] Functional variants of proteins or peptides may contain conservative amino acid substitutions compared to their native (i.e., unmutated physiological sequence). These amino acid sequences and their encoding nucleotide sequences particularly fall within the term "functional variant" as defined herein. Substitutions in which amino acids from the same category are replaced with each other are called conservative substitutions. Specifically, these amino acids include those with aliphatic side chains, positively or negatively charged side chains, amino acids with aromatic groups in their side chains, or amino acids whose side chains can form hydrogen bonds, such as side chains with hydroxyl functional groups. This means, for example, that an amino acid with a polar side chain is replaced by another amino acid with a similarly polar side chain, or, for example, an amino acid with a hydrophobic side chain is replaced by another amino acid with a similarly hydrophobic side chain (e.g., serine (threonine) is replaced by threonine (serine), leucine (isoleucine) is replaced by isoleucine (leucine)). Truncations, insertions, and substitutions are possible, especially at sequence positions that do not alter the three-dimensional structure or do not affect the binding region. Modifications to the three-dimensional structure caused by insertions or deletions can be easily determined, for example, using circular dichroism (CD) spectroscopy (Urry, 1985, Absorption, Circular Dichroism and ORD of Polypeptides, in: Modern Physical Methods in Biochemistry, Neuberger et al. (ed.), Elsevier, Amsterdam).

[0594] Furthermore, functional variants of proteins or peptides as defined herein may also contain sequences in which the nucleotides of the nucleic acid are replaced according to the degeneracy of the genetic code without resulting in a change in the respective amino acid sequence of the protein or peptide, i.e., the amino acid sequence or at least a part thereof does not differ from the original sequence in terms of mutations within one or more of the above meanings.

[0595] Recovery / Harvest

[0596] One or more oligosaccharides produced using the two-strain system (co-culture) method described herein can be recovered from the culture medium and / or the cells of the process. In this context, "recovery" and "harvest" are used interchangeably. "Recovery" and "harvest" in the context both refer to collecting the produced HMOs from the culture (cells) / culture broth (culture medium) after the end of the process. In the two-strain mixing process, it is preferred to discard the biomass (cells) as they only contain the first and second oligosaccharides, while the target oligosaccharides are complex fucosylated or sialylated oligosaccharides produced by enzymatic reactions in the culture medium.

[0597] The separation of cells from the culture medium can be carried out using any method well-known to those skilled in the art, such as any suitable type of centrifugation or filtration. The separation of cells from the culture medium can be carried out immediately after harvesting the fermentation broth or after storing the fermentation broth under appropriate conditions.

[0598] After recovering the mixed process medium, the HMO mixture / composition can be used for further processing and purification. It may be necessary to separate individual HMOs from the HMO mixture to obtain, for example, purified or enriched sialylated and / or fucosylated oligosaccharides (containing at least four monosaccharides). Alternatively, the HMO mixture produced by the mixed process can be purified to remove lactose and other non-HMO metabolic by-products (e.g., by ultrafiltration and / or nanofiltration), and then the HMO mixture or HMO composition can be used directly.

[0599] The purification of specific components in the HMO composition or HMO mixture can be carried out according to methods known to those skilled in the art. For example, HMOs can be purified according to methods known in the art, such as those described in WO2015 / 188834, WO2017 / 182965 or WO2017 / 152918, where WO2017 / 152918 describes a method for purifying HMOs.

[0600] Oligosaccharide mixtures, such as HMOs and their compositions

[0601] The methods described herein can produce one or more oligosaccharides. Generally, the methods produce a mixture of oligosaccharides. Using the co-culture method, two HMOs are mixed in the desired ratio, and this mixture may also contain a relatively small amount of by-product oligosaccharides (e.g., the intermediate LNT-II produced during the production of LNT and LNnT, or 2’FL and 3FL produced during the production of DFL), for example less than 20%, less than 15% or less than 10%. After fermentation, the content of by-product oligosaccharides may also be high, but it can be reduced to the desired level during the downstream purification process to produce the desired oligosaccharide composition, such as HMO.

[0602] Complex fucosylated and / or sialylated HMOs, such as the LST-c and LST-a mixture, have high application value as nutritional supplements or therapeutic agents, as described in the section on the uses of HMO mixtures in this application.

[0603] One aspect of the present disclosure is an HMO mixture or composition consisting essentially of:

[0604] a) at least 40 wt% of LST-c, less than 25 wt% of LNnT, less than 25 wt% of 6’SL and less than 10 wt% of lactose, or

[0605] b) at least 60 wt% of LST-a, less than 30 wt% of LNT, less than 15 wt% of 3'SL and less than 2 wt% of lactose, and

[0606] wherein the total composition constitutes 100 wt% of the respective components, and the composition is a mixture of at least two components.

[0607] In an embodiment, the composition or mixture of HMOs consists essentially of:

[0608] a) at least 50 wt% of LST-c, 15 to 25 wt% of LNnT, 15 to 25 wt% of 6'SL and 0 to 7 wt% of lactose;

[0609] b) at least 60 wt% of LST-a, 15 to 30 wt% of LNT, 0 to 15 wt% of 3'SL and 0 to 2 wt% of lactose;

[0610] wherein the total composition constitutes 100 wt% of the respective components.

[0611] In other embodiments, the HMO mixture described herein consists essentially of 45 - 55 wt% of LST-c, 20 - 30 wt% of LNnT and 20 - 30 wt% of 6'SL, wherein the total composition constitutes 100 wt% of the respective components.

[0612] In other embodiments, the HMO mixture described herein consists essentially of 50 wt% of LST-c, 25 wt% of LNnT and 25 wt% of 6'SL.

[0613] In other embodiments, the HMO mixture described herein consists essentially of 60 to 80 wt% of LST-a, 20 to 30 wt% of LNT and 5 to 15 wt% of 3'SL, wherein the total composition constitutes 100 wt% of the respective components.

[0614] In other embodiments, the HMO mixture described herein consists essentially of 65 wt% of LST-a, 25 wt% of LNT and 10 wt% of 3'SL.

[0615] As shown in the examples, compared with traditional in vitro processes, the mixing process of the present disclosure can increase the ratio of the desired complex fucosylated or sialylated HMOs to the donor and / or acceptor HMOs. In traditional in vitro methods, the ratio between individual HMOs is limited by the kinetic barriers of enzymatic reactions, thereby preventing the balance between the components of the in vitro enzymatic reaction from shifting.

[0616] In other embodiments, the molar ratio of LST-c:LNnT in the HMO composition and / or mixture is higher than 2.5:1, and the molar ratio of LST-c:6’SL is higher than 2.5:1.

[0617] In other embodiments, the molar ratio of LST-a:LNT in the HMO composition and / or mixture is higher than 1.5:1, and the molar ratio of LST-a:3’SL is higher than 8:1.

[0618] In some embodiments, the composition comprising the HMO mixture is a nutritional composition. The nutritional composition is, for example, an infant formula or a medical nutritional composition.

[0619] In some embodiments, the composition comprising the HMO mixture is a dietary supplement.

[0620] In some embodiments, the composition comprising the HMO mixture is a pharmaceutical composition.

[0621] Use of the HMO mixture and composition

[0622] Infant clinical data show that human milk oligosaccharide supplements can serve as a food source for beneficial gut bacteria, thus helping to form the desired flora. HMOs are naturally present in breast milk and have evolved over thousands of years. HMO research (clinical and preclinical) has now shown that specific HMOs can bring unique health benefits to us at appropriate supplementation levels. In particular, human milk oligosaccharide supplements may help enhance immunity and gut health, including maintaining a balanced microbiota, and may play a role in cognitive development, which may open up future innovation opportunities.

[0623] Accordingly, in an embodiment, the present invention relates to the use of the mixture or composition disclosed herein in infant nutrition.

[0624] The present invention also relates to the use of the mixture or composition disclosed herein as a dietary supplement, a medical nutritional product or a pharmaceutical composition.

[0625] The HMO mixture or composition produced according to the method described herein can be used to enhance the beneficial flora in the gut microbiota. The beneficial bacteria are, for example, bacteria of the genus Bifidobacterium, Lactobacillus, or Barnesiella sp. The enhancement of beneficial bacteria, in turn, can lead to an increase in the production of short-chain fatty acids (SCFAs) such as acetate, propionate, and butyrate, which have been shown to have many benefits for infants and young children, such as inhibiting pathogenic bacteria, preventing infections and diarrhea, and reducing the risk of allergies and metabolic disorders (see, for example, WO2006 / 130205, WO 2017 / 129644, WO2017 / 129649).

[0626] The HMO mixtures or compositions produced according to the methods described herein can be used to reduce the numbers of harmful viruses and bacteria in the gut microbiota. Examples of pathogenic bacteria and viruses that can be reduced by the HMO mixtures described herein include Candida albicans, Clostridioides difficile, Enterococcus faecium, Escherichia coli, Helicobacter pylori, Streptococcus agalactiae, Shigella dysenteriae, Staphylococcus aureus, norovirus, and rotavirus. Each of the compositions described herein can also be used to treat and / or reduce the risk of various bacterial infections in humans.

[0627] The HMO mixtures or compositions produced according to the methods described herein can be used to improve the regeneration and viability of lyophilized probiotics (including probiotics of the genera Bifidobacterium and Lactobacillus), particularly to improve regeneration and / or viability and / or shelf life in an acidic environment (such as the stomach or acidic foods), which is an advantage of using the HMO mixtures described herein. Examples of Bifidobacterium that can have increased regeneration ability and viability are Bifidobacterium animalis subsp. lactis BB12 DSM 32269, Bifidobacterium animalis BIF6, Bifidobacterium longum subsp. longum DSM 32946, Bifidobacterium longum BB536, Bifidobacterium bifidum DSMZ 32403, Bifidobacterium infantis, Bifidobacterium breve DSM 33789, Bifidobacterium infantis SP37 DSM 32687, Bifidobacterium adolescentis DSM 34065, and / or Bifidobacterium animalis subsp. lactis DSM 16284. Examples of Lactobacillus that can have increased regeneration ability and viability are Lactobacillus rhamnosus GG DSM 32550, Lactobacillus rhamnosus 19070-2 DSM 26357, Lactobacillus rhamnosus GG, Lactobacillus rhamnosus LBrGG, Lactobacillus reuteri DSM 12246, Lactobacillus plantarum TIFN101, Lactobacillus gasseri Lg-36 200B FloraFitDanisco, Lactobacillus casei DSM 32382, Lactobacillus paracasei, Lactobacillus plantarum PS128, Lactobacillus plantarum (Sacco) DSM32383, Lactococcus lactis PAREVE, Lactobacillus paracasei subsp. paracasei, and / or Lactobacillus reuteri S12 DSM 33752. Lactobacillus reuteri S12 DSM 33752.

[0628] In the present application, "regeneration" refers to the process of restoring / reviving the viability of dried bacteria (i.e., "reviving" the bacterial cells by rehydration, where "rehydration" refers to restoring the liquid). This process is sometimes also referred to as "reconstitution".

[0629] In the present application, "viability" refers to the ability of bacterial cells to survive and perform the functions of living cells. One method of determining the viability of bacterial cells is to plate them on an agar plate containing a suitable growth medium and count the number of colonies formed after incubating for a predetermined time (plate count). Alternatively, FACS analysis can also be used.

[0630] In the present application, "improving the regeneration of Bifidobacterium and / or Lactobacillus bacteria" means increasing the number of Bifidobacterium and / or Lactobacillus. Bacteria that are successfully regenerated / restored compared to the corresponding control (i.e., the number / quantity of Bifidobacterium and / or Lactobacillus without the addition of HMO).

[0631] In the present application, "increasing the viability of Bifidobacterium and / or Lactobacillus" means increasing the number (quantity) of viable Bifidobacterium and / or Lactobacillus compared to the corresponding control (i.e., the number / quantity of Bifidobacterium and / or Lactobacillus without the addition of HMO).

[0632] In the present application, "acidic" means a pH value below 7.0 (e.g., pH ≤ 6.0, ≤ 5.0, ≤ 4.0, ≤ 3.0, or in the range of 1.0 - 6.0, e.g., 2.0 to 5.0). The pH value measured in the stomach is in the range of about 1.5 - 3.5. The pH value measured in a healthy vagina is in the range of about 3.8 - 5.0. The pH value of fruit juice is in the range of about 2.0 - 4.5.

[0633] The HMO mixture or composition produced according to the methods described herein can be used to extend the shelf life of probiotics (e.g., Bifidobacterium and / or Lactobacillus).

[0634] One embodiment of the present invention is a composition comprising the HMO mixture described herein (particularly in the "HMO mixture" section) and one or more probiotics. Preferably, the probiotics are Bifidobacterium and / or Lactobacillus, such as any of the specific strains mentioned above.

[0635] The HMO mixture or composition produced according to the methods described herein can be used to improve the flowability of powders or reduce the viscosity of liquids.

[0636] The HMO mixture or composition produced according to the methods described herein can be used in nutritional compositions. Nutritional compositions include, for example, infant formula, oral rehydration salts, or dietary maintenance, medical nutrition, or supplements for the elderly or immunocompromised. Such anti-infection compositions may also contain macronutrients such as dietary fats, carbohydrates, and proteins. Dietary fats include, for example, coconut oil, soybean oil, and monoglycerides and diglycerides. Carbohydrates include, for example, glucose, edible lactose, and hydrolyzed corn starch. Proteins include, for example, soy protein, whey, and skim milk. Vitamins and minerals (e.g., calcium, phosphorus, potassium, sodium, chloride, magnesium, manganese, iron, copper, zinc, selenium, iodine, and vitamins A, E, D, C, and B complex) may also be included in such anti-infection compositions.

[0637] Sequence Listing

[0638] The present application contains a sequence listing in text format and electronic format, which is incorporated herein by reference.

[0639] The following table shows an overview of the SEQ ID NOs used in this application.

[0640] Summary of the sequences listed in the application:

[0641]

[0642]

[0643] Embodiments

[0644] The following embodiments of the invention can be used in combination with any other embodiments described herein.

[0645] 1. A method for producing one or more oligosaccharides having at least three monosaccharide units, the method comprising the step of co-culturing first and second genetically modified microbial cells in a culture medium, wherein,

[0646] a) the first genetically modified microbial cell is capable of producing a disaccharide or a first oligosaccharide of at least three monosaccharide units, and wherein the genetically modified cell

[0647] i) is capable of growing on a first carbon source while being restricted or not growing on a second carbon source, and

[0648] ii) comprises one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase; and

[0649] iii) comprises at least one pathway for producing an activated sugar nucleotide from the first carbon source;

[0650] And

[0651] iv) is preferably capable of exporting the first oligosaccharide into the culture medium; and

[0652] b) the second genetically modified microbial cell is capable of producing a second oligosaccharide of at least three monosaccharide units, and wherein the genetically modified cell

[0653] i) is capable of growing on the second carbon source while being restricted or not growing on the first carbon source; and

[0654] ii) comprises one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase; and

[0655] iii) comprises a biosynthetic pathway for producing an activated sugar nucleotide from the second carbon source;

[0656] And

[0657] iv) is preferably capable of exporting the second oligosaccharide into the culture medium.

[0658] 2. The method according to item 1, wherein if a disaccharide is produced in step a), the disaccharide is not lactose.

[0659] 3. The method according to item 1 or 2, wherein if a disaccharide is produced in step a), the disaccharide is lacto-N-biose (LNB) or N-acetyl lactosamine (LacNAc).

[0660] 4. The method according to items 1 to 3, wherein the first carbon source is selected from glucose, glycerol, sucrose, maltose, galactose, fructose, sorbitol, arabinose, and maltose, and the second carbon source is selected from sucrose, glycerol, galactose, maltose, fructose, sorbitol, arabinose, and glucose, and wherein the first carbon source and the second carbon source are different.

[0661] 5. The method according to any one of the foregoing, wherein one of the genetically modified microbial cells is capable of growing on sucrose and comprises one or more nucleic acid sequences encoding a PTS-dependent sucrose utilization system or a nucleic acid encoding an invertase or a sucrose hydrolase that enables the cell to assimilate sucrose.

[0662] 6. The method according to any one of item 5, wherein the PTS-dependent sucrose utilization system is encoded by scrY (SEQ ID NO:97), scrA (SEQ ID NO:98), scrB (SEQ ID NO:99), and optionally scrR (SEQ ID NO:100) or by the cscABKR gene cluster (SEQ ID NO:110), and the invertase is encoded by SacC_Agal (SEQ ID NO:111) or Bff (SEQ ID NO:112) or a functional variant of any of these sequences.

[0663] 7. The method according to item 5 or 6, wherein the ability of the cell to grow on one or more carbon sources selected from glucose, glycerol, galactose, maltose, sorbitol, arabinose, and fructose is reduced or lost.

[0664] 8. The method according to items 1 to 3, wherein one of the genetically modified microbial cells is capable of growing on glucose and comprises one or more nucleic acids encoding one or more glucose transport systems.

[0665] 9. The method according to item 8, wherein the glucose transport system is a PTS-dependent glucose transport system selected from the following:

[0666] i) Glucose PTS complex component IICB Glc ;

[0667] ii) The β-glucoside PTS complex component - IIABC Bgl ;

[0668] iii) The mannose PTS complex component - IICD Man ;

[0669] iv) The N-acetylglucosamine PTS complex component - IIABC Nag ; and

[0670] v) The maltose / maltodextrin PTS complex - IICB malX .

[0671] 10. The method according to item 8, wherein the glucose transport system is selected from the following:

[0672] i) Galactose:H+ symporter GalP;

[0673] ii) Glucose uptake protein GlcU;

[0674] iii) Sodium / glucose transporter family (SGLT);

[0675] iv) Galactose / glucose ABC transporter (mglABC) system;

[0676] v) Trehalose / maltose / sucrose / palatinose (TMSP)-ABC transporter (malEFG) system;

[0677] vi) Glucose / mannose ABC transporter (glcEFG) system;

[0678] vii) Glucose proton symporter (glcP);

[0679] viii) Glucose facilitator (glf); and

[0680] ix) Hexose transporter (HXT).

[0681] 11. The method according to items 8 to 10, wherein the ability of the cell to grow on one or more carbon sources selected from sucrose, glycerol, galactose, maltose, sorbitol, arabinose, and fructose is reduced or lost.

[0682] 12. The method according to items 1 to 3, wherein the genetically modified microbial cell capable of growing on glycerol comprises one or more nucleic acids encoding one or more glycerol transport systems.

[0683] 13. The method according to item 12, wherein the glycerol transport system is selected from glycerol facilitator or glycerol / H+ symporter.

[0684] 14. The method according to item 12 or 13, wherein the ability of the cells to grow on one or more carbon sources selected from sucrose, glucose, galactose, maltose, sorbitol, arabinose, and fructose is reduced or lost.

[0685] 15. The method according to any one of the preceding items, wherein the first genetically modified microbial cell grows on sucrose, while the second genetically modified microbial cell grows on glucose or glycerol.

[0686] 16. The method according to any one of items 1 to 14, wherein the first genetically modified microbial cell grows on glucose or glycerol, while the second genetically modified microbial cell grows on sucrose.

[0687] 17. The method according to items 1 to 3, wherein the genetically modified microbial cell capable of growing on galactose comprises one or more nucleic acids encoding one or more galactose transport systems.

[0688] 18. The method according to item 17, wherein the galactose transport system is selected from galactose:H+ symporter, galactose / glucose ABC transporter (mglABC) system, PTSLac (lacFE) system, and / or sodium / glucose transporter family (sglT).

[0689] 19. The method according to item 17 or 18, wherein the ability of the cells to grow on one or more carbon sources selected from sucrose, glucose, glycerol, maltose, sorbitol, arabinose, and fructose is reduced or lost.

[0690] 20. The method according to items 1 to 3, wherein the genetically modified microbial cell capable of growing on fructose comprises one or more nucleic acids encoding one or more fructose transport systems.

[0691] 21. The method according to item 20, wherein the galactose transport system is selected from PTS complex component IIABC Fru 、glucose PTS complex component IICB Glc and fructose transporter FruP.

[0692] 22. The method according to item 20 or 21, wherein the ability of the cells to grow on one or more carbon sources selected from sucrose, glucose, glycerol, maltose, sorbitol, arabinose, and galactose is reduced or lost.

[0693] 23. The method according to items 1 to 3, wherein the genetically modified microbial cell capable of growing on maltose comprises one or more nucleic acids encoding one or more maltose transport systems.

[0694] 24. The method according to item 23, wherein the galactose transport system is selected from the MalFGK ABC superfamily transport system and / or the maltose / maltodextrin PTS complex.

[0695] 25. The method according to item 23 or 24, wherein the ability of the cell to grow on one or more carbon sources selected from sucrose, glucose, glycerol, fructose, sorbitol, arabinose, and galactose is reduced or lost.

[0696] 26. The method according to items 1 to 3, wherein the genetically modified microbial cell capable of growing on arabinose comprises one or more nucleic acids encoding one or more arabinose transport systems.

[0697] 27. The method according to item 26, wherein the arabinose transport system is selected from the AraFGH ABC superfamily transport system and / or the arabinose-proton symporter AraE.

[0698] 28. The method according to item 26 or 27, wherein the ability of the cell to grow on one or more carbon sources selected from sucrose, glucose, glycerol, fructose, sorbitol, and galactose is reduced or lost.

[0699] 29. The method according to items 1 to 3, wherein the genetically modified microbial cell capable of growing on sorbitol comprises one or more nucleic acids encoding one or more sorbitol transport systems.

[0700] 30. The method according to item 30, wherein the sorbitol transport system is selected from the sorbitol-PTS system (EIIBC srl ).

[0701] 31. The method according to item 30 or 31, wherein the ability of the cell to grow on one or more carbon sources selected from sucrose, glucose, glycerol, fructose, arabinose, and galactose is reduced or lost.

[0702] 32. The method according to any one of items 5 to 7, items 12 to 14, items 17 to 19, items 20 to 22, or items 23 to 25, wherein the functionality of one or more endogenous proteins involved in glucose import and utilization in the cell is reduced or eliminated, and wherein the proteins are selected from:

[0703] i) Glucose PTS complex component IICB Glc (ptsG)

[0704] ii) β-glucoside PTS complex component IIABC Bgl (bglF)

[0705] iii) Mannose PTS complex component - IICD Man (manX),

[0706] iv) N - acetylglucosamine PTS complex component - IIABC Nag (nagE)

[0707] v) Maltose / maltodextrin transport system (malX)

[0708] vi) Galactose / glucose high - affinity ABC transporter component (mglC)

[0709] vii) Trehalose / maltose / sucrose / palatinose (TMSP) - ABC transporter malF;

[0710] viii) Trehalose / maltose / sucrose / palatinose (TMSP) - ABC transporter malG ix) Galactose permease (galP);

[0711] x) Glucose - proton symporter (glcP);

[0712] xi) Glucose facilitator (glf);

[0713] xii) Glucose uptake protein (glcU);

[0714] xiii) Sodium / glucose transporter family (sglT)

[0715] xiv) Hexose transporter (HXT); and / or

[0716] xv) Glucokinase (glk).

[0717] 33. The method according to item 32, wherein the functionality of proteins involved in glucose and / or glycerol import and utilization is reduced or eliminated by completely or partially inactivating one or more genes selected from ptsG, bglF, manX, nagE, malX, malF, malG, mglC, and glk.

[0718] 34. The method according to any one of items 5 to 7, items 8 to 11, items 17 to 19, items 20 to 22, or items 23 to 25, wherein the functionality of proteins involved in glycerol import and utilization is reduced or eliminated by completely or partially inactivating one or more genes selected from glpF, stl1, and glpk.

[0719] 35. A method according to any one of items 8 to 11, 12 to 14, 17 to 19, 20 to 22 or 23 to 25, wherein the cell does not contain a gene encoding a protein involved in sucrose import and utilization, or wherein the functionality of one or more endogenous proteins in the cell is reduced or eliminated, and wherein the protein is selected from the PTS-dependent sucrose utilization system, in particular sucrose permease, or invertase and sucrose hydrolase.

[0720] 36. A method according to any one of items 5 to 7, 8 to 11, 12 to 14, 20 to 22 or 23 to 25, wherein the functionality of the protein involved in galactose import and utilization is reduced or eliminated by completely or partially inactivating one or more genes selected from galP, lacF, mglA, mglC, sglT and galK.

[0721] 37. A method according to any one of items 5 to 7, 8 to 11, 12 to 14, 17 to 19, or 23 to 25, wherein the functionality of the protein involved in fructose import and utilization is reduced or eliminated by completely or partially inactivating one or more genes selected from fruP, ptsG and fruA.

[0722] 38. A method according to any one of items 5 to 7, 8 to 11, 12 to 14, 17 to 19, or 20 to 22, wherein the functionality of the protein involved in maltose import and utilization is reduced or eliminated by completely or partially inactivating one or more genes selected from malF, malG, malK and malX.

[0723] 39. A method according to item 38, wherein by complete or partial inactivation, the functionality of the protein involved in glucose import and utilization according to items 8 to 11 is also reduced or eliminated.

[0724] 40. A method according to any one of the preceding items, wherein one or more oligosaccharides are harvested from the co-culture.

[0725] 41. A method according to any one of the preceding items, wherein the one or more oligosaccharides are one or more human milk oligosaccharides (HMOs).

[0726] 42. A method according to item 41, wherein the one or more human milk oligosaccharides are a mixture of at least two human milk oligosaccharides.

[0727] 43. The method according to any one of the foregoing, wherein one or more recombinant nucleic acids in the first and second genetically modified microbial cells independently encode at least one glycosyltransferase selected from β-1,3-N-acetylglucosaminyltransferase, β-1,3-galactosyltransferase, β-1,4-galactosyltransferase, α-1,2-fucosyltransferase, α-1,3-fucosyltransferase, α-1,3 / 4-fucosyltransferase, α-1,4-fucosyltransferase, α-2,3-sialyltransferase, and α-2,6-sialyltransferase.

[0728] 44. The method according to any one of the foregoing, wherein the first and second genetically modified microbial cells independently are capable of producing one or more disaccharides or oligosaccharides selected from the group consisting of: LNB, LacNAc, 2’FL, 3FL, 2’FLacNAc, 2’FLNB, Lewis A, Lewis X, 3’SL, 6’SL, 3’SLacNAc, 3’SLNB, sialyl-Lewis A, sialyl-Lewis X, DFL, LNT-II, LNT, LNnT, LNFP-I, LNFP-III, LNFP-IV, LNFP-V, LNFP-VI, FSL, LST-a, LST-b, LST-c, LST-d, LNDFH-II, and LNDFH-III, DSLNT, pLNH, pLNnH, LNH, LNnH, (D)F-LNH-I, (D)F-LNH-II, (D)F-LNH-III, F-para-LNH-I, DF-para-LNH, DF-para-LNnH, TF-LNH, S-LNFP-I (FLSTb), S-LNFP-II (FLSTa), S-LNH, S-LNnH-I, FS-LNH, FS-LNnH-I, DS-F-LNH-II.

[0729] 45. The method according to item 44, wherein the one or more oligosaccharides are HMOs independently selected from 2’FL, 3FL, 3’SL, 6’SL, DFL, LNT-II, LNT, LNnT, LNFP-I, LNFP-III, LNFP-V, LNFP-VI, LST-a, LST-c, LNDFH-II, and LNDFH-III.

[0730] 46. The method according to any one of items 1 to 38 or 40 to 45, wherein the first genetically modified microbial cell comprises β-1,4-galactosyltransferase, the enzyme allowing galactosylation of free glucose monosaccharide to generate lactose intracellularly, and wherein the glucose kinase activity that converts glucose to glucose-6-phosphate in the cell is reduced or eliminated.

[0731] 47. A method according to any one of the preceding claims, wherein the first genetically modified microbial cell is capable of producing LacNAc, LNB, Lewis A, Lewis X, 2’FL, 3FL LNT-II, LNT, LNnT LNFP-I, LST-c or LST-a, preferably without adding lactose to the culture medium.

[0732] 48. A method according to claim 47, wherein the HMO produced by the first genetically modified cell is the most abundant HMO produced by the cell.

[0733] 49. A method according to any one of the preceding claims, wherein the first genetically modified microbial cell comprises a recombinant nucleic acid sequence encoding a transporter capable of exporting the disaccharide or first oligosaccharide product into the extracellular culture medium.

[0734] 50. A method according to item 49, wherein the transporter is a sugar efflux transporter or a major facilitator superfamily (MFS) transporter, preferably selected from setA, yberC, nec, vag, marc, bad and fred.

[0735] 51. A method according to any one of the preceding claims, wherein the first genetically modified microbial cell produces the intermediate LNT-II and comprises

[0736] a) a recombinant nucleic acid sequence encoding a β-1,3-N-acetylglucosaminyltransferase, and

[0737] b) a biosynthetic pathway for preparing UDP-GlcNAc from a carbon source assimilated by the first genetically modified microbial cell, and

[0738] c) optionally, a recombinant nucleic acid sequence encoding an MFS transporter capable of exporting LNT-II into the extracellular culture medium.

[0739] 52. A method according to any one of the preceding claims, wherein the first genetically modified microbial cell produces the intermediate LNT and comprises:

[0740] a) a recombinant nucleic acid sequence encoding a β-1,3-N-acetylglucosaminyltransferase, and

[0741] b) a recombinant nucleic acid sequence encoding a β-1,3-galactosyltransferase, and

[0742] c) a biosynthetic pathway for preparing UDP-GlcNAc and UDP-Gal from a carbon source assimilated by the first genetically modified microbial cell, and

[0743] d) Optionally, a recombinant nucleic acid sequence encoding an MFS transporter capable of exporting LNT into the extracellular culture medium.

[0744] 53. The method according to any one of the preceding claims, wherein the first genetically modified microbial cell produces the intermediate LNnT and comprises:

[0745] a) a recombinant nucleic acid sequence encoding a β-1,3-N-acetylglucosaminyltransferase; and

[0746] b) a recombinant nucleic acid sequence encoding a β-1,4-galactosyltransferase; and

[0747] c) a biosynthetic pathway for preparing UDP-GlcNac and UDP-Gal from a carbon source assimilated by the first genetically modified microbial cell; and

[0748] d) Optionally, a recombinant nucleic acid sequence encoding an MFS transporter capable of exporting LNnT into the extracellular culture medium.

[0749] 54. The method according to any one of the preceding claims, wherein the first genetically modified microbial cell produces the intermediate 2'-FL or 3-FL and comprises:

[0750] a) a recombinant nucleic acid sequence encoding an α-1,2-fucosyltransferase or an α-1,3-fucosyltransferase; and

[0751] b) a biosynthetic pathway for preparing GDP-fucose from a carbon source assimilated by the first genetically modified microbial cell; and

[0752] c) Optionally, a recombinant nucleic acid sequence encoding an MFS transporter capable of exporting 2'-FL or 3-FL into the extracellular culture medium; and

[0753] d) Optionally, a β-1,4-galactosyltransferase that allows free glucose monosaccharide to be galactosylated to generate lactose intracellularly.

[0754] 55. The method according to any one of the preceding claims, wherein the first genetically modified microbial cell produces the intermediate 3'SL or 6'SL and comprises:

[0755] a) a recombinant nucleic acid sequence encoding an α-2,3-sialyltransferase or an α-2,6-sialyltransferase; and

[0756] b) a biosynthetic pathway for preparing CMP-N-acetylneuraminic acid (CMP-Neu5Ac) from a carbon source assimilated by the first genetically modified microbial cell; and

[0757] c) Optionally, a recombinant nucleic acid sequence encoding an MFS transporter capable of exporting 3'SL or 6'SL into the extracellular culture medium; and

[0758] d) Optionally, a β-1,4-galactosyltransferase that allows free glucose monosaccharides to be galactosylated to produce lactose intracellularly.

[0759] 56. The method according to any one of the preceding claims, wherein the second genetically modified microbial cell comprises at least one nucleic acid sequence encoding a protein or protein complex capable of importing a disaccharide or oligosaccharide produced by the first genetically modified microbial cell.

[0760] 57. The method according to claim 56, wherein the protein or protein complex capable of importing the disaccharide or oligosaccharide produced by the first genetically modified microbial cell is selected from Table 1 or Table 2.

[0761] 58. The method according to any one of the preceding claims, wherein the second genetically modified microbial cell comprises

[0762] a) one or more recombinant nucleic acid sequences encoding glycosyltransferases selected from β-1,3-galactosyltransferase, β-1,4-galactosyltransferase, α-1,2-fucosyltransferase, α-1,3-fucosyltransferase, α-1,3 / 4-fucosyltransferase, α-1,4-fucosyltransferase, α-2,3-sialyltransferase and α-2,6-sialyltransferase, and

[0763] b) optionally, a non-functional or deleted lactose permease, and

[0764] c) optionally, a recombinant nucleic acid sequence encoding an MFS transporter capable of exporting the oligosaccharide produced by the second genetically modified microbial cell.

[0765] 59. The method according to claim 58, wherein the MFS transporter is selected from marc, nec, yberC and vag.

[0766] 60. The method according to any one of items 56 to 59, wherein one or more HMOs produced by the microbially modified cells of the second gene modification have at least three monosaccharide units, such as at least four monosaccharide units, and are selected from Lewis A, Lewis X, sialyl-LacNAc, sialyl-LNB, sialyl-Lewis X, sialyl-Lewis A, Lewis B, Lewis Y, DFL, FSL, LNT, LNnT, LST-a, LST-b, LST-c, LST-d, LNFP-I, LNFP-II, LNFP-III, LNFP-V, LNFP-VI, LNDFH-I, LNDFH-II, LNDFH-III, DSLNT, pLNH, pLNnH, LNH, LNnH, (D)F-LNH-I, (D)F-LNH-II, (D)F-LNH-III, F-para-LNH-I, DF-para-LNH, DF-para-LNnH, TF-LNH, S-LNFP-I (FLSTb), S-LNFP-II (FLSTa), S-LNH, S-LNnH-I, FS-LNH, FS-LNnH-I, DS-F-LNH-II, or a mixture thereof.

[0767] 61. The method according to any one of the foregoing, wherein one or more oligosaccharides produced are harvested from the cell culture.

[0768] 62. The method according to any one of items 1 to 50, wherein the microbially modified cells of the first gene modification according to item 54 are co-cultured with microbially modified cells of the second gene modification, the microbially modified cells of the second gene modification comprising:

[0769] a) a recombinant nucleic acid sequence encoding a transporter capable of importing oligosaccharides from the microbially modified cells of the first gene modification, and

[0770] b) one or more recombinant nucleic acid sequences encoding a glycosyltransferase selected from α-1,2-fucosyltransferase, α-1,3-fucosyltransferase, α-1,3 / 4-fucosyltransferase, and α-2,3-sialyltransferase, and

[0771] c) optionally, a recombinant nucleic acid sequence encoding an MFS transporter, and

[0772] d) wherein the HMO product produced by the microbially modified cells of the second gene modification is DFL or FSL.

[0773] 63. The method according to item 62, wherein the nucleic acid sequence encoding a transporter capable of importing an intermediate product is a lactose permease capable of importing fucosyllactose, and preferably, the lactose permease is overexpressed.

[0774] 64. The method according to item 62 or 63, wherein the MFS transporter is selected from nec, marc, fred, setA, and bad.

[0775] 65. A method for producing LNT or LNnT, the method comprising co-culturing:

[0776] a) a first genetically modified microbial cell according to item 51, and

[0777] b) a second genetically modified microbial cell according to any one of items 56 to 59, wherein

[0778] i) the nucleic acid sequence encoding a protein or protein complex capable of importing an oligosaccharide produced by the first genetically modified microbial cell is recombinant and encodes a transporter capable of importing LNT-II, the transporter being selected from mutant lacY transporters, MFS transporters (such as Blon_0962), and ABC transporters (such as Blon_2177, 2176, 2175, or Blon_0883-0884-0885-08836, BBPC_1775-1776-1777, and Bbr_0527-0528-0530-0531, or RHOM_04095-04100-04105); and,

[0779] ii) the recombinant nucleic acid sequence encoding a glycosyltransferase encodes a β-1,4-galactosyltransferase or a β-1,3-galactosyltransferase; and

[0780] iii) the recombinant nucleic acid encoding an MFS transporter capable of exporting an oligosaccharide encodes vag, nec, Edic1, or YberC; and

[0781] c) harvesting the LNnT produced in the co-culture, and

[0782] wherein, compared with the LNnT produced by a single cell, the LNT-II produced by the method is significantly less, and / or the pLNnH by-product is extremely few or absent.

[0783] 66. A method for producing LNFP-I, LNFP-II, LNFP-V, LNDFH-I, and / or LNDFH-II, the method comprising co-culturing:

[0784] a) a first genetically modified microbial cell according to item 52, and

[0785] b) A second genetically modified microbial cell according to any one of items 56 to 59, wherein said cell comprises:

[0786] i) a nucleic acid sequence encoding a protein or protein complex capable of importing the oligosaccharide produced by the first genetically modified microbial cell, which is recombinant and encodes a transporter capable of importing LNT; and

[0787] ii) a recombinant nucleic acid sequence encoding a glycosyltransferase, which encodes a fucosyltransferase selected from α-1,2-fucosyltransferase, α-1,3-fucosyltransferase, α-1,3 / 4-fucosyltransferase, α-1,4-fucosyltransferase, and

[0788] c) Harvesting LNFP-I, LNFP-II, LNFP-V, LNDFH-I and / or LNDFH-II produced in the co-culture,

[0789] wherein, compared with LNFP-I, LNFP-II and / or LNFP-V produced by a single cell, the LNT, pLNH2 and LNT-II by-products, and their fucosylated derivatives produced by said method are significantly less.

[0790] 67. A method for producing LNFP-III, LNFP-VI and / or LNDFH-III, said method comprising co-culturing

[0791] a) a first genetically modified microbial cell according to item 53, and

[0792] b) a second genetically modified microbial cell according to any one of items 56 to 59, wherein said cell comprises

[0793] i) a nucleic acid sequence encoding a protein or protein complex capable of importing the oligosaccharide produced by the first genetically modified microbial cell, which is recombinant and encodes a transporter capable of importing LNnT, and

[0794] ii) a recombinant nucleic acid sequence encoding a glycosyltransferase, which encodes α-1,3-fucosyltransferase or α-1,3 / 4-fucosyltransferase, and

[0795] c) Harvesting LNFP-III, LNFP-VI and / or LNDFH-III produced in the co-culture, wherein, compared with LNFP-III, LNFP-VI and / or LNDFH-III produced by a single cell, the LNnT, pLNH2 and LNT-II by-products, and their fucosylated derivatives produced by said method are significantly less.

[0796] 68. A method according to any one of items 1 to 50, wherein one or more oligosaccharides produced by the method are oligosaccharides produced from a donor oligosaccharide and a receptor oligosaccharide produced by first and second genetically modified cells, and the method further comprises the steps of:

[0797] a) making an enzyme having transglycosidase activity available in a culture medium; and

[0798] b) incubating a disaccharide or a first oligosaccharide and the second oligosaccharide produced in co-culture together with the transglycosidase in the culture medium to form a third oligosaccharide in the culture medium.

[0799] 69. A method according to item 68, wherein the oligosaccharide produced is a complex oligosaccharide of at least 4 monosaccharide units, such as at least 5 monosaccharide units.

[0800] 70. A method according to any one of items 1 to 50, wherein one or more oligosaccharides produced by the method are sialylated and / or fucosylated oligosaccharides of at least four monosaccharide units, and wherein at least the first or second genetically modified microbial cell produces a sialylated or fucosylated donor oligosaccharide, while the other cell produces a receptor oligosaccharide, and the method further comprises the steps of:

[0801] a) making an enzyme having transglycosidase activity available in a culture medium, wherein the enzyme having transglycosidase activity is

[0802] i) a fucosyltransferase if the donor oligosaccharide is a fucosylated oligosaccharide; or

[0803] ii) a sialyltransferase if the donor oligosaccharide is a sialylated oligosaccharide; and

[0804] b) incubating the first oligosaccharide and the second oligosaccharide produced in co-culture together with the transglycosidase in the culture medium to form a third oligosaccharide of at least four monosaccharide units in the culture medium.

[0805] 71. A method according to item 68, wherein the transglycosidase is either added to the culture medium during the culture process, or expressed by a recombinant nucleic acid in one of the genetically modified cells, or expressed by a third genetically modified cell in the same culture medium as the first and second genetically modified strains.

[0806] 72. A method according to item 71, wherein the third genetically modified cell grows on one of the carbon sources already present in the culture.

[0807] 73. A method according to item 71, wherein the third genetically modified cell grows on a third carbon source.

[0808] 74. The method according to any one of items 68 to 73, wherein the transglycosidase is selected from α-1,2-fucosyltransferase, α-1,3-fucosyltransferase, α-1,3 / 4-fucosyltransferase, α-2,3-sialyltransferase, α-2,6-sialyltransferase, trans-lacto-N-diosidase, β-N-acetylglucosaminidase, and trans-β-galactosidase.

[0809] 75. The method according to any one of items 68 to 74, wherein the first and second genetically modified microbial cells comprise a lactose importer, such as lactose permease.

[0810] 76. The method according to any one of items 68 to 75, wherein lactose is added as a substrate to the co-culture for the production of the first and second oligosaccharides.

[0811] 77. The method according to item 68 or any one of items 71 to 76, wherein the donor oligosaccharide is selected from LNT, LNFP-I, and LST-a.

[0812] 78. The method according to any one of items 68 to 76, wherein the donor oligosaccharide is selected from 2’FL, 3FL, DFL, LNT, LNnT, FSL, sialyl-LacNAc, sialyl-LNB, 3’SL, and 6’SL.

[0813] 79. The method according to item 77, wherein the donor oligosaccharide is produced by the cells according to item 54 or 55.

[0814] 80. The method according to any one of items 68 to 77, wherein the acceptor oligosaccharide is selected from LNB, LacNAc, Lewis A, Lewis X, 2’FL, 3-FL, 3’SL, LNT, LNnT, LNFP-I, LNFP-II, LNFP-III, LNFP-V, LNFP-VI, LST-a, and LST-c.

[0815] 81. The method according to any one of items 68 to 80, wherein the donor oligosaccharide and the acceptor oligosaccharide are different.

[0816] 82. The method according to any one of items 68 to 81, wherein one or more oligosaccharides produced by incubating sialidase with donor and acceptor molecules have at least three monosaccharide units, such as at least four monosaccharide units, and are selected from Lewis X, Lewis A, 3’SLacNAc, 3’SLNB, 6’SLNB, 6’SlacNAc, GlcNAc(1-3)-3FL, LewisY, Lewis B, sialyl-Lewis X, sialyl-Lewis A, DFL, LNFP-I, LNFP-II, LNFP-III, LNFP-V, LNFP-IV, LNDFH-I, LNDFH-II, LNDFH-III, DF-para-LNnH, FLSTa(S-LNFP-II), FSL, LSTa, FLSTa, LSTc, FLSTc, 6’SLN, FLSTb(S-LNFP-I), LSTb, DSLNT, para-LNH, gal-LNnT, F-p-LNH, S-p-LNH.

[0817] 83. The method according to any one of the preceding claims, wherein one or more oligosaccharides produced are harvested from a cell culture.

[0818] 84. The method according to any one of the preceding items, wherein the first and second genetically modified microbial cells are selected from Escherichia coli, Corynebacterium glutamicum, Lactococcus lactis, Bacillus subtilis, Streptomyces lividans, Pichia pastoris, and Saccharomyces cerevisiae.

[0819] 85. The method according to item 84, wherein the first and second genetically modified microbial cells are selected from the same species.

[0820] 86. The method according to item 84 or 85, wherein the genetically modified microbial cell is Escherichia coli.

[0821] 87. Use of one or more oligosaccharides produced by the method according to claims 1 to 86 in the production of a nutritional composition.

[0822] Examples

[0823] Background strain – MDO

[0824] The strain (genetically engineered cell) constructed in this application is based on Escherichia coli K-12 DH1 (Escherichia coli K-12 DH1), and its genotype is: Fˉ, gyrA96, recA1, relA1, endA1, thi-1, hsdR17, supE44. The Escherichia coli K-12 DH1 strain was further modified to generate the MDO strain with the following modifications: lacZ: deletion of 1.5 kbp, lacA: deletion of 0.5 kbp, nanKETA: deletion of 3.3 kbp, melA: deletion of 0.9 kbp, wcaJ: deletion of 0.5 kbp, mdoH: deletion of 0.5 kbp, and the Plac promoter was inserted upstream of the gmd gene. In all the following examples, the MDO strain was used as the background strain.

[0825] The method of inserting a gene of interest into the Escherichia coli genome is well known to those skilled in the art. Inserting a gene cassette into the Escherichia coli chromosome can be achieved by gene phage technology (see, for example, Herring and Blattner 2004 J. Bacteriol. 186: 2673-81 and Warming et al. 2005 Nucleic Acids Res. 33(4): e36) using specific selectable marker genes and screening methods.

[0826] Example 1: Production of HMO mixture by co-culturing 3'SL and LNT producing strains and co-fermenting the two strains

[0827] This example illustrates that two strains growing on different carbon sources can be co-cultured to produce an HMO mixture.

[0828] Strain

[0829] The above MDO strain was further modified to generate a 3'SL producing strain and an LNT producing strain capable of growing on glucose, and their genotypes are shown in Table 14.

[0830] Table 14: Genotypes of strains used in the examples

[0831]

[0832] 1 D29nst – the gene encoding the α-2,3-sialyltransferase of SEQ ID NO: 91, controlled by the Plac promoter (SEQ ID NO: 92).

[0833] 2DnadC – Deletion of the quinolinate phosphoribosyltransferase of WP_101348535.1. For more details, please refer to WO2017101958.

[0834] 3 pBS-nadC-Plac-neuBCA – A plasmid expressing neuBCA (SEQ ID NO:94) and nadC. For more details, please refer to WO 2017 / 101958.

[0835] 4 lgtA – Three genomic inserts of the gene copy encoding β-1,3-N-acetylglucosaminyltransferase (SEQ ID NO:94), controlled by the PglpF promoter (SEQ ID NO:93).

[0836] 5 galTK – Two genomic inserts of the gene encoding β-1,3-galactosyltransferase (SEQ ID NO:96), controlled by the PglpF promoter (SEQ ID NO:93).

[0837] 6 scrYA, scrBR two operons, encoding the sequences of SEQ ID NO:97 and 98 and SEQ ID NO:99 and 100 respectively, controlled by the PglpF_SD1 promoter (SEQ ID NO:101) and the Pscr promoter (SEQ ID NO:102) respectively.

[0838] 7 YberC – A genomically integrated MFS transporter (SEQ ID NO:103), controlled by the Plac promoter (SEQ ID NO:92).

[0839] 8 lacY – An additional genomically integrated copy of lacY (SEQ ID NO:1), controlled by the PglpF promoter (SEQ ID NO:93).

[0840] 9 ΔptsG – Deletion of the glucose-specific EIICB component of the PTS system to restrict growth on glucose.

[0841] 10 Δglf – Deletion of the native glucose facilitator to restrict its growth on glucose.

[0842] 11 ΔlacI – Deletion of the lac repressor, making the use of IPTG ineffective.

[0843] Fermentation

[0844] Escherichia coli strains were cultured in a mineral medium in a bioreactor, which consisted of 15 g / L glucose, 15 g / L sucrose, lactose monohydrate, (NH4)2HPO4, KH2PO4, MgSO4×7H2O, KOH, NaOH, citric acid, trace element solution, antifoaming agent, and thiamine. The dissolved oxygen level was maintained at 20% by a cascading step of first stirring and then passing air flow (starting from 1000 rpm (maximum 2000 rpm) and 1 VVM (maximum 3 VVM)). The pH value was maintained at 6.8 by titration with 10% NH4OH solution. Cultivation started from 1% (v / v) inoculum of each strain, and these inocula were from precultures grown in a similar medium containing glucose (3'SL strain) or sucrose (LNT strain). After glucose and sucrose were depleted in the batch medium, a feed solution containing 1:1 (w / w) glucose and sucrose (sterilized separately with minerals), MgSO4×7H2O, trace metals, and antifoaming agent was continuously fed at a constant profile to limit the carbon content in the culture. The temperature was initially 34 °C, but decreased linearly to 28 °C in 1 hour after 3 hours of feeding. The growth, metabolic activity, and state of the cells were tracked by online measurement of stirring, ammonium hydroxide base addition, temperature, pH, respiratory quotient, and CO2 evolution rate.

[0845] Results

[0846] The fermentation results are shown in Table 15 and Figure 3 as follows, where the mole percentage (mole%) of each HMO is the percentage of the total HMO (mM) + lactose.

[0847] Table 15: Formation of HMOs in the co-fermentation of 3'SL and LNT strains

[0848]

[0849]

[0850] As can be seen from Table 15, within the first 42 hours of fermentation, the production rate of LNT was slightly faster than that of 3'SL. In the remaining time of fermentation, the yields of the two molecules were close to 1:1, which was also reflected in Figure 3 where the yields of 3'SL and LNT followed very similar curves. This clearly indicates that the growth and yield of each strain can be controlled by controlling the addition of two different carbon sources, and one strain cannot outperform the other.

[0851] This actually provides an option to produce a controlled HMO mixture in a single fermentation, and its potential advantage lies in co-producing different products whose yields alone are not sufficient to support a full batch.

[0852] Example 2: Synthesis of LST-c using a dual-strain mixing process

[0853] In this example, two strains were co-cultured. One strain produces LNnT from lactose (substrate) and sucrose (carbon source), and the other strain produces 6’SL from lactose (substrate) and glucose (carbon source). 6’SL serves as the sialic acid donor substrate, and LNnT serves as the acceptor substrate for subsequent sialic acid transfer reactions. LST-c was generated in the fermentation medium by adding α2,6-sialyltransferase. This process is also known as the dual-strain mixing process.

[0854] Strain

[0855] The MDO strain described in the Background Strain section above was further engineered to generate the 6’SL-producing strain and the LNnT-producing strain, and their genotypes are shown in Table 16.

[0856] Table 16: Genotypes of the strains used in the examples

[0857]

[0858] 1ST 6, Pd2 – Gene encoding the α2,6-sialyltransferase of SEQ ID NO: 104.

[0859] 2 DnadC – Deletion of the quinolinic acid phosphoribosyltransferase of WP_101348535.1. For more details, see WO2017 / 101958.

[0860] 3 pBS-nadC-Plac-neuBCA – Plasmid expressing neuBCA (SEQ ID NO: 94) and nadC. For more details, see WO 2017 / 101958.

[0861] 4 lgtA-PglpF – Two genomic insert copies of the gene encoding β-1,3-N-acetylglucosaminyltransferase (SEQ ID NO: 95), controlled by the PglpF promoter (SEQ ID NO: 93).

[0862] 5 galT – One genomic insert copy of the gene encoding β-1,4-galactosyltransferase (SEQ ID NO: 105), controlled by the PglpF promoter (SEQ ID NO: 93).

[0863] 6The scrYA and scrBR operons, encoding the sequences of SEQ ID NO:97 and 98 and SEQ ID NO:99 and 100 respectively, are controlled by the PglpF_SD1 promoter (SEQ ID NO:101) and the Pscr promoter (SEQ ID NO:102) respectively.

[0864] 7 The vag gene encoding a heterologous major facilitator superfamily (MFS) transporter (SEQ ID NO:106) is controlled by the PglpF promoter (SEQ ID NO:93).

[0865] 8 ΔptsG – Deletion of the glucose-specific EIICB component of the PTS system to restrict growth on glucose

[0866] 9 ΔlacI – Deletion of the lac repressor, rendering the use of IPTG ineffective.

[0867] Fermentation

[0868] The dual-strain mixed process was carried out in a 2L Sartorious B-stat bioreactor starting from 1000 g of mineral medium, which consisted of: 15 g / kg glucose (sterilized separately) and 15 g / kg sucrose (sterilized separately), lactose monohydrate (sterilized separately), in an amount sufficient to produce the required amount of oligosaccharides produced by the cells, but not exceeding the amount that the cells can convert into the required oligosaccharides produced by the cells, for example in the range of 10 - 80 g / kg, (NH4)2HPO4, KH2PO4, MgSO4×7H2O (sterilized separately), KOH, NaOH, citric acid, trace element solution, antifoaming agent and thiamine (filtered sterilized). The dissolved oxygen level was maintained at 20% by a cascade step of first stirring and then passing air at 1 VVM (up to 3 VVM). The pH was maintained at 6.8 by titration with an NH4OH solution. The culture started from a 1% (v / v) inoculum of each preculture grown to an OD 600 of 2.5 - 5, where the 6’SL strain used glucose as the carbon source and the LNnT strain used sucrose as the carbon source.

[0869] After about 14 hours, the glucose and sucrose in the batch medium were exhausted, and then a feeding solution containing glucose and sucrose at a ratio of 1:2 (w / w) (sterilized separately with minerals), MgSO4 x 7H2O, H3PO4, trace metals, and an antifoaming agent was continuously added in a constant control curve to limit the carbon content in the culture. The initial temperature was set at 34 °C and linearly decreased to 28 °C over 1 hour after 3 hours of feeding. The growth and status of the cells were monitored by measuring the wet biomass (weight of the cell pellet / weight of the culture broth after centrifugation at 14,000 g for 3 minutes), the optical density at 600 nm, and the online measurement of the CO2 release rate, agitation, alkali addition, dissolved oxygen, and temperature.

[0870] At 113 hours after the start of fermentation, 3.39 mg / ml of α-2,6-transsialidase from Photobacterium eiognathid JT-SHIZ-119 (PITS-197, SEQ ID NO:88) was added to initiate the sialyltransferase reaction, at which point almost all lactose had been converted to 6’SL and LNnT. The process was monitored by measuring the concentrations of the substrates and products by HPLC. The lactose and oligosaccharide concentrations were determined by HPLC one or more times per day. Samples for HPLC analysis were heat-treated at 90 °C for 20 minutes at the time of collection to stop the enzymatic process.

[0871] Results

[0872] Figure 4 Concentration data showing the process progression are presented as weight % relative to the total weight of the substrates and products.

[0873] It can be seen that until 113 hours, when all lactose was consumed and the transsialidase was added to the culture, the 6’SL and LNnT produced by the LNnT and 6’SL strains were close to an equimolar ratio. After the addition of the transsialidase, the formation of LST-c was initiated by the consumption of 6’SL and LNnT, as evidenced by the decrease in their amounts. The formation of LST-c also produced an equimolar amount of lactose as the leaving group in the sialylation reaction. However, the lactose was taken up and reused by the LNnT and 6’SL strains to form 6’SL and LNnT.

[0874] In addition to Example 1, this example shows the stable formation of two different products obtained by co-culturing two strains on two different carbon sources, which can be used in a further process (sialyltransferase) to form a third product (see Figure 4) It also shows that the large difference in the product yields (gLNnT or 6’SL / g carbon source) of the two strains can be compensated by feeding at a modified ratio to achieve the desired product ratio. In this case, the glucose:sucrose ratio is 1:2 (g / g) to achieve a 6’SL:LNnT ratio of 1:1.

[0875] To illustrate the differences in the dual-strain mixed system, a conventional in vitro enzymatic process was also carried out. Using 6’SL as the sialic acid donor, the sialic acid transfer of LNnT was catalyzed by α-2,6-sialyltransferase PITS-197. A substrate solution was prepared with 116.8 mM LNnT and 116.8 mM 6’SL, and the pH value was 6.87. The sialic acid transfer reaction was initiated by adding 1.57 mg / ml PITS-197 at 25 °C. The reaction progress was monitored by measuring the concentrations of the substrate and product by HPLC. Samples for HPLC determination were collected and treated at 95 °C for 5 minutes to denature the PITS-197 enzyme and thus terminate the reaction. Figure 5 shows the reaction progress curve of the in vitro synthesis of LST-c. From Figure 5 it can be seen that, compared with the dual-strain mixed process that results in a very low lactose concentration (as Figure 4 shown), the in vitro synthesis process produces an equal molar amount of lactose and LST-c. Table 17 compares the final product compositions of the mixed process and the in vitro process for the synthesis of LST-c.

[0876] Table 17: Comparison of the final product compositions of the dual-strain mixed process and the in vitro process for the synthesis of LST-c starting from the acceptor (LNnT) to the donor (6’SL) in an equimolar ratio.

[0877] Process Lactose (wt.%) 6’-SL (wt.%) LNnT (wt.%) LST-c (wt.%) In vitro 11.9 17.3 31.8 38.8 Mixing 5.8 20.0 19.5 54.6

[0878] As can be seen from Table 17, due to the higher conversion rate of LNnT, the production amount of LST-c in the dual-strain mixed process increased significantly. In addition, the lactose content in the final product mixture obtained by the dual-strain mixed process was lower. Therefore, the advantage of the dual-strain mixed process is that it can obtain a higher level of LST-c and reduce the lactose content. In addition, the advantage of this mixed process is that it can replace the expensive starting materials 6’SL and LNnT with cheap lactose, and compared with the in vitro process, it can provide a higher conversion rate of LNnT to the desired product LST-c. Therefore, overall, it constitutes a more economical process.

[0879] Example 3: Co-culturing 2’FL and 3FL strains to produce DFL

[0880] In this example, the 2’FL strain capable of growing on sucrose was co-cultured with the 3FL-producing strain that grows on glucose. It shows how the 2’-fucosyllactose (2’FL) produced by the 2’FL strain is taken up by the 3-fucosyllactose (3FL) strain, where an additional fucose is added at the 3-position of 2’FL to form 2’,3-difucosyllactose (DFL).

[0881] strain

[0882] The MDO strain described in the background strain section above was further engineered to generate the 2’FL-producing strain and the 3FL-producing strain, and their genotypes are shown in Table 18.

[0883] Table 18: Genotypes of the strains used in this example

[0884]

[0885] 1 futC – The gene encoding the α-1,2-fucosyltransferase of SEQ ID NO:108.

[0886] 2 CA = An additional colanic acid gene cluster (gmd-wcaG-wcaH-wcaI-manC-manB, SEQ ID NO:109), controlled by the PglpF promoter and located at a locus different from the native locus.

[0887] 3 nec The gene encoding a heterologous major facilitator superfamily (MFS) transporter (SEQ ID NO:107), controlled by the PglpF promoter (SEQ ID NO:93).

[0888] 6 The scrYA and scrBR operons, encoding the sequences of SEQ ID NO:97 and 98 and SEQ ID NO:99 and 100 respectively, are controlled by the PglpF_SD1 promoter (SEQ ID NO:101) and the Pscr promoter (SEQ ID NO:102) respectively.

[0889] 5 futA – Three independent genomic copies of the gene encoding the α-1,3-fucosyltransferase (SEQ ID NO:90), controlled by the PglpF promoter (SEQ ID NO:93).

[0890] 6Gene of the major facilitator superfamily (MFS) transporter of the MARC encoding heterologous (SEQ ID NO:90).

[0891] 7 Extra - genomic integration copy of lacY–lacY (SEQ ID NO:1), controlled by the PglpF promoter (SEQ ID NO:93).

[0892] 8ΔglpR – Deletion of the glpR repressor.

[0893] 9 ΔptsG – Deletion of the EIICB component specific for glucose in the PTS system to limit growth on glucose.

[0894] Fermentation

[0895] The two - strain fermentation process was carried out in the same manner as in Example 1. The cultivation started with a 1% (v / v) inoculum of each strain from a preculture grown in a similar medium containing glucose (3FL strain) or sucrose (2’FL strain).

[0896] Results

[0897] Figure 6 The progress of the process is shown, expressed as weight % relative to the total weight of the substrate and the product.

[0898] It can be seen that before the complete consumption of lactose, the yield of 2’FL far exceeds that of 3FL, which reflects the different carbon source yields of different strains. At this time, DFL has started to accumulate but is still a minor product. However, once lactose is depleted, the production rate of DFL increases significantly. The final composition is 86 wt% DFL, 8.3 wt% 2’FL, and 5.8% 3FL, indicating that DFL can be produced as the main product by co - culturing 2’FL and 3FL strains.

[0899] This example demonstrates how a strain can produce a precursor molecule and secrete it into the culture medium, which is subsequently taken up by a second strain for product formation, thus forming a product. In addition, this example also shows that the product can be obtained in large excess from its precursor molecule.

[0900] Example 4 Comparative Example - Synthesis of LST - c using a single - strain mixing process

[0901] In addition to the two - strain mixing process described in Example 2, this example also describes a method for synthesizing LST - c using a single - strain mixing process to compare the performance of the two - strain system with a simpler single - strain system.

[0902] In this single-strain mixed process, the acceptor substrate LNnT is produced in situ from lactose by the LNnT strain, which has a genotype similar to the LNnT strain in Table 16 but the ptsG gene is not deleted. Instead of using a second strain for the production of 6'-SL as in Example 2, a purified sialic acid donor substrate 6'-SL is added during the cultivation. The sialic acid transfer reaction is catalyzed by adding α-2,6-sialyltransferase from Photobacterium eiognathid JT-SHIZ-119 (PITS-197, SEQ ID NO:88). The cultivation starts with 700 g of mineral medium as described in the method section above, which contains lactose and 25 g / kg of sucrose. The sucrose in the batch medium is depleted after about 15 hours, and then a feed solution containing sucrose and minerals is continuously added using a control curve that limits the carbon content of the culture. The initial sucrose feed rate starts at 1.43 g / h and linearly increases to 2.93 g / h over 5 hours, and then remains constant. The temperature is initially set at 33 °C and linearly decreases to 30 °C over 1 hour after 3 hours of feeding. Sterile 6'SL is fed separately at a constant rate starting 15 hours after the start of the fed-batch phase (equivalent to about 30 hours after inoculation / start of fermentation) for 24 hours.

[0903] At 69.5 hours after inoculation (start of fermentation), a fermentation broth containing 115 mg / L of α-2,6-sialyltransferase from Photobacterium eiognathid JT-SHIZ-119 (PITS-197, SEQ ID NO:41) is added to initiate the sialic acid transfer reaction, at which time almost all of the lactose has been converted to LNnT and 6'SL has also been added. Additional enzyme solutions are added at 99 hours (366 mg / L), 121 hours (281 mg / L), and 146 hours (542 mg / L) respectively.

[0904] Figure 7 The process progress curve is shown as the mass fraction of substrates and products in the total mass of substrates and products. As shown, before the addition of the enzyme, lactose is almost completely converted to LNnT by the E. coli strain. The increase in the 6'-SL level is due to the continuous addition of the 6'-SL solution until the enzyme is added. Since the addition of the enzyme at 69.5 hours, the results of the enzymatic sialic acid transfer can be seen in the formation of LST-c and the consumption of 6'-SL. In addition, due to the rapid recycling of the lactose by-product released by the sialic acid transfer reaction to form LNnT in vivo, the lactose concentration remains at a stable low level. The performance of the single-strain mixed production of LST-c is compared with that of the dual-strain mixed system and the performance of the in vitro synthesis of LST-c from 6'-SL and LNnT (as described in Example 2, as Figure 4 and Figure 5A comparison was made (as shown). Table 19 shows the comparison of the final product compositions of LST-c synthesized by different processes.

[0905] Table 19: Comparison of the final product compositions of LST-c synthesized by single-strain mixed process, two-strain process, and in vitro process

[0906]

[0907] As shown in Table 19, the single-strain mixed LST-c process and the two-strain mixed process are very similar in terms of LST-c production. This clearly shows that the two-strain process is a very stable process, and growing two different strains in the same culture does not affect their ability to produce the desired product. The advantage of the two-strain process is that there is no need to separately produce and purify the second HMO and then return it to the fermentation process, which makes the two-strain process cheaper and simpler than the single-strain process.

[0908] Example 5: Synthesis of LST-a using the two-strain mixed process

[0909] This example describes a method for synthesizing LST-a using the two-strain mixed process. Two strains were co-cultured. One strain utilized lactose (substrate) and sucrose (carbon source) to produce LNT, and the second strain utilized lactose (substrate) and glucose (carbon source) to produce 3'SL. The produced 3'SL was used as the sialyl donor substrate, and LNT was used as the acceptor substrate. The subsequent trans-sialylation reaction was catalyzed by adding α-2,3-trans-sialidase (TcTS, SEQ ID NO: 113) to the culture medium, thereby generating LST-a.

[0910] The synthesis of LST-a in the two-strain mixed process used the LNT-producing strain (MF2, Table 14, Example 1) and the 3'SL-producing strain (MF1, Table 14, Example 1). In this example, the strains were co-cultured according to the method in Example 2 with the following changes. The starting culture medium was 700 g of medium, supplemented with lactose, 15 g / kg of glucose, and 15 g / kg of sucrose. The co-culture started with an inoculum of 1% (v / v) of each strain, and these strains were grown in a similar medium in pre-culture. The medium for the LNT strain contained sucrose, and the medium for the 3'SL strain contained glucose, both growing to OD 600It was 2.5 - 5. After about 18 hours, glucose and sucrose in the batch medium were exhausted, and then glucose and sucrose mineral feed solutions were continuously supplemented at rates of 1.17 g glucose / hour and 1.17 g sucrose / hour respectively to limit the carbon content in the co - culture. The initial temperature was set at 33°C and linearly decreased to 28°C in 1 hour at the start of the fed - batch phase. 68 hours after inoculation / fermentation started, a fermentation broth of 339 mg / L sterile - filtered α - 2,3 - sialyltransferase (TcTS, SEQ ID NO:113) was added to initiate the sialylation reaction of LNT, and the enzyme solution was added again at 90 hours.

[0911] Figure 8 Shows the production amounts of lactose, LNT, 3’SL, and LST - a (expressed as weight % relative to the total weight of substrates and products). Until 68 hours into the process (i.e., when the enzyme was added), the data showed that the LNT strain and the 3’SL strain converted lactose into LNT and 3’SL respectively, without the formation of LST - a. After adding the enzyme at 68 hours, as the formation of LST - a increased, the contents of LNT and 3’SL decreased. In addition, even though lactose was released as a by - product of the enzymatic reaction, its concentration continued to decrease because lactose was rapidly recycled by the corresponding strains into LNT and 3’SL.

[0912] The process of generating LST - a by mixing the two strains was compared with the conventional in vitro enzymatic process for generating LST - a described herein.

[0913] In vitro comparative experiments were carried out using purified LNT and 3’SL as substrates. A substrate solution composed of 150 mM LNT and 150 mM 3’SL was prepared at pH 6.5 and 25°C. The sialic acid transfer reaction was initiated by adding 0.51 mg / mL α - 2,3 - sialyltransferase (TcTS, SEQ ID NO:60). The reaction progress was monitored by measuring the concentrations of substrates and products by HPLC. Samples for HPLC analysis were collected and immediately heat - treated at 90°C for 5 minutes to terminate the reaction. Figure 9 Shows the process of in vitro sialylation of LNT to synthesize LST - a using 3’SL as the sialic acid donor. Compared with the mixed process that can also achieve complete conversion of 3’SL when starting with a 3’SL level twice that of lactose (mol / mol), the conversion rate of 3’SL in the in vitro process was only 57% when starting with a 3’SL / LNT ratio of 1:1 mol / mol. Table 20 compares the final product compositions of the two - strain mixing process and the in vitro process for synthesizing LST - a.

[0914] Table 20: Comparison of the final product compositions of the two - strain mixing process and the in vitro process for synthesizing LST - a

[0915]

[0916] As can be seen from Table 20, compared with the pure in vitro LST-a process, the dual-strain mixed process for synthesizing LST-a achieved full utilization of lactose and obtained a higher LST-a component (as shown in Table 20). In addition, the amount of 3'SL in the final mixture produced by the dual-strain mixed process was 2.5 times lower, which is advantageous in terms of purification. Separating 3'SL from LST-a is more challenging than separating it from neutral LNT.

[0917] Example 6: Controlled co-fermentation of 3'SL and LNT production strains with different carbon source ratios

[0918] In Example 1, the LNT strain (MP2) that can grow on sucrose but has a reduced ability to grow on glucose was co-cultured with the 3'SL strain (MF1) that can grow on glucose but cannot grow on sucrose. These strains were cultured at a ratio of 1:1 in terms of the carbon source in the batch phase and the feed, as well as the inoculation ratio (i.e., the amount of each strain added at the start of the culture). Since the unit carbon source yields (mole / g) of the two strains are similar, the formation ratio of LNT and 3'SL in the culture medium is 1:1.

[0919] This example investigated whether changes in the carbon source ratio would result in a mixture of LNT and 3'SL that reflects the ratio change.

[0920] Two fermentation examples were carried out using the fermentation procedure described in Example 1, but the changes in the carbon source dosage (batch phase and feeding phase) and inoculation volume are shown in Table 21.

[0921] Table 21: Experimental settings for changing the carbon source and inoculation volume

[0922]

[0923] If the strain growth rates are similar, the inoculation ratio usually matches the carbon source ratio to achieve a similar batch phase length. However, in the sucrose / glucose 66 / 73% experiment, the inoculation ratio of the two strains remained at 50 / 50% as in Example 1 to accommodate the fact that the MF2 strain, despite deleting the ptsG gene, still has limited growth on glucose. If the inoculation ratio of the sucrose strain is higher than that of the glucose strain, it is expected that the sucrose strain will consume a large amount of glucose in the batch phase and cause the strain ratio to deviate from the expected 66 / 33 ratio. Therefore, the glucose strain (MF1) that cannot grow on sucrose obtained a leading advantage by inoculating at a ratio of 50% (instead of 33% glucose).

[0924] Results

[0925] The fermentation results are shown in Tables 22 and 23 below and Figure 10 as shown in A and 10B, where the mole % of each HMO is the percentage of the total HMO (mM) + lactose.

[0926] Table 22: Formation of HMOs during co-fermentation of LNT strain (Suc) and 3'SL strain (Glc) at a Suc / Glc ratio of 66:33% and an LNT / 3'SL inoculation ratio of 50:50%

[0927]

[0928] Table 23: Formation of HMOs during co-fermentation of LNT strain (Suc) and 3'SL strain (Glc) at a Suc / Glc ratio of 25:75% and an LNT / 3'SL inoculation ratio of 25:75%

[0929]

[0930] As can be seen from Table 22, the ratio of LNT / 3'SL at the end of fermentation was close to 61:39, which is very close to the carbon source ratio of 66:33. This clearly shows that by changing the carbon source ratio and adjusting the inoculation ratio to accommodate the still limited growth of the LNT strain on the second carbon source (glucose), the content of LNT and 3'SL in the co-culture can be changed in a predictable manner.

[0931] Table 23 clearly shows that the amount of LNT and 3'SL produced in the co-culture can be reversed by changing the carbon source ratio such that the content of glucose is higher than sucrose during fermentation, thereby producing more 3'SL. In fact, the final ratio was 28:72, which is very close to the carbon source ratio of 25:75% used in the fermentation.

[0932] Combined with Example 1, this basically shows that in a single fermentation, a controlled mixture of HMOs can be produced from two strains. This provides production flexibility in terms of strain design and fermentation scale. For example, it is possible to produce the desired product mixture without expressing multiple glycosyltransferases in a single strain.

[0933] Example 7 - Regeneration and Activity of Lyophilized Lactobacillus

[0934] Probiotics can be consumed in the form of live bacteria or dried products (such as freeze-dried). Regardless of the drying method, rehydration is an important step in the recovery process of dehydrated bacteria; an inadequate rehydration / regeneration step may lead to a decrease in cell viability and low final survival rates. Therefore, rehydration is a crucial step in the reconstitution process of freeze-dried cultures. For live and rehydrated bacteria, their survival ability under acidic conditions is crucial because they need to pass through the acidic environment of the stomach and may also face storage (shelf life) in acidic foods.

[0935] This example tested whether an HMO mixture similar to the HMO mixtures produced by the mixing processes described in Examples 2 and 5 could benefit the rehydration (regeneration) and viability of probiotics. The test was conducted under acidic conditions to simulate the survival conditions of bacteria when passing through the stomach or when added to acidic beverages.

[0936] The freeze-dried probiotic Lactobacillus rhamnosus DSM 32550 (0.4 mg / ml), used alone (control) or in combination with the HMO mixture (5% w / v) shown in Table 24, was dissolved in sterile phosphate buffer (PBS, pH = 3), warmed to 37°C, and mixed vigorously for approximately 30 seconds until no visible clumps remained. The tubes were incubated at 37°C for 3 hours. The samples were further diluted, 100 μl was taken, and plated in duplicate on MRS agar plates, which were then placed in an anaerobic incubator and cultured at 37°C. After 72 hours of culture, the regeneration ability and activity of the probiotics were determined by counting the number of colonies on the plates. The experimental setup is shown in Figure 11 .

[0937] Table 24: HMO Compositions Tested in This Example

[0938] HMO Control (A) B C D LST-a 0 55 65 0 LNT 0 45 25 0 3'SL 0 0 10 0 LNnT 0 0 0 25 LST-c 0 0 0 50 6'SL 0 0 0 25

[0939] CFU / ml was calculated based on the number of colonies counted (average of two plates) after 72 hours of culture. The E-2 dilution plates were used to count Mix B and Mix C, and the results are shown in Figure 12 A (containing LST-a); the E-4 dilution plates were used to count Mix D, and the results are shown in Figure 12 B. Figure 13 This is a plate picture of Lactobacillus rhamnosus DSM 32550 colonies, which was taken after 6 days of culture to obtain larger colonies.

[0940] Compared with the control group without the HMO mixture (survival rate was 0), the freeze-dried Lactobacillus rhamnosus DSM 32550 dissolved in the HMO mixture described herein exhibited enhanced regenerative ability and viability. These data clearly show that in low pH conditions (such as in the stomach or acidic beverages), the presence of any HMO mixture can improve the regenerative ability and vitality of Lactobacillus rhamnosus DSM 32550. In addition, it can also be seen that the combination of a higher dose of LST-a and a small amount of 3'SL is more effective than the mixture containing only LST-a and LNT. Mixture D (LST-c, LNnT, and 6'SL) is more effective than mixtures B and C in terms of the regenerative ability and [1] viability of Lactobacillus rhamnosus DSM 32550.

[0941] To our knowledge, there has been no evidence indicating that the tested mixtures can improve the regeneration and viability of probiotics in acidic environments.

[0942] Example 8 - Regeneration and Viability of Freeze-Dried Bifidobacterium

[0943] Similar to Example 7 above, we also tested the ability of the HMO mixtures in Table 24 to promote the regeneration and viability of Bifidobacterium longum DSM 32946 in acidic environments.

[0944] The freeze-dried probiotic Bifidobacterium longum DSM 32946 (0.4 mg / ml) was used alone or in combination with the HMO mixtures shown in Table 24 (5% w / v), dissolved in sterile water at pH 3.0, heated to 37°C, and mixed vigorously for about 30 seconds until no visible clumps remained. The tubes were incubated at 37°C for 30 minutes. Then, 100 μl was taken and plated in duplicate on MRS cysteine agar plates and incubated at 37°C in an anaerobic incubator for 48 hours. The regeneration and viability of the probiotics were determined by counting the number of colonies on the plates after 48 hours of incubation.

[0945] Figure 14 A plate picture of the colonies of Bifidobacterium longum DSM 32946 after 2 days of incubation is shown. CFU / ml was calculated based on the average number of colonies (average of two plates) counted on the undiluted plates after 48 hours of culture. The results are shown in Table 25.

[0946] Table 25: Average CFU / ml of Bifidobacterium longum DSM 32946 after 30 minutes of acid treatment and then cultured at 37°C for 48 hours

[0947]

[0948] As can be seen from Table 25, these mixtures enabled some Bifidobacterium longum subsp. longum DSM 32946 strains to survive under acid treatment, while the survival rate of the control group was 0.

[0949] Example 9: Synthesis of para-LNH using a dual-strain mixed process and trans-lacto-N-biosidase

[0950] This example illustrates that two strains (LNT-S2 and LNnT-S1) growing on different carbon sources (sorbitol and sucrose) can be co-cultured to produce an HMO mixture in the extracellular medium and in situ convert it into larger oligosaccharides (pLNH) through an enzymatic reaction.

[0951] In this example, LNT serves as the LNB donor substrate and LNnT as the acceptor substrate for the in situ trans-lacto-N-biosidase reaction occurring in the fermentation broth. Specifically, trans-lacto-N-biosidase transfers the non-reducing end of LNT (LNB, gal-β1,3-glcNAc) to the non-reducing end of LNnT, forming hexaoigosaccharides, para-LNH, and lactose (the leaving group). Lactose is taken up by the LNT and LNnT strains and processed into more donor and acceptor substrates.

[0952] Strain

[0953] The MDO strains described in the background strain section above were further engineered to generate LNT-producing strains and LNnT-producing strains, and their genotypes are shown in Table 26.

[0954] Table 26: Genotypes of the strains used in the examples

[0955]

[0956] 1 galT – A genomically inserted gene encoding β-1,4-galactosyltransferase (SEQ ID NO: 105), controlled by the PglpF promoter (SEQ ID NO: 93).

[0957] 2 Edic1 – A genomically integrated MFS transporter (SEQ ID NO: 124), controlled by the PglpF promoter (SEQ ID NO: 93).

[0958] 3 ΔlacI – Deletion of the lac repressor, rendering the use of IPTG ineffective.

[0959] 4lgtA – A genomic insertion of a gene copy encoding β-1,3-N-acetylglucosaminyltransferase (SEQ ID NO:94), under the control of the PglpF promoter (SEQ ID NO:93).

[0960] 5 galTK – Two genomic insertions of genes encoding β-1,3-galactosyltransferase (SEQ ID NO:96), under the control of the PglpF promoter (SEQ ID NO:93).

[0961] 6 scrYA, scrBR two operons, encoding sequences of SEQ ID NO:97 and 98 and SEQ ID NO:99 and 100 respectively, under the control of the PglpF_SD1 promoter (SEQ ID NO:101) and the Pscr promoter (SEQ ID NO:102) respectively.

[0962] 7 nec A gene encoding a heterologous major facilitator superfamily (MFS) transporter (SEQ ID NO:107), under the control of the PglpF promoter (SEQ ID NO:93).

[0963] 8 lacY – An additional genomic integration copy of lacY (SEQ ID NO:1), under the control of the PglpF promoter (SEQ ID NO:93).

[0964] 9 ΔsrlA – Deletion of the sorbitol PTS complex IICB slr – Restricts growth on sorbitol.

[0965] 10 ΔglpR - Deletion of the glpR repressor

[0966] Fermentation

[0967] The dual-strain mixing process was carried out in a 250 mL fermenter (Ambr 250 bioreactor system, Sartorius), starting from 100 mL of mineral medium, which contained 15 g / kg sorbitol (sterilized separately), 15 g / kg sucrose (sterilized separately), 40 g / kg lactose monohydrate (sterilized separately), (NH4)2HPO4, KH2PO4, MgSO4×7H2O (sterilized separately), KOH, NaOH, citric acid, trace element solution, antifoaming agent and thiamine (filter sterilized). The dissolved oxygen level was maintained at 20% by first starting stirring at 300 rpm (up to 3000 rpm) and then in a cascading step with an air flow of 1 VVM (up to 3 VVM). The pH was maintained at 6.8 by titrating with NH4OH solution. The pre-cultures were grown in shake flasks in a similar medium with glucose (LNT-S2, sorbitol strain) or sucrose (LNnT-S1, sucrose strain) respectively. The total inoculum at the start of the culture in the bioreactor was 2% (v / v). Since the growth rate of the LNnT-S1 strain on sucrose was significantly faster than that of the LNT-S2 strain on sorbitol, the inoculation ratio of the LNT strain to the LNnT strain was 20:1 to try to achieve the same batch length.

[0968] An increase in pH value (>6.85) triggered the start of the feeding stage, indicating the depletion of sorbitol and sucrose in the batch medium. The feeding solution contained sorbitol and sucrose in a 1:1 (w / w) ratio (sterilized separately with the minerals), MgSO4×7H2O, H3PO4, trace metals and antifoaming agent, and was fed continuously with a constant control curve to limit the carbon content in the culture broth. The temperature was initially set at 34 °C and decreased linearly to 28 °C over 1 hour after 3 hours of feeding. The growth, metabolic activities and status of the cells were tracked by on-line measurement of stirring, ammonium hydroxide base addition, temperature, reflectance, pH value, respiratory quotient and CO2 evolution rate.

[0969] At 48 hours and 96 hours of fermentation, 0.2 mL of the stock solution of lacto-N-diosidase (LNbX, SEQ ID NO:122) from Bifidobacterium longum at 100 mg / mL was added to initiate the lacto-N-diosidase reaction. The lactose and oligosaccharide concentrations in the samples collected during the fermentation were determined by HPLC. The samples for HPLC analysis were heated at 90 °C for 20 minutes at the time of collection to terminate the enzymatic reaction.

[0970] Results

[0971] Table 27 shows the concentration data illustrating the process, expressed as weight % relative to the total weight of the substrates and products involved in the trans-galacto-N-digosidase reaction occurring in the culture medium. In the HPLC method used for analyzing the samples, LNT and LNnT could not be distinguished because they eluted at the same time point.

[0972] Table 27: Weight % relative to the total weight of substrates, products, and by-products measured during fermentation.

[0973]

[0974]

[0975] As can be seen from Table 27, it is possible to produce pLNH as a product during the co-fermentation of the two strains.

[0976] The disaccharide LNB is likely produced by the hydrolytic side activity of trans-galacto-N-digosidase, which is likely able to cleave LNT into lactose and LNB.

[0977] As described in the fermentation section above, the growth rates of the LNT-S2 and LNnT-S1 strains differed significantly, so a seeding ratio of 20:1 was used. Despite the different seeding rates, the sorbitol strain producing LNT consumed all the sorbitol very slowly, so the yield of LNT was likely low during the first 50 - 60 hours of fermentation. Therefore, the amount of LNT + LNnT converted at the end of fermentation was lower than expected (84% remained in the fermentation broth), but a large amount of pLNH was still produced, indicating the possibility of producing pLNH through a combined fermentation and enzymatic process. To our knowledge, this is the first time pLNH has been produced by a fermentation process without initial fermentation and purification of the donor and acceptor substrates.

[0978] The concentrations of the substrates and products shown in Table 27 can be adjusted by optimizing the process, for example, by adjusting the inoculation ratio of the two strains or creating a head start for the LNT-S2 strain, and it may also be possible to adjust the sucrose-sorbitol ratio in the fermentation so that the yields of the two strains respectively reach sufficient amounts of LNT while maintaining an excess of LNnT relative to LNT (preferably a ratio of LNT:LNnT of 1:5).

[0979] Example 10: Synthesis of sialyl-Lewis X using a co-strain process and trans-sialidase.

[0980] This example demonstrates that two strains (S-LacNAc and Lewis X) grown on different carbon sources (sorbitol and sucrose, respectively) can be co-cultured with LacNAc, using LacNAc as a substrate to produce a mixture of non-HMO oligosaccharides in the extracellular medium. These oligosaccharides react with trans-sialidase to catalytically generate larger sialylated oligosaccharides in the fermentation medium (in situ).

[0981] In this example, sialyl-LacNAc serves as the sialic acid donor substrate and Lewis X as the acceptor substrate for an in situ trans-sialylation reaction in the fermentation broth. Specifically, trans-sialidase transfers the sialic acid group on the α-2,3 linkage to the galactose moiety of the Lewis X oligosaccharide, generating sialyl-Lewis X. After sialyl-Lewis X is formed, LacNAc is released (the leaving group) and recycled by the S-LacNAc and Lewis X strains for processing into more donor and acceptor substrates.

[0982] Fermentation using the S-LacNAc and Lewis X strains (genotypes shown in Table 28) was the same as described in Example 9 (inoculation ratio of Lewis X:S-LacNAc was 20:1), except that LacNAc was used instead of lactose as the initial substrate, and the enzyme LNbX was replaced with α-2,3-trans-sialidase (TcTS, SEQ ID NO:113), but the dosage was the same.

[0983] Strain

[0984] The MDO strain described in the background strain section above was further engineered to generate the S-LacNAc-producing strain and the LewisX-producing strain, with genotypes as shown in Table 28.

[0985] Table 28: Genotypes of strains used in the examples

[0986]

[0987] 1 poral – one genomic copy of the gene encoding α-2,3-sialyltransferase (SEQ ID NO:125), under the control of the PglpF promoter (SEQ ID NO:93).

[0988] 2 The scrYA and scrBR operons, encoding the sequences of SEQ ID NO:97 and 88 and SEQ ID NO:99 and 100, respectively, under the control of the Pscr promoter (SEQ ID NO:102).

[0989] 3The nec gene encodes a heterologous major facilitator superfamily (MFS) transporter (SEQ ID NO: 107) and is controlled by the PglpF promoter (SEQ ID NO: 93).

[0990] 4 neuA encodes a genomic copy of the Campylobacter jejuni CMP-Neu5Ac synthase (GenBank AAK91728.1) gene

[0991] 5 neuB encodes a genomic copy of the Campylobacter jejuni Neu5Ac synthase (GenBank AAK91726.1) gene

[0992] 6 neuC encodes a genomic copy of the Campylobacter jejuni UDP-GlcNAc 2-epimerase (GenBank AAK91727.1) gene

[0993] 7 ΔsrlA - Deletes the sorbitol PTS complex IICB slr – Restricts growth on sorbitol

[0994] 8 CA = Additional klebsanate gene cluster (gmd - wcaG - wcaH - wcaI - manC - manB, SEQ ID NO: 109), controlled by the PglpF promoter and located at a locus different from the native locus.

[0995] 9 futA – Three independent genomic copies of the gene encoding α - 1,3 - fucosyltransferase (SEQ ID NO: 90), controlled by the PglpF promoter (SEQ ID NO: 93).

[0996] 10 marc encodes a gene for a heterologous major facilitator superfamily (MFS) transporter (SEQ ID NO: 90).

[0997] 11 ΔlacI – Deletes the lac repressor, rendering the use of IPTG ineffective.

[0998] Results

[0999] Table 29 shows the concentration data illustrating the process progression, expressed as weight % relative to the total weight of the substrate and product in the sialidase reaction occurring in the culture medium.

[1000] Table 29: Weight % relative to the total weight of the substrate and product measured during fermentation.

[1001]

[1002] As can be seen from Table 29, the 20:1 inoculation ratio of the two strains was insufficient to offset the difference in growth rates on sorbitol and sucrose, resulting in an imbalance between the donor and the acceptor, and the yield of sialyl-LacNAc (sucrose strain) was much higher than that of Lewis X (sorbitol strain). The sucrose-growing sialyl-LacNAc strain completed batch growth after 24 hours and triggered the fed-batch phase, while the sorbitol-growing Lewis X strain first consumed all the accumulated sorbitol after 85 hours and then became carbon-limited. Due to the limited amount of Lewis X produced within the first 85 hours, the enzymatic reaction in the culture medium only started after 85 hours. Once the Lewis X strain began to produce Lewis X, sialyl-Lewis X rapidly accumulated from the enzymatic reaction. Since the accumulated LacNAc was very small compared to the sialyl-Lewis X produced, this indicated that LacNAc was recycled to produce more sialyl-LacNAc and Lewis X.

[1003] If more sialyl-Lewis X is required, the process can be optimized, for example, by further adjusting the inoculation ratio of the two strains, or creating a head start for the Lewis X strain, or by adjusting the ratio of sucrose to sorbitol during fermentation so that the two strains produce equal amounts of Lewis X and sialyl-LacNAc respectively. In addition, the process can be designed to completely convert the charged sialyl-LacNAc by stopping the sucrose feed used by the S-LacNAc strain before the end of the whole process. Therefore, the purification of charged sialyl-Lewis X is easier because the only remaining product will be neutral Lewis X.

[1004] Example 11: Production of sialyl-LacNAc using a dual-strain system with intermediate disaccharide uptake ability

[1005] In this example, the LacNAc strain (strain 1) capable of growing on sorbitol was co-cultured with the strain expressing α-2,3-sialyltransferase (2,3-ST strain / strain 2) growing on sucrose. This process corresponded to the process shown in Figure 1 , except that the first strain directly produced the precursor disaccharide (LacNAc) from the first carbon source (sorbitol) without taking up any additional initial substrate (such as lactose), as shown in Figure 1 . Then, the second strain took up LacNAc (the precursor disaccharide), and this strain expressed a sialyltransferase that modified the 3-position of the galactose moiety in LacNAc to generate the non-HMO oligosaccharide sialyl-LacNAc (3’S-LacNAc).

[1006] Co-fermentation of LacNAc and the 2,3-ST strain (genotype shown in Table 30) was carried out according to the method in Example 9 (inoculation ratio of LacNAc:2,3-ST was 20:1), except that no initial substrate (lactose) was added during the fermentation process, nor was any enzyme added.

[1007] Strain

[1008] The MDO strain described in the background strain section above was further engineered to generate a LacNAc-producing strain and a strain expressing α-2,3-sialyltransferase (2,3-ST strain), and its genotype is shown in Table 30. LU et al. further described in Biotechnology Notes Vol. 3, pp. 15 - 25, 2022 how to efficiently engineer a strain capable of overproducing GlcNAc, which is required for the LacNAc-producing strain.

[1009] Table 30: Genotypes of strains used in the examples

[1010]

[1011] 1 poral – One genomic copy of the gene encoding α-2,3-sialyltransferase (SEQ ID NO: 125), controlled by the PglpF promoter (SEQ ID NO: 93).

[1012] 2 The scrYA and scrBR operons, encoding the sequences of SEQ ID NO: 97 and 88 and SEQ ID NO: 99 and 100 respectively, are controlled by the Pscr promoter (SEQ ID NO: 102).

[1013] 3 nec – A gene encoding a heterologous major facilitator superfamily (MFS) transporter (SEQ ID NO: 107), controlled by the PglpF promoter (SEQ ID NO: 93).

[1014] 4 neuA – One genomic copy of the gene encoding Campylobacter jejuni CMP-Neu5Ac synthase (GenBank AAK91728.1)

[1015] 5 neuB – One genomic copy of the gene encoding Campylobacter jejuni Neu5Ac synthase (GenBank AAK91726.1)

[1016] 6One genomic copy of the neuC gene encoding Campylobacter jejuni UDP-GlcNAc 2-epimerase (GenBank AAK91727.1)

[1017] 7 ΔsrlA - Deletion of the IIBC component of the sorbitol PTS complex slr – Restricts growth on sorbitol

[1018] 8 GNA1 – One genomic copy of the gene encoding glucosamine-phosphate N-acetyltransferase 1 (GenBank NP_116637.1), under the control of the PglpF promoter (SEQ ID NO:93).

[1019] 9 GlnA - One genomic copy of the gene encoding glutamine synthetase 1 (GenBank WP_001271717.1), under the control of the PglpF promoter (SEQ ID NO:93).

[1020] 10 galT - One genomic insertion of the gene encoding β-1,4-galactosyltransferase (SEQ ID NO:105), under the control of the PglpF promoter (SEQ ID NO:93).

[1021] 11 yqaB - One genomic copy of the gene encoding fructose-1-phosphate phosphatase (GenBank WP_000273290.1), under the control of the PglpF promoter (SEQ ID NO:93).

[1022] 12 glmS * 54 - A glucosamine synthase variant with three mutations (A38T, R249C, G...

Claims

1. A method for producing one or more oligosaccharides having at least three monosaccharide units, the method comprising the step of co-culturing first and second genetically modified microbial cells in a culture medium, wherein, a) the first genetically modified microbial cell is capable of producing a disaccharide or a first oligosaccharide of at least three monosaccharide units, and wherein the genetically modified cell i) is capable of growing on a first carbon source and is growth-limited or non-growing on a second carbon source, and ii) comprises one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase; and iii) comprises at least one pathway for producing activated sugar nucleotides from the first carbon source; and b) the second genetically modified microbial cell is capable of producing a second oligosaccharide of at least three monosaccharide units, and wherein the genetically modified cell i) is capable of growing on the second carbon source and is growth-limited or non-growing on the first carbon source; and ii) comprises one or more recombinant nucleic acid sequences encoding at least one glycosyltransferase; and iii) comprises a biosynthetic pathway for producing activated sugar nucleotides from the second carbon source.

2. The method according to claim 1, wherein the first carbon source is selected from glucose, glycerol, sucrose, fructose, sorbitol, arabinose, galactose, and maltose, and the second carbon source is selected from sucrose, glycerol, fructose, sorbitol, arabinose, galactose, maltose, and glucose, and wherein the first carbon source and the second carbon source are different.

3. The method according to claim 1 or 2, wherein the genetically modified microbial cell capable of growing on sucrose comprises one or more nucleic acid sequences encoding a PTS-dependent sucrose utilization system, or a nucleic acid encoding an invertase or a sucrose hydrolase that enables the cell to assimilate sucrose.

4. The method according to any one of claim 3, wherein the PTS-dependent sucrose utilization system is encoded by scrY, scrA, scrB, and optionally scrR, or by the cscABKR gene cluster, and the invertase is encoded by SacC_Agal or Bff.

5. The method according to claim 3 or 4, wherein the ability of the cell to grow on glucose and / or glycerol is reduced or lost.

6. The method according to claim 1 or 2, wherein the genetically modified microbial cell capable of growing on glucose comprises one or more nucleic acids encoding one or more glucose transport systems.

7. The method according to claim 6, wherein the ability of the cell to grow on sucrose and / or glycerol is reduced or lost.

8. The method according to any one of the preceding claims, wherein one or more of the recombinant nucleic acids in the first and second genetically modified microbial cells independently encode at least one glycosyltransferase selected from the group consisting of β-1,3-N-acetylglucosaminyltransferase, β-1,3-galactosyltransferase, β-1,4-galactosyltransferase, α-1,2-fucosyltransferase, α-1,3-fucosyltransferase, α-1,3 / 4-fucosyltransferase, α-1,4-fucosyltransferase, α-2,3-sialyltransferase, and α-2,6-sialyltransferase.

9. The method according to any one of the preceding claims, wherein the first and second genetically modified microbial cells are capable of independently producing one or more disaccharides or oligosaccharides selected from the group consisting of LNB, LacNAc, 2’FL, 3FL, 2’FLacNAc, 2’FLNB, Lewis A, Lewis X, 3’SL, 6’SL, 3’SLacNAc, 3’SLNB, sialyl-Lewis A, sialyl-Lewis X, DFL, LNT-II, LNT, LNnT, LNFP-I, LNFP-III, LNFP-IV, LNFP-V, LNFP-VI, FSL, LST-a, LST-b, LST-c, LST-d, LNDFH-II, and LNDFH-III, DSLNT, pLNH, pLNnH, LNH, LNnH, (D)F-LNH-I, (D)F-LNH-II, (D)F-LNH-III, F-para-LNH-I, DF-para-LNH, DF-para-LNnH, TF-LNH, FLST-b, FLST-a, FLST-c, S-LNH, S-LNnH-I, FS-LNH, FS-LNnH-I, DS-F-LNH-II.

10. The method according to any one of the preceding claims, wherein one or more of the oligosaccharides produced by the method are one or more human milk oligosaccharides, such as a mixture of at least two human milk oligosaccharides (HMO).

11. The method according to any one of the preceding claims, wherein at least the first genetically modified microbial cell comprises a recombinant nucleic acid sequence encoding a transporter capable of exporting the disaccharide or first oligosaccharide product into the extracellular culture medium.

12. The method according to any one of the preceding claims, wherein the first genetically modified microbial cell is capable of producing LacNAc, LNB, Lewis A, Lewis X, 2’FL, 3FL, LNT-II, LNT, LNnT, LNFP-I, LST-c, or LST-a as the most abundant disaccharide or oligosaccharide.

13. The method according to any one of the preceding claims, wherein the second genetically modified microbial cell comprises at least one nucleic acid sequence encoding a protein or protein complex capable of importing a disaccharide or oligosaccharide produced by the first genetically modified microbial cell.

14. The method according to claim 13, wherein the protein or protein complex capable of importing a disaccharide or oligosaccharide produced by the first genetically modified microbial cell is selected from Table 1 or Table 2.

15. The method according to any one of the preceding claims, wherein the second genetically modified microbial cell comprises a) one or more recombinant nucleic acid sequences encoding a glycosyltransferase selected from β-1,3-galactosyltransferase, β-1,4-galactosyltransferase, α-1,2-fucosyltransferase, α-1,3-fucosyltransferase, α-1,3 / 4-fucosyltransferase, α-1,4-fucosyltransferase, α-2,3-sialyltransferase and α-2,6-sialyltransferase, and b) optionally, a non-functional or deleted lactose permease, and c) optionally, a recombinant nucleic acid sequence encoding an MFS transporter capable of exporting an oligosaccharide produced by the second genetically modified microbial cell.

16. The method according to any one of claims 13 to 15, wherein one or more oligosaccharides produced by the second genetically modified microbial cell have at least three monosaccharide units, such as at least four monosaccharide units, and are selected from Lewis A, Lewis X, sialyl-LacNAc, sialyl-LNB, sialyl-Lewis X, sialyl-Lewis A, Lewis B, Lewis Y, DFL, FSL, LNT, LNnT, LST-a, LST-b, LST-c, LST-d, LNFP-I, LNFP-II, LNFP-III, LNFP-V, LNFP-VI, LNDFH-I, LNDFH-II, LNDFH-III, DSLNT, pLNH, pLNnH, LNH, LNnH, (D)F-LNH-I, (D)F-LNH-II, (D)F-LNH-III, F-para-LNH-I, DF-para-LNH, DF-para-LNnH, TF-LNH, S-LNFP-I (FLSTb), S-LNFP-II (FLSTa), S-LNH, S-LNnH-I, FS-LNH, FS-LNnH-I, DS-F-LNH-II, or a mixture thereof.

17. The method according to any one of claims 1 to 12, wherein one or more oligosaccharides produced by the method are oligosaccharides produced from a donor oligosaccharide and a receptor disaccharide or receptor oligosaccharide, the donor oligosaccharide and the receptor disaccharide or receptor oligosaccharide being produced by the first and second genetically modified cells, and the method further comprises the following steps: a) making an enzyme with transglycosidase activity available in a culture medium; and b) incubating the first oligosaccharide or disaccharide and the second oligosaccharide produced in co-culture together with the transglycosidase in the culture medium to form a third oligosaccharide of at least four monosaccharide units in the culture medium.

18. The method according to claim 17, wherein the transglycosidase is either added to the culture medium during the culture process, or expressed by recombinant nucleic acid in the first or second genetically modified cell, or expressed by recombinant nucleic acid in a third genetically modified cell available in the culture medium.

19. The method according to any one of claims 17 or 18, wherein the first and second genetically modified microbial cells comprise a lactose importer, such as lactose permease.

20. The method according to claim 17 or 18, wherein the transglycosidase is selected from α-1,2-fucosyltransferase, α-1,3-fucosyltransferase, α-1,3 / 4-fucosyltransferase, α-2,3-sialyltransferase, α-2,6-sialyltransferase, trans-lacto-N-diosidase, β-N-acetylglucosaminidase, and trans-β-galactosidase.

21. The method according to any one of claims 17 to 19, wherein the donor oligosaccharide is selected from Lewis A, Lewis X, 2’FL, 3FL, DFL, sialyl-LacNAc, sialyl-LNB, FSL, 3’SL, 6’SL, LNT, LNnT, LNFP-I, and LST-a.

22. The method according to any one of claims 17 to 21, wherein the acceptor disaccharide or acceptor oligosaccharide is selected from LacNAc, LNB, Lewis A, Lewis X, 2’FL, 3FL, LNT-II, LNT, LNnT, Para-LNnH, LNFP-I, LNFP-II, LNFP-III, LNFP-IV, LNFP-V, LNFP-VI, 3’SL, 6’SL, LST-a, and LST-c.

23. The method according to any one of claims 17 to 22, wherein the one or more oligosaccharides produced by incubating sialidase with donor and acceptor molecules have at least three monosaccharide units, such as at least four monosaccharide units, and are selected from DFL, LNFP-I, LNDFH-I, Lewis-Y, Lewis-B, LNFP-II, LNFP-V, DFL, LNFP-III, LNFP-IV, LNDFH-I, LNDFH-II, LNDFH-III, LNDFH-III, LNDFH-II, Lewis-X, Lewis-A, DF-para-LNnH, FLSTa (S-LNFP-II), FSL, LSTa, FLSTa, 3’SLacNAc, 3’SLNB, LSTc, FLSTc, 6’SLN, FLSTb (S-LNFP-I), LSTb, DSLNT, 6’SLNB, 6’SLacNAc, para-LNH, gal-LNnT, LNFP-V, F-p-LNH, S-p-LNH, GlcNAc(1-3)-3FL, sialyl-Lewis X, sialyl-Lewis A.

24. The method according to any one of the preceding claims, wherein the one or more oligosaccharides produced are harvested from a cell culture.

25. The method according to any one of the preceding claims, wherein the first and second genetically modified microbial cells are selected from Escherichia coli, Corynebacterium glutamicum, Lactococcus lactis, Bacillus subtilis, Streptomyces lividans, Pichia pastoris and Saccharomyces cerevisiae.

26. The method according to claim 25, wherein the first and second genetically modified microbial cells are selected from the same species.

27. The method according to claim 25 or 26, wherein the genetically modified microbial cell is Escherichia coli.

28. Use of one or more oligosaccharides produced by the method according to claims 1 to 27 in the production of a nutritional composition.

Images