Alpha-1,3-fucosyltransferase polypeptides for the production of lacto-n-fucopentaose iii

Abstract

Disclosed are alpha-1, 3-fucosyltransferase polypeptides comprising defined amino acid motifs for the use in the production of lacto-N-fucopentaose III (LNFP-III), nucleic acid molecules comprising at least one nucleotide sequence which encodes one of the variant alpha-1, 3-fucosyltransferase polypeptides, genetically engineered microbial cells containing at least one of the alpha-1, 3-fucosyltransferase polypeptides and / or one of the nucleotide sequences encoding one of the alpha- 1, 3-fucosyltransferase polypeptides, the use of the al pha-1, 3-fucosyltransferase polypeptides and / or the microbial cells for the production of LNFP-III and a method for producing LNFP-III.

Description

[0001] Alpha-1, 3-fucosyltransferase polypeptides for the production of lacto-Af- fucopentaose III

[0002] The present invention relates to an alpha-1 , 3-fucosyltransferase polypeptide for the production of lacto- / V-fucopentaose III (LNFP-III), a nucleic acid molecule comprising a nucleotide sequence that encodes such polypeptide, a genetically engineered microbial cell comprising such polypeptide and a method of producing LNFP-III.

[0003] Background

[0004] Approximately 200 structurally distinct human milk oligosaccharides (HMOs) have been identified so far. Said HMOs are based on the disaccharide lactose and bear additional monosaccharide residues which are based on N-acetyl-glucosamine, fucose, sialic acid, and galactose. The concentration and composition of HMOs in human milk varies between individuals and during the lactation period from up to 20 g / L in the colostrum to 5-10 g / L in the mature milk.

[0005] Human milk oligosaccharides are not digested during their transit through the intestine of infants. Due to their persistence in the infant’s gut, they exhibit beneficial effects to the children. More specifically, HMOs have been shown to be prebiotic as they serve as carbon source for commensal microorganisms of the genera Bifidobacterium, Bacteroides and Lactobacillus. Therefore, HMOs support proliferation of these microorganisms in infants’ guts.

[0006] Milk of women belonging to the so-called “secretor phenotype” contains a high content of a-1 ,2-fucosylated HMOs. These women express the FUT2 gene encoding the so-called “fucosyltransferase 2”. The most abundant HMOs in their milk are 2’-fucosyllactose (2’-FL; Fuc(a1-2)Gal(pi-4)Glc) and Lacto- / V- fucopentaose-l (LNPF-I; Fuc(a1-2) Gal(pi-3)GlcNAc(pi-3)Gal(pi-4)Glc).

[0007] HMOs include several structurally distinct lactofucopentaoses. The best characterized lactofucopentaoses are LNFP-I, lacto- / V-fucopentaose II (LNPF-II; Gal(pi ,3)(Fuc(a1 ,4))GlcNAc(pi ,3)Gal(pi ,4)Glc), lacto- / V-fucopentaose III (LNFP-III; Gal(pi ,4)(Fuc(a1 ,3))GlcNAc(pi ,3)Gal(pi ,4)Glc), and lacto- / V-fucopentaose V (LNFP-V;

[0008] Gal(pi ,3)GlcNAc( i ,3)Gal(pi ,4)(Fuc(a1 ,3))Glc). LNFP-III is known to be an immunomodulatory agent and it is therefore of interest to develop an economical and efficient production method. LNFP-III can be synthesized in an enzymatic reaction from lacto-ZV-neotetraose (LNnT; Gal(pi ,4)GlcNAc(pi ,3)Gal(pi ,4)Glc). A suitable fucosyltransferase catalyzes the addition of a fucose residue via a-1 ,3 linkage to the / V-acetylglucosamine moiety to synthesize LNFP-III. However, to produce LNFP-III in a genetically modified cell, a fucosyltransferase is required that has a high substrate specificity for LNnT and that specifically catalyzes the conversion of LNnT to LNFP-III.

[0009] A fermentation process that gives access to LNFP-III in high purity requires a fucosyltransferase that uses specifically LNnT as a substrate. In an in-cell synthesis starting from lactose, several enzymes and reaction steps are required (an exemplary enzymatic synthesis of LNFP-III is shown in Fig. 1). Starting from lactose, the first step in the synthesis of LNFP-III is the transfer of / V-acetylglucosamine (GIcNAc) to lactose via a p-1 ,3 linkage to yield lacto- / V-triose (LNT2). This reaction is catalyzed by an / V-acetylglucosaminyltransferase, such as for example LgtA. A galactose is added via a p-1 ,4 glycosidic bond to the terminal N-acetylglucosamine residue of LNT2 to yield LNnT. This reaction is catalyzed by a p-1 , 4- galactosyltransferase such as Lex1 or LgtB. In the last step of the synthesis, fucose is transferred to the 3-position of the GIcNAc unit by an a-1 ,3-fucosyltransferase (1 ,3-FucT).

[0010] The a-1 ,3-fucosyltransferase, can however also use substrates different from LNnT, namely lactose which will react to 3-fucosyllactose (3-FL). Another side product observed in the synthesis of LNFP-III is lacto- / V-neo-difucohexaose II (LNnDFHII), which is particularly problematic in the whole-cell biosynthesis of LNFP-III as it is toxic in higher concentrations. Therefore, the described undesired side reactions have to be minimized. Thus, an a- 1 ,3- fucosyltransferase is desired with high specificity for LNnT.

[0011] The production of LNnT in metabolically engineered cells with heterologous N- acetylglucosaminyltransferase LgtA and heterologous 1 ,4-galactosyltransferase LgtB has been described before. For the synthesis of LNFP-III, guanosinediphospate-L-fucose (GDP-Fuc) is required as a donor substrate. GDP- fucose can either be provided endogenously or by the addition of fucose and a fucokinase, such as the Bacteroides fragilis fucokinase-1P-guanyltransferase (Fkp). The use of two alpha-1 , 3-fucosyltransferases FutA and FutB from Helicobacter pylori has been described to transfer fucose from GDP-Fuc to the 3-position of GIcNAc within LNnT (Dumon et al. in Biotechnol. Prog. 2004, 20, 412-419; Dumon et al. in Glycoconjugate Journal 18, 465-474, 2001). However, with FutA no LNFP-III formation was observed, and with FutB no more than 20% of LNFP-III was obtained. Other products formed were LNnDFH-ll, LNFP-VI, 3-FL and LNnT. Thus, these enzymes turned out not to be suited for the synthesis of LNFP-III in high purity.

[0012] Huang Yu et al. (ACS Catalysis 2019, 9,11794-11800) have characterized an 1 ,3 / 4 fucosyltransferase of Bacteroides fragilis. They used purified enzyme to incubate LNnT in vitro with a 1.2 or 2.2-fold excess of GDP-Fuc. With a 1.2-fold excess of GDP-Fuc, LNFP-III was the main product (52% LNFP-III; 10% LNFP-VI, 18% LNnDFH-ll). In human milk, the ratio of LNnDFH-ll / LNFP-III is approximately 1 to 10 (see Ren et al. in Nutrients 2023, 15, 1408), which is significantly lower than achieved here. In order to mimic the natural concentration in a final product, it is desirable to achieve a lower ratio of LNnDFH-ll I LNFP-III. Furthermore, a high concentration of LNnDFH-ll can be toxic for bacteria used in fermentation as it is a large free carbohydrate molecule that is naturally not produced by the bacteria.

[0013] Bai et al. (Carbohydrate Research 2019, 480, 1-6) used a 1,3- fucosyltransferase from Helicobacter pylori for in vitro fucosylation of lactose (Lac), N- acetyllactosamine (LacNAc) and LNnT. With low GDP-Fuc excess, mainly LNFP-III was built - and not LNFP-VI; whereas with a higher excess of GDP-Fuc, LNnDFH-ll was accumulated. These findings were limited to in vitro experiments only.

[0014] Sugita et al. (Journal of Biotechnology 2023, 361 , 110-118) describe the production of LNFP-III in E. coli using a 1,3-fucosyltransferase from Parabacteroides goldsteinii. This fucosyltransferase reacts preferably with a GIcNAc moiety and not with lactose and therefore does not produce 3-FL from lactose. Further, two variants each of fucosyltransferases of Parabacteroides goldsteinii and Bacteroides fragilis were generated and their reactivity with the GIcNAc moiety of the LacNAc motif and with lactose was investigated. They found that substitution of amino acid residue Asn17 in the fucosyltransferase of Parabacteroides goldsteinii or of the corresponding residue Asp28 in the fucosyltransferase of Bacteroides fragilis alters the reactivity to the respective substrate. Mutation of amino acid residue Ser93 in the fucosyltransferase of Parabacteroides goldsteinii and amino acid residue Leu104 in the fucosyltransferase of Bacteroides fragilis resulted in severely compromised enzyme activity. The use of the fucosyltransferase of Parabacteroides goldsteinii in vivo lead to a comparably higher level of LNFP-III production, no 3-FL formation and a comparably lower amount of further byproducts.

[0015] However, there is the need for more efficient and specific fucosyltransferases for the selective synthesis of LNFP-III.

[0016] Summary

[0017] The present disclosure provides alpha-1, 3-fucosyltransferase polypeptides comprising a hydrophobic amino acid residue selected from Tryptophan and Methionine in a first amino acid motif for the use in the production of lacto-A / - fucopentaose III (LNFP-III), nucleic acid molecules comprising at least one nucleotide sequence which encodes one of the alpha-1 , 3-fucosyltransferase polypeptides, genetically engineered microbial cells containing at least one of the alpha-1, 3-fucosyltransferase polypeptides and I or at least one of the nucleotide sequences encoding one of the alpha-1 , 3-fucosyltransferase polypeptides, the use of the alpha-1 , 3-fucosyltransferase polypeptides and I or the microbial cells for the production of LNFP-III and a method for producing LNFP-III wherein the method utilizes one of the alpha-1 , 3-fucosyltransferase polypeptides or a microbial host cell containing at least one of the alpha-1, 3-fucosyltransferase polypeptides.

[0018] In a first aspect, provided are alpha-1 , 3-fucosyltransferase polypeptides, wherein the amino acid sequences of the alpha-1 , 3-fucosyltransferase polypeptides comprise a. a first amino acid motif of [R / K / N]xx[R / W]xPzx (SEQ ID NO: 12), b. in a distance of 5 to 300 amino acid residues in C-terminal direction of the first amino acid motif a second amino acid motif of FCx and c. in a distance of 1 to 10 amino acid residues in C-terminal direction of the second amino acid motif, a third amino acid motif of [S / T][D / N], and d. in a distance of 5 to 200 amino acid residues in C-terminal direction of the third amino acid motif, a fourth amino acid motif of ENx, and e. in a distance of 5 to 20 amino acid residues in C-terminal direction of the fourth amino acid motif, a fifth amino acid motif of [S / T]EKx, and f. in a distance of 2 to 20 amino acid residues in C-terminal direction of the fifth amino acid motif, a sixth amino acid motif of PxYxG.

[0019] The abbreviation x stands for any naturally occurring alpha-amino acid. The abbreviation z stands for a hydrophobic amino acid selected from Tryptophan or Methionine. The letters RKNWPFCSTDEYGM are abbreviations according to the commonly known 1 -letter amino acid code as set out below.

[0020] In a second aspect, provided are nucleic acid molecules comprising a nucleotide sequence that encodes one of the alpha-1 , 3-fucosyltransferase polypeptides which comprises a Tryptophan residue or a Methionine residue in the first amino acid motif at amino acid residue z.

[0021] In a third aspect, provided are genetically engineered microbial cells comprising one of the alpha-1 , 3-fucosyltransferase polypeptides which comprises a Tryptophan residue or a Methionine residue in the first amino acid motif at amino acid residue z.

[0022] In a fourth aspect, provided is the use of one of the alpha-1, 3-fucosyltransferase polypeptides which comprises a Tryptophan residue or a Methionine residue in the first amino acid motif at amino acid residue z or the use of genetically engineered microbial cells comprising one of the alpha-1, 3-fucosyltransferase polypeptides which comprises a Tryptophan residue or a Methionine residue in the first amino acid motif at amino acid residue z.

[0023] In a fifth aspect, provided is a method for producing LNFP-III, the method comprising culturing a population of genetically engineered microbial cells comprising one of the alpha-1, 3-fucosyltransferase polypeptides which comprises a Tryptophan residue or a Methionine residue in the first amino acid motif at amino acid residue z in a culture medium and under conditions that are suitable for the microbial cells to synthesize LNFP-III. Brief description of the drawings

[0024] Fig. 1 shows an exemplary synthesis scheme of LNFP-III (4) when starting from lactose (1), via LNT2 (2) and LNnT (3). Also, the possible side reaction leading to the formation of 3-FL (5) is shown. A 1 ,3-fucosyltransferase (A) catalyzes the transfer of fucose from the donor substrate GDP-Fuc (6) to the respective acceptor substrate. An N-acetylglucosaminyltransferase (B), such as LgtA, catalyzes the transfer of GIcNAc from the donor substrate UDP-GIcNAc (7) to the acceptor substrate lactose (1) to yield LNT2 (2). A 1 ,4-galactosyltransferase (C), such as Lex1 , catalyzes the transfer of galactose from the donor substrate UDP-Gal (8) to the acceptor substrate LNT2 (2).

[0025] Fig. 2a and Fig. 2b show an alphafold2 model of the Bacteroides fragilis alpha-1 ,3- fucosyltransferase with GDP-Fuc (6) and lactose (1) docked in the lactose binding site. Fig. 2b shows an enlarged view of the part surrounded by the dashed line in FIG. 2a. The amino acid residues E75, W28 and L104 are indicated. The arrow in the binding site indicates the distance between the 3-hydroxy group of Glucose and the C1 of Fucose.

[0026] Fig. 3a to Fig. 3c show an alignment of sequences of a selection of nine alpha-1 ,3- fucosyltransferases with first (1.) to seventh (7.) amino acid motifs indicated below the alignments, wherein the amino acid residue z in the first amino acid motif is not Tryptophan or Methionine. Here, amino acid sequences of different alpha-1 , 3- fucosyltransferases are shown that are suitable as starting points for optimizing selectivity by introducing Tryptophan or Methionine into the first amino acid motif. Identical amino acid residues are highlighted by using a black background.

[0027] Fig. 4a and Fig. 4b display the nucleotide sequence of the Bacteroides fragilis alpha-1 , 3-fucosyltransferase BfFucT-encoding gene bfFucT (Fig. 4a, SEQ ID NO:

[0028] 10), and the deduced amino acid sequence of the Bacteroides fragilis alpha-1 , 3- fucosyltransferase BfFucT (Fig. 4b, SEQ ID NO: 1).

[0029] Fig. 5a and Fig. 5b display the nucleotide sequence of the Bacteroides gallinaceum alpha-1 , 3-fucosyltransferase BgFucT-encoding gene bgFucT (Fig. 5a, SEQ ID NO:

[0030] 11), and the deduced amino acid sequence of the Bacteroides gallinaceum alpha- 1 , 3-fucosyltransferase BgFucT (Fig. 5b, SEQ ID NO: 2). Detailed description

[0031] According to the first aspect, provided are alpha-1 , 3-fucosyltransferase polypeptides for the use in the production of LNFP-III. An alpha-1, 3-fucosyltransferase polypeptide according to the invention comprises the listed amino acid motifs: a. a first amino acid motif of [R / K / N]xx[R / W]xPzx (SEQ ID NO: 12), b. a second amino acid motif in a distance of 5 to 300 amino acid residues in C-terminal direction of the first amino acid motif, the second amino acid motif being FCx, c. a third amino acid motif of [S / T][D / N] in a distance of 1 to 10 amino acid residues in C-terminal direction of the second amino acid motif, d. a fourth amino acid motif of ENx in a distance of 5 to 200 amino acid residues in C-terminal direction of the third amino acid motif, e. a fifth amino acid motif of [S / T]EKx in a distance of 5 to 20 amino acid residues in C-terminal direction of the fourth amino acid motif, and f. in a distance of 2 to 20 amino acid residues in C-terminal direction of the fifth amino acid motif, the sixth amino acid motif of PxYxG.

[0032] The capital letters in the amino acid motives stand for the one-letter abbreviations of amino acids according to IIIPAC.

[0033] The letter x is a place holder for any of the 20 naturally occuring canonical alphaamino acids and z stands for a hydrophobic amino acid selected from Tryptophan or Methionine. These amino acids possess hydrophobic bulky side chains.

[0034] The inventors have surprisingly found that when taking an alpha-1-3- fucosyltransferase as starting point (starting alpha-1 , 3-fucosyltransferase) that does not comprise a Methionine or Tryptophan residue in the position of amino acid residue z, and they substitute the amino acid residue z by Methionine or Tryptophan, they can obtain an alpha-1 , 3-fucosyltransferase polypeptide that is more suitable for the use in the production of LNFP-III than the starting alpha-1 ,3- fucosyltransferase. The specificity of the alpha-1, 3-fucosyltransferase polypeptide according to the invention for the subtrate Lacto-ZV-neotetraose (LNnT) is higher than the specificity of the starting alpha-1, 3-fucosyltransferase for LNnT. Therefore, less undesired side products such as 3-fucosyllactose (3-FL) and lacto-ZV-neo- difucohexaose II (LNnDFH-ll) are produced and the production of LNFP-III is improved.

[0035] The abbreviations for the 20 naturally occuring canonical alpha-amino acids according to the IIIPAC code including three-letter abbreviations and one-letter abbreviations are given here:

[0036] Alanine (Ala, A), Cysteine (Cys, C), Aspartic Acid (Asp, D), Glutamic Acid (Glu, E), Phenylalanine (Phe, F), Glycine (Gly, G), Histidine (His, H), Isoleucine (lie, I), Lysine (Lys, K), Leucine (Leu, L), Methionine (Met, M), Asparagine (Asn, N), Proline (Pro, P), Glutamine (Gin, Q), Arginine (Arg, R); Serine (Ser, S), Threonine (Thr, T), Valine (Vai, V), Tryptophan (Trp, W), Tyrosine (Tyr, Y).

[0037] Amino acids are grouped according to the properties of their side chains. Nonpolar or hydrophobic amino acids have hydrophobic side chains (Ala, Gly, lie, Leu, Met, Trp, Phe, Pro, Vai). Polar or hydrophilic amino acids have hydrophilic side chains (Cys, Ser, Thr, Tyr, Asn, Gin). Amino acids charged at neutral pH value are grouped into positively charged or polar basic amino acids (His, Lys, Arg) and negatively charged or polar acidic amino acids (Asp, Glu).

[0038] Ambiguities in the sequences of the amino acid motifs are indicated by listing the acceptable amino acids for a given position between square parentheses [ ] and separated by I . For example, [S / T]EKx stands for Serine or Threonine, followed by Glutamic Acid, followed by Lysine followed by any of the 20 natural alpha-amino acids.

[0039] An alpha-1, 3-fucosyltransferase polypeptide catalyzes the introduction of a fucosyl (Fuc) residue from the donor substrate guanosine diphosphate fucose (GDP-Fuc) in an alpha-1, 3-linkage to a suitable accepting residue within an oligosaccharide. In the synthesis of LNFP-III, the fucosyl residue is transferred to the N-acetylglucosamine within the acceptor substrate LNnT. If the alpha-1 , 3-fucosyltransferase polypeptide accepts acceptor substrates different from LNnT, undesired side products are formed. One exemplary undesired acceptor subtrate is lactose, which yields 3- fucosyllactose when the reaction with GDP-Fuc is catalyzed by the alpha-1 , 3- fucosyltransferase polypeptide.

[0040] LNFP-III and LNFP-3 are used interchangeably herein for lacto-ZV-fucopentaose-lll (Gal(beta1-4)[Fuc(alpha1-3)]GlcNAc(beta1-3)Gal(beta1-4)Glc).

[0041] LNT-II and LNT2 are used interchangeably herein for lacto-N-triose II (GlcNAc(pi ,3)Gal(pi ,4)Glc).

[0042] LNnT is used for lacto-N-neotetraose (Gal(pi,4)GlcNAc(pi,3)Gal(pi,4)Glc). LNnDFH-ll and LNnDFH-2 are used interchangeably herein for lacto-N-neo- difucohexaose II (Gal(beta1 ,4)[Fuc(alpha1- 3)]GlcNAc(beta1 ,3)Gal(beta1 ,4)[Fuc(1 , 3]]Glc).

[0043] The monosaccharides are abbreviated as follows: Galactose (Gal), Fucose (Fuc), N- Acetylglucosamine (GIcNAc) and Glucose (Glc).

[0044] In another embodiment, the alpha-1 , 3-fucosyltransferase polypeptide according to the invention has a first amino acid motif, which is RxxRxPzx (SEQ ID NO: 13).

[0045] In another embodiment, the alpha-1 , 3-fucosyltransferase polypeptide comprises a seventh amino acid motif of GENx in a distance of 5 to 50 amino acid residues of the first amino acid motif in N-terminal direction. Further, the distance between the first amino acid motif and the second amino acid motif is between 5 and 100 amino acid residues and the distance between the third amino acid motif and the fourth amino acid motif is between 5 and 100 amino acid residues. Further, the distance between the fourth and fifth amino acid motif is 2 to 10 amino acid residues and the fifth amino acid motif is GYx[S / T]EKx (SEQ ID NO: 14). In C-terminal direction after the sixth amino acid motif there are 50 to 300 amino acid residues.

[0046] In another embodiment, the alpha-1 , 3-fucosyltransferase polypeptide exhibits a higher substrate specificity for LNnT over lactose as compared to the substrate specificty of the starting alpa-1 , 3-fucosyltransferase. Thus, in reactions catalyzed by the alpha-1 , 3-fucosyltransferase polypeptide according to the invention, more LNFP- III and less 3-fucosyllactose is formed than in reactions catalyzed by the starting alpha-1, 3-fucosyltransferase under the same conditions. The substrate specificity of the alpha-1 , 3-fucosyltransferase polypeptides can be tested by monitoring the formation of the expected products and undesired byproducts (3-FL; LNFP3; LNnDFH-2) in reactions catalyzed by the alpha-1 , 3- fucosyltransferase polypeptide and a starting alpha-1, 3-fucosyltransferase that does not have the amino acid residues Methionin or Tryptophan at the position of amino acid residue z. The substrate specificity can be tested in vitro under the same reaction conditions, such as buffer, concentration of GDP-fucose and enzyme, and temperature. The concentration of the substrates lactose and LNnT should be identical. The formation of expected products can be followed over time to estimate kinetic parameters, or the yields can be determined at a specific point in time and compared. The alpha-1, 3-fucosyltransferase polypeptide can be used in isolated form or as cell extract of cells expressing the respective alpha-1 ,3- fucosyltransferase polypeptide. The design of suitable experiments is well-known in the art.

[0047] The substrate specificity can further be tested in vivo by comparing the yields of the expected product and the undesired by-products in reactions taking place under identical conditions in microbial cells expressing the al pha-1 , 3-fucosyltransferase polypeptide according to the invention or the starting alpha-1 , 3-fucosyltransferase.

[0048] In another embodiment, the alpha-1 , 3-fucosyltransferase polypeptide according to the invention has a sequence identity to the amino acid sequence as set forth in SEQ ID NO: 1 of at least 80%, wherein the amino acid residue z corresponds to the amino acid residue at position 104 of SEQ ID NO: 1 and wherein the amino acid residue z at position 104 is one of Tryptophan or Methionine. The amino acid sequence of the alpha-1, 3-fucosyltransferase of SEQ ID NO: 1 possesses a Leucine (L) residue at position 104 and is based on the amino acid sequence of the naturally occuring alpha-1, 3-fucosyltransferase of Bacteroides fragilis. To get to the alpha- 1 ,3- fucosyltransferase polypeptide according to the invention, the amino acid at position 104 is mutated to one of Tryptophan or Methionin. In a preferred embodiment, the alpha-1 , 3-fucosyltransferase polypeptide according to the invention has a sequence identity to the amino acid sequence as set forth in SEQ ID NO: 1 of at least 85%, of at least 90%, of at least 95%, of at least 98% or of at least The term “sequence identity of [a certain] %” in the context of two or more nucleotide sequences or amino acid sequences refers to a relationship between the sequences of two polypeptides or polynucleotides, as determined by sequence comparison (alignment). As used herein, “sequence identity” is determined across the entire length of a sequence. “Sequence identity” means that the two or more sequences have nucleotides or amino acid residues in common in the given percentage when compared and aligned. Identity measures the percent of identical matches between the smaller of two or more sequences with gap alignments (if any) addressed by a particular mathematical model, algorithms, or computer program.

[0049] Percent sequence identity of nucleotide sequences or amino acid sequences can be readily calculated by any of the methods known to one of ordinary skill in the art. In preferred embodiments, the “percent identity” of two sequences (e.g., polynucleotide or amino acid sequences) is determined using the algorithm of Karlin and Altschul Proc. Natl. Acad. Sci. USA 87:2264-68, 1990, modified as in Karlin and Altschul Proc. Natl. Acad. Sci. USA 90:5873-77, 1993. Such an algorithm is incorporated into the NBLAST® and XBLAST® programs (version 2.0) of Altschul et al., J. Mol. Biol. 215:403-10, 1990. Where gaps exist between two sequences, Gapped BLAST ® can be utilized, for example, as described in Altschul et al., Nucleic Acids Res. 25(17):3389-3402, 1997. When utilizing BLAST and Gapped BLAST® programs, the default parameters of the respective programs (e.g., XBLAST® and NBLAST®) can be used, or the parameters can be adjusted appropriately as would be understood by one of ordinary skill in the art.

[0050] Another local alignment technique which may be used, for example, is based on the Smith-Waterman algorithm (Smith, T. F. & Waterman, M. S. (1981) J. Mol.

[0051] Biol. 147:195-197). A general global alignment technique which may be used, for example, is the Needleman-Wunsch algorithm (Needleman, S. B. & Wunsch, C. D. (1970) J. Mol. Biol. 48:443-453), which is based on dynamic programming.

[0052] Software for performing BLAST analyses is publicly available through the National Center for Biotechnology Information (http: / / www.ncbi.nlm.nih.gov / ).

[0053] For multiple sequence alignments, computer programs including Clustal Omega (Sievers et al., Mol Syst Biol. 2011 Oct. 11 ; 7:539) may be used. In some embodiments, a sequence, including a polynucleotide or amino acid sequence, is found to have a specified percent identity to a reference sequence, such as a sequence disclosed in this application and / or recited in the claims when sequence identity is determined using Clustal Omega (Sievers et al., Mol Syst Biol. 2011 Oct. 11; 7:539).

[0054] Examples of sequence alignment algorithms are CLUSTAL Omega (http: / / www.ebi. ac.uk / Tools / msa / clustalo / ), EMBOSS Needle (http: / / www.ebi.ac.uk / Tools / psa / emboss_needle / ), MAFFT (http: / / mafft.cbrc.jp / alignment / server / ) or MUSCLE (http: / / www.ebi.ac.uk / Tools / msa / muscle / ).

[0055] For purposes of the present invention, 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) as 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 later (available at https: / / www.ebi.ac.uk / Tools / psa / emboss needle / ). The parameters used are gap open penalty of 10, gap extension penalty of 0.5, and the EBLOSUM62 (EMBOSS version of 30 BLOSUM62) substitution matrix. The output of Needle labeled "longest identity" (obtained using the -nobrief option) is used as the percent identity and is calculated as follows: (Identical Residues x 100) / (Length of Alignment - Total Number of Gaps in Alignment).

[0056] In another embodiment, the alpha-1 , 3-fucosyltransferase polypeptide according to the invention has a sequence identity of at least 80% to the amino acid sequence as set forth in SEQ ID NO: 2, wherein the amino acid residue at position z corresponds to the amino acid residue at position 91 of SEQ ID NO: 2 and wherein the amino acid residue at position 91 is selected from Tryptophan or Methionine. The amino acid residue of the alpha-1, 3-fucosyltransferase of SEQ ID NO: 2 possesses a Leucine (L) residue at position 91 and is based on the amino acid sequence of the naturally occuring alpha-1, 3-fucosyltransferase of Bacteroides gallinaceum. To get to the alpha-1 , 3-fucosyltransferase polypeptide according to the invention, the amino acid at position 91 is mutated to one of Tryptophan or Methionin. In a preferred embodiment, the alpha-1 , 3-fucosyltransferase polypeptide has a sequence identity to the amino acid sequence as set forth in SEQ ID NO: 2 of at least 85%, of at least 90%, of at least 95%, of at least 98% or of at least 99%.

[0057] In another embodiment, the alpha-1 , 3-fucosyltransferase polypeptide according to the invention comprises at position z as a hydrophobic amino acid residue Tryptophan (W). This amino acid has a hydrophobic and bulky side chain, which leads to a change of substrate specificity in favor of LNnT versus lactose.

[0058] In some embodiments, the alpha-1, 3-fucosyltransferase polypeptide comprises functional fragments of any one of the alpha-1, 3-fucosyltransferase polypeptides described herein before. Said functional fragments comprise variants of the alpha- 1, 3-fucosyltransferase polypeptides that are truncated by one or more amino acid residues at the N-terminal and / or C-terminal end as compared to any one of the various alpha-1 , 3-fucosyltransferase polypeptides.

[0059] The term “functional” indicates that said variants and fragments are also capable of catalysing the transfer of a fucose residue from a donor substrate to an acceptor molecule, i.e. they can possess fucosyltransferase activity.

[0060] According to the second aspect, provided are nucleic acid molecules comprising a nucleotide sequence that encodes an alpha-1, 3-fucosyltransferase polypeptide as disclosed herein above. The alpha-1, 3-fucosyltransferase polypeptide comprises at least a first amino acid motif of [R / K / N]xx[R / W]xPzx; in a distance of 5 to 300 amino acid residues in C-terminal direction a second amino acid motif of FCx; in a distance of 1 to 10 amino acid residues in C-terminal direction a third amino acid motif of [S / T][D / N]; in a distance of 5 to 200 amino acid residues in C-terminal direction a fourth amino acid motif of ENx; in a distance of 5 to 20 amino acid residues in C- terminal direction a fifth amino acid motif of [S / T]EKx; and in a distance of 2 to 20 amino acid residues in C-terminal direction a sixth amino acid motif of PxYxG.

[0061] The nucleic acid molecule is either a linear nucleic acid molecule or a circular nucleic acid molecule. Exemplary linear nucleic acid molecules are ribonucleic acids, chromosomes, and gene duplicates. Examples of circular nucleic acid molecules are plasmids, cosmids, bacterial artificial chromosomes, and yeast artificial chromosomes; bacterial chromosomes. In some embodiments, the nucleotide sequence encoding the alpha-1 , 3- fucosyltransferase polypeptide according to the invention is operably linked to at least one expression control sequence. The term "operably linked" as used herein, shall mean a functional linkage between a nucleic acid expression control sequence (such as a promoter, signal sequence, or array of transcription factor binding sites) and a second nucleotide sequence such as a nucleotide sequence encoding an alpha-1 , 3-fucosyltransferase polypeptide, wherein the expression control sequence effects transcription and / or translation of the second nucleotide sequence. Accordingly, the term "promoter" designates nucleotide sequences which usually "precede" a protein-coding nucleotide sequence in a polynucleotide and e.g. provide a site for initiation of the transcription into mRNA. "Regulator" sequences, also usually located "upstream" of (i.e. , preceding) a protein-coding nucleotide sequence in a given polynucleotide, bind proteins that determine the frequency (or rate) of transcriptional initiation. Collectively referred to as "promoter / regulator" or "control" nucleotide sequence, these sequences which precede a selected protein-coding nucleotide sequence (or series of protein-coding nucleotide sequences) in a nucleic acid molecule cooperate to determine whether the transcription (and eventual expression) of a protein-coding nucleotide sequence will occur. Nucleotide sequences which "follow" a protein-coding nucleotide sequence in a polynucleotide and provide a signal for termination of the transcription into mRNA are referred to as transcription "terminator" sequences. It is understood that the nucleotide sequence encoding the alpha-1 , 3-fucosyltransferase polypeptide and being operably linked to expression control sequences will be transcribed - and eventually expressed - under permissive conditions.

[0062] In some embodiments, the promoter is an inducible promoter, i.e. a promoter enabling to switch on expression of the second nucleotide sequence by the presence of an inductor such as a small molecule. In alternative embodiments, the promoter is a negatively regulated promoter, i.e. a promoter whose activity is negatively regulated in response to a small molecule.

[0063] Being operably linked to expression control sequences enables expression of the nucleotide sequence encoding the alpha-1, 3-fucosyltransferase polypeptide in vitro or in vivo, and thereby enables synthesis of the alpha-1, 3-fucosyltransferase polypeptide for its use in the synthesis of LNFP-III.

[0064] According to the third aspect, provided are genetically engineered microbial cells comprising at least one alpha-1, 3-fucosyltransferase polypeptide as described herein above and I or comprising a nucleic acid molecule encoding for the alpha- 1, 3-fucosyltransferase polypeptide as described herein above. The alpha-1, 3- fucosyltransferase polypeptide comprises at least a first amino acid motif of [R / K / N]xx[R / W]xPzx; in a distance of 5 to 300 amino acid residues in C-terminal direction a second amino acid motif of FCx; in a distance of 1 to 10 amino acid residues in C-terminal direction a third amino acid motif of [S / T][D / N]; in a distance of 5 to 200 amino acid residues in C-terminal direction a fourth amino acid motif of ENx; in a distance of 5 to 20 amino acid residues in C-terminal direction a fifth amino acid motif of [S / T]EKx; and in a distance of 2 to 20 amino acid residues in C- terminal direction a sixth amino acid motif of PxYxG.

[0065] A “genetically engineered microbial cell" is understood as a microbial cell which has been transformed or transfected or is capable of transformation or transfection by an exogenous polynucleotide. Thus, the nucleotide sequences as used in the invention, may, e.g., be comprised in a vector which is to be stably transformed / transfected or otherwise introduced into host microbial cells. A great variety of expression systems can be used to produce polypeptides. Such vectors include, among others, chromosomal, episomal and virus-derived vectors, e.g., vectors derived from bacterial plasmids, from bacteriophage, from transposons, from yeast episomes, from insertion elements, from yeast chromosomal elements, from viruses, and vectors derived from combinations thereof, such as those derived from plasmid and bacteriophage genetic elements, such as cosmids and phagemids. The expression system constructs may contain control regions that regulate as well as engender expression. Generally, any system or vector suitable to maintain, propagate or express polynucleotides and to synthesize a polypeptide in a microbial cell may be used for expression in this regard. The appropriate polynucleotide may be inserted into the expression system by any of a variety of well-known and routine techniques, such as, for example, those set forth in Sambrook et al., “Molecular Cloning, A laboratory Manual,” 2ndEdition, Cold Spring Harbor Laboratory Press, 1989. Preferably, polynucleotides containing the appropriate nucleotide sequence(s) are stably introduced into the genome of the microbial cell. Genomic integration can be achieved by recombination or transposition.

[0066] The genetically enginieered microbial cell is either a prokaryotic cell or a eukaryotic cell. Suitable microbial cells include bacterial cells, archaebacterial cells, yeast cells, and fungal cells.

[0067] In some embodiments, the genetically engineered microbial cell is a prokaryotic cell, more specifically a bacterial cell. The bacterial cell can be selected from the group consisting of the genera of Bacillus, Lactobacillus, Lactococcus, Enterococcus, Bifidobacterium, Sporolactobacillus spp., Micromonospora spp., Micrococcus spp., Rhodococcus spp., and Pseudomonas. Suitable bacterial species are Bacillus subtilis, Bacillus licheniformis, Bacillus coagulans, Bacillus thermophilus, Bacillus laterosporus, Bacillus megaterium, Bacillus mycoides, Bacillus pumilus, Bacillus lentus, Bacillus cereus, Bacillus circulans, Bifidobacterium longum, Bifidobacterium infantis, Bifidobacterium bifidum, Citrobacter freundii, Clostridium cellulolyticum, Clostridium ljungdahlii, Clostridium autoethanogenum, Clostridium acetobutylicum, Corynebacterium glutamicum, Enterococcus faecium, Enterococcus thermophiles, Escherichia coli, Erwinia herbicola (Pantoea agglomerans), Lactobacillus acidophilus, Lactobacillus salivarius, Lactobacillus plantarum, Lactobacillus helveticus, Lactobacillus delbrueckii, Lactobacillus rhamnosus, Lactobacillus bulgaricus, Lactobacillus crispatus, Lactobacillus gasseri, Lactobacillus casei, Lactobacillus reuteri, Lactobacillus jensenii, Lactococcus lactis, Pantoea citrea, Pecto bacterium carotovorum, Proprionibacterium freudenreichii, Pseudomonas fluorescens, Pseudomonas aeruginosa, Streptococcus thermophiles and Xanthomonas campestris. In another preferred embodiment, the bacterial species is Escherichia coli.

[0068] Bacterial cells can be genetically engineered by well-established moelcular biological methods, have little demands on the nutritional composition of the culture medium, multiply rapidly, and fermentation conditions for in vivo production can be controlled. In other embodiments, the genetically engineered microbial cell is a eukaryotic cell such as a yeast cell. The yeast cell may be selected from the group consisting of Saccharomyces sp., in particular Saccharomyces cerevisiae, Saccharomycopsis sp., Pichia sp. (aka Hanensula sp.), in particular Pichia pastoris, Kluyveromyces sp., Yarrowia sp., Rhodotorula sp., and Schizosaccharomyces sp.

[0069] Although not multiplying as rapidly as bacterial cells and being more demanding with respect to their culture medium, eukaryotic microorganisms can be genetically engineered and fermentation conditions for in vivo production can be controlled.

[0070] In some embodiments, the genetically engineered microbial cell contains a plasmid which comprises the nucleotide sequence encoding at least one of the alpha-1 , 3- fucosyltransferase polypeptides according to the invention. In additional or alternative embodiments, the nucleotide sequence encoding the alpha-1 , 3- fucosyltransferase polypeptide is integrated into the genomic DNA of the microbial cell.

[0071] In certain embodiments, the genetically engineered microbial cell is a microbial cell for the production of an oligosaccharide of interest, wherein the oligosaccharide of interest is LNFP-III. LNFP-III contains an N-acetylglucosamine. Thus, such microbial cell is able to synthesize an N-acetylglucosamine-containing (GIcNAc-containing) oligosaccharide intracellularly. In particular, the microbial cell is able to synthesize the GIcNAc-containing oligosaccharides LNT2 and LNnT, which are intermediate oligosaccharides in the synthesis of LNFP-III.

[0072] The terms “is capable of’ and “is able to” are used synonymously and are to be understood such that the microbial cell synthesizes the GIcNAc-containing oligosaccharide of interest intracellularly when cultured in a medium and under conditions that are permissive for the microbial cell to synthesize the GIcNAc- containing oligosaccharide of interest.

[0073] In additional and / or alternative embodiments, the genetically engineered microbial cell for producing a GIcNAc-containing oligosaccharide of interest further comprises one or more exogenous and / or heterologic nucleotide sequences that each, independently, encode one or more enzymes that is / are necessary for the metabolic pathway to synthesize the GIcNAc-containing oligosaccharide of interest.

[0074] For intracellular biosynthesis of a GIcNAc-containing oligosaccharide of interest, the genetically engineered microbial cell possesses a metabolic pathway for intracellular biosynthesis of UDP-GIcNAc. In some embodiments, the metabolic pathway for intracellular biosynthesis of UDP-GIcNAc comprises the enzymatic activities for the de novo biosynthesis of UDP-GIcNAc from fructose-6-phsophate. This de novo biosynthesis pathway for UDP-GIcNAc comprises the enzymatic activities of a glucosamine-fructose-6-phosphate transaminase, a phosphoglucosamine mutase, and an N-acetylglucosamine-1-phosphate uridyltransferase. Providing a microbial cell that possesses a de novo biosynthesis pathway for UDP-GIcNAc is advantageous, because it is not necessary to culture the microbial cell in the presence of exogenous GIcNAc.

[0075] A glucosamine-fructose-6-phosphate transaminase catalyzes the conversion of D- fructose-6-phosphate and L-glutamine to D-glucosamine-6-phosphate and L- glutamate. An exemplary glucosamine-fructose-6-phosphate transaminase that can be used in the microbial cells disclosed herein is E. coli GlmS (UniProtKB entry number P17169). In certain embodiments, the genetically engineered microbial cell overexpresses and / or expresses an exogenous glucosamine-fructose-6-phosphate transaminase gene, such as the E. coli glmS gene or glucosamine-fructose-6- phosphate transaminase gene from a different species.

[0076] A phosphoglucosamine mutase catalyzes the conversion of glucosamine-6- phosphate to glucosamine-1-phosphate. An exemplary phosphoglucosamine mutase that can be used in the microbial cells disclosed herein is E. coli GlmM (UniProtKB entry number P31120). All references to UniProtKB Entry Nos. refer to UniProtKB (www.uniprot.org) Release 2023_03 released on June 28, 2023. In certain embodiments, the genetically engineered microbial cell overexpresses and / or expresses an exogenous phosphoglucosamine mutase gene, such as the E. coli glmM gene or a phosphoglucosamine mutase gene from a different species.

[0077] An N-acetylglucosamine-1 -phosphate uridyltransferase catalyzes the transfer of an acetyl group from acetyl coenzyme A to glucosamine-1-phosphate to produce N- acetylglucosamine-1-phoshate, and then converts N-acetylglucosamine-1-phoshate to UDP-GIcNAc utilizing UTP. An exemplary N-acetylglucosamine-1 -phosphate uridyltransferase that can be used in the microbial cells disclosed herein is E. coli Gimli (UniProtKB entry number P0ACC7).

[0078] In certain embodiments, the genetically engineered microbial cell overexpresses and / or expresses an exogenous N-acetylglucosamine-1 -phosphate uridyltransferase gene, such as the E. coli glmU gene or an N-acetylglucosamine-1-phosphate uridyltransferase gene from a different species.

[0079] The E. coli N-acetylglucosamine-1 -phosphate uridyltransferase is a bifunctional protein, wherein the C-terminal domain catalyzes the first enzymatic reaction, and the N-terminal domain catalyzes the latter enzymatic reaction.

[0080] In an alternative, the two enzymatic steps for converting GlcNAc-1-P to UDP - GIcNAc can be performed by two distinct polypeptides, a glucosamine-1 -phosphate acetyltransferase for transferring an acetyl group to GlcNAc-1-P, and an UDP-N- acetylglucosamine diphosphorylase to catalyze the formation of UDP-GIcNAc from N-acetylglucosamine-1 -phosphate and UTP. In some embodiments, the microbial cell has been genetically engineered to possess an UDP-N-acetylglucosamine diphosphorylase or to increase the UDP-N-acetylglucosamine diphosphorylase activity within the microbial cell.

[0081] In some embodiments, the genetically engineered microbial cell comprises a beta- 1,3-N-acetylglucosaminyltransferase, which transfers a GIcNAc moiety from UDP- GIcNAc onto a lactose molecule to generate LNT2.

[0082] An exemplary beta-1 , 3-N-acetylglucosaminyltransferase that can be used in the microbial cells disclosed herein is LgtA from Neisseria meningitidis (UniProtKB Entry No. Q9JXQ6).

[0083] In some embodiments, the genetically engineered microbial cell comprises an LNT2-accepting beta-1 , 4-galactosyltransferase for galactosylating LNT-2. Genetically engineering the microbial cell to possess an LNT2-accepting (3-1,4- galactosyltransferase allows to provide a genetically engineered microbial cell and a method for producing LNnT. Another exemplary LNT2-accepting p-1,4-galactosyltransferase that can be used in the microbial cells disclosed herein is a Neisseria meningitidis LgtB (UniProtKB entry number Q51116 and others). Another exemplary LNT2-accepting beta-1 ,4- galactosyltransferase that can be used in the microbial cells disclosed herein is a beta-1, 4-galactosyltransferase from Aggregatibacter aphrophilus Lex1 (NCBI Reference Sequence Database (https: / / www.ncbi.nlm.nih.gov / protein) RefSeq protein record WP_005701792.1 of August 20, 2018).

[0084] In some embodiments, the genetically engineered microbial cell possesses a metabolic pathway for the biosynthesis of UDP-galactose (UDP-Gal). In some embodiments, the genetically engineered microbial cell has been genetically engineered to possess a metabolic pathway for the biosynthesis of UDP-galactose or for enhancing the metabolic pathway for the biosynthesis of UDP-galactose.

[0085] An exemplary metabolic pathway for intracellular biosynthesis of UDP-galactose starts from glucose-6-phosphate. This metabolic pathway comprises the enzymatic activities of a phosphoglucomutase, an UTP-glucose-1 -phosphate uridylyltransferase, and an UDP-glucose 4-epimerase. One or more of these enzymatic activities can be provided by a heterologous enzyme.

[0086] In some embodiments, the genetically engineered microbial cell possesses at least one exogenous nucleotide sequence that encodes one or more of the enzymes that are involved in the intracellular biosynthesis of UDP-galactose.

[0087] The enzyme phosphoglucomutase converts glucose-1-phosphate to glucose-6- phosphate. The phosphoglucomutase is encoded by a phosphoglucomutase gene.

[0088] A suitable example of a phosphoglucomutase gene is the pgm gene of E. coli (strain K12) encoding the E. coli phosphoglucomutase PgM (UniProtKB Entry No. P36938).

[0089] The enzyme UTP — glucose-1 -phosphate uridylyltransferase converts glucose-1 - phosphate to UDP-glucose. The UTP — glucose-1 -phosphate uridylyltransferase is encoded by a UTP — glucose-1 -phosphate uridylyltransferase gene. A suitable example of a UTP — glucose-1 -phosphate uridylyltransferase gene is the galU gene of E. coli (strain K12) encoding the E. coli UTP — glucose-1 -phosphate uridylyltransferase GalU (UniProtKB Entry No. P0AEP3). The enzyme UDP-glucose 4-epimerase converts UDP-glucose to UDP-galactose. The UDP-glucose 4-epimerase is encoded by a UDP-glucose 4-epimerase gene. An example of a suitable UDP-glucose 4-epimerase gene is the galE gene of E. coli (strain K12) encoding the E. coli UDP-glucose 4-epimerase GalE (UniProtKB entry No. P09147).

[0090] Expression of one or more exogenous and / or endogeneous genes encoding one or more of the enzymes necessary for intracellular UDP-galactose biosynthesis implements or enhances intracellular GDP-galactose biosynthesis in the genetically engineered microbial cell and provides the donor substrate for the galactosyltransferases involved in the biosynthesis of LNnT.

[0091] In some embodiments, the genetically engineered microbial cell possesses a metabolic pathway for the biosynthesis of GDP-L-fucose, herein also called GDP- fucose (GDP-Fuc). The metabolic pathway for intracellular GDP-fucose biosynthesis may be a salvage pathway or a de novo pathway.

[0092] The salvage pathway for intracellular biosynthesis of GDP-fucose comprises the enzymatic activities of fucose kinase for phosphorylation of fucose to provide fucose-1-phosphate as well as of a fucose- 1 -phosphate guanylyltransferase to convert fucose- 1 -phosphate to GDP-fucose. Hence, the genetically engineered microbial cell possesses a fucose kinase and a fucose-1-phosphate guanylyltransferase. In certain embodiments, the enzymatic activities of the fucose kinase and the fucose-1-phosphate guanylyltransferase are provided by a bifunctional enzyme that exhibits both enzymatic activities.

[0093] Example of a suitable bifunctional fucose kinase I fucose-1 -phosphate guanylyltransferase is encoded by the fkp gene of Bacteroides fragilis (UniProtKB Entry No. Q58T34). Alternatively, genes encoding a fucose kinase, a fucose-1- phosphate guanylyltransferase and / or a bifunctional fucose kinase / a fucose-1 - phosphate guanylyltransferase can be obtained from the genera Lentisphaera, Ruminococcus, Solibacter, Arabidopsis, Oryza, Physcomitrella, Vitis, Danio, Bos, Equus, Macaca, Pan, Homo, Rattus, Mus and Xenopus. Possessing the salvage pathway for intracellular biosynthesis of GDP-L-fucose enables the genetically engineered microbial cell to utilize intracellular free L-fucose for GDP-L-fucose biosynthesis.

[0094] In some embodiments, the genetically engineered microbial cell possessing the salvage pathway for intracellular biosynthesis of GDP-L-fucose further comprises a fucose permease for internalization of exogenous L-fucose by the microbial cell. A suitable fucose permease is the L-fucose-proton symporter FucP of E. coli (strain K12) (UniProt Entry No. P11551. Also GenBank acc no CP000948).

[0095] In some embodiments, the genetically engineered microbial cell possesses a de novo pathway for the intracellular biosynthesis of GDP-L-fucose. The de novo pathway for intracellular biosynthesis of GDP-L-fucose starts with the isomerisation of fructose-6-phosphate to mannose-6-phosphate by a mannose-6-phosphate isomerase. Mannose-6-phosphate is then converted to mannose-1-phosphate by a phosphomannomutase. Mannose- 1 -phosphate is reacted with GTP by a mannose- 1-phosphate guanylyltransferase to yield GDP-alpha-D-mannose which is further converted by a GDP-mannose-4,6-dehydratase to GDP-4-keto-6-deoxymannose. GDP-4-keto-6-deoxymannose is then converted to GDP-L-fucose by two steps involving an epimerase activity and a reductase activity, which are both provided by a GDP-L-fucose synthase.

[0096] Hence, a microbial cell that possesses the de novo pathway for intracellular biosynthesis of GDP-L-fucose comprises a mannose-6-phosphate isomerase phosphomannomutase, a mannose-1 -phosphate guanylyltransferase, a GDP- mannose-4,6-dehydratase and a GDP-L-fucose synthase.

[0097] A suitable mannose-6-phosphate isomerase is e.g. the mannose-6-phosphate isomerase of E. coli (strain K12) (UniProtKB entry No. P00946), encoded by the E. coli (K12) manA gene.

[0098] A suitable phosphomannomutase is e.g. the phosphomannomutase of E. coli (strain K12) (UniProtKB Entry No. P24175), encoded by the E. coli (K12) manB gene. A suitable mannose-1 -phosphate guanylyltransferase is e.g. the mannose-1- phosphate guanylyltransferase of E. coli (strain K12) (UniProtKB entry No. P24174), encoded by the E. coli (K12) manC gene.

[0099] A suitable GDP-mannose-4,6-dehydratase is e.g. the GDP-mannose-4,6- dehydratase of E. coli (strain K12) (UniProt KB entry No. P0AC88), encoded by the E. coli (K12) gmd gene.

[0100] A suitable GDP-L-fucose synthase is e.g. the E. coli (strain K12) GDP-L-fucose synthase (UniProtKB Entry No. P32055), encoded by the E. coli (K12) i / caG gene.

[0101] Thus, in some embodiments the genetically engineered microbial cell has been genetically engineered to contain and express one or more genes encoding at least one of the enzymes involved in the salvage pathway and / or the de novo pathway for intracellular biosynthesis of GDP-L-fucose.

[0102] In some embodiments, one or more of the genes encoding at least one of the enzymes involved in the salvage pathway and / or the de novo pathway for intracellular biosynthesis of GDP-L-fucose is a heterologous gene.

[0103] In additional and / or alternative embodiments, the genetically engineered microbial cell comprises a lactose permease or a lactose importer for the internalization of exogenous lactose. In certain embodiments, lactose permease or lactose importer is a non-endogenous or heterologous permease or importer. In some embodiments the genetically engineered microbial cell contains a polynucleotide comprising a nucleotide sequence that encodes the non-endogenous or heterologous lactose permease or lactose importer.

[0104] Possessing a lactose permease enables the genetically engineered microbial cell to internalize exogenous lactose and utilize it e.g. as an educt in the intracellular biosynthesis of the oligosaccharide of interest. A suitable lactose permease is E. coli (K12) lactose permease LacY (UniProtKB Entry No. P02920) as encoded by the E. coli K12 lacY gene.

[0105] In an additional and / or alternative embodiment, the genetically engineered microbial cell contains a nucleic acid molecule which comprises and expresses a nucleotide sequence encoding a lactose permease, such as e.g. the E. coli K12 lactose permease LacY or functional fragments or functional variants thereof.

[0106] In some embodiments, the genetically engineered microbial cell synthesizes lactose intracellularly.

[0107] In an additional and / or alternative embodiment, the genetically engineered microbial cell has reduced, deleted or functionally impaired enzymatic activity that hydrolyzes lactose, such as e.g. p-galactosidase activity. For example, in E. coli (strain K12) the P-galactosidase is LacZ (UniProtKB Entry No. P00722) as encoded by the lacZ gene.

[0108] Having deleted or at least reduced the intracellular p-galactosidase activity improves yield of the desired oligosaccharide as the intracellular pool of lactose not diminished by intracellular p-galactosidase activity. Hence, more lactose - regardless of whether internalized lactose or lactose being produced intracellularly by the genetically engineered microbial cell that produces the oligosaccharide of interest - is available for production of the oligosaccharide of interest.

[0109] In additional and / or alternative embodiments, the genetically engineered microbial cell has reduced, deleted or functionally impaired enzymatic activities which acetylates lactose. Deletion or at least reduction of galactoside O-acetyltransferase activity improves yield of the desired oligosaccharide as the intracellular pool of lactose is not diminished. For example, in E. coli (strain K12) the galactoside O- acetyltransferase is LacA (UniProt Entry No. P07464) as encoded by the lacA gene.

[0110] In additional and I or alternative embodiments, the genetically engineered microbial cell has reduced, deleted or functionally impaired enzymatic activities of the enzymes L-fucose isomerase (fuel) and I or L-fuculokinase (fucK). This prevents the bacteria from the utilization of fucose for purposes other than the synthesis of the desired HMO.

[0111] According to the fourth aspect, provided is the use of an alpha-1, 3- fucosyltransferase polypeptide as described herein before for the synthesis of LNFP-III. According to the fifth aspect, provided is a method of producing LNFP-III, the method comprising culturing a population of genetically engineered microbial cells as described herein before. The population of genetically engineered microbial cells are cultured in a culture medium and under conditions that are permissive for the microbial cells to intracellularly synthesize LNFP-III.

[0112] The expression “culture medium that is permissive for the genetically engineered microbial cell to intracellularly synthesize LNFP-III” as used herein refers to a culture medium that is adapted to cell growth, the culture medium comprising nutrients and growth factors essential to growth of the genetically engineered microbial cells and is free of toxic agents that can cause cell death, such as cell growth inhibitors, or wherein toxic agents are present in an amount that is not lethal to said cells. In addition, the culture medium contains factors essential to the intracellular synthesis of the desired oligosaccharide such as, e.g. educts and / or precursors of the oligosaccharide of interest.

[0113] The expression “conditions that are permissive for the genetically engineered microbial cell to intracellularly synthesize LNFP-III are understood to be conditions relating to physical and / or chemical parameters including but not limited to temperature, pH, pressure, osmotic pressure and product / precursor / acceptor concentration.

[0114] In some embodiments, the method comprises culturing the population of genetically engineered microbial cells in the presence of exogenous lactose. Lactose is a disaccharide. Lactose is added to the culture medium if the genetically engineered microbial cell can internalize exogenous lactose and / or is not able to intracellularly synthesize lactose. The internalized lactose can serve as educt for the intracellular biosynthesis of LNFP-III as is shown for example in Fig. 1.

[0115] In some embodiments, the method comprises culturing the population of genetically engineered microbial cells wherein the genetically engineered microbial cells express a beta-1 , 3-N-acetyl hexaminyltransferase, a beta-1 ,4 galactosyltransferase and a pathway to synthesize GDP-fucose, preferably a de novo pathway or a salvage pathway. Preferably the genetically engineered microbial cells express a lactose permease and the genetically engineered microbial cells are cultured in the presence of lactose in the culture medium.

[0116] In additional and / or alternative embodiments, the method comprises culturing the genetically engineered microbial cell in the presence of at least one of glucose, fructose, sucrose and glycerol. Any one of these compounds serves as a carbon source and energy source for the microbial cells. In some embodiments, the genetically engineered microbial cells are cultivated in the presence of a mixture of glucose and fructose. The mixture may be an equimolar mixture of glucose and fructose and may be obtained by hydrolysis of sucrose.

[0117] The method, optionally, further comprises recovering LNFP-III from the culture medium and / or the genetically engineered microbial cells.

[0118] The recovery of LNFP-III comprises at least one of the following steps: clarification, centrifugation, microfiltration, ultrafiltration, nanofiltration, diafiltrations, two-phase partitioning, reverse osmosis, microfiltration, activated charcoal or carbon treatment, treatment with non-ionic surfactants, enzymatic digestion, tangential flow high- performance filtration, tangential flow ultrafiltration, affinity chromatography, ion exchange treatment, hydrophobic interaction chromatography and / or gel filtration, ligand exchange chromatography, electrodialysis, drying such as e.g. rotary evaporation, free-fall evaporation, freeze-drying, spray-drying, drum-drying, rollerdrying, vacuum-drying.

[0119] The present invention will be described with respect to particular embodiments and with reference to drawings, but the invention is not limited thereto but only by the claims. Furthermore, the terms first, second and the like in the description and in the claims, are used for distinguishing between similar elements and not necessarily for describing a sequence, either temporally, spatially, in ranking or in any other manner. It is to be understood that the terms so used are interchangeable under appropriate circumstances and that the embodiments of the invention described herein are capable of operation in other sequences than described or illustrated herein.

[0120] It is to be noticed that the term “comprising”, used in the claims, should not be interpreted as being restricted to the means listed thereafter; it does not exclude other elements or steps. It is thus to be interpreted as specifying the presence of the stated features, integers, steps or components as referred to, but does not preclude the presence or addition of one or more other features, integers, steps or components, or groups thereof. Thus, the scope of the expression “a device comprising means A and B” should not be limited to devices consisting only of components A and B. It means that with respect to the present invention, the only relevant components of the device are A and B.

[0121] Reference throughout this specification to “one embodiment” or “an embodiment” means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, appearances of the phrases “in one embodiment” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment but may. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner, as would be apparent to one of ordinary skill in the art from this disclosure, in one or more embodiments.

[0122] Similarly, it should be appreciated that in the description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.

[0123] Furthermore, while some embodiments described herein include some, but not other features included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the invention, and form different embodiments, as would be understood by those in the art. For example, in the following claims, any of the claimed embodiments can be used in any combination. In the description and drawings provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In other instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

[0124] The invention will now be described by a detailed description of several embodiments of the invention. It is clear that other embodiments of the invention can be configured according to the knowledge of persons skilled in the art without departing from the true spirit or technical teaching of the invention, the invention being limited only by the terms of the appended claims.

[0125] Examples

[0126] Example 1 : Genetic engineering of LNFP-lll-producing E.coli strains bearing alpha-1, 3-fucosyltransferase polypeptides of SEQ ID NO: 1 or SEQ ID NO: 2

[0127] A bacterial production strain producing LNT-II was generated by insertion of the 1,3- N-acetylglucosaminyl-transferase gene IgtA from Neisseria meningitidis into the araBA locus of an E. coli DH1 strain in which the genes wcaJ, lacZ, lacA, nagB, waaB, and mdoH were deleted. In addition, the E. coli lacY gene was expressed heterologously in this strain. The bacterium’s metabolic pathway to generate UDP- galactose was enhanced by chromosomal integration of E. coli genes pgm, galE and galU. Furthermore, the strain possesses a p-1,4-galactosyltransferase gene lex- 1 from Aggregatibacter segnis and Aggregatibacter aphrophilus, which was placed under control of a constitutive promoter and a rrnB terminator. The resulting strain is designated “E. coli LNnT”.

[0128] To prevent E. coli from the utilization of fucose, the neighboring genes of fucose isomerase fuel and L-fuculokinase fucK were knocked out via A-Red recombination using the flanking sequences of the KEIO collection primer set. A dual expression construct with the tetracycline promoter expressed malE.BfFucT and the phage T5 promoter expressed fkp was integrated into the E coli genome via Ez TN5 transposition. BfFucT corresponds to SEQ ID NO: 1 , which is based on the sequence of the fucosyltransferase of Bacteroides fragilis.

[0129] A similar dual expression construct encompasses the tetracycline promoter expressed BgFucT with the phage T5 promoter expressed fkp and was integrated into the genome via Mariner transposition with the Himar transposase. BgFucT corresponds to SEQ ID NO: 2, which is based on the sequence of the fucosyltransferase of Bacteroides gallinaceum.

[0130] Within the genome-integrated constructs, the original protein sequences of BfFucT (SEQ ID NO: 1) and BgFucT (SEQ ID NO: 2) were altered to the BfFucT-L104W and BgFucT-L91W variants via A-Red mediated single-stranded DNA recombineering using a 90-nt DNA oligonucleotide bearing the desired mutation (Table 1). Individual clones were screened with a primer complementary at the 3'- end only to the L104W or L91W variant DNA sequence and a flanking primer to give a 500 bp PCR product if the ssDNA recombination had been incorporated. Each variant candidate was verified via Sanger sequencing of a PCR product amplified over the mutated region.

[0131] Table 1 : MAGE primers used to generate BfFucT-L104W and BgFucT-L91W

[0132] Example 2: Substrate specificity of 1,3-fucosyltransferase polypeptides (in vivo)

[0133] Clonal lines of “E coli LNFP-lll-BfFucT L104W’ and “E. coli LNFP-lll-BgFucT L91W” bearing the desired alpha-1 , 3-fucosyltransferase polypeptides were subjected to a microti terp I ate screening assay for oligosaccharide production. After 24 hours in the main culture containing 10 mM lactose and 40 mM fucose, the supernatants were harvested and a 1 / 20 dilution of each supernatant was used for LC-MS quantification of LNFP-111 , LNDFH-I I , LNnT, 3-FL and LNT2. The results are shown in Table 2 below. concentration produced by E. coli strains which differ in the amino acid sequence of their alpha-1, 3-fucosyltransferase polypeptides.

[0134] When comparing the data for BfFucT and its variant BfFucT L104W, it is evident that the substrate specificity for LNnT over lactose is increased significantly, as the ratio of LNFP-III : 3-FL has been increased from 9 / 36=0,25 to 13 / 3=4,3. Thus, the enzyme is preferring LNnT as a substrate over lactose and therefore has a higher substrate specificity.

[0135] Importantly, the relative formation of undesired LNnDFH-2 has been reduced at the same time. The relevant ratio of LNFP-III : LNnDFH-2 has been increased from 9 / 36=0,25 to 13 / 1=13,0, which is a factor of about 50. Thus, the formation of undesired byproduct is observed significantly less. Thus, the enzyme has been significantly improved to be used in the industrial production of LNFP-III and mixtures containing LNFP-III.

[0136] When comparing the data for BgFucT and its variant BgFucT L91W, it is evident that the formation of LNnDFH2 is reduced to a very low level. The ratio of

[0137] LNFP - III : LNnDFH-2 has been increased from 69 / 7 to 50 / 0, thus the formation of LNnDFH-2 was not detected anymore. Such a reduction of undesired byproduct is very beneficial for the use of the enzyme in the production for LNFP-III and its mixtures that are to be used for human consumption. When using BfFucT, 36% of 3-FL is formed compared to only 2% in the case of BgFucT. Thus, in the case of BgFucT, the substrate specificity for LNFP-III is higher than that of BfFucT. Even though the formation of 3-FL is rather low in the case of BgFucT, it is even lower in case of the variant BgFucT L91W. Thus, the substrate specificity was improved after introducing the mutation L91W. The variant BgFucT L91 W has a lower activity than the BgFucT, as can be seen in the ratio of LNFP-III : LNnT, which is higher for BgFucT. LNnT is the substrate for BgFucT and has thus not been completely converted in the course of the experiment. However, in the fermentative production of LNFP-III, this lower activity can be rather easily overcome by enhancing the expression of the variant BgFucT L91 W.

[0138] Culture medium

[0139] The culture medium used to grow the cells for the production of the desired oligosaccharide contained: 3 g / L KH2PO4, 12 g / L K2HPO4, 5 g / L (NH4)SC>4, 0.3 g / L citric acid, 0.1 g / L NaCI, 2 g / L MgSCLx7 H2O and 0.015 g / L CaChx6 H2O, supplemented with 10 mg / L thiamine and 1 ml / L trace element solution (54.4 g / L ammonium ferric citrate, 9.8 g / L MnChx4 H2O, 1.6 g / L C0CL2x6 H2O, 1 g / L CuChx2 H2O, 1.9 g / L MnChx4 H2O, 1.1 g / L Na2MoO4x2 H2O, 1.5 g / L Na2SeO3, 1.5 g / L NiSCLx6 H2O). The pH of the medium was adjusted to 7.0. For synthesis of the desired oligosaccharide lactose was added to the medium to a concentration of 10 mM.

[0140] LC / MS analytics

[0141] Product identity was determined by multiple reaction monitoring (MRM) using an LC triple-quadrupole MS detection system (Shimadzu LCMS-8050). Precursor ions were selected and analyzed in quadrupole 1 , followed by collision-induced fragmentation (CID) with argon and selection of fragment ions in quadrupole 3. Selected transitions and collision energies for intermediates and end-product metabolites are listed in Table 3. LNT2, LNnT, LNFP-III, LNDFH-I I, LNnT, and 3-FL in particle free culture supernatant or cytoplasmic fractions were separated by high performance liquid chromatography (HPLC) using a Waters XBridge Amide HPLC column (3.5 pm, 2.1 A 50 mm). Neutral sugars were eluted with isocratic flow of H2O with 0.1% (v / v) ammonium hydroxide. Before LC / MS analysis, the samples of neutral oligosaccharides were prepared by filtration (0.22 pm pore size) and cleared by solid-phase extraction on a Strata ABW ion exchange matrix (Phenomenex). The HPLC system encompasses a Shimadzu Nexera X2 SIL-30ACMP autosampler run at 8 °C, a Shimadzu LC-20AD pump, and a Shimadzu CTO-20AC column oven running at 35 °C for elution of neutral sugars. For the analysis of neutral sugars 1 pl was injected into the instrument. Flow rate was set 0,3 mL / min (neutral), corresponding to a run time of 5 min. Whereas neutral sugars were analyzed in negative ion mode. The mass spectrometer was operated at unit resolution. Collision energy, Q1 and Q3 pre-bias were optimized for each analyte. Quantification methods were established using commercially available standards (Carbosynth and Elicityl).

[0142]

[0143] Table 3 List of diagnostic MRM transitions and parameters for the identification of 3-FL, LNT-II, LNnT, LNFP-III, LNFP-VI, LNnDFH-ll (CE = Collision energy) of the Shimadzu LCMS-8050 Triple Quadrupole (QQQ) Mass Analyzer. Example 3: Alphafold2-model of alpha-1, 3-fucosyltransferase from Bacteroides fragilis

[0144] For the Alphafold2 model of Example 3, the Alphafold2model of the 1 ,3- fucosyltransferase from Bacteroides fragilis with the UniProtKB entry Q5L9S6 was obtained from the AlphaFold Protein Structure Database (https: / / alphafold.ebi.ac.uk / ) using the protein identifier Q5L9S6. Protein variants of the 1 , 3-fucosyltransferase from Bacteroides fragilis were submitted as protein fasta files to a local installed Alphafold2 program (https: / / github.com / google"- The relaxed model with the highest score was taken for further studies. Within open-source PyMol, the Alphafold2model of the 1 ,3- fucosyltransferase from Bacteroides fragilis and chain B of H pylori futA with GDP- Fucose (PDB 2nzy; were structurally aligned.

[0145] The PDB coordinates of the aligned GDP-Fucose were ligand-named GDF and copied as ATOM coordinates into the Alphafold2model of the 1 , 3-fucosyltransferase from Bacteroides fragilis. Lactose docking was performed within DockingPie (GitHub - paiardinZDockingPie: A Consensus Docking Plugin for PyMOL; https: / / github.com / paiardin / DockingPie) using the structure of the 1 ,3- fucosyltransferase with GDP-Fucose obtained before and a mol2 file of lactose with Smira. Catalytically active conformations of lactose were manually checked: Within the catalytically active confirmation, the glucose 3-OH (within lactose) need to be near the active site glutamate E75 and close to C1 of fucose (within GDP-Fucose).

[0146] Fig. 2a and 2b show the model obtained as described above of the 1 ,3- fucosyltransferase from Bacteroides fragilis with GDP-Fucose as donor and lactose as acceptor close to each other with the catalytically active site E75 in-between. Glutamate E75 is close to GDP-Fucose. The alpha-1 , 3-fucosyltransferase polypeptides according to the invention possess a hydrophobic amino acid residue selected from Tryptophan or Methionine at the position corresponding to L104. Without being bound, the inventors believe that this mutation prevents binding of the lactose acceptor in this binding pocket and therefore increases the substrate specificity for LNnT. It is considered that there is a second binding pocket for LNnT to enter the catalytically active site. Example 4: Homology of alpha-1, 3-fucosyltransferases and identification of amino acid motifs

[0147] Fig 3 shows a comparison of amino acid sequences of nine selected 1 ,3- fucosyltransferases that do not carry Tryptophan or Methionine at the position of amino acid residue z. The identical amino acid residues are highlighted in black. The amino acid motifs characterizing the invention are annotated with their respective numbers from the 1stamino acid motif (1) to the seventh amino acid motif (7). It can be seen from the comparison that even though the nine selected 1 ,3- fucosyltransferases do not have a particular high sequence identity, they are all characterized by several amino acid sequences, including the first amino acid motif, which is the one comprising amino acid residue z as described above.

Claims

CLAIMS1. An alpha-1 ,3-fucosyltransferase polypeptide for the use in the production of lacto- / V-fucopentaose III (LNFP-III), wherein the alpha-1, 3-fucosyltransferase polypeptide comprises a. a first amino acid motif of [R / K / N]xx[R / W]xPzx (SEQ ID NO: 12), and b. in a distance of 5 to 300 amino acid residues in C-terminal direction of the first amino acid motif, a second amino acid motif of FCx, and c. in a distance of 1 to 10 amino acid residues in C-terminal direction of the second amino acid motif, a third amino acid motif of [S / T][D / N]; and d. in a distance of 5 to 200 amino acid residues in C-terminal direction of the third amino acid motif, a fourth amino acid motif of ENx; and e. in a distance of 5 to 20 amino acid residues in C-terminal direction of the fourth amino acid motif, a fifth amino acid motif of [S / T]EKx; and f. in a distance of 2 to 20 amino acid residues in C-terminal direction of the fifth amino acid motif, a sixth amino acid motif of PxYxG, wherein x stands for any naturally occurring alpha-amino acid and z stands for an amino acid selected from Tryptophan or Methionine.

2. The alpha-1, 3-fucosyltransferase polypeptide according to claim 1, wherein the first amino acid motif is RxxRxPzx (SEQ ID NO: 13).

3. The alpha-1, 3-fucosyltransferase polypeptide according to claim 2, comprising- a seventh amino acid motif of GENx in N-terminal direction in a distance of 5 to 50 amino acid residues of the first amino acid motif,- wherein the distance between the first amino acid motif and the second amino acid motif is between 5 and 100 amino acid residues,- wherein the distance between the third amino acid motif and the fourth amino acid motif is between 5 and 100 amino acid residues,- wherein the distance between the fourth and fifth amino acid motif is 2 to 10 amino acid residues and the fifth amino acid motif is GYx[S / T]EKx (SEQ ID NO: 14), and- in C-terminal direction after the sixth amino acid motif there are 50 to 300 amino acid residues.

4. The alpha-1, 3-fucosyltransferase polypeptide according to claim 1, wherein the alpha-1, 3-fucosyltransferase polypeptide has a sequence identity to the amino acid sequence as set forth in SEQ ID NO: 1 of at least 80%, wherein the amino acid residue z is corresponding to the amino acid residue at position 104 of SEQ ID NO: 1 and wherein the amino acid residue at position 104 is Tryptophan or Methionine.

5. The alpha-1, 3-fucosyltransferase polypeptide according to claim 1, wherein the alpha-1, 3-fucosyltransferase polypeptide has a sequence identity of at least 80% to the amino acid sequence as set forth in SEQ ID NO: 2, wherein the amino acid residue z is corresponding to the amino acid residue at position 91 of SEQ ID NO: 2 and wherein the amino acid residue at position 91 is Tryptophan or Methionine.

6. The alpha-1, 3-fucosyltransferase polypeptide according to any one of claims 1 to 5, wherein said hydrophobic amino acid is Tryptophan (W).

7. A nucleic acid molecule comprising a nucleotide sequence which encodes an alpha-1 , 3-fucosyltransferase polypeptide according to any one of claims 1 to 6.

8. A genetically engineered microbial cell comprising an alpha-1, 3- fucosyltransferase polypeptide according to any one of claims 1 to 6 and / or a nucleic acid molecule according to claim 7.

9. A genetically engineered microbial cell according to claim 8, wherein the genetically engineered microbial cell expresses a beta-1 , 3-N-acetyl hexaminyltransferase and a beta-1,4-galactosyltransferase.

10. Use of an alpha-1 ,3-fucosyltransferase polypeptide of any one of claims 1 to 6 or of a genetically engineered microbial cell of claims 8 or 9 for the synthesis of LNFP-III.

11. A method of producing LNFP-III, the method comprising the step of culturing a population of genetically engineered microbial cells according to claim 8 or 9 in a culture medium and under conditions that are permissive for the microbial cells to intracellularly synthesize LNFP-III.

12. Method according to claim 11, wherein the genetically engineered microbial cells expresses a beta-1, 3-N-acetyl hexaminyltransferase, a beta-1,4- galactosyltransferase, a lactose permease and a pathway to synthesize GDP- fucose and wherein the genetically engineered microbial cells are cultured in the presence of lactose.

Images