Enzymatic method for preparation of UDP-galactose

Abstract

The present invention relates to an enzyme-catalyzed process for producing UDP-galactose from low-cost substrates uridine monophosphate and D-galactose in a single reaction mixture. The process can be operated (semi)continuously or in batch mode. The process can be extended to uridine as starting material instead of uridine monophosphate. Further, the process can be adapted to produce galactosylated molecules and biomolecules including saccharides, proteins, peptides, glycoproteins or glycopeptides, particularly human milk oligosaccharides (HMO) and (monoclonal) antibodies.

Claims

1. A method for producing uridine 5 ′-diphospho-a-D-galactose comprising the following steps: ##STR00003## A) providing a solution comprising (i) uridine monophosphate and D-galactose represented by the following formulae ##STR00004## (ii) polyphosphate, and adenosine triphosphate; and providing a set of enzymes comprising a glucose-1-phosphate uridylyltransferase, a galactokinase, a polyphosphate kinase, and a uridine monophosphate kinase; B) producing uridine 5 ′-diphospho-α-D-galactose from uridine monophosphate and D-galactose in the presence of the set of enzymes, polyphosphate, and adenosine triphosphate.

2. The method according to claim 1, wherein the set of enzymes further comprises a pyrophosphatase.

3. The method according to claim 1, wherein the set of enzymes further comprises a one-domain polyphosphate kinase 2.

4. The method according to claim 1, wherein the set of enzymes further comprises a two-domain polyphosphate kinase 2.

5. The method according to claim 1, wherein at least one enzyme of the set of enzymes is immobilized on a solid support.

6. The method according to claim 1, wherein the set of enzymes is co-immobilized on a solid support.

7. The method according to claim 6, wherein the set of enzymes is directly co-immobilized on a solid support from fermentation broth, crude cell lysate, purified cell lysate or cell homogenate.

8. The method according to claim 1, wherein the concentration of uridine monophosphate and D-galactose in the solution provided in A) is in the range of 0.2 mM to 15,000 mM.

9. The method according to claim 1, wherein the polyphosphate is a long-chain polyphosphate having at least 25 phosphate residues.

10. The method according to claim 1, wherein the uridine 5′-diphospho-α-D-galactose is produced in a single reaction mixture.

11. The method according to claim 1, wherein the uridine monophosphate in A) is obtained from uridine, adenosine triphosphate and a uridine kinase; or from uracil, 5-phospho-α-D-ribose 1-diphosphate and a uracil phosphoribosyltransferase; or from orotic acid, 5-phospho-α-D-ribose 1-diphosphate, an orotate phosphoribosyltransferase and a UMP transferase.

12. The method according to claim 1, further comprising producing a galactosylated saccharide, galactosylated glycopeptide, galactosylated glycoprotein galactosylated protein, galactosylated peptide, galactosylated bioconjugate or galactosylated small molecule from uridine 5′-diphospho-α-D-galactose and a saccharide, glycopeptide, glycoprotein, protein, peptide, bioconjugate or small molecule by forming an O-glycosidic bond between uridine 5′-diphospho-α-D-galactose and an available hydroxyl group of the saccharide, glycopeptide, glycoprotein, protein, peptide, bioconjugate or small molecule in the presence of a galactosyltransferase.

13. The method according to claim 12, wherein the saccharide, glycopeptide, glycoprotein, protein, peptide, bioconjugate or small molecule is an antibody or a monoclonal antibody; or a human milk oligosaccharide or a bioconjugate.

14. The method according to claim 12, further comprising recycling of uridine diphosphate formed from the producing a GlcNAcylated saccharide, a GlcNAcylated glycopeptide, a GlcNAcylated glycoprotein, a GlcNAcylated protein, a GlcNAcylated peptide, a GlcNAcylated bioconjugate or a GlcNAcylated small molecule from uridine 5′-diphospho-N-acetylglucosamine and a saccharide, glycopeptide, glycoprotein, protein, peptide, bioconjugate or small molecule by forming an O-glycosidic bond between uridine 5′-diphospho-N-acetylglucosamine and an available hydroxyl group of the saccharide, glycopeptide, glycoprotein, protein, peptide, bioconjugate or small molecule in the presence of an N-acetylglucosaminyltransferase to obtain uridine triphosphate.

15. The method according to claim 12, wherein the saccharide, glycopeptide, glycoprotein, protein, peptide, bioconjugate or small molecule is a carbohydrate conjugate vaccine or an antibody drug conjugate.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1: shows the multi-enzyme cascade through which UDP-galactose is enzymatically synthesized from low-cost substrates galactose, polyphosphate and UMP. The reaction cascade consists of (a) the formation of galactose-1-phosphate (Gal-1P) from D-galactose and ATP, (b) the formation of uridine triphosphate (UTP) from UMP and polyphosphate, and (c) the reaction of galactose-1-phosphate with uridine triphosphate to UDP-galactose. Optionally an inorganic diphosphatase (PmPpa) can added to the reaction cascade in order to hydrolyze pyrophosphate PP.sub.i which inhibits the enzyme glucose 1-phosphate uridylyltransferase. The cascade can also be extended by adding a 1D-PPK2 to assist the conversion of ADP to ATP. Also, the cascade can be extended by adding a 2D-PPK2 in order to activate phosphorylation of AMP to ADP. Moreover, the cascade can be extended by adding a 1D-PPK2 and a 2DPPK2 in order to inhibit frequent hydrolysis of adenosine phosphates.

(2) FIG. 2: shows an exemplary reaction scheme of the inventive method for producing UDP-galactose starting from uridine or uracil and 5-phospho-α-D-ribose 1-diphosphate. The formation of UMP from uridine is catalyzed by uridine kinase and the formation of UMP from uracil is catalyzed by uracil phosphoribosyltransferase.

(3) FIG. 3: shows the SDS-gel of the purified enzyme mix obtained by Expression mode B.

(4) FIG. 4A: shows the reaction time course of all measured compounds.

(5) FIG. 4B: shows the HPAEC-UV chromatogram of the feed solution after 0 min reaction time.

(6) FIG. 4C: shows the HPAEC-UV chromatogram of aliquots taken after a reaction time of 370 min.

(7) FIG. 5A: shows the reaction time course of all measured compounds.

(8) FIG. 5B: shows the HPAEC-UV chromatogram of the feed solution after 0 min reaction time.

(9) FIG. 5C: shows the HPAEC-UV chromatogram of aliquots taken after a reaction time of 540 min.

(10) FIG. 6: shows substrate, metabolite and product concentrations after a reaction time of 14 h as measured by HPAEC-UV/PAD.

(11) FIG. 7: shows a workflow scheme for the complete UDP-galactose cascade starting from mixing the biomasses containing the overexpressed enzymes to carrying out the synthesis reaction of UDP-galactose on a solid support. The workflow is also suitable for screening various solid supports for enzyme immobilization.

(12) FIG. 8: Results of the solid support screening of the UDP-galactose synthesis. Concentrations were measured by HPAEC-UV.

(13) FIG. 9: shows the reaction scheme of the UDP-galactose cascade coupled to GalT to glycoengineer commercial antibodies such as Rituximab or Herceptin.

(14) FIG. 10: shows electropherogram (CGE-LIF analysis) of Rituximab (A) and galactosylated Rituximab (B) prepared by the inventive method.

(15) FIG. 11: shows electropherogram (CGE-LIF analysis) of Rituximab galactosylated in a one stage process of the inventive method.

(16) FIG. 12: shows electropherogram (CGE-LIF analysis) of Rituximab galactosylated in a two stage process of the inventive method.

(17) FIG. 13A: shows a process scheme for the inventive galactosylation of molecules, such as glycoproteins or antibodies, in a two reactor setup.

(18) FIG. 13B: shows a process scheme for the inventive galactosylation of molecules, such as glycoproteins or antibodies, in a one-step one reactor setup. D-Galactose polyphosphate and UMP in catalytic amounts are added to a reactor containing a substrate to be galactosylated, beads loaded with the enzymes of the inventive UDP-Gal cascade and a galactosyltransferase. The galactosyltransferase may also be present in solution and not immobilized on the beads. Only catalytic amounts of UMP are required since the UDP-Gal consumed in the galactosylation reaction is continuously regenerated in the presence of the beads loaded with the enzymes of the inventive UDP-Gal cascade, galactose and polyphosphate.

(19) FIG. 14: shows exemplary galactosylated human milk saccharides.

(20) FIG. 15: shows exemplarily the reusability of UDP-Gal enzyme cascade co-immobilized on methacrylate beads functionalized with epoxy groups.

(21) FIG. 16 shows intermediates and product formed in the UDP-Gal cascade of Experiment H. (A) UDP-Gal and uridine; (B) UMP, UDP and UTP; (C) ADP, AMP and ATP. The experiments were carried out in triplicate; error bars represent standard deviation.

(22) FIG. 17 shows educts, intermediates and product formed in the UDP-Gal scale-up experiment of Example I in direct comparison with small-scale run. (A) uridine; (B) UDP-Gal; (C) UMP; (D) UDP; (E) UTP; (F) ADP; (G) ATP; and (H) AMP.

(23) FIG. 18 shows (A) chromatogram of reaction products containing LNnT and (B) MS/MS spectrum of the reaction product.

(24) FIG. 19 shows the MS/MS spectrum of the reaction product of the formation of para-Lacto-N-neohexaose (para-LNnH) (experiment K).

(25) The following examples are included to demonstrate preferred embodiments of the invention. It should be appreciated by those skilled in the art that the techniques disclosed in the examples, which follow represent techniques discovered by the inventor to function well in the practice of the invention, and thus can be considered to constitute preferred modes for its practice. However, those skilled in the art should, in light of the present disclosure, appreciate that many changes can be made in the specific embodiments, which are disclosed and still obtain a like or similar result without departing from the spirit and scope of the invention.

(26) Further modifications and alternative embodiments of various aspects of the invention will be apparent to those skilled in the art in view of this description. Accordingly, this description is to be construed as illustrative only and is for the purpose of teaching those skilled in the art the general manner of carrying out the invention. It is to be understood that the forms of the invention shown and described herein are to be taken as examples of embodiments. Elements and materials may be substituted for those illustrated and described herein, parts and processes may be reversed, and certain features of the invention may be utilized independently, all as would be apparent to one skilled in the art after having the benefit of this description of the invention. Changes may be made in the elements described herein without departing from the spirit and scope of the invention as described in the following claims.

EXAMPLES

(27) Abbreviations and Acronyms ADP adenosine 5′-diphosphate AMP adenosine 5′-monophosphate ATP adenosine 5′-triphosphate dH.sub.2O deionized water IPTG isopropyl β-D-thiogalactopyranoside LGTB Lacto-N-neotetraose biosynthesis glycosyltransferase UDP uridine 5′-diphosphate UMP uridine 5′-monophosphate UTP uridine 5′-triphosphate GTP guanosine 5′-triphosphate PolyP polyphosphate PPi pyrophosphate Pi phosphate PPK2 polyphosphate kinase 2 PPK3 polyphosphate kinase 3 1D-PPK2 1-domain polyphosphate kinase 2 2D-PPK2 2-domain polyphosphate kinase 2 GalU glucose 1-phosphate uridylyltransferase GalT UDP-galactosyltransferase BiGalK galactokinase GalK galactokinase URA6 uridine monophosphate kinase UPP uracil phosphoribosyltransferase PmPpA Pasteurella multocida inorganic pyrophosphatase

(28) Chemicals & Reagents

(29) Unless otherwise stated, all chemicals and reagents were acquired from Sigma-Aldrich, and were of the highest purity available. Solid supports were obtained from Resindion, ChiralVision, Röhm GmbH & Co. KG and micromod GmbH.

Example 1: Preparation of Enzymes

(30) The genes encoding for the enzymes BiGalK, URA6, PPK3, GalU, 1D-PPK2, 2D-PPK2 and PmPpA were cloned into standard expression vectors as listed in Table 1.

(31) TABLE-US-00001 TABLE 1 Enzymes used in this example Enzyme Abbreviation EC class Origin SEQ ID glucose 1-phosphate GalU 2.7.7.9 E. coli K-12 MG1655 4 uridylyltransferase Galactokinase BiGalk 2.7.1.6 Bifidobacterium 1 infantis ATCC 15697 Polyphosphate kinase 3 PPK3 2.7.4.1 Ruegeria pomeroyi 3 Uridine monophosphate kinase URA6 2.7.4.14 Arabidopsis thaliana 2 Inorganic diphosphatase PmPpa 3.6.1.1 Pasteurella 5 multocida Pm70 1-domain polyphosphate 1D-PPk2 2.7.4.1 Pseudomonas aeruginosa 6 kinase 2 2-domain polyphosphate 2D-PPK2 2.7.4.1 Pseudomonas aeruginosa 7 kinase 2

(32) Transformation, Cultivation, Expression

(33) For all gene expressions E. coli BL21 Gold (DE3) was used as a host organism unless stated otherwise.

(34) Gene Expression: One-Enzyme, One-Cultivation (Expression Mode A).

(35) Plasmids and Stock Cultures

(36) Stock solutions of E. coli cultures carrying the plasmids (pET28a with kanamycin resistance) with the gene sequences of GalU, PPK3, URA6, PmPpa, 1D-PPK2 were available from earlier studies (see [1, 2]). The stock solutions contained 50% glycerol and were kept at −20° C.

(37) Gene synthesis and cloning of the gene sequence of BiGalK into expression vector pET100/D-TOPO with an antibiotic resistance against ampicillin were carried out by a commercial supplier and according to earlier published literature [3].

(38) The purchased plasmid was transferred into E. coli by transferring 1 μl of the plasmid stock solution into a culture E. coli BL21 Gold (DE3). The solution was than kept on ice for 1 h, followed by heat shocking the cells for 1 min at 42° C. Subsequently, 500 μL of LB media were added and the mix was incubated for 20 min at 37° C. followed by centrifuging the solution at 6000 g and 4° C. for 10 min. The supernatant was discarded and the cell pellet dissolved in 100 μl deionized H.sub.2O (dH.sub.2O.) and spread on LB agar plates containing ampicillin. The agar plate was incubated at 37° C. Stock solutions of E. coli cells containing the plasmid were generated in 2 mL slant media.

(39) Enzyme Expression

(40) For heterologous gene expression, aliquots were removed from the stock solutions and spread on LB agar plates containing the according antibiotic. The plates were cultivated overnight at 37° C. Single cultures were used to inoculate precultures (containing 50 μg/mL kanamycin and 100 μg/mL ampicillin, respectively) in shaker flasks with baffles. Cultures were typically grown to an OD.sub.600 of about 4.2. Main expression cultures containing 50 μg/mL kanamycin and 100 μg/mL ampicillin, respectively, were typically inoculated with 1% preculture and cultivated at 37° C. to an OD.sub.600 of around 0.6-0.8. The temperature was then changed to 16-20° C. and the expression was induced with typically 0.4 mM IPTG. After, typically, 20 h. The cultures were harvested typically by 6000 xg for 30 min at 4° C. Media used were autoinduction (AI) media, LB and TB media. More details on the media used in the experiments are given in table 2 below.

(41) TABLE-US-00002 TABLE 2 The content of growth media for E. coli is detailed. All media were autoclaved before use. Media Content Luria-Bertani (LB) 10 g tryptone 5 g yeast extract 5 g NaCl in 1 L dH.sub.2O Terrific broth (TB) 24 g yeast extract 12 g tryptone 5 g glycerol 89 mM Phosphate buffer (added after autoclaving) in 1 L dH.sub.2O Auto induction (AI) See [5] Slant 20 g tryptone 10 g yeast extract in 1 L dH.sub.2O with glycerol (50% v/v)

(42) Enzyme Purification

(43) The plasmids pET28a and pET100/D-TOPO harbor a N-terminal His6-tag and the enzyme are, thus, purified with Ion metal affinity chromatography using the ÄKTA™ start system and HisTrap High-Performance or Fast-Flow columns (1 mL column volume) from GE Healthcare. For the purification of enzymes the cells were lysed by sonication in lysis buffer (50 mM HEPES (pH 7.5), 10 mM Mg.sup.2+, 300 mM NaCl, 10 mM imidazole and 5% glycerol).

(44) Imidazole (500 mM) was used as eluent in isocratic elutions (50 mM HEPES (pH 7.5), 10 mM Mg.sup.2+, 300 mM NaCl, 500 mM imidazole and 5% glycerol). Standard conditions as recommended by the manufactures were used. After purification the enzyme concentrations were tested by BCA assays and evaluated by SDS-gels.

(45) Gene Expression: All Enzymes, One Cultivation (Expression Mode B).

(46) For the gene expression described in this section the LOBSTR E. coli expression strain (based on E. coli BL21 Gold (DE3)) from Kerafast Inc was used. Two gene sequences were cloned into one specific expression vector each. An E. coli strain was created carrying all three expression plasmids.

(47) Cloning

(48) The resistance markers and restriction sites for the used expression vectors are detailed in Table 3.

(49) pACZDuet vector harboring the gene sequences for URA6 and PPK3 was bought from a commercial supplier. The gene sequences of GalU and PmPpA were cut by enzymatic digestion form isolated pET28a vectors and cloned into pCDFDuet expression vector. Standard protocols for enzymatic digestion, PCR and ligation were used for the cloning. GalK from pET100-D/TOPO and NahK from pET28a were cloned into expression vector pRSFDuet1. Empty expression vectors pCDFDuet and pRSFDuet1 were purchased from a commercial supplier.

(50) The gene constructs were confirmed by gene sequencing by a commercial supplier.

(51) TABLE-US-00003 TABLE 3 Gene sequences with restriction sites and expression vector for Expression Mode B. Template Restriction sites Destination vector (Res.) GalK NcoI, NotI pRSFDuet1 (kanamycin) GalU NcoI, NotI PmPpA NdeI, KpnI pCDFDuet (spectinomycin) URA6 NcoI, NotI PPK3 NdeI, KpnI pACYCDuet (chloramphenicol)

(52) Transformation

(53) All plasmids were transformed into LOBSTR E. coli cells by heat shock (as described in the “Gene expression: One enzyme, one cultivation” section) and then plated on LB agar plates containing all selection markers (chloramphenicol, spectinomycin, kanamycin). Thus, only those cells carrying all three vectors could grow on the agar plates.

(54) Enzyme Expression

(55) For the expression described here TB media was used containing the following concentrations of antibiotics (34 μg/mL chloramphenicol, 50 μg/mL spectinomycin, and 30 μg/mL kanamycin). The cells were precultured in 15 mL at 30° C. overnight, and main cultures of 200 mL were inoculated with 1% preculture and cultivated at 30° C. up to OD.sub.600=0.8. The temperature was lowered to 16° C. and the expression was induced by adding 0.5 mM IPTG. After 20 h the cells were harvested by centrifuging at 6000 xg for 30 min at 4° C. Cell were lysed by sonication in lysis buffer (50 mM HEPES (pH 7.5), 10 mM Mg.sup.2+, 300 mM NaCl, 10 mM imidazole and 5% glycerol).

(56) Purification

(57) As described in the section “Gene expression: One enzyme, one cultivation”. The enzyme concentrations were tested by the BCA assay and the purification was evaluated by a SDS-gel (see FIG. 3).

(58) Measurements

(59) High-performance anion exchange chromatography (HPAEC) with UV (260 nm) and pulsed amperometric detection (PAD) was utilized to measure concentrations of reactants. For analyte separation and quantification, a step gradient elution method was developed and validated. Chromatographic separation was performed at a system flow of 0.5 mL/min using a non-porous pellicular column CarboPac™ PA1 (250×2 mm). The HPAEC system (ICS5000) as well as all columns, components and software were purchased from Thermo Scientific (Waltham, USA).

(60) Saccharides on antibodies were analyzed by PNGase F treatment and CGE-LIF analysis. Standard protocols were followed for the analysis.

Example 2: Homogeneous Preparation of UDP-Galactose—Experiments A, B and C

(61) Purified enzymes from the Expression mode B were used for these experiments. The synthesis was carried out without 1D-PPK2 (Experiment A, see FIG. 1 for the pathway) and with 1D-PPK2 (Experiment B). The reaction volumes were 12 μL for Experiment A and 34 μL for Experiment B in HEPES (pH 7.5) buffered aqueous solutions. The reaction temperature was 30° C. The amount of purified enzymes used was 400 μg. Initial substrate, buffer, co-factor concentrations and the amount of enzymes used are detailed in Table 4.

(62) Reaction aliquots for reaction time course measurements were quenched as follows. For Experiment A 2 μL of the reaction were aliquoted and diluted in 298 μL of 90° C. dH.sub.2O, for Experiment B 5 μL were diluted in 315 μL of 90° C. dH.sub.2O.

(63) TABLE-US-00004 TABLE 4 Substrate, co-factor and buffer concentrations as well as amount of enzymes used in Experiment A and B. Compound Experiment A Experiment B Experiment C HEPES (mM) 35 25 50 MgCl.sub.2 (mM) 13 10 Purified enzyme 400 400 various mix (μg) 1D-PPK2 (μg) — 5.2 various UMP (mM) 3.5 2.8 2.5 ATP (mM) 1.8 1.3 2 D-galactose (mM) 3.5 2.8 2.5 PolyP.sub.25 (mM) 5 3.75 6

(64) The reaction time course of Experiment A is shown in FIG. 4A-C. After 370 min a UDP-galactose yield of 100% was achieved with respect to UMP and galactose. This result shows that this combination of enzymes can achieve full conversion of substrates to UDP-galactose. There is no apparent enzyme inhibition and no side reactions take place. AMP is detected showing that ADP is partly hydrolyzed.

(65) The reaction time course of Experiment B is shown in FIG. 5A-C. After 540 min a UDP-galactose yield of 100% was achieved with respect to UMP and galactose. This result shows that this combination of enzymes can achieve full conversion of substrates to UDP-galactose. There is no apparent enzyme inhibition and no side reactions take place. However, no AMP was detected showing that in the presence of 1D-PPK2, ADP was converted back to ATP fast enough before detectable amounts of ADP were hydrolyzed to AMP (see Experiment A).

(66) In Experiment C various enzyme concentrations were tested and the effect of this combination on the productivity was investigated. Enzymes were expressed as detailed in the section “Gene expression: One enzyme, one cultivation (Expression mode A)”. The reaction volumes were 100 μl. Reactions were quenched after a reaction time of 14 h by taking an aliquot of 20 μl and diluting it in 480 μl dH.sub.2O (90° C.).

(67) TABLE-US-00005 TABLE 5 Enzyme concentrations used for the 4 reactions in Experiment C. Series 1 Series 2 Series 3 Series 4 Enzyme (μg) (μg) (μg) (μg) Galk 300 300 300 300 URA6 300 300 300 300 PPK3 440 440 440 440 GalU 870 870 870 870 PmPpA 440 440 440 440 1D-PPK2 0 6.5 0 6.5 2D-PPK2 0 0 160 160

(68) Results of Experiment C are depicted in FIG. 6. UDP-galactose was successfully formed in all reactions.

Example 3: Heterogeneous Preparation of UDP-Galactose—Experiment D

(69) In Experiment D, a wide range of commercially available solid supports (Table 7) were screened for the co-immobilization of the enzymes used in the inventive UDP-galactose synthesis (see FIG. 1) and their effect on the synthesis of UDP-galactose was evaluated. As depicted in FIG. 7, the following protocol was used for the experiment: Biomasses from different cultivations were mixed together [see FIG. 7, step 1] and centrifuged 6000 xg for 30 min at 4° C. [step 2]. The composition of cultures used is detailed in Table 8. Cell pellets were resuspended in 60 mL buffer A [step 3] (see Table 6). Cells were lysed by sonication [step 4]. After sonication—the slurry was centrifuged 12 000 xg for 45 min at 4° C. [step 5] to remove cell debris, followed by filtration through 1.2 μm and 0.8 μm filters. After centrifugation, the supernatant was removed and kept on ice. A given mass of each immobilizer (see Table 8) was added to a 2 mL low-binding tube. After approximately 2 h of incubation with buffer A, the supernatant was removed. Afterwards, 0.5 mL of cell lysate were added to each tube and incubated overnight (˜12 h) at 4° C. [step 6]. The beads were washed (3 times) with lysis buffer B [step 7] and blocking buffer (2 M glycine) was added followed by incubation for 24 h [step 8]. Afterwards, the blocking buffer was discarded and the beads were washed with buffer B (see Table 6) three times.

(70) To test the multi-enzyme cascade on various enzyme loaded beads, 100 μL of the feed solution (see Table 9) containing substrates was transferred to each tube containing the beads. The reactions were carried out for around 20 h at 30° C. and under shaking (400 rpm). The UDP-galactose concentrations were then measured by HPAEC-UV/PAD. The results are depicted in FIG. 8. It is shown that the enzymes are active when co-immobilized on a wide variety of commercially available beads.

(71) TABLE-US-00006 TABLE 6 Buffer compositions for Experiment D. Buffer A Buffer B HEPES pH 7.5 (mM) 75 200 MgCl.sub.2 (mM) 20 20 NaCl (mM) 300 500 Glycerol (% v/v) 5 5 Protease Inhibitor (Roche, 3 tablets 3 tablets EDTA-free “cOmplete ™”)

(72) TABLE-US-00007 TABLE 7 Table of solid supports tested in Experiment D. Mass used in Solid support Experiment D (mg) EC-EP 66 EP403/M 68 IB-COV1 53 IB-COV2 58 IB-COV3 50 Eupergit ® CM 49 ECR8215F 52 ECR8204F 51 ECR8209F 52 ECR8285 52 EP403/S 54 EP400/SS 62

(73) TABLE-US-00008 TABLE 8 Overall volumes of cultures containing the overexpressed enzymes in E. coli used for Experiment D. AI media (mL) LB media (mL) TB media (mL) GalK 160 80 PPK3 80 120 URA6 120 GalU 200 PmPpa 40 1D-PPK2 80

(74) TABLE-US-00009 TABLE 9 Concentrations of the feed solution used in Experiment D. Compound Conc. (mM) Galactose 4.5 UMP 10.1 ATP 30 PolyP.sub.25 30 MgCl.sub.2 57 HEPES (pH 7.5) 100

Example 4: Galactosylation of Antibodies—Experiments F and G

(75) In Experiment F and G the UDP-galactose cascade immobilized on solid support ECR8285 was coupled to a soluble UDP-galactosyltransferase (GalT) bought from Sigma-Aldrich to galactosylate the commercially available antibody Rituximab (purchased from Evidentic GmbH).

(76) Pretests

(77) In pretests (Experiment E) the glycoprofile of Rituximab was analyzed by PNGase F digest and CGE-LIF (see section “Measurements”). The results are depicted in FIG. 10A. It can be seen that only a small fraction of the glycans of the Fc-region of the antibody are fully galactosylated. To engineer the glycoprofile of the antibody, 100 μg Rituximab were incubated with bought UDP-Gal (purchased from Sigma-Aldrich, Order no. U4500) 25 milliunits of GalT (from Sigma-Aldrich, Order no. G5507) in 50 mM HEPES and 10 mM MnCl.sub.2 overnight at 30° C. The results are depicted in FIG. 10B. The CGE-LIF analysis showed that after the reaction all detected glycans were galactosylated.

(78) Coupling the cascade to GalT

(79) One-Stage Coupling

(80) In Experiment F feed solution (250 μL, see Table 9) was added to the ECR8285 beads (52 mg of solid support, weight measured before enzymes were immobilized on the bead) (from Experiment D) harboring the immobilized cascade. Immediately afterwards, 100 μg of Rituximab, GalT (25 milliunits) and 10 mM of MnCl.sub.2 were added. After an incubation time of 24 h at 30° C. and shaking at 550 rpm, the supernatant of the reaction was then analyzed by PNGase F digest and CGE-LIF analysis (see section “Measurements”). The results are depicted in FIG. 11. It can be seen that all glycans were galactosylated. Most glycans are fully galactosylated while the smaller fraction exhibited only one galactose moiety on one of the two branches, indicating that achieving full galactosylation is a matter of incubation time and optimization of reactions conditions, i.e. reactant concentrations, respectively.

(81) Two-Stage Coupling—See FIG. 13

(82) In Experiment G, 100 μg Rituximab were galactosylated in a two stage process. In the first stage the enzyme cascade immobilized on ECR8285 was used to produce UDP-galactose. 52 mg of the beads (weight measured before enzymes were immobilized) were incubated with the feed solution (100 μL) at 30° C. for 24 h. In the second stage the supernatant was transferred to another reactor containing 100 μg Rituximab, 25 miliunits GalT and 10 mM MnCl.sub.2. The reactants were then incubated at 30° C. and 550 rpm for around 24 h. The supernatant of reaction was analyzed by PNGase F digest and CGE-LIF analysis (see section “Measurements”) (see FIG. 12). Identical to Experiment F, all glycans were galactosylated with most glycans exhibiting galactose moieties on both branches, indicating that achieving full galactosylation is a matter of incubation time and optimization of reactions conditions, i.e. reactant concentrations, respectively.

Example 5: Synthesis of UDP-Gal Starting from Uridine—Experiments H, I, J and K

(83) Production and Purification of the Enzymes

(84) The list of the plasmid used in this study is shown in Table 10. LOBSTR E. coli competent cells (Kerafast, US) were used as the expression host. Cells were transformed based on heat-shock protocol. The fermentation carried out in TB media supplement with 1.5 mM MgSO.sub.4 and corresponding antibiotics. The cells were cultivated at 37° C. until OD.sub.600 of 0.8-1.0, afterwards, induction carried out with 0.4 mM IPTG, followed by 20-24 h cultivation at 16° C. All the chemicals used are from Carbosynth Ltd.

(85) At the end of the cultivation, cells were harvested by centrifugation (7000×g, 20 minutes) and cell pellets were resuspended in lysis buffer (50 mM MOPS buffer, 300 mM NaCl, 10 mM MgCl.sub.2, 10 mM imidazole and 5% glycerol at pH 7.4) and were disrupted by high-pressure homogenization (Maximator, Germany) (3 times passage at 800-1000 psi. Enzymes were purified using immobilized metal affinity chromatography (ÄKTAstart, GE Health care Life Sciences, Uppsala, Sweden) in combination with 1 mL or 5 mL HisTrap HP (GE Health care Life Sciences, Sweden) columns. The binding buffer contained 50 mM MOPS buffer, 300 mM NaCl, 10 mM MgCl.sub.2, 10 mM imidazole and 5% glycerol at pH 7.4. And the elution buffer consisted of 50 mM MOPS buffer, 300 mM NaCl, 10 mM MgCl.sub.2, 250 mM imidazole and 5% glycerol at pH 7.4.

(86) To remove imidazole from the elution buffer and to concentrate the enzyme solution, buffer exchange was performed with Amicon® Ultra-15 Centrifugal Filter Unit—3 KDa MW cutoff (Merck, Germany). The exchange buffer contained 50 mM MOPS buffer, 300 mM NaCl, 10 mM MgCl.sub.2, 5% glycerol at pH 7.4. Afterwards, the retentate solution (concentrated enzyme) was mixed 1:1 with glycerol to have the final enzyme solution in 50% glycerol. Enzymes were stored at −20° C.

(87) TABLE-US-00010 TABLE 10 Enzymes used in this example. Gene Abbr. Enzyme Uniprot. No. Origin Plasmid SEQ ID No galK GALK Galactokinase B3DTF0 Bifidobacterium pET100/D- 8 longum TOPO galu GALU UTP-glucose- P0AEP3 Escherichia coli pET-28a(+) 4 1-phosphate (strain K12) uridylyl- transferase ppa PmPpA Inorganic P57918 Pasteurella pET-28a(+) 5 diphosphatase multocida udk UDK uridine/cytidine P0A8F4 Escherichia coli pET-28a(+) 9 kinase (strain K12) UMK3 URA6 UMP/CMP O04905 Arabidopsis pACYCDuet 2 kinase thaliana SPO1727 PPK3 NDP Q5LSN8 Ruegeria pACYCDuet 3 kinase/polyP.sub.n pomeroyi kinase

Experiment H—Synthesis of UDP-Gal Starting from Uridine Using Purified Enzymes

(88) The reactions contained 150 mM Tris-HCl (pH 8.5), 75 mM MgCl.sub.2, 52 mM uridine, 54 mM Gal, 0.6 mM ATP, 20 mM PolyP.sub.n, 0.07 μg/μL UDK, 0.12 μg/μL URA6/PPK3, 0.17 μg/μL GALK, 0.12 μg/μL GALU, 0.06 μg/μL PmPpa in the total volume of 250 μL. The successful production of UDP-Gal and concentration of intermediates are shown in FIG. 16. Yields are approaching 100% after 24 hours.

Experiment I—Large Scale Production of UDP-Gal from Uridine and Gal Using Cell Lysate

(89) For the preparation of cell-lysate the following biomasses were mixed: UDK, 3.46 g; URA6/PPK3, 5.20 g; GALK, 5.54 g; GALU, 5.70 g; PmPpA, 1.7 g in 120 mL of 50 mM HEPES buffer (pH 8.1), 400 mM NaCl, and 5% glycerol. The mixture was passed three times through a high-pressure homogenizer. The cell-free extract was centrifuged at 10,000×g for 45 min. Afterwards, small scale (200 μL) preliminary experiments were carried out to find a suitable lysate amount for the UDP-Gal synthesis. It was found out 10% of v/v of cell lysate is sufficient to perform the synthesis. These findings were directly used for the 1 liter scale synthesis which correlate to 5000× scaling factor.

(90) To carry out the 1 liter experiment, a spinner flask equipped with a propeller type impeller was chosen to mimic the condition of a stirred tank reactor. The synthesis conditions were as follows: 150 mM Tris-HCl (pH 8.5), 58 mM uridine, 55 mM Gal, 6.2 mM ATP, 20 mM PolyP.sub.n, and 75 mM MgCl.sub.2. The reaction was carried out at 37° C. and 60 rpm. To understand the effect of the scale-up on the performance of the cascade, a parallel 200 μL experiment was carried out. The time courses of cascade intermediates are shown in FIG. 17.

Experiment J—One-Pot Production of Lacto-N-Neotetraose (LNnT)

(91) In these experiments, HMOS are synthesized using recombinant Leloir glycosyltransferases and nucleotide sugar modules (UDP-Gal and UDP-GlcNAc). The nucleotide sugar is synthesized first and subsequently, the reaction mixture is combined with the glycosyltransferases and substrates to produce the target HMO.

(92) UDP-Gal was produced based on the condition described in Table 11. All the reactions were carried out with an incubation time of around 24 hours, 550 rpm shaking at 37° C. Afterwards, 75 μL of the reaction module containing the product UDP-Gal was transferred to a new vial containing 0.2 μg/μL LGTB (Lacto-N-neotetraose biosynthesis glycosyltransferase, from Neisseria meningitidis serogroup B (strain MC58), expressed in E. coli BL21), 20 units of alkaline phosphatase (AP), 150 μL of Lacto-N-triose (LNT II), and 156 mM of MES buffer (pH 5.5). The chromatogram of the reaction product and MS/MS results, after overnight incubation, are shown in the FIG. 18.

(93) TABLE-US-00011 TABLE 11 Reaction conditions for production of UDP-Gal. Reaction Mixture Concentration UDK 0.06 μg/μL URA6/PPK3 0.1 μg/μL GALK 0.16 μg/μL GALU 0.12 μg/μL PPA 0.06 μg/μL Gal 52 mM Uridine 50 mM ATP 2.5 mM PolyP.sub.25 20 mM Tris-HCl (8.5) 150 mM MgCl.sub.2 75 mM Total Volume 20 mL

Experiment K—One-Pot Production of Para-Lacto-N-Neohexaose (Para-LNnH)

(94) 50 μL of a para-Lacto-N-neopentaose containing solution was mixed with 40 μL of the UDP-Gal reaction mixture as listed in Table 11 containing UDP-Gal as a reaction product, and 20 units of AP and 0.2 μg/μL of LGTB in MES buffer (240 mM-pH 6.5) in a total volume of 210 μL. The MS/MS spectrum of the reaction product is shown in FIG. 19.

REFERENCES

(95) 1. Mahour, R., et al., Establishment of a five-enzyme cell-free cascade for the synthesis of uridine diphosphate N-acetylglucosamine. Journal of Biotechnology, 2018. 283: p. 120-129. 2. Rexer, T. F. T., et al., One pot synthesis of GDP-mannose by a multi-enzyme cascade for enzymatic assembly of lipid-linked oligosaccharides. Biotechnology and Bioengineering, 2018. 115(1): p. 192-205. 3. Li, L., et al., A highly efficient galactokinase from Bifidobacterium infantis with broad substrate specificity. Carbohydrate Research, 2012. 355: p. 35-39. 4. Warnock, D., et al., In vitro galactosylation of human IgG at 1 kg scale using recombinant galactosyltransferase. Biotechnology and Bioengineering, 2005. 92(7): p. 831-842. 5. Li, Z., et al., Simple defined autoinduction medium for high-level recombinant protein production using T7-based Escherichia coli expression systems. Applied Microbiology and Biotechnology, 2011. 91(4): p. 1203.