WO2023044368A1 - Method for producing glycosylated therapeutics by using an immobilized enzyme preparation - Google Patents

Abstract

Methods are disclosed for producing glycosylated therapeutics using immobilized uridine diphosphate glycosyltransferases (UGTs). Also disclosed are methods of using multiple immobilized enzymes, such as UGTs and sucrose synthases (SuSy), in sequential reactions that are more efficient and cost effective when compared to the lone UGT reaction. Both approaches exhibit satisfactory properties in the glycosylation process, such as high rates of conversion and low labor costs and should be widely usable with UGTs from multiple species with similar effects.

Description

METHOD FOR PRODUCING GLYCOSYLATED THERAPEUTICS BY USING AN IMMOBILIZED ENZYME PREPARATION

RELATED APPLICATION

[0001] This application claims the benefit of U.S. Provisional Application No. 63/245,723, filed on September 17, 2021. The entire teachings of the above application are incorporated herein by reference.

INCORPORATION BY REFERENCE OF MATERIAL IN XML

[0002] This application incorporates by reference the Sequence Listing contained in the following extensible Markup Language (XML) file being submitted concurrently herewith: a) File name: 57671003001. xml; created September 8, 2022, 8,153 Bytes in size.

BACKGROUND

[0003] Glycosylation is a potential strategy for improving the efficacy, solubility, potency, pharmacokinetic (PK), and/or pharmacodynamic (PD) attributes of small-molecule therapeutics. International patent applications PCT/US2021/022416, PCT/US2021/022410, and PCT/US2021/022414 disclose glycosylation of ivacaftor, enasidenib, and etoposide, respectively. However, glycosylation of small molecule drugs is expensive and time consuming.

SUMMARY

[0004] Described herein are methods for producing glycosylated therapeutics. Also described herein are reaction systems for using uridine diphosphate glycosyltransferases (UGTs) to glycosylate a target compound, such as a compound having therapeutic properties when administered to a mammal.

[0005] Non-covalent affinity-based immobilization is a less common technique for enzyme immobilization than chemical crosslinking for enzyme immobilization. In general, a wide variety of resin types can be used for affinity -based immobilization, such as nitrocellulose membranes or polylysine-coated slides [1], Affinity immobilization is beneficial due to the ease with which the protein is recoverable. A metal affinity tag on the enzyme that associates with a nickel resin does not negatively affect the UGT enzyme, which is not always the case when adding a metal affinity tag to an enzyme. Accordingly, a nickel resin allows for one step preparation and immobilization of a UGT rather than multiple steps for immobilizing the UGT enzyme, as is common.

[0006] Also disclosed herein are reaction systems using a uridine diphosphate glycosyltransferase (UGTs) and a sucrose synthase (SuSy) to glycosylate a target substrate. High concentrations of sucrose (e.g., above 200 millimolar) have been shown to stabilize the folded (active) conformation of the UGTs and sucrose synthases by inhibiting the rate of protein unfolding. This is in tandem with the inherent stabilizing properties of enzyme immobilization [2, 3], Compared to immobilized UGTs alone, using immobilized UGTs and sucrose synthase increases the rate of glycosylating the therapeutic, reduces UDP-dependent UGT inhibition, and uses sucrose (an inexpensive reagent) as a substrate to create UDP- glucose in situ [4], In some embodiments, the UGT and sucrose synthase enzymes can be in separate immobilization vessels, which allows for the modular removal of one enzyme without disturbing the other, thereby allowing for hot swapping of UGTs (e.g., replacing one UGT for another UGT).

[0007] Enzyme immobilization can be readily scaled by using larger resin beds/columns and greater pumping/flow capacity. The most significant economic determinants in scaling- up are processing costs compared to alternatives [5],

[0008] Compared to methods of glycosylation using free enzymes (e.g., UGT and/or SuSy dissolved in solution), the methods described herein provide a more efficient enzymatic reaction, a more stable enzyme less prone to aggregation, reduce the amount of time required to generate new enzyme material, and reduce the amount of time required to manage the glycosylation reaction. Smaller amounts of inexpensive reagents are used, thereby reducing the cost of large-scale glycosylation.

[0009] Described herein is a method of glycosylating a compound. The method can include a) contacting an aqueous solution comprising a uridine diphosphate glycosyltransferase (UGT) having an affinity tag with an affinity resin so that the UGT becomes immobilized on the affinity resin; and b) contacting an aqueous solution comprising UDP-ghicose and a compound having a nucleophilic center with the affinity resin with UGT immobilized thereon; thereby glycosylating the compound.

[0010] Described herein is a method of glycosylating a compound. The method can include a) contacting an aqueous solution comprising a uridine diphosphate glycosyltransferase (UGT) having an affinity tag with an affinity resin so that the UGT becomes immobilized on the affinity resin; b) contacting a solution comprising a sucrose synthase (SuSy) having an affinity tag with an affinity resin so that the SuSy become immobilized on the affinity resin; c) contacting an aqueous solution comprising uridine diphosphate (UDP), sucrose, and a compound having a nucleophilic center with the affinity resin with UGT immobilized thereon and with the affinity resin having the SuSy immobilized thereon; thereby glycosylating the compound.

[0011] The affinity tag can be a metal affinity tag and the affinity resin can be a metal affinity resin. The metal affinity resin can include nickel or manganese. The metal affinity tag can be a poly-histidine tag. The metal affinity tag can be on the C-terminus of the UGT or the SuSy.

[0012] The aqueous solution with the UDP -glucose and the compound can be incubated with the affinity resin with UGT immobilized thereon. The aqueous solution with the UDP- glucose and the compound can be flowed over the affinity resin with UGT immobilized thereon.

[0013] The UGT and SuSy can be immobilized on separate affinity resins.

BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The foregoing will be apparent from the following more particular description of example embodiments, as illustrated in the accompanying drawings in which like reference characters refer to the same parts throughout the different views. The drawings are not necessarily to scale, emphasis instead being placed upon illustrating embodiments.

[0015] FIGs. 1 A-B are representations of a fixed bed immobilized enzyme reactor. FIG. 1 A is an embodiment with immobilized UGTs. The black dots represent the enzyme fixed to a medium and packed into a column. Substrate and product are circulated from the reaction vessel through the column FIG. IB is an embodiment with multiple columns placed in series within the circulating medium, allowing for sequential reactions in the same system. The second enzyme, represented by orange dots, performs a second reaction, and allows for a different mixture of reagents to create the same product. For example, the first column can be immobilized UGT and the second column can be immobilized SuSy.

[0016] FIGs. 2A-B show that the immobilized enzymes are at least as active as their nonimmobilized counterparts. FIG. 2 A is a chart showing that sucrose synthase from glycine max remains highly active for at least one week or more while affixed to nickel-sepharose beads. UDP-Glc concentrations produced per cycle fluctuated about 100 pM between days. The average rate of UDP-Glc production in mg/(mL * min) using this method was approximately 0.02 (+/- 25%). On average, this is approximately as active as the nonimmobilized enzyme but benefits from the same enzyme stock being able to perform multiple days of reactions. FIG. 2B is a chart showing that the UGT from Streptomyces antibioticus is more active than the immobilized counterpart and remains so for days. The enzyme remains highly active, near starting levels, for more than a week.

[0017] FIG. 3 shows that UGT from Bacillus cereus remains active for at least 5 days while affixed to nickel-sepharose beads. Complete conversion of etoposide to etoposide-3"- O-D-glucoside is seen after 5 days. Trace (1) is the sample loaded onto the column at the beginning of the fifth day while trace (2) is the same reaction assessed by HPLC the next morning. Elution at approximately 4.7 minutes is etoposide-3"-O-D-glucoside. Elution at approximately 5.25 minutes is etoposide.

[0018] FIG. 4 shows that sucrose synthase and a UGT are a highly active pair when placed in series such as the UGT from UGT from Streptomyces antibioticus (DRB0753; SEQ ID NO: 2). After a single day, all the available parent compound was converted into ivacaftor-5-di-O-D-glucoside and ivacaftor-5-tri-O-D-glucoside (trace (4), retention time 3.7 minutes and 3.6 minutes respectively) which is much faster than the rate seen with immobilized UGT alone, where the majority peak was ivacaftor-5-di-O-D-glucoside (trace (3), retention time 3.7 minutes). Trace (2) shows the elution retention time for ivacaftor-5- mono-O-D-glucoside (retention time 4.2 minutes). Trace (1) shows the elution retention time for ivacaftor (retention time 5.15 minutes).

[0019] FIG. 5A-B show that the immobilized UGT/SuSy pair perform better than the equivalent reaction involving soluble, non-immobilized enzymes. FIG. 5A is a chromatograph showing the production of ivacaftor glucoside after one day of sustained glycosylation reaction. Trace (1) shows the elution retention time for the parent compound ivacaftor (retention time 5.15 minutes). Trace (2) is the product profile of the nonimmobilized enzyme reaction. The vast majority of ivacaftor has been converted to the ivacaftor-5-di-O-D-glucoside state (retention time 3.7 minutes) with a small population of ivacaftor-5-tri-O-D-glucoside (retention time 3.6 minutes). Trace (3) is the product profile of the immobilized enzyme reaction. A much larger portion of total compound has gone to the triglycosylated state in Trace (3) (46.23%) compared to Trace (2) (16.97%). FIG. 5B is a chromatograph of the same samples after an additional day of reaction. Both samples have continued to produce triglycosylated ivacaftor, yet the non-immobilized free enzyme reaction (Trace (2)) has still not produced the amount of triglycosylated compound seen in the immobilized enzyme sample (Trace (3), 41.55% versus 54.54%) or even the previous day’s immobilized enzyme sample (FIG. 5A Trace (3), 41.55% versus 46.23%).

DETAILED DESCRIPTION

[0020] A description of example embodiments follows.

Overview

[0021] In industrial settings, the choice to use free soluble enzyme versus enzymes in an insoluble or immobilized form can depend on the application. Limitations such as reaction volumes or a specific enzyme’s physical attributes can influence whether to use a free soluble enzyme or an immobilized enzyme. Immobilizing enzymes on a fixed bed can be advantageous because the enzymes can be recovered and reused, and the enzymes can be stabilized enough to maintain activities longer than in soluble form. While immobilization may affect the activity or selectivity of an enzyme, depending on the immobilization method employed, simple and cost-effective methods are often preferred in industrial settings. Compared to chemical catalysts, enzymatic synthesis methods often have much higher overall synthesis yields and lack the need for dangerous organic solvents. Compared to whole cell processes, the methods described herein do not require keeping microbes in optimal living conditions, nor do they require engineering transporters to move the therapeutic in- and-out of the cell. Compared to soluble enzymes, the methods described herein include ways of potentially stabilizing the enzyme and/or increasing its activity.

[0022] However, the potential benefits of enzyme immobilization are hard to predict and can be outweighed by downsides, making some enzymes unsuitable for industrial-scale immobilization. Depending on the enzyme, enzyme activity when immobilized can be significantly less than when dissolved due to the restricted movement of the immobilized enzyme. Industrial considerations include the costs for the immobilization medium and the labor involved with its safe disposal. The upfront investments are another important factor: immobilized enzyme reactors in the food industry can be scaled to very high-throughput, able to move multiple tons per hour through a column bed and produce thousands of kilograms of product per kilogram of immobilized enzyme, but at a high initial cost (large physical bed volumes, pressure safety issues, etc). In the pharmaceutical industry, reactions are typically performed at a smaller scale, but the reaction system can still scale from hundreds of milligrams to hundreds of kilograms per kilogram of enzyme. [0023] One example in an industrial setting is immobilization of glucose isomerases during production of high-fructose corn syrup. The glucose isomerase can be recycled, which reduces cost and labor [6-9], While sugar-active enzymes, such as glucose isomerases, have been used in industrial settings, use of immobilized UGT has not been widely reported, but are generally limited to only a small subset of very specialized UGTs [10, 11], Knowledge on the benefits of UGT immobilization and its effect on glycosyltransferase reactions is therefore limited.

Glycosylation

[0024] A potential strategy for improving or modulating the efficacy, safety, and/or PK/PD profile of a small molecule-based therapeutic is modification by glycosylation. The small molecule, or aglycone, is modified by the addition of one or more sugar groups or chains of two or more sugar groups (called oligosaccharides) to nucleophilic centers of the aglycone. These sugar groups can be naturally occurring sugars such as glucose, fructose, rhamnose, mannose, galactose, fucose, xylose, arabinose, glucuronic acid, or N- acetylglucosamine, or they can be synthetically synthesized sugars (e.g., 6-Br-D-glucose, 2- deoxy-D-glucose, 5 -thio-D-glucose). These sugars can be attached to the small molecule or to other sugar groups by either an alpha or beta glycosidic bond.

[0025] In general, glycosylation of a small molecule can lead to increased aqueous solubility, altered interactions with proteins and membranes, altered absorption and excretion, changes in metabolic stability, and other changes in PK/PD characteristics [12-14], [0026] Glycosylation can enhance or block the transport of a glycoside into specific tissues or organs. Glycosylation can enhance uptake through interaction between the glycoside moiety and lectins or glucose transporters on the cell surface.

[0027] In some cases, glycosylation alters the pharmacological activity of the drug, either by enhancing or decreasing potency or even by changing the mechanism of action [12-14], [0028] The identity of the sugar and the stereochemistry of the glycosidic bond can also affect the pharmacological activity or PK/PD profile of a glycoside.

[0029] Glycosylation is also a potential strategy for developing prodrugs and compounds for targeted drug delivery to specific tissues. Glycosidases are enzymes that catalyze the hydrolysis of glycosidic bonds and that are specifically expressed in different tissues and organs including blood plasma, the colon, the intestines, and the gut microflora. Glycosidases exhibit substrate specificity towards different glycosidic bond stereochemistry or towards different monosaccharides. A glycosylated drug could function as a prodrug or as a targeted drug if it is preferentially cleaved by a tissue-specific glycosidase. This has been demonstrated by Zipp et al: the alpha-glycosidic bonds in cannabinoid glycosides have been shown to be preferentially cleaved by glycosidases present in the large intestine of mice and not by other chemical or enzymatic processes that may be present in the small intestine, stomach, blood plasma, or brain [15, 16],

[0030] In summary, glycosylation of a small molecule may improve aqueous solubility, but can also alter interactions with proteins and membranes, pharmacological activity, and/or PK/PD characteristics in ways that are unexpected.

Glycosyltransferases

[0031] Traditional methods for glycosylating small molecules are non-selective, and it is particularly difficult to control the stereo- and regiospecificity of glycosylation [17, 18], There is often more than one position on the aglycone that will react with the reagent used to make the desired modification. This makes it necessary to chemically ‘block’ or render temporarily unreactive, the other positions on the molecule in order to selectively modify the desired position. A typical modification will require multiple protection and de-protection steps using the standard methods of synthetic organic chemistry.

[0032] Glycosyltransferases (GTs) are a class of enzymes with the potential to act as the catalyst for the generation of novel glycosylated therapeutic small molecules. GTs catalyze the transfer of a sugar from an activated sugar donor molecule to an acceptor molecule (Lairson et al. 2008). They are a large and well-characterized family found in viruses, archaea, bacteria, and eukaryotes. Greater than 600,000 GTs categorized into approximately 110 families are described in the Carbohydrate-active Enzymes Directory (www.cazy.org), and greater than 150 GT structures are reported (www.rcsb.org) [19, 20], The majority of GTs utilize nucleotide-activated sugar donors and are referred to as Leloir GTs, although lipid phosphate and phosphate-activated sugar donors are also used [21, 22], GT acceptors include proteins, lipids, oligosaccharides, and small molecules.

[0033] GTs offer several advantages as a potential tool in a general small molecule glycosylation platform [12, 14, 23, 24], GTs are often characterized by very high conversion efficiencies (up to 100%). As a result, lower concentrations of potentially expensive or difficult to synthesize substrates are required for GT-catalyzed reactions. GTs are able to glycosylate a wide variety of acceptor structures, with many GTs exhibiting promiscuity towards the sugar donor and acceptor. Furthermore, GTs can catalyze the formation of O-, N- , S-, and even C-gly cosides. As a result of these characteristics, GTs are generally amenable to both in vitro and in vivo bioengineering efforts.

Uridine diphosphate GTs (UGTs)

[0034] Uridine diphosphate GTs (UGTs) utilize uridine diphosphate (UDP) sugar donors, and form the largest group of Leloir GTs in plants [23], Recently, the identification and characterization of new UGTs, especially in plants and bacteria, has exploded as part of an increased interest in characterizing natural product biosynthetic pathways. This method is described by Torens-Spence et al. [25], In this paper, 33 UGT enzyme-encoding genes were cloned from a Golden root plant, expressed in yeast, and screened for regiospecific activity in modifying tyrosol to produce salidroside or icariside D2, which are tyrosol metabolites in the plant’s native salidroside biosynthetic pathway. Another group identified naturally occurring enzymes having promiscuous N- and O- glycosyltransferase activity by mining the expressed genes of Carthamus tinctorius. Xie et al. describes the identification of a promiscuous glycosyltransferase (UGT71E5) from C. tinctorius which contains A-glycosylase activity towards multiple diverse nitrogen-heterocyclic aromatic compounds [26], Zhang et al. describes the identification of three new UGTs (UGT 84A33, UGT 71 AE1 and UGT 90A14) from C. tinctorius having promiscuous ( -gly cosy 1 transferase activity against benzylisoquinoline alkaloids and their use in making glycosylated derivatives [27], With the continuing technological improvements and decreasing costs of genome and transcriptome sequencing and analysis, it is becoming easier to identify and characterize naturally occurring GTs for the development of novel small molecule diversity generating platforms.

[0035] UGTs are described in international patent applications PCT/US2021/022416, PCT/US2021/022410, and PCT/US2021/022414. UGTs can be produced by cultivation of a microorganism having a gene encoding for and expressing a UGT enzyme derived from a strain including, but not limited to, those species belonging to the Slreplomyces. Bacillus, Glycine, or Galega genera.

[0036] UGTs can catalyze the addition of many different monosaccharide from an activated nucleotide sugar (also known as the "glycosyl donor") to a nucleophilic glycosyl acceptor molecule, the nucleophile of which can be oxygen- carbon-, nitrogen-, or sulfurbased. In general, the target compound must have at least one nucleophilic center available for glycosylation. Monosaccharides, Disaccharides, Trisaccharides, and Oligosaccharides

[0037] Glycosyltransferases can catalyze the addition of many different monosaccharides to nucleophilic centers of the aglycone. In general, suitable monosaccharides include, but are not limited to, open and closed chain monosaccharides. The monosaccharides can be in the L- or D- configuration. Typically, the monosaccharides have 5, 6, or 7 carbons (a pentose monosaccharide, hexose monosaccharide, or heptose monosaccharide, respectively).

[0038] Suitable monosaccharides include allose, apiose, arabinose, fructose, fucitol, fucose, galactose, glucose, glucuronic acid, mannose, 7V-acetylglucosamine, N- acetylgalactosamine, rhamnose, and xylose. Other suitable monosaccharides include glucosamine, galactosamine, mannosamine, 5-thio-D-glucose, nojirimycin, deoxynojirimycin, 1,5-anhydro-D-sorbitol, 2,5-anhydro-D-mannitol, 2-deoxy-D-galactose, 2- deoxy -D-glucose, 3 -deoxy -D-glucose, arabinitol, galactitol, glucitol, iditol, lyxose, mannitol, L-rhamnitol, 2-deoxy-D-ribose, ribose, ribitol, ribulose, xylulose, altrose, gulose, idose, levulose, psicose, sorbose, tagatose, talose, galactal, glucal, fucal, rhamnal, arabinal, xylal, 3,4-di-O-acetyl-L-fucal, 3,4-di-O-acetyl-L-rhamnal, 3,4-di-O-acetyl-D-arabinal, 3,4-di-O- acetyl-D-xylal, valienamine, validamine, valiolamine, valienol, valienone, galacturonic acid, mannuronic acid, /'/-acetylneuraminic acid, A -acetyl muramic acid, gluconic acid D-lactone, galactonic acid gamma-lactone, galactonic acid delta-1 actone, mannonic acid gamma-lactone, D-altro-heptulose, D-manno-heptulose, D-glycero-D-manno-heptose, D-glycero-D-gluco- heptose, D-allo-heptulose, D-altro-3 -heptulose, D-glycero-D-manno-heptitol, and D-glycero- D-altro-heptitol.

[0039] Suitable oligosaccharides include, but are not limited to, carbohydrates having from 2 to 10 or more monosaccharides linked together (e.g., 2, 3, 4, 5, 6, 7, 8, 9, or 10 monosaccharides linked together). The constituent monosaccharide unit may be, for example, a pentose monosaccharide, a hexose monosaccharide, or a pseudosugar (including a pseudoamino sugar). Oligosaccharides do not include bicyclic groups that are formed by fusing a monosaccharide to a benzene ring, a cyclohexane ring, or a heterocyclic ring. Pseudosugars that may be used in the invention are members of the class of compounds wherein the ring oxygen atom of the cyclic monosaccharide is replaced by a methylene group. Pseudosugars are also known as “carba-sugars.”

[0040] The glycosyltransferases can catalyze addition of a monosaccharide to a target compound, and the bond between the monosaccharide and target can be either an alpha or beta glycosidic bond. Disaccharides, trisaccharides, and oligosaccharides are formed by serial enzymatic additions of two or more monosaccharides to the target compound. When more than one monosaccharide is added by serial enzymatic reactions, successive monosaccharides can be bonded to the preceding monosaccharide by either an alpha or beta glycosidic bond.

Sucrose Synthase

[0041] Sucrose synthase (SuSy) is a glycosyl transferase belonging to the GT4 subfamily of glycosyltransferases (GTs). SuSy catalyzes the reversible cleavage of sucrose into fructose and either uridine diphosphate glucose (UDP-Glc) or adenosine diphosphate glucose (ADP- G). The enzyme plays a key role in sugar metabolism with its products being found in many metabolic pathways [28], Of the sucrose-cleaving enzymes, it is the only one capable of catalyzing both the synthesis and cleavage of sucrose in a reversible manner while remaining almost energy neutral [29-31], SuSy regulates sucrose flux within the cell by rapidly altering its cellular location from the cytosol to the plasma membrane and/or plastids [32-35], In some plants, more than one SuSy isoform is expressed (e.g., six different are known to exist within Arabidopsis thaliana) [36, 37],

[0042] In recent years, SuSy has been used to generate UDP-Glc, which can be expensive, from UDP and sucrose, which are relatively inexpensive [38-40], The conversion of a bulk of inexpensive chemicals to a high-priced reagent in a one-step reaction provides a key aspect for industrial interest. SuSy has gained interest as an industrial biocatalyst because of the ability to couple the synthesis of UDP-Glc to the glycosylation of a target compound by including a sucrose synthase within a UGT reaction [38-40], However, this coupled enzyme reaction has mostly been used in the food and agricultural industries for the synthesis of sweetener compounds or natural product supplements like nothofagin, rebaudiosides, davidiosides, confusosides, or ginsenoside [41],

Immobilized Enzyme Reactors

[0043] FIG. 1A is a schematic of a fixed bed immobilized enzyme reactor (e.g., a chromatography column). A container 110, which is preferably a stirred container, contains a reaction medium. The reaction medium circulates from the container 110 to the enzyme bed 120 and back to the container 110. The reaction medium includes substrates 115 for glycosylation in the enzyme bed 120. Typically, the substrates are therapeutic compound. The enzyme bed 120 includes enzymes 125 that are fixed to a medium, such as a nickelsepharose resin. The enzymes convert the substrate 115 into a product 135. Typically, the fixed bed is in the form of a column.

[0044] FIG. IB is a schematic of a reaction system having multiple columns in series, allowing for sequential reactions in the same system. The reaction medium circulates from the container 110, to a first enzyme bed 120a (e.g., having UGTs (enzymes 125a)), to second enzyme bed 120b (e.g., having sucrose synthase (enzumes 125b) , and back to the container 110. The reaction medium includes substrates 115 for glycosylation in the enzyme bed 120. The enzymes 125a and 125b convert the substrate 115 into a product 135.

[0045] A liquid medium comprising an enzyme of interest (either a UGT or an SuSy) is recirculated through the fixed bed until enough enzyme has associated with the resin. In some cases, this is a solution of purified enzyme containing only the enzyme and a stabilizing salt buffer. In others, the enzyme remains in the host cell’s lysate until passing through the column, without prior need to purify the enzyme before the immobilization process. The enzyme containing liquid is pumped through the resin until the column is “primed” with enzyme. Multiple enzyme-containing columns, primed with separate enzymes, can be linked in series to perform sequential reactions within the same system. This also allows for modularity such that if a different UGT is needed for a new small molecule to be glycosylated, primed columns can easily be swapped out for one another without disrupting the rest of the system. Once the column is primed, a solution containing the reagents needed for the glycosylation reaction are pumped over the column which allows substrates to contact the enzyme and efficient glycosylation to occur.

[0046] To form the lysate, UGT-expressing cells (e.g., UGT-expressing yeast cells) are lysed and the insoluble part is discarded by centrifugation so that the lysate is cell-free. In other embodiments, the cell-free lysate is not required. For example, in some embodiments, recombinant UGTs can be used. In other embodiments, purified UGTs can be used. Similar procedures can be applied to generate a lysate for the sucrose synthase.

[0047] The resin (e.g., a nickel-sepharose resin) non-covalently associates with an affinity tag (e.g., a metal affinity tag) at the N-terminus or C-terminus of the UGT or SuSy. The affinity tag of the enzyme non-covalently associates with the resin bed so that the enzyme can be recovered as needed without requiring additional chemical treatment or cleavage steps. [0048] In the examples described herein, a metal affinity tag is used. Metal affinity tags are well-known in the art. One example of a metal affinity tag is a poly-histidine tag, which can be in the form of approximately sixteen histidine amino acids residues at the N-terminus or C-terminus of the UGT or SuSy. Other types of tag-and-resin combinations are also suitable. The methods do not involve covalently crosslinking an enzyme to a support material.

[0049] Leaching of enzyme from the resin during a prolonged reaction process is deemed largely irrelevant because it is readily removed during downstream purification.

[0050] An additional advantage of the immobilized enzyme reactors is that they can be reused without losing much efficiency. In a non-immobilized (free enzyme) reaction, the target compound may be 90+% glycosylated within about one day. At that time, the enzymes are difficult to separate and recover for use in further glycosylation reactions. In contrast, the immobilized enzyme reactors can be reused, and the reaction can proceed with approximately similar efficiency.

[0051] One of skill in the art will appreciate that additional features, such as a filter or pump, can be placed in series with the enzyme bed(s).

Immobilized Uridine Diphosphate Glycosyltransferases (UGTs)

[0052] The devices and methods described herein relate to immobilized UGTs for glycosylating compounds, such as small-molecule therapeutics. Many different UGTs can be used, and each UGT can glycosylate many different compounds. Examples of glycosylation of small molecule compounds are disclosed in international patent applications

PCT/US2021/022416, PCT/US2021/022410, and PCT/US2021/022414, which disclose glycosylation of ivacaftor, enasidenib, and etoposide, respectively.

[0053] Compared to dissolved UGTs, immobilized UGTs for glycosylation of smallmolecule therapeutics can lead to increased rates of catalysis and an easy setup for the recovery of the enzyme for further reactions. Highly insoluble or hydrophobic drugs often need large volumes of buffered reaction mixture in order to glycosylate them. By passing the drug-containing reaction mixture through a small, fixed resin bed loaded with UGT, the drugcontaining mixture passes through a very high effective concentration of enzyme, ensuring faster conversion rates with relatively low amounts of enzyme, despite an overall large reaction volume. The added benefit of a generally more stable enzyme, due to immobilization and/or dissolved sucrose, means that overall, immobilization of a UGT can be beneficial in manufacturing of glycosylated pharmaceuticals.

[0054] After a period of time (e.g., from 1 to 72 hours), the compound is converted to a monosaccharide, disaccharide, trisaccharide, or oligosaccharide thereof.

Immobilized Sucrose Synthase

[0055] Other embodiments relates to devices and methods with an immobilized sucrose synthase (SuSy) enzyme in combination with an immobilized UGT enzyme. Compared to an embodiment with only an immobilized UGT, an embodiment with an immobilized UGT and an immobilized SuSy glycosylates the small molecule therapeutic with one or more of an increased rate of reaction, an increased extent of reaction, and by using less expensive reagents because UDP-Glc is substituted for low amounts of UDP plus inexpensive sucrose. The combination sucrose synthase and sucrose allows for rapid and cost-effective production of UDP-glucose, a primary substrate for glycosylation by UGTs.

[0056] In some embodiments, the UGT and the SuSy are immobilized on the same resin. In other embodiments, the UGT and SuSy are immobilized on separate resins which are configured in series, as in FIG. IB.

[0057] Sucrose provides an additional benefit of stabilizing the active conformation of UGTs and sucrose synthases. While glycerol and glucose are commonly used as cryoprotectants, sucrose can also act as a cryoprotectant in protein crystallography, stabilizing and protecting a protein as it is frozen. It is not seen as a protein stabilizer while immobilizing enzymes.

EXEMPLIFICATION

Example 1: Immobilized sucrose synthase remains stable for at least one week.

[0058] 500 micrograms of recombinantly expressed sucrose synthase from plant Glycine max (GmSuSy; SEQ ID NO: 5) with a C-terminal poly-histidine tag (SEQ ID NO: 4) was loaded onto a 3 mL column containing nickel-Sepharose resin using gravity flow. A stock of 500 millimolar sucrose combined with 1 millimolar UDP was made to be used as substrate. A sample consisting of 1 column volume (3 mL) containing the UDP and sucrose was flowed over the resin and collected after 10 minutes of contact. Four samples were collected each day. There was a period of 30 minutes between each sample incubation, column washing, and reloading of fresh sample. Each cycle eluent was analyzed for fructose concentration to give effective monitoring of UDP-Glc production in mg/(mL * min). Sucrose synthase activity was monitored in this way for 1 week with negligible loss in enzyme activity (FIG.

2A).

Example 2: Immobilized UGTs remain active for days throughout repeated etoposide reaction.

[0059] 2 milligrams of recombinantly expressed UDP -glycosyltransferase from the bacterium Streptomyces antibioticus (DRB0753; SEQ ID NO: 2) with a C-terminal polyhistidine tag (SEQ ID NO: 4) was loaded onto a nickel-sepharose resin as in Example 1. UDP-Glc and etoposide were used as substrates. The UGT remains active for at least 5 days with minor loss in activity (FIG. 2B). The formation of etoposide-3 "-O-D-glucoside by the enzymatic reaction was monitored by HPLC for 5 days, each day comparing the substrates prior to exposure to the column (FIG. 3, Trace (1)) to the reaction mix on the following morning (FIG. 3, Trace (2), Day 5). All etoposide has been converted to a glucoside form. Elution at approximately 4.7 minutes is etoposide-3 "-O-D-glucoside. Elution at approximately 5.25 minutes is etoposide.

Example 3: Sucrose synthase and UGTs together perform better with ivacaftor than those UGTs alone.

[0060] GmSuSy and a UGT from Bacillus cereus (DRB0754; SEQ ID NO: 3), each having a with a C-terminal poly-histidine tag (SEQ ID NO: 4), were immobilized onto two separate columns as in Examples 1 and 2 and then placed in series such that eluent from Column 1, containing GmSuSy, flows directly into Column 2, containing DRB0754. Coupling the reactions greatly increased the efficiency of the glycosylation system. After a single day, non-glycosylated ivacaftor (FIG. 4, Trace (1)) was completely converted into ivacaftor-5-di-O-D-glucoside and ivacaftor-5-tri-O-D-glucoside when using immobilized SuSy and UGT together (FIG. 4, Trace (4)). Ivacaftor-5-tri-G-D-glucoside, which has a longer sugar chain requiring multiple rounds of glycosylation, was found in barely detectable amounts in the reaction using immobilized UGT alone (FIG. 4, Trace (3)). While a small population of monoglycosylated ivacaftor was still present in the UGT-only preparation (FIG. 4, Trace (3) with Trace (2) as a reference standard), none was detected in the coupled enzyme reaction. Elution at approximately 3.6 minutes is ivacaftor-5-tri-O-D-glucoside. Elution at approximately 3.7 minutes is ivacaftor-5-di-O-D-glucoside. Elution at approximately 4.2 minutes is ivacaftor-5-mono-O-D-glucoside. Elution at approximately 5.15 minutes is ivacaftor.

Example 4: The immobilized UGT/SuSy system performs better than the nonimmobilized counterparts.

[0061] GmSuSy and a UGT from Bacillus cereus (DRB0754; SEQ ID NO: 2) were immobilized and the same procedure as Example 3 was performed, again using ivacaftor as a substrate to form the products ivacaftor-5-di-O-D-glucoside (retention time 3.7 minutes) and ivacaftor-5-tri-O-D-glucoside (retention time 3.6 minutes). A reaction mixture involving nonimmobilized UGT and SuSy performed the reaction for an equal amount of time and produced far less triglycosylated compound (FIG. 5A, Trace (2), 16.97% of total product signal) compared to the immobilized preparation (FIG. 5A Trace (3), 46.23% of total product signal). After a further 24 hours, the non-immobilized enzyme mixture (FIG. 5B, Trace (2)) still lagged behind the immobilized reaction (FIG. 5B Trace (3)) and in fact still had not reached the 46.23% triglycosylated ivacaftor that was achieved by the immobilized enzyme preparation.

REFERENCES

[0062] 1. Homaei, A.A., et al., Enzyme immobilization: an update. J Chem Biol, 2013. 6(4): p. 185-205.

[0063] 2. Haas, D.J. and M.G. Rossmann, Crystallographic studies on lactate dehydrogenase at-75 degrees C. Acta Cry stall ogr B, 1970. 26(7): p. 998-1004.

[0064] 3. Moehlenbrock, M.J. and S.D. Minteer, Introduction to the Field of Enzyme Immobilization and Stabilization. Methods Mol Biol, 2017. 1504: p. 1-7.

[0065] 4. Schmblzer, K., et al., Sucrose synthase: A unique glycosyltransferase for biocatalytic glycosylation process development. Biotechnol Adv, 2016. 34(2): p. 88- 111.

[0066] 5 Weetail, H.H. and W.H. Pitcher, Jr., Scaling up an immobilized enzyme system. Science, 1986. 232(4756): p. 1396-403.

[0067] 6. Basso, A. and S. Serban, Industrial applications of immobilized enzymes — A review. Molecular Catalysis, 2019. 479: p. 110607.

[0068] 7. Tate & Lyle Introduces DOLCIA PRIMA® Crystalline Allulose. 2017, Tate & Lyle PLC. [0069] 8. Zittan, L., P.B. Poulsen, and S.H. Hemmingsen, Sweetzyme — A New Immobilized Glucose Isomerase. Starch - Starke, 1975. 27(7): p. 236-241.

[0070] 9. Jorgensen, O.B., et al., A New Immobilized Glucose Isomerase with High Productivity Produced by a Strain of Streptomyces murinus. Starch - Starke, 1988. 40(8): p. 307-313.

[0071] 10. Nedwin, G.E.D., CA, US), Sharma, Vivek (North Liberty, IA, US), Shetty, Jayarama K. (Pleasanton, CA, US), Alpha-amylase blend for starch processing and method of use thereof. 2013, Danisco US Inc. (Palo Alto, CA, US): United States.

[0072] 11. Kiiski, I., et al., Microfluidic Immobilized Enzyme Reactors for Prediction of Drug Clearance In Vivo. 2019.

[0073] 12. Gantt, R.W., P. Peltier-Pain, and J.S. Thorson, Enzymatic methods for glyco(diversification/randomization) of drugs and small molecules. Nat. Prod. Rep., 2011. 28(11): p. 1811-1853.

[0074] 13. Kfen, V., Glycoside vs. Aglycon: The Role of Glycosidic Residue in Biological Activity. Glycoscience, 2008: p. 2589-2644.

[0075] 14. De Bruyn, F., et al., Biotechnological advances in UDP -sugar based glycosylation of small molecules. Biotechnol. Adv., 2015. 33(2): p. 288-302.

[0076] 15. Zipp, B.J., J.M. Hardman, and R.T. Brooke, Cannabinoid glycoside prodrugs and methods of synthesis, in Patent. 2018.

[0077] 16. Hardman, J.M., R.T. Brooke, and B.J. Zipp, Cannabinoid glycosides: In vitro production of a new class of cannabinoids with improved physicochemical properties. 2017.

[0078] 17. Zhu, X. and R.R. Schmidt, New principles for glycoside-bond formation. Angew. Chem. Int. Ed Engl., 2009. 48(11): p. 1900-1934.

[0079] 18. Gu, X., et al., Direct glycosylation of bioactive small molecules with glycosyl iodide and strained olefin as acid scavenger. J. Org. Chem., 2014. 79(3): p. 1100- 1110.

[0080] 19. Lombard, V., et al., The carbohydrate-active enzymes database (CAZy) in 2013. Nucleic Acids Res., 2014. 42(D1): p. D490-D495.

[0081] 20. Berman, H.M., The Protein Data Bank. Nucleic Acids Research, 2000. 28(1): p. 235-242. [0082] 21. Breton, C., S. Fournel-Gigleux, and M.M. Palcic, Recent structures, evolution and mechanisms of glycosyltransferases. Curr. Opin. Struct. Biol., 2012. 22(5): p. 540- 549.

[0083] 22. Lairson, L.L., et al., Glycosyltransferases: Structures, Functions, and Mechanisms. Annu. Rev. Biochem., 2008. 77(1): p. 521-555.

[0084] 23. Yonekura-Sakakibara, K. and K. Hanada, An evolutionary view of functional diversity in family 1 glycosyltransferases. Plant J., 2011. 66(1): p. 182-193.

[0085] 24. Schmid, J., et al., Bacterial Glycosyltransferases: Challenges and Opportunities of a Highly Diverse Enzyme Class Toward Tailoring Natural Products. Front. Microbiol., 2016. 7: p. 182.

[0086] 25. Torrens- Spence, M.P., et al., Complete Pathway Elucidation and Heterologous Reconstitution of Rhodiola Salidroside Biosynthesis. Mol. Plant, 2018. 11(1): p. 205- 217.

[0087] 26. Xie, K., et al., EnzymaticN-Glycosylation of Diverse Arylamine Aglycones by a Promiscuous Glycosyltransferase fromCarthamus tinctorius. Adv. Synth. Catal., 2017. 359(4): p. 603-608.

[0088] 27. Zhang, Y., et al., Enzymatic biosynthesis of benzylisoquinoline alkaloid glycosides via promiscuous glycosyltransferases from Carthamus tinctorius. Chin. Chem. Lett., 2019. 30(2): p. 443-446.

[0089] 28. Stein, O. and D. Granot, An Overview of Sucrose Synthases in Plants. Frontiers in Plant Science, 2019. 10(95).

[0090] 29. Koch, K., Sucrose metabolism: regulatory mechanisms and pivotal roles in sugar sensing and plant development. Curr Opin Plant Biol, 2004. 7(3): p. 235-46.

[0091] 30. Salerno, G.L. and L. Curatti, Origin of sucrose metabolism in higher plants: when, how andwhy? Trends Plant Sci, 2003. 8(2): p. 63-9.

[0092] 31. Winter, H. and S.C. Huber, Regulation of sucrose metabolism in higher plants: localization and regulation of activity of key enzymes. Crit Rev Biochem Mol Biol, 2000. 35(4): p. 253-89.

[0093] 32. Hennen-Bierwagen, T.A., et al., Proteins from multiple metabolic pathways associate with starch biosynthetic enzymes in high molecular weight complexes: a model for regulation of carbon allocation in maize amyloplasts. Plant Physiol, 2009. 149(3): p. 1541-59. [0094] 33. Cai, G., et al., Distribution of callose synthase, cellulose synthase, and sucrose synthase in tobacco pollen tube is controlled in dissimilar ways by actin fdaments and microtubules. Plant Physiol, 2011. 155(3): p. 1169-90.

[0095] 34. Fujii, S., T. Hayashi, and K. Mizuno, Sucrose synthase is an integral component of the cellulose synthesis machinery. Plant Cell Physiol, 2010. 51(2): p. 294-301.

[0096] 35. Haigler, C.H., et al., Carbon partitioning to cellulose synthesis. Plant Mol Biol, 2001. 47(1-2): p. 29-51.

[0097] 36. Kleczkowski, L.A., S. Kunz, and M. Wilczynska, Mechanisms ofUDP-glucose synthesis in plants. Critical reviews in plant sciences, 2010. 29(4): p. 191-203.

[0098] 37. Bieniawska, Z., et al., Analysis of the sucrose synthase gene family in Arabidopsis. Plant J, 2007. 49(5): p. 810-28.

[0099] 38. Schmolzer, K., et al., Sucrose synthase: A unique glycosyltransferase for biocatalytic glycosylation process development. Biotechnology Advances, 2016. 34(2): p. 88-111.

[00100] 39. Dai, L., et al., One-Pot Synthesis of Ginsenoside Rh2 and Bioactive Unnatural

Ginsenoside by Coupling Promiscuous Glycosyltransferase from Bacillus subtilis 168 to Sucrose Synthase. Journal of Agricultural and Food Chemistry, 2018. 66(11): p. 2830-2837.

[00101] 40. Hu, Y., et al., Biocatalytic synthesis of ginsenoside Rh2 using Arabidopsis thaliana glucosyltransferase-catalyzed coupled reactions. Journal of Biotechnology, 2020. 309: p. 107-112.

[00102] 41. Gutmann, A., et al., Glycosyltransferase cascades for natural product glycosylation: Use of plant instead of bacterial sucrose synthases improves the UDP- glucose recycling from sucrose and UDP. Biotechnol J, 2017. 12(7).

INCORPORATION BY REFERENCE; EQUIVALENTS

[00103] The teachings of all patents, published applications, and references cited herein are incorporated by reference in their entirety.

[00104] While example embodiments have been particularly shown and described, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the embodiments encompassed by the appended claims.

Claims

CLAIMS What is claimed is:

1. A method of glycosylating a compound, the method comprising: a) contacting an aqueous solution comprising a uridine diphosphate glycosyltransferase (UGT) having an affinity tag with an affinity resin so that the UGT becomes immobilized on the affinity resin; and b) contacting an aqueous solution comprising UDP -glucose and a compound having a nucleophilic center with the affinity resin with UGT immobilized thereon; thereby glycosylating the compound.

2. The method of claim 1, wherein the affinity tag is a metal affinity tag, and wherein the affinity resin is a metal affinity resin.

3. The method of claim 2, wherein the metal affinity resin comprises nickel.

4. The method of claim 2, wherein the metal affinity resin comprises manganese.

5. The method of claim 2, wherein the metal affinity tag is a poly-histidine tag.

6. The method of claim 2, wherein the metal affinity tag is C-terminal of the UGT.

7. The method of claim 1, wherein the aqueous solution comprising the UDP-glucose and the compound is incubated with the affinity resin with UGT immobilized thereon.

8. The method of claim 1, wherein the aqueous solution comprising the UDP-glucose and the compound is flowed over the affinity resin with UGT immobilized thereon.

9. A method of glycosylating a compound, the method comprising: a) contacting an aqueous solution comprising a uridine diphosphate glycosyltransferase (UGT) having an affinity tag with an affinity resin so that the UGT becomes immobilized on the affinity resin; b) contacting a solution comprising a sucrose synthase (SuSy) having an affinity tag with an affinity resin so that the SuSy become immobilized on the affinity resin; and c) contacting an aqueous solution comprising uridine diphosphate (UDP), sucrose, and a compound having a nucleophilic center with the affinity resin with UGT immobilized thereon and with the affinity resin having the SuSy immobilized thereon; thereby glycosylating the compound. The method of claim 9, wherein the UGT and SuSy are immobilized on separate affinity resins. The method of claim 9, wherein the affinity tag is a metal affinity tag, and wherein the affinity resin is a metal affinity resin. The method of claim 11, wherein the metal affinity resin comprises nickel. The method of claim 11, wherein the metal affinity resin comprises manganese. The method of claim 11, wherein the metal affinity tag is a poly-histidine tag. The method of claim 11, wherein the metal affinity tag is C-terminal of the UGT. The method of claim 11, wherein the metal affinity tag is C-terminal of the SuSy.