Chemical modification of protein preparations by amino acid modification reaction and protein preparations stabilized by chemical modification

Abstract

A tyrosine residue modification method has been developed, which includes: mixing a protein with a glycosylation agent having a sugar conjugated to 1-methyl-4-aryl-urazole (MAUra); and adding laccase to the mixture to modify a tyrosine residue with the MAUra compound. This method prevents protein aggregation and can provide stable protein preparations. The tyrosine modification method using laccase and MAUra is a mild method and enables easy glycosylation without any alteration of an amino acid sequence for sugar conjugation.

Description

Chemical modification of protein formulations by amino acid modification reactions, and protein formulations stabilized by chemical modification

[0001] The present invention relates to a method for stabilizing protein preparations by amino acid modification reactions, and to chemically modified protein preparations, particularly insulin preparations in which tyrosine is chemically modified.

[0002] Various proteins are used as pharmaceuticals, including proteins secreted in the body such as insulin and growth hormone, and antibodies against proteins specific to diseases such as cancer. Human insulin was the first recombinant protein drug approved in 1982. Since then, various genetically modified insulin analogs have been sold, including ultra-rapid-acting, rapid-acting, intermediate-acting, and sustained-acting types, which differ in duration of action and time to onset of effect.

[0003] Proteins, which are active ingredients, fold intramolecularly through hydrogen bonds and van der Waals forces resulting from the main and side chains of the protein, forming a three-dimensional structure. Proteins are typically stabilized in their native structure, but external conditions such as high temperature, pH, and light can cause them to transition to a mutated structure, known as a denatured state. Denaturation of proteins can lead to physical instability, aggregation, and reduced efficacy. Furthermore, protein formulations maintained at high concentrations for effective administration are prone to aggregation, which can lead to problems such as reduced physiological activity.

[0004] Several methods have been used to maintain the efficacy of protein formulations. Protein formulations can be stabilized by adding additives such as amino acids (e.g., arginine, glutamic acid) or sugars to improve hydrophilicity, or by maintaining a constant pH in a buffer solution to prevent aggregation or precipitation. Furthermore, protein formulations must be stored under certain conditions, such as refrigerated storage until use, to avoid exposure to high temperatures.

[0005] For example, it is recommended that insulin preparations be stored in the refrigerator (2-8°C) before opening, and at room temperature (below 30°C) after opening. Depending on the product, it should be used within four to eight weeks after opening. If the product is removed from the refrigerator and allowed to reach room temperature after each use, condensation will form on the injector, leading to malfunction, so it must be stored at room temperature after opening. It is also said that freezing in the refrigerator or exposure to temperatures above 37°C while stored at room temperature will cause the product to lose its effectiveness. Therefore, care must be taken when storing insulin preparations in the summer, even in countries with high average temperatures, such as Japan.

[0006] As mentioned above, many genetically modified insulin analogs have been produced. However, all commercially available insulin analogs are intended to modify the onset and duration of action. Meanwhile, insulin preparations that have been developed to stabilize protein formulations have also been disclosed (Patent Documents 1 to 7). Patent Documents 1 to 5 disclose methods for maintaining physical and chemical stability by controlling the pH and ionic environment using buffers and additives. Patent Documents 6 and 7 disclose insulin preparations that have high thermodynamic stability and are less prone to self-association even at high protein concentrations by modifying the amino acid sequence. Attempts to improve protein stability have also been made through hydration via glycosylation. Non-Patent Document 1 reports that insulin with various sugar structures attached at different binding positions was chemically synthesized, resulting in the synthesis of 12 glycosylated mutants, and the improved in vitro stability was confirmed. Non-Patent Document 2 reports that introducing a cysteine, which serves as a scaffold for glycosylation, into the N-terminus of the B chain by chemical synthesis and then glycosylation can maintain hypoglycemic activity while reducing aggregation. Furthermore, Non-Patent Document 3 discloses introducing a thiol to the N-terminus of the B chain by chemical modification, and achieving glycosylation through a disulfide bond using sialic acid bound to the thiol. Non-Patent Document 4 discloses glycosylation by crosslinking a primary amine-linked branched oligosaccharide with transglutaminase to a mutant insulin in which phenylalanine at the N-terminus of the B chain has been replaced with glutamine by genetic recombination. All of the methods disclosed in Non-Patent Documents 1 to 4 have shown that excellent stability was achieved by glycosylation.

[0007] Japanese Patent Application Laid-Open No. 2015-71608 Special Publication No. 2017-502052 Japanese Patent Application Laid-Open No. 2020-73534 Special Publication No. 2002-500196 Special Publication No. 2010-529566 Special Publication No. 2022-507627 Special Publication No. 20185505874

[0008] Guan, X. et al., ACSChem. Biol. 2018, 13 (1), 73-81. https: / / doi.org / 10.1021 / acschembio.7b00794.Hossain, M. A.,et al., J. Am. Chem. Soc. 2020, 142 (3), 1164-1169. https: / / doi.org / 10.1021 / jacs.9b11424.Kabotso, D. E. K.et al., J. Med. Chem. 2020, 63 (11), 6134-6143. https: / / doi.org / 10.1021 / acs.jmedchem.0c00266.Sato,M. et al., J. Am. Chem. Soc. 2004, 126 (43), 14013 14022. https: / / doi.org / 10.1021 / ja046426l.Wang,L. et al., Science Advances 2023, 9 (37), eadi1057, https: / / doi.org / 10.1126 / sciadv.adi1057.Ivanova,M. I. et al., Proceedings of the National Academy of Sciences 2009, 106 (45),18990-18995. https: / / doi.org / 10.1073 / pnas.0910080106.Sato,S., et al., Chemical Communica ions 2018, 54 (46), 5871-5874. https: / / doi.org / 10.1039 / C8CC02891E.Sato,S. , et al., Org. Biomol. Chem. 2020, 18 (19), 3664-3668.https: / / doi.org / 10.1039 / D0 B00650E.Ban,H., et al., J. Am. Chem. Soc. 2010, Vol.132(5), 1523-1525, DOI:10.1021 / ja909062qBan,H., et al., Bioconjugate Chem. 2013, Vol.24(4),pp.520-532, DOI:10.1021 / bc300665tSato,S. & Nakamura, H. Angew. Chem., Int. Ed. 2013, Vol.52(33), pp.8681-8684,DOI: 10.1002 / anie.201303831Sato,S. et al., ACS Chem. Biol.,2015, Vol.10(11), pp.2633-2640, DOI:10.1021 / acschembio.5b00440Sato,S.et al., ChemBioChem 2017, Vol.18(5), pp.475-478, DOI 10.1002 / cbic.201600649Sato,S. et al., Bioconjugate Chem. 2020, Vol.31(5), pp.1417-1424, DOI:10.1021 / acs.bioconjchem.0c00120Alvarez-Dorta,D.et al., J. Am. Fabric. Soc. 2018, Vol.140(49), pp.17120- 17126, DOI:10.1021 / jacs.8b09372Song, C. et al., Chem. Ski. 2019, Vol.10(34),pp.7982-7987, DOI: 10.1039 / C9SC02218JLi,BX et al., Nat. Fabric. 2021, Vol.13, pp.902-908, DOI:10.1038 / s41557-021-00733-yTilley,SD & Francis, MB, J. Am. Fabric. Soc. 2006, Vol.128(4), pp.1080- 1081,DOI: 10.1021 / ja057106kChoi,EJ et al., Chem.-Eur. J. 2018, Vol.24(43), pp.10948-10952, DOI:10.1002 / chem.201802380Maruyamaet al., J. Am. Fabric. Soc., 2021, Vol.143, pp.19844-19855, DOI:10.1021 / jacs.1c09066Ohata, J. et al., Angew. Chem., Int.Ed. 2018, Vol.57(11), 2827-2830, DOI: 10.1002 / anie.201711868Struck, A.-W., J. Am. Chem. Soc. 2016,Vol.138 (9),pp.3038-3045, DOI: 10.1021 / jacs.5b10928Bruins, JJ etal., Bioconjugate Chem. 2017, vol.28(4), pp.1189-1193Montanari, E. etal., Bioconjugate Chem. 2018, Vol.29(8), pp.2550-2560, DOI:10.1021 / acs.bioconjchem.8b00227Marmelstein, AM, et al., J. Am. Chem. Soc. 2020, Vol.142(11), pp.5078-5086, DOI:10.1021 / jacs.9b12002Lobba, MJ etal., ACS Cent. Sci. 2020, vol.6(9), pp.1564-1571, DOI:10.1021 / acscentsci.0c00940.

[0009] As mentioned above, various methods have been attempted to produce insulin preparations that maintain physical and chemical stability. However, although controlling pH and ionic environment is a common protein stabilization method, it does not involve structural modification and therefore cannot completely prevent aggregation. Methods that increase stability by amino acid substitution through genetic recombination or that introduce scaffolds for glycosylation may not be applicable to all existing insulin analogs because they may weaken the effects of amino acid mutations introduced to change the insulin's onset time or duration of action. In addition, chemical modification methods require many complicated steps, such as the introduction of functional groups.

[0010] It has been reported that glycosylation has minimal effect on the secondary structure of proteins, and activity is maintained as long as the modified amino acid position is not involved in activity. The objective of the present invention is to provide a method for stabilizing protein preparations by simple glycosylation, and a stable protein preparation. Here, using insulin preparations as an example, we demonstrate that a chemically and physically stable insulin preparation can be obtained by simple glycosylation without changing the insulin sequence. However, it goes without saying that this method can also be applied to protein preparations other than insulin. Furthermore, because glycosylation can be easily performed without changing the amino acid sequence, it can also be applied to various insulin analogs. This modification method chemically and physically stabilizes protein preparations by glycosylation of tyrosine. Therefore, it can be applied to all protein preparations in which tyrosine is present at an appropriate position.

[0011] The present invention relates to the chemical modification of protein preparations using the following amino acid modification reaction, and to chemically modified protein preparations, particularly insulin preparations. (1) A method for inhibiting and stabilizing aggregation of proteins contained in protein preparations, in which tyrosine present in the protein is modified to stabilize the protein preparation. The method of modifying tyrosine residues with a sugar-containing modifying agent is very simple because it does not require the introduction of a functional group for modification. Any modification method that modifies tyrosine may be used, including but not limited to the methods described in the following examples.

[0012] (2) The method for stabilizing a protein formulation according to (1), characterized in that the method for modifying tyrosine comprises adding a modifying agent, which is a compound having a urazole structure to which a sugar is bound, to the protein and reacting it with laccase to modify the tyrosine present in the protein. The tyrosine residue modification method using a compound having a urazole structure as a modifying agent is an enzymatic reaction that can be carried out under very mild conditions, so there is no risk of denaturing the protein.

[0013] (3) A method for stabilizing a protein preparation according to (1), wherein the protein contained in the protein preparation is insulin. Since insulin was approved as the first recombinant protein preparation, various insulin analogs with different durations of action and time to onset of effect have been commercially available. This method does not require the introduction of functional groups, so existing insulin analogs can be modified. Therefore, stabilization can be achieved while maintaining the effects of the insulin analog, such as its duration of action.

[0014] (4) A method for stabilizing a protein preparation according to (3), which is a stabilization method by modifying TyrA14 of the insulin A chain. The present inventors have found that modification of TyrA14 can stabilize insulin and inhibit aggregation. Chemical modification of TyrA14 can stabilize insulin preparations.

[0015] (5) A protein formulation in which a protein in the protein formulation is modified and stabilized using a modifying agent in which a sugar is bound to a compound having a urazole structure. As shown in the following examples, the aggregation of the protein formulation can be suppressed and stabilized by using a modifying agent in which a sugar is bound to a compound having a urazole structure. The stabilized protein formulation in which protein aggregation is suppressed can also be stored for long periods at room temperature.

[0016] 1 shows the insulin sequence. 2 shows a schematic diagram of the tyrosine residue modification method using 1-methyl-4-aryl-urazole (MAUra) and laccase. 3 shows the compound used as the tyrosine modifier. 4 shows a schematic diagram of the glycosylation reaction scheme. 5 shows the results of MALDI-TOF-MS analysis of the sample after glycosylation of insulin with compounds 1 to 3 and the reaction. 6 shows the results of MALDI-TOF-MS analysis of monoadduct insulin (a molecule in which a glycomodifier and insulin are bound in a 1:1 ratio is called a monoadduct) after HPLC purification. 7 shows the CD spectra of the monoadduct insulin and insulin after HPLC purification. 8 shows the insulin peptide sequence after DTT reduction and Glu-C digestion. 9 shows the analysis results of modified tyrosine residues. 10 shows the results of MALDI-TOF-MS analysis of insulin modified with compounds 1 to 3 after DTT reduction and of the Glu-C digest. 1 shows the results of a ThT assay of mono-adduct insulin modified with compounds 1 to 3. 2 shows the results of analyzing aggregate formation by filter-trap assay of mono-adduct insulin modified with compounds 1 to 3. 3 shows a photograph of a membrane. 4 shows a graph showing the average staining intensity on a membrane. 5 shows the results of evaluating the physiological activity of glycosylated insulin. 6 shows the results of analyzing insulin receptor activation by AKT phosphorylation using Western blotting. 7 shows a graph showing that insulin modified with compound 3 increases AKT phosphorylation in a concentration-dependent manner, indicating activation. 8 shows the results of evaluating the physiological activity of modified insulin in vivo. 9 shows a compound that can modify proteins by forming a covalent bond at the ortho position of a tyrosine residue. 10 shows the results of glycosylation of insulin with compounds 8, 12, 15, 19, 23, and 25, followed by MALDI-TOF-MS analysis of the post-reaction sample. 11 shows the results of a ThT assay of mono-adduct insulin modified with compound 8. 1 shows the results of analyzing aggregate formation by filter trap assay of monoadduct insulin modified with compound 8. 2 is a graph showing the average staining intensity on the membrane.

[0017] As disclosed herein, protein aggregation can be suppressed and stabilization can be achieved by modifying tyrosine residues. There are many methods for modifying tyrosine, as exemplified below (Non-Patent Documents 7 to 26). Specifically, tyrosine residue modification using MAUra (Non-Patent Documents 7 and 8), tyrosine ligation reaction (Non-Patent Document 9), 4-phenyl-3H-1,2,4 Many methods are known, including click reactions using triazoline-3,5(4H)-diones (PTADs) (Non-Patent Document 10), chemical modification using ligand-linked electrooxidation catalysts (Non-Patent Document 11), modification methods using luminol derivatives and hemin (Non-Patent Documents 12-14), electrochemical methods using urazole or phenothiazine derivatives (Non-Patent Documents 15, 16), photoredox catalytic reactions (Non-Patent Document 17), alkylation reactions using π-allylpalladium complexes (Non-Patent Document 18), sulfur-fluorine exchange reactions (Su-FEx) (Non-Patent Document 19), modification with iminoxyl radicals (Non-Patent Document 20), a method in which boronic acid is attached to tyrosine residues using rhodium(III) salts and arene complexes (Non-Patent Document 21), and modification methods using tyrosinase-mediated oxidation reactions (Non-Patent Documents 22-26). In this specification, the tyrosine residue modification method using MAUra is used to stabilize the protein formulation, but tyrosine residues may also be modified using other methods such as those exemplified above, as they are believed to have similar effects.

[0018] The glycosylation method described in this embodiment is a method of binding a compound having a sugar chain attached to a tyrosine residue, and is a very simple method that simply involves mixing a glycosylation agent with a protein and adding a catalytic enzyme. As shown in the following examples, this method does not require the introduction of an amino acid that serves as a scaffold for glycosylation or modification of the amino acid sequence, and therefore can be widely applied as a modification method for existing protein formulations, regardless of whether amino acid modifications are present. Below, we will disclose a method for stabilizing proteins by glycosylation using insulin as an example, but it goes without saying that similar methods can be used to stabilize proteins that do not have a tyrosine residue to be modified at a position related to activity.

[0019] Furthermore, the physical and chemical stability of proteins glycosylated by this method can be further maintained by controlling the pH and ionic environment through the addition of buffer solutions and additives as disclosed in, for example, Patent Documents 1 to 5. Therefore, by adding appropriate buffer solutions and additives to glycosylated proteins, more stable protein formulations can be provided.

[0020] In the following examples, a modifying agent in which the monosaccharide glucose or the disaccharide maltose is added to 1-methyl-4-aryl-urazole (MAUra) is used, but any sugar having hydrophilicity may be used, including, but not limited to, monosaccharides and disaccharides, larger trisaccharides and oligosaccharides. Examples of such sugars include monosaccharides such as galactose, fructose, ribose, allose, altrose, talose, gulose, idose, glucuronic acid, sialic acid, arabinose, xylose, sorbose, glucosamine, and N-acetylglucosamine, disaccharides such as lactose and sucrose, and trisaccharides such as raffinose and maltotriose.

[0021] As shown in the following examples, the more negative the ΔCLogP value of the added modifying agent, the more aggregation can be suppressed, so that a modifying agent to which a sugar that makes the ΔCLogP value negative upon modification can be preferably used. Furthermore, since solubility also changes depending on the isoelectric point, a sugar that can change the isoelectric point can also be preferably selected.

[0022] Insulin is a protein consisting of 51 amino acids, with two polypeptide chains, the A chain (SEQ ID NO: 1) and the B chain (SEQ ID NO: 2), connected by disulfide bonds. The A chain and the B chain each contain two tyrosines, for a total of four (Figure 1). Native insulin has a secondary structure rich in α-helices, but it has been shown that in aggregates this changes to a structure rich in β-sheets (Non-Patent Document 5). It has been reported that insulin fibrils, the aggregates of insulin, have various morphologies, but are thought to be composed of a continuous fibrillar cross-β-sheet structure formed by antiparallel interactions of β-sheet structures. It has been suggested that the paired structure of the LVEALYL (B11-B17, SEQ ID NO: 9) sequence, which is part of the B chain, may cause aggregation, and that the LVEALYL sequence and the SLYQLENY (A12-A19, SEQ ID NO: 10) sequence of the A chain may form a β-sheet, interacting via tyrosine (Non-Patent Document 6). It has also been shown that the interaction between insulin molecules is enhanced by hydrophobic interactions and π-π stacking involving the side chains of PheB24 (phenylalanine, the 24th amino acid of the B chain; similarly, the terms "amino acid residue," "chain," and "position from the N-terminus") and TyrB26 and the side chains of IleA2, SerA12, and TyrA14 (Non-Patent Document 5). Since the sequence involved in aggregation contains tyrosine, glycosylation of tyrosine may lead to aggregation inhibition by utilizing the hydration effect and steric hindrance of the sugar structure. Therefore, we investigated tyrosine modification of insulin with sugars.

[0023] [Tyrosine Modification of Insulin] The tyrosine modification method used was the MAUra-based tyrosine residue modification method (Non-Patent Documents 7 and 8). In this method, a modifying agent with a urazole structure undergoes one-electron oxidation by the oxidation catalyst laccase, forming a covalent bond at the ortho position of the tyrosine residue (Figure 2A). Because there are two ortho positions per tyrosine residue, up to two molecules of the modifying agent can be added. This reaction can highly efficiently modify tyrosine, and because it is a mild reaction that proceeds in water at 37°C and pH 6.0, it does not cause protein denaturation. Therefore, we adopted the MAUra-based tyrosine residue modification method.

[0024] Since the addition of sialic acid to insulin has been shown to inhibit aggregation (Non-Patent Document 3), the addition of monosaccharides is also thought to be effective. Therefore, we performed glycosylation using a compound in which a tyrosine residue modification structure was linked to glucose or maltose as a tyrosine modifier (Figure 2B). Compound 1 is a glycosylation modifier in which glucose is linked to the tyrosine modifier MAUra, and compound 2 is a structure in which the glucose moiety of compound 1 has been removed. This compound was used as a control because the MAUra structure may exhibit aggregation inhibition due to steric hindrance. Compound 3 is a glycosylation modifier in which maltose is linked to MAUra. It has a larger sugar structure than compound 1 and is expected to be more hydrophilic due to the replacement of the benzene ring with a triazole ring. When the LogP, an index of compound hydrophobicity, was predicted using ChemDraw (CLogP), compound 3 was estimated to have the lowest hydrophobicity (Table 1). Furthermore, using this index, the change in the hydrophobicity of tyrosine when compounds 1 to 3 bind to tyrosine was estimated by calculating the difference (ΔC Log P) between the Log P of the phenol alone, which is the side chain structure of tyrosine, and the C Log P of the structure after one compound 1 to 3 is added to the phenol (Table 1). It was predicted that compound 3 would also reduce the hydrophobicity of tyrosine the most in terms of ΔC Log P.

[0025]

[0026] Compounds 1-3 were synthesized according to the methods described in Non-Patent Documents 7 and 8. The glycosylation reaction was performed as follows (Figure 3A). Compounds 1-3 were reacted with insulin at a 1:1 ratio. Compounds 1-3 were each dissolved in DMSO to a concentration of 10 mM, and 2.5 μL was added to 500 μL of a 50 μM insulin solution (10 mM sodium phosphate buffer, pH 6.0) and mixed thoroughly (the final concentrations of compounds 1-3 were 50 μM, 0.5% DMSO). Laccase (Amano Enzyme Inc.) was added to a final concentration of 250 nM, mixed thoroughly, and shaken at 37°C for 1 hour. The mixture was shaken while exposed to air.

[0027] The reaction samples were analyzed by MALDI-TOF-MS to confirm the progress of the modification reaction. It was found that all samples modified with compounds 1 to 3 contained high proportions of 0-adducts (unmodified insulin) and 1-adducts (a molecule in which a glycomodifier and insulin are linked in a 1:1 ratio) (Figure 3B). Because this allows evaluation of changes in physical properties due to the addition of a single molecule and because the fewer types of glycomodifier adducts produced in the reaction mixture, the easier it is to purify, we decided to produce and purify modified insulin using these reaction conditions. The reaction mixtures of compounds 1 to 3 were analyzed by HPLC, and the peak of the 1-adduct insulin was identified and separated.

[0028] The glycosylation reaction sample was purified by reverse-phase HPLC. The solution containing the modified insulin was purified using an AX-C18 column (InertSustain, inner diameter x length: 4.6 x 150 mm, particle size: 5 μm, GL Sciences Inc.). The program was as follows: linear gradient, gradient rate: 30-100% B / 14 min; eluent A: 10 mM ammonium formate aqueous solution; eluent B: 10 mM ammonium formate (water / acetonitrile = 90 / 10); flow rate: 1 ml / min, 40°C. Since the maximum absorption wavelength of insulin is around 276 nm, detection was performed at a wavelength of 275 nm.

[0029] The samples were stored at -80°C until further analysis. The purified samples were analyzed by MALDI-TOF-MS to identify the molecules contained in the samples (Figure 4). The results confirmed that the mono-adducts of each compound accounted for the majority of the purified samples.

[0030] [Estimation of Secondary Structure by CD Spectroscopy] Circular dichroism (CD) spectroscopy was performed to determine whether the purified monoadduct insulins of compounds 1 to 3 maintained the same secondary structure as native insulin. The compounds are indicated by the compound numbers shown in Figure 2B. For example, an insulin with compound 1 attached is referred to as mono-modified insulin. In this example, the reaction was performed and purified under conditions that yielded monoadduct insulin, so insulin with one compound attached was used for all of the following analyses. After HPLC purification, each of the mono-adduct insulins (1 to 3) was diluted with sodium phosphate buffer (pH 7.4, 10 mM) to a final concentration of 10 μM. CD spectra were obtained using a spectrophotometer (J-720WI, Jasco Co.) by injecting 100 μL of sample into a cylindrical microquartz cell with a 10 mm optical path length, and integrating four times in the wavelength range from 180 nm to 350 nm at a scan speed of 100 nm / min and a resolution of 0.5 nm. The spectrum of the sodium phosphate buffer solution was subtracted from the spectrum of each sample.

[0031] CD spectra of proteins exhibit characteristic spectra derived from their higher-order structures (α-helix, β-sheet, random coil), allowing us to estimate their higher-order structure in solution from the shape of the spectrum. The CD spectrum of native insulin exhibits a typical α-helical spectrum with negative maxima at 208 nm and 222 nm. However, when the insulin transforms into a β-sheet-rich structure that exhibits aggregation, a negative maximum is observed only at approximately 218 nm. Figure 5 shows the CD spectra of human insulin and the three modified insulins. The CD spectra of insulin modified with compounds 1-3 exhibited a typical α-helical spectral pattern with negative maxima near 210 nm and 224 nm. This suggests that the secondary structure of the insulins modified with compounds 1-3 remains the same as that of the unmodified insulin.

[0032] [Identification of glycosylation sites] To identify which of the four tyrosines in the purified 1-3 modified insulin was glycosylated, the insulin was fragmented using digestive enzymes and reducing agents and analyzed by mass spectrometry. The two disulfide bonds bridging the two peptide chains of insulin were reduced with the reducing agent dithiothreitol (DTT), separating the two peptide chains (Figure 6A, top). Specifically, 20 μL of 5 μM 1-3 modified insulin (10 mM phosphate buffer, pH 7.4) was mixed with 1 μL of 200 mM DTT solution and incubated at 37°C for at least 1 hour. MALDI-TOF-MS analysis of the reduced sample detected the unmodified B chain in the 1-3 modified insulin (Figure 6B, top panel). These results suggest that the glycosylation site is TyrA14 or TyrA19 in the A chain.

[0033] Next, insulin was fragmented using the digestive enzyme Glu-C. Glu-C is a serine protease that specifically cleaves the C-terminus of aspartic acid or glutamic acid residues, and in insulin, it cleaves at the C-terminus of the four glutamic acid residues (Figure 6A, bottom). Glu-C (Promega) was added to the insulin solution to a final protein:enzyme ratio of 200:1 to 20:1 (w / w) and incubated overnight at 37°C. MALDI-TOF-MS analysis of the Glu-C-digested samples revealed two fragments: A(5-17)-B(1-13) containing TyrA14 and B(22-30) containing TyrB26 (Figure 6B, bottom panel). In all 1-, 2-, and 3-modified insulins, the A(5-17)-B(1-13) mono-adduct and the B(22-30) o-adduct were detected. Since no unmodified fragment of A(5-17)-B(1-13) was detected, it was concluded that TyrA14 contained in this fragment was the main modification site.

[0034] [Evaluation of Aggregation Stability] 1. Evaluation by ThT Assay To evaluate the aggregation stability of glycosylated insulin, a ThT (thioflavin T) assay was performed. As described above, insulin aggregates in various forms of cross-β-sheet structure. ThT is a molecule that binds to cross-β-sheet structures and enhances fluorescence. Based on preliminary studies, the conditions for evaluating aggregation stability were determined as follows: 1-3 modified insulin or insulin was dissolved in 10 mM sodium phosphate and 100 mM sodium chloride (pH 7.4) to a concentration of 10 μM. 3 μL of 1 mM ThT (DMSO solution, final concentration 10 μM) was added to 300 μL of insulin solution. 50 μL of this solution was dispensed into each well of a clear-bottom 96-well plate (Greiner). The samples were sealed with a seal and incubated in a plate reader set to aggregation conditions (40°C, 1000 rpm) to induce aggregation. The time course of ThT fluorescence intensity was measured using a fluorometer (Perkin-Elmer, EnVision) by exciting at 430 nm and measuring fluorescence at 470 nm. Unmodified insulin began to increase in fluorescence 2 hours after the start of the assay, whereas the 1-3 modified insulins showed a delayed onset of aggregation (Figure 7). Experiments were performed in sextuplicate; each line represents the mean value, and error bars represent the standard error of six samples.

[0035] The onset of aggregation was delayed by 9 hours for 2-modified insulin without a glycosylated structure compared to unmodified insulin. The fact that 2-modified insulin inhibited aggregation suggests that steric hindrance due to chemical modification of TyrA14 is effective in inhibiting aggregation. Furthermore, 1-modified insulin initiated aggregate formation 14 hours after the start of the assay, while 3-modified insulin inhibited aggregation for at least 15 hours. The 3-modified insulin exhibited a stronger aggregation-inhibiting effect than 1-modified insulin, which is thought to be correlated with the high hydrophilicity of the modifying agent structure.

[0036] 2. Evaluation by Filter Trap Assay Using samples immediately after the ThT assay, a filter trap assay was performed to detect insoluble proteins in the sample. By filtering the sample through a membrane with a defined pore size, unaggregated proteins and smaller molecules pass through the membrane, while aggregated proteins are captured by the membrane. The filter trap assay allows the detection of amorphous aggregates that cannot be detected by ThT and aggregates that are just large enough to form fibrils, enabling a more detailed evaluation of aggregation stability.

[0037] A 0.45 μm pore size cellulose acetate membrane (Cytiva) was hydrophilized by soaking in methanol and sandwiched between dot blotters (Scie-Plotters). 100 μL of TBS buffer was added to each well and aspirated. The ThT assay sample was then added to each well and aspirated. The membrane was removed from the dot blotter, washed with TBST, and then immersed in Ponceau S staining solution (Beacle Inc.) and shaken at room temperature for 15 minutes for staining. The membrane was then immersed in Milli-Q to wash away excess staining solution. Ponceau staining allowed visualization of aggregates on the membrane (Figure 8A).

[0038] Ponceau staining allows the detection of total protein by reversibly binding a negatively charged dye molecule to the positively charged amino groups of proteins. Darker spots were observed on the membrane, indicating the possible presence of large aggregates. For mono-modified insulin, which was shown to be in the initial stage of aggregation in the ThT assay, the membrane was stained only in samples that showed high fluorescence intensity at the end of the assay (Figure 8A). No aggregates were detected with triply modified insulin. These results indicate that mono- and triply modified insulins did not form amorphous or small aggregates that might not be detected by the ThT assay, confirming the strong aggregation-stabilizing effects of mono- and triply modified insulins. The aggregates on the stained membrane were quantified using ImageJ (Figure 8B). Filter trap assay results showed that unmodified insulin and di-insulin, which reached maximum aggregation in the ThT assay, showed high intensity.

[0039] [Physiological activity evaluation] We analyzed whether glycosylated insulin activates the insulin receptor (IR) using 3-modified insulin. When insulin binds to the receptor, a conformational change occurs in the receptor, leading to autophosphorylation of tyrosine in the cytoplasmic domain and activation of downstream signal transduction pathways. Therefore, the IR-activating ability of 3-modified insulin was evaluated by detecting the activation (phosphorylation) of AKT, which is present downstream of the signal.

[0040] Insulin or tri-modified insulin was added to HEK293 cells at concentrations ranging from 0.2 to 50 nM. After 15 minutes, cells were harvested and p-AKT was detected by Western blot analysis (Figure 9A). The primary antibody used was anti-p-Akt (T308) antibody, and the control was anti-vinculin antibody (both from Cell Signaling Technology). The secondary antibody used was anti-rabbit StarBright™ Blue 700 fluorescently labeled secondary antibody (Bio-Rad). Visualization was performed using ChemiDoc (Bio-Rad). Quantification of each band revealed no difference in the intensity of p-AKT detection between insulin and tri-modified insulin (Figure 9B). This indicates that the IR activation ability of tri-modified insulin is maintained.

[0041] [In vivo physiological activity evaluation] The effects of modified insulin were evaluated using mice. C57BL / 6N mice (SLC Japan) were maintained under conditions with free access to commercially available pelleted feed (CRF-1, Oriental Yeast Co., Ltd.) and drinking water (tap water) for at least one week prior to use in the experiment. After a 3-hour fast, the mice were intraperitoneally administered insulin or 3-modified insulin (15.7 nmol / kg body weight). Tail vein blood glucose levels were measured before fasting, before administration (0 min), and 30 and 60 min after administration using a Nova StatStrip Express 900 (Siemens Healthcare). The modified insulin had an effect on blood glucose levels comparable to that of insulin, indicating that it maintained its physiological activity (Figure 10).

[0042] Next, we show compounds (Figure 11) that have a urazole structure and can modify proteins by forming a covalent bond at the ortho position of a tyrosine residue using laccase, as described above, and their synthesis method. The numbers written under the compounds shown below correspond to the compound numbers shown in Figure 11 and the synthesis method described below.

[0043]

[0044] Synthesis of Benzyl 1-methylhydrazine-1-carboxylate (Compound 1). Under a nitrogen atmosphere, methylhydrazine (2.98 g, 64.6 mmol, 5 equivalents) and triethylamine (5.40 mL, 38.8 mmol, 3 equivalents) were dissolved in anhydrous dichloromethane (DCM, 10 mL), and benzyl chloroformate (1.82 mL, 12.9 mmol, 1 equivalent) was added dropwise at 0°C. The mixture was stirred at room temperature for 1 hour, filtered, and evaporated. The residue was purified by flash column chromatography (ethyl acetate / hexane, 10% to 77% ethyl acetate) to obtain the target compound 1 (1.24 g, 53% yield) as a colorless oil.

[0045] Synthesis of Benzyl 1-methyl-2-(prop-2-yn-1-ylcarbamoyl)hydrazine-1-carboxylate (Compound 2). Under a nitrogen atmosphere, 1,1-carbodiimidazole (1.21 g, 7.45 mmol, 1.1 equivalents) and the synthesized compound 1 (1.22 g, 6.77 mmol, 1.0 equivalents) were stirred in anhydrous tetrahydrofuran (THF, 15 ml) at room temperature for 30 minutes. Propargylamine (720 mg, 0.84 ml, 13.1 mmol, 1.93 equivalents) was added to the mixture, which was then stirred at 60°C for 24 hours. The mixture was evaporated, and the resulting residue was purified by flash column chromatography (ethyl acetate / hexane, 10% to 100% ethyl acetate) to give compound 2 (1.49 g, 84%) as a colorless oil.

[0046] Synthesis of 1-methyl-4-(prop-2-yn-1-yl)-1,2,4-triazolidine-3,5-dione (Compound 3). Compound 2 (1.49 g, 5.68 mmol, 1 equivalent) obtained in the above synthesis was stirred in water (50 ml) and 4 M aqueous KOH solution (957 mg, 17.1 mmol, 3 equivalents) at 80°C for 15 hours. This mixture was neutralized with 1 M aqueous HCl solution and concentrated under reduced pressure. The crude product was purified by flash column chromatography (chloroform / methanol, 9% to 76% methanol) to obtain compound 3 (204 mg, 23%) as a yellow oil.

[0047]

[0048] [(2S,3R,4S,5R,6R)-6-(acetoxymethyl)-5-(((2R,3R,4S,5R,6R)-3 ,4-diacetoxy-6-(acetoxymethyl)-5-(((2R,3R,4S,5R,6R)-3,4,5- triacetoxy-6-(acetoxymethyl)tetrahydro-2H-pyran-2-yl)oxy)t etrahydro-2H-pyran-2-yl)oxy)tetrahydro-2H-pyran-2,3,4-triyl Synthesis of triacetate (compound 5)] Maltotriose (378 mg, 0.750 mmol, 1 equiv.), Pd(PhCN)Cl (63.3 mg, 165 μmol, 22 mol%), and silver trifluoromethanesulfonate (84.8 mg, 330 μmol, 44 mol%) were dissolved in anhydrous DCM under a nitrogen atmosphere, and acetic anhydride (3.12 mL, 33.0 mmol, 44 equiv.) was added. The mixture was stirred at room temperature for 12 h, quenched with saturated NaHCO (6 mL), and extracted with DCM (10 mL × 3). The residue was purified by flash column chromatography (chloroform / methanol, 3% to 24% methanol) to give compound 5 (521 mg, 72%) as a white solid.

[0049] [(2R,3R,4S,5R,6R)-2-(acetoxymethyl)-6-(((2R,3R,4S,5R,6R)-4, 5-diacetoxy-2-(acetoxymethyl)-6-(((2R,3R,4S,5R,6R)-4,5-diacet toxy-2-(acetoxymethyl)-6-azidotetrahydro-2H-pyran-3-yl)oxy) tetrahydro-2H-pyran-3-yl)oxy)tetrahydro-2H-pyran-3,4,5-triyl Synthesis of triacetate (compound 6)] Iron(III) chloride (81.1 mg, 0.500 mmol, 1 equiv.) in anhydrous DCM (1 mL) and compound 5 in anhydrous DCM (0.5 mL) were stirred at room temperature for 5 minutes under a nitrogen atmosphere. Azidotrimethylsilane (86.4 mg, 0.750 mmol, 1.5 equiv.) was added to the mixture, which was stirred at room temperature for 28 hours. The reaction was quenched with saturated NaHCO (10 mL), extracted with DCM (5 mL x 3), and dried over MgSO. The residue was purified by flash column chromatography (ethyl acetate / hexane, 10% to 100% ethyl acetate) to give compound 6 (244 mg, 51%) as a white solid.

[0050] [(2R,3R,4S,5R,6R)-2-(acetoxymethyl)-6-(((2R,3R,4S,5R,6R)-4,5-diacetoxy-2-(ace diacetoxy-2-(acetoxymethyl)-6-(((2R,3R,4S,5R,6R)-4,5-diacetoxy-2-(acetoxymethyl)-6-(4-((1-meth) Synthesis of (hydroxymethyl)-1H-1,2,3-triazol-1-yl)tetrahydro-2H-pyran-3-yl)oxy)tetrahydro-2H-pyran-3-yl)oxy)tetrahydro-2H-pyran-3,4,5-triyl triacetate (Compound 7) 20 mol% iron(III) chloride (8.11 mg, 0.05 mmol, 0.2 equivalents) and Cu powder (15.9 mg, 0.25 mmol, 1.0 equivalents) were placed in a 10 mL round-bottom flask and dissolved in tBuOH:H ２A 1:1 mixture of Compound 6 (237 mg, 0.25 mmol, 1 equiv.) in tBuOH (1 mL) was added. The resulting mixture was stirred at ambient temperature for 5 minutes, and then a solution of Compound 6 (237 mg, 0.25 mmol, 1 equiv.) in tBuOH (1 mL) was added, followed by Compound 3 (57.4 mg, 0.375 mmol, 1.5 equiv.). The resulting reaction mixture was stirred at ambient temperature and the progress of the reaction was monitored by TLC. After 18 hours, complete disappearance of the starting material was confirmed, and the organic volatiles were removed under reduced pressure. The resulting mixture was then immersed in H ２ The mixture was partitioned between 0 and DCM (5 mL each). The organic layer was separated, and the aqueous layer was further extracted with DCM (3 × 3 mL). The organic layers were combined, washed with brine (10 mL), dried over MgSO, filtered, filtered through Celite to remove inorganic salts, and concentrated under reduced pressure. The resulting crude product was purified by reverse-phase silica gel column chromatography (water / CHCN) to give compound 7 (161 mg, 58%) as a pink solid.

[0051] [4-((1-((2R,3R,4R,5S,6R)-5-((2R,3R,4R,5S,6R)-3,4-dihydroxy-6-(hydroxymethyl)-5- ((2R,3R,4S,5S,6R)-3,4,5-trihydroxy-6-(hydroxymethyl)tetrahydr ro-2H-pyran-2-yl)oxy)tetrahydro-2H-pyran-2-yl)oxy)-3,4-dihydro xy-6-(hydroxymethyl)tetrahydro-2H-pyran-2-yl)-1H-1,2,3-triazo Synthesis of l-4-yl)methyl)-1-methyl-1,2,4-triazolidine-3,5-dione (compound 8)] Compound 7 (98.8 mg, 33 μmol, 1 equiv.) dissolved in anhydrous methanol (1.5 ml) at 0° C. was added with 28% sodium methylate solution (in methanol, 275 μl) and stirred at room temperature for 1 hour. The mixture was neutralized with Amberlyst™ 15 ion exchange resin (H+ form), filtered, and evaporated. The residue was purified by reversed-phase column chromatography (H ２ O / methanol) and preparative HPLC (H ２0 / methanol) to give compound 8 (9.9 mg, 16%) as a white solid.

[0052]

[0053] [Synthesis of (2R,5R)-2-azido-5-(hydroxymethyl)tetrahydrofuran-3,4-diol (Compound 9)] 2-Chloro-1,3-dimethylimidazolinium chloride (DMC, 875 mg, 5.00 mmol) was added to D-ribose (250 mg, 1.67 mmol), triethylamine (2.32 ml, 16.7 mmol), and an aqueous solution (6 ml) of NaN3 (1.08 g, 16.7 mmol), and the reaction mixture was stirred at 0°C for 3 hours. The reaction mixture was concentrated, 10 ml of ethanol was added, and the solid was removed by filtration. The filtrate was concentrated under vacuum and purified by reverse phase column chromatography (H ２ 0 / methanol) and concentrated in vacuo to give compound 9 (132 mg, 45%).

[0054] Synthesis of 4-((1-((2R,5R)-3,4-dihydroxy-5-(hydroxymethyl)tetrahydrofuran-2-yl)-1H-1,2,3-triazol-4-yl)methyl)-1-methyl-1,2,4-triazolidine-3,5-dione (Compound 10). Copper(II) sulfate pentahydrate (8.64 mg, 34.6 μmol) and sodium ascorbate (13.7 mg, 69.2 μmol) in degassed deionized water (1 mL) were added to compound 9 (20.2 mg, 0.115 mmol) in degassed ethanol (0.4 mL), and the reaction mixture was heated at 80° C. for 15 minutes. Compound 3 (19.4 mg, 0.127 mmol) was added to the reaction mixture, which was heated at 50° C. for 16 hours and concentrated under reduced pressure. The residue was purified by reverse-phase column chromatography (water / methanol) and then by preparative HPLC (water / methanol) to give compound 10 (12.0 mg, 32%) as a white solid.

[0055]

[0056] [(2R,3S,4R,5R,6R)-5-acetamido-2-(acetoxymethyl)-6-(4-((1-methyl-3,5-dioxo-1,2 ,4-triazolidin-4-yl)methyl)-1H-1,2,3-triazol-1-yl)tetrahydro-2H-pyran-3,4-diyl Synthesis of diacetate (compound 11)] 2-Acetamido-3,4,6-tri-O-acetyl-2-deoxy-β-D-glucopyranosyl azide (100 mg, 0.269 mmol), compound 3 (45.2 mg, 0.295 mmol), copper(II) sulfate pentahydrate (10.5 mg, 41.9 μmol), and sodium ascorbate (26.6 mg, 0.134 mmol) were added sequentially to the reaction flask. ２ A mixture of 0.25O / tBuOH / DCM (1:2:1, 4 mL) was added, and the reaction flask was fitted with a reflux condenser, covered with aluminum foil, and heated to 60°C. After stirring at this temperature for 24 h, the mixture was cooled to room temperature and the solvent was evaporated. The resulting crude product was purified by reversed-phase silica gel column chromatography (H ２ 0 / methanol) to give compound 11 (112 mg, 79%).

[0057] Synthesis of N-((2R,3R,4R,5S,6R)-4,5-dihydroxy-6-(hydroxymethyl)-2-(4-((1-methyl-3,5-dioxo-1,2,4-triazolidin-4-yl)methyl)-1H-1,2,3-triazol-1-yl)tetrahydro-2H-pyran-3-yl)acetamide (Compound 12) Compound 11 (107 mg, 204 μmol, 1 equivalent) in absolute methyl alcohol (5 mL) was added to a solution of 28% sodium methylate (595 μL) at 0° C., and the mixture was stirred at room temperature for 1 hour. After neutralization with Amberlyst 15 ion exchange resin (H+ form), filtration and evaporation were carried out. The residue was purified by reversed-phase column chromatography (H ２ O / methanol) and preparative HPLC (H ２0 / methanol) to give compound 12 (49.9 mg, 61%) as a white solid.

[0058]

[0059] [(2R,3S,4S,5R,6S)-2-(acetoxymethyl)-6-(((2R,3R,4S,5R,6R)-4,5-diacetoxy-2-(acet oxymethyl)-6-azidotetrahydro-2H-pyran-3-yl)oxy)tetrahydro-2H-pyran-3,4,5-triyl Synthesis of triacetate (compound 13)] Under a nitrogen atmosphere, lactose (856 mg, 2.50 mmol), Pd(PhCN)Cl (76.7 mg, 200 μmol, 8 mol%), and silver trifluoromethanesulfonate (103 mg, 400 μmol, 16 mol%) were dissolved in 8 mL of anhydrous DCM, and acetic anhydride (7.56 mL, 80.0 mmol) was added. The mixture was stirred at room temperature for 2 hours and quenched with saturated NaHCO (6 mL). The mixture was then extracted with DCM (10 mL × 3), and the residue was purified by flash column chromatography (chloroform / methanol) to give compound 13′ (1.61 g, 95%) as a white solid.

[0060] Iron(III) chloride (243 mg, 1.50 mmol) in anhydrous DCM (5 mL) and compound 13' in anhydrous DCM (2.5 mL) were stirred at room temperature for 5 minutes under a nitrogen atmosphere. Azidotrimethylsilane (259 mg, 2.25 mmol) dissolved in anhydrous DCM (2.5 mL) was added to the mixture, which was stirred at room temperature for 36 hours. Saturated NaHCO ３ The reaction was quenched with 10 ml of HCl. Then, it was extracted with DCM (5 ml × 3) and dried over MgSO. The residue was purified by flash column chromatography (ethyl acetate / hexane) to give compound 13 (992 mg, 59%) as a white solid.

[0061] [(2R,3S,4S,5R,6S)-2-(acetoxymethyl)-6-(((2R,3R,4S,5R,6R)-4,5-diacetoxy-2-(acetoxymethyl)-6-(4-((1-methyl-3,5-dioxo -1,2,4-triazolidin-4-yl)methyl)-1H-1,2,3-triazol-1-yl)tet rahydro-2H-pyran-3-yl)oxy)tetrahydro-2H-pyran-3,4,5-triyl Synthesis of triacetate (compound 14)] Compound 13 (100 mg, 0.151 mmol), compound 3 (25.5 mg, 0.166 mmol), copper(II) sulfate pentahydrate (5.89 mg, 23.6 μmol), and sodium ascorbate (15.0 mg, 75.6 μmol) were added sequentially to the reaction flask. ２ A mixture of 0H2O / tBuOH / DCM (1:2:1, 4 mL) was added, and the reaction flask was fitted with a reflux condenser, covered with aluminum foil, and heated to 60°C. After stirring at this temperature for 24 h, the mixture was cooled to room temperature and the solvent was evaporated. The resulting crude product was purified by reversed-phase silica gel column chromatography (H ２ 0 / methanol) to give compound 14 (99.5 mg, 81%).

[0062] [4-((1-((2R,3R,4R,5S,6R)-3,4-dihydroxy-6-(hydroxymethyl)-5-(( (2S,3R,4S,5R,6R)-3,4,5-trihydroxy-6-(hydroxymethyl)tetrahydro -2H-pyran-2-yl)oxy)tetrahydro-2H-pyran-2-yl)-1H-1,2,3-triazol -4-yl)methyl)-1-methyl-1,2,4-triazolidine-3,5-dione (compound 15)] Compound 14 (93.6 mg, 115 μmol, 1 equivalent) was dissolved in anhydrous methanol (5 ml) at 0° C., and 28% sodium methylate solution (in methanol, 784 μl) was added and stirred at room temperature for 1 hour. After neutralization with Amberlyst 15 ion exchange resin (H+ form), the mixture was filtered and evaporated. The residue was purified by reversed-phase column chromatography (H ２ O / methanol) and then purified by preparative HPLC (H ２ 0 / methanol) to give compound 15 (38.2 mg, 64%) as a white solid.

[0063]

[0064] Synthesis of (2S,3S,4S,5R,6R)-6-(acetoxymethyl)tetrahydro-2H-pyran-2,3,4,5-tetrayl tetraacetate (Compound 16): Under a nitrogen atmosphere, mannose (1.8 g, 10.0 mmol), Pd(PhCN)Cl (192 mg, 500 μmol, 5 mol%), and silver trifluoromethanesulfonate (257 mg, 1.00 mmol, 10 mol%) were dissolved in 20 mL of anhydrous DCM, and acetic anhydride (18.9 mL, 200 mmol) was added. The mixture was stirred at room temperature for 2 hours, quenched with saturated NaHCO (20 mL), and extracted with DCM (10 mL × 3). The residue was purified by flash column chromatography using chloroform / methanol to give compound 16 (3.78 g, 97%) as a colorless oil.

[0065] Synthesis of (2R,3R,4S,5S,6S)-2-(acetoxymethyl)-6-azidotetrahydro-2H-pyran-3,4,5-triyl triacetate (Compound 17) Iron(III) chloride (906 mg, 5.58 mmol) in anhydrous DCM (5 mL) and Compound 16 in anhydrous DCM (2.5 mL) were stirred at room temperature for 5 minutes under a nitrogen atmosphere. Azidotrimethylsilane (965 mg, 8.38 mmol) dissolved in anhydrous DCM (2.5 mL) was added to the mixture, which was stirred at room temperature for 12 hours. The reaction was then quenched with saturated NaHCO (10 mL). The mixture was then extracted with DCM (5 mL x 3) and dried over MgSO. The residue was purified by flash column chromatography (ethyl acetate / hexane) to give compound 17 (1.55 g, 74%) as a white solid.

[0066] Synthesis of (2R,3R,4S,5S,6S)-2-(acetoxymethyl)-6-(4-((1-methyl-3,5-dioxo-1,2,4-triazolidin-4-yl)methyl)-1H-1,2,3-triazol-1-yl)tetrahydro-2H-pyran-3,4,5-triyl triacetate (Compound 18) Compound 17 (100 mg, 0.268 mmol), compound 3 (50.1 mg, 0.295 mmol), copper(II) sulfate pentahydrate (10.4 mg, 41.8 μmol), and sodium ascorbate (26.5 mg, 134 μmol) were added sequentially to a reaction flask. ２ A mixture of 0H2O / tBuOH / DCM (1:2:1, 4 mL) was added, and the reaction flask was fitted with a reflux condenser, covered with aluminum foil, and heated to 60°C. After stirring at this temperature for 36 h, the mixture was cooled to room temperature and the solvent was evaporated. The resulting crude product was purified by reversed-phase silica gel column chromatography (H ２ 0 / methanol) to give compound 18 (104 mg, 74%).

[0067] Synthesis of 1-methyl-4-((1-((2S,3S,4S,5S,6R)-3,4,5-trihydroxy-6-(hydroxymethyl)tetrahydro-2H-pyran-2-yl)-1H-1,2,3-triazol-4-yl)methyl)-1,2,4-triazolidine-3,5-dione (Compound 19) Compound 18 (104 mg, 198 μmol) was dissolved in anhydrous methanol (5 ml) at 0°C, and 28% sodium methylate solution (in methanol, 770 μl) was added and stirred at room temperature for 1 hour. After neutralization with Amberlyst 15 ion exchange resin (H+ form), the mixture was filtered and evaporated. The residue was purified by reversed-phase column chromatography (H ２ O / methanol) and then purified by preparative HPLC (H ２ 0 / methanol) to give compound 19 (46.0 mg, 65%) as a white solid.

[0068]

[0069] Synthesis of (3R,4S,5S,6R)-6-(acetoxymethyl)tetrahydro-2H-pyran-2,3,4,5-tetrayl tetraacetate (Compound 20): Galactose (1.80 g, 10.0 mmol), Pd(PhCN)Cl (192 mg, 500 μmol, 5 mol%), and silver trifluoromethanesulfonate (257 mg, 1.00 mmol, 10 mol%) were dissolved in 20 mL of anhydrous DCM under a nitrogen atmosphere, and acetic anhydride (18.9 mL, 200 mmol) was added. The mixture was stirred at room temperature for 2 hours, quenched with saturated NaHCO (20 mL), and extracted with DCM (10 mL × 3). The residue was purified by flash column chromatography with chloroform / methanol to give compound 20 (3.56 g, 91%) as a colorless oil.

[0070] Synthesis of (2R,3S,4S,5R,6R)-2-(acetoxymethyl)-6-azidotetrahydro-2H-pyran-3,4,5-triyl triacetate (Compound 21): Under a nitrogen atmosphere, iron(III) chloride (1.47 g, 9.08 mmol) in anhydrous DCM (5 mL) and compound 20 in anhydrous DCM (2.5 mL) were stirred at room temperature for 5 minutes. Azidotrimethylsilane (1.57 g, 13.6 mmol) dissolved in anhydrous DCM (2.5 mL) was added to the mixture, which was stirred at room temperature for 36 hours. The reaction was then quenched with saturated NaHCO (10 mL). The mixture was then extracted with DCM (5 mL x 3) and dried over MgSO. The residue was purified by flash column chromatography (ethyl acetate / hexane) to give compound 21 (2.09 g, 61%) as a white solid.

[0071] Synthesis of (2R,3S,4S,5R,6R)-2-(acetoxymethyl)-6-(4-((1-methyl-3,5-dioxo-1,2,4-triazolidin-4-yl)methyl)-1H-1,2,3-triazol-1-yl)tetrahydro-2H-pyran-3,4,5-triyl triacetate (Compound 22) Compound 17 (100 mg, 0.268 mmol), compound 3 (50.1 mg, 0.295 mmol), copper(II) sulfate pentahydrate (10.4 mg, 41.8 μmol), and sodium ascorbate (26.5 mg, 134 μmol) were added sequentially to a reaction flask. ２ A mixture of 0.25O / tBuOH / DCM (1:2:1, 4 mL) was added, and the reaction flask was fitted with a reflux condenser, covered with aluminum foil, and heated to 60°C. After stirring at this temperature for 36 h, the mixture was cooled to room temperature and the solvent was evaporated. The resulting crude product was purified by reversed-phase silica gel column chromatography (H ２ 0 / methanol) to give compound 22 (119 mg, 85%).

[0072] Synthesis of 1-methyl-4-((1-((2R,3R,4S,5R,6R)-3,4,5-trihydroxy-6-(hydroxymethyl)tetrahydro-2H-pyran-2-yl)-1H-1,2,3-triazol-4-yl)methyl)-1,2,4-triazolidine-3,5-dione (Compound 23) Compound 22 (110 mg, 209 μmol) was dissolved in anhydrous methanol (5 ml), and 28% sodium methylate solution (in methyl alcohol, 813 μl) was added at 0°C. The mixture was stirred at room temperature for 1 hour. After neutralization with Amberlyst 15 ion exchange resin (H+ form), the mixture was filtered and evaporated. The residue was purified by reversed-phase column chromatography (H ２ O / methanol) and then purified by preparative HPLC (H ２ 0 / methanol) to give compound 23 (38.8 mg, 52%) as a white solid.

[0073]

[0074] Synthesis of (2R)-5-azido-2-(hydroxymethyl)tetrahydrofuran-3-ol (Compound 24) Deoxy-D-ribose (250 mg, 1.86 mmol), triethylamine (2.60 ml, 18.6 mmol), and NaN3 (1.21 g, 18.6 mmol) were added with DMC (945 mg, 5.59 mmol), and the reaction mixture was stirred at 0°C for 3 hours. The reaction mixture was concentrated, 10 ml of ethanol was added, and the solid was removed by filtration. The filtrate was concentrated under vacuum and purified by reverse phase column chromatography (H ２ O / methanol) and concentrated in vacuo to give compound 24 (176 mg, 59%).

[0075] Synthesis of 4-((1-((2R,5R)-4-hydroxy-5-(hydroxymethyl)tetrahydrofuran-2-yl)-1H-1,2,3-triazol-4-yl)methyl)-1-methyl-1,2,4-triazolidine-3,5-dione (Compound 25) Compound 24 (50 mg, 0.314 mmol), compound 3 (48.1 mg, 0.314 mmol), copper(II) sulfate pentahydrate (12.2 mg, 49.0 μmol), and sodium ascorbate (26.5 mg, 157 μmol) were added sequentially to a reaction flask. ２ A mixture of 0H2O / tBuOH / DCM (1:2:1, 4 mL) was added, and the reaction flask was fitted with a reflux condenser, covered with aluminum foil, and heated to 60°C. After stirring at this temperature for 36 hours, it was cooled to room temperature and the solvent was evaporated. The resulting crude product was purified by reversed-phase silica gel column chromatography (H ２ 0 / methanol) to give compound 25 (34.2 mg, 35%).

[0076] The synthesized compounds shown in Figure 11 were confirmed to modify insulin in the same way as compounds 1 to 3 shown in Figure 2B. Glycosylation was carried out essentially as described above (see Figure 3A). The compounds were reacted with insulin at a 1:1 ratio. The compounds were diluted in DMSO:H2O to 10 mM each. ２ The compounds were dissolved at a ratio of 0:9, and 2.5 μL was added to 500 μL of a 50 μM insulin solution (10 mM sodium phosphate buffer, pH 6.0) and mixed well (the final concentrations of compounds 1 to 3 were 50 μM, with 0.05% DMSO). Laccase (Amano Enzyme Inc.) was added to a final concentration of 250 nM, mixed well, and shaken at 37°C for 1 hour. The mixture was shaken while exposed to air.

[0077] After the reaction, the samples were purified by HPLC and analyzed by MALDI-TOF-MS to confirm the modification reaction. The results are shown in Figure 12. Both compounds were able to modify insulin through a mild reaction using laccase.

[0078] To evaluate the aggregation stability of glycosylated insulin, a ThT assay was performed as described above. Aggregation was induced in the same manner using insulin modified with compound 8 at 10 μM or 50 μM concentrations, or insulin, and the change in ThT fluorescence intensity over time was measured using a fluorometer. Six experiments were performed in replicates; each line represents the mean value, and the error bars represent the standard error of six samples. While the fluorescence of unmodified insulin began to increase 2 hours after the start of the assay, no aggregation was observed with the 8-modified insulin at 24 hours (Figure 13A).

[0079] Using samples immediately after the ThT assay, a filter trap assay to detect insoluble proteins in the sample was performed in the same manner as described above. Supporting the results of the ThT assay, the filter trap assay also showed that unmodified 50 μM insulin produced a large amount of aggregates ( Figure 13B ). Both results indicated that insulin modified with compound 8 was stable and did not aggregate.

[0080] As shown above, the aggregation of protein preparations can be suppressed by modifying tyrosine. Here, we have used insulin as an example to demonstrate the inhibition of aggregation and stabilization of protein preparations by tyrosine modification, but the method described herein can be applied to any protein preparation that does not contain tyrosine at a site involved in activity. Furthermore, tyrosine can be modified by a mild method using laccase with any compound that has a urazole structure, in addition to the compounds shown in Figure 11.

[0081] Furthermore, the present inventors have demonstrated that the tyrosine involved in insulin aggregation is TyrA14. Modification of TyrA14 was found to inhibit insulin aggregation with all compounds used for modification, suggesting that modification of TyrA14 can inhibit aggregation and produce stabilized insulin preparations. Modification with the highly hydrophilic compound 3 is particularly advantageous compared to compound 1, as it requires fewer synthesis steps and has been reported that the addition of a triazole structure does not affect protein-protein interactions.

Claims

1. A method for inhibiting aggregation and stabilizing a protein contained in a protein formulation, which method modifies tyrosine present in the protein to stabilize the protein formulation.

2. A method for stabilizing a protein preparation according to claim 1, characterized in that the method of modifying tyrosine comprises adding a modifying agent, which is a compound having a urazole structure to which a sugar is bound, to the protein and reacting it with laccase to modify the tyrosine present in the protein.

3. A method for stabilizing a protein preparation according to claim 1, characterized in that the reaction is carried out at a temperature in the range of 20°C to 50°C.

4. A method for stabilizing a protein preparation according to any one of claims 1 to 3, wherein the protein contained in the protein preparation is insulin.

5. A method for stabilizing a protein preparation according to claim 4, which is a stabilization method by modifying TyrA14 of the insulin A chain.

6. A protein preparation in which the protein in the protein preparation is modified and stabilized using a modifying agent in which a sugar is bound to a compound having a urazole structure.

7. The stabilized protein preparation according to claim 6, wherein the protein in the protein preparation is insulin or an insulin analogue.

8. The insulin preparation according to claim 7, wherein TyrA14 of the insulin A chain is glycosylated.

Images