Glycoprotein synthesis

(Cross-reference with related applications) This application applies to US Provisional Patent Application No. 60 / 419,265 (Filing Date October 16, 2002), US Provisional Patent Application No. 60 / 420,990 (Filing Date October 23, 2002), and US Provisional Patent Application No. 60 / Claim the priority of No. 441,450 (filed on January 16, 2003) and incorporate the entire disclosure of that specification into this specification.

(Statement of Invention Rights Created from Federal Government-Supported Research and Development) The present invention was created under government grants as grant numbers GM44154, GM62159 and GM66494 issued by the National Institutes of Health and grant numbers DE-FG03-00ER45812 issued by the Department of Energy (DOE). It was. The United States Government reserves the right to the present invention.

(Technical field of invention) The present invention relates to the fields of glycopeptides, glycoproteins, and related mimetics and methods for synthesizing glycopeptides, glycoproteins, and related mimetics.

Post-translational modifications of proteins by glycosylation can affect protein folding and stability, alter the intrinsic activity of the protein, and alter its interaction with other biomolecules. See, for example, Non-Patent Document 1. Since natural glycoproteins often exist as a population of many different sugar forms, it is difficult to analyze glycan structure and study glycosylation effects on protein structure and function. Therefore, systematic clarification of glycan function and development of improved glycoprotein therapeutic agents require methods for the synthesis of homogeneously glycosylated natural and non-natural proteins.

One of the conventionally known approaches for producing proteins with the desired glycosylation pattern is to use glycosidases to convert heterogeneous natural sugar proteins into simple homogeneous cores, allowing the sugars to be sequentially grafted with glycosyltransferases. There is. See, for example, Non-Patent Document 2. This approach has the disadvantage that the major glycosylation site is pre-determined by the cell line expressing the protein. Alternatively, a glycopeptide containing a desired glycan structure can be synthesized by solid phase peptide synthesis. Coupling of this glycopeptide with other peptides or recombinant protein fragments by natural chemical ligation (see, eg, Non-Patent Document 3), expressed protein ligation (see, eg, Non-Patent Document 4), or recombinant protease is greater. It can be a sugar protein. See, for example, Non-Patent Document 5. Both natural chemical ligation and expressed protein ligation are most effective for small proteins and require a cysteine residue at the N-terminus of the glycopeptide. When proteases are used to ligate peptides to each other, the ligation site should be located away from the glycosylation site for good coupling yield. See, for example, Non-Patent Document 5. The third approach is to directly modify the protein with sugar using a chemical method. Coupling a haloacetamide sugar derivative with a thiol group of cysteine gives good selectivity (eg, Non-Patent Document 6; and Non-Patent Document 7), but this method is a protein with two or more cysteine residues. Then it can be a problem.

Therefore, there is a need for improved methods for producing glycoproteins with the desired glycosylation pattern. The present invention satisfies the above and other needs, as will be apparent from the disclosure below.<nplcit num="1"><text>Varki, A. (1993) Glycobiology 3: 97-130</text></nplcit><nplcit num="2"><text>Witte, K. et al., (1997) J.Am.Chem.Soc.119: 2114-2118</text></nplcit><nplcit num="3"><text>Shin, Y. et al., (1999) J.Am.Chem.Soc.121:11684-11689</text></nplcit><nplcit num="4"><text>Tolbert, TJand Wong, C.-H. (2000) J.Am.Chem.Soc.122: 5421-5428</text></nplcit><nplcit num="5"><text>Witte, K. et al., (1998) J.Am.Chem.Soc.120: 1979-1989</text></nplcit><nplcit num="6"><text>Davis, NJand, Flitsch, SL (1991) Tetrahedron Lett. 32: 6793-6796</text></nplcit><nplcit num="7"><text>Macmillan, D .; et al., (2002) Org Lett 4: 1467-1470</text></nplcit>

<p> The present invention provides a method for synthesizing a glycoprotein. In a predetermined embodiment, these methods include the step of incorporating an unnatural amino acid containing a first reactive group into a protein; the sugar moiety containing a second reactive group is brought into contact with the protein to form a first reactive group. It involves the step of reacting with a second reactive group to bind the sugar moiety to an unnatural amino acid. Glycoproteins produced by these methods are also included in the present invention. The first reactive group is, in certain embodiments, an electrophilic moiety (eg, a keto moiety, an aldehyde moiety, and / or an equivalent) and the second reactive group is a nucleophilic moiety. In certain embodiments, the first reactive group is a nucleophilic moiety and the second reactive group is an electrophilic moiety (eg, keto moiety, aldehyde moiety, and / or equivalent). For example, the electrophilic moiety is attached to the sugar moiety and the nucleophilic moiety is attached to an unnatural amino acid. The sugar moiety can contain a single sugar moiety, or the sugar moiety can contain more than one sugar moiety.</p>

<p> In certain embodiments, the method further comprises the glycosyltransferase, the glycosyltransferase moiety, and other reactants required for glycosyltransferase activity under sufficient time and suitable conditions to transfer the sugar from the sugar donor moiety to the sugar moiety. Includes the step of contacting the sugar moieties. The product of this reaction can optionally be contacted with at least a second glycosyltransferase and a suitable sugar donor moiety.</p><p> In certain embodiments, the method further comprises the sugar moiety β1-4N-acetylglucosaminyltransferase, α1,3 fucosyltransferase, α1,2 fucosyltransferase, α1,4 fucosyltransferase, β1-4galactosyltransferase, sialyltransferase, and / or. Including the step of contacting with one or more of the equivalents to form a bilateral or trilateral oligosaccharide structure.</p><p> In one embodiment, the sugar moiety comprises the terminal GlcNAc, the sugar donor moiety is UDP-Gal, and the glycosyltransferase is β-1,4-galactosyltransferase. In one embodiment, the sugar moiety comprises the terminal GlcNAc, the glycosyl donor moiety is UDP-GlcNAc, and the glycosyltransferase is β1-4N-acetylglucosaminyltransferase. Optionally, the method further comprises contacting the product of the N-acetylglucosaminyltransferase reaction with β1-4 mannosyltransferase and GDP-mannose to form a sugar moiety containing Manβ1-4GlcNAc β1-4GlcNAc-. Optionally, the method further contacts the Manβ1-4GlcNAcβ1-4GlcNAc- moiety with α1-3 mannosyltransferase and GDP-mannose to form a sugar moiety containing Manα1-3Manβ1-4GlcNAcβ1-4GlcNAc-. Including stages. Optionally, the method further contacts the Manα1-3Manβ1-4GlcNAcβ1-4GlcNAc-part with α1-6 mannosyltransferase and GDP-mannose to form a sugar moiety containing Manα1-6 (Manα1-3) Manβ1-4GlcNAcβ1-4GlcNAc-. Including stages. In some cases, the method further contacts the Manα1-6 (Manα1-3) Manβ1-4GlcNAcβ1-4GlcNAc-part with β1-2N-acetylglucosaminyltransferase and UDP-GlcNAc and Manα1-6 (GlcNAcβ1-2Manα1-3) Manβ1. -4GlcNAc β1 Includes the step of forming a sugar moiety containing -4GlcNAc-. In some cases, the method further contacts the Manα1-6 (GlcNAcpl-2Manα1-3) Manβ1-4GlcNAcβ1-4GlcNAc-part with β1-2N-acetylglucosaminyltransferase and UDP-GlcNAc to contact GlcNAcβ1-2Manα1-6 (GlcNAcβ1-2Manα1). -3) Manβ1-4GlcNAc Including the step of forming a sugar moiety containing β1-4GlcNAc-.</p><p> The step of incorporating an unnatural amino acid containing the first reactive group into a protein uses an orthogonal tRNA / orthogonal aminoacyl tRNA synthetase (O-tRNA / O-RS) pair in a predetermined embodiment, and the O-tRNA recognizes a selector codon. Then, in response to the selector codon, the unnatural amino acid is integrated into the protein, and O-RS preferentially aminoacylates the O-tRNA with the unnatural amino acid. For example, O-RS comprises an amino acid sequence comprising any one of SEQ ID NOs: 1, 2, or 3. In some cases, O-tRNA is a mutRNA^Tyr_CUAincluding. In certain embodiments, the unnatural amino acid is incorporated into the polypeptide in vivo.</p><p> The present invention further provides a glycoprotein containing a sugar moiety and a polypeptide. In certain embodiments, in the glycoprotein of the invention, the sugar moiety is the reaction generation of a nucleophilic reaction of a first reactive group attached to an unnatural amino acid present in the polypeptide and a second reactive group attached to the sugar moiety. It is bound to the polypeptide by the substance. In certain embodiments, the first reactive group is an electrophilic moiety (eg, keto moiety, aldehyde moiety, and / or equivalent) and the second reactive group is a nucleophilic moiety.</p><p> In certain embodiments, the nucleophilic portion of the invention includes, but is not limited to, hydrazide, hydroxylamine, semicarbazide, carbohydrazide, sulfonylhydrazide and the like. For example, the nucleophilic part is not limited, but -NR¹-NH₂(Hydrazide), -NR¹(C = O) NR²NH₂(Semicarbazide), -NR¹(C = S) NR²NH₂(Thiosemcarbazide),-(C = O) NR¹NH₂(Carbonyl hydrazide),-(C = S) NR¹NH₂(Thiocarbonyl Hydrazide),-(SO₂) NR¹NH₂(Sulfonyl Hydrazide), -NR¹NR²(C = O) NR³NH₂(Carbazide), -NR¹NR²(C = S) NR³NH₂(Thiocarbazide), and -O-NH₂(Hydroxylamine) etc. are included, and each R in the above formula¹, R², And R³Is independently H, or an alkyl having 1 to 6 carbon atoms.</p><p> In certain aspects of the invention, the reaction products of the invention include, for example, oxime, amide, hydrazone, reduced hydrazone, carbohydrazone, thiocarbohydrazone, sulfonylhydrazone, semicarbazone, thiosemicarbazone and the like.</p><p> Another aspect of the present invention is a method for synthesizing a sugar protein by incorporating an unnatural amino acid containing a sugar moiety into the protein. The glycoprotein produced by the above method is also a feature of the present invention. In certain embodiments, the integration step uses an orthogonal tRNA / orthogonal aminoacyl tRNA synthetase (O-tRNA / O-RS) pair, where the O-tRNA recognizes the selector codon and responds to the selector codon with a sugar moiety (eg, β). Non-natural containing -O-GlcNAc-L-serine, tri-acetyl-β-GlcNAc-serine, tri-O-acetyl-GalNAc-α-threonine, or α-GalNAc-L-threonine, and / or equivalent) Incorporating amino acids into proteins, O-RS preferentially aminoacyl O-tRNA with unnatural amino acids Turn into a le. In one aspect, the integration step is performed in vivo. For example, O-RS comprises an amino acid sequence comprising any one of SEQ ID NOs: 4, 5, or 6 or a polynucleotide comprising any one of the polynucleotide sequences of SEQ ID NO: 8, 9, or 10. Coded. In some cases, O-tRNA is a mutRNA^Tyr_CUAincluding. These methods further transfer glycosyltransferases, glycosyltransferase moieties, and other reactants and sugar moieties required for glycosyltransferase activity in sufficient time and under appropriate conditions to transfer the sugar from the sugar donor moiety to the sugar moiety. Can include the steps of contacting.</p><p> In certain embodiments, the method further comprises contacting the product of the glycosyltransferase reaction with at least a second glycosyltransferase and a second sugar donor moiety. In one embodiment, the sugar moiety comprises the terminal GlcNAc, the glucose donor portion is UPD-GlcNAc, and the glycosyltransferase is β1-4N-acetylglucosaminyltransferase. In another aspect, the sugar moiety comprises the terminal GlcNAc, the sugar donor moiety is UDP-Gal, and the glycosyltransferase is β-1,4-galactosyltransferase. Additive sugar can be added.</p><p> In certain embodiments, the glycosyltransferases of the invention are not limited, but galactosyltransferases, fucosyltransferases, glucosyltransferases, N-acetylgalactosaminyltransferases, N-acetylglucosaminyltransferases, glucuronyltransferases, sialyltransferases, mannosyltransferases, Includes glucuronic acid transferase, galacturonic acid transferase, oligosaccharyl transferase and the like.</p><p> The present invention also provides host cells useful for the synthesis of glycoproteins (eg, mammalian cells, yeast cells, bacterial cells, plant cells, fungal cells, progenitor cells, insect cells, and / or equivalents). These host cells include a) unnatural amino acids containing sugar moieties and; b) orthogonal tRNAs that recognize selector codons; c) orthogonal aminoacyl-tRNA synthetases (O-RS) that catalyze the binding of unnatural amino acids to orthogonal tRNAs. And; d) a polynucleotide encoding a glycosyltransferase and; e) a polynucleotide sequence encoding a polypeptide and containing at least one selector codon.</p><p> Compositions comprising a translation system are also provided by the present invention. The translation system contains orthogonal tRNA (O-tRNA) and orthogonal aminoacyl tRNA synthetase (O-RS), where O-RS is the sugar moiety (eg β-O-GlcNAc-L-serine, tri-acetyl-β-GlcNAc- Preferentially aminoacylate O-tRNA with unnatural amino acids including serine, tri-O-acetyl-GalNAc-α-threonine, α-GalNAc-L-threonine, and / or equivalents, with at least 1 O-tRNA. Recognize individual selector codons. In certain embodiments, the O-RS comprises an amino acid sequence comprising any one of SEQ ID NOs: 4, 5, or 6, or a poly comprising a polynucleotide sequence of any one of SEQ ID NOs: 8, 9, or 10. Encoded by nucleotides. In some cases, O-tRNA is a mutRNA^Tyr_CUAincluding.</p><p> Artificial (eg, artificially made non-natural) polypeptides and polynucleotides are also features of the invention. For example, the artificial polypeptide of the present invention is, for example, a polypeptide containing the amino acid sequence shown in any one of SEQ ID NOs: 4 to 6; (b) the polynucleotide sequence shown in any one of SEQ ID NOs: 8 to 10. A polypeptide comprising the amino acid sequence encoded by; (c) (a), or a polypeptide specifically immunoreactive to an antibody specific for the polypeptide of (b); and (d) (a). , (B), or (c) contains an amino acid sequence containing a conservative variant. Antibodies or antisera that are specifically immunoreactive with the artificial polypeptides of the invention are also provided by the invention. The artificial polynucleotide of the present invention is, for example, (a) a polynucleotide containing the nucleotide sequence set forth in any one of SEQ ID NOs: 8 to 10; A polynucleotide that is complementary to or encodes a tide sequence; (c) A polynucleotide that encodes a polypeptide containing the amino acid sequence set forth in any one of SEQ ID NOs: 1 to 6 or a conserved variant thereof; d) A polynucleotide encoding an artificial polypeptide; (e) hybridizes to a polynucleotide of (a), (b), (c), or (d) under high stringent conditions over substantially the entire length of the nucleic acid. Nucleic acid; polynucleotides that match at least 98% of the polynucleotides of (f) (a), (b), (c), (d), or (e); and (h) (a), (b), ( Includes polynucleotides containing conserved variants of c), (d), (e), or (f). [<u style="single">Definition</u>〕 </p><p> Before describing the present invention in detail, it should be understood that the present invention is not limited to specific devices or biological systems, and of course can be applied to various things. Similarly, it should be understood that the terms used herein are for the purpose of describing only certain aspects and are not limiting. The singular form used herein and in the claims also includes the plural, unless it is clear from the content that this is not the case. Thus, for example, the term "cell" includes a combination of two or more surfaces, the term "bacteria" includes a bacterial mixture, and so on for other terms.</p><p> Orthogonal: The term "orthogonal" as used herein is a molecule used inefficiently by a corresponding molecule inherent in a cell or other translation system (eg, an orthogonal tRNA (O-tRNA) and / or an orthogonal aminoacyl-tRNA synthetase). (O-RS)). Orthogonal means that the orthogonal tRNA cannot function with the endogenous tRNA synthetase or can only function with low efficiency in the translation system, or that the orthogonal RS cannot function with the endogenous tRNA or can only function with low efficiency, for example 20 It means that it can only function with an efficiency of less than%, less than 10%, less than 5%, or less than 1%. For example, the efficiency with which an orthogonal tRNA in the translation system is aminoacylated by the arbitrary endogenous RS of the translation system is low or zero compared to the efficiency with which the endogenous tRNA is aminoacylated by the endogenous RS. In another example, the efficiency with which the orthogonal RS aminoacylates any endogenous tRNA in the translation system is low or zero compared to the efficiency with which the endogenous RS aminoacylates the endogenous tRNA.</p><p> Priority aminoacylation: The term "priority aminoacylation" is the starting material that O-RS uses to make natural tRNAs or O-tRNAs with the efficiency of aminoacylating O-tRNAs with unnatural amino acids. It means that it is, for example, about 70%, about 75%, about 85%, about 90%, about 95%, or about 99% or more as compared with. In that case, the unnatural amino acid is incorporated into the growing polypeptide chain with high fidelity, for example, an efficiency of about 75% or more with a given selector codon, an efficiency of about 80% or more with a given selector codon, and a given selector codon. The efficiency is about 90% or more, the efficiency is about 95% or more with a given selector codon, or the efficiency is 99% or more with a given selector codon.</p><p> Selector codons: The term "selector codons" means codons that are recognized by O-tRNAs during the translation process, but are generally not recognized by endogenous tRNAs. The O-tRNA anticodon loop recognizes the selector codon on the mRNA and integrates its amino acid (eg, an unnatural amino acid) into this site within the polypeptide. Selector codons include, for example, nonsense codons (eg, stop codons such as amber, ocher and opal codons), codons of 4 or more bases, codons derived from natural or unnatural base pairs and / or equivalents. In a given system, the selector codon can also contain one of the natural 3-base codons if the endogenous system does not use the native 3-base codon, such as a system that lacks a tRNA that recognizes the natural 3-base codon. , A system in which the natural 3-base codon is a rare codon corresponds to this.</p><p> Suppressor tRNA: Suppressor tRNA is a messenger RNA in a given translation system. It is a tRNA that modifies the reading of mRNA). Suppressor tRNAs can skip, for example, stop codons, 4-base codons, rare codons, and / or equivalents.</p><p> Translation system: The term "translation system" means the components required to incorporate natural amino acids into a growing polypeptide chain (protein). Examples of the components of the translation system include ribosome, tRNA, synthetase, mRNA and the like. The components of the invention can be added to the translation system both in vivo and in vitro. The translation system can be cells, which can be prokaryotic cells (eg E. coli cells, protozoal cells, etc.) or eukaryotic cells (eg yeast, mammals, plants, insects, etc.).</p><p> Unnatural Amino Acids: As used herein, the term "unnatural amino acids" means one of 20 natural amino acids or any amino acid other than selenocysteine or pyrrolidine, modified amino acids, and / or amino acid analogs.</p><p> Sugar moiety: As used herein, the term "sugar moiety" means a natural and unnatural sugar moiety (ie, modifying an unnatural sugar moiety, such as a hydroxyl or amino position (eg, dehydroxylation, deamination, etc.). The esterified sugar moiety, such as 2-deoxyGal, is an example of an unnatural sugar moiety). The term "sugar" is a general formula (CH)₂O)_nExamples include, but are not limited to, monosaccharides, disaccharides, oligosaccharides, and polysaccharides. Oligosaccharides are chains composed of sugar units (also called sugars). The sugar units can be arranged according to arbitrary rules, and the bond between the two sugar units can be any of about 10 different methods.</p><p> The following abbreviations are used herein: Ara = Arabinosil; Fru = Fructsil; Fuc = Fukosil; Gal = galactosyl; GalNAc = N-Acetylgalactosaminyl; Glc = Glucosyl; GlcNAc = N-acetylglucosaminyl; Man = Mannosil; and NeuAc = sialyl (generally N-acetylneuraminyl).</p><p> Oligosaccharides are considered to have a reducing end and a non-reducing end, regardless of whether the reducing end sugar is actually a reducing sugar. According to accepted nomenclature, oligosaccharides have a non-reducing end on the left side and a reducing end on the right side. The total oligosaccharides described herein describe the glycosidic bond (α or β), the configuration of the ring bond, and the ring position of the reducing sugar involved in the bond, following the name or abbreviation (for example, Gal) of the non-reducing sugar. , Followed by the name or abbreviation of the reducing sugar (eg GlcNAc). The bond between the two sugars can be expressed as, for example, 2,3,2 3,2-3, or (2,3). Natural or non-natural bonds between two sugars (eg 1-2, 1-3, 1-4, 1-6, 2-3, 2-4, 2-6, etc.) are also included in the present invention. Each sugar is pyranose.</p><p> The term "sialic acid" (abbreviated as "Sia") means any member of the 9-carbon carboxylated sugar family. The most common member of the sialic acid family is N-acetyl-neuraminic acid (2-keto-5-acetamide-3,5-dideoxy-D-glycero-D-galactonuropyranose-1-onic acid) (Neu5Ac, It is often abbreviated as NeuAc or NANA). The second member of this family is N-glycolyl-neuraminic acid (Neu5Gc or NeuGc), which is the hydroxylated N-acetyl group of NeuAc. The third sialic acid family member is 2-keto-3-deoxy-nonulosonic acid ( KDN) (Nadano et al. (1986) J.Biol.Chem.261: 11550-11557; Kanamori et al. (1990) J.Biol.Chem.265:21811-21819). 9-OC₁-C₆Also included are 9-substituted sialic acids such as acyl-Neu5Ac (eg 9-O-lactyl-Neu5Ac or 9-O-acetyl-Neu5Ac), 9-deoxy-9-fluoro-Neu5Ac and 9-azido-9-deoxy-Neu5Ac. .. For the sialic acid family, see, for example, Varki (1992) Glycobiology 2: 25-40; Sialic Acids: Chemistry. Metabolism and Function, R. Schauer, Ed. (Springer-Verlag, New York (1992). Synthesis of sialic acid compounds). And its use in the sialylation method is described, for example, in the international application WO 92/16640 (published October 1, 1992).</p><p> The donor substrate for glycosyltransferases is an activated nucleotide sugar. Such activated sugars are generally composed of nucleoside diphosphate or derivatives of sugars having a monophosphate as a leaving group, uridine, guanosine diphosphate and cytidine monophosphate. Bacterial, plant, and fungal systems may optionally use other active nucleotide sugars.</p><p> Unless defined in this section and elsewhere in the specification, all scientific and technological terms used herein have the meanings commonly understood by those skilled in the art to which the present invention belongs. [Simple description of drawings]</p><p> FIG. 1 schematically shows an example of two types of schemes (sequential pathway and convergence pathway) for binding a sugar moiety to a polypeptide containing an unnatural amino acid.</p><p> Figure 2 shows HPLC analysis of the 7-hour and 26-hour coupling reactions of mutant Z-domain protein I (Figure 1) containing aminooxysaccharide 1 (Figure 1) and p-acetyl-L-phenylalanine.</p><p> Figure 3 shows the high-resolution MALDI-FTICR MS spectra of mutant Z-domain proteins I (Figure 1), glycoprotein mimetics II, III, and IV (Figure 1). 2 of each spectrum⁺Shows isotope clusters.</p><p> Figure 4 shows the expression of the Gly4 A mutant myoglobin (~ 18.5 kD). Protein Ni²⁺It was purified by affinity chromatography and degraded by SDS-PAGE. The gel was silver stained.</p><p> Figure 5 shows a MALDI-TOF analysis of the molecular weight of the Gly4 A mutant myoglobin.</p><p> FIG. 6 shows the characterization of purified mutant myoglobin, where A, B and C contain glycosylated amino acids. A shows the binding of GlcNAc-specific lectin Banderiraea simplicifolia II (BSII) to wild-type myoglobin and sugar myoglobin. B is UDP- [H³] Shows on-blot galactosyltransferase labeling of the sugar myoglobin with galactose. C showed a quantitative analysis of the galactosyltransferase reaction performed in solution and standardized the radiolabeled galactose so that 1.0 corresponds to 100% metastasis.</p>

Post-translational modifications of proteins regulate a number of biological processes, including metabolism, signal transduction, and gene expression. However, the combination associated with the production of a homogeneous population of selectively modified proteins Developmental problems have hampered detailed studies of the effects of these modifications on protein structure and function. For example, glycosylation is one of the most common post-translational modifications of proteins in eukaryotes, affecting a wide range of protein functions from folding and secretion to biomolecule recognition and serum half-life. See, for example, RADwek, (1996) Chem. Rev. 96: 683. Although the effects of glycosylation have been elucidated considerably, the specific role of oligosaccharide chains and the relationship between their structures and functions have only just begun to be elucidated. For example, CR Bertozzi, &amp; LLKiess1ing, (2001) Science See 291: 2357. The main problem is that it is difficult to isolate the intrinsic sugar form from a natural source because the sugar protein is generally produced as a mixture of sugar forms. Although various methods have been developed for synthesizing sugar forms of defined structure, there are significant drawbacks in the size, quantity, and / or quality of the sugar proteins produced. For example, P. Sears, &amp; CH Wong, (2001) Science 291: 2344; M. Wacker et al., (2002) Science 298: 1790; BG Davis, (2002) Chem. Rev. 102: 579; and H CHang, &amp; CR Bertozzi, (2001). ) See Acc.Chem.Res.34:727. The present invention solves this and other problems and provides glycoproteins and glycoprotein mimetics, as well as methods for synthesizing glycoproteins with the desired glycosylation pattern. The glycoproteins and glycoprotein mimetics of the present invention are useful for facilitating research on the production of therapeutic glycoproteins in homogeneous sugar forms and / or the structure and function of glycosylated proteins. [<u style="single">Glycosylation</u>〕

The present invention provides a method for synthesizing a glycoprotein. In certain embodiments, these methods involve incorporating an unnatural amino acid, including a first reactive group, into a protein; contacting and sharing a second reactive group attached to a sugar moiety and a first reactive group. It comprises the step of forming a bond and binding the sugar moiety to the protein.

A wide variety of suitable reactive groups are known to those of skill in the art. Suitable reactive groups include, for example, amino, hydroxyl, carboxyl, carboxylic acid, carbonyl, alkenyl, alkynyl, aldehyde, ester, ether (eg thio-ether), amide, amine, nitrile, vinyl, sulfide, sulfonyl. , Phosphoryl, or similar chemical reactive groups. Other suitable reactive groups include, but are not limited to, maleimide, N-hydroxysuccinimide, sulfone-N-hydroxysuccinimide, nitrilotriacetic acid, activated hydroxyl, haloacetyl (eg, bromoacetyl, iodoacetyl), activated carboxyl, hydrazide, Epoxy, aziridine, sulfonyl chloride, trifluoromethyldiazilidine, pyridyl disulfide, N-acyl-imidazole, imidazole carbamate, vinyl sulfone, succinimidyl carbonate, aryl azide, anhydride, diazoacetic acid, benzophenone, isothiocyanic acid, isocyanic acid, imide Examples include esters, fluorobenzenes, biotins and avidins.

In certain embodiments, one of the reactive groups is an electrophilic moiety and the second reactive group is a nucleophilic moiety. The nucleophilic or electrophilic moiety can be attached to the side chain of the unnatural amino acid, in which case the corresponding group is attached to the sugar moiety. Suitable electrophilic moieties that react with nucleophilic moieties to form covalent bonds are known to those of skill in the art. Examples of such an electrophilic moiety include, but are not limited to, a carbonyl group, a sulfonyl group, an aldehyde group, a ketone group, a hindered ester group, a thioester group, a stable imine group, an epoxide group, an aziridine group and the like. Suitable nucleophilic moieties capable of reacting with electrophilic moieties are known to those of skill in the art. Examples of such a nucleophilic moiety include aliphatic or aromatic amines (eg, ethylenediamine). In other embodiments, the reactive group is -NR.¹-NH₂(Hydrazide), -NR¹(C = O) NR²NH₂(Semicarbazide), -NR¹(C = S) NR²NH₂(Thiosemcarbazide),-(C = O) NR¹NH₂(Carbonyl hydrazide),-(C = S) NR¹NH₂(Thiocarbonyl Hydrazide),-(SO₂) NR¹NH₂(Sulfonyl Hydrazide), -N R¹NR²(C = O) NR³NH₂(Carbazide), -NR¹NR²(C = S) NR³NH₂(Thiocarbazide), -O-NH₂(Hydroxylamine) and / or equivalent, and in the above formula, each R¹, R², And R³Is independently H, or an alkyl having 1 to 6 carbon atoms, preferably H. In one aspect of the invention, the reactive groups are hydrazide, hydroxylamine, semicarbazide, carbohydrazide, sulfonylhydrazide and the like.

The reaction products of the nucleophilic and electrophilic moieties generally incorporate the atoms originally present in the nucleophilic moiety. The nucleophilic moiety used as a typical bond obtained by reacting the nucleophilic moiety with an aldehyde or ketone and the nucleophilic moiety to react with the nucleophilic moiety (eg, aldehyde, ketone, and / or equivalent). Depending on the product), reaction products such as oxime, amide, hydrazone, reduced hydrazone, carbohydrazone, thiocarbohydrazone, sulfonylhydrazone, semicarbazone, thiosemicarbazone, or similar functional groups can be mentioned. Bonding with a carboxylic acid is commonly referred to as carbohydrazide or hydroxamic acid. The binding to sulfonic acid is commonly referred to as sulfonyl hydrazide or N-sulfonyl hydroxylamine. The resulting bond can then be stabilized by chemical reduction.

In certain embodiments, the glycoprotein is synthesized by incorporating an unnatural amino acid bound to the sugar moiety into the polypeptide. For example, an orthogonal O-tRNA / O-RS can be used that integrates an unnatural amino acid with a sugar moiety into a growing polypeptide chain in response to a selector codon. See, for example, the section "Manufacturing proteins with unnatural amino acids" herein. [<u style="single">Glycosyltransferase</u>〕

The present invention provides a method for further glycosylating an unnatural amino acid containing a sugar moiety or a sugar moiety bound to an amino acid. The glycosylation step is preferably carried out enzymatically using, for example, glycosyltransferases, glycosidases, or other enzymes known to those of skill in the art. In certain embodiments, multiple enzymatic steps are performed with a single reaction mixture containing two or more different glycosyltransferases. For example, the galactosylation step and the sialylation step can be carried out simultaneously by adding both sialyltransferase and galactosyltransferase to the reaction mixture.

In enzymatic glycosynthesis involving a glycosyltransferase reaction, the recombinant cells of the invention optionally contain at least one heterologous gene encoding a glycosyltransferase. Numerous glycosyltransferases and their polynucleotide sequences are known. See, for example, The WWW Guide To Cloned Glycosyltransferases (see World Web www.vei.co.uk/TGN/gt guide.htm). Glycosyltransferases Amino acid sequences and nucleotide sequences encoding glycosyltransferases capable of estimating amino acid sequences are also available from various public databases such as GenBank, Swiss-Prot, and EMBL.

Galactosyltransferases, fucosyltransferases, glucosyltransferases, N-acetylgalactosaminyltransferases, N-acetylglucosaminyltransferases, glucuronyltransferases, sialyltransferases, but not limited to glycosyltransferases that can be used in the cells of the present invention. Examples thereof include mannosyl transferase, glucuronic acid transferase, galacturonic acid transferase, oligosaccharyl transferase and the like. Suitable glycosyltransferases include those obtained from eukaryotes and prokaryotes.

Glycosyltransferase receptors are present on glycoproteins that are modified by the methods of the invention. Suitable receptors include, for example, Galβ1,4GalNAc-, Galβ1,3Ga. Examples include galactosyl receptors such as lNAc-, lacto-N-tetraose-, Galβ1,3GlcNAc-, Galβ1,4GlcNAc-, Galβ1,3Ara-, Galβ1,6GlcNAc-, and Galβ1,4Glc- (lactose). Other receptors are also known to those of skill in the art (see, eg, Paulson et al. (1978) J. Biol. Chem. 253: 5617-5624). In general, the receptor forms part of a sugar moiety bound to a glycoprotein.

The amount or concentration of enzyme is expressed in units of activity, which is a measure of the initial catalytic rate. One active unit catalyzes the formation of 1 μmol of product per minute at a given temperature (generally 37 ° C) and pH value (generally 7.5). Therefore, 10 units of enzyme is the catalytic amount of enzyme when 10 μmol of substrate is converted to 10 μmol of product in 1 minute at a temperature of 37 ° C. and a pH value of 7.5. The enzyme may be released in a solution and used, or may be fixed to a support such as a polymer. Therefore, the reaction mixture is initially substantially homogeneous, although some precipitates may form during the reaction.

The glycosylation reaction involves the appropriate glycosyltransferases and receptors, as well as activated nucleotide sugars that act as sugar donors for the glycosyltransferases. The reaction can further include other components that promote glycosyltransferase activity. These components include divalent cations (eg Mg)⁺²Or Mn⁺²), Materials required for ATP regeneration, phosphate ions, and organic solvents. The concentration or amount of the various reactants used in the process depends on the reaction conditions such as temperature and pH values and a number of factors such as the selection and amount of receptor sugar to be glycosylated. As the reaction medium, a dissolution auxiliary surfactant (for example, Triton or SDS) and an organic solvent such as methanol or ethanol may be further added, if necessary.

Oligosaccharides produced using the method of the present invention can be analyzed by methods known to those skilled in the art. For example, the sugar unit can be released from the sugar moiety by, for example, alkali β elimination, and separated from the polypeptide by gel filtration. The resulting oligosaccharides can then be separated from each other and completely analyzed using one or more standard techniques such as gel filtration, HPLC, thin layer chromatography, and ion exchange chromatography or a combination thereof. .. Complete structural analysis of purified oligosaccharide units requires determination of monosaccharide units, their ring morphology, configuration (D or L), anomeric bonds (α or β), position of bonds between sugars and their sequences. .. Furthermore, if a substituent is present, its position is confirmed. Methylation analysis can be used to determine the location of glycosidic bonds between monosaccharides. Anomer configuration of sugar residues is, for example¹It can be assigned using 1 NMR spectroscopy. The conditions and methods used to perform full structural sugar analysis are Beeley, Laboratory Techniques in Biochemistry and Molecular Biology, eds. Burdon and Knippenberg, Elsevier, Amsterdam (1985), Hounsell, Glycoanalysis Protocols, Meth. Mol. Biol. Vol.76, 1998, and El Rassi, Carbohydrate Analysis: High Performance Liquid Chromatography and Capillary It is outlined in Electrophoresis, Elsevier Science Ltd, Vol. 58 (1994).

Other techniques for fully characterizing oligosaccharide sugars include FAB-MS (fast atom bombardment-mass spectrometry), HPAE (high pH anion exchange chromatography) and NMR (nuclear magnetic resonance spectroscopy, especially¹H-NMR and¹³C-NMR) can be mentioned. These techniques are complementary. Examples of methods using these techniques to completely characterize the structure of oligosaccharides are Spellman et al., (1989) J. Biol. Chem. 264: 14100, and Stanley et al. (1988) J. Biol. Chem. 263: 11374. Other methods include positive ion fast atom bombardment mass spectrometry (FAB-MS) and gas chromatography-electron impact mass spectrometry (G). C / EI-MS) methylation analysis (see EPO application No. 89305153.2). [<u style="single">In vivo synthesis of glycoproteins</u>〕

In order to synthesize a glycoprotein in vivo, a polynucleotide encoding the relevant polypeptide may be introduced into an expression vector. The polynucleotide further comprises one or more selector codons at the desired positions of the sugar moiety. With unnatural amino acids (eg, unnatural amino acids containing moieties that can bind sugar moieties (eg, aldehyde or keto derivatized amino acids) or unnatural amino acids containing sugar moieties); with orthogonal tRNAs that recognize selector codons;; unnatural The expression vector is introduced into a host cell containing an orthogonal aminoacyl-tRNA synthetase (O-RS) that catalyzes the binding of an amino acid to an orthogonal tRNA. O-RS binds an unnatural amino acid to an orthogonal tRNA and then introduces the unnatural amino acid into an unfinished protein.

In certain embodiments, the host cell further comprises one or more polynucleotides encoding a glycosyltransferase. Such host cells can catalyze the addition of one or more sugars to the glycosylated moiety bound to an unnatural amino acid.

Several well-known methods are available as methods for introducing the target nucleic acid into the host cell, and any of them can be used in the present invention. These methods include fusion of DNA-containing bacterial protoplasts with recipient cells, electroporation, infection with gene guns and viral vectors, and the like. Bacterial cells can be used to amplify the number of plasmids containing the DNA constructs of the invention. By growing bacterial cells to the logarithmic phase, plasmids within the bacterium can be isolated by a variety of methods known in the art (see, eg, Sambrook, see below). In addition, a number of kits are commercially available for purifying plasmids from bacteria (eg EasyPrep®, FlexiPrep® (both Pharmacia Biotech); StrataClean® (Stratagene); and QIAprep ( See Registered Trademark) (Qiagen)). The isolated and purified plasmids are further manipulated to create other plasmids that can be used to transfect cells or integrate into the relevant vector to infect an organism.

Recombinant host cells can be cultured in conventional nutrient media appropriately modified to suit operations such as screening steps, promoter activation or transformation cell selection. These cells can optionally be cultured in transgenic organisms.

For example, other useful literature on cell isolation and culture (eg, subsequent nucleic acid isolation) includes Freshney (1994) Culture of Animal Cells, a Manual of Basic Technique, 3rd Edition, Wiley-Liss, New York and its references. Literature; Payne et al. (1992) Plant Cell and Tissue Culture in Liquid Systems, John Wiley &amp; Sons, Inc. New York, NY; Gamborg and Phillips (eds) (1995) Plant Cell, Tissue and Organ Culture; Fundamental Methods Springer Lab Manual, Springer-Verlag (Berlin Heidelberg New York) and Atlas and Parks (eds) The Handbook of Microbiological Media (1993) CRC Press, Boca Raton, FL.

Berger and Kimmel, Guide to Molecular Cloning Techniqu es, Methods in Enzymology volume 152 Academic Press, Inc., San Diego, CA (Berger); Sambrook et al., Molecular Cloning-A Laboratory Manual (3rd Edition)), Vol.1-3, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 2001 ("Sambrook") and Current Protocols in Molecular Biology, FM Ausubel et al., Current Protocols, a joint venture between Greene Publishing Associates, Inc. and John Wiley &amp; Sons, Inc., (2003 Addendum) ("Ausubel")). These textbooks describe mutagenesis, the use of vectors, promoters and many other related matters, such as selector codons for producing proteins containing unnatural amino acids, orthogonal tRNAs, orthogonal synthetases and their pairs. It also describes the preparation of genes containing it. [<u style="single">Production of proteins with unnatural amino acids</u>〕

Unnatural amino acid (for example, de coupling the sugar moiety also producing glycoproteins comprising a non-natural amino acids, including unnatural amino acids (e.g., an aldehyde or keto-derivatized amino acid) or a sugar moiety containing a portion as possible) a feature of the present invention is there. In certain embodiments, the present invention relates to the production of a glycoprotein comprising one or more unnatural amino acids linked to a suitable reactive group capable of forming a covalent bond when reacted with a second reactive group. In certain embodiments, the unnatural amino acid comprises an electrophilic moiety (eg, an aldehyde or keto-derivatized amino acid), the aldehyde or keto moiety is reacted with the nucleophilic moiety and the sugar moiety is attached to the polypeptide or protein. Proteins containing unnatural amino acids are synthesized by cells whose protein biosynthesis mechanism has been modified to accept genetically encoded additional amino acids using orthogonal tRNA / aminoacyl-tRNA synthetase (O-tRNA / O-RS) pairs. Will be done. In particular, cells include orthogonal tRNAs that recognize selector codons (eg, stop codons, 4-base codons, etc.) and orthogonal aminoacyl-tRNA synthetases capable of binding aldehydes or keto-derivatized amino acids to orthogonal tRNAs.

In certain embodiments, the present invention relates to the production of a glycoprotein containing one or more unnatural amino acids, including a sugar moiety. Proteins containing unnatural amino acids are synthesized by cells whose protein biosynthesis mechanism has been modified to accept genetically encoded additional amino acids using orthogonal tRNA / aminoacyl-tRNA synthetase (O-tRNA / O-RS) pairs. Will be done. In particular, cells contain orthogonal tRNAs that recognize selector codons (eg, stop codons, 4-base codons, etc.) and orthogonal aminoacyl-tRNA synthetases capable of binding unnatural amino acids with sugar moieties to orthogonal tRNAs.

With this technique, unnatural amino acids can be directly and site-specifically incorporated into proteins in vivo. The important point is that unnatural amino acids are added to the genetic repertoire rather than being replaced by one of the 20 standard amino acids. A protein can have one or more (identical or different) unnatural amino acids at specific positions in the protein. Unlike the initial method for derivatizing a protein, the use of an O-tRNA / O-RS pair does not derivatize this particular amino acid at each position in the protein where the particular amino acid is present, but rather in the protein. It is possible to produce a protein having an unnatural amino acid at only one position where a specific amino acid is present.

To produce glycoproteins, host cells and organisms suitable for in vivo integration of unnatural amino acids with orthogonal tRNA / RS pairs can be used. Host cells are genetically modified (eg, transformed, transduced or transfected) with one or more vectors expressing a vector encoding an orthogonal tRNA, an orthogonal tRNA synthetase, and a protein to be derivatized. Each of these components may be placed in the same vector, or each may be placed in a separate vector. The two components may be placed in the same vector and the third component may be placed in the second vector. The vector can be, for example, in the form of a plasmid, bacterium, virus, naked polynucleotide or polynucleotide conjugate.

The coding regions of orthogonal tRNAs, orthogonal tRNA synthetases, and proteins to be derivatized are functionally linked to functional gene expression control elements in the desired host cell. A typical vector contains a transcription and translation terminator, a transcription and translation initiation sequence, and a promoter useful for regulating the expression of a particular target nucleic acid. The vector optionally contains at least one independent terminator sequence, a sequence that allows replication of the cassette in eukaryotes or prokaryotes or both (eg, shuttle vectors), and selectable markers for both prokaryotes and eukaryotes. Includes a comprehensive expression cassette. Vectors are suitable for replication and / or integration in prokaryotes, eukaryotes, or preferably both. Giliman &amp; Smith, Gene 8:81 (1979); Roberts et al., Nature, 328: 731 (1987); Schneider, B. et al., Protein Expr. Purif. 6435:10 (1995); Berger and See Kimmel, supra; Sambrook, supra, and Ausubel, supra. A catalog of bacteria and bacteriophages useful for cloning is available, for example, from the ATCC, such as The ATCC Catalog of Bacteria and Bacteriophage (1992) Gherna et al. (Edited) published by the ATCC. Other basic procedures and theories of other aspects of sequencing, cloning and molecular biology are also described in Watson et al. (1992) Recombinant DNA Second Edition Scientific American Books, NY. Proteins and Polypeptides of Interest.

For example, a method for producing a glycoprotein includes a step of growing a cell containing at least one selector codon and a nucleic acid encoding the protein in a suitable medium, and a portion capable of binding an unnatural amino acid (for example, a portion capable of binding a sugar moiety). A step of providing an unnatural amino acid containing or a sugar moiety) and a step of producing a protein by incorporating the unnatural amino acid at a specific position of the protein during translation of the nucleic acid by at least one selector codon. Including. Cells also function with orthogonal tRNAs (O-tRNAs) that recognize selector codons; unnatural amino acids (eg, unnatural amino acids that contain moieties that can bind sugar moieties or unnatural amino acids that include sugar moieties. ) Includes orthogonal aminoacyl-tRNA synthetase (O-RS) that preferentially aminoacylates O-tRNA. Published WO 2002/085923, title of invention "IN VIVO INCORPORATION OF UNNATURAL AMINO ACIDS) describes this method and is incorporated herein by reference. For example, when an O-tRNA / O-RS pair is introduced into a host, the pair produces an unnatural amino acid that can be externally added to the growth medium in response to a selector codon (eg, a moiety that can bind a sugar moiety). Induces in vivo protein integration of unnatural amino acids containing or unnatural amino acids containing sugar moieties. In some cases, the composition of the present invention may be introduced into an in vitro translation system or an in vivo system. Application for reference dated October 15, 2003, incorporated herein by reference, title of invention "Site Specific Incorporation of Keto Amino Acids into Proteins" (agent reference number 54-000170PCT) ) Also see.

The cells of the present invention are capable of synthesizing large amounts of useful amounts of glycoproteins. On one side, the composition may optionally be, for example, at least 10 μg, at least 50 μg, at least 75 μg, at least 100 μg, at least 200 μg, at least 250 μg, at least 500 μg, at least 1 mg, at least 10 mg or more, or an amount achievable by the in vivo protein production method. Contains glycoproteins (see book for more information on recombinant protein production and purification) Described in the specification). On the other side, the protein may optionally be in, for example, a cell lysate, buffer, pharmaceutical buffer, or other suspension (eg, in any volume of about 1 nl to about 100 L), eg, at least 10 μg protein / l, at least. Concentrations of 50 μg protein / l, at least 75 μg protein / l, at least 100 μg protein / l, at least 200 μg protein / l, at least 250 μg protein / l, at least 500 μg protein / l, at least 1 mg protein / l, or at least 10 mg protein / l Is present in the composition. Large amounts (eg, other methods, eg in vitro) of proteins incorporating at least one unnatural amino acid (eg, an unnatural amino acid containing a moiety capable of binding a sugar moiety or an unnatural amino acid containing a sugar moiety). Production of (more than is generally possible in translation) is also a feature of the invention.

Incorporation of unnatural amino acids (eg, unnatural amino acids containing moieties capable of binding sugar moieties or unnatural amino acids containing sugar moieties) includes, for example, dimensions, acidity, nucleophilicity, hydrogen bonding, hydrophobicity, protease targeting sites. It can be carried out, for example, to modify the protein structure and / or function to modify accessibility, target approach to the protein moiety, etc. Proteins containing unnatural amino acids (eg, unnatural amino acids containing a moiety capable of binding a sugar moiety or unnatural amino acids containing a sugar moiety) can enhance catalytic or physical properties or can be entirely novel. For example, by incorporating an unnatural amino acid (for example, an unnatural amino acid containing a portion capable of binding a sugar moiety or an unnatural amino acid containing a sugar moiety) into a protein, toxicity, biological distribution, structural properties, and spectroscopy may be possible. It modifies properties such as protein properties, chemical and / or photochemical properties, catalytic ability, half-life (eg, serum half-life), and reactivity with other molecules (eg, shared or non-shared). Compositions containing proteins incorporating at least one unnatural amino acid (eg, an unnatural amino acid containing a moiety capable of binding a sugar moiety or an unnatural amino acid containing a sugar moiety) are, for example, novel therapeutic, diagnostic, catalytic enzymes. , Industrial enzymes, binding proteins (eg, antibodies), and, for example, useful for studying protein structure and function. See, for example, Doherty, (2000) Unnatural Amino Acids as Probes of Protein Structure and Function, Current Opinion in Chemical Biology, 4: 645-652.

In one aspect of the invention, the composition is at least one, eg, at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10. , Or at least one unnatural amino acid incorporating 11 or more unnatural amino acids (eg, an unnatural amino acid containing a moiety capable of binding a sugar moiety or an unnatural amino acid containing a sugar moiety) and / or another unnatural amino acid. Contains protein. The unnatural amino acids may be the same or different, eg 1, 2, 3, 4, 5, 6, 7, 8, 9, or 1, 2, 3, 4, containing 10 or more different unnatural amino acids. 5, 6, 7, 8, 9, or 10 or more different sites can be present in the protein. In another aspect, the composition contains less than all of the specific amino acids present in the protein, but at least one unnatural amino acid (eg, an unnatural amino acid containing a portion capable of binding a sugar moiety or an unnatural containing a sugar moiety. Contains proteins substituted with (amino acids). In a given protein that incorporates two or more unnatural amino acids, the unnatural amino acids may be the same or different (eg, the protein may incorporate two or more different types of unnatural amino acids, or two. The same unnatural amino acid may be incorporated). In a given protein that incorporates three or more unnatural amino acids, the unnatural amino acids may be the same or different, or may be a combination of multiple unnatural amino acids of the same species and at least one other unnatural amino acid. Good.

Using the compositions and methods described herein, unnatural amino acids (eg, unnatural amino acids containing a moiety that binds a sugar moiety or an unnatural amino acid containing a sugar moiety) (and, eg, comprising one or more selector codons). Almost any protein (or portion thereof) incorporating the corresponding arbitrary coding nucleic acid) can be produced. Don't try to identify hundreds of thousands of known proteins Or any of the known proteins to incorporate one or more unnatural amino acids, for example by adjusting the available arbitrary mutation method to incorporate one or more suitable selector codons into the translation system. Can be modified. General sequence depository institutions for known proteins include GenBank, EMBL, DDBJ and NCBI. Other depository institutions can also be easily confirmed by searching the Internet.

In general, proteins are any available protein (eg, therapeutic protein, diagnostic protein, industrial enzyme, or a portion thereof, etc.) and, for example, at least 60%, at least 70%, at least 75%, at least 80%, at least 90%, It matches at least 95%, or at least 99% or more, and contains one or more unnatural amino acids. Examples of therapeutic, diagnostic, and other proteins that can be modified to incorporate one or more unnatural amino acids, such as unnatural amino acids that contain a sugar moiety or an unnatural amino acid that contains a sugar moiety. Is not limited, but WO2002 / 085923, It is described above. Examples of therapeutic, diagnostic, and other proteins that can be modified to incorporate one or more unnatural amino acids, including amino acids that bind sugar moieties, and / or unnatural amino acids, including sugar moieties, are not limited. , For example α1 antitrypsin, angiostatin, antihemopathic factor, antibody (details of the antibody will be described later), apolypoprotein, apoprotein, atrial sodium diuretic factor, atrial sodium diuretic polypeptide, atrial peptide, CXC Chemokines (eg T39765, NAP-2, ENA-78, Gro-a, Gro-b, Gro-c, IP-10, GCP-2, NAP-4, SDF-1, PF-4, MIG), calcitonin, CC chemokines (eg, monocytic chemoattractant protein-1, monocytic chemoattractant protein-2, monocytic chemoattractant protein-3, monocytic inflammatory protein-1α, monocytic inflammatory protein-1β, RANTES, I309, R83915, R91733, HCC1, T58847, D31065, T64262), CD40 ligand, C kit ligand, collagen, colony stimulator (CSF), complement factor 5a, complement inhibitor, complement receptor 1, cytokines (eg epithelial neutrophils) Activated peptide-78, GROα / MGSA, GROβ, GROγ, MIP-1α, MIP-1δ, MCP-1), epidermal growth factor (EGF), erythropoetin (EPO, by incorporation of one or more unnatural amino acids Suitable targets for modification), epidermal exfoliation toxins A and B, IX, VII, VIII, X, fibroblast growth factor (FGF), fibrinogen, fibronectin, G-CSF, GM-CSF, glucocerebrosidase, Gonadotropin, growth factor, hedgehog protein (eg Sonic, Indian, Desert), hemoglobin, hepatocellular growth factor (HGF), hirudin, human serum albumin, insulin, insulin-like growth factor (IGF), interferon (eg IFN-α, IFN-β, IFN-γ), interleukins (eg IL-1, IL-2, IL-3, IL-4, IL-5,

A composition for incorporating an unnatural amino acid described herein (for example, an unnatural amino acid containing a moiety capable of binding a sugar moiety or an unnatural amino acid containing a sugar moiety) in vivo. Transcription modulators and some of them are examples of proteins that can be produced using these methods. Examples of transcriptional modulators include genes that regulate cell proliferation, differentiation, regulation, etc. and transcriptional modulator proteins. Transcription modulators are present in prokaryotes, viruses and eukaryotes (eg, animals including fungi, plants, yeasts, insects, and mammals) and provide a wide range of therapeutic targets. As is self-evident, expression and transcriptional activators are, for example, binding to receptors, stimulation of signal transduction cascades, regulation of transcription factor expression, binding to promoters and enhancers, binding to proteins that bind to promoters and enhancers, DNA unwinding. Transcription is regulated by a number of mechanisms such as pre-promoter splicing, RNA polyadenylation and RNA degradation.

As one of the proteins of the present invention (for example, one or more unnatural amino acids containing an amino acid that binds a sugar moiety and / or a protein incorporating an unnatural amino acid containing a sugar moiety), an expression activator (for example, cytokine, inflammation) is used. Sex molecules, growth factors, their receptors, and tumor gene products such as interleukins (eg IL-1, IL-2, IL-8, etc.), interferons, FGF, IGF-I, IGF-II, FGF, PDGF, TNF, TGF-α, TGF-β, EGF, KGF, SCF / c-kit, CD40L / CD40, VLA-4 / VCAM-1, ICAM-1 / LFA-1 and hyalulin / CD44), signaling molecules and Corresponding tumor gene products (eg Mos, Ras, Raf and Met) and transcriptional activators and suppressors (eg p53, Tat, Fos, Myc, Jun, Myb, Rel, and steroid hormone receptors (eg estrogen, progesterone, testosterone) , Aldosterone, LDL receptor ligand and corticosterone)).

An enzyme (eg, an industrial enzyme) incorporating at least one unnatural amino acid (eg, an unnatural amino acid containing a portion that binds a sugar moiety or an unnatural amino acid containing a sugar moiety) or a portion thereof is also provided by the present invention. .. Examples of enzymes include, but are not limited to, amidase, amino acid racemase, acylase, dehalogenase, dioxygenase, diarylpropanperoxidase, epimerase, epoxide hydrolase, esterase, isomerase, kinase, glucose isomerase, glycosidase, glycosyl transferase, haloperoxidase, monooxygenase. (Eg p450s), lipase, lignin peroxidase, nitrile hydratase, nitrilase, protease, phosphatase, subtilisin, transamidase and nuclease.

Numerous proteins that can be modified by the present invention are commercially available (eg, Sigma). The BioSciences 2002 catalog and price list), the corresponding protein sequences and genes, and in general many of their variants are well known (see, eg, Genbank). For example, by inserting one or more unnatural amino acids containing an amino acid that binds a sugar moiety or an unnatural amino acid containing a sugar moiety according to the present invention so as to modify the protein with respect to one or more applicable therapeutic properties. Any can be modified. Examples of treatment-related properties are serum half-life, storage half-life, stability, immunogenicity, therapeutic activity, detectability (eg by addition of reporter groups (eg, label or label binding site) to unnatural amino acids), LD.₅₀Alternatively, reduction of other side effects, introduction into the body through the stomach (for example, oral availability), and the like can be mentioned. Examples of diagnostic-related properties include storage half-life, stability, diagnostic activity, detectability, specificity and the like. Examples of the relevant enzyme properties include storage half-life, stability, enzyme activity, production ability, specificity and the like.

Various other proteins can also be modified to contain one or more unnatural amino acids of the invention. For example, the present invention replaces one or more natural amino acids of one or more vaccine proteins with one or more unnatural amino acids containing an amino acid that binds a sugar moiety, or infects, for example, an unnatural amino acid containing a sugar moiety. Sex fungi (eg Aspergillus, Candida species); fine Fungi, especially Escherichia coli and medically important bacteria that can be used as pathogenic bacterial models (eg Staphylococci (eg aureus) or Streptococci (eg pneumoniae)); Entamoeba) and whipworms (Trypanosoma, Leishmania, Trichomonas, Giardia, etc.); Viruses (eg, (+) RNA viruses (eg, poxvirus (eg, vaccinia), picornavirus (eg, polio), togavirus (eg, ruin), Flavivirus (eg HCV) and Coronavirus), (-) RNA virus (eg Labdvirus (eg VSV), Paramixovirus (eg RSV), Orthomixovirus (eg Influenza), Bunyavirus and Arenavirus), dsDNA It can be incorporated into proteins derived from viruses (eg, leovirus), RNA DNA viruses (ie, retroviruses, eg HIV and HTLV), and certain DNA RNA viruses (eg, hepatitis B).

Insect resistant proteins (eg Cry protein), starch and lipid producing enzymes, plants and insect toxins, toxin resistant proteins, mycotoxin detoxifying proteins, plant growth enzymes (eg ribulose 1,5-bisphosphate carboxylase / oxygenase, "RUBISCO"), lipoxygenase Agricultural proteins such as (LOX) and phosphoenolpyruvate (PEP) carboxylase are also suitable targets for modification by unnatural amino acid integration and / or sugar addition of the present invention.

In certain embodiments, the protein or polypeptide (or portion thereof) in the methods and / or compositions of the invention is encoded by a nucleic acid. In general, a nucleic acid is at least one selector codon, at least two selector codons, at least three selector codons, at least four selector codons, at least five selector codons, at least six selector codons, and at least seven. Includes selector codons, at least 8 selector codons, at least 9 selector codons, and 10 or more selector codons. [<u style="single">Peptide characterization by immunoreactivity</u>〕

The sugar polypeptide of the present invention (for example, in the case of a protein synthesized by the translation system described herein, contains an unnatural amino acid containing an amino acid capable of binding a sugar moiety or an unnatural amino acid containing a sugar moiety. Or, for example, in the case of novel synthetases, it provides a variety of novel polypeptide sequences (including novel sequences of standard amino acids), so that glycopolypeptides also provide novel structural features that can be recognized, for example, in immunoassays. Preparation of an antiserum that specifically binds to the polypeptide of the present invention and a polypeptide bound to such an antiserum are also features of the present invention. The term "antibody" as used herein is not limited, but is a polypeptide substantially encoded by one or more immunoglobulin genes or fragments thereof that specifically bind to and recognize a sample (antigen). Means. Examples include polyclonal, monoclonal, chimeric, single-strand antibodies and the like. Immunoglobulin fragments such as Fab fragments and fragments produced by expression libraries containing phage displays are also included in the term "antibody" as used herein. Regarding antibody structure and terminology, for example, Paul, Fundamental See Immunology, 4th Ed., 1999, Raven Press, New York.

For example, the present invention specifically binds or specifically reacts to an antibody or antiserum prepared against an immunogen containing a synthetase amino acid sequence selected from one or more of the various sequences described herein. Contains the sex synthetase protein. To eliminate cross-reactivity with other homologues, use available synthetases such as wild-type M. janaschii tyrosyl synthetase (TyrRS) or known artificial synthetases such as those described in WO 2002/085923. Subtract antibody or antiserum. Wild-type Methanococcus jannaschii (M.janaschii) Tyrosyl synthetase ( If the TyrRS) or conventional sequence corresponds to a nucleic acid, optionally a nucleic acid-encoded polypeptide is prepared and used for antibody / antiserum subtraction purposes.

In one example of a typical format, the immunoassay is directed to one or more of the synthetase sequences described herein or to one or more polypeptides comprising a substantial subsequence thereof (ie, at least about 30% of the full length sequence described). The polyclonal antiserum prepared for this is used. Potential polypeptide immunogens derived from these sequences are collectively referred to as "immunogenic polypeptides" in the text below. The resulting antisera were optionally selected to be less cross-reactive to control synthetase homologues (wild TyrRs and / or synthetases as described in WO 2002/085923) and the polyclonal antisera were selected prior to use in immunoassays. Such cross-reactivity is eliminated, for example, by immunoadsorption of one or more control synthetase homologues.

To prepare antisera for immunoassays, one or more immunogenic polypeptides are prepared and purified as described herein. For example, recombinant proteins can be produced in recombinant cells. Immunize incoherent mice with a standard adjuvant (eg Freund's adjuvant) according to a standard mouse immunoprotocol (antibody production, immunoassay) For a standard description of the formats and conditions that can be used to measure specific immunoreactivity, see, for example, Harlow and Lane (1988) Antibodies. A Laboratory Manual, Cold Spring Harbor Publications, New. See York. Other aspects of the antibody are also described herein and can be applied to characterize the polypeptide by immunoreactivity in this case). Alternatively, one or more synthetic or recombinant polypeptides derived from the sequences disclosed herein are conjugated to a carrier protein and used as an immunogen. Other details regarding proteins, antibodies, antisera, etc. are described in WO 2002/085923, supra.

Polyclonal sera are collected and titers against immunogenic polypeptides by immunoassay (eg, solid phase immunoassay using one or more immunogenic proteins immobilized on a solid support). Ten⁶A polyclonal antiserum having the above titers is selected, pooled, and subtracted with a control synthetase polypeptide to prepare a high titer polyclonal antiserum subtraction pool.

High titer polyclonal antiserum subtraction pools are tested for cross-reactivity to control homologues in a comparative immunoassay. In this comparative assay, we set discriminatory binding conditions for subtraction high-titer antiserum and compared the signal-to-noise ratio of high-titer polyclonal antiserum to immunogenic synthetase binding to that of control synthetase homologues. And at least about 5 to 10 times. That is, the stringency of the binding reaction is adjusted by adding a non-specific competitor such as albumin or skim milk powder and / or adjusting the salt conditions, temperature, and / or equivalent. These binding conditions are used in late assays to determine if the test polypeptide (the polypeptide compared to the immunogenic and / or control polypeptide) specifically binds to the subtraction polyclonal antiserum pool. .. In particular, test polypeptides that exhibit a signal-to-noise ratio of at least 2-5 times that of control synthetase homologues and at least about 1/2 the signal-to-noise ratio of immunogenic polypeptides under discriminatory binding conditions are known to be synthetases. It is the polypeptide of the present invention because it has substantial structural similarity to the immunogenic polypeptide in comparison.

In another example, a competitive binding format immunoassay is used to detect the test polypeptide. For example, as described above, cross-reactive antibodies are removed from the antiserum mixture pool by immunoadsorption of the control polypeptide. The immunogenic polypeptide is then immobilized on a solid support and the support is exposed to the subtraction antiserum pool. Add the test protein to the assay and sub Compete for binding to the traction antiserum pool. Compare the ability of the test protein to compete for binding of the immobilized protein to the subtraction antiserum pool with the ability of the immunogenic polypeptide added to the assay to compete for binding (the immunogenic polypeptide is the antiserum pool). Effectively competes with immobilized immunogenic polypeptides). Calculate the cross-reactivity percentage of the test protein using standard calculations.

In a parallel assay, the ability of the control protein to compete for binding to the subtraction antiserum pool is optionally compared to the ability of the immunogenic polypeptide to compete for binding to the antiserum. Again, standard calculations are used to calculate the cross-reactivity percentage of the control polypeptide. A test polypeptide has a subtraction pit when the cross-reactivity percentage of the test polypeptide is at least 5-10 times that of the control polypeptide or when the binding of the test polypeptide is approximately in the same range as the binding of the immunogenic polypeptide. It is said to specifically bind to the serum pool.

In general, in the competitive binding immunoassays described herein, immunoadsorption antiserum pools can be used to compare optional test polypeptides to immunogenic and / or control polypeptides. To make this comparison, immunogenicity, test and control polypeptides are each assayed over a wide concentration range and inhibit 50% of the binding of subtraction antisera to, for example, immobilized control, test or immunogenic proteins. The amount of each polypeptide required to do so is determined by standard techniques. A test polypeptide is immunogenic when the amount of test polypeptide required for binding in a competitive assay is less than twice the amount required for an immunogenic polypeptide and at least about 5-10 times that of a control polypeptide. It is said to specifically bind to the antibody produced against the protein.

As an additional test of specificity, an antiserum pool may be used until little or no binding of the resulting immunogenic polypeptide subtraction antiserum pool to the immunogenic polypeptide used for immunoadsorption can be detected (in control polypeptides). Fully immunoadsorbed to immunogenic polypeptides. This fully immunoadsorbed antiserum is then tested for reactivity with the test polypeptide. If little or no reactivity is observed (ie, less than twice the signal-to-noise ratio observed in the binding of fully immunoadsorbed antiserum to immunogenic polypeptide), the test polypeptide is an immunogenic protein. It specifically binds to the antiserum induced by. [<u style="single">Pair of Orthogonal TRNA and Orthogonal Aminoacyl TRNA Synthetase</u>〕

A translation system suitable for the production of proteins containing one or more unnatural amino acids is International Patent Application No. WO 2002/086075, the title of the invention "Methods and compositions for making orthogonal tRNA-aminoacyl tRNA synthetase pairs (METHODS AND). COMPOSITION FOR THE PRODUCTION OF ORTHOGONAL tRNA-AMINOACYL tRNA SYNTHETASE PAIRS) and WO 2002/085923, supra. Each of these applications incorporates the entire disclosure content herein as reference material. Such translation systems generally include orthogonal tRNAs (O-tRNAs), orthogonal aminoacyl-tRNA synthetases (O-RSs), and unnatural amino acids (eg, aldehydes or unnatural amino acids that contain moieties capable of binding sugar moieties. Contains cells containing keto-derivative amino acids) or unnatural amino acids containing sugar moieties), and O-RS aminoacylates O-tRNA with unnatural amino acids. Cells use these components to integrate unnatural amino acids into the growing polypeptide chain.

Orthogonal pairs consist of O-tRNAs (eg suppressor tRNAs, frameshift tRNAs, etc.) and O-RSs. The O-tRNA is not acylated by the endogenous synthetase as described above and can decode the selector codon. O-RS recognizes O-tRNAs, for example, in an extended anticodon loop, and prioritizes O-tRNAs over unnatural amino acids (eg, unnatural amino acids containing moieties that can bind sugar moieties or unnatural amino acids containing sugar moieties). Aminoacylate. Various codons due to the development of multiple orthogonal tRNA / synthetase pairs Allows the simultaneous incorporation of multiple unnatural amino acids. See Example 5 for representative O-tRNA and O-RS sequences.

O-tRNAs and O-RSs may be naturally occurring or may be induced by mutating native tRNAs and / or RSs from various organisms listed by origin and host name. In various embodiments, the O-tRNA and O-RS are derived from at least one organism. In another aspect, the O-tRNA is derived from a naturally occurring or naturally occurring mutated tRNA from the first organism, and the O-RS is naturally derived from the second organism. Derived from mutated RSs that are present or naturally occurring.

Specifically, these methods include (a) creating a library of tRNAs derived from at least one tRNA derived from the first organism, and (b) RS derived from the first organism. The step of providing a pool of tRNAs by negative selection of a library of tRNAs that are aminoacylated by aminoacyl-tRNA synthetase (RS) derived from a second organism in the absence, and (c) the introduced orthogonal RS (O-). It comprises providing at least one recombinant O-tRNA by selecting a member that is aminoacylated by RS) from a pool of tRNAs. Recombinant O-tRNAs recognize selector codons, are not efficiently recognized by RS derived from the second organism, and are preferentially aminoacylated by O-RS. The method further includes (d) the step of creating a library of mutant RS derived from at least one aminoacyl-tRNA synthetase (RS) derived from a third organism, and (e) unnatural and natural amino acids. To provide a pool of active RS by selecting members from the RS library that preferentially aminoacylate recombinant O-tRNAs in the presence of (f) at least one species in the absence of unnatural amino acids. Including the step of providing a specific O-tRNA / O-RS pair by negatively selecting a pool of active RSs that preferentially aminoacylate the recombinant O-tRNA of the specific O-tRNA / O-RS pair. Contains at least one recombinant O-RS and recombinant O-tRNA specific for an unnatural amino acid (eg, an unnatural amino acid containing a moiety capable of binding a sugar moiety or an unnatural amino acid containing a sugar moiety). ..

One of the strategies to create orthogonal pairs is to create a mutant library and screen and / or select O-tRNA or O-RS from this library.

The second strategy for creating orthogonal tRNA / synthetase pairs is to introduce heterologous tRNA / synthetase pairs (eg, pairs of different biological origin) into the host cell. The properties of the heterologous synthetase candidate include, for example, not loading the host cell tRNA, and the properties of the heterologous tRNA candidate include, for example, not being acylated by the host cell synthetase. In addition, heterologous tRNAs derived from heterologous tRNAs are orthogonal to all host cell synthetases. [<u style="single">Orthogonal Aminoacyl-tRNA Synthetase (O-RS)</u>〕

The O-RS of the present invention preferentially in vitro or in vivo aminoacylation of O-tRNA with an unnatural amino acid (for example, an unnatural amino acid containing a moiety capable of binding a sugar moiety or an unnatural amino acid containing a sugar moiety). To do. The O-RS of the present invention can be provided to a translation system (for example, a cell or in vivo translation system) by a polypeptide containing O-RS and / or a polynucleotide encoding O-RS or a part thereof. For example, O-RS comprises the amino acid sequences set forth in SEQ ID NOs: 1-6 or conserved variants thereof. In another example, O-RS, or a portion thereof, is encoded by a polynucleotide sequence encoding an amino acid comprising SEQ ID NOs: 1-6 or a complementary polynucleotide sequence thereof, or SEQ ID NOs: 8, 9, or 10. Examples thereof include O-RS encoded by a polynucleotide containing a polynucleotide sequence containing any one of the above, or a part thereof. For example, the sequence of typical O-RS molecules is described in detail. See Table 2 and Example 5 of the detailed document. See also the section "Nucleic Acids and Polypeptide Sequences and Variants" herein.

To prepare O-RS, after preparing a pool of mutant synthetase from the skeleton of wild-type synthetase, the specificity for unnatural amino acids having, for example, an aldehyde or keto moiety or sugar moiety is compared with 20 standard amino acids. It is based on the selection of mutant RS. To isolate such synthetases, the selection method of the present invention can (i) reduce the activity of the desired synthetase from the first selection, is sensitive due to the small population, and (ii) each time. It is "adjustable" because it is desirable to change the selection stringency in the selection of, and (iii) it is versatile and can be used for a variety of unnatural amino acids.

The method for producing an orthogonal aminoacyl-tRNA synthetase includes, for example, a step of mutating synthetase at various sites by combining various synthetase domains, for example, a synthetase active site, a synthetase editing mechanism site, and a step of applying a selection process. Use a combined strategy that makes negative selections after positive selections. In positive selection, cells survive under positive selection pressure when the selector codon introduced at a non-essential position of the positive marker is suppressed. Thus, in the presence of both natural and unnatural amino acids, living cells encode an active synthetase that loads the orthogonal suppressor tRNA with natural or unnatural amino acids. In negative selection, synthetase with natural amino acid specificity is eliminated when the selector codon introduced at the non-essential position of the negative marker is suppressed in the absence of non-natural amino acids. Negatively and positively selected surviving cells encode synthetase that aminoacylates (loads) orthogonal suppressor tRNAs with unnatural amino acids only. These synthetases can then be further mutagenic, for example by DNA shuffling, other recursive mutagenesis methods, and / or equivalents.

A library of mutant RS can be made using various mutagenesis techniques known in the art. For example, mutant RS can be produced by site-specific mutation, random point mutation, homologous recombination, chimera construction, or the like. A chimeric library of RS is also included in the present invention.

Positive selection can be based on suppression of the selector codon in a positive selection marker that contains a selector codon (eg, an amber stop codon) in the selectable marker gene. Antibiotics or other selective substances can be applied as positive selective pressure. In addition, selectable markers can be used for both positive and negative markers as described herein in the presence and absence of unnatural amino acids. Optionally, a selectable marker gene containing a selector codon is used for positive selection and a negative selectable marker containing at least one selector codon (eg, a toxicity marker such as a balnase gene) is used for negative selection.

Positive selection can also be based on suppression of the selector codon at a non-essential position of the β-lactamase gene that makes cells resistant to ampicillin, using negative selection using ribonuclease balnase as a negative marker. The cytoplasmic localized chloramphenicol acetyltransferase (CAT) gene can also be used for β-lactamase secreted by the periplasm, and ampicillin is bactericidal, while chloramphenicol is bacteriostatic. It is sex.

After performing positive and negative selection / screening strategies on synthetases, these synthetases can be further mutagenic. For example, nucleic acids encoding O-RS can be isolated; (eg, random mutagenesis, site-specific bumps). However, a set of polynucleotides encoding mutant O-RS can be made from nucleic acids (by mutagenesis, recombination, or any combination thereof); O-tRNA can be attached to unnatural amino acids (eg, sugar moieties). These individual steps or combinations of these steps can be repeated until a mutant O-RS that preferentially aminoacylates with an unnatural amino acid containing a capable moiety or an unnatural amino acid containing a sugar moiety) is obtained. In one aspect of the invention, the steps are performed multiple times, eg, at least twice. In some cases, the concentration of the selected substance is varied.

In the methods of the invention, additional levels of selection / screening stringency can also be used to make O-tRNAs, O-RSs, or pairs thereof. Selection or screening stringency can be varied at one or both stages of the method for making O-RS. This can be, for example, a variation in the amount of selection / screening substance used. Additional positive and / or negative selections can also be made. Selection or screening can further include one or more of changes in amino acid permeability, changes in translation efficiency, changes in translation fidelity, and the like. In general, one or more changes are based on mutations in one or more genes in organisms that use orthogonal tRNA-tRNA synthetase pairs to produce proteins.

See WO 2002/086075, supra for other details regarding the production of O-RS, modification of the substrate specificity of synthetase, and other examples of O-RS. [<u style="single">Orthogonal tRNA (O-tRNAS)</u>〕

Orthogonal tRNAs (O-tRNAs) of the invention can bind unnatural amino acids (eg, sugar moieties) to proteins encoded by polynucleotides containing selector codons that are recognized, for example, in vivo or in vitro by O-tRNAs. It mediates the integration of moiety-containing unnatural amino acids (eg, aldehyde or keto derivatized amino acids) or unnatural amino acids containing a sugar moiety.

An example of the O-tRNA of the present invention is SEQ ID NO: 7. See Table 2 and Example 5 herein for the sequences of representative O-tRNA and O-RS molecules. See also the section "Nucleic Acids and Polypeptide Sequences and Variants" herein. In the tRNA molecule, thymine (T) is replaced with uracil (U). Additional modifications may be made to the base. The present invention also includes conserved variants of O-tRNA. For example, as a conserved variant of O-tRNA, a molecule that functions in the same manner as the O-tRNA of SEQ ID NO: 7 and maintains the tRNA L-form structure, but is not the same sequence (and is not a wild-type tRNA molecule). Can be mentioned. See also the section "Nucleic Acids and Polypeptide Sequences and Variants" herein.

A method for producing recombinant orthogonal tRNA (O-tRNA) is described in International Patent Application No. WO 2002/086075, supra.

For example, in order to improve tRNA orthogonality while maintaining its affinity for the desired RS, the method selected negative and positive selections by the mutant suppressor tRNA library in the absence and presence of cognate synthetase, respectively. Combine. Negative selection introduces a selector codon at a non-essential position on a marker gene (eg, a toxic gene such as a balnase gene). Selector codons (eg, amber codons), for example, if members of a mutant tRNA library derived from Methanococcus jannaschii are aminoacylated by an endogenous host (eg, E. coli) synthetase (ie, non-orthogonal to a host, eg, E. coli synthetase). ) Is suppressed, and the toxic gene product produced leads to cell death. Cells with orthogonal tRNAs or non-functional tRNAs survive. Then, the surviving cells are positively selected and the selector codon (for example, amber codon) is placed on a positive marker gene (for example, a drug resistance gene such as β-lactamase gene). .. These cells also contain an expression vector with cognate RS. These cells are grown in the presence of a selective substance (eg ampicillin). Then, a tRNA that can be aminoacylated by co-expressed cognate synthetase and can insert an amino acid in response to this selector codon is selected. Cells containing non-functional tRNAs or tRNAs that cannot be recognized by the synthetase are sensitive to antibiotics. Therefore, (ii) aminoacylation can be performed by the synthetase, which is not a substrate of the endogenous host (eg, Escherichia coli) synthetase, and (iii) the tRNA functional in translation survives after selection of both.

Build a library of mutant tRNAs. Introduce mutations into specific positions (eg, non-conserved or conserved positions), random positions, or combinations of both of the desired loops of tRNA (eg, anticodon loops (D arm, V loop, TΨC arm) or loop combinations or whole loops) be able to. A chimeric library of tRNAs is also included in the present invention. In some cases, a library of tRNA synthetases derived from various organisms (eg, microorganisms such as eubacteria or primordial bacteria) (eg, a library containing natural diversity) (eg, US Pat. No. 6,238,884 (Short et al.); US Pat. No. 5,756,316 (Schallenberger et al.); US Pat. No. 5,783,431 (Petersen et al.); US Pat. No. 5,824,485 (Thompson et al.); See US Pat. No. 5,958,672 (Short et al.)) May be constructed and screened for orthogonal pairs. ..

Specific positions (eg, non-conserved or conserved positions) of the desired loop or region of the tRNA (eg, anticodon loop, acceptor stem, D-arm or loop, variable loop, TΨC arm or loop, other region or combination of tRNA molecules). , Random positions or combinations of both can be introduced. In general, mutations in tRNA include mutations in the anticodon loop of each member of the library of mutant tRNAs that allow recognition of selector codons. This method can further include the addition of an additional sequence (CCA) to the 3'end of the O-tRNA. In general, O-tRNAs have improved orthogonality to the desired organism as compared to starting materials (eg, multiple tRNA sequences) while maintaining their affinity for the desired RS.

For example, in negative selection, negative selection markers, such as enzymes that confer antibiotic resistance (eg β-lactamase), enzymes that confer detectable products (eg β-galactosidase, chloramphenicol acetyltransferase (CAT)), For example, a selector codon is introduced at a non-essential position of a polynucleotide encoding a toxic substance (eg, balnase) (eg, a position that still produces functional balnase). Screening / selection is optionally performed by growing the cell population in the presence of a selective substance (eg, an antibiotic such as ampicillin). In one embodiment, the concentration of the selected substance is varied.

For example, to measure the activity of suppressor tRNA, in vivo suppression of selector codons (eg nonsense or frameshift mutations introduced into a polynucleotide encoding a negative selectable marker (eg β-lactamase gene (bla))). Use a selection system based on. For example, construct a polynucleotide variant (eg, a bla variant that contains a selector codon in place). Cells (eg bacteria) are transformed with these polynucleotides. In the case of orthogonal tRNAs that cannot be efficiently loaded by endogenous E. coli synthetase, antibiotic resistance (eg ampicillin resistance) should be equal to or less than that of non-plasmid transformed bacteria. Higher levels of antibiotic (eg ampicillin) resistance are observed when the tRNA is not orthogonal or when heterologous synthetases capable of loading the tRNA are co-expressed in the system. Select cells that cannot grow on LB agar plates with antibiotic concentrations comparable to those that have not been transformed with the plasmid (eg, bacteria).

In the case of toxic substances (eg, ribonuclease balnase), when multiple potential tRNA members are aminoacylated by an endogenous host (eg, E. coli) synthetase (ie, not orthogonal to the host, eg, E. coli synthetase). Selector codons are suppressed and the toxic polynucleotide products produced lead to cell death. Cells containing orthogonal tRNAs or non-functional tRNAs survive. In some cases, two or more amber codons may be added to the ribonuclease balnase gene. Surviving cells can be selected, for example, by using a comparative ratio cell density assay.

In one embodiment, a tRNA pool orthogonal to the desired organism is positively selected and the selector codon is placed on a positive selection marker encoded by a drug resistance gene, such as the β-lactamase gene. Positive selection is performed on cells containing a polynucleotide encoding or containing a member of a tRNA pool orthogonal to the cell, a polynucleotide encoding a positive selection marker, and a polynucleotide encoding Cognate RS. In certain embodiments, the second cell population comprises cells that have not been eliminated by negative selection. The polynucleotide is expressed in the cell and the cell is grown in the presence of a selective substance (eg ampicillin). Next, a tRNA that is aminoacylated by co-expressed cognate synthetase and capable of inserting an amino acid in response to this selector codon is selected. In general, these cells exhibit higher repressive efficiency than cells with non-functional tRNAs or tRNAs that cannot be efficiently recognized by the synthetase. Cells containing non-functional tRNAs, or tRNAs that are not efficiently recognized by the synthetase, are sensitive to antibiotics. Thus, (i) not a substrate for an endogenous host (eg, E. coli) synthetase; (ii) can be aminoacylated by the synthetase in question; (iii) translationally functional tRNAs survive after selection of both.

The stringency of selection in the above method (eg, positive selection, negative selection or both positive and negative selection) may optionally include variation in selective stringency. For example, since balnase is a highly toxic protein, the stringency of negative selection can be controlled by introducing different numbers of selector codons into the balnase gene and / or by using an inducible promoter. In another example, the concentration of the selected or screening substance (eg ampicillin) is varied. In one aspect of the invention, the stringency is varied because the desired activity can be reduced in the initial round. That is, the low stringency selection criteria are applied in the early rounds, and the high stringency criteria are applied in the late rounds. In certain embodiments, negative selection, positive selection or both positive and negative selection are repeated multiple times. Multiple different negative selection markers, positive selection markers or both positive and negative selection markers can be used. In certain embodiments, the positive and negative selectable markers can be the same.

In the present invention, for example, an unnatural amino acid (for example, an unnatural amino acid containing a moiety capable of binding a sugar moiety (for example, an aldehyde or keto derivatized amino acid) or an unnatural amino acid containing a sugar moiety) is used, for example, O-tRNA, O. Other type selections can also be used to make -RS, and O-tRNA / O-RS pairs. For example, a reporter can be used for both the positive selection stage, the negative selection stage or both the positive selection stage and the negative selection stage, and the reporter is detected by fluorescence activated cell sorting (FACS). For example, a positive selection marker (eg, chloramphenicol acetyltransferase (CAT) gene) is used, a selector codon (eg, amber stop codon) is added to the CAT gene, positive selection is first performed, and then negative selection screening is performed. It is possible to select a position in a negative marker (for example, T7 RNA polymerase gene) that transcribes another gene (for example, GFP) that cannot suppress the selector codon (for example, two or more). In one aspect, a positive selectable marker and a negative selectable marker Can be placed in the same vector (eg, plasmid). Negative marker expression induces expression of reporters (eg, green fluorescent protein (GFP)). The stringency of selection and screening can be varied, for example the light intensity required to fluoresce the reporter. In another embodiment, a reporter screened by the FAC is used as a positive selection marker to perform positive selection followed by negative selection screening, with selector codons (eg, two or more) at positions within the negative marker (eg, balnase gene). Can be selected that cannot be suppressed. See also Example 4 herein, for example.

In some cases, the reporter is presented on the cell surface, phage display, etc. The cell surface display (eg, the OpA cell surface display system) depends on the expression of specific epitopes (eg, poliovirus C-peptide fused to the outer membrane porin OppA) on the E. coli cell surface. Epitopes are presented to the cell surface only when the selector codon in the protein message is suppressed during translation. Thus, the peptides presented contain amino acids recognized by one of the mutant aminoacyl-tRNA synthetases in the library and can be isolated from cells containing the corresponding synthetase gene along with antibodies to peptides containing specific unnatural amino acids. it can. The OmpA cell surface display system was developed and improved by Georgiou et al. As an alternative to phage display. Francisco, JA, Campbell, R., Iverson, BL &amp; Georgoiu, G. Production and fluorescence-activated cell sorting of Escherichia coli expressing a functional antibody fragment on the external surface.Proc Natl Acad Sci See US A.90: 10444-8 (1993).

The selection step can also be performed in vitro. The selected components (eg synthetase and / or tRNA) can then be introduced into cells for in vivo integration of unnatural amino acids.

Other methods for producing recombinant orthogonal tRNA are also described, for example, in International Patent Application WO 2002/086075, supra. Forster et al., (2003) Programming peptidomimetic synthetases by translating genetic codes designed de novo PNAS 100 (11): 6353-6357; and Feng et al., (2003), Expanding tRNA recognition of a tRNA synthetase by a single amino acid change, PNAS 100 (10): 5676-5681 See also. [<u style="single">Resources and host organisms</u>〕

The translational components for producing the glycoproteins of the present invention are generally derived from non-eukaryotes. For example, orthogonal O-tRNAs are non-enuclear organisms (or combinations of organisms) such as Methanococcus jannaschii, Methanopyrus thermoautotrophicum, Halobacterium (eg Haloferax volcanii and Halobacterium species NRC-1), Archaeoglobus fulgidus, Pyrococcus furiosus, Pyrococcus horikoshii, Aeropyrum per. maripaludis, Methanopyrus kandleri, Methanosarcina mazei (Mm), Pyrobaculum aerophilum, Pyrococcus abyssi, Sulfolobus solfataricus (Ss), Sulfolobus tokodaii, Thermo plasma acidophilum, Thermoplasma It can be derived from primordial bacteria such as volcanium and eubacteria such as Escherichia coli, Thermos thermophilus, Bacillus stearothermphilus, and orthogonal O-RS is a non-enuclear organism (or combination of organisms) such as Methanococcus jannaschii, Methanopyrus thermoautotrophicum, Halobacterium ( For example, Haloferax volcanii and Halobacterium species NRC-1), Archaeoglobus fulgidus, Pyrococcus furiosus, Pyrococcus horikoshii, Aeropyrum pernix, Methanococcus maripaludis, Methanopyrus kandleri, Methanosarcina mazei, Pyrobaculum aerophilum, Pyrobaculum aerophilum It can be derived from primordial bacteria such as tokodaii, Thermoplasma acidophilum and Thermoplasma volcanium, and eubacteria such as Escherichia coli, Thermus thermophilus and Bacillus stearothermphilus. In one embodiment, eukaryotes such as plants (eg, monocotyledons, or complex plants such as dicotyledons), algae, protozoa, fungi, yeasts, animals (eg, mammals, insects, arthropods, etc.) Resources can also be used as O-tRNA and O-RS resources.

The individual components of the O-tRNA / O-RS pair may be from the same organism or from different organisms. Orthogonal tRNA / O-RS pairs can be used in a variety of host organisms (eg, second organisms). In one aspect, the O-tRNA / O-RS pair is from the same organism. Alternatively, the O-tRNA / O-RS pair of O-tRNA and O-RS are from different organisms. [<u style="single">Selector codon</u>〕

The selector codons of the present invention extend the genetic codon framework of the protein biosynthesis mechanism to incorporate unnatural amino acids (eg, unnatural amino acids containing a moiety capable of binding a sugar moiety or an unnatural amino acid containing a sugar moiety). For example, selector codons include, for example, unique 3-base codons, nonsense codons (eg, stop codons such as amber codons or opal codons), unnatural codons, codons of at least 4 bases, rare codons and the like. For example, a large number of selector codons such as one or more, two or more, three or more can be introduced into a desired gene.

The 64 genetic codons encode 20 amino acids and 3 stop codons. Since only one stop codon is required to terminate translation, the other two can be used primarily to encode non-protein-producing amino acids. The amber stop codon UAG has been successfully used in in vitro biosynthesis systems and Xenopus oocytes to integrate unnatural amino acids. Of the three stop codons, UAG is the least frequently used stop codon in E. coli. Some E. coli strains contain a natural suppressor tRNA that recognizes UAG and inserts natural amino acids. In addition, these amber suppressor tRNAs have also been used in conventional protein mutagenesis. In certain embodiments of the invention, other stop codons are used in the invention.

In one embodiment, the method uses a stop codon, the selector codon, for in vivo integration of unnatural amino acids. For example, an O-tRNA that recognizes a stop codon (eg, UAG) is prepared and aminoacylated with the desired unnatural amino acid by O-RS. This O-tRNA is not recognized by the native aminoacyl-tRNA synthetase. A stop codon (eg, TAG) can be introduced at the site of interest within a protein gene using conventional site-specific mutagenesis. For example Sayers, JR, Schmidt, W. Eckstein, F See .5', 3'Exonuclease in phosphorothioate-based oligonucleotide-directed mutagenesis. Nucleic Acids Res, 791-802 (1988). In vivo fusion of O-RS, O-tRNA and mutant genes results in the incorporation of unnatural amino acids in response to UAG codons, resulting in proteins containing unnatural amino acids at specific positions.

In vivo integration of unnatural amino acids can be performed without significantly affecting the host (eg E. coli). For example, in non-eukaryotic cells such as Escherichia coli, the suppression efficiency of UAG codons is O-tRNA (eg, amber suppressor tRNA) and release factor 1 (which binds to UAG codon and initiates the release of growing peptides from the ribosome). Since it depends on RF1) competition, suppression efficiency can be regulated, for example by increasing the expression level of O-tRNAs (eg suppressor tRNAs) or by using RF1-deficient strains.

An unnatural amino acid (eg, an unnatural amino acid containing a moiety capable of binding a sugar moiety (eg, an aldehyde or keto derivatized amino acid) or an unnatural amino acid containing a sugar moiety) can also be encoded by a rare codon. For example, when the arginine concentration is lowered in an in vitro protein synthesis reaction, the rare arginine codon AGG has been found to be effective in inserting Ala by synthetic tRNA acylated with alanine. See, for example, Ma et al., Biochemistry, 32: 7939 (1993). In this case, the synthetic tRNA competes with the native tRNAArg, which is present as a small amount in E. coli. Some organisms do not use the full triplet codon. The codon AGA, which is not assigned by Micrococcus luteus, is used for amino acid insertion into in vitro transcription / translation extracts. See, for example, Kowal and Oliver, Nucl. Acid. Res., 25: 4685 (1997). The components of the invention can be made for in vivo use of these rare codons.

Selector codons also include codons of 4 or more bases (eg, codons of 4, 5, 6 or more bases). Examples of 4-base codons include AGGA, CUAG, UAGA, CCCU and the like. Examples of the 5-base codon include AGGAC, CCCCU, CCCUC, CUAGA, CUACU, UAGGC and the like. For example, in the presence of a mutant O-tRNA (eg, a special frameshift suppressor tRNA) with an anticodon loop (eg, at least 8-10 nt anticodon loop), codons of 4 or more bases are read as a single amino acid. In other embodiments, the anticodon loop can decode, for example, at least 4 base codons, at least 5 base codons, or at least 6 or more base codons. Since 256 types of 4-base codons can be considered, multiple unnatural amino acids can be encoded in the same cell by using codons of 4 bases or more. Anderson et al., Exploring the Limits of Codon and Anticodon Size, Chemistry and Biology, Vol.9,237-244 (2002); and Magliery, Expanding the Genetic Code: Selection of Efficient Suppressors of Four-base Codons and Identification of Shifty Four-base Codons with a Library See Approach in Escherichia coli, J. Mol. Biol. 307: 755-769 (2001).

The method of the present invention uses extended codons based on frameshift suppression. For codons of 4 or more bases, for example, one or many unnatural amino acids can be inserted into the same protein. For example, 4 base codons are used to incorporate unnatural amino acids into proteins using in vitro biosynthesis. For example, Ma et al., Biochemistry, 1993, 32, 7939 (1993); and Hohsaka et al., J. Am. Chem. Soc., 121: See 34 (1999). CGGG and AGGU have been used to simultaneously in vitro integrate NBD derivatives of 2-naphthylalanine and lysine into streptavidin with two chemically acylated frameshift suppressor tRNAs. See, for example, Hohsaka et al., J. Am. Chem. Soc., 121: 12194 (1999). In vivo studies, Moore et al. TRNA with NCUA anticodon^LeuDerivatives tested their ability to suppress UAGN codons (N can be U, A, G or C) and the quadruple UAGA is a tRNA with a UCUA anticodon.^LeuIt was found that it can be decoded with an efficiency of 13 to 26%, but it can hardly be decoded at 0 or -1 frame. See Moore et al., (2000) J. Mol. Biol., 298: 195. In one aspect, extended codons based on rare codons or nonsense codons can be used in the present invention to reduce missense skipping and frameshift suppression at other undesired sites.

An unnatural amino acid (eg, an unnatural amino acid containing a moiety capable of binding a sugar moiety or an unnatural amino acid containing a sugar moiety) can also be incorporated into the desired polypeptide using a translation bypass system. 1 In the translation bypass system, a large sequence is inserted into a gene but not translated into a protein. This sequence contains a structure that acts as a signal for the ribosome to jump over the sequence and resume translation downstream of the insertion.

A trans-translation system is used as a substitute or in combination with the other above methods for incorporating an unnatural amino acid (eg, an unnatural amino acid containing a moiety capable of binding a sugar moiety or an unnatural amino acid containing a sugar moiety) into a polypeptide. be able to. This system utilizes a molecule called tmRNA that exists in E. coli. This RNA molecule is structurally closely related to alanyl tRNA and is aminoacylated by alanyl synthetase. The difference between tmRNA and tRNA is that the anticodon loop is replaced with a special large sequence. This sequence can use the open reading frame encoded in the tmRNA as a template to allow the ribosome to resume translation on the stopped sequence. In the present invention, it is possible to prepare an orthogonal tmRNA that is preferentially aminoacylated with orthogonal synthetase and loaded with an unnatural amino acid. By transcribing genes that use this system, the ribosome ceases at a specific site, unnatural amino acids are introduced at this site, and then translation resumes using the sequence encoded within the orthogonal tmRNA.

In a given system, the selector codon can further include one of the native tribase codons if the endogenous system uses (or rarely uses) the native tribase codon. For example, a system lacking a tRNA that recognizes a natural 3-base codon and / or a system in which the 3-base codon is a rare codon corresponds to this.

Selector codons optionally contain unnatural base pairs. These unnatural base pairs further extend the existing genetic alphabet. With one base pair increase, the number of triplet codons increases from 64 to 125. The properties of the third base pair are stable and selective base pairing, high fidelity and efficient enzyme integration into DNA by polymerase, and efficient persistence after synthesis of unfinished unnatural base pairs. Primer extension can be mentioned. See also, for example, Hirao et al., An unnatural base pair for incorporating amino acid analogues into protein, Nature Biotechnology, 20: 177-182 (2002) for non-natural base pairs applicable to methods and compositions. Other related literature is listed below.

For in vivo use, unnatural nucleosides are membrane permeable and phosphorylated to form the corresponding triphosphates. Moreover, the increased genetic information is stable and is not destroyed by cellular enzymes. Previous reports by Benner et al. Used different hydrogen bond patterns than the canonical Watson click pair, the most notable of which is the iso C: iso G pair. Is. For example, Switzer et al., (1989) J. Am. Chem. Soc., 111: 8322; and Piccirilli et al., (1990) Nature, 343: 33; Kool, (2000) Curr. Opin. Chem. Biol .. 4: 602. reference. These bases are generally mispaired to some extent with natural bases and cannot be enzymatically replicated. Kool et al. Have demonstrated that base pair formation can be induced by replacing hydrogen bonds with hydrophobic packing interactions between bases. Kool, (2000) Curr. Opin. Chem. Biol., 4: 602; and Guckian and See Kool, (1998) Angew.Chem.Int.Ed.Engl., 36,2825. In an attempt to develop unnatural base pairs that meet all of the above requirements, Schultz, Romerberg et al. Systematically synthesized and tested a series of unnatural hydrophobic bases. PICS: PICS self-pairs are more stable than natural base pairs and can be efficiently integrated into DNA by the Klenow fragment (KF) of E. coli DNA polymerase I. See, for example, McMinn et al., (1999) J.Am.Chem.Soc., 121: 11586; and Ogawa et al., (2000) J.Am.Chem.Soc., 122: 3274. 3MN: 3MN autopairs can be synthesized by KF with sufficient efficiency and selectivity for biological function. See, for example, Ogawa et al., (2000) J. Am. Chem. Soc., 122: 8803. However, both bases only function as chain terminators for late replication. A mutant DNA polymerase that can be used to replicate the PICS autopair has recently been developed. In addition, 7AI self-pairs can also be duplicated. See, for example, Tae et al., (2001) J. Am. Chem. Soc., 123: 7439. A novel metallo base pair Dipic: Py has also been developed that forms a stable pair when bound to Cu (II). See Meggers et al., (2000) J. Am. Chem. Soc .. 122: 10714. Since extended codons and unnatural codons are inherently orthogonal to natural codons, the methods of the invention can take advantage of this property to make their orthogonal tRNAs. [<u style="single">Unnatural amino acids</u>〕

The unnatural amino acids used herein are selenocysteine and / or pyrrolidine and the following 20 genetically encoded α-amino acids, namely alanine, arginine, aspartic acid, aspartic acid, cysteine, glutamine, glutamic acid, glycine, It means any amino acid other than histidine, isoleucine, leucine, lysine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine, valine, modified amino acid or amino acid analog. The general structure of the α amino acid is formula I:<chemistry num="1"><img file="JP4752001B2_D0001.tif" /></chemistry>Represented by.

Unnatural amino acids are generally arbitrary structures with formula I, in which the R group is an arbitrary substituent other than that used in the 20 natural amino acids. For the structure of 20 natural amino acids, see, for example, L. Stryer, Biochemistry, 3rd Edition, 1988, Freeman and Company, New York. The unnatural amino acid of the present invention may be a natural compound other than the above 20 kinds of α-amino acids.

In some cases, the unnatural amino acid of the present invention differs from the natural amino acid only in the side chain, so that it forms an amide bond with another amino acid (for example, a natural or unnatural amino acid) like a natural protein. On the other hand, unnatural amino acids have side chain groups different from those of natural amino acids.

In the production of the glycoprotein of the present invention, in formula I, R is a protein containing an unnatural amino acid and a sugar moiety. Of particular importance are unnatural amino acids that contain a moiety that can react with a reactive group that is attached to a sugar moiety to bind. Suitable R groups include, for example, keto, azide, hydroxyl, hydrazine, cyano, halo, aminooxy, alkenyl, alkynyl, carbonyl, ether, thiol, seleno, sulfonyl, boric acid, boric acid, phospho, phosphono, phosphine, heterocycle. , Enon, imine, aldehyde, ester, thioic acid, thioester, hindered ester, hydroxylamine, amine and the like, or any combination thereof. In certain embodiments, the unnatural amino acid has a photocrosslinking group.

In addition to unnatural amino acids containing novel side chains, unnatural amino acids may optionally be, for example, formulas II and III :.<chemistry num="2"><img file="JP4752001B2_D0002.tif" /></chemistry>In the equation, Z is generally OH, NH, including a modified backbone structure as represented by the structure of.₂, SH, NH-R'or S-R', X and Y may be the same or different, generally S or O, R and R'may be the same or different, and generally have the formula I. The unnatural amino acid is selected from the same groups as the R group described above and hydrogen. For example, the unnatural amino acids of the invention optionally contain substitutions at amino or carboxyl groups as represented by formulas II and II. Examples of this type of unnatural amino acid include, but are not limited to, α-hydroxy acids, α-thioic acids, and α-aminothiocarboxylates having side chains or unnatural side chains corresponding to 20 standard natural amino acids. .. In addition, α-carbon substitutions optionally include L, D or α, α-di-substituted amino acids (eg, D-glutamic acid, D-alanine, D-methyl-O-tyrosine, aminobutyric acid, etc.). Other alternative structures include cyclic amino acids (eg proline analogs and 3, 4, 6, 7, 8 and 9-membered ring proline analogs), β and γ amino acids (eg substituted β-alanine and γ-aminobutyric acid). Can be mentioned.

For example, many unnatural amino acids are based on natural amino acids (eg, tyrosine, glutamine, phenylalanine, etc.). Tyrosine analogs include para-substituted tyrosine, ortho-substituted tyrosine, and meta-substituted tyrosine, which are acetyl, benzoyl, amino, hydrazine, hydroxylamine, thiol, carboxy, isopropyl, and methyl groups. C₆-C₂₀Includes straight-chain or branched-chain hydrocarbons, saturated or unsaturated hydrocarbons, O-methyl groups, polyether groups, nitro groups and the like. In addition, multiple substituted aryl rings are also conceivable. Glutamine analogs of the present invention include, but are not limited to, α-hydroxy derivatives, γ-substituted derivatives, cyclic derivatives and amide-substituted glutamine derivatives. Examples of phenylalanine analogs include, but are not limited to, meta-substituted, ortho-substituted, and / or para-substituted phenylalanines, the substituents including hydroxy, methoxy, methyl, allyl, aldehyde, keto and the like. ..

Specific examples of unnatural amino acids are not limited, but p-acetyl-L-phenylalanine, O-methyl-L-tyrosine, L-3- (2-naphthyl) alanine, 3-methylphenylalanine, O-4-allyl-. L-tyrosine, 4-propyl-L-tyrosine, tri-O-acetyl-GlcNAc β-serine, β-O-GlcNAc-L-serine, tri-O-acetyl-GalNAc-α-threonine, α-GalNAc-L- Threonine, L-Dopa, Fluorinated Phenylalanine, Isopropyl-L-Phenylalanine, p-Azide- L-Phenylalanine, p-acyl-L-Phenylalanine, p-benzoyl-L-Phenylalanine, L-phosphoserine, phosphonoserine, phosphonotyrosine, p-iodophenylalanine, p-bromophenylalanine, p-amino-L-phenylalanine, isopropyl- Examples include L-Phenylalanine, which is described below or elsewhere herein. The structures of various unnatural amino acids are shown, for example, in FIGS. 17, 18, 19, 26 and 29 of WO 2002/085923.

Examples of unnatural amino acids suitable for use in the method of the present invention include those in which the sugar moiety is bound to the amino acid side chain. In one embodiment, the unnatural amino acid having a sugar moiety includes Man, GalNAc, Glc, Fuc, or serine or threonine having a Gal moiety. Examples of unnatural amino acids containing sugar moieties are not limited, but are, for example, tri-O-acetyl-GlcNAc β-serine, β-O-GlcNAc-L-serine, tri-O-acetyl-GalNAc-α-threonine, α- GalNAc-L-threonine, O-Man-L-serine, tetra-acetyl-O-Man-L-serine, O-GalNAc-L-serine, tri-acetyl-O-GalNAc-L-serine, Glc-L- Serine, tetraacetyl-Glc-L-serine, fuc-L-serine, tri-acetyl-fuc-L-serine, O-Gal-L-serine, tetra-acetyl-O-Gal-L-serine, β-O -GlcNAc-L-serine, tri-acetyl-β-GlcNAc-L-serine, O-Man-L-serine, tetra-acetyl-O-Man-L-serine, O-GalNAc-L-threonine, tri-acetyl -O-GalNAc-L-serine, Glc-L-serine, tetraacetyl-Glc-L-serine, fuc-L-serine, tri-acetyl-fuc-L-serine, O-Gal-L-serine, tetra- Examples include acetyl-O-Gal-serine. The present invention includes unprotected and acetylated forms of the unnatural amino acids. WO2003 / 031464A2, Invention title "Remodeling and Glycoconjugation of Peptides"; and US Pat. No. 6,331,418, Invention title "Sugar compositions, Saccharide Compositions, [<u style="single">Chemical synthesis of unnatural amino acids</u>〕

Many of the above unnatural amino acids are commercially available, for example, from Sigma (USA) and Aldrich (Milwaukee, WI, USA). Those not on the market may be synthesized by using the description of the following examples or a standard method known to those skilled in the art. For organic synthesis techniques, for example, Fessendon and Fessendon, Organic Chemistry (1982, 2nd Edition, Willard Grant Press, Boston Mass.); March, Advanced Organic Chemistry (3rd Edition, 1985, Wiley and Sons, New York); and Carey. And Sundberg, Advanced Organic Chemistry (3rd Edition, Parts A and B, 1990, Plenum Press, New York). See WO 2002/085923 for other synthesis of unnatural amino acids.

For example, meta-substituted phenylalanine is synthesized in a procedure as summarized in WO 2002/085923 (see, eg, Figure 14 of the publication). Generally, NBS (N-bromosuccinimide) is added to a meta-substituted methylbenzene compound to obtain meta-substituted benzyl bromide, which is then reacted with a malonic acid compound to obtain meta-substituted phenylalanine. Typical substituents used at the meta position include, but are not limited to, ketones, methoxy groups, alkyls, acetyls and the like. For example, 3-acetylphenylalanine is produced by reacting NBS with a solution of 3-methylacetophenone. See the examples below for details. Similar synthesis It is used to produce 3-methoxyphenylalanine. In that case, the R group at the meta position of benzyl bromide is -OCH.₃Is. See, for example, Matsoukas et al., J. Med. Chem., 1995, 38, 4600-4669.

In certain embodiments, the design of the unnatural amino acid is modified by known information about the active site of the synthetase (eg, the orthogonal tRNA synthetase used to aminoacylate the orthogonal tRNA). For example, a nitrogen-substituted derivative of the amide (1), a methyl group at the γ position (2), and NC.^γ-Three glutamine analogs of the cyclic derivative (3) are provided. For example, the mutation from the active site Phe233 to a hydrophobic small amino acid is C of Gln.^γComplements a series of side chain mutations of residues within 10 Å of the glutamine side chain, based on the X-ray crystal structure of Escherichia coli GlnRS whose major binding site residues are homologous to yeast GlnRS, as they can be complemented by increased conformational bulk. The analog was designed to do so.

For example, N-naphthaloyl-L-glutamic acid 1,5-anhydride (Compound No. 4 in FIG. 23 of WO 2002/085923) is optionally used to synthesize glutamine analogs with substituents on the nitrogen of the amide. For example, King, FE &amp; Kidd, DAAA New Synthesis of Glutamine and of γ-Dipeptides of Glutamic Acid from Phthylated Intermediates. J.Chem.Soc., 3315-3319 (1949); Friedman, OM &amp; Chatterrji, R. Synthesis of Derivatives of Glutamine as Model Substrates for Anti-Tumor Agents.J.Am.Chem.Soc.81,3750-3752 (1959); Craig, JC et al., Absolute Configuration of the Enantiomers of 7-Chloro-4- [[4- (diethylamino) -1-methylbutyl] amino] quinoline (Chloroquine). J.Org.Chem.53,1167-1170 (1988); and Azoulay, M., Vilmont, M. &amp; Frappier, F. Glutamine analogues as See Potential Antimalarials, .Eur.J.Med.Chem.26,201-5 (1991). The anhydride is generally made from glutamic acid, where the amine is first protected as a phthalimide and then refluxed in acetic acid. The anhydride is then ring-opened with a number of amines to give the amide various substituents. Deprotecting the phthaloyl group with hydrazine gives free amino acids as shown in Figure 23 of WO 2002/085923.

Substitution of the γ position is generally carried out by alkylation of glutamic acid. See, for example, Koskinen, AMP &amp; Rapoport, H. Synthesis of 4-Substituted Prolines as Conformationally Constrained Amino Acid Analogues. J. Org. Chem. 54, 1859-1866 (1989). For example, a protected amino acid as shown in Compound No. 5 of FIG. 24 of WO 2002/085923 may first alkylate the amino moiety with 9-bromo-9-phenylfluorene (PhflBr) (eg Christie, BD &amp; Rapoport, H. Synthesis of Optically). Pure Pipecolates from L-Asparagine.Application to the total Synthesis of (+)-Apovincamine through Amino Acid Decarbonylation and Iminium Ion Cyclization.J.Org.Chem.1989,1859-1866 (1986)) followed by esterification of the acid moiety with O-tert-butyl-N, N'-diisopropylisourea. .. KN (Si (CH)₃)₃)₂Is added to site-selectively deprotonate the α-position of the methyl ester to form an enolate, and then alkylated with various alkyl iodides as the case may be. .. Hydrolysis of the t-butyl ester and Phfl group gave the desired γ-methylglutamine analog (Compound No. 2 in FIG. 24 of WO 2002/085923).

NC as shown by compound number 3 in Figure 25 of WO 2002/085923^γ-Cyclic analogs are optionally prepared from Boc-Asp-Ot-Bu in four steps as previously described. For example, Barton et al., Synthesis of Novel α-Amino-Acids and Derivatives Using Radical Chemistry: Synthesis of L-and D-α-Amino-Adipic Acids, L-α-aminopimelic Acid and Appropriate Unsaturated Derivatives.Tetradedron Lett.43,4297- 4308 (1987) and Subasinghe et al., Quisqualic acid analogues: synthesis of beta-heterocyclic 2-aminopropanoic acid derivatives and their activity at a novel quisqualate-sensitized site.J.Med.Chem.35 See 4602-7 (1992). The Michael adduct is obtained by adding compound 7 as shown in FIG. 25 after the formation of the anion of Nt-Boc-pyrrolidinone, pyrrolidinone, or oxazolidone. Free amino acids are then obtained by deprotection with TFA.

In addition to the above unnatural amino acids, a tyrosine analog library was also designed. Based on the crystal structure of B. stearothermophilus TyrRS, whose active site is highly homologous to M. jannashii synthetase, residues within 10 Å of the aromatic side chain of tyrosine were mutated (Y32, G34, L65, Q155, D158, A167, Y32 and D158). A tyrosine analog library as shown in Figure 26 of WO 2002/085923 was designed to complement the series of substitutions of these active site amino acids. These include various phenyl substitution patterns that impart various hydrophobicity and hydrogen bonding properties. Tyrosine analogs are optionally prepared using the general strategy indicated by WO 2002/085923 (see, eg, Figure 27 of the publication). For example, sodium ethoxide is optionally used to produce the enolate of diethyl acetamide malonate. The desired tyrosine analog can then be produced by adding the appropriate benzyl bromide and then hydrolyzing. [<u style="single">Cellular uptake of unnatural amino acids</u>〕

Incorporation of unnatural amino acids is one of the problems commonly considered when designing and selecting unnatural amino acids, for example for incorporation into proteins. For example, α-amino acids have a high charge density, so these compounds are unlikely to become cell permeable. Natural amino acids are taken up by bacteria via a series of protein transport systems that exhibit varying degrees of amino acid specificity. Therefore, the present invention provides rapid screening to determine which unnatural amino acids are taken up when they are taken up by cells.

For example, various unnatural amino acids are optionally screened for cytotoxicity in minimal media. Toxicity is generally (1) NOAEL with no significant change in doubling time, (2) Low toxicity with less than about 10% increase in doubling time, and (3) Approximately 10% to about 50% increase in doubling time. It is classified into 5 categories: addictive, (4) highly toxic with an increase in doubling time of about 50% to about 100%, and (5) extremely toxic with an increase in doubling time of more than about 100%. See, for example, Liu, DR &amp; Schultz, PG Progress toward the evolution of an organism with an expanded genetic code. PNAS, USA 96,4780-4785 (1999). The toxicity of amino acids, which are generally classified as highly toxic or extremely toxic, is measured as a function of their concentration, and IC₅₀Get the value. In general, amino acids that are closely related analogs of natural amino acids or that exhibit reactive functional groups are the most toxic. This tendency is that the toxic mechanism of these unnatural amino acids pros to natural amino acids. It suggests the possibility of protein integration or inhibition of the essential enzyme to be sed.

In order to identify possible uptake pathways for toxic amino acids, IC may be in the medium supplemented with, for example, excess structurally similar natural amino acids.₅₀Repeat the level toxicity assay. For toxic amino acids, the presence of excess natural amino acids generally supplements the ability of cells to proliferate in the presence of toxins, because natural amino acids effectively eliminate competition with toxins for cell uptake or binding to essential enzymes. is expected. In these cases, toxic amino acids are optionally assigned to possible uptake pathways and are referred to as "lethal alleles" that need to be complemented for cell survival. These lethal alleles are extremely useful in assaying the ability of cells to take up non-toxic, unnatural amino acids. Complementarity of toxic alleles is determined by recovery of cell proliferation, suggesting that non-toxic amino acids are taken up by cells by the same uptake pathway, presumably assigned to lethal alleles. No conclusions have been reached regarding the lack of complementarity. For example, see Examples below for trials and conclusions.

The results obtained (eg, as described in Examples below) demonstrate that complementarity of lethal unnatural amino acid alleles is an efficient method for quantitatively testing amino acid uptake. This method is generally less laborious than radiolabeling a large number of compounds, and is therefore a more advantageous method for analyzing the unnatural amino acid. This general strategy is optionally used to rapidly assess cellular uptake of a wide range of molecules such as nucleobase analogs, sugar analogs or peptide analogs. For example, this strategy is optionally used to assess the cellular uptake of the unnatural amino acids described herein.

The present invention also provides a method independent of the total amino acid uptake pathway as a general method for delivering unnatural amino acids. This general method relies on uptake by peptide permease, which transports dipeptides and tripeptides through the cell membrane. Peptide permease is not sahod side chain specific and its KD value for its substrate is equivalent to the KD amino acid of amino acid permease, for example from about 0.1 mM to about 10 mM. For example, Nickitenko et al., A structure of DppA, a periplasmic depeptide transport / chemosensory receptor.Biochemistry 34,16585-16595 (1995) and Dunten, P., Mowbray, SL Crystal structure of the dipeptide binding protein from Escherichia coli involved in active transport and chemotaxis.Protein Science See 4,2327-34 (1995). The unnatural amino acid is then taken up as a conjugate of a natural amino acid such as lysine and released into the cytoplasm after hydrolysis of the dipeptide by one of the endogenous E. coli peptidases. To test this approach, several Unn-Lys and Lys-Unn dipeptides were synthesized by solid phase synthesis and growth of lysine biosynthesis-deficient E. coli strains in the presence and absence of these dipeptides in lysine minimal medium. Tested. The only source of lysine available to these cells is a dipeptide containing unnatural amino acids. The uptake of phosphonoserine, phosphonotyrosine, pentafluorophenylalanine and caged serin was analyzed in this way. Growth was observed at dipeptide concentrations of 10 mM or higher in all four cases. Although uptake is readily analyzed by the methods described herein, another method of designing unnatural amino acids available for the cellular uptake pathway is to provide a biosynthetic pathway that produces amino acids in vivo. [<u style="single">Biosynthesis of unnatural amino acids</u>〕

Many biosynthetic pathways are naturally present in cells to produce amino acids and other compounds. It seems that a method for biosynthesizing a specific unnatural amino acid does not exist in nature (for example, in a cell), and the present invention provides such a method. For example, a biosynthetic pathway for unnatural amino acids is formed in E. coli, optionally by adding a novel enzyme or modifying an existing E. coli pathway. With Depending on the case, the new enzyme may be a natural enzyme or an artificially produced enzyme. For example, the biosynthesis of p-aminophenylalanine (as described in WO 2002/085923) depends on the addition of a combination of known enzymes from other organisms. The genes for these enzymes can be introduced into cells by transforming cells (eg, E. coli cells) with a plasmid containing this gene. When a gene is expressed in a cell, it provides an enzymatic pathway for synthesizing the desired compound. Examples of enzyme species added in some cases are described in the following examples. The lyase sequence is registered in Genbank, for example. In some cases, artificially prepared enzymes are also added to the cells. In this way, the cellular mechanisms and resources are manipulated to produce unnatural amino acids.

Various methods are available for producing new enzymes used in biosynthetic pathways or for evolving existing pathways. For example, in some cases, new enzymes and pathways are developed using recursive recombination, such as that developed by Maxygen, Inc. (see World Web www.maxygen.com). For example, Stemmer 1994, Rapid evolution of a protein in vitro by DNA shuffling, Nature Vol.370 No.4: Pg.389-391; and Stemmer, 1994, DNA shuffling by random fragmentation and reassembly: In vitro recombination for molecular. evolution, Proc.Natl.Acad.Sci.USA.Vol.91:Pg.10747-10751. Similarly, the Metabolic Path of DesignPath®, which was developed by Genencor (see world web genencor.com) in some cases. Used for recombination, eg, recombining pathways to produce unnatural amino acids in E. coli. This technique uses, for example, a combination of functional genomics and novel genes identified by molecular evolution and design to route existing pathways in host organisms. Reconstruct. Diversa Corporation (see World Web diversa.com) also provides techniques for rapid screening of genetic libraries and pathways, for example to create new pathways.

In general, the biosynthetic methods of the invention (eg, the pathway for producing p-aminophenylalanine (pAF) from chorismic acid) do not affect the concentration of other amino acids produced in the cell. For example, the pathway used to produce pAF from chorismic acid produces pAF intracellularly, but the concentration of other aromatic amino acids commonly produced from chorismate cells remains substantially unchanged. Generally, the unnatural amino acids produced by the recombinant biosynthesis pathway of the present invention are produced at a concentration sufficient for efficient protein biosynthesis (for example, the amount of natural cells), but the concentration of other amino acids is changed or the cell resources are changed. It will not be exhausted. The typical concentration thus produced in vivo is from about 10 mM to about 0.05 mM. If you have transformed the bacterium with a plasmid containing the gene used to produce the desired enzyme in a particular pathway to produce the 21st amino acid (eg pAF, dopa, O-methyl-L-tyrosine, etc.) By using in vivo selection, the production of unnatural amino acids is further optimized for both ribosomal protein synthesis and cell proliferation. [<u style="single">Nucleic acid and polypeptide sequence variants</u>〕

As described above and below, the present invention provides compositions and methods comprising nucleic acid polynucleotide sequences (eg, O-tRNA and O-RS) and polypeptide amino acid sequences (eg, O-RS), eg, said sequences. To do. Examples of such sequences (eg, O-tRNA and O-RS) are disclosed herein (see Table 2, eg, SEQ ID NOs: 1-10). However, as will be apparent to those skilled in the art, the present invention is not limited to the sequences disclosed herein (eg, Examples). As will be appreciated by those skilled in the art, the present invention also provides a number of unrelated sequences (eg, encoding O-tRNA or O-RS) having the functions described herein.

The present invention relates to polypeptides (eg, O-RS) and polynucleotides (eg, O-tRNA). , O-RS or a polynucleotide encoding an portion thereof, an oligonucleotide used for isolating an aminoacyl-tRNA synthetase clone, and the like. Examples of the polynucleotide of the present invention include polypeptides or proteins incorporating the unnatural amino acids of the present invention. The polypeptide of the present invention further comprises an artificial polypeptide, for example, (a) a polypeptide containing the amino acid sequence shown in any one of SEQ ID NOs: 4 to 6; (b) any one of SEQ ID NOs: 8 to 10. A polypeptide containing an amino acid sequence encoded by the polypeptide sequence shown; a polypeptide specifically immunoreactive to an antibody specific for the polypeptide of (c) (a) or (b); and (d). ) An amino acid sequence containing a conservative variant of (a), (b), or (c). Antibodies or antisera specifically immunoreactive to the artificial polypeptides of the invention are also provided. In one aspect, the composition comprises the polypeptides of the invention and excipients such as buffers, water, pharmaceutically acceptable excipients and the like.

Examples of the polynucleotide of the present invention include those encoding the relevant protein or polypeptide of the present invention containing one or more selector codons. The polynucleotide of the present invention also includes a polynucleotide containing any one of SEQ ID NOs: 8, 8, or 10 or a conserved variant thereof. The polynucleotides of the present invention include polynucleotides encoding the amino acid sequences of SEQ ID NOs: 1-6. The polynucleotides of the invention also include polynucleotides encoding the polypeptides of the invention. Similarly, an artificial nucleic acid (other than a natural polynucleotide) that hybridizes with the polynucleotide under high stringent conditions over substantially the entire length of the nucleic acid is also the polynucleotide of the present invention. Artificial polynucleotides are artificial polynucleotides that do not exist in nature.

In certain embodiments, the polynucleotide of the invention is introduced into a vector (eg, plasmid, cosmid, phage, virus, etc.). In one embodiment, the vector is an expression vector. In another aspect, the expression vector comprises a promoter operably linked to one or more of the polynucleotides of the invention. In another embodiment, a vector incorporating the polynucleotide of the invention is introduced into a cell.

As will be apparent to those skilled in the art, a large number of variants of the disclosed sequences are also included in the present invention. For example, a conserved variant of a disclosed sequence that is functionally identical is also included in the present invention. Variants of nucleic acid polynucleotide sequences that hybridize to at least one disclosed sequence are also included in the invention. For example, sequences that are considered to be unique subsequences of the sequences disclosed herein by standard sequence comparison methods are also included in the present invention. [<u style="single">Conserved mutation</u>〕

Due to the degeneracy of the genetic code, "silent substitutions" (ie, substitutions of nucleic acid sequences that do not change the encoded polypeptide) are implicit features of the entire nucleic acid sequence encoding the amino acid. Similarly, a "conserved amino acid substitution" is one in which one or several amino acids in an amino acid sequence are replaced with another amino acid having highly similar properties, and such substitutions are also highly similar to the disclosed constructs. Is self-evident. Such conservative variations of each disclosed sequence are characteristic of the present invention.

A "conservative mutation" in a particular nucleic acid sequence means a nucleic acid that encodes the same or essentially the same amino acid sequence, and if the nucleic acid does not encode an amino acid sequence, it means essentially the same sequence. As is obvious to those skilled in the art, individual amino acids in the encoded sequence or individual amino acids with a low percentage (generally less than 5%, more generally less than 4%, 2% or less than 1%) are substituted, added or deleted. If one amino acid is deleted, one amino acid is added, or one amino acid is replaced by one chemically similar amino acid as a result of substitution, deletion or addition, these The modification is a "conservative modification mutation". Therefore, the "conservative variation" of the polypeptide sequence of the present invention is a conservative unnatural amino acid containing an amino acid that binds a sugar moiety and / or an amino acid of the polypeptide sequence that is an unnatural amino acid containing a sugar moiety of the same conservative substituent. Low percentage, generally 5% Less than, more generally less than 2% or less than 1% may be replaced. Finally, the addition of sequences that do not alter the encoded activity of the nucleic acid molecule, such as the addition of non-functional sequences, is also a conservative mutation of the basic nucleic acid.

Conservative substitution tables showing functionally similar amino acids are well known in the art. An example of a group containing natural amino acids that mutually contain "conservative substitutions" is shown below.<tables num="1"><img file="JP4752001B2_D0003.tif" /></tables> 〔<u style="single">Nucleic acid hybridization</u>〕

Comparative hybridization can be used to identify nucleic acids of the invention (eg, SEQ ID NOs: 7, 8, 9, or 10) that contain conservative variants of the nucleic acids of the invention, and this comparative hybridization method presents the nucleic acids of the invention. Is a suitable method for identifying. Furthermore, a target nucleic acid that hybridizes to the nucleic acid shown in SEQ ID NO: 7, 8, 9, or 10 under high, ultra-high and ultra-high stringency conditions is also a feature of the present invention. Examples of such nucleic acids include those with one or several silent or conserved nucleic acid substitutions compared to a given nucleic acid sequence.

At least about 5 of the signal-to-noise ratio observed when hybridizing to a probe at a rate of at least 1/2 compared to a perfectly matched complementary target, ie hybridizing with any of the unmatched target nucleic acids. Test nucleic acids have a signal-to-noise ratio of at least 1/2 compared to probe-target hybridization under conditions where a fully matched probe at a factor of 10 is bound to a perfectly matched complementary target. It is said to specifically hybridize to probe nucleic acids.

Nucleic acids generally "hybridize" when associated in solution. Nucleic acids hybridize by a variety of well-characterized physicochemical forces such as hydrogen bonding, solvent exclusion, and base stacking. For detailed guidance on nucleic acid hybridization, see Tijssen (1993) Laboratory Techniques in Biochemistry and Molecular Biology--Hybridization with Nucleic Acid Probes part I chapter 2, Overview of principles of hybridization and The strategy of nucleic acid probes assays, (Elsevier, New York) and Ausubel, supra. Hames and Higgins (1995) Gene Probes 1 IRL Press at Oxford University Press, Oxford, England, (Hames and Higgins 1) and Hames and Higgins (1995) Gene Probes 2 IRL Press at Oxford University Press, Oxford, England (Hames and Higgins 2) synthesize and label oligonucleotide-containing DNA and RNA. , Detection and quantification are described in detail.

One example of stringent hybridization conditions for hybridization of complementary nucleic acids with more than 100 complementary residues on the filter in Southern or Northern blots is 50% formalin plus 1 mg of heparin at 42 ° C. Overnight hybridization. One example of stringent wash conditions is Sambrook, supra, for a description of SSC buffer at 65 ° C, 0.2 x SSC for 15 minutes. In many cases Remove background probe signals with low stringency lavage prior to high stringency lavage. One example of low stringency washes is at 40 ° C, 2 x SSC for 15 minutes. In general, specific hybridization is considered to be detected if the signal-to-noise ratio is 5 times (or more) than that observed for unrelated probes in a specific hybridization assay.

For nucleic acid hybridization experiments such as Southern and Northern hybridization, "stringent hybridization wash conditions" are sequence-dependent and will vary with various environmental parameters. Detailed guidance on nucleic acid hybridization can be found in Tijssen (1993), supra and Hames and Higgins 1 and 2. Stringent hybridization and washing conditions can be readily determined empirically for any test nucleic acid. For example, to determine high stringent hybridization and washing conditions, until the overall selection criteria are met (eg, temperature rise in hybridization or washing, salt concentration decrease, surfactant concentration increase and / or organic solvent concentration such as formalin). Gradually increase hybridization and wash conditions (by increasing). Hybridization and wash conditions are incremented until the probe binds to a perfectly matched complementary target, eg, at least about 5 times the signal-to-noise ratio observed in hybridization of the unmatched target to the probe.

The "super stringent" condition is chosen to be equal to the thermal melting point (Tm) of a particular probe. Tm is the temperature at which 50% of the test sequence hybridizes to a perfectly matched probe (under specified ionic strength and pH). For the purposes of the present invention, "high stringent" hybridization and washing conditions are generally selected such that the specified ionic strength and pH are approximately 5 ° C lower than the Tm of the particular sequence.

"Ultra-high stringency" hybridization and washing conditions are the signal pairs observed in hybridization of a perfectly matched complementary target nucleic acid with any of the target nucleic acids whose signal-to-noise ratio does not match when binding the probe. It is a condition that increases the stringency of hybridization and washing conditions until it becomes at least 10 times the noise ratio. A target nucleic acid is said to bind to a probe under ultra-high stringency conditions when hybridized to the probe under the above conditions at least 1/2 of the signal-to-noise ratio of a perfectly matched complementary target.

Similarly, higher levels of stringency can be determined by increasing the hybridization and / or wash conditions of the hybridization assay in question. For example, the signal-to-noise ratio when binding a probe to a perfectly matched complementary target nucleic acid is at least 10- to 20-fold the signal-to-noise ratio observed in hybridization of any unmatched target nucleic acid. Increase the stringency of hybridization and washing conditions up to 50-fold, 100-fold, or 500-fold or higher. A target nucleic acid is said to bind to a probe under ultra-high stringency conditions when hybridized to the probe under the above conditions at least 1/2 of the signal-to-noise ratio of a perfectly matched complementary target.

Nucleic acids that do not hybridize to each other under stringent conditions are substantially identical if the polypeptides encoded by these nucleic acids are substantially identical. This is the case, for example, when making a copy of a nucleic acid using the maximum codon degeneracy allowed for the genetic code. [<u style="single">Unique sub-array</u>〕

On one side, the invention is selected from the sequences of O-tRNA and O-RS disclosed herein. A nucleic acid containing a unique subsequence is provided. Unique subsequences are unique compared to the nucleic acids corresponding to any known O-tRNA and O-RS nucleic acid sequences. For example, alignment can be performed using BLAST set as the default parameter. The arbitrary unique subsequence is useful, for example, as a probe for identifying the nucleic acid of the present invention.

Similarly, the present invention comprises a polypeptide comprising a unique subsequence in a polypeptide selected from the sequences of O-RS disclosed herein. In this case, the unique subsequence is unique compared to the polypeptide corresponding to any of the known polypeptide sequences.

The invention also provides a target nucleic acid that hybridizes under stringent conditions with a unique coding oligonucleotide that encodes a unique subsequence in a polypeptide selected from the sequence of O-RS, in which case the unique subsequence is a control poly. It is unique compared to the polypeptide corresponding to any of the peptides (eg, the parent sequence mutated to obtain the synthetase of the invention). The unique sequence is determined as described above. [<u style="single">Sequence comparison, degree of coincidence and degree of homology</u>〕

The term "match" or "match" percentage for two or more nucleic acid or polypeptide sequences contrasts and aligns the two or more sequences or subsequences to the maximum extent, and the sequence comparison algorithm described below. Means that the percentages of amino acid residues or nucleotides that are identical or identical to each other when used or visually measured by one of (or other algorithms available to those of skill in the art) are specific. To do.

The term "substantially matched" with respect to two or more nucleic acids or polypeptides (eg, DNA encoding an O-tRNA or O-RS or O-RS amino acid sequence) maximizes two or more sequences or subsequences. Contrast and alignment to correspond to, and nucleotide or amino acid residue concordance of at least about 60%, preferably 80%, most preferably 90-95% when used or visually measured using a sequence comparison algorithm. It means that there is. Such "substantially matching" sequences are generally considered to be "homologous" even if the actual origin is not stated. There is preferably a "substantial match" over the region of the sequence having a length of at least about 50 residues, more preferably the region having a sequence length of at least about 100 residues, over at least about 150 residues or the overall length of the two sequences being compared. Most preferably, the sequences are substantially identical.

Proteins and / or protein sequences are "homologous" when naturally or artificially derived from a common ancestral protein or protein sequence. Similarly, nucleic acids and / or nucleic acid sequences are homologous when naturally or artificially derived from a common ancestral nucleic acid or nucleic acid sequence. For example, any native nucleic acid can be modified by available arbitrary mutagenesis methods to add one or more selector codons. When expressed, this mutagenic nucleic acid encodes a polypeptide containing one or more unnatural amino acids. The mutation method can, of course, further mutate one or more standard codons and also mutate one or more standard amino acids in the resulting mutant protein. Homology is generally estimated from the sequence similarity between two or more nucleic acids or proteins (or their sequences). The exact percentage of sequence similarity that is useful for determining homology varies depending on the nucleic acid and protein, but usually a low sequence similarity of about 25% is used to determine homology. Higher levels of sequence similarity, such as 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95% or 99% or higher, can also be used to determine homology. Methods for determining sequence similarity percentages (eg, BLASTP and BLASTIN using default parameters) are described herein and are generally available.

For sequence comparison and homology determination, one sequence is generally used as a reference sequence and the test sequence is compared with this. Compare. When using the sequence comparison algorithm, enter the test sequence and reference sequence into the computer, specify subarray coordinates as needed, and specify the sequence algorithm program parameters. The sequence comparison algorithm then calculates the sequence match percentage of the test sequence with respect to the reference sequence based on the specified program parameters.

The optimal sequence alignment for comparison is, for example, the local homology algorithm of Smith &amp; waterman, Adv.Appl.Math.2: 482 (1981), the homology of Needleman &amp; Wunsch, J.Mol.Biol.48: 443 (1970). Gender Alignment Algorithms, Pearson &amp; Lipman, Proc.Nat'l Acad.Sci.USA 85: 2444 (1988) Similarity Search Methods, Computer Software for These Algorithms (Wisconsin Genetics Software Package, Genetics Computer Group, 575 Science Dr. , Madison, WI GAP, BESTFIT, FASTA and TFASTA), or visually (generally Ausubel et al., See below).

An example of an algorithm suitable for determining sequence matching and sequence similarity percentage is the BLAST algorithm described in Altschul et al., J. Mol. Biol. 215: 403-410 (1990). Software for performing BLAST analysis is National Center for Biotechnology It is publicly available from Information (www.ncbi.nlm.nih.gov/). This algorithm is first high by identifying short words of length W in the query sequence that match or satisfy a given positive threshold score T when aligned with words of the same length in the database array. Identify score sequence pairs (HSPs). T is called the adjacent word score threshold (Altschul et al., Supra). Start the search with these initial adjacent word hits as a seed and search for longer HSPs containing these words. The word hits are then extended in both directions along each sequence as long as the cumulative alignment score can be increased. For nucleotide sequences, the parameters M (reward score for a pair of match residues, always&gt; 0) and N (penalty score for mismatch residues, always &lt;0) are used to calculate the cumulative score. For amino acid sequences, the scoring matrix is used to calculate the cumulative score. When the cumulative alignment score drops by an amount X from its maximum reached, or the cumulative score drops below zero due to the accumulation of one or more negative score residue alignments, or reaches the end of either sequence, in each direction. Stop extending word hits. The BLAST algorithm parameters W, T and X determine the sensitivity and speed of the alignment. The BLASTIN program (for nucleotide sequences) uses word length (W) 11, expected value (E) 10, cutoff 100, M = 5, N = 4, and double-strand comparison as defaults. For amino acid sequences, the BLASTP program uses word length (W) 3, expected value (E) 10, and BLOSUM62 scoring matrix as defaults (Henikoff &amp; Henikoff (1989) Proc.Natl.Acad.Sci.USA 89: See 10915).

In addition to calculating the sequence match percentage, the BLAST algorithm also performs a statistical analysis of the similarity between the two sequences (see, eg, Karlin &amp; Altschul, Proc. Nat'l. Acad.Sci.USA 90: 5873-5787 (1993)). ). One measure of similarity provided by the BLAST algorithm is the minimum total probability (P (N)), which indicates the probability of an accidental match between two nucleotide or amino acid sequences. For example, a nucleic acid is considered to be similar to a reference nucleic acid if the minimum total probability of comparing the test nucleic acid to the reference nucleic acid is less than about 0.1, more preferably less than about 0.01, and most preferably less than about 0.001. [<u style="single">Mutagenesis and other molecular biology techniques</u>〕

The polynucleotides and polypeptides of the invention and the polynucleotides and polypeptides used in the present invention can be manipulated using molecular biology techniques. General textbooks describing molecular biology techniques include Berger and Kimmel, supra; Sambrook, supra; and Ausubel, supra. These textbooks describe mutagenesis, the use of vectors, promoters and many other related matters, such as selector codons for producing glycoproteins of the invention, including orthogonal tRNAs, orthogonal synthetases and their pairs. It also describes the preparation of genes containing it.

For example, to mutate a tRNA molecule, to make a library of tRNAs, to make a library of synthetases, for example, to make an unnatural amino acid (eg, an unnatural amino acid or sugar moiety containing a moiety capable of binding a sugar moiety). Various mutagenesis are used in the present invention to insert a selector codon encoding (an unnatural amino acid containing) into the protein or polypeptide of interest. Examples of these are, but are not limited to, site-specific, random point mutagenesis, homologous recombination, DNA shuffling or other recursive mutagenesis, chimeric construction, mutagenesis using uracil-containing templates, oligos. Included are nucleotide-specific mutagenesis, phosphorothioate-modified DNA mutagenesis, mutagenesis using gap-duplex DNA, or any combination thereof. Other available methods include point mismatch repair, mutagenesis using repair-deficient host strains, restriction-selection and restriction-purification, deletion mutation induction, mutation induction by complete gene synthesis, double-strand break repair. And so on. Mutagenesis using, for example, chimeric constructs is also included in the present invention. In one embodiment, mutagenesis can be carried out with known information (eg, sequence, sequence comparison, physical properties, crystal structure, etc.) of the natural molecule or the modified or mutated natural molecule.

Host cells are genetically modified (eg, transformed, transduced or transfected) with the polynucleotides of the invention or constructs incorporating the polynucleotides of the invention (eg, vectors of the invention such as cloning vectors or expression vectors). For example, a coding region for a protein to be derivatized with an orthogonal tRNA, an orthogonal tRNA synthetase, and, for example, an unnatural amino acid containing a moiety capable of binding a sugar moiety (eg, an aldehyde or keto derivatized amino acid) or an unnatural amino acid containing a sugar moiety. Is functionally linked to a functional gene expression control element in the desired host cell. A typical vector contains a transcription and translation terminator, a transcription and translation initiation sequence, and a promoter useful for regulating the expression of a particular target nucleic acid. The vector optionally contains at least one independent terminator sequence, a sequence that allows replication of the cassette in eukaryotes or prokaryotes or both (eg, shuttle vectors), and selectable markers for both prokaryotes and eukaryotes. Includes a comprehensive expression cassette. Vectors are suitable for replication and / or integration in prokaryotes, eukaryotes, or preferably both. Giliman &amp; Smith, Gene See 8:81 (1979); Roberts et al., Nature, 328: 731 (1987); Schneider, B. et al., Protein Expr. Purif. 6435:10 (1995); Ausubel, Sambrook, Berger (all above). The vector can be, for example, in the form of a plasmid, bacterium, virus, naked polynucleotide or polynucleotide conjugate. Vectors are small by electroporation (From et al., Proc. Natl. Acad. Sci. USA 82,5824 (1985)), infection by viral vectors, nucleic acids embedded in a matrix of small beads or particles or attached to the surface. Introduce into cells and / or microorganisms by fast injection in particle form (Klein et al., Nature 327, 70-73 (1987) and / or standard method of equivalents.

A catalog of bacteria and bacteriophages useful for cloning is available, for example, from the ATCC, such as The ATCC Catalog of Bacteria and Bacteriophage (1992) Gherna et al. (Edited) published by the ATCC. The basis for other aspects of other sequencing, cloning and molecular biology This procedure and basic theoretical matters are also included in Sambrook (above), Ausubel (above) and Watson et al. (1992) Recombinant DNA Second Edition. Listed in Scientific American Books, NY. In addition, Midland Certified Reagent Company (Midland, TX mcrc.com), The Great American Gene Company (Ramona, CA, see World Web genco.com), ExpressGen Inc. (Chicago, IL, see World Web expressgen.com), Operon Technologies Almost arbitrary nucleic acids (and almost arbitrary labeled nucleic acids, standard or non-standard) can be made-to-order or standard-ordered from Inc. (Alameda, CA) and many other distributors .

Recombinant host cells can be cultured in conventional nutrient media appropriately modified to suit operations such as screening steps, promoter activation or transformation cell selection. These cells can optionally be cultured in transgenic organisms. For example, for other useful literature on cell isolation and culture (eg for late nucleic acid isolation), see Freshney (1994) Culture of Animal Cells, a Manual of Basic Technique, 3rd Edition, Wiley-Liss, New York and its references; Payne et al. (1992) Plant Cell and Tissue Culture in Liquid Systems John Wiley &amp; Sons, Inc. New York, NY; Gamborg and Phillips (eds) (1995) Plant Cell, Tissue and Organ Culture; Fundamental Methods Springer Lab Manual, Springer-Verlag (Berlin Heidelberg New York) and Atlas and Parks (eds) The Handbook of Microbiological Media (1993) CRC Press, Boca Raton, FL. [<u style="single">kit</u>〕

The kit is also a feature of the present invention. For example, a kit for producing a glycoprotein containing at least a sugar moiety is provided, which comprises a polynucleotide sequence encoding an O-tRNA and / or a polynucleotide sequence encoding an O-tRNA and / or an O-RS, and / Or includes O-RS. In one embodiment, the kit further comprises an unnatural amino acid having a sugar moiety, or an unnatural amino acid having a moiety for binding the sugar moiety. In another aspect, the kit further comprises instructions for producing the glycoprotein.

Hereinafter, the present invention will be illustrated by examples, but the present invention is not limited by these examples. As a matter of course, the examples and embodiments described herein are for purposes of illustration only, and various modifications or modifications have been conceived by those skilled in the art in view of these descriptions, and such modifications or modifications are also described in the present application. It shall be included in the spirit and scope and the scope of claims.

〔<u style="single">A system for incorporating keto functional groups into proteins</u>] This example describes a system for producing p-acetyl-L-phenylalanine and incorporating this unnatural amino acid into a protein.

The genetic code of most known organisms encodes the same 20 standard amino acids as a building block for protein biosynthesis. In rare cases, selenocysteine (eg Bock, A. et al., (1991) Mol. Microbiol. 5: 515-520) or pyrrolidine (eg Srinivasan, G. et al., (2002) Science 296: 1459-1462; Hao, See B. et al., (2002) Science 296: 1462-1466 ) Is added. The side chains of standard amino acids consist of a surprisingly small number of functional groups--nitrogen bases, carboxylic acids and amides, alcohols and thiol groups, and residual simple alkane or hydrophobic groups. If novel amino acids (eg, amino acids with metal chelating, fluorescent, redox activity, photoactivity or spin-labeled side chains) can be added to the genetically encoded amino acids, then the structure and function of the protein and perhaps the organism itself. It is thought that the operability of the will be significantly increased. Recently, we have added a new component to the translation mechanism of Escherichia coli to place a large number of unnatural amino acids into proteins with high fidelity and site-specifically. It was reported that it could be incorporated vivo (eg Wang, L. et al. (2001) Science 292: 498-500; Wang, L. et al. (2002) J. Am. Chem. Soc. 124: 1836-1837; and Zhang, See Z. et al. (2002) Angew.Chem.Int.Ed.Engl.41: 2840-2842). This example can extend this approach to add keto-containing amino acids to the genetic code of an organism (eg E. coli) and selectively in vitro proteins with a variety of substances using the unique reactivity of keto groups. Prove that it can be modified.

Keto groups are ubiquitous in organic chemistry and are involved in numerous reactions from addition reactions to aldol condensation. In addition, the unique reactivity of the keto group allows it to be selectively modified with hydrazide and hydroxylamine derivatives in the presence of other amino acid side chains. For example, Cornish, VW et al. (1996) J.Am.Chem.Soc.118: 8150-8151; Geoghegan, KF &amp; Stroh, JG (1992) Bioconjug.Chem.3: 138-146; and Mahal, LK et al. (1997) Science See 276: 1125-1128. This important functional group is a cofactor (eg Begley, TP et al. (1997) in Top.Curr.Chem..eds.Leeper, FJ &amp; Vederas, JC (Springer-Verlag, New York), Vol.195, pp.93- (See 142) and metabolites (see, eg, Diaz, E. et al. (2001) Microbiol. Mol. Biol. Rev. 65: 523-569), and as post-translational modifiers of proteins (eg, Okeley, NM &amp; van). der Donk, WA (2000) Chem.Biol.7: R159-R171) Exists, but not on the side chain of standard amino acids. To genetically encode this functional group as p-acetyl-L-phenylalanine in E. coli, a tRNA capable of site-specific insertion of this amino acid into a protein in E. coli in response to an amber nonsense codon (only). -Developed synthetase. Importantly, this tRNA-synthase pair is orthogonal to its corresponding portion of the 20 standard amino acids, i.e., the orthogonal synthetase (only) is obtained by aminoacylating the orthogonal tRNA (only) with only unnatural amino acids. Acylated tRNAs insert unnatural amino acids in response only to amber codons.

〔<u style="single">Materials and methods</u>] Production of p-Acetyl-L-Phenylalanine: Fmoc-4-Acetyl-L-Phenylalanine was purchased from RSP Amino Acid Analogues, Inc. (Worcester, MA). 4 mL of piperidine (20% in dimethylformamide (DMF)) was added to this compound (1.0 g, 2.3 mmol), and the mixture was stirred for 2 hours at room temperature. Evaporation of the solvent gave a white powder. The solids were then resuspended in 10 mL of cold water (0.1% trifluoroacetic acid (TFA)) and the supernatant was collected by filtration. The desired product was separated from the reaction mixture using preparative reverse phase HPLC (Microsorb C18, Rainin Instrument Co., Inc., Woburn, MA) (0.1% TFA addition H).₂O Medium 5 30% CH₃30 minutes using CN). Eluate (t_R= 12 minutes) was lyophilized to give a white solid (0.45 g, 88%).¹1 H NMR (400MHz, D₂O): δ7.85-7.28 (m, 4H), 4.23 (dd, 1H, 5.4Hz), 3.2 (m, 2H), 2.7 (s, 3H). MS Electrospray Ionization (ESI): C₁₁H₁₃NO₃Calculated value [M + 1]⁺208.09, measured value (ESI): 208.47.

Synthesis of p-acetyl- (±) -phenylalanine (see, eg, Cleland, GH (1969) J. Org. Chem. 34: 744-747): N-Bromosuccinimide (NBS) (18.5 g, 105 mmol) 4-methyl After adding acetophenone (13.4 g, 100 mmol) to a solution of carbon tetrachloride (400 mL) under stirring, 2', 2'-azobisisobutyronitrile (AIBN) (0.43 g, 2.5 mmol) was added. The reaction mixture was then heated to reflux for 4 hours. After the reaction (TLC: 8: 1 / hexane: EtOAc) is complete, the solution is water (1 x 100 mL), 1M HCl aqueous solution (3 x 100 mL), 0.5% LVDS.₃Washed with aqueous solution (3 x 100 mL) and brine (1 x 100 mL). Anhydrous PEG combined with organic layers₄Drying in and evaporating the solvent gave a yellow solid, and recrystallization with hexane gave the desired 1- (4-bromoethyl-phenyl) ethanone as a solid (16.8 g, 78%). Absolute ethanol (50 ml) was added dropwise to the sodium pieces (2.3 g, 0.1 mol) washed with pentane under an argon atmosphere for 15 minutes, and the solution was stirred for another 15 minutes. Next, diethyl solid acetamide malonate (2.7 g, 10 mmol) was added under stirring for 30 minutes, and then 1- (4-bromoethyl-phenyl) etanone (2.1 g, 10 mmol) in absolute ethanol was added dropwise over 90 minutes. The mixture was heated to reflux overnight, cooled, and then diethyl ether (150 mL) and water (100 mL) were added to the solution. Separate the organic layer and 0.5% LVDS₃The cells were washed sequentially with (3 x 100 mL) and brine (1 x 100 mL). Anhydrous DDL₄After drying with, the solvent was removed under reduced pressure to obtain a brown gum-like solid. Hexane-dichloromethane (4: 1) was added to the residue, the insoluble material was filtered off, and thoroughly washed with 10: 1 dichloromethane-benzene, 2-acetylamino-2- (4-acetyl-benzyl) malonate diethyl ester. Was obtained as a yellow solid (3.3 g, 95% crude yield). 4M HCl in dioxane was added to this compound and the mixture was stirred overnight. The mixture was then evaporated to dryness and recrystallized from water to give p-acetyl- (±) -phenylalanine (13.2 g, 64% total yield) as a white solid.¹HNMR (400MHz, D₂O): δ7.85-7.28 (m, 4H), 4.27 (dd, 1H, 5.4HZ), 3.30 (m, 2H), 2.68 (s, 3H).¹³C NMR (400MHz, D₂O): δ 195.8,174.3,145.9,133.1,128.9,127.8,60.2,38.3,26.5. MS (ESI): C₁₁H₁₃NO₃Calculated value [M + 1]⁺208.09, measured value 208.07.

Mutant synthetase evolution: Positive selection uses plasmid pYC-J17 and mutRNA^Tyr_CUAThe gene and the chloramphenicol acetyltransferase (CAT) gene, which has a TAG stop codon in Asp112, were expressed. See, for example, Wang, L. et al. (2001) Science 292: 498-500. Supercoil DNA encoding the tyrosine tRNA synthetase (TyrRS) library was transformed into pYC-J17-introduced E. coli DH10B competent cells. Add 17 μg / mL tetracycline, 25 μg / mL kanamycin, 60 μg / mL chloramphenicol, and 1 mM p-acetyl-L-phenylalanine to a minimal medium (GMML) plate containing 1% glycerol and 0.3 mM leucine to cultivate the cells. Plated. After 40 hours incubation at 37 ° C, colonies were pooled and plasmids were isolated. A plasmid encoding mutant synthetase (pBK plasmid) was isolated from pYC-J17 using gel electrophoresis and transformed into E. coli DH10B competent cells into which pLWJ17B3 for negative selection was introduced. The plasmid pLWJ17B3 is a mutRNA under the control of the lpp promoter and rrnC terminator.^Tyr_CUAAnd expresses the balnase gene with 3 amber codons in Gln2, Asp44, and Gly65 under the control of the arabinose promoter. Transformed cells were grown on LB (Luria-Bertani) plates supplemented with 0.2% arabinose, 50 μg / ml kanamycin, and 35 μg / ml chloramphenicol. After 8 hours, cells were removed from the plate and the pBK plasmid was purified with a few more selections. 2nd and 3rd positive In the selection, the concentration of chloramphenicol was increased to 80 and 100 μg / mL, respectively. After alternating 3 positive and 2 negative selections, 9 μg / ml chloramphenicol and p-acetyl-L-phenylalanine in the absence of p-acetyl-L-phenylalanine in an in vivo CAT assay. 120 μg / ml chloramphenicol IC in the presence of₅₀Eleven mutant TyrRS showing values were identified. See, for example, Wang, L. &amp; Schultz, PG (2001) Chem.Biol. 8: 883-890. Although the codon usage of each mutant TyrRS is different, the protein sequences of these mutant TyrRS converged on three independent clones LW1, LW5 and LW6.

Protein expression and purification: Bacteriophage T5 promoter and t using plasmid pLEIZ₀A Z domain gene with an amber codon and a COOH-terminal His6 tag is expressed at position 7 under the control of the terminator, and mutRNA is controlled by the lpp promoter and rrnC terminator.^Tyr_CUAThe gene was expressed. The mutant synthetase gene (LW1RS) isolated from clone LW1 was encoded by the plasmid pBKLW1RS under the control of the constitutive E. coli GlnRS promoter and terminator. 25 μg / mL kanamycin, 34 μg / mL chloramphenicol, and 1.0 mM p-acetyl- (±) -phenylalanine were added to the minimum medium (GMML medium) containing 1% glycerol and 0.3 mM leucine, and pLEIZ and pBK- Escherichia coli DH10B cells co-transformed with LW1RS were grown. Cell is OD₆₀₀When = 0.5 was reached, isopropyl-pD-thiogalactopyranoside (IPTG) (1 mM) was added to induce protein expression. After 5 hours, the cells were pelleted and the protein Ni under denaturing conditions according to the manufacturer's protocol (Qiagen, Valencia, CA).²⁺Purified by affinity chromatography. The protein was then desalted on a PD-10 column (Amersham Pharmacia, Piscataway, NJ) and eluted in water. Protein yields were measured by the Bradford assay (BCA kit, Biorad, Hercules, CA). Protein aliquots were used for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and mass spectrometry.

In vitro protein modification with fluorescein hydrazide and biotin hydrazide: Purified wild-type (wt) and mutant Z-domain proteins were replaced by dialysis with phosphate buffered saline (100 mM potassium phosphate, pH 6.5, 0.5 M sodium chloride). .. Fluorescein hydrazide 1 (Molecular Probe, Eugene, OR) or biotin hydrazide 2 (Molecular Probe, Eugene, OR) was dissolved in DMF and added to 0.07 μmol of each protein in a silane-coated Eppendorf tube to a final concentration of 1 mM. PBS buffer (pH 6.5) was added to a final volume of 0.5 ml. The reaction mixture was maintained at 25 ° C for 18 hours. An unreacted dye or biotin was removed from the protein using a PD-10 column (Amersham Pharmacia, Piscataway, NJ) and the protein was eluted with PBS buffer. To measure labeling efficiency, eluted protein samples were then analyzed by reverse phase HPLC (ZORBAX). SB-C 18,4.6 mm × 250 mm, flow rate 1.0 mL / min, aqueous 50 mM triethylamine acetate buffer, pH 7.0 in 10 40% CH₃70 minutes using CN, Agilent, Palo Alto, CA). Retention time of unlabeled mutant Z domain (t)_R) Is 39.3 minutes, t of the fluorescein hydrazide-labeled mutant Z domain_RIs 40.7 minutes, t of the biotin hydrazide-labeled mutant Z domain_RWas 40.9 minutes.

Fluorescence spectrum measurements: total fluorescence emission spectra are excited 490 nm; excitation and emission bands 4 nm and 4 nm, respectively; photomultiplier tube voltage 950 V; and scanning speed 1 nm / sec FluoroMax-2 Spectrofluorometer (Instruments SA, Inc., Edison, Recorded using NJ). 10 ng of each labeled protein was used. The spectrum reported represents the average of three scans.

〔<u style="single">Results and discussion</u>] Ketogenic amino acids: Ketogenic groups can participate in the addition reaction to the carbonyl group or the acidic Cα position, thus providing a unique chemical reactivity not found in the 20 standard amino acids. This group can be substituted for the natural amino acid cysteine by selective modification of proteins with various chemical reagents. Reactive thiol groups of cysteine are widely used to bind various biophysical probes to proteins. For example, Creight on, TE (1986) Methods Enzymol. 131: 83-106; Altenbach, C. et al., (1990) Science 248: 1088-1092; Brinkley, M. (1992) Bioconjug. Chem. 3: 2-13; Giuliano, KA et al. (1995) Annu. Rev.Biophvs.Biomol.Struct.24: 405-434; Mannuzzu, LM et al., (1996) Science 271: 213-216; Griffin, BA et al. (1998) Science 281: 269-272; Llopis, J. et al., ( 2000) Methods See Enzymol. 327: 546-564; and Gaietta, G. et al., (2002) Science 296: 503-507. However, labeling of single cysteine residues is often difficult due to the presence of two or more reactive groups in proteins and the exchange reactions that occur in the presence of free thiols when using disulfide bonds. Therefore, if a non-protein-producing amino acid having orthogonal reactivity can be used, it is possible to selectively modify a protein even when a single cysteine cannot be selectively labeled or when two different labels are required. Keto easily reacts with hydrazide, hydroxylamine, and semicarbazide in aqueous solution under mild conditions to form stable hydrazone, oxime, and semicarbazone bonds, respectively, under physiological conditions. See, for example, Jencks, WP (1959) J.Am.Chem.Soc.81: 475-481; and, Shao, J. &amp; Tam, JP (1995) J.Am.Chem.Soc.117: 3893-3899.

Several methods have been developed to selectively incorporate carbonyl groups into peptides and small proteins. First, an aldehyde was introduced into the N-terminal of the peptide by oxidizing N-terminal serine or threonine with periodate. The aldehyde group was coupled to a protein fragment containing biotin and a fluorescent reporter (see, eg, Geoghegan, KF &amp; Stroh, JG (1992) Bioconjug. Chem. 3: 138-146) or a COOH-terminated hydrazide via a hydrazone bond (eg,). , Gaertner, HF et al., (1994) J. Biol. Chem. 269: 7224-7230). However, the carbonyl group introduced by this method is limited to the N-terminus and the protein must be stable to oxidation. The peptide segment containing hydrazide or hydroxylamine is then produced using solid phase peptide synthesis (SPPS) and then reacted with a branched chain aldehyde core matrix to form peptide dendrimers (eg Shao, J. &amp; Tam, JP (1995) J.Am.Chem.Soc.117: 3893-3899; and Rose, K. (1994) J.Am.Chem.Soc.116: 30-33), reaction with keto-containing peptide segments Methods have been implemented to form synthetic proteins (see, eg, Canne, LE et al., (1995) J. Am. Chem. Soc. 117: 2998-3007). Although SPPS can introduce keto groups throughout the protein, there are inherent problems with the synthesis of large peptides or proteins. In some cases, this size constraint can be resolved by expressed protein ligation (EPL), which chemically ligates the synthetic peptide to the COOH terminal of the recombinant protein. For example, Muir, TW et al. (1998) Proc.Natl.Acad.Sci.US See A 95: 6705-6710. A peptide containing a ketone group was produced by SPPS and ligated with the Src homology 3 domain of the Abelson protein tyrosine kinase. See, for example, Ayers, B. et al., (1999) Biopolymers 51: 343-354.

In vitro biosynthesis methods have also been used to incorporate keto groups into proteins. See, for example, Cornish, VW et al. (1996) J. Am. Chem. Soc. 118: 8150-8151. In this method, unnatural amino acids containing keto groups are chemically acylated into amber suppressor tRNAs. Fusing an acylated tRNA with a mutant gene in an in vitro extract capable of assisting protein biosynthesis selectively integrates unnatural amino acids in response to UAG codons. This method chemically in suppressor tRNA with unnatural amino acids The protein yield is low because it needs to be aminoacylated in vitro and the acylated tRNA is consumed as a stoichiometric reagent during translation and cannot be regenerated. By developing an orthogonal tRNA-synthetase pair with specificity for p-acetyl-L phenylalanine, ketogenic amino acids can be directly integrated into proteins, for example, in living E. coli cells in response to the UAG codon. As long as it can be expressed in an organism (for example, Escherichia coli), there is no size restriction on the target protein, and it is considered that a large amount of mutant protein can be expressed. Furthermore, it is believed that the label can be selectively introduced into all cells as long as the labeling reagent is cell permeable and non-toxic.

Evolution of mutant synthetase with specificity for p-acetyl-L-phenylalanine: Methanococcus jannaschii tyrosyl tRNA synthetase (TyrRS) and mutant tyrosine amber suppressor tRNA (mutRNA)^Tyr_CUA) Was used as a starting point for making orthogonal tRNA-synthetase pairs. Traditionally, this pair has been shown to be orthogonal in E. coli. See, for example, Wang, L. &amp; Schultz, PG (2001) Chem.Biol.8: 883-890; and Wang, L. et al. (2000) J.Am.Chem.Soc.122: 5010-5011. A library of M. jannaschii TyrRS mutants was prepared and screened to alter the amino acid specificity of TyrRS so that it was loaded with p-acetyl-L-phenylalanine and not loaded with 20 standard amino acids. Residues within 6.5 Å of para position on the aryl ring of bound tyrosine using the crystal structure of homologous Bacillus stearothermophilus TyrRS (see, eg, Brick, P. et al. (1989) J. Mol. Biol. 208: 83-98). The group was identified. Five corresponding residues (Tyr32, Glu107, Asp158, Ile159 and Leul62) of the active site of M.jannaschii TyrRS were randomly mutated by polymerase chain reaction (PCR) and 1.6 × 10⁹A library of the size of (eg, Wang, L. et al. (2001) Science 292: 498-500. This TyrRS mutant library was first introduced into E. coli with plasmid pYC-J17 (eg, Wang, L). (See 2001) Science 292: 498-500) 1 mM p-acetyl-L based on suppression of amber termination codons at non-essential positions (Asp112) of the chloramphenicol acetyltransferase (CAT) gene encoded above. -Positive selection in the presence of phenylalanine. Surviving cells in chloramphenicol are mutRNAs^Tyr_CUAShould encode a mutant synthetase that aminoacylates with a standard amino acid or p-acetyl-L-phenylalanine. Therefore, the DNA encoding the mutant synthetase was isolated and transformed into a negative selection strain expressing the toxic protein balnase gene containing three amber codons at the permissible site (encoded on plasmid pLWJ17B3). mutRNA^Tyr_CUACells encoding mutant synthetase, which loads natural amino acids into, produce balnase and die. Since p-acetyl-L-phenylalanine was not added to the growth medium in the negative selection, the living cells should encode synthetase with specificity for unnatural amino acids. After alternating 3 positive and 2 negative selections with increasing chloramphenicol concentration, a large number of clones whose survival in chloramphenicol depends on the addition of p-acetyl-L-phenylalanine Appeared. These TyrRS were administered using an in vivo assay based on suppression of the Asp112TAG codon within the CAT gene. The characteristics were determined. See, for example, Wang, L. &amp; Schultz, PG (2001) Chem.Biol. 8: 883-890. Eleven TyrRS mutants were identified. Selected synthetase and mutRNA^Tyr_CUACells expressing 1% glycerol and 0.3 mM leucine on a minimal medium plate (GMML plate) survived at 9 μg / ml chloramphenicol in the absence of p-acetyl-L-phenylalanine and of this unnatural amino acid. In presence, cells survived in 120 μg / ml chloramphenicol on GMML plates. This result suggests that the selected mutant synthetase has higher activity on p-acetyl-L-phenylalanine than the native amino acid. Sequencing the DNA of these mutants revealed that the amino acid codon usage was different, but at the protein level, it converged on three independent mutants (LW1, LW5, and LW6). Table 1 shows the active site mutations of mutant synthetase. According to the crystal structure of homologous TyrRS derived from B.stearothermophilus, M.jannaschii The conserved side chains of Tyr32 and Asp158 appear to form hydrogen bonds with the hydroxyl groups of the substrate tyrosine. In mutant synthetase, Tyr32 is mutated to Leu or Ala and Asp158 is mutated to Gly158. It seems that these mutations are detrimental to tyrosine binding and at the same time allow room for the methyl group of p-acetyl-L-phenylalanine. Determining the X-ray crystal structure of mutants will elucidate the exact role of these mutants.<tables num="2"><img file="JP4752001B2_D0004.tif" /></tables>

Typing of mutant proteins incorporating p-acetyl-L-phenylalanine: evolved synthetase and mutRNA^Tyr_CUATo test the ability of p-acetyl-L-phenylalanine to selectively integrate into proteins, staphylococcal protein A tagged with COOH-terminated His6 (eg, Nilsson, B. et al. (1987) Protein Eng. 1: 107) The permissible site (Lys7) of the Z domain gene (see -113) was replaced with an amber termination codon. Since the Z domain has a molecular weight of about 7.9 kD, its mass can be measured with very high accuracy using ion cyclotron resonance (ICR) mass spectrometry. mutRNA^Tyr_CUA, LW1RS and Z domain gene (Lys7TAG) transformed cells were grown in the presence of 1 mM p-acetyl- (±) -phenylalanine. The addition of unnatural amino acids did not affect the growth rate of cells. Mutant protein Ni²⁺Purified by affinity chromatography, the total isolated yield was 3.6 mg / L in the minimum medium. For comparison, when wild-type (wt) TyrRS was used instead of mutant TyrRS, the yield of Z domain was 9.2 mg / L in the minimum medium. p-Acetyl- (±) -Phenylalanine, mutRNA^Tyr_CUAAlternatively, the Z domain was not obtained in the absence of LW1RS, and it was found that the fidelity of unnatural amino acid integration into this site was very high. We also succeeded in incorporating p-acetyl-L phenylalanine into other proteins (eg Cdc42).

mutRNA^Tyr_CUA/ WT TyrRS expressed wt Z domain protein and mutRNA^Tyr_CUABoth mutant Z-domain proteins expressed by / LW1RS were analyzed by electrospray ionized Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS). In the wt Z domain protein, three peaks were observed in mass corresponding to the acetylated form of the intact protein, the protein without the first methionine, and the protein without the first methionine (N-terminal trypsin). Digestive protein Confirmed by tandem mass spectrometry of rugment). For mutant Z-domain proteins, the experimental monoisotopic mass of the intact protein was 7949.893Da, within 2.2 ppm from the theoretical mass of 7949.874Da. The other two peaks are each the first methionine-free protein (M)_Experimental= 78188.838Da, M_Theoretical= 7818.833Da) and its acetylated form (M)_Experimental= 7860.843Da, M_Theoretical= 7860.844Da) corresponds. No peaks corresponding to mutant proteins with other amino acids at the amber codon position were observed in the spectrum. The signal-to-noise ratio above 1500 observed in the intact protein mass spectrum means that the fidelity of p-acetyl-L-phenylalanine integration exceeds 99.8%. Liquid chromatography tandem mass spectrometry of trypsin digests, NH₂The sequence of the terminal peptide was confirmed. NH₂A 606.23 Da precursor ion corresponding to the double-loaded molecular ion of the terminal trypsin-digesting peptide MTSVDNY * INK was isolated and fragmented on an ion trap mass spectrometer (ITMS). Fragment ion masses were clearly assigned, confirming site-specific integration of p-acetyl-L-phenylalanine. As these results clearly show, evolved synthetase is a mutRNA.^Tyr_CUAIn cooperation with, p-acetyl-L-phenylalanine instead of the natural amino acid is incorporated at the position encoded by the amber codon rather than at other positions.

Site-Specific Protein Modification with Fluorescein Hydrazide: Next, we investigated whether the keto group of p-acetyl-L-phenylalanine could function as a chemical handle for in vitro site-specific protein modification. Purified mutant p-acetyl-L-phenylalanine Z-domain protein (mutant Z-domain) and wt Z-domain protein were treated with 1 mM fluorescein hydrazide in phosphate buffer (Scheme 1 at 25 ° C. for 18 hours). The protein was later separated from excess fluorescein hydrazide by size exclusion chromatography and analyzed by dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE). First, the gel was imaged with a fluorescence imaging system and then silver stained. Mutations. The band of the body Z domain shows a fluorescence signal, but fluorescence cannot be detected from the wt Z domain. The fluorescence spectrum was measured with 490 nm excitation using aliquots of these two proteins. P-acetyl-L -Only Z-domain proteins containing phenylalanine show a fluorescence spectrum similar to fluorescein. Since no fluorescent signal was detected in the Z domain, it was found that the labeling reaction occurred only between the hydrazide and the ketone, and that the wt protein had no functional group. Labeled products were analyzed by quadrupole time-of-flight mass spectrometry (QTOF MS). 8425.160 Da (M_TheoreticalAn experimental monoisotopic mass of = 8424.958Da) was obtained, and it was confirmed that fluorescein hydrazide reacted with the mutant Z domain protein at a molar ratio of 1: 1. The reaction mixture was separated by high performance liquid chromatography (HPLC) to determine the degree of labeling. The ratio of the peak area of the labeled Z domain to the peak area of the unlabeled Z domain was 90 ± 5%.<chemistry num="3"><img file="JP4752001B2_D0005.tif" /></chemistry>

Site-specific protein modification with biotin hydrazide: To demonstrate the versatility of this approach, the Z domain was further labeled with a biotin hydrazide derivative (Structure C). Purify mutant and wt Z domain in phosphate buffer at 25 ° C with 1 mM biotin hydrazide 18 Time processed. After dialyzing with phosphate buffer to remove excess biotin hydrazide, the protein was subjected to SDS-PAGE. The separated protein was transferred to a nitrocellulose membrane and probed with a biotin-specific avidin-HRP conjugate. As expected, only the mutant Z domain containing p-acetyl-L-phenylalanine was detected and found to be labeled with biotin hydrazide. No signal was observed in the wt Z domain. When the labeling efficiency was measured by HPLC analysis as described in the fluorescein labeling experiment, it was 80 ± 10%. The labeled protein is QTOF MS (M)_Experimental= 8416.236, M_Theoretical= 8416.146Da) confirmed that it was a product formed from one molecule of biotin hydrazide and one molecule of mutant Z domain. These experiments demonstrate the excellent specificity of the ketone handle for in vitro protein modification.<chemistry num="4"><img file="JP4752001B2_D0006.tif" /></chemistry>

In short, a keto group, which is a novel chemical functional group, was incorporated into the protein in a site-specific manner in vivo. This functional group can be selectively and efficiently labeled in vitro with, for example, fluorescein and biotin by a specific chemical reaction between a keto group and a hydrazide derivative. Using this approach, a variety of other hydrazides or hydroxylases can be used as probes of protein structure and function, to produce proteins with enhanced catalytic or therapeutic properties, or to perform bioassays using proteins. Proteins can be selectively labeled with amine derivatives (sugars, spin labels, metal chelating agents, cross-linking agents, polyethers, fatty acids and toxins, etc.). Small molecules for in vivo imaging of protein localization, protein migration and protein conformational changes during molecular degradation, if unique chemical handles can be directly and site-specifically integrated into proteins in living cells. Fluorophores allow in vivo modification of proteins. This technique can also be used to in vivo label proteins containing p-acetyl-L-phenylalanine in E. coli with a fluorophore.

Application for reference dated October 15, 2003, incorporated into this specification, title of invention "SITE SPECIFIC INCORPORATION OF KETOAMINO ACIDS INTO PROTEINS" (agent reference number 54A-000170PCT) See also.

〔<u style="single">In vivo integration of meta-tyrosine analogs</u>] mtRNA^Tyr_CUAOrthogonal TyrRS was made to aminoacylate (described in Example 1 of WO 2002/085923) with meta-tyrosine analogs.

Preparation of mutant TyrRS library plasmid: Constructed a plasmid library encoding the meta-substituted tyrosine derivative-specific mutant M. jannaschii TryRS, largely following the method described in Example 1 of WO 2002/085923. did. In summary, 6 of the active site of M. jannaschii TyrRS within 6.9 Å of the meta position of the aryl ring of bound tyrosine in the crystal structure of Bacillus stearothermophilus TyrRS using the NNK codon scheme as described in Example 1 above. Residues (Tyr³², Ala⁶⁷, His⁷⁰, Gln¹⁵⁵, Asp¹⁵⁸, Ala¹⁶⁷) Was mutated to all 20 amino acids at the DNA level. The constructed plasmid library pBK-lib is about 1 × 10.⁹Included independent clones.

Evolution of m-acetylphenylalanine integration orthogonal tRNA-synthetase pairs: 5 candidates whose survival in chloramphenicol depends on the addition of unnatural amino acids after 3 positive and 2 negative selections Clone (SEQ ID NOs: 17-21 of WO2002 / 085923 and SEQ ID NOs: 49-53 of WO2002 / 085923) appeared. In the absence of m-acetylphenylalanine, chloramphenicol-resistant ICs in cells into which one of the three mutant TyrRS plasmids has been introduced.₅₀Is 20 μg / ml. Chloramphenicol-resistant ICs of the same cell in the presence of m-acetylphenylalanine₅₀Is 100 μg / ml. The major difference between these two numbers is that the selected synthetase can more specifically integrate m-acetylphenylalanine than the natural amino acid in the cell. The data for m-methoxyphenylalanine was similar, and 5 clones were isolated (SEQ ID NOs: 22-26 of WO2002 / 085923 and SEQ ID NOs: 54-58 of WO2002 / 085923).

Protein expression of DHFR incorporating unnatural amino acids: DHFR in response to amber codon as described in Example 1 of WO 2002/085923 using m-methoxyphenylalanine and m-acetylphenylalanine synthetase selected above. The relevant unnatural amino acid was incorporated. As a negative control, cells introduced with both an orthogonal pair of tRNA / synthetase and an amber mutant vector encoding DHFR were grown in the absence of unnatural amino acids. The results of protein expression are shown in FIG. 10 of WO 2002/085923. From these results, it was clearly proved that the orthogonal pair of tRNA / synthetase specifically incorporates unnatural m-methoxyphenylalanine and m-acetylphenylalanine. The expression yield of DHFR protein was about 0.5 mg / L in both cases.

In one embodiment, a compound (eg, a hydrazide derivative) can be used to label the protein in vivo with at least one unnatural amino acid (eg, a meta-tyrosine analog).

〔<u style="single">Synthesis of glycoprotein mimetics</u>] Selective chemical modification of proteins is greatly facilitated if non-protein-producing functional groups with unique reactivity are available. The keto group is such a chemical handle, absent in the side chains of natural amino acids, and easily and selectively reacts with hydrazide and hydroxylamine derivatives in the presence of standard amino acids under mild conditions. See, for example, Cornish, VW et al., (1996) J. Am. Chem. Soc. 118: 8150-8151 and their references. The keto group is inserted into the peptide by solid phase peptide synthesis and is coupled with the nucleophilic sugar derivative to construct a neosaccharide peptide. See, for example, Rodriguez, EC et al., (1998) J. Org. Chem. 63: 7134-7135. We have recently developed a general method that allows unnatural amino acids to be directly site-specifically incorporated into proteins in living cells (eg WO 2002/085923 and October 15, 2003, which are incorporated herein as reference material). Date-corresponding application, title of invention "Site-specific protein integration of ketoamino acids (SITE) SPECIFIC INCORPORATION OF KETO AMINO ACIDS INTO PROTEINS) , see agent reference number 54-000170PCT). See also Wang, L. et al., (2001) Science 292: 498-500, for example. In response to the amber nonsense codon, we succeeded in incorporating the keto-containing amino acid p-acetyl-L-phenylalanine with translation fidelity exceeding 99.8%. See, for example, Wang, L. et al., (2003) Proc. Natl. Acad. Sci. USA 100: 56-61. This example describes the production of homogeneous sugar protein mimetics using genetically encoded keto functional groups and aminooxysaccharide derivatives.

Two pathways were investigated to generate glycoprotein mimetics (see Figure 1). In the first approach, the sugar derivatized with an aminooxy group is first coupled with the keto group, and the added sugar is enzymatically bound with a glycosyltransferase. In the more convergent second pathway, glycans with a defined structure are produced as aminooxy derivatives and coupled directly to the protein in one step. Since the Z domain of staphylococcal protein A has a relatively small size (molecular weight 7.9 kD), it is very accurate and easy to determine mass spectrometric characteristics, so this was used as a model protein (for example, Nilsson, B. et al., (1987). ). See Protein Eng. 1: 107-113).

The 7th codon of the corresponding gene was mutated to the amber stop codon TAG, and a His6 tag was attached to the C-terminus to facilitate protein purification. A mutant Z-domain protein was obtained by incorporating p-acetyl-L-phenylalanine at the amber position using a previously reported protocol. See, for example, Wang, L. et al., (2003) Proc. Natl. Acad. Sci. USA 100: 56-61. About 3.6 mg / L of protein was obtained after nickel affinity chromatography. Next, a β-linked aminooxy analog of N-acetylglucosamine (GlcNAc) 1 shown in FIG. 1 was synthesized according to the publication method. For example, Cao, S. et al., (1995) Tetrahedron See 51: 6679-6686. Mutant Z domain protein (10 mg / mL) and aminooxysaccharide 1 (21 mM) were mixed in aqueous 100 mM sodium acetate buffer (pH 5.5) and incubated at 37 ° C for 7-26 hours. The reaction mixture was analyzed by reverse phase high performance liquid chromatography (HPLC) by monitoring the absorbance at 280 nm (see Figure 2). Only two main peaks were observed, and the corresponding eluate was characterized by matrix-assisted laser desorption / ionization-Fourier transform ion cyclotron resonance mass spectrometry (MALDI-FTICR MS) (see Figure 3). From the obtained monoisotopic mass, one peak (t)_R= 44.8 minutes) is the unreacted mutant Z domain (M)_theoretical= 7818.833Da, M_experimental= 7818.836Da) and the other peak (t)_R= 43.2 minutes) is the mutant Z domain (M) derivatized with aminooxysaccharide 1._theoretical= 8036.924 Da, M_theoretical= 8036.914 Da) was found to correspond. When expressed in E. coli, the Z-domain protein has three forms: an intact protein, a protein without the first methionine, and an acetylated form of the protein without the first methionine. The intact protein can be separated from the other two forms using reverse phase HPLC. To simplify mass spectrometry, this example used a purified fraction containing the first methionine-free Z domain and its acetylated form. As shown in the spectra of III and IV in FIG. 2, two peaks corresponding to these two forms can be observed in the total mass spectrum. See Figure 1 for the structure. As a control, no sugar derivatized protein is observed when tyrosine is integrated at position 7 of the Z domain. From this fact and the high accuracy mass observed in the sugar-modified Z domain (error &lt;1.2 ppm), it was confirmed that aminooxysaccharide 1 was selectively bound to the keto group. Coupling efficiency increases over time (determined by the area of HPLC peaks corresponding to the starting material and product), and the conversion rate from starting material to product is 42% after 7 hours and 95% after 26 hours. Exceeded (see Figure 2).

Next, it was investigated whether or not the second sugar could be enzymatically coupled to the first sugar. UDP-galactoside in 150 mM HEPES (N-2-hydroxyethylpiperazin-N'-2-ethanesulfonic acid) buffer (pH 7.4) with purified adduct II (5 mg / mL) (see Figure 1 for structure). Incubated at ambient temperature for 48 hours in the presence of (UDP-Gal) (16 mM) and β-1,4-galactosyltransferase (0.4 units / mL). It is known that β-1,4-galactosyltransferase transfers galactose from a sugar nucleotide to the 4-position of the GlcNAc moiety to form Galβ1,4GlcNAc. See, for example, Schanbacher, FL, and Ebner, KE (1970) J. Biol. Chem. 245: 5057-5061. After separation by HPLC New peak (t_R= 42.5 minutes) was identified. Monoisotopic mass of eluate measured by MALDI-FTICR MS (M)_theoretical= 8198.977, M_experimentalFrom = 8198.969), it was confirmed that galactose couples with GlcNAc to form adduct III (see Fig. 3). See Figure 1 for the structure. The coupling efficiency measured by HPLC analysis was about 60%, close to previously reported values for β-1,4-galactosyltransferase. See, for example, Witte, K. et al., (1997) J. Am. Chem. Soc. 119: 2114-2118. This result indicates that the unnatural binding of the first sugar to the protein does not significantly affect the glycosyltransferase reaction. This disaccharide-labeled protein is further reacted with CMP-sialic acid and α-2,3-sialyltransferase (see, eg, Kitagawa, H., and Paulson, JC (1994) J. Biol. Chem. 269: 1394-1401). Then, sialic acid is added to galactose and becomes IV (t)._R= 41.7 minutes) was confirmed by MALDI-FTICR MS (M)_theoretical= 8490.072, M_experimental= 8490.014) (see Figure 3). The coupling efficiency of the conversion from III to IV was 65% according to HPLC analysis. See Figure 1 for the structure.

Protein mimetics III and IV were also made using the convergence pathway. refer graph1. Aminooxy GlcNAc (0.05M) is 2 in 150 mM HEPES buffer (pH 7.4) with a total yield of 70% using β-1,4-galactosyltransferase (0.75 units / mL) and glycosyl donor UDP-galactose. Was converted to. Aminopropyl Silica Gel After purification by HPLC, sialic acid is added to 2 (0.03M) using α-2,3-sialyl transferase (0.22 units / mL) and CMP-sialic acid (0.03M), and the yield is about. At 80%, 3 was obtained in the same buffer as above. Purified aminooxy analogs 2 and 3 (13 and 7.2 mM, respectively) in Z-domain protein (5 mg / mL) containing p-acetyl-L-phenylalanine in 100 mM aqueous sodium acetate buffer (pH 5.5) at ambient temperature. Coupling yielded glycoprotein mimetics III and IV, respectively. refer graph1. The resulting III and IV are consistent with the corresponding adducts made by the first sequential pathway for HPLC and MALDI-FTICR. Confirmed by MS analysis. Under the same reaction conditions, the coupling efficiencies of 2 and I and 3 and I for 26 hours were about 76% and 60%, respectively. The yield was lower than the 1-I coupling (95%), presumably due to the increased steric effect as the glycans became more complex.

In short, we have demonstrated a general method for synthesizing homogeneous sugar protein mimetics containing defined sugar substituents.

〔<u style="single">Experimental materials and methods</u>] Overview: UDP-Gal, CMP-NeuAc, β-1,4-galactosyltransferase (β-1,4-GalT) and α-2,3-sialyltransferase (α-2,3-SialT) were purchased from Calbiochem. .. Unless otherwise specified, all chemicals were obtained from Aldrich, Acros or Sigma and used without further purification. The reaction was monitored by thin layer chromatography (TLC) using ninhydrin or cerium molybdate color former as a developing reagent. The total non-aqueous reaction was carried out in an oven-dried glass container under an Ar atmosphere. All non-aqueous solvents were distilled before use. NMR spectra were recorded with a Bruker AMX-400, AMX-500 or AMX-600MHz spectrometer and residual solvent peaks (CDCl).₃:¹Hδ7.24,¹³Cδ77.0; CD₃OD:¹Hδ3.30,¹³Cδ49.0; D₂O:¹It was collated with Hδ4.76).

Compound 2: Newly prepared MnCl in Figure 1.₂Compound 1 (5 mg, 0.021 m) of FIG. 1 in 350 μL of HEPES buffer (150 mM, pH 7.4) to which the solution (2 mmol) was added. mol) and UDP-Gal (21 mg, 0.032 mmol) were dissolved. β-1,4-GalT (0.3U, 0.1UμL)^-1) And alkaline phosphatase (0.5U, 1UμL)^-1) Was added, and the reaction mixture was gently shaken at ambient temperature for 2 days. Centrifuge the reaction mixture and flow at a flow rate of 1 mL min.^-190 minutes at 100: 0 A: B 50:50 A: B (A = MeCN and B = H)₂The supernatant was purified by aminopropyl silica gel HPLC using the gradient elution of O). The residence time of the desired product was 53 minutes. Lyophilization of the column fraction gave pure compound 2 (6 mg, 70%) of Figure 1 as a white powder;¹H NMR (D₂O, 600MHz) δ4.58 (d, J = 6.12,1H), 4.42 (d, J = 7.44,1H), 3.96 (d, J = 11.88 1H), 3.87 (m, 1H), 3.78 (dd, J = 4.83,12.3,1H), 3.72-3.69 (m, 6H), 3.62 (dd, J = 3.06, 10.08, 1H), 3.56 (m, 1H), 3.50 (m, 1H), 1.98 (s, 3H) ..¹³C NMR (D₂O, 150MHz) δ175.18,103.98,103.31,78.63,75.78,75.13,72.92,72.82,71.39,68.99,61.46,60.43,53.80,22.55. HR-FTMS (pos) C₁₄H₂₆N₂O₁₁Calculated value of [M + Na]⁺= 421.1429, measured value 421.1448.

Compound 3: Newly Prepared MnCl in Figure 1:₂Compound 2 (5.3 mg, 0.013 mmol) and CMP-NeuAc (10 mg, 0.016 mmol) in FIG. 1 were dissolved in 450 μL of HEPES buffer (150 mM, pH 7.4) to which the solution (5 mmol) was added. α-2,3-SialT (22mU, 3.7mUμL)^-1) And alkaline phosphatase (50mU, 50mUμL)^-1) Was added, and the reaction mixture was gently shaken at ambient temperature for 2 days. Centrifuge the reaction mixture and flow at a flow rate of 1 mL min.^-130 minutes at 100: 0 A: B 0: 100 A: B (A = MeCN and B = H)₂The supernatant was purified by aminopropyl silica gel HPLC using the gradient elution of O). The corresponding fractions (27 minutes) were collected and lyophilized to give a white powder (7 mg, 76%).¹H NMR (D₂O, 600MHz) δ4.55 (d, J = 8.34,1H), 4.48 (d, J = 7.86,1H), 4.04 (dd, J = 3.06,9.60,1H), 3.58-3.96 (m, 17H), 3.51 (m, 1H), 2.67 (dd, J = 4.80,12.72,1H), 1.98 (s, 3H), 1.96 (s, 3H), 1.75 (t, J = 12.30,1H). ES-MS (neg) C₂₅H₄₃N₃O₁₉Calculated value [MH]^-= 688, measured value 688.

General procedure for coupling an aminooxysaccharide derivative to a mutant Z-domain protein: In a typical reaction, an aminooxysaccharide derivative (500 μg) and a mutant Z-domain protein ~ 1 mg are added to 100 mM NaO Ac buffer (pH 5.5). ) Was dissolved. Water was added to a total volume of 100 μL and the reaction mixture was shaken at 37 ° C. for 26 hours. The mixture is then centrifuged and the flow rate is 1 mL min.^-170 minutes 90:10 A: B 60:40 A: B (A = H)₂The supernatant was purified by reverse phase HPLC on an Agilent ZORBAX SB-C18 4.6 mm × 250 mm column utilizing gradient elution of 0.1% TFA in O and 0.1% TFA in B = MeCN). Column fractions were neutralized with TrisCl buffer (pH 7.0) and desalted on a size exclusion column. After elution with water, the eluate was lyophilized to give the pure II, III, and IV of FIG. 1 as white powder in 96%, 76%, and 60% yields, respectively.

Production of Glycoprotein Mimetics III and IV (Figure 1) Using the Sequential Route: Newly Prepared MnCl to Produce III in Figure 1.₂II (~ 0.5 mg) and UDP-Gal (1 mg) in FIG. 1 were dissolved in 90 μL of 150 mM HEPES buffer (pH 7.4) to which the solution (0.5 mmol) was added. β-1,4-GalT (40mU, 40mUμL)^-1) And alkaline phosphatase (50mU, 50mUμL)^-1) Was added, and the reaction mixture was gently shaken at ambient temperature for 2 days. The reaction mixture is centrifuged and the supernatant is purified by reverse phase HPLC. Ta. Newly prepared MnCl to produce IV in Figure 1.₂III (~ 0.5 mg) and CMP-NeuAc (0.5 mg) in FIG. 1 were dissolved in 90 μL of 150 mM HEPES buffer (pH 7.4) to which the solution (0.5 mmol) was added. α-2,3-SialT (10mU, 3.7mUμL)^-1) And alkaline phosphatase (50mU, 50mUμL)^-1) Was added, and the reaction mixture was gently shaken at ambient temperature for 2 days. The reaction mixture was centrifuged and the supernatant was purified by reverse phase HPLC.

MALDI-FTICR MS: The MALDI-FTICR MS experiment used a Bruker Daltonics (Billerica, MA) home-building device with an APEX II console and a 9.4T magnet. Sugar moieties tend to disintegrate when conventional MALDI sample preparations containing TFA are used. We used a low-sensitivity, low-temperature matrix. The matrix is 3-hydroxypicolinic acid (20 mg mL)^-1) And diammonium citrate (1 mg mol)^-1) Is a mixture. FTICR's special medium-pressure MALDI source, which reduces metastable degradation by providing collision cooling to the ion source, further minimizes glycoprotein degradation.

〔<u style="single">Another Strategy for Glycoprotein Synthesis</u>] In one aspect of the invention, another strategy has been developed for synthesizing homogeneous glycoproteins in an organism (eg, E. coli) by translating glycosylated amino acids. For example, myoglobin containing β-GlcNAc-serine at a defined position can be expressed in E. coli with good yield and high fidelity. The β-GlcNAc moiety can be recognized by a sugar-binding protein or can be late modified with a galactosyltransferase. This approach appears to be applicable to other post-translational modifications (eg, protein phosphorylation, acetylation, methylation, etc.).

Glycosylation is one of the most common post-translational modifications of proteins in eukaryotes, affecting a wide range of protein functions from folding and secretion to biomolecule recognition and serum half-life. See, for example, RADwek, (1996) Chem. Rev. 96: 683. Although the effects of glycosylation have been elucidated considerably, the specific role of oligosaccharide chains and the relationship between their structures and functions have only just begun to be elucidated. See, for example, CR Bertozzi, &amp; LLKiessling, (2001) Science 291: 2357. The main problem is that it is difficult to isolate the intrinsic sugar form from a natural source because the sugar protein is generally produced as a mixture of sugar forms. Although various methods have been developed for synthesizing sugar forms of defined structure, there are significant drawbacks in the size, quantity, and / or quality of the sugar proteins produced. For example, P. Sears, &amp; CH Wong, (2001) Science 291: 2344; M. Wacker et al., (2002) Science 298: 1790; BG Davis, (2002) Chem. Rev. 102: 579; and H CHang, &amp; See CR Bertozzi, (2001) Acc. Chem. Res. 34: 727. In this example, it was used to produce a unique sugar form in E. coli, including the development of an orthogonal synthetase-tRNA pair that genetically encodes a glycosylated amino acid in response to a selector codon (eg, amber codon, TAG). Describe the strategy and ingredients. Direct genetic protein integration of the above and other sugar-modified amino acids can significantly enhance the ability to analyze and manipulate glycoprotein structures and functions.

Amino acids with novel chemical and physical properties are used in Escherichia coli (eg L. Wang et al., (2001) Science 292: 498; L. Wang et al., (2002) J. Am. Chem. Soc. 124: 1836; Z. Zhang Et al., (2002) Angew.Chem.Int.Ed.Engl.41: 2840; JWChin et al., (2002) J.Am.Chem.Soc.124: 9026; JWChin et al., (2002) Proc .Natl.Acad.Sci.USA 99:11020; SW Antoro et al., (2002) Nat.Biotechnol.20: 1044; L. Wang et al., (2003), Proc.Natl.Acad.Sci.USA 100:56; and Z .Zhang et al., (2003) Biochemistry (See 42: 6735)) and several methods have been previously developed that first made it possible to systematically add to the genetic code of yeast (see, eg, JWChin et al., Science, (printed in 2003)). In this approach, the amber suppressor M. jannaschii TyrRS-mutRNA, which does not cross-react with endogenous tRNA and synthetase.^Tyr_CUAWe have developed pairs and evolved them to load only the desired unnatural amino acids. This method can also incorporate glycosylation, phosphorylation, or methylated amino acids directly into the protein (see, eg, T. Arslan et al., (1997) J. Am. Chem. Soc. 119: 10877), a selective enzyme for the protein. Alternatively, there is no need for post-chemical translation modification. Attempts have been made to site-specifically incorporate β-O-GlcNAc-L-serine (Compound A, GlcNAc: N-acetylglucosamine) into proteins in Escherichia coli. O-GlcNAc modification is ubiquitous in almost all eukaryotic cells, is involved in cell signaling, protein trafficking and regulation of cell proliferation, and is also a substrate for producing more complex sugars. For example, L. Wells et al., (2001) Science 291: 2376; and N. Lamarre-Vincent, &amp; See L.Hsieh-Wilson, (2003) J.Am.Chem.Soc.125:6612. However, since sugar derivatives having free hydroxyl groups do not easily cross the membrane of eukaryotic cells, substrate compound A does not appear to be cell permeable. See, for example, AKSarkar et al., (1995), Proc. Natl. Acad. Sci. USA 92: 3323. On the other hand, it has been shown that the acetylation of the hydroxyl group of the sugar promotes the passage through the cell membrane, and that the hydroxyl acetyl group can be introduced into the cell and then deacetylated by the non-specific cytosol esterase. See, for example, N. Lamarre-Vincent, &amp; L. Hsieh-Wilson, (2003) J. Am. Chem. Soc. 125: 6612. Therefore, in these experiments, the precursor N-Fmoc-tri-acetyl-β-GlcNAc-serine was used as a commercially available acetylated derivative tri-acetyl-β-GlcNAc-serine (Compound B). The compound is represented by the following formula.<chemistry num="5"><img file="JP4752001B2_D0007.tif" /></chemistry>

Orthogonal mutRNA in E. coli from a library of active site mutants using a series of positive and negative selections^Tyr_CUATyrRS, which specifically loads β-GlcNAc-serine, was isolated. Based on the X-ray structure of homologous Bacillus stearothermophilus TyrRS, it is encoded by two libraries with randomly placed active site residues, namely the plasmid pBK-lib-m, and the residue Tyr.³², Ala⁶⁷, His⁷⁰, Gln¹⁵⁵, Asp¹⁵⁸, And Ala¹⁶⁷The residue Tyr, encoded by the plasmid pBK-lib, with a first library of randomly placed³², Glu¹⁰⁷, Asp¹⁵⁸, Ile¹⁵⁹, And Leu¹⁶²We built a second library with randomly placed. All of these residues are within 6.9 Å of the phenyl ring and are the major residues that form the substrate binding pocket. Approximately 2.6 x 10 independent clones in total in the library⁹And said. The library was then positively selected to select TyrRS mutants capable of incorporating glycosylated amino acids based on the suppression of amber codons introduced into Asp112 of the chloramphenicol acetyltransferase (CAT) gene. Cells that survive in the presence of high concentrations of chloramphenicol are thought to express the mutant TyrRS, which has the ability to insert β-GlcNAc-serine or endogenous amino acids in response to the Asp112TAG amber codon. Next, three toxic balnase genes Negative selection based on amber codon repression was used to eliminate the mutant TyrRS incorporating endogenous amino acids from selected clones. After 5 positive and 4 negative selections, 3 clones emerged that survived in the presence of high concentrations of chloramphenicol. These clones and their mutations are as follows. S1-90 (Glu)¹⁰⁷ Pro¹⁰⁷, Asp¹⁵⁸ Cys¹⁵⁸, Ile¹⁵⁹ Tyr¹⁵⁹, Leu¹⁶² Arg¹⁶²), S4-5 (Tyr³² Gly³², Glu¹⁰⁷-Gly¹⁰⁷, Asp¹⁵⁸ Cys¹⁵⁸, Leu¹⁶² His¹⁶²), S1-5 (Glu)¹⁰⁷ Cys¹⁰⁷, Asp¹⁵⁸ His¹⁵⁸, Ile¹⁵⁹ Asp¹⁵⁹, Leu¹⁶² Met¹⁶²). 30 μg / ml chloramphenicol using 1 mM serine, α-tri-acetyl-GalNAc-threonine, α / β-tri-acetyl-GalNAc-serine or β-tetra-acetyl-Glu-asparagin instead of compound B All of these clones appear to be highly selective for β-GlcNAc-serine, as no cell proliferation is obtained above Cole. From these in vivo genetic results, the newly selected mutant TyrRS is considered to have excellent specificity for β-GlcNAc-L-serine.

To test the integration efficiency and fidelity of compound B, a mutant myoglobin gene (Gly4TAG) containing an amber codon and a C-terminal His6 tag at position 4 was generated. See, for example, SW Antoro et al., (2002) Nat. Biotechnol. 20: 1044. Mutant synthetase S1-90 in minimal medium in the presence of compound B mutRNA^Tyr_CUAAnd when co-expressed with the Gly4TAG myoglobin gene, a full-length mutant myoglobin 1 mg / L was produced (see FIG. 4). For comparison, wild-type myoglobin 5.5 mg / L was produced under similar conditions, and the suppression level of S1-90 was found to be good. S-90, mutRNA^Tyr_CUAOr, in the absence of compound B, no expression of full-length myoglobin was observed by silver-stained SDS-PAGE (see FIG. 4).

Figure 4 shows the expression of Gly4 compound A mutant myoglobin (~ 18.5 kD). Protein Ni²⁺-Purified by affinity chromatography and degraded by SDS-PAGE. The gel was silver stained. Lane 1 shows that myoglobin was expressed in the presence of orthogonal tRNA, synthetase S1-90, and compound B. The ~ 18kDa band corresponds to the full length myoglobin. Lane 2 shows the protein eluted after expression in the presence of orthogonal tRNA and synthetase S1-90 and in the absence of substrate compound B. Lane 3 shows the protein eluted after expression in the presence of orthogonal tRNA and substrate compound B and in the absence of synthetase S1-90. Lane 4 shows the protein eluted after expression in the presence of synthetase S1-90 and substrate compound B and in the absence of orthogonal tRNA. Lane 5 contains purified wild-type myoglobin for comparison.

According to high resolution MALDI-TOF analysis, the monoisotopic mass of His6 tag purified mutant myoglobin is 18430.1 Da, Glc (OH).₃Theoretical mass of myoglobin containing Nac-serine and deficient in methionine (M)_theoretical= 1.8429.5 Da) and within 32ppm. See Figure 5. Note that deletion of the N-terminal Met is common in E. coli. In addition, no signal corresponding to O-acetylated myoglobin or wild-type myoglobin was observed. Mass spectral data confirmed the high specificity of myoglobin integration of GlcNAc-serin (&gt; 96%).

Several addition experiments were performed to further characterize the mutant myoglobin. First, an ELISA-like assay was used to analyze the binding of GlcNAc-specific lectin Bandeiraea simplicifolia II (BSII) (see, eg, S. Ebisu et al., (1978), Carbohydr. Res. 61: 129) to wild-type myoglobin and sugar myoglobin. did. See Figure 6A. Figure 6A shows the GlcNAc-specific lectin Bande The binding of riraea simplicifolia II (BSII) to wild-type myoglobin and sugar myoglobin is shown. Wild-type myoglobin, sugar myoglobin, and negative control (without lectin) A₄₀₅Indicates the value. Gly4 Compound A mutant myoglobin (200 ng) and wild-type myoglobin (200 ng) were immobilized in microtiter plate wells and then incubated in the presence of biotinylated BSII and streptavidin-alkaline phosphatase conjugate. Wells were incubated in the presence of p-nitrophenyl phosphate and monitored by measuring absorbance at 405 nm. Two forms of myoglobin were immobilized in microtiter plate wells and then incubated in the presence of biotinylated BSII, streptavidin-alkaline phosphatase conjugate, and p-nitrophenyl phosphate, respectively. Wells containing wild-type myoglobin generated signals equivalent to negative control wells. On the other hand, wells containing sugar myoglobin generated at least 200 times more signal than wild-type myoglobin, demonstrating selective recognition by GlcNAc-specific lectins. Furthermore, since this lectin is highly selective for GlcNAc, the results also show that the sugar is not modified to other isomers such as GalNAc and ManNAc (eg S. Ebisu et al., (1978), See Carbohydr. Res. 61: 129).

We also investigated whether the O-GlcNAc-serine residue of myoglobin could be selectively modified with galactosyltransferase. It is known that β-1,4-galactosyltransferase transfers galactose (Gal) from the nucleotide UDP-Gal to the 4-position of N-acetylglucosamine (GlcNAc) to form Galβ1,4GlcNAc. To investigate whether O-glycosylated myoglobin could be modified with UDP-Gal, both wild-type and O-glycosylated myoglobin were degraded by SDS-PAGE and transferred to PVD membranes. The membrane is then milk galactosyltransferase and radioactive UDP- [H³]-Incubated for 24 hours at room temperature in the presence of galactose. See, for example, K. Kamemura et al., (2002), J. Biol. Chem. 277: 19229. By exposing the film to an X-ray film [H³]-Monitored Gal uptake. Only the sugar myoglobin was labeled and no detectable signal was observed with wild-type myoglobin. See Figure 6B. Figure 6B shows UDP- [H³] Shows on-blot galactosyltransferase labeling of the sugar myoglobin with galactose. Wild-type myoglobin (1 μg) and Gly4 compound A mutant myoglobin (1 μg) were degraded by 12% SDS-PAGE and transferred to PVD membranes. Then the membrane is milk galactosyltransferase (1U), UDP- [H³] -Treatment with galactose (0.5 μCi) and bovine intestinal alkaline phosphatase (1 U) for 24 hours at room temperature. After thorough cleaning, the film was exposed to X-ray film using enhanced autoradiography.

Further glycosyl transfer reactions were performed in solution for quantitative analysis. See, for example, K. Witte et al., (1997) J. Am. Chem. Soc. 119: 2114. A 72% disaccharide yield was obtained based on the radioactive label present after incubation at room temperature for 48 hours. See Figure 6C. Figure 6C shows a quantitative analysis of the galactosyltransferase reaction performed in solution and standardized the radiolabeled galactose so that 1.0 corresponds to 100% metastasis. HPLC purified wild myoglobin (100 μg) and Gly4 compound A mutant myoglobin (100 μg) in a solution containing pyruvate kinase (5U), UDP-glucose pyrophosphorylase (1U), inorganic pyrophosphorylase (10U), galactose- 1-phosphate-uridyl transferase (1U), milk galactosyl transferase (2U), glucose-1-phosphate (3 μmol), uridine diphosphate (3 μmol), phosphoenolpyruvate (0.01 mmol), and DTT (2 μmol) ) Was added. After adjusting the reaction solution to pH 7.2, [H³] -Galactose-1-phosphate (0.01 mmol) was added. The reaction was carried out at room temperature for 48 hours. Protein products were separated on a PD-10 Sephadex 25 column. Radioactive label uptake measured with a liquid scintillation analyzer Ta.

As a result of these tests, it was proved that β-GlcNAc-L-serine can be incorporated into a protein at the time of translation in Escherichia coli with excellent specificity and good yield. The incorporated β-GlcNAc-serine can function as a major glycosylation site to which sugars can be sequentially added with glycosyltransferases. For example, K. Kamemura et al., (2002), J. Biol. Chem. 277: 19229.

〔<u style="single">Materials and methods</u>] Directed evolution of the mutant TyrRS enzyme. General procedures for positive and negative selection have been previously reported. See, for example, Z. Zhang et al., (2003) Biochemistry, 42: 6735. In summary, the plasmid pBK-lib-m (eg, Z. Zhang et al.,) Was introduced into the competent E. coli DH10B into which the plasmid pRep (2) / YC (see, eg, SW Antoro et al., (2002) Nat. Biotechnol. 20: 1044) was introduced. A combination of (2003) Biochemistry 42: 6735) and pBK-lib (see, eg, L. Wang et al., (2001) Science 292: 498) was transformed. GMML medium (1% glycerol, 0.3 mM leucine, 1 mM MgCl) supplemented with 40 μg / ml tetracycline, 50 μg / ml kanamycin, 68 μg / ml chloramphenicol, and 1 mM compound B.₂, 0.1 mM CaCl₂Transformed cells were grown at 37 ° C for 60 hours in 500 ml of 1 × M9 minimal medium containing 0.5% NaCl. The plasmid (pBK) was purified from living cells and transformed into E. coli DH10B into which pLWJ17B3 (see, eg, L. Wang et al., (2001) Science 292: 498) was introduced to initiate negative selection. The cells were then plated on LB (Luria-Bertani) plates supplemented with 40 μg / ml chloramphenicol, 50 μg / ml kanamycin, and 0.02% L-arabinose and incubated at 37 ° C for 8 hours. The plasmid pBK was purified from living cells and used for subsequent positive and negative selection. After 5 positive and 4 negative selections, 3 candidate orthogonal tRNA-synthetase pairs conferring substrate-dependent chloramphenicol resistance were isolated and sequenced.

Expression and characterization of mutant myoglobin. Kanamycin, tetracycline, 0.02% L-arabinose, 5 μM FeCl₃, And pBAD / JYAMB-4TAG (see, eg, SW Antoro et al., (2002) Nat. Biotechnol. 20: 1044) and pS1-90-introduced DH10B cells were grown in 500 ml of GMML medium supplemented with 0 or 1 mM compound B. .. Cells are pelleted, lysed and Ni under natural conditions²⁺-Proteins were purified by affinity chromatography using NTA beads. Proteins were analyzed by 12% SDS-PAGE and stained with silver. High-resolution mass spectrometry of aliquots of purified protein was performed. Molecular weight of proteins was measured using matrix-assisted laser desorption / ionization (MALDI) by a time-of-flight (TOF) mass spectrometer (Voyager DE-STR, Applied Biosystems, Foster City, CA). The protein sample was desorbed and ionized after irradiation with a 337 nm nitrogen laser. Sinapinic acid was used as the MALDI matrix. Lectin binding and according to established protocols (see, eg, K. Kamemura et al., (2002), J. Biol. Chem. 277: 19229; and K. Witte et al., (1997) J. Am. Chem. Soc. 119: 2114). A glycosyltransferase reaction was performed.

〔<u style="single">Typical O-RS sequence</u>] Representative O-RSs that can be used in the present invention include SEQ ID NOs: 1 to 6 (see Table 2), and typical O-tRNAs that can be used in the present invention include SEQ ID NO: 7. I can get rid of it. Representative polynucleotides encoding O-RS include SEQ ID NOs: 8-10.

As a matter of course, the examples and embodiments described herein are for purposes of illustration only, and various modifications or modifications have been conceived by those skilled in the art in view of these descriptions, and such modifications or modifications are also described in the present application. It shall be included in the spirit and scope and the scope of claims.

Although the present invention has been described in some detail so as to be clearly understood, it is obvious to those skilled in the art from the above disclosure that various changes can be made to the form and details without departing from the true scope of the present invention. Is. For example, all the above techniques and devices can be used in various combinations. All publications, patents, patent applications, and / or other documents cited herein incorporate the entire disclosure as reference material for all purposes, and each publication, patent, patent application, and / or other document. Is treated as an individual description when it is incorporated as a reference material for all purposes.<tables num="3-1"><img file="JP4752001B2_D0008.tif" /></tables><tables num="3-2"><img file="JP4752001B2_D0009.tif" /></tables><tables num="3-3"><img file="JP4752001B2_D0010.tif" /></tables>

<figref num="1">An example of two schemes (sequential pathway and convergence pathway) for binding a sugar moiety to a polypeptide containing an unnatural amino acid is schematically shown.</figref><figref num="2">Shown is an HPLC analysis of the 7-hour and 26-hour coupling reactions of mutant Z-domain protein I (Fig. 1) containing aminooxysaccharide 1 (Fig. 1) and p-acetyl-L-phenylalanine.</figref><figref num="3">High resolution MALDI-FTICR MS spectra of mutant Z domain proteins I (Fig. 1), glycoprotein mimetics II, III, and IV (Fig. 1) are shown. 2 of each spectrum⁺Shows isotope clusters.</figref><figref num="4">It shows the expression of Gly4 A mutant myoglobin (~ 18.5kD). Protein Ni²⁺It was purified by affinity chromatography and degraded by SDS-PAGE. The gel was silver stained.</figref><figref num="5">MALDI-TOF analysis of the molecular weight of the Gly4 A mutant myoglobin is shown.</figref><figref num="6-1">The characterization of purified mutant myoglobin containing glycosylated amino acids is shown. A shows the binding of GlcNAc-specific lectin Banderiraea simplicifolia II (BSII) to wild-type myoglobin and sugar myoglobin. B is UDP- [H³] Shows on-blot galactosyltransferase labeling of the sugar myoglobin with galactose.</figref><figref num="6-2">The characterization of purified mutant myoglobin containing glycosylated amino acids is shown. C showed a quantitative analysis of the galactosyltransferase reaction performed in solution and standardized the radiolabeled galactose so that 1.0 corresponds to 100% metastasis.</figref>