WO2020116659A1 - Preparation method for disulfide-rich peptide library, cross-linked polypeptide capable of binding with target polypeptide, and method for preparing same - Google Patents

Abstract

The purpose of the present invention is to provide: easy and convenient preparation of a polypeptide having a structure which is not easily destroyed by an enzyme; a peptide library that can be used in preparation of such a polypeptide; and a method for preparing the peptide library. Provided is a preparation method that is for a library of disulfide-rich peptides having a structure similar to any polypeptide selected from the group consisting of conotoxins and analogs thereof, and having the ability to bind to a desired target protein, the method comprising a preparation step for preparing a cross-linked protein having a specific structure in a solution containing 10-90% of an organic solvent, by causing a protein having a desired sequence to undergo a crosslinking reaction using a crosslinking agent.

Description

Method for preparing disulfide-rich peptide library, cross-linked polypeptide capable of binding target polypeptide, and method for preparing the same

The present invention relates to expanding the structural diversity of molecules used in an in vitro selection experiment by using a non-aqueous buffer solution as a chemical cross-linking agent or reaction conditions.

Although antibody drugs have been able to realize the specific and effective treatment that was a problem of small molecule drugs, they also have therapeutic problems.
The first problem is that antibody drugs are very expensive compared to other drugs. This is mainly because the antibody is a protein having a relatively large molecular weight (polypeptide, about 150 kDa) composed of a plurality of polypeptide chains, and the production cost is high. In addition, the antibody has features such as (1) low intracellular targeting ability and tissue permeability, (2) relatively low thermal stability, and (3) difficult site-specific modification.

Another issue is antigenicity. Antibodies, which are giant glycoproteins, may be recognized as foreign to the living body because of their size. For this reason, it is necessary to change to a chimeric antibody or a humanized antibody so that the antibody is not recognized as a foreign substance. At present, several antibody engineering techniques have been developed to solve this problem, and antibody drugs can be put to practical use.
However, the problems of low intracellular targeting ability of antibodies and low tissue permeability still remain. The low intracellular targeting ability described above is directly linked to the problem that the target antigen is limited, and depletion of the target antigen can be a major obstacle to the further development of antibody drugs.

Under these circumstances, peptide drugs have been attracting attention as post-antibody drugs. A peptide is a medium molecule with a molecular weight of about 10,000 to 10,000 kDa, and is larger than a small molecule and smaller than an antibody. Compared with low molecular weight compounds, peptides have the advantage that they can be expected to have improved target affinity and specificity due to the diversity of their amino acid sequences. It is also easy to mimic or inhibit protein-protein interactions.
Compared with antibodies, peptide drugs have the advantage of being excellent in tissue penetration due to their small molecular size and generally not antigenic. Furthermore, peptide drugs can be produced by chemical synthesis or can be site-specific chemically modified. In addition, since it is a molecule of a size that can be handled by various display technologies described later, it has an advantage that rapid research and development can be expected.
There is also an advantage that a wider variety of molecules can be targeted. Examples of the target include not only biomolecules, but also inorganic surfaces (metals, semiconductors and metal oxides), organic molecules (nanocarbons and polymers), and the like. Artificial peptides have been designed that specifically bind.

By the way, it is known that naturally there are peptides that have a three-dimensional structure with a disulfide bond and have high molecular recognition ability, high affinity, and stability. Examples of such peptides include ω-conotoxin, which is a kind of neurotoxin produced by Conus, and peptide such as cyclotide, which is often contained in plants and has an antibacterial action. Cycloviolacin having a disulfide bond at three positions is often contained in the Rubiaceae, Violetaceae, and Cucurbitaceae plants (see FIGS. 1A and 1B).

Conotoxin is a protein composed of 11 to 30 amino acids, and is a neurotoxin that is used when predating small fish, which are fed by conus shells. Although general peptides are easily degraded by proteolytic enzymes and have low stability in blood, conotoxins are not degraded in blood, so when they are injected into the target organism, they strongly stimulate various ion channels and acetylcholine receptors. Block and paralyze this target. The reason why it is not degraded in blood is that conotoxin has multiple disulfide bonds, and the structure is restricted and stabilized by the disulfide bonds. In other words, conotoxins are excellent peptide aptamers for ion channels and acetylcholine receptors.

In addition, conotoxins include α-conotoxins whose action points are acetylcholine receptors transmitted from nerves to muscles, δ-conotoxins whose action points are voltage-dependent sodium ion channels, and κ-conotoxins whose action points are potassium channels. It is known that there are five types, μ-conotoxin whose action point is a voltage-dependent sodium ion channel in muscle, and ω-conotoxin whose action point is an N-type calcium ion channel.
Since N-type calcium ion channels in the spinal cord are involved in the transmission of pain sensation, conotoxin is administered as a powerful analgesic when administered intrathecally, and its analgesic effect is said to be two to three orders of magnitude stronger than morphine. Therefore, it is expected to be used as a therapeutic drug for pain in patients with terminal cancer who have become resistant to morphine. Furthermore, the conotoxin scaffold molecule is also used as a drug.

Wilson A.J. (2009)Inhibition of protein-protein interactions using designed molecules.Chem. Soc.Rev., 38, 3289-3300 Kates S. A., Sole N. A., Johnson C. R., Hudson D., BaranyG., AlbericioF. (1993) A novel, convenient, three-dimensional orthogonal strategy for solid-phase synthesis of cyclic peptide , Tetrahedron Lett., 34, 1549-1552.

As mentioned above, biopharmaceuticals are shifting from antibodies to peptides with smaller molecular weight in recent years. Since an antibody is composed of a plurality of polypeptide chains, it can have a higher-order structure, and it is difficult to expose the recognition site for digestive enzymes or proteolytic enzymes. Therefore, there is an advantage that it is difficult to be decomposed by such an enzyme. On the other hand, peptides have the advantages of high intracellular targeting ability and high thermostability and easy site-specific modification, and are cheaper in production cost than antibodies.

However, peptides have the problem that they are less stable in blood than the above antibodies. This is because the peptide is a low-molecular weight compound and does not easily have a complicated structure, so that the recognition sites of digestive enzymes and proteolytic enzymes are exposed, and these enzymes are easily decomposed.

Some peptides have already been put to practical use as first-generation biopharmaceuticals, but due to the property that they are easily decomposed by the above-mentioned enzymes, when orally administered, they are rapidly decomposed by the above-mentioned enzymes, so that the action is sustained. It is known that there are problems that it is extremely short, it is difficult to penetrate the gastrointestinal mucosa, and side effects may occur with antibody production. For this reason, it is said that the development of peptides has been delayed even though they are expected as therapeutic agents.

Among these, the issue of particular concern is the stability of the peptide. Peptides have low resistance to proteolysis, and are decomposed within minutes by proteases present in blood and tissues, and then excreted out of the body. Therefore, in order for the peptide to act stably as a drug in the body, it is necessary to extend the half-life of the peptide until the function of the peptide is exerted. That is, the development of peptide drugs requires a strategy to improve the stability of peptides.

-One of the means to improve the stability of such peptides is to make the peptides cyclic. Proteases that decompose peptides are divided into exoproteases that cleave 1-2 amino acids from the end of the protein and endoproteases that cleave from the center of the protein, depending on the cleavage site. Cyclization of peptides is effective in inhibiting degradation by exoproteases, since exoproteases cannot degrade cyclic peptides. For example, a cyclic peptide in which the amino terminus and the carboxyl terminus of the peptide are linked by a peptide bond and a cyclic peptide using a chemical cross-linking agent having a thiol group of cysteine as a reactive group have been obtained by screening. Although there are naturally occurring cyclic peptides that are resistant to digestion, the mechanism of cyclic peptide formation in cells has not been clarified.

In order to solve the above problems and improve the stability of peptides, N-methylation (improved membrane permeability and stability), incorporation of unnatural amino acids (improved specificity and stability), and addition of PEG ( Various chemical modification protocols such as reduction of reductive clearance), cyclization (improvement of stability) and other structural constraints (for example, construction of steric structure by disulfide bond) have been developed. Particularly, drug design approaches based on structural constraints including cyclization have been most actively studied. Most of the peptides actually used as pharmaceuticals are naturally-occurring cyclic peptides and include insulin, cyclosporin A, gramicidin S, oxytocin, vancomycin and the like. FIG. 2 shows the chemical structures of typical cyclic peptides, cyclosporin (A) and oxytocin (B).

In order to stabilize the structure of the peptide, formation of a bicyclic peptide using halogen, or substitution of a disulfide bond with another bond in chemically synthesizing a naturally occurring peptide can be mentioned.
Cyclic peptides have improved resistance to proteolysis when compared to linear peptides. Furthermore, since it is expected that a high affinity for a specific molecule or a high specificity for a target molecule can be expected, peptides are emerging as a template for developing therapeutic agents and diagnostic agents.

However, the cyclic peptide has a complicated structure and cannot be easily synthesized. In addition, when peptide synthesis is performed in an aqueous solvent, there is a problem that the crosslinked structure is broken by protease and other enzymes. For this reason, there has been a strong social demand for the preparation of peptides which are simple and whose structure is not easily broken by an enzyme.

Moreover, it is known that peptides have structural diversity. "Structural diversity of peptides" means that peptides having the same sequence have different three-dimensional structures. For example, when a peptide is cross-linked in an aqueous solvent using a known chemical cross-linking agent, the hydrophilic surface is exposed to the aqueous solvent and cross-linking is performed so as to maintain such a structure. That is, structural diversity is unlikely to occur because only normal modes of crosslinking occur. On the other hand, when the peptide is crosslinked in an organic solvent or other non-aqueous solvent, the peptide is exposed to the solvent on the hydrophobic surface, and the peptide is crosslinked so as to maintain such a state. That is, cross-linking occurs that does not occur in an aqueous solvent, and structural diversity is likely to be generated. Then, such a crosslinked peptide may have a different folding from that in the case of being crosslinked under an aqueous solvent condition (see FIG. 3 ).
However, peptides cross-linked with the hydrophobic surface exposed to the solvent cannot be synthesized in aqueous solvents. Therefore, there has been a strong social demand for synthesizing such peptides.

Also, when such a peptide has a function of binding to a target protein (target polypeptide), it can be used as a drug or a diagnostic agent. For this reason, there was a strong social interest in creating a peptide library that could be used for the preparation of such peptides.

The inventors of the present invention have earnestly studied under the above circumstances and completed the present invention.
That is, the present invention provides a DNA library preparation step for preparing a disulfide-rich peptide library; a transcription step for preparing and purifying mRNA from the DNA library; and a step of binding the purified mRNA obtained in the transcription step and a linker. a binding step for forming an mRNA-linker conjugate; a translation step for translating the mRNA-linker conjugate in a cell-free translation system to form an mRNA-peptide conjugate; a mRNA to a cDNA in the mRNA-peptide conjugate And a reverse transcription step of forming an mRNA-peptide-cDNA conjugate that binds to the mRNA-peptide conjugate.

A chemical cross-linking library preparing step of chemically cross-linking the mRNA-peptide-cDNA conjugate to prepare a chemical cross-linking library; a target polypeptide immobilizing step of immobilizing a target polypeptide on resin particles; An in vitro selection step of performing in vitro selection using the immobilized target polypeptide to obtain a polypeptide having the desired sequence; and using a disulfide bond or a cross-linking agent in a solution containing an organic solvent, And a cross-linking step of cross-linking the polypeptide having the desired sequence to form a cross-linked polypeptide having a specific structure, which is a method for preparing a cross-linked polypeptide capable of binding to a target polypeptide.

Here, the “unique structure” means that a structure cross-linked with a hydrophobic surface exposed to a solvent, a recognition site of a digestive enzyme, a protease or another enzyme that decomposes a peptide is inside the peptide. A folded structure or the like.
The purified mRNA and the linker are preferably bound by an enzyme or photocrosslinking, and the binding of the purified mRNA and the linker is carried out by ligase when a cyanovinylcarbazole-free linker is used, and cyanovinylcarbazole is included. When a linker is used, it is preferably performed by UV irradiation.

When a linker containing cyanovinylcarbazole is used, the time required for cross-linking is as short as 5 minutes or less, and the DNA constituting the main chain of the linker and the RNA bound to this linker can be treated without damage. Because they can be combined. In addition, the working time is higher than that in the case where ligase is used because the protein is eaten by several tenths. In the translation step, it is preferable to remove the ribosome bound to the mRNA-peptide conjugate after translation. A chelating agent is preferably used for removing the ribosome, and the chelating agent is preferably ethylenediaminetetraacetic acid or glycoldiaminetetraacetic acid.

The solution containing the organic solvent preferably contains 10 to 90% by volume of the organic solvent, and the organic solvent is preferably any solvent selected from the group consisting of dimethylformamide, acetonitrile, and ethanol. Further, the cross-linked polypeptide having the desired sequence preferably contains three or more cysteines and can bind to the target polypeptide. Here, the target polypeptide is preferably any polypeptide selected from the group consisting of an antigen protein, an in vivo signal transduction substance, an in vivo signal transduction receptor, and a tumor marker protein. Further, the cross-linking agent is preferably any one selected from the group consisting of bismaleimideethane, bismaleimide propane, and α,α′-dibromo-o-xylene.

Yet another aspect of the present invention is a cross-linked polypeptide capable of binding the target polypeptide, prepared by the above method. Here, it is preferable that the cross-linking polypeptide has any polypeptide selected from the group consisting of peptides containing 3 or more cysteines, or a structure similar thereto. In addition, it is preferable that the crosslinked polypeptide has the sequence described in any one of SEQ ID NOs: 1 and 2 in the following sequence listing from the viewpoint of stability of the crosslinked structure.

According to the method for preparing a crosslinked polypeptide of the present invention, there is a great advantage that a crosslinked polypeptide having a specific structure capable of binding to a target polypeptide can be prepared in a solution containing an organic solvent.

Further, in vitro selection is performed using the disulfide-rich peptide library prepared by the above-mentioned preparation method to obtain a polypeptide having the desired sequence, thereby obtaining various cross-linked polypeptides capable of binding to the target polypeptide. There is also an advantage that you can.

Further, the cross-linking polypeptide capable of binding to the target polypeptide of the present invention has a fundamental difference in structure from the cross-linking polypeptide obtained in an aqueous solution containing no organic solvent, and thus has the advantage of high protease resistance. ..

FIG. 1 is a schematic diagram showing the structures of ω-conotoxin (A) and cycloviolacin O1 (B). FIG. 2 is a diagram showing the chemical structures of typical cyclic peptides, cyclosporine (A) and oxytocin (B). FIG. 3 is a schematic diagram showing the difference between the structure (A) of a crosslinked peptide synthesized in an aqueous solvent and the structure (B) of the structure of a crosslinked peptide synthesized in a non-aqueous solvent when transferred to an aqueous solvent. Is.

FIG. 4 is a diagram showing the structure and active site of midkine. FIG. 5 is a schematic diagram showing the structure of a linker that does not contain anovinylcarbazole (cnvK) for cDNA display (hereinafter sometimes referred to as “SBP-rG linker”). FIG. 6 is a gel electrophoresis image showing the result of PCR amplification of the initial library.

FIG. 7 is a gel electrophoresis image showing the purification results of the ligation product, translation product, and reverse transcription product in which mRNA and SBP linker were linked. FIG. 8 is a gel electrophoresis image showing the result of His-tag purification of the peptide translated by cell-free translation. FIG. 9 is a gel electrophoresis image for confirming the immobilization of midkine on Sepharose resin.

FIG. 10 is a gel electrophoresis image for confirming immobilization of midkine on streptavidin magnetic beads. FIG. 11 is a gel showing each washing solution and eluate from the first round to the fourth round of a system using a peptide library crosslinked with bismaleimideethane (hereinafter, sometimes referred to as “BMOE”). It is an electrophoretic image.

Figure 12: Gel electrophoresis showing a library (T7+) in which the T7 region promoter region was added, the DNA was restored to full construct by adopting the overlap extension method, and then the already obtained DNA was amplified by PCR using the template. It is a statue. FIG. 13 is a diagram schematically showing the upshift assay. Figure 14 shows that after cross-linking the clones in 20% acetonitrile (hereinafter abbreviated as "ACN") using TBMB at a final concentration of 10 μM, gel electrophoresis (8M urea denaturation 6%) was performed. It is a gel electrophoresis image showing the result of SDSPAGE, 20 mA, 2 hours).

Figure 15 shows the results of cross-linking clones using a final concentration of 10 mM TBMB in 80% ACN and gel electrophoresis (8 M urea denaturation 6% SDS PAGE, 20 mA, 2 hours). It is a gel electrophoresis image. FIG. 16 is a gel electrophoresis image after upshift assay using TBMB (final concentration 10 mM) and clones in 80% ACN. FIG. 17 is a gel electrophoresis image showing the result of an upshift assay using BMOE (final concentration 4 mM) in a phosphate buffer and S-body library.

FIG. 18 is a gel electrophoresis image showing that IL6R could be immobilized on Sepharose resin. FIG. 19 is a gel electrophoresis image showing that IL6R could be immobilized on streptavidin magnetic beads.

FIG. 20 is a gel electrophoresis image showing the condition of in-vitro selection in each round. FIG. 21 is a schematic diagram showing a pull-down assay scheme. FIG. 22 is a gel electrophoresis image showing the result of a pull-down assay of the peptide group selected in vitro.

FIG. 23 is a graph showing the amount of IL6R bound to the peptide group before and after in vitro selection. FIG. 24 is a diagram showing the amino acid sequence of conotoxin. (A) shows the natural conotoxin, (B) shows the amino acid sequences of the peptides used for the analysis of peptide cross-linking under different solvent conditions, and these are modified from the sequences shown in (A) above. Is.

FIG. 25 is a schematic diagram showing the structure of a linker containing cyanovinylcarbazole (cnvK) for cDNA display (hereinafter sometimes referred to as “cnvK linker”). FIG. 26 is a gel electrophoresis image showing the result of amplification by PCR1. FIG. 27 is a gel electrophoresis image showing the result of amplification by PCR2. FIG. 28 is a gel electrophoresis image showing the result of amplification by PCR3.

FIG. 29 is a schematic diagram (A) of the structure of the PCR3 product and a graph (B) showing the occurrence rate of amino acids in the cysteine expression region. FIG. 30 is a gel electrophoresis image when the above PCR3 product is transferred. FIG. 31 is a gel electrophoresis image showing that the cnvK linker and the above mRNA were crosslinked. FIG. 32 is a gel electrophoresis image showing that the ligation product was translated, immobilized on the above magnetic beads, and purified by His-tag. FIG. 33 is a gel electrophoresis image showing the effect of a solvent on the reverse transcription of the above product and the crosslinking of the reverse transcription product (under 90% DNF or under 4% DMF).

FIG. 34 is a gel electrophoretic image showing the results of examining the difference in crosslink formation due to the crosslinker. FIG. 35 is a diagram showing the results of examining the resistance of streptavidin magnetic beads to organic solvents. FIG. 36 is an HPLC chromatogram showing changes in the peptide structure before and after cross-linking when peptide 1 shown in FIG. 24 is cross-linked in 90% DMF or in a conjugation buffer that is an aqueous solvent containing 4% DMF. is there.

FIG. 37 is an HPLC chromatogram showing the structural change of the peptide before and after crosslinking when peptide 1 was replaced with peptide 2. FIG. 38 is an HPLC chromatogram showing changes in the peptide structure before and after crosslinking when peptide 1 was replaced with peptide 3. FIG. 39 is a HPLC chromatograph showing changes in the peptide structure before and after cross-linking when Peptide 1 shown in FIG. 24 is cross-linked in a conjugation buffer which is an aqueous solvent containing 90% ACN or 10% ACN and 4% DMF. It is gram.

FIG. 40 is an HPLC chromatogram showing changes in the peptide structure before and after crosslinking when peptide 1 was replaced with peptide 2. FIG. 41 is an HPLC chromatogram showing changes in the peptide structure before and after crosslinking when peptide 1 was replaced by peptide 3.

FIG. 42 is a mass spectrogram when the peptide 1 was cleaved with trypsin. (A) shows the results when crosslinked in 90% DMF, and (B) shows the results when crosslinked in the conjugation buffer. FIG. 43 is a mass spectrum gram when the above peptide 1 was replaced with peptide 2. FIG. 44 is a mass spectrogram when the peptide 1 is replaced with the peptide 3.

Hereinafter, the present invention will be described in more detail with reference to the drawings as necessary.
The present invention comprises a target polypeptide comprising a preparation step of preparing a cross-linked polypeptide having a specific structure in a solution containing an organic solvent by carrying out a cross-linking reaction of a polypeptide having a desired sequence with a cross-linking agent. A method for preparing a cross-linked polypeptide capable of binding to

Further, the present invention includes (a1) a DNA library preparation step, (a2) a transcription step, (a3) a binding step for forming an mRNA-linker conjugate, (a4) a translation step, and (a5) reverse transcription. Step, (a6) crosslinked library preparation step, (a7) target polypeptide immobilization step, (a8) in vitro selection step, and (a9) in a solution containing an organic solvent, obtained in the above in vitro selection step And a cross-linking step of cross-linking the obtained polypeptide.

In the DNA library preparation step (a1) above, a cDNA display method is used to prepare a DNA library from mRNA containing a desired nucleotide sequence under normal conditions. Then, in the transcription step (a2), mRNA is prepared from the obtained DNA library according to a standard method and purified.

Subsequently, in the binding step of (a3) above, the mRNA purified in the transcription step of (a2) above is bound to a linker to form an mRNA-linker conjugate. The linker used in this step preferably has at least an mRNA binding site, a reverse transcription primer binding region, and a peptide binding site. For example, a linker as shown in FIG. 4 may be used. It is preferable in terms of work efficiency after the translation process described below. If necessary, a linker as shown in FIG. 5 can be used.

Next, in the translation step (a4), the mRNA-linker conjugate obtained in the binding step (a3) is translated in a cell-free translation system to form an mRNA-peptide conjugate. As the cell-free translation system, use of rabbit reticulocyte lysate (RRL), wheat germ extract, insect cells (SF9 or SF21, etc.), etc. is effective for ribosome, translation factor and tRNA synthesis. It is preferable because it contains all required cell macromolecules such as.

Subsequently, in the reverse transcription step of (a5) above, cDNA is produced from the mRNA in the mRNA-peptide conjugate using a reverse transcriptase, and the cDNA is bound to the mRNA-peptide conjugate to bind mRNA-peptide-cDNA. Let the body form. As the reverse transcriptase, PrimeScript series (manufactured by Takara Bio Inc.) or the like can be used.

Next, in the cross-linking library formation step of (a6) above, the mRNA-peptide-cDNA conjugate obtained as described above is allowed to form a disulfide bond or is cross-linked using a chemical cross-linking agent to form a cross-linking library. Let it form.

Subsequently, in order to examine the binding property with the crosslinked library obtained in the crosslinked library forming step of (a6), the following target substances are immobilized on the resin in the (a7) step according to a conventional method. As such a resin, it is preferable to use various magnetic beads, since the mRNA-peptide-cDNA conjugate can be released from the resin-bound mRNA-peptide-cDNA conjugate in a lump. The mRNA, peptide, or cDNA can then be individually cleaved from the linker.
Then, using the crosslinked library and the target polypeptide immobilized on the resin, for example, in vitro selection using a cDNA display method can be performed to obtain a polypeptide having the desired sequence.

As the cross-linking agent used for forming the cross-linked polypeptide having the specific structure, bismaleimidoethane or bismaleimide propane can be used, and using these, a disulfide-rich peptide library can be efficiently prepared at low cost. be able to. "
Subsequently, in order to examine the binding property with the crosslinked library obtained in the step (a6) of forming the crosslinked library, a target substance as described below is immobilized on the resin. Then, using the crosslinked library and the target polypeptide immobilized on the resin, in vitro selection can be performed using, for example, a cDNA display method to obtain a polypeptide having the desired sequence.

Examples of the target substance (target polypeptide) include, as described above, antigen proteins and fragments thereof (eg, IgG, scV, Fab, Fc, VHH, etc.), various signal transducing substances in vivo (eg, hormone, Cyclic AMP (cAMP), second messenger such as cyclic GMP (cGMP), interleukin, midkine and other cytokines), in vivo signal receptor (cAMP receptor, cGMP receptor, interleukin receptor, etc.) , And tumor marker proteins (scc, CA-125, CEA, PSA, etc.) and the like.

As such a target molecule, for example, midkine or interleukin 6 (IL6) can be used. Midkine is a basic low-molecular-weight protein (polypeptide) that is classified into cytokines and growth factors. There are 5 pairs of disulfide bonds, 3 pairs at the N-terminal side and 2 pairs at the C-terminal side. have. It is known that the activity of midkine is at the C-terminus, and is most expressed in the metaphase embryo in humans, but it is restricted to vascular epithelium and specific mucosal epithelium after adulthood. .. FIG. 4 shows the structure and active site of midkine.

It is also known that midkine increases in serum and/or urine of patients with various cancers (blood concentration is 0.15 ng/mL in healthy subjects and 0.5 ng/mL or more in cancer patients). It has been attracting attention as a tumor marker. In addition, midkine is an esophageal cancer, gastric cancer, colon cancer, liver cancer, pancreatic cancer, thyroid cancer, lung cancer, breast cancer, bladder cancer, uterine cancer, ovarian cancer, prostate cancer, neuroblastoma. It is suitable as a target molecule because it shows an increase in blood concentration in patients with cytomas, glioblastoma, etc.

Interleukin 6 (hereinafter also referred to as “IL6”), which is a type of cytokine, is used in various cells such as T cells, B cells, monocytes, fibroblasts, keratinocytes, endothelial cells, mesangial cells, and fat. It is known that it is produced by cells and some tumor cells, and that it is mainly expressed by T cells, monocytes, activated B cells, neutrophils and other hematopoietic cells. It is also known that an increase in the amount of IL6 produced leads to the development of various autoimmune diseases, inflammatory diseases and the like. There are two types of IL6R, which is a receptor for IL6, and they are classified into a membrane-bound type (hereinafter sometimes referred to as “mIL6R”) and a soluble type (hereinafter sometimes referred to as “sIL6R”). IL6 exhibits biological activity via IL6R and gp130. IL6 bound to membrane-bound IL6R induces a homodimer of gp130, forming a high-affinity receptor complex consisting of IL6, IL6R, and a trimer of gp130. SIL6R bound to IL6 also forms a trimer with gp130.

As a molecular targeting drug targeting IL6R, tocilizumab (humanized human IL6R monoclonal antibody: TCZ) can be mentioned. TCZ inhibits both mIL6R and sIL6R and blocks IL6-mediated signal transduction, and thus is used to improve symptoms such as rheumatoid arthritis, Castleman's disease, and generalized juvenile idiopathic arthritis.

Then, in the target polypeptide immobilization step of (a7), the target polypeptide is immobilized on the resin, and in the in vitro selection step of (a8), the crosslinked library and the immobilized target polypeptide are reacted. Then, in vitro selection is performed to obtain a polypeptide having the desired sequence.
Then, mRNA may be obtained from the cDNA of the obtained polypeptide, and the steps (a3) to (a8) may be repeated a desired number of times. This repetition is called a round, and by increasing the number of rounds, it is possible to obtain a nucleotide sequence of a peptide that has converged to some extent.

Next, the obtained peptide is crosslinked in a solution containing an organic solvent to prepare a crosslinked polypeptide having a unique structure. In the present invention, an aqueous solvent containing an organic solvent is used, as shown in FIG. 3(A), even if the peptide does not change its structure in the aqueous solvent (buffer aqueous solution), it is a non-aqueous solvent. This is because the structure taken inside may change in the aqueous solvent due to the change in the polarity of the solvent (see FIG. 3(B)).

The cross-linked peptide obtained as described above is preferably a polypeptide containing three or more cysteines, because it has a wide range of applications as a pharmaceutical scaffold. Specifically, the cross-linking polypeptide contains three or more cysteines and can bind to a target polypeptide. Examples of such a peptide include the above-mentioned conotoxin or its analogs. .. The crosslinked polypeptide having such a structure is preferably produced in a solution containing any organic solvent selected from the group consisting of dimethylformamide, acetonitrile and ethanol. When the content of the organic solvent in the solution is 10 to 90% by volume, the cross-linked polypeptide having the unique structure can be efficiently prepared while maintaining the diversity, and thereby the desired disulfide-rich can be obtained. Peptide libraries can also be prepared.

The formation of the cross-linked polypeptide having the above-mentioned specific structure may be formed by the cysteine contained in the desired position of the above peptide, or may be formed by using a cross-linking agent. As the cross-linking agent used for forming the cross-linked polypeptide, bismaleimide ethane, bismaleimide propane, α,α′-dibromo-o-xylene and the like can be used. Use of these cross-linking agents makes it possible to efficiently prepare a disulfide-rich peptide library at low cost.

Here, the conotoxin exemplified above is a peptide in which 11 to 30 amino acids are connected and a peptide having a unique structure having three SS bonds, as shown in FIG. 1(A), Not easily decomposed by enzymes. Therefore, a peptide having 3 or more cysteines is expected to be less likely to be decomposed by an enzyme, and thus can be preferably used.

Note that "conotoxin" shall collectively include the above-mentioned α-conotoxin, δ-conotoxin, κ-conotoxin, μ-conotoxin, and ω-conotoxin. In addition, the "analog of conotoxin" refers to a compound having a cyclic cystine knot (CCK) motif. That is, it refers to a compound in which the C-terminus and the N-terminus are linked to form a ring and which has three disulfide bonds in the molecule. And it is known that conotoxin has six kinds of superfamilies.

Cyclotide compounds include bracelet cyclotide, which is the most common subgroup, Mobius cyclotide containing cis-proline in loop 5 shown in Fig. 1(B), and trypsin having a sequence similar to that of an acyclic trypsin inhibitor. • Inhibitors-including Knottons.

The cross-linked peptide can be prepared as described above. In addition, the binding between such a cross-linking peptide and the target peptide can be analyzed by using gel electrophoresis or another analysis method after reacting under desired conditions. Examples of such an analysis technique include gel electrophoresis and upshift assay. Further, the change in the structure of the crosslinked peptide prepared as described above can be confirmed to some extent by the change in the HPLC chromatogram.

(Example 1) Screening for midkine using S-body library and cross-linking agent BMOE (1) Target polypeptide etc. Since the structural change of cross-linking peptide can be detected, midkine (ATGEN) And manufactured by IL6R (manufactured by MBL Science).

(2) Gel electrophoresis (2-2) Nucleic acid polyacrylamide gel electrophoresis (PAGE)
In the PAGE performed in this example, the influence of the three-dimensional structure of the nucleic acid was made negligible by adding urea as a denaturant and setting the temperature of the thermostatic bath during the migration to 60°C. As a result, the migration rate of the nucleic acid was changed only by the difference in the size. Therefore, the nucleic acid molecules were fractionated by the size.

(2-3) Preparation of reagents The composition of 10 x TBE, 2 x gel loading buffer, and 40% acrylamide solution (acrylamide:bisacrylamide = 19:1) was as shown in Tables 2 to 4 below.

(2-4) Preparation of Nucleic Acid Polyacrylamide Gel The stains on the glass plate with spacers, the comb, and the gasket were rinsed with MilliQ water, and dust was wiped off with a Pro wipe. After sufficiently drying the above three members, a gasket was installed along the spacer, and then both ends of the glass plate were fixed with clips. Next, a nucleic acid PAGE solution having the composition shown in Table 5 below was prepared, and the nucleic acid PAGE solution was injected into the gap between the assembled glass plates. A comb was quickly inserted into the gel injected into the gap of the glass plate, and the gel was left to stand until it solidified.

Gel electrophoresis was performed according to the following procedure. After the above gel hardened, the surface of the glass plate was washed with MilliQ water, and the comb and the lower member of the gasket were removed from the glass plate. The electrophoresis plate was placed on the holder and fitted into the upper tank, and the lower tank of the electrophoresis tank was filled with the buffer for electrophoresis. 0.5X TBE was prepared by diluting 10X TBE 20 times and used as a buffer for electrophoresis.

Next, while being careful not to let air bubbles enter the lower tank, I attached it to the migration tank containing the migration buffer and filled the upper tank with migration buffer. After this, pre-running was performed for 9 minutes at a constant voltage of 200V. After the completion of pre-running, the electrophoretic sample was applied to the gel, and a voltage was applied for a time depending on the size of the sample and the hardness of the gel. During the migration, a constant temperature bath was used to maintain the temperature of the lower bath of the bath at 60°C.

(2-5) Analysis of migration results by imager When the nucleic acid of the sample is labeled with a fluorescent substance such as FITC, the fluorescence intensity of the fluorescent substance is first read, and then the nucleic acid is stained with STBR-Gold to determine the light intensity. Read For the SYBR-Gold labeling, the stock solution diluted 10,000 times was used.

A solution having the composition shown in Table 6 or 7 below was prepared for polypeptide polyacrylamide gel electrophoresis (SDS-PAGE).

10x SDS-running buffer was diluted 10 times to make 1xSDS-running buffer, which was used as a buffer for electrophoresis. Electrophoresis was performed at a constant voltage of 20 mA in the same manner as for nucleic acid polyacrylamide electrophoresis. After completion of electrophoresis, labeling was performed using FITC fluorescence of SBP linker, SYPRO Ruby, and Lumitein, and the band position was confirmed by observing the fluorescence.

(3) Screening In order to compare the effect of the cross-linking structure by the chemical cross-linking agent and the cross-linking structure by the conventional disulfide bond on the binding ability of the molecule, affinity screening for midkine was performed. Screening for midkine was performed in parallel using a chemically crosslinked peptide (hereinafter referred to as “S-body”) library and an S-body library crosslinked by disulfide bonds.

For the crosslinking reaction, 1x conjugation buffer (0.2 M sodium phosphate (pH 7.2), 0.5 M TCEP-HCl, and bismaleimideethane (BOME), Thermo Fisher Scientific), oxidative folding buffer ( 0.1 M Tris (pH 7.0), 1 mM EDTA, 2 mM GSSG, 7 mM GSH, 0.05% Tween 20), dimethyl sulfoxide (Nacalai Tesque, Inc.), reduced glutathione and non-reduced glutathione (Sigma-Aldrich), EZ-Link ^™ maleimide-PEG11-biotin (manufactured by Thermo Fisher Scientific) was used.

For fixation of midkine, Illustra Micro Spin Empty Column (manufactured by GE Healthcare), Amicon Ultra-0.5 centrifugal filter device 3K (manufactured by MERCKMILLIPORE), NHS-activated Sepharose 4 fast flow (manufactured by GE Healthcare) , EZ-Link Sulfo-NHS-SS-Biotin (manufactured by Thermo Fisher Scientific) and Ts beads (FG beads) (manufactured by Tamagawa Seiki) were used.

Also, as a construct for the S-body library, T7Ω-Library(MGC(27AA))-HisX6-C(Ytag) having the following sequence was created. In the sequences below, X represents any nucleotide. As the puromycin linker, the SBP linker having the structure shown in Table 8 below and FIG. 5 was used.

(SEQ ID NO: 9 in the sequence listing)
GATCCCGCGA AATTAATACG ACTCACTATA GGGGAAGTAT TTTTACAACA ATTACCAACA ACAACAACAA ACAACAACAA CATTACATTT TACATTCTAC AACTACAAGC CACCATGGGC TGCNNNNNNN NNNNNNNNNN NNNNNNNNNN NNNNNNNNNN NNNNNNNNNN NNNNNNNNNN NNNNNNNNNN NNNNNNNNNN NNNNGGGGGA GGCAGCCATC ATCATCATCA TCACGGCGGA AGCAGGACGG GGGGCGGCGG GGAAA

(4) Preparation of cDNA display from peptide library (4-1) Amplification by PCR DNA synthesis of disulfide-rich peptide library was requested to Eurofin Genomics. The library was purified by performing PCR using the solutions shown in Table 9 below so that the diversity of the initial library was not lost and subbands did not appear.

The PCR program consisted of incubation at 98° C. for 1 minute, incubation at 98° C. for 10 seconds, followed by incubation at 68° C. for 20 seconds for 5 cycles, incubation at 68° C. for 1 minute, and cooling to 4° C. The PCR product was ethanol-precipitated and concentrated, and then the primer and enzyme were removed using the PCR-Clean-up Mini Kit. At this stage, the PCR product was confirmed by the above-described 8 M urea-denatured nucleic acid polyacrylamide gel electrophoresis (electrophoresis performed on nucleic acids such as DNA and RNA; hereinafter also referred to as “PAGE”).
The library DNA prepared by PCR was confirmed by 6% PAGE so that the diversity of the initial library was not lost and subbands did not appear. The DNA was stained with SYBR Gold, and the results are shown in Fig. 6.

(4-2) Transcription DNA was then transcribed into RNA using RiboMAX-Large Scale RNA production systems. Table 10 below shows the reaction composition and reaction conditions for one tube.

Transferring reaction was performed by incubating at 37℃ for 3 hours in the reaction solution with the above composition. After that, 1 μL of RQ1 RNase-Free DNase was added and further incubated at 37°C for 15 minutes. After 15 minutes, immediately after that, it was purified using the After Tri Reagent RNA Clean-up Kit according to the attached protocol. At this stage, transcripts and peptide libraries were confirmed by the above 8M urea denaturing PAGE.

(4-3) Ligation Next, the purified mRNA was ligated to the SBP linker. The composition of the ligation reaction solution for one tube is shown in Table 11 below.

The annealing conditions were 90°C for 2 minutes, 1 minute to 70°C, 70°C for 1 minute, 15 minutes to 25°C, and 25°C for 30 seconds. After annealing, 1.0 μL of T4 RNA ligase and 0.5 μL of T4 polynucleotide kinase were added to each sample and incubated at 25°C for 30 minutes. The ligation product was subjected to 8M urea denaturation 6% PAGE (200V, 25 minutes) using mRNA as a control. Gel electrophoresis was stained with SYBR Gold. The results are shown in Fig. 7.

(4-4) Cell-free translation Next, the mRNA-linker conjugate was translated in a cell-free translation system in a solution for cell-free translation having the composition shown in Table 12 below to give the mRNA-protein (polypeptide) conjugate. Formed.

25 μL of the reaction solution having the composition shown in Table 12 was dispensed into each tube and then incubated at 30° C. for 25 minutes. Then, 12 μL of 3 M KCl and 6 μL of 1 M MgCl ₂ were added, and the mixture was further incubated at 37° C. for 60 minutes. The mRNA-linker conjugate to be added was 120 pmol in the first round, 24 pmol in the second round, and 18 pmol in the third and subsequent rounds. The solution composition in each round was adjusted to an amount suitable for carrying out cell-free translation.

(4-5) Purification of mRNA-Protein (Polypeptide) Conjugate (In Vitro Virus: IVV) Subsequently, RNase was removed from streptavidin (SA) magnetic beads (Dynabeads MyOne Streptavidin C1). 300 μL of Dynabeads MyOne Streptavidin C1 was placed in a 1.5 mL tube and left on a magnetic stand. All the subsequent operations for removing the supernatant were performed on a magnetic stand. The supernatant was discarded, resuspended in 1x binding buffer and the supernatant removed again.

Next, because the residual ribosome may hinder the steric hindrance during the immobilization of magnetic beads, the ribosome bound to the post-translational mRNA-peptide conjugate was removed as follows. First, 65 μL of 0.5 M EDTA (final concentration 70 mM) was added as a chelating agent. To this, an equal amount of 2x binding buffer was added, mixed with washed Dynabeads MyOne Streptavidin C1 and incubated at room temperature for 25 minutes, and the supernatant was removed. This supernatant was subjected to 8M urea denaturation 6% SDS-PAGE (20mA, 2 hours) together with the ligation product as Sup.1 to confirm the content. The gel was stained with FITC. The results are shown in Fig. 8.

(4-6) Reverse transcription Next, unstable mRNA was reverse transcribed into cDNA. First, Dynabeads MyOne Streptavidin C1 used for the purification of IVV was washed with 1 x ReverTraAce buffer. Then, a solution having the composition shown in Table 13 below was prepared on Dynabeads MyOne Streptavidin C1.

Next, this solution and Dynabeads MyOne Streptavidin C1 were mixed and incubated at 42°C for 45 minutes, and after the incubation was completed, the supernatant was removed to obtain a cDNA display library peptide. The solution containing the obtained cDNA peptide was subjected to 8M urea-denatured 6% SDS-PAGE (20mA, 2 hours) to confirm the content. The gel was stained with FITC. The results are shown in Fig. 8.

(4-7) Cross-linking reaction Next, a cross-linking reaction was performed using a chemical cross-linking agent in order to form a cross-linked structure in the peptides of the cDNA display library. The reduction reaction of the disulfide bond in the peptide obtained as described above was performed using 235 μL of 1× conjugation buffer containing 5 μL of 0.5 M TCEP.

The above conjugation buffer and Dynabeads MyOne Streptavidin C1 were mixed and incubated at 25°C for 5 minutes. Then, the cross-linking agent BMOE was added to a final concentration of 4 mM and suspended, and the mixture was incubated at 25°C overnight. Subsequently, in order to prepare an S-body library crosslinked by disulfide bonds, oxidative folding treatment was performed. This treatment makes it possible to surely form a disulfide bond. After completion of reverse transcription, 180 μL of oxidation folding buffer was added to Dynabeads MyOne C1 and incubated overnight at 4°C.

(4-8) Purification of uncrosslinked peptide In order to remove uncrosslinked cDNA display, the following treatment was performed. Prior to biotinylation of uncrosslinked cDNA display, it was mixed with 1 mM biotin to block the biotin binding site on Dynabeads MyOne Streptavidin C1 and incubated for 15 minutes at room temperature.
Then, the supernatant was discarded, and 0.1 M maleimide-PEG11-biotin (which binds to a thiol group) and uncrosslinked cDNA display were incubated at 25°C for 30 minutes to allow binding, and removed by Dynabeads MyOne Streptavidin C1.

(4-9) RNase treatment Next, the cDNA display was cut off from Dynabeads MyOne Streptavidin C1. First, the oxidation folding buffer was removed, washed with His-tag wash buffer, and the supernatant was discarded. 75 μL of RNase treatment buffer containing 74 μL of His-tag wash buffer and 1 μL of RNase T1 (1,000 U/μL) was prepared on Dynabeads MyOne Streptavidin C1.
Next, this solution was incubated at 37°C for 15 minutes, and then vigorously stirred on a shaker for 10 minutes to separate the cDNA display from Dynabeads MyOne Streptavidin C1.

(4-10) His-tag purification At this stage, among the cDNA displays recovered in the previous procedure, those in which the peptide was translated were recovered. First, 20 μL of His-mag Sepharose Ni was taken and washed with His-tag washing buffer. The cDNA display obtained by the procedure described above was put into the washed His-mag Sepharose Ni, and incubated at 4°C for 2 hours or more.

After incubation, the supernatant was confirmed as Sup. by electrophoresis. Next, the mixed solution of cDNA display and His-mag Sepharose Ni was washed twice with His-tag washing buffer. After adding the His-tag elution buffer and stirring vigorously for 10 minutes or more with a shaker, the supernatant was collected.

Next, column purification was performed using Micro BioSpin 6 Column. First, to remove the preservation solution, the Micro BioSpin 6 Column was centrifuged at 1,000 xg for 2 minutes. Next, the column was washed 4 times with 1X selection buffer. Next, the His-tag-purified cDNA display was placed in the column and centrifuged at 1,000 g for 4 minutes to recover.

(5) Preparation of Carrier with Immobilized Midkine (5-1) Immobilization of Midkine Different beads were used in each round for immobilization of midkine. To prevent reaction inhibition during immobilization, midkine was used after buffer exchange after the second round. First, Amicon Ultra 0.5-MWCO=3K was washed with ultrapure water and a buffer according to each round. Next, midkine and buffer were added in appropriate amounts. Centrifugation was performed at 14,000 xg at 4°C for 30 minutes, and concentration and washing by buffer exchange were repeated 3 to 4 times. After washing, the column was installed upside down in a new tube and centrifuged at 1,000 xg for 5 minutes at 4°C.
To calculate the efficiency of buffer exchange at this time, 15% SDS-PAGE was performed.

(5-2) Immobilization on sepharose resin After immobilizing midkine on the resin, NHS-activated Sepharose 4 Fast Flow was used in the first round in order to destroy the unreacted groups of the resin. First, 100 μL of NHS-activated Sepharose 4 Fast Flow was put in an empty Illustra ^™ MicroSpin ^™ columns, and 200 μL of well-cooled 1 mM HCl was added and washed 6 times. Next, 200 μL of 20 mM Tris-HCl (pH 8.0) was added for washing. Subsequently, 50 μg of midkine was added and the volume was adjusted to 50 μL with 20 mM Tris-HCl (pH 8.0). Then, it was incubated overnight at 4°C.

After incubation, washed 6 times with 20 mM Tris-HCl (pH 8.0) and incubated overnight at 4°C. After incubation, equilibrated with 1x selection buffer and stored. The midkine before fixation and the flow-through after fixation were subjected to 15% SDS PAGE (20 mA, 2 hours) to confirm whether or not they were immobilized on Sepharose resin. The gel was stained with Coomassie Brilliant Blue (CBB). The results are shown in Fig. 9.

From Fig. 9, it was confirmed that no band was confirmed at the band position of the control midkine after F.T. and the midkine could be immobilized with sufficient efficiency.

(5-3) Immobilization on streptavidin magnetic beads Dynabeads MyOne Streptavidin C1 was used in the second and subsequent rounds. First, in order to biotinize midkine, a biotin buffer was added to 300 pmol of midkine and 450 pmol of EZ-link Sulfo-NHS-SS-Biotin to prepare a 60 μL reaction solution.

-This reaction solution was incubated on ice for 2 hours or more. During the incubation of this reaction solution, DynabeadsMyOne Streptavidin C1 was washed with biotinylated buffer. After incubation, the required amount of midkine was taken, mixed with washed DynabeadsMyOne Streptavidin C1 and stored at 4°C. The midkine before immobilization and the supernatant after immobilization were subjected to 15% SDS-PAGE (20 mA, 2 hours) to confirm whether or not they were immobilized on streptavidin magnetic beads. The gel was stained with SYPRO Ruby. The results are shown in Fig. 10.

From Fig. 10, it was confirmed that midkine could be immobilized with sufficient efficiency because no band was seen in the band position of midkine as a control after F.T.

(6) In vitro selection experiment for midkine using cross-linked peptide (selection)
Prior to the above selection, preselection was performed to remove cDNA display that nonspecifically binds to other than the target molecule. For preselection, the beads without midkine added, which had been treated by the procedure described above, were used. First, in the first round, NHS-activated Sepharose 4 Fast Flow in which midkine was not immobilized was mixed with cDNA display, and the cDNA display was eluted with 1x selection buffer. The eluted cDNA display was used for selection. From the second round onward, pre-selection was similarly performed for each fixed bead.

In this selection, midkine was bound to the cDNA display, and then the cDNA display was eluted. The first round was as follows. First, the preselected cDNA display was added to NHS-activated Sepharose 4 Fast Flow with immobilized midkine, and incubated at 4°C for 1 hour. Then, the cells were washed with 1x selection buffer and the washing solution was collected (hereinafter, this washing solution may be referred to as "flow-through" or "F.T."). Washing was further repeated several times by the same procedure, and the washing liquid for each time was collected (hereinafter, may be referred to as “wash 1 to 10” or “Wash 1-n”).

Next, elution operation was performed using 1% SDS containing dithiothreitol or glycine-hydrochloric acid buffer (pH 3.4), and the eluates were collected respectively. The amounts of cDNA display and midkine, incubation time and number of washes used in each round are shown in Table 14 below. In the PCR program, incubation was carried out at 98°C for 1 minute, then at 98°C for 10 seconds, followed by cooling to 68°C for 20 cycles of 25 cycles, incubation at 68°C for 1 minute, and then cooling to 4°C. As described above, a crosslinked peptide library (hereinafter sometimes referred to as “S-body library”) was obtained.

The state of the in-vitro selection experiment in each round was subjected to 8M urea denaturation 6% PAGE (200V, 25 minutes) and confirmed. The gel was stained with STBR Gold. From FIG. 11, it was confirmed that the band of the last wash lane and the band of the first elution lane were compared with each other, and the elution lane was darker regardless of crosslinking, as the rounds were repeated.

From this, it was considered that the progress of selection may not be changed by the disulfide bond or the chemical cross-linking agent, and the ratio of sequences after selection may be the same. Furthermore, it was considered that screening could be performed by expanding the structural diversity of molecules by further increasing the valency of the crosslinking agent or by crosslinking in a non-aqueous buffer such as an organic solvent.

(7) Preparation of library for the next round In the above (6), DNA containing no T7 promoter region required for transcription was obtained, but a new T7 region and promoter region were added to prepare a library for the next selection. Had to do. Therefore, the overlap extension method was adopted to restore the DNA to the full construct.

-Using the DNA obtained by the above procedure as a template, PCR was performed in a solution having the same composition as in Table 14 according to the same PCT program as in (6) above to obtain amplified DNA. 25 μL of each DNA amplified by PCR was subjected to 8M urea denaturing PAGE and confirmed.

(8) Results Library DNA prepared by amplification by PCR so as not to lose the diversity of the initial library and to prevent generation of subbands was subjected to 8 M urea denaturing 6% PAGE (200 V, 25 minutes). The results of staining the gel with SYBR Gold are shown in FIG.
From FIG. 12, it was confirmed that a band was present near the full length of the library of 265 bp. Since this band is library DNA, it was confirmed by PCR that the DNA could be amplified without problems. Further, using the DNA obtained by the above-mentioned procedure as a template, the DNA amplified by PCR was subjected to 8 M urea denaturing 6% gel electrophoresis (200 V, 25 minutes) in 25 μL aliquots. The gel was stained with SYBR Gold. The results are shown in Figure 12. From FIG. 12, an upshifted band near 265 bp was detected, and it was observed that the T7 region was ligated.

(Example 2) Development of screening system using high concentration organic solvent (1) Materials The cross-linking reaction contains 1 x TBMB reaction buffer (20 mM ammonium hydrogen carbonate, 5 mM EDTA and 0.05% Tween 20 (pH 8.0). )), 1,3,5-tris(bromomethyl)benzene (all manufactured by Tokyo Chemical Industry Co., Ltd.), and acetonitrile (HPLC grade, manufactured by Wako Pure Chemical Industries, Ltd.) were used. In addition, Streptavidin from Streptomyces avidinii (Sigma-Aldrich) was used for the upshift assay. The following clones obtained from the S-body library obtained as described above were used as controls and samples.

Clone No. 45 MGCRTDYRSFREYYRECWRSFRIHDWSDNF (SEQ ID NO: 10: positive control)
Clone No. 67 MGCRNSPNPRQHSYSYCPRDRGYHVFACYP (SEQ ID NO: 11: sample)

Also, cDNA display was obtained from the cloned DNA in the same manner as the above cDNA display preparation method, except that three types of clones were used instead of the peptide library and the crosslinking reaction conditions were excluded.

(2) Chemical cross-linking of cDNA display in high-concentration organic solvent and its efficiency Upshift assay was performed to calculate the efficiency of chemical cross-linking reaction by TBMB. The efficiency of chemical cross-linking reaction of S-body library by BMOE was also calculated by the same upshift assay.

(2-1) Chemical cross-linking reaction In order to prepare a clone cross-linked by a chemical cross-linking agent, a disulfide bond reduction reaction was performed. First, the biotin binding site on Dynabeads MyOne Streptavidin C1 was blocked with 1 mM biotin to stabilize the streptavidin immobilized on Dynabeads MyOne Streptavidin C1.

Next, a reaction buffer containing 10 mM TCEP dissolved therein was added, and the mixture was incubated at 42°C for 10 minutes. After that, the TBMB concentration, the reaction time, and the acetonitrile (hereinafter, sometimes abbreviated as “ACN”) concentration settings were changed, and the crosslinking reaction was performed to obtain the crosslinking efficiency. The crosslinking reaction of the S-body library by BMOE was performed under the same conditions as in Example 2 above.
The results are shown in Table 15 below. It was shown that higher chemical cross-linking efficiency is advantageous in maintaining library size. In particular, it was revealed that the library size was not impaired when it was about 80% or more.

　*表中、N.D.は、検出不可を表す。

*In the table, ND means not detectable.

(2-2) Upshift Assay for Calculation of Crosslinking Efficiency Each cDNA display was prepared under the chemical crosslinking conditions shown in Table 15 above. Then, the cross-linking efficiency was calculated for each of the combination of the clone and TBMB and the combination of the peptide library and BMOE.

　Crosslinked cDNA display is not biotinylated, so streptavidin does not bind, and uncrosslinked cDNA display is biotinylated, so streptavidin binds. Utilizing this, an upshift assay in which a thiol group which was not bound by a cross-linking agent was biotinylated and streptavidin was bound was performed to calculate the cross-linking efficiency. The upshift assay is shown schematically in FIG.

Figure 14 shows the results of gel electrophoresis (8M urea denaturation 6% SDSPAGE, 20mA, 2 hours) after cross-linking clones using TBMB at a final concentration of 10μM in 20% ACN. .. The gel was stained with FITC. Since the band intensity ratio of the uncrosslinked peptide and the crosslinked peptide was 8:2, the crosslinking efficiency was 20% when the reaction time was 1 hour.

Calculating in the same way when the reaction time was overnight, the crosslinking efficiency was 33%. Since the cross-linked peptide had a single band, it was considered that intermolecular cross-linking was present in a very small amount or could not be confirmed by SDS polyacrylamide gel electrophoresis. In any case, the intermolecularly crosslinked molecules were not judged to be a problem amount when introduced into the screening system.

Next, clones were cross-linked using a final concentration of 10 mM TBMB in 80% ACN and subjected to gel electrophoresis (8 M urea denaturation 6% SDS PAGE, 20 mA, 2 hours). The gel was stained with FITC. The results are shown in Figure 15. From FIG. 15, the crosslinking efficiency was calculated in the same manner as in 20% ACN. As a result of comparison between the case where the final concentration of TBMB is 10 μM and the case where the final concentration is 10 mM, it was revealed that the final concentration of TBMB is required to be 10 mM to improve the crosslinking efficiency.

On the other hand, when the reaction was performed overnight using TBMB 10 mM, the band in the lane disappeared. This was considered to be due to the denaturation of streptavidin on the streptavidin magnetic beads due to long-term exposure to high-concentration ACN. When the streptavidin on the beads is denatured and the biotin in the SBP linker is dissociated from the beads and dissociated from the beads, the cDNA display is peeled off from the magnetic beads, and the recovery efficiency of the cDNA display is reduced. If this happens, you will not be able to see the peptides on the cDNA display by SDS-PAGE. From this, it was considered that it is better to avoid exposing streptavidin on the magnetic beads to 80% acetonitrile for a long time, and the reaction time should be as short as possible.

Next, in order to investigate the shortest reaction time that maintains a crosslinking efficiency of 80%, we conducted a crosslinking experiment in 80% acetonitrile. An upshift assay using TBMB (final concentration 10 mM) in 80% ACN and a clone was performed, and subjected to 8M urea denaturing 6% SDS PAGE (20 mA, 2 hours). The gel was stained with FITC. The results are shown in Figure 16. When the band intensity ratio was calculated from FIG. 16, the cross-linking efficiency exceeded 80% at any reaction time, and was highest at a reaction time of 20 minutes. From the above, it was shown that the shortest reaction time that can maintain the crosslinking efficiency of 80% is 10 minutes.

(3) Crosslinking experiment using BMOE and S-body library Next, an upshift assay was performed using BMOE (final concentration 4 mM) in a phosphate buffer and S-body library, and gel electrophoresis (8 M urea denaturation 6% SDS-PAGE, 20 mA, 2 hours). The gel was stained with FITC. The results are shown in Figure 17.

From FIG. 17, it was confirmed that the cross-linking efficiency was 80% and that the cross-linking reaction could be performed without intermolecular cross-linking even in the peptide library. From these experiments, it is possible to chemically crosslink peptide clones or peptide libraries in 80% organic solvent or with a divalent or trivalent cross-linking agent, and when the content of organic solvent is high (high concentration organic solvent), cross-linking is performed. It was shown that the reaction time had to be shortened.
From the above, it was suggested that the cross-linking reaction or cross-linking agent in a high-concentration organic solvent may extend the structural diversity of peptide clones or peptide libraries.

(Example 3) Screening for IL6R using S-body library and BMOE In this experiment, an affinity screen for IL6R was performed using an S-body library cross-linked with BMOE in 50% DMF.

(1) Materials, etc. N,N-dimethylformamide (for molecular biology, manufactured by Wako Pure Chemical Industries, Ltd.) was used for the crosslinking reaction. Recombinant human IL6R protein (manufactured by ACRO Biosystems) was used for immobilization of IL6R. Preparation of cDNA display from the peptide library was performed in the same manner as in Example 2 above except that the crosslinking reaction conditions were different, the uncrosslinked cDNA display was not purified, and the reactive group of the crosslinking agent was blocked. ..

(2) Chemical cross-linking reaction In order to prepare a Cys3 library cross-linked by a chemical cross-linking agent, disulfide bond reduction was performed. First, the biotin binding site on Dynabeads MyOne Streptavidin C1 was blocked with 1 mM biotin to stabilize the streptavidin immobilized on Dynabeads MyOne Streptavidin C1. Next, a reaction buffer containing 10 mM TCEP dissolved therein was added, and the mixture was incubated at 42°C for 5 minutes. Thereafter, the supernatant was removed, a reaction solution having the composition shown in Table 16 below was added, and the mixture was incubated at 25°C for 15 minutes.

After the incubation, the supernatant was removed, 10 mM mercaptoethanol was added, and the mixture was incubated at 25°C for 15 minutes to block the maleimide group of the chemical crosslinking agent that did not bind to the thiol group.

(3) Preparation of carrier immobilized with IL6R IL6R used in each round was immobilized on sepharose resin in the first and fifth rounds and immobilized on strept-conjugated beads in the second to fourth rounds. used.

(3-1) Immobilization on Sepharose Resin NHS-activated Sepharose 4 Fast Flow was used in the first and fifth rounds. After fixing the IL6R, the unreacted groups of the resin were crushed. First, 200 μL of NHS-activated Sepharose 4 Fast Flow was put in an empty Illustra ^™ MicroSpin ^™ columns, and 3 mL of well cooled 1 mM HCl was added to wash.

Next, 200 μL of PBS-T (PBS containing 0.5% Tween 20) was added and washed. Next, IL6R was added and the volume was adjusted to 100 μL with PBS-T. Then, it was incubated overnight at 4°C. After the incubation, the plate was washed 6 times with 20 mM Tris-HCl (pH 8.0) and further incubated overnight at 4°C. After this incubation, the reactive groups were blocked with 0.5 M ethanolamine and incubated overnight at 4°C. After that, the supernatant was removed, and 1× selection buffer was added and stored. The IL6R before fixation and the flow-through after fixation were subjected to 10% SDS PAGE (20 mA, 2 hours) to confirm fixation on Sepharose resin. The gel was stained with SYPRO Ruby. The results are shown in Figure 18.
From FIG. 18, it was confirmed that IL6R could be immobilized on Sepharose resin with sufficient efficiency.

(3-2) Immobilization on streptavidin magnetic beads From the second round to the fourth round, Dynabeads MyOne Streptavidin C1 was used. First, in order to biotinylate IL6R, ypmol IL6R and 1.5 ypmol EZ-link Sulfo-NHS-SS-biotin were dissolved in a biotinylation buffer to prepare 60 μL of an immobilization solution.

This immobilization solution was incubated on ice for 2 hours or more, and Dynabeads MyOne Streptavidin C1 was washed with a biotinylation buffer during the incubation. After the incubation, a required amount of IL6R was taken, mixed with washed Dynabeads MyOne Streptavidin C1, and stored at 4°C. The IL6R before fixation, the supernatant after fixation, and the eluate of IL6R with 50 mM DTT were subjected to 10% SDS-PAGE (10% SDS-PAGE, 20 mA, 2 hours) to immobilize them on streptavidin magnetic beads. confirmed. The gel was stained with Lumitein. The results are shown in Figure 19.
From FIG. 19, it was confirmed that IL6R can be immobilized on streptavidin magnetic beads with sufficient efficiency, and IL6R can be eluted by DTT.

(4) In-vitro selection experiment The products obtained from the first round to the fourth round were subjected to affinity screening in the same manner as in Example 2 above. The product obtained in the 5th round was incubated to bind the IL6R to the cross-linked S-body library, followed by continued washing for 14 hours (Fig. 20). After completion of washing, 1% SDS was used to denature IL6R, and the cDNA display remaining on the column was recovered. The recovered solution was subjected to 6% PAGE (200 V, 25 minutes), and the progress of in-vitro selection in each round was confirmed. The result of staining the gel with SYBR Gold is shown in FIG.

As shown in FIG. 20, when the band intensities of the library at the 5th round were compared between cross-linking in 50% DMF and cross-linking in phosphate buffer, 9 bands were found in the library in 50% DMF. It was twice as high. This is considered to reflect the number of molecules of the peptide that binds to IL6R in the 5th round, suggesting that the structural diversity of the peptide may have been expanded by crosslinking in 50% DMF.

(5) Pull-down assay of peptides selected in vitro: RNA was removed by RNase H treatment in a cell-free translation system after translation, and reverse transcription was not performed after immobilization on streptavidin magnetic beads. Except for transfer, ligation, immobilization on Dynabeads MyOne C1 Streptavidin, and chemical cross-linking were performed in the same manner as in Example 2.
Peptide-linker conjugates for pulldown assays were prepared as described above and pulldown assays were performed for the presence or absence of interaction between the library peptides obtained above and the sCD40 ligand. The reaction scheme is shown schematically in FIG.

50 μL of 0.62 μM IL-6R (31 μM) and 1 μL of 3.1 mM NHS-fluorescein were mixed (51 μL in total), and this reaction solution was incubated at room temperature for 60 minutes, where 1 μL of 5 mM ethanolamine was added. In addition, it was incubated at room temperature for 30 minutes. After that, 0.6 μM fluorescein-labeled IL-6R was added, and this solution was analyzed by gel electrophoresis (8M urea-denatured SDS-PAGE (20mA, 2 hours). The results of detecting the FITC label are shown in FIG.

From FIG. 22, the band of the peptide group before and after selection under each condition was confirmed as the lower band, and IL6R bound to the peptide contained in the peptide group was confirmed as the upper band. In order to compare the amount of IL6R bound to the peptide group before and after selection under each cross-linking condition, the band intensity ratio of IL6R in each lane was measured. FIG. 23 shows a graph of the rate of increase obtained by the following equation (1).

Increasing rate = (band intensity ratio of library after selection-band intensity ratio of library before selection)/band intensity ratio of library before selection…(1)

As shown in FIG. 23, when the band intensities of the libraries under each cross-linking condition were compared before and after selection, it was shown that the selection of molecules that bind to the target molecule is advanced although the sequence convergence is not sufficient. The initial library increased by 45% compared to the pre-selection library. From this, it was shown that the IL6R binding sequence contained in the library was enriched by 1.5 times by the in vitro selection experiment under the condition of crosslinking in the presence of 50% DMF.
In contrast, a library cross-linked without DMF was similarly compared with -30%, indicating that cross-linking without DMF resulted in a lower percentage of cross-linked IL6R binding sequences than in the initial library.

Based on the above results, by conducting cross-linking in the presence of 50% DMF and conducting a selection experiment in an aqueous buffer solution, the structural diversity of the library was expanded compared with the conventional peptide library cross-linked in the aqueous buffer solution. It has been shown. In addition, it was expected that the crosslinked three-dimensional structure in 50% DMF would have a different three-dimensional structure from that in the conventional aqueous buffer solution.

(Example 4) Analysis of peptide cross-linking under different solvent conditions (1) Preparation of sample for studying peptide cross-linking (1-1) Preparation of conotoxin-based peptide fragment Comparison among known sequences of natural conotoxins The following two amino acid sequences (SEQ ID NOS: 12 and 13 in the sequence listing) were selected as those having a relatively short length and containing four cysteines (see FIG. 24). In these peptides, C1 and C3 and C2 and C4 (the numbering of cysteine is the one that appears first from the N-terminus as 1 and the order is 2, 3 and 4 below). Disulfide bridge.
(SEQ ID NO: 12 in the Sequence Listing) MCPPLCKPSCTNC
(Sequence ID No. 13 in the sequence listing) DCPPHPVPGMHKCVCLKTC

In order to fragment it with protease after cross-linking and make it a sequence that can be analyzed by TOF-MS, we designed the following three types of peptides incorporating lysine (K, a substrate for protease) in the above sequence and synthesized them at Toray Research Center. Was requested (see Figure 24).
(SEQ ID NO: 14 of Sequence Listing) MCPKLCKPSCKNC
(SEQ ID No. 15 in the sequence listing) DCPPHPVPGMHKCKCLKTC
(Sequence ID No. 16 in the sequence listing) DCPPHKPVCPGMHKCLKTC
(2) Construction of S-body library preparation construct Further, as a S-body library preparation construct, T7Ω-Library (MGC(28AA))-HisX6-C(Ytag) having the following sequence was prepared. In the sequences below, X represents any nucleotide. As the puromycin linker, cnvK linker having the structure shown in Table 8 below and FIG. 25 was used. Further, the nucleotide sequence of the construct for S-body library in which the nucleotide sequence of the library shown in Table 8 below is represented by N is shown as SEQ ID NO: 9 below. Here, N represents any nucleotide of A, T, G, and C.

(SEQ ID NO: 9 in the sequence listing)
5'-GATCCCGCGAAATTAATACGACTCACTATAGGGGAAGTATTTTTACAACAATTACCAACAACAACAACAAACAACAACAACATTACATTTTACATTCTACAACTACAAGCCACCNNNNNNNNNNNNNNNNNNGANNCANNCANNCANNACGGAGCAGTAATTAATCAACAATTACCAACAACAACAACAAACAACAACAACATTACATTTTACATTCTACAACTACAAGCCACCNNNNNNGGNNGCNNNNNNGGNAGANCGANNCATCATGACGA

Using the above S-body library construct, selection was performed in vitro under the same conditions as above, and a clone A having the following sequence was obtained.

(SEQ ID NO:17 in the sequence listing)
5'-GATCCCGCGAAATTAATACGACTCACTATAGGGGAAGTATTTTTACAACAATTACCAACAACAACAACAAACAACAACAACATTACATTTTACATTCTACAACTACAAGCCACCATGGGCTGCAGCTATTATCATGGCCGTCACGGTGGCGCCGCATCATCATCATCATCATCGGTCATACTGCCACCATAGTCACTGCCCCTCGGGCATGGCATCAACACACTAGTAGACTACTCCCCCTCATGGGAGCGGCAGCACCATTAGATACTGCCCCTCATGGGAGCGGCAGCAG

(Example 5) Preparation of C-random-C library (1) Preparation of cDNA display from peptide library (1-1) Amplification by PCR 1
DNA synthesis of the disulfide-rich peptide library was commissioned to Tsukuba Oligo. PCR1 (Extension PCR 1) was performed using the solutions shown in Table 18 below so that the diversity of the initial library was not lost and subbands did not appear, and the library was purified.

The PCR1 program was as follows: (t1-1) 98°C for 1 minute, (t1-2) 98°C for 10 seconds, (t1-3) followed by incubation at 63°C for 5 seconds, (t1-4) After incubating at 72° C. for 10 seconds, (t1-1) to (t1-4) were carried out for 5, 7, or 9 cycles, incubated at 72° C. for 2 minutes, and then cooled to 4° C. The obtained PCR product was confirmed by 8M urea-denatured nucleic acid polyacrylamide gel electrophoresis (electrophoresis performed on nucleic acids such as DNA and RNA; hereinafter also referred to as “PAGE”).

After that, the primer and enzyme were removed using PCR-Clean-up Mini Kit to obtain Extension product 1. Figure 26 shows the results of gel electrophoresis (8M urea 6% PAGE). In this example, the 6% gel for gel electrophoresis had the composition shown in Table 19 below. From FIG. 26, it was confirmed that the desired PCR product 1 was obtained using the C-random-C library and His(Ytag-cnvK). Also, the amount of PCR product 1 did not increase significantly depending on the number of cycles.

(1-2) Amplification 2 by PCR
The Extension products 1 obtained in (4-1) above (collected from the products obtained in 5 to 9 cycles of PCR product 1) were used as a template for the following PCR, and the solutions shown in Table 19 below were used. PCR2 (Extension PCR 2) was carried out.

The PCR2 program consists of (t2-1) incubation at 98°C for 1 minute, (t2-2) incubation at 98°C for 10 seconds, (t2-3) followed by incubation at 58°C for 5 seconds, (t2-4). After incubating at 72°C for 15 seconds, (t2-1) to (t2-4) were carried out for 7, 10 or 13 cycles, incubated at 72°C for 2 minutes, and then cooled to 4°C. The obtained PCR product 2 (Extension product 2) was confirmed by the above-mentioned 8M urea-denatured nucleic acid PAGE. After that, the primer and the enzyme were removed using PCR-Clean-up Mini Kit to obtain Extension product 2. The results of gel electrophoresis of these are shown in FIGS. 27 and 28.

From FIG. 27, it was confirmed that Extension product 1 used as a point template for PCR2 was amplified by T7Ω, and the desired PCR product 2 (235 bp) was obtained. Also, the amount of PCR product 2 did not increase significantly depending on the number of cycles.

(1-3) Amplification by PCR 3
The Extension products 2 obtained in the above (4-2) (collecting products obtained by 7 to 13 cycles of PCR products 2) were used as a template for the following PCR, and the composition shown in Table 20 below was used. PCR3 was performed using the solution.

The PCR3 program consists of (t3-1) incubation at 98°C for 1 minute, (t3-2) incubation at 98°C for 10 seconds, (t3-3) followed by incubation at 67°C for 5 seconds, (t3-4). After incubating at 72°C for 15 seconds, 25 cycles of (t3-1) to (t3-4) were performed, and after incubating at 72°C for 2 minutes, the temperature was lowered to 4°C. The obtained PCR product 3 was confirmed by the above-mentioned 8 M urea-denatured nucleic acid PAGE. After that, the primer and the enzyme were removed using PCR-Clean-up Mini Kit to obtain PCR product 3 (235 bp). The purified product (25 ng/μL) obtained by column-purifying the 10-fold to 1000-fold diluted PCR products (4 to 6 in FIG. 28) was used for transcription. The resulting PCR3 product was subjected to 8 M urea-denatured nucleic acid PAGE for final confirmation (see FIG. 28).
A sequence analysis of this library was requested to Eurofin Genomics, Inc., and it was confirmed that the libraries shown in Table 21 below (C-random-C library) were synthesized. This library was as designed.

The nucleotide sequence of the C-random-C library preparation construct (SEQ ID NO: 18 in the sequence listing) is shown below.
(SEQ ID NO:18 in the sequence listing)
5'－GATCCCGCGAAATTAATACGACTCACTATAGGGGAAGTATTTTTACAACAATTACCAACAACAACAACAAACAACAACAACATTACATTTTACATTCTACAACTACAAGCCACCATGGGCTGCXYZXYZXYZXYZXYZXYZXYZGGAZCATZCATGAGCAACATZ

In the above C-random-C library preparation construct, (a) amino acid is biased as little as possible, (b) the frequency of stop codons is suppressed, (c) about 2 residues in 16 residues in the random region The nucleotide sequence of the random region was adjusted so as to satisfy the requirement that the cysteine of γ. Here, XYZ in the above SEQ ID NO: 18 means that any nucleotide of A, T, G, and C corresponding to the start codon to the random region in Table 21 above appears at the frequency shown in Table 22 below. Indicates.

When the frequency of appearance of cysteine in the random region is 12.4%, 0.124×16=1.984, and stochastically about 2 cysteines appear in the random region. Then, when translated as described above, about 4 residues of cysteine came to appear in one molecule.
As a result of preparing the nucleotide sequence of the random region as described above, the appearance rate of amino acids in the random region is 30.5% for hydrophobic amino acids, 35.4% for polar (uncharged) amino acids, and 10.5% for hydrophilic (basic) amino acids. , Hydrophilic (acidic) amino acids became 5.1%. The appearance rate of each amino acid is shown in FIG. 29 together with a schematic diagram of the peptide of the present invention.

(1-4) Transcription Then, using RiboMAX-Large Scale RNA production systems, the DNA (PCR product 3 (C-random-C library)) obtained in (4-3) above was used as a template, Was transcribed to obtain a transcript (RNA). The reaction composition and reaction conditions for one tube are shown in Table 23 below.

Transferring reaction was performed by incubating at 37℃ for 3 hours in the reaction solution with the above composition. After that, RQ1 RNase-Free DNase was added at 2.5 μL and further incubated at 37°C for 15 minutes. Immediately after 15 minutes, the transcript was purified using Agencourt RNA Cleane XP (Beckman Coulter, Inc.) according to the attached protocol. At this stage, the transcript and the C-random-C library were confirmed by the above 8M urea denaturing PAGE. Results are shown in FIG. By this amplification, the target product, 205 bp mRNA, was obtained. The stock solution of transcript (No. 2 in FIG. 30) was purified and used as a template for ligation below.

(1-5) Ligation Next, the purified mRNA obtained in (1-4) above was ligated to the cnvK linker. The composition of the ligation reaction solution for one tube is shown in Table 24 below.

Regarding the annealing conditions, after incubating at 90°C for 2 minutes, the temperature was lowered to 70°C at 0.1°C/minute. After incubating at 70°C for 1 minute, the temperature was lowered to 25°C at 0.01°C/minute, and the mixture was incubated at 25°C for 30 seconds. After annealing, using a CL-1000 Ultraviolet Crosslinker, ultraviolet light of 365 nm was irradiated for 405 seconds (405 mJ/cm ² ), and the above cnvK linker and the transcription product were photocrosslinked to obtain a ligation product. Then, the obtained ligation product and uncrosslinked mRNA (control) were subjected to the above 8 M urea denaturing PAGE (200 V, 25 minutes).

After gel electrophoresis, the gel was stained with FITC or SYBR Gold. The results are shown in Fig. 31. From FIG. 31, it was confirmed that the cnvK linker and the above-mentioned mRNA were crosslinked by shifting the position of the detected band to a higher molecular side than the band of the mRNA alone.

(1-6) Cell-Free Translation Next, the mRNA-linker conjugate was translated in a cell-free translation system in a solution for cell-free translation having the composition shown in Table 25 below to give the mRNA-protein (polypeptide) conjugate. Formed.

25 μL of the reaction solution having the composition shown in Table 26 was dispensed into each tube and then incubated at 30° C. for 40 minutes. Then, 24 μL of 3 M KCl and 6 μL of 1 M MgCl ₂ were added, and the mixture was further incubated at 37° C. for 80 minutes. After this, as shown in Table 26 above, cell-free translation was performed from the amount of mRNA-cnvK linker at the start point (6 pmol) to examine the formation efficiency and cross-linking efficiency of the finally obtained cDNA display. ..

(1-7) Purification of mRNA-Protein (Peptide) Conjugate (In Vitro Virus: IVV) Subsequently, RNase was removed from streptavidin (SA) magnetic beads (Dynabeads MyOne Streptavidin C1). 60 μL of Dynabeads MyOne Streptavidin C1 was placed in a 1.5 mL tube and left on a magnetic stand. All the subsequent operations for removing the supernatant were performed on a magnetic stand. The supernatant was discarded, resuspended in 1.5x binding buffer and the supernatant removed again.

Next, because the residual ribosome may hinder stericization during immobilization of magnetic beads, the ribosome bound to the post-translational mRNA-peptide conjugate was removed as follows. First, 8 μL of 0.5 M EDTA (final concentration 70 mM) was added as a chelating agent. To this, an equal amount of 2x binding buffer was added, mixed with washed Dynabeads MyOne Streptavidin C1 and incubated at room temperature for 25 minutes to remove the supernatant. This supernatant was subjected to 8M urea denaturing PAGE (20mA, 2 hours) together with the ligation product as Sup.1 to confirm the content. The gel was stained with FITC. The results are shown in Figure 32. From FIG. 32, it was confirmed that the ligation product was translated, immobilized on the above magnetic beads, and purified by His-tag.

(1-9) Reverse transcription Next, unstable mRNA was reverse transcribed into cDNA. First, Dynabeads MyOne Streptavidin C1 used for the purification of IVV was washed with 1 x ReverTraAce buffer. Then, a solution having the composition shown in Table 26 below was prepared.

Next, Dynabeads MyOne Streptavidin C1 was added to this solution and mixed, incubated at 42°C for 45 minutes, and after the incubation was completed, the supernatant was removed and cDNA was obtained from the cDNA display library. The solution containing the obtained cDNA peptide was subjected to 8M urea denaturing PAGE (20mA, 2 hours) to confirm the contents. The gel was stained with FITC. The results are shown in Figure 33.

(1-9) RNase treatment Next, the cDNA display was cut off from Dynabeads MyOne Streptavidin C1. First, the reverse transcription buffer was removed, washed with His-tag washing buffer, and the supernatant was discarded. An RNase-treated solution was prepared on Dynabeads MyOne Streptavidin C1 by adding 19.8 μL of His-tag washing buffer and 20 μL of RNase T1 (1,000 U/μL)-containing buffer for RNase treatment to each tube.
Next, this RNase-treated solution was incubated at 37° C. for 20 minutes, and then the tube was set on a magnetic stand to collect the solution containing the cDNA display.

(1-10) Purification of His-tag Among the cDNA displays recovered in the above procedure, those in which the peptide was translated were recovered using His-tag. First, 20 μL of His-mag Sepharose Ni was placed in a tube and washed with a His-tag binding buffer. The above cDNA display was added to washed His-mag Sepharose Ni and incubated at 4° C. for 1 hour or more.

After the incubation, the supernatant was confirmed as sup. by PAGE. Next, the His-tag washing buffer was added to the tube containing the mixed solution of cDNA display and His-mag Sepharose Ni, and the tube was washed twice. After adding the His-tag elution buffer and stirring for 20 minutes with a shaker, the supernatant was collected. As described above, a purified cDNA display constituting the C-random-C library was obtained. The results are shown in Figure 33. FIG. 33(A) shows the results obtained using the clones obtained as described above, and FIG. 33(B) shows the results obtained using the non-cloned library.
It was speculated that the fact that the clone was obtained in the presence of 50% DMF might have affected the purification. As shown in FIG. 34, there was a difference in crosslinking depending on the crosslinking agent used.

(Example 5)
(1) Examination of organic solvent resistance of streptavidin magnetic beads First, the organic solvent resistance of streptavidin magnetic beads was examined. As the organic solvent, it was decided to use a polar solvent while considering the effect on the peptide. Dimethylsulfoxide (hereinafter sometimes referred to as "DMF"), which is an aprotic polar solvent, has a dielectric constant of 36.7 and boiling point of 153°C, and acetonitrile (dielectric constant of 37.5, boiling point of 82°C), and ethanol which is a protic polar solvent (dielectric A rate of 25.8 and a boiling point of 78°C) was selected. The concentration of each organic solvent was set to 50%, and the change in fluorescence intensity of fluorescent biotin (free biotin) released from streptavidin magnetic beads was examined with the reaction time.

The results are shown in Figure 35. As shown in FIG. 35, the fluorescence intensity of the above-mentioned free biotin greatly increased with time in 50% acetonitrile, and it was determined that the solvent resistance of the streptavidin magnetic beads had a problem depending on the solvent. A similar trend was observed with ethanol, although the rise was small. On the other hand, with DNF, no significant change was observed in the fluorescence intensity of free biotin. Therefore, DMF was selected as the solvent for studying the crosslinked structure in view of the solvent resistance of the magnetic beads. The reaction time was set to 30 minutes from the result of FIG.

(3) Examination of changes in cross-linking structure in organic solvent (3-1) Examination of changes in cross-linking structure in DMF The peptides shown in SEQ ID NOs: 14 to 16 were respectively added to 0.1% acetic acid aqueous solution and 100% DMF. Was dissolved at a concentration of 1 mg/mL and incubated in a solvent having the composition shown in Table 27 below at room temperature for 1 hour to crosslink. As a cross-linking agent, bismaleimideethane (BOME) was dissolved in DMF and used. Further, tris(2-carboxyethyl)phosphine hydrochloric acid (TCEP-HCl) was dissolved in ultrapure water and used as a reducing agent for cleaving the disulfide bond in the protein or between the proteins.

After the incubation for 1 hour, the sample reacted in a 90% DMF solution (hereinafter referred to as "DMF90 sample") was subjected to vacuum centrifugation (2 hours at 40°C) to remove the solvent, and 100 μL 1x conjugation buffer was added. The above-mentioned vacuum centrifugation was not performed for the sample reacted in the 4% DMF solution (hereinafter referred to as "DMF04 sample").
Then, each sample was diluted 5-fold with 0.1% TFA aqueous solution, 100 μL was taken from the diluted solution and injected into HPLC for analysis. As a control, a sample containing DMF90 alone, DMF4%+crosslinking agent, peptide alone, or neither crosslinking agent nor peptide was used.

For HPLC, 1260 Hip ALS Agilent 1260, Degasser Agilent 1200, Quat Pump Agilent 1200, DAD Agilent 1200 and TCC Agilent 1200 (all from Agilent Technology) and Shiseido C18 (4.6 mm id × 250 mm) as analytical column (Shiseido) or Inert sustain c18 (GL Science) was used.
The eluents were trifluoroacetic acid (hereinafter sometimes abbreviated as "TFA"), acetonitrile and methanol (both manufactured by Fuji Film Wako Pure Chemical Industries, Ltd.) in the proportions shown in Table 28 below. The gradient elution was performed. The grades of these reagents were for HPLC except TFA. The stop time was fixed at 36 minutes and the eluent flow rate was fixed at 1 mL/min. The measurement was performed at λ=220 nm.

　上記表中、A液は0.1%TFA、B液は100%アセトニトリルである。

In the above table, solution A is 0.1% TFA and solution B is 100% acetonitrile.

The results of HPLC analysis of cross-linked peptide 1 (SEQ ID NO: 14 in the sequence listing) are shown in FIGS. 36(A) and 36(B). FIG. 36(A) shows the results when crosslinked in 4% DMF (including the conjugation buffer), and the same (B) shows the results when crosslinked in 90% DMF. In addition, the results of HPLC analysis of cross-linked peptide 2 (SEQ ID NO: 15 in the sequence listing) are shown in FIGS. 37(A) and 37(B). The solvent used in FIGS. 37(A) and 37(B) is the same as that in FIG. 37. Further, the results of HPLC analysis of cross-linked peptide 3 (SEQ ID NO: 16 in the sequence listing) are shown in FIGS. 38(A) and 38(B). The solvent used in FIGS. 38(A) and (B) is the same as that in FIG.

In the chromatograms shown in FIGS. 36 to 38, the very large peak observed at a retention time of 17 minutes was BMOE. In each of the above-mentioned cross-linked peptides 1 to 3, the peak of the non-cross-linked peptide decreased after the cross-linking, and a peak was observed at a position that could not be confirmed before the cross-linking. From the retention time, the peak detected after cross-linking was considered to be a cross-linking peptide (reaction product of cross-linking agent and peptide).
However, the difference between the peaks detected in the chromatogram of the sample containing the crosslinked product in the conjugation buffer (HEPES buffer, 4% DMF) and the sample containing the crosslinked product in 90% DMF can be confirmed in the above-mentioned crosslinked peptides 1 to 3. There wasn't. From this, it was not possible to confirm whether the position for cross-linking was changed.

(3-2) Examination of Change in Crosslinking Structure in Acetonitrile The organic solvent was acetonitrile (hereinafter sometimes abbreviated as “ACN”), and the crosslinker was α,α′-dibromo-o-xylene (hereinafter, “ACN”). , Sometimes abbreviated as “DBX”), and prepared different solutions (see Table 29 below), and incubated as in (3-1) above.

For the sample crosslinked in 90% CAN (ACN90 sample), the solvent was removed by vacuum centrifugation at room temperature for 20 minutes, and 150 μL of 20 mM NH4HCO ₃ buffer (pH 8.2) was added. From this solution, 100 μL was taken and injected into HPLC for analysis. The HPLC conditions were the same as those in (3-1) above, except that the eluent was a gradient shown in Table 30 below and the stop time was 63 minutes.

The results are shown in Figures 39-42. As a result of cross-linking the above-mentioned cross-linked peptide 1 using acetonitrile and DBX, a single peak was detected after a retention time of 21 minutes in the sample containing only cross-linked peptide 1. On the other hand, when cross-linked in a solvent containing 10% ACN, a sharp single peak was detected after 15 minutes and a very large peak was detected after 20 minutes. Furthermore, a broad low peak was detected at 25 to 27 minutes. When crosslinked in 90% ACN solvent, the peaks at retention times after 15 minutes and after 20 minutes decreased, suggesting that DBX was used for crosslinking. In addition, four peaks appeared at 23 to 27 minutes, and it was suggested that the peak was high and sharp after 25 to 27 minutes, and the peptide was cross-linked.

As a result of cross-linking the above-mentioned cross-linked peptide 2 under the same conditions as the above-mentioned cross-linked peptide 1, a small single peak was detected just before the retention time of 23 minutes in the sample containing only the above-mentioned cross-linked peptide 2. On the other hand, when cross-linked in a solvent containing 10% ACN, a sharp single peak was detected after 15 minutes and a very large peak was detected after 21 minutes. Furthermore, two small peaks were detected at 25 to 26 minutes. When crosslinked in 90% ACN solvent, the peaks at retention times after 15 minutes and after 21 minutes decreased, suggesting that DBX was used for crosslinking. In addition, the peak that appeared at the same 25-26 minutes was higher than that when cross-linked in a solvent containing 10% ACN. A peak was detected after 27 minutes, which had not been detected until now.

As a result of cross-linking the above-mentioned cross-linked peptide 3 under the same conditions as the above-mentioned cross-linked peptide 1, a single peak was detected just before the retention time of 23 minutes in the sample containing only the above-mentioned cross-linked peptide 3. On the other hand, when cross-linked in a solvent containing 10% ACN, a sharp single peak was detected after 15 minutes and a very large peak was detected after 20 minutes. Furthermore, two peaks were detected at 25 minutes to 26 minutes. When crosslinked in 90% ACN solvent, the peaks at retention times after 15 minutes and after 20 minutes decreased, suggesting that DBX was used for crosslinking. In addition, the peak that appeared at the same 25-26 minutes was higher than that when cross-linked in a solvent containing 10% ACN. In this sample as well, a peak appeared 27 minutes after the detection, which had not been detected so far, and thus this peak was estimated to be derived from DBX.

(3-3) Analysis of peptide cross-linking structure using TOF/MS In the above (3-1), peptides 1 to 3 cross-linked with BMOE were digested with trypsin/Lys-C Mix. , MALDI/TOF-MS and analyzed. Combinations in which peaks were considered to appear in each peptide were listed up, and the obtained mass spectrum was collated to confirm whether fragments were obtained. Only the fragments confirmed by mass spectrum are shown in Table 31. The mass spectrum is shown in Figures 42-44.

In Table 25 above, the numbers with [] indicate the values confirmed in the mass spectrum of the sample that was crosslinked in 90% DMF. In addition, the numerical value with () indicates that confirmed in the mass spectrum of the sample crosslinked in the conjugation buffer.

As a result of collation, in the above-mentioned cross-linked peptides 1 to 3, some of the smallest fragments that the cross-linking agent reacted and the smallest fragments that were not cross-linked were confirmed. However, almost no peak derived from intramolecular crosslinking could be confirmed. In addition, 44 peaks at equal intervals were observed around 1,000 to 1,600 [m/z], which were considered to be the peaks of the surfactant (derived from Tween 20) contained in the buffer. From the above, a peptide that crosslinks in an organic solvent was obtained.

The present invention is useful in the technical fields of protein engineering, pharmacy, medicine and diagnostics.

SEQ ID NO: 1 amino acid sequence of cross-linked polypeptide (1)
SEQ ID NO: 2: Amino acid sequence of cross-linked polypeptide (2)
SEQ ID NO: 3: Amino acid sequence of cross-linked polypeptide (3)
SEQ ID NO: 4: Base sequence of T7 promoter SEQ ID NO: 5: Base sequence of Ω sequence SEQ ID NO: 6: Base sequence of GGGS sequence SEQ ID NO: 7: Base sequence of inosine primer SEQ ID NO: 8: Base sequence of primer (NewYtag) SEQ ID NO: 9 : N-base sequence of construct for preparing S-body SEQ ID NO: 10: Amino acid sequence of positive control (clone 45) SEQ ID NO: 11: Amino acid sequence of sample (clone 67) SEQ ID NO: 12: Nucleotide sequence 1 of natural conotoxin
SEQ ID NO: 13: Base sequence 2 of natural conotoxin
SEQ ID NO: 14: Base sequence 1 in which a base for protease digestion is incorporated into the above natural conotoxin
SEQ ID NO: 15: Base sequence 2 in which a base for protease digestion is incorporated into the above natural conotoxin
SEQ ID NO: 16: Base sequence 3 in which a base for protease digestion is incorporated into the above natural conotoxin
SEQ ID NO: 17: nucleotide sequence of clone No. 25 obtained from S-body library SEQ ID NO: 18: nucleotide sequence of C-random-C library preparation construct

Claims (12)

A DNA library preparation step for preparing a disulfide-rich peptide library;
A transcription step of preparing and purifying mRNA from the DNA library;
A binding step of binding the purified mRNA obtained in the transcription step and a linker to form an mRNA-linker conjugate;
A translation step in which the mRNA-linker conjugate is translated in a cell-free translation system to form an mRNA-peptide conjugate;
A reverse transcription step of preparing a cDNA from the mRNA in the mRNA-peptide conjugate and forming an mRNA-peptide-cDNA conjugate that binds to the mRNA-peptide conjugate.
A chemical cross-linking library preparing step of chemically cross-linking the mRNA-peptide-cDNA conjugate to prepare a chemical cross-linking library;
A target polypeptide immobilization step of immobilizing the target polypeptide on a resin;
An in vitro selection step of performing in vitro selection using the chemical cross-linking library and the target polypeptide immobilized on a resin to obtain a polypeptide having the desired sequence;
A cross-linking step of cross-linking the polypeptide having the desired sequence to form a cross-linked polypeptide having a specific structure using a disulfide bond or a cross-linking agent in a solution containing an organic solvent;
A method for preparing a cross-linked polypeptide capable of binding to a target polypeptide, comprising: The method for preparing a crosslinked protein according to claim 1, wherein in the binding step, the purified mRNA and the linker are bound by an enzyme or photocrosslinking. The purified mRNA is bound to the linker by ligase when a cyanovinylcarbazole-free linker is used, and by ultraviolet irradiation when a cyanovinylcarbazole-containing linker is used, Item 3. A method for preparing a crosslinked protein according to Item 1 or 2. The method for preparing a crosslinked protein according to claim 1, wherein the ribosome bound to the mRNA-peptide conjugate after translation is removed in the translation step. The method for preparing a crosslinked polypeptide according to claim 4, wherein a chelating agent is used for removing the ribosome. The method for preparing a crosslinked polypeptide according to claim 5, wherein the chelating agent is ethylenediaminetetraacetic acid or glycoldiaminetetraacetic acid. The solution containing the organic solvent contains 10 to 90% by volume of the organic solvent, and the organic solvent is any solvent selected from the group consisting of dimethylformamide, acetonitrile, and ethanol. 1. The method for preparing the crosslinked polypeptide according to 1. The method for preparing a crosslinked polypeptide according to claim 1, wherein the crosslinked polypeptide having the desired sequence contains three or more cysteines and can bind to the target polypeptide. 2. The crosslinked poly according to claim 1, wherein the cross-linking agent is any one selected from the group consisting of bismaleimideethane, bismaleimide propane, and α,α′-dibromo-o-xylene. Methods for preparing peptides. A cross-linked polypeptide capable of binding to a target polypeptide, which is prepared by the method according to any one of claims 1 to 9. The crosslinked polypeptide according to claim 10, wherein the crosslinked polypeptide is selected from the group consisting of peptides containing 3 or more cysteines. 11. The cross-linked polypeptide according to claim 10, wherein the cross-linked polypeptide has any one of sequences 1 to 3 in the sequence listing below.
[SEQ ID NO: 1]
MGCSYYHGRHGDPPVHCHHSHCPYGRYH
[SEQ ID NO: 2]
MGCHSCNPSNSTDVWSNAYYPLSHSSCRNP

Images