WO2019198316A1 - Method for improving pharmacokinetics of middle molecule having biological activity and method for producing middle molecule library using improved pharmacokinetics - Google Patents

Abstract

[Problem] To provide a method for imparting membrane permeability to a middle molecular compound such that binding with an intracellular target becomes possible. [Solution] A peptide sequence that binds with a protein is fused with a middle molecule having a molecular weight of 500 to 5,000, such protein being present in a target cell and having activity to pull in middle molecules from a lipid bilayer.

Description

A method for improving the pharmacokinetics of medium molecules with physiological activity, and a method for producing a medium molecule library using improved pharmacokinetics

The present invention relates to a method of permeabilizing medium molecules having a molecular weight of 500 to 5000, and more specifically, binds to the medium molecule with a protein that exists in the target cell and has an activity of pulling out the medium molecule from the lipid bilayer membrane. The present invention relates to a method for permeating the medium molecule by fusing a transmembrane facilitating peptide sequence, and a cyclic peptide library having acquired membrane permeation ability by the method.

Organic compounds having a molecular weight of up to about 500 have been developed as many pharmaceuticals so far because they are low in production cost, can be administered orally, and have little immunotoxicity. However, this low molecular drug has low specificity and has become increasingly difficult in recent years due to the problem of side effects caused by off-targets. In contrast, antibody and protein-derived high molecular weight drugs with molecular weights exceeding 10,000 are increasing in sales due to their high specificity and few side effects, but they are expensive, cannot be administered orally, and exhibit immunotoxicity. There is a problem that there is a possibility.

So-called medium molecular weight drugs such as peptides, cyclic peptides, or natural products having a molecular weight of about 500 to 5000 are inexpensive and have low immunotoxicity, and have high specificity for target molecules and few side effects. It has the advantages of high-molecular-weight pharmaceuticals and is expected to have a wide range of applications. However, when medium molecular drugs target intracellular molecules, a method for permeating the cell membrane to reach the desired intracellular target molecule has not been established. Moreover, in order to show high bioavailability by oral administration, it is also necessary to improve stability such as protease resistance.

Cyclosporin A (CsA), which is one of such medium molecular drugs, is an immunosuppressant that targets intracellular calcineurin, such as interleukin-2 (IL-2) and interferon-γ by T lymphocytes. Specifically, reversibly suppresses the production and release of cytokines. This action dephosphorylates activated T-cell nuclear factor (NFAT) that controls IL-2 production and inhibits intracellular signal transduction by calcineurin, an intracellular calcium-dependent dephosphorylation enzyme that activates T cells. I know that. As shown in FIG. 1, CsA is an 11-residue cyclic peptide containing a non-protein special amino acid, and 7 of amino groups forming a peptide bond are N-methylated. This has acquired cell membrane permeability and resistance to proteases.

Until now, many peptide sequences that are known to have membrane permeation activity are characterized by being rich in linear basic residues, and this sequence is added to peptides with the desired physiological activity. (See, for example, Patent Documents 1 to 5). However, such a membrane-penetrating peptide does not impart membrane-permeating activity to any physiologically active substance, and its membrane-penetrating mechanism is not clear.

Also, attempts have been made to improve membrane permeability of cyclic peptides by imparting a specific structure to the middle molecule. For example, the invention described in Patent Document 6 improves the membrane permeability and / or metabolic stability of the cyclic peptide by introducing a cyclopropane amino acid unit having a specific structure into the ring of the cyclic peptide. Furthermore, in the invention described in Patent Document 7, at least two amino acids which are not adjacent to each other in the cyclic peptide have hydrophobic side chains, and in a hydrophilic environment, these hydrophobic side chains are located inside the ring of the cyclic peptide. By interacting, a peptide that can penetrate the cell membrane and reach the inside of the cell is obtained.

WO2011 / 126010 WO2016 / 136708 WO2016 / 136707 WO2013 / 061818 WO2010 / 134537 Japanese Unexamined Patent Publication No. 2017-43615 JP2015-42159A

Middle molecular compounds are expected as new drug discovery materials that inhibit intracellular protein-protein interactions, especially due to their strong physiological activity. However, since many of them do not exhibit membrane permeability, a method for obtaining a medium molecular compound that can act on a desired intracellular target has not yet been established. An object of the present invention is to impart a membrane permeability to a medium molecular compound and to allow binding to an intracellular target.

The inventors of the present invention have focused on the ability to bind to peptidylprolyl isomerase (PPI), which is one of the proteins abundant in cells, as a feature common to medium molecules exhibiting membrane permeability. . The membrane-permeability fusion thus obtained is made by fusing a peptide sequence that binds to a protein that is present in cells typified by PPI and has an activity of pulling out the middle molecule from the lipid bilayer with the middle molecule. The present inventors have found that the ability of the body to move into cells can be improved, thereby completing the present invention.

In one embodiment of the present invention, the method of membrane permeation of a medium molecule is a peptide that binds to a medium molecule having a molecular weight of 500 to 5000, which is present in a target cell and has an activity of pulling out the medium molecule from the lipid bilayer membrane. The step is characterized in that a sequence is fused to obtain a membrane-permeable fusion, and the membrane-permeable fusion and a cell are brought into contact with each other.
A protein that is present in the target cell and has an activity of pulling out a medium molecule from the lipid bilayer membrane is one of the abundant proteins in the cell and is weakly localized in the lipid bilayer membrane. It is a presentation protein that has the activity of binding to medium molecules. This protein is preferably a protein having peptidylprolyl isomerase activity, more preferably peptidylprolyl isomerase A (PPIA), but if it has an activity to pull out a medium molecule from a lipid bilayer membrane, It is not limited to these. The peptide sequence that binds to the protein having peptidylprolyl isomerase activity is an amino acid residue sequence represented by the following formula (I):
Leu-Leu-Val-Bmt-Abu (I)
Wherein Bmt represents (4R) -4-[(E) -2-butenyl] -4-methyl-L-threonine; Abu represents L-2-aminobutyric acid and forms a peptide bond The amino group may be N-alkylated.) Or a homologous amino acid residue sequence thereof.

In another embodiment, the method for producing a membrane-permeable fusion includes preparing a medium molecule having a molecular weight of 500 to 5000, and having the medium molecule in the target cell to extract the medium molecule from the lipid bilayer membrane. It is characterized by fusing a peptide sequence that binds to a protein having

In yet another embodiment, the membrane-permeable cyclic peptide of the present invention has the following formula (II):
Cyclo [(X) _n -Leu-Leu-Val-Bmt-Abu]
... (II)
(In the formula, n X's are each independently any natural or non-natural amino acid or derivative thereof, n is an integer of 5 to 50, and Bmt is (4R) -4-[(E ) -2-butenyl] -4-methyl-L-threonine; Abu represents L-2-aminobutyric acid, and the amino group forming the peptide bond may be N-alkylated.)
It consists of the amino acid sequence represented by these, or its homologous amino acid residue. However, the cyclic peptide excludes cyclosporin A.

Yet another embodiment of the present invention is a method for screening a membrane-permeable cyclic peptide capable of binding to a target molecule, wherein the random amino acids are represented by n Xs represented by (X) _n in the above formula (II). A step of constructing a cyclic peptide library into which is introduced, a step of bringing the cyclic peptide library into contact with a target molecule, and a step of selecting a membrane-permeable cyclic peptide bound to the target molecule.

According to the method of the present invention, various intermediate molecules can be permeated through a membrane by fusing a peptide sequence that binds to a protein that exists in the target cell and has an activity of extracting the intermediate molecule from the lipid bilayer membrane. In addition, by using a cyclic peptide library composed of membrane-permeable fusions, which are various fusion peptides, it is possible to select medium molecules that have membrane permeability and have the ability to bind to desired intracellular targets. Is possible.

FIG. 1 shows the structure of cyclosporin A (CsA). FIG. 2 shows the three-dimensional structure of a complex of calcineurin (Cn), cyclosporin A (CsA), and PPIA. FIG. 3A shows the three-dimensional structure of Cn, FK506 and FKBP complex. FIG. 3B shows the three-dimensional structure of the rapamycin target protein (mTOR), rapamycin (Rap) and FKBP complex. FIG. 4 is a schematic diagram showing how the middle molecules embedded in the cell membrane are extracted by PPI and migrate into the cell. FIG. 5A is a schematic diagram showing how PPIA pulls out CsA encapsulated in DPC micelles. FIG. 5B shows the result of NMR analysis of the bond between CsA and PPIA encapsulated in DPC micelles. FIG. 6A shows the chemical shift change of each residue due to the binding of CsA and PPIA added in the state of being encapsulated in DPC micelles. FIG. 6B shows the mapping onto the PPIA / CsA complex (in the absence of micelles) structure. FIG. 7 shows the result of NMR analysis of the bond between DPC micelle not containing CsA and PPIA. FIG. 8 shows the results of a liposome addition experiment for PPIA. FIG. 9 shows the results of a CsA-containing liposome addition experiment for PPIA. FIG. 10 is a graph showing changes in signal intensity of CsA-bound and non-bound PPIA when CsA-containing liposomes are added to PPIA.

Next, preferred embodiments of the present invention will be described with reference to the drawings. Note that the embodiments described below do not limit the invention according to the claims, and all the elements and combinations thereof described in the embodiments are essential to the solution means of the present invention. Not necessarily.

[Molecular molecular permeation method]
The method according to the present embodiment is a medium molecular membrane permeation method. In the present specification, “medium molecule” means a peptide or macrolide compound having a molecular weight of about 500 to 5000, which is neither a low molecule (an organic compound having a molecular weight of up to about 500) nor a polymer (a protein having a molecular weight of 10,000 or more). , Nucleic acids and natural products or their derivatives. A peptide having a molecular weight of about 500 to 2000, that is, a linear or cyclic peptide having about 5 to 20 amino acid residues is preferable. As the peptide, a cyclic peptide is preferable, and details thereof will be described later. The macrolide compound is a macrocyclic lactone and is a general term for compounds having 12 or more ring members. Examples thereof include FK506 and rapamycin. The nucleic acid includes DNA and RNA, and includes, for example, short interfering RNA (siRNA), double stranded RNA (dsRNA), micro RNA (miRNA), short hairpin RNA (shRNA), and nucleic acid aptamer.
By using such a medium molecule as a pharmaceutical product, it is expected to develop a pharmaceutical product that is inexpensive, less immunogenic, and highly specific. A peptide sequence (sometimes referred to as a “membrane permeation promoting peptide sequence”) that binds to a protein that is abundant in the target cell and has an activity of pulling out the medium molecule from the lipid bilayer membrane. By fusing, it is possible to realize a medium molecular drug that can penetrate the cell membrane and reach the target molecule in the details.

In the present specification, the “target cell” means a cell into which a medium molecule is introduced by the method of this embodiment, and is a eukaryotic cell, prokaryotic cell, animal cell, plant cell, fungal cell, archaeal cell, eubacteria Includes cells. Cells include eukaryotic cells such as yeast cells, plant cells and animal cells. Specific cells include mammalian cells. In addition, cells include any cell that is beneficial or desirable to introduce medium molecules. Such cells may include specific cells that cause disease or an adverse health condition. For example, it is a tumor cell or an immune cell, and by introducing a medium molecule into these cells, cell proliferation can be suppressed, or enhancement of immune reaction can be suppressed. Alternatively, medium molecules can be introduced into harmful bacterial cells to inhibit their growth or kill them. In this way, therapeutic treatment is provided by the methods described herein. The type of cell is not particularly limited, and by the method of the present embodiment, various tissues (eg, liver, kidney, pancreas, lung, spleen, heart, blood, muscle, bone, brain, stomach, small intestine, large intestine, skin, It is possible to introduce a fusion in which a medium molecule and a membrane permeation facilitating peptide sequence are fused into cells in adipose tissue or the like). In the present specification, the “membrane” means a plasma membrane such as a cell or a lipid bilayer membrane.

In the present specification, “a protein that is present in a target cell and has an activity of extracting a middle molecule from the lipid bilayer membrane” is not particularly limited, but it is abundant in the target cell and is present in the lipid bilayer membrane. Proteins with the activity of pulling medium molecules from, for example, proteins with peptidylprolyl isomerase activity, such as a group of enzymes that catalyze cis-trans isomerization of proline residues in protein molecules and thyroid hormone receptors Intracytoplasmic receptors are mentioned. In general, the peptide bond between amino acids is much more stable in the trans form than the cis form (low energy state), and this state is achieved naturally. However, the proline residue has a relatively stable N-side peptide bond as a cis isomer due to its unique structure (exactly an amino acid, not an amino acid). These need to be somewhere for the correct folding of the protein. However, since the activation energy required for cis-trans isomerization of this bond is relatively high at about 20 kcal / mol, this bond is not naturally isomerized, and the isomerization of the proline residue needs to be catalyzed for folding. There is. Prolyl isomerase works here and can therefore be called one of the chaperones. About 6000 types of PPIs from prokaryotes to eukaryotes have been identified and are present in large amounts in cells. For example, peptidylprolyl isomerase A (also referred to as PPIA or cyclophilin A) is the 22nd most abundant protein in mammalian cells and is presumed to be abundant in cells at a concentration of μM level ( Schwanhausser B et al., Global quantification of mammalian gene expression control., Nature. 2011 May 19; 473 (7347): 337-42). PPIs are classified into three subfamilies based on their amino acid sequence homology: the cyclophilin family, the FKBP (FK506 binding protein) family, and the parvulin family. In addition, a part of cytoplasmic receptors such as thyroid hormone receptors can be used for the same purpose because a hormone having a molecular weight exceeding 500 can be extracted from the cell membrane and transferred into the cytoplasm or nucleus.

Cyclophilin A is known to be involved in AIDS virus HIV-1 infection, and cyclophilin D is involved in myocardial infarction and cerebral infarction. There is a report that FKBP12 functions in a cell proliferation signal transduction system. Cyclophilin A and FKBP12 are also direct receptors for immunosuppressants, and they are also called immunophilins. On the other hand, the Parbulin subfamily Pin1 has high specificity for the phosphorylated Ser / ThrPro sequence and functions in various aspects such as transcription, apoptosis, and regulation of amyloid β production. It is thought that these PPIs may control activation / deactivation by catalyzing cis / trans isomerization reaction in cells to change the structure around the proline peptide bond of the substrate protein. Yes. However, the catalytic mechanism of the cis-trans isomerization reaction by PPI is still unclear. Prolyl isomerase may include subunits or modules of different functions, such as modules that exhibit catalytic activity and modules that exhibit chaperone or binding activity.

Therefore, the medium molecule obtained by the method of the present embodiment is a peptide sequence (membrane permeation promoting peptide sequence) that is present in the target cell and binds to a protein (for example, PPI) having an activity of pulling out the medium molecule from the lipid bilayer membrane. It is considered that the protein is extracted from the cell membrane by PPI or the like and presented to the target protein in the cell. As used herein, “binding” includes the formation of a complex between two molecules non-covalently. “Fusion” means that two molecules are made into one molecule by a covalent bond. Fusion of the medium molecule with the membrane permeation enhancing peptide sequence may form one or more covalent bonds. Further, it is preferable that the medium molecule and the membrane permeation promoting peptide sequence are fused by two or more covalent bonds to form a ring. In a preferred embodiment, peptidylprolyl isomerase A (PPIA or cyclophilin A) can be used as the protein having the peptidylprolyl isomerase activity. In a more preferred embodiment, the peptide sequence of this embodiment can use the binding site of cyclosporin A (CsA) and PPIA (cyclophilin A).

FIG. 2 schematically shows a three-dimensional structure of a complex (Cn / CsA / PPIA complex) of calcineurin (Cn), cyclosporin A (CsA), and PPIA, which is a model of the present invention. . Cn consists of two subunits. In FIG. 2, CnA represents a catalytic subunit, and CnB represents a calcium-binding subunit. In the Cn / CsA / PPIA complex, most of the three contact surfaces are carried by CsA, and PPIA almost only presents CsA to Cn. Furthermore, CsA interacts with PPIA to fix the three-dimensional structure, and acquires the binding ability to Cn on the surface opposite to the binding site with PPIA.

As another model example, FIG. 3 shows the binding mode between FKBP and FK506 (A) or rapamycin (B). FKBP, a kind of PPI, has an interesting feature that it binds to macrolide compounds such as FK506 and rapamycin (Rap). Since FK506 and Rap have different intracellular target proteins from Cn or mTOR, even if they are medium molecules introduced into the cell by the same PPI, the structure other than the binding site to PPI is different. It shows that it can bind to the target protein.

From this, the presumed mechanism by which CsA present in the cell membrane is extracted by PPI and moves into the cell is schematically shown in FIG. CsA, which is highly hydrophobic, has a high affinity for the cell membrane, and is considered to easily migrate to the membrane when absorbed into the body (arrow A in FIG. 4). However, since it is poorly soluble in an aqueous solution, it cannot be transferred from the membrane into the cell as it is. At this time, PPIA present in the cell assists in membrane permeation. PPIA is weak and shows binding to the membrane as shown later (arrow B in FIG. 4). Since PPIA bound to the membrane directs the CsA binding surface in the membrane direction, it efficiently forms a complex with PPI and moves from the cell membrane into the cytoplasm (arrow C in FIG. 4). As shown in FIG. 2, this complex is presented in the cytoplasm to the target protein calcineurin. At this time, either PPI abundantly present in the cell directly pulls out with CsA existing in the membrane, or catches CsA slightly distributed in the cell from the membrane and transfers it from the membrane to the cell. Good.

The present invention relates to an intracellular antigen-presenting protein that concentrates and presents in a cell a protein group that can be extracted in a cell, such as PPIA, and other proteins that can extract a medium molecule in a membrane. By using this, CsA increases the membrane permeability of medium molecules by a mechanism similar to that obtained by PPIA. As demonstrated in the examples described later, since CsA shifts from the membrane fraction to the soluble fraction by the addition of PPIA, it is reliably predicted that it will pass through the membrane and enter the cytoplasm by the method of the present invention. it can.

In one embodiment of the present invention, the membrane permeation facilitating peptide sequence consists of a peptide having 5 or more residues, including natural or non-natural amino acids having no side chain charge or derivatives thereof. Non-charged amino acids in the side chain include naturally occurring amino acids such as glycine, alanine, valine, leucine, isoleucine, methionine, proline, phenylalanine, and tryptophan. Among these, hydrophobic nonpolar amino acids such as glycine, alanine, valine, leucine, isoleucine, proline and phenylalanine are preferable.

For example, an amino acid residue represented by the following formula (I) (SEQ ID NO: 1):
Leu-Leu-Val-Bmt-Abu (I)
Wherein Bmt represents (4R) -4-[(E) -2-butenyl] -4-methyl-L-threonine; Abu represents L-2-aminobutyric acid and forms a peptide bond The amino group may be N-alkylated.). The amino acid residues represented by the above formula (I) include five amino acid residues that CsA is presumed to bind to PPIA: MeLeu-MeLeu-MeVal-MeBmt-Abu (SEQ ID NO: 3), but binding to PPIA As long as it has an ability, it may be substituted by a homologous amino acid residue as found in other cyclosporin families. As used herein, “homologous amino acid residues” are amino acid residues that share similar characteristics or properties, for example, residues that are similar in polarity, charge, size, aromaticity and / or hydrophobicity. Say. For example, Bmt may be MeBmt, MeLeu, desoxy-MeBmt or methylaminooctanoic acid, Abu may be L-Ala, L-Thr, L-Val, L-norvaline, L-Ala. Also included are optical isomers such as D-Ala. In this specification, the notation such as MeBmt and MeLeu indicates N-methyl Bmt and N-methyl Leu.

In addition, since the amino acid sequence represented by the above formula (I) is highly hydrophobic, it is considered that a medium molecule containing this sequence easily moves to the cell membrane only by mixing with cells. In particular, N-alkylation of the peptide backbone has been reported as a method to increase peptide membrane permeability and resistance to proteolysis (eg Yan T, Feringa BL, Barta K, Direct N-alkylation of unprotected amino acids with alcohols. Sci Adv. 2017 Dec 8; 3 (12): eaao6494 etc.). Of the N-alkylations, N-methylation is preferred.

The method for fusing the medium molecule and the membrane permeation-enhancing peptide sequence is not particularly limited, but as a method that can be generally applied to any medium molecule, separately prepared medium molecule and membrane permeation-enhancing peptide sequence, There is a method of fusing by well-known and usual chemical reaction means. For example, various homobifunctional and / or heterobifunctional crosslinker reagents such as bis (sulfosuccinimidyl) suberate, bis (diazobenzidine), dimethyl adipimidate, dimethyl pimelic acid, dimethyl suberimidate, suberin Disuccinimidylate, glutaraldehyde, m-maleimidobenzoyl-N-hydroxysuccinimide, sulfo-m-maleimidobenzoyl-N-hydroxysuccinimide, sulfosuccinimidyl 4- (N-maleimidomethyl) cyclohexane-1-carboxylate, sulfosuccin Nimidyl 4- (p-maleimido-phenyl) butyrate and (1-ethyl-3- (3-dimethyl-aminopropyl) carbodiimide can be used, with an appropriate linker between the middle molecule and the membrane permeation promoting peptide sequence. Insertion may be.

As another embodiment, a DNA encoding a medium molecule consisting of a peptide and a DNA encoding a membrane permeation facilitating peptide sequence or a variant thereof are linked in a form that can function in frame, and the fusion DNA is expressed as an expression vector. The recombinant peptide can be produced by genetic engineering by introducing the vector into an appropriate host cell and expressing the vector. At this time, elements such as a promoter and a terminator may be linked in a functionable manner. In addition, when a linker composed only of genetically encoded amino acids is arranged, the above linker peptide is inserted between the DNA encoding the medium molecule and the DNA encoding the membrane permeation promoting peptide sequence or a variant thereof. Alternatively, DNA encoding may be interposed.

More specifically, when the medium molecule is a peptide or a derivative thereof, a known peptide synthesis method can be used for fusion with the membrane permeation promoting peptide sequence, details of which will be described later. When the medium molecule is a macrolide compound, a known chemical synthesis method similar to the peptide synthesis method can be used. For example, a convergent synthesis method has been reported in which multiple parts called building blocks are created in advance and the desired macrolide compounds are synthesized by combining them (A platform for the discovery of new macrolide antibiotics, Nature volume 533, pages 338-345 (19 May 2016)). A functional group that reacts with the above-mentioned homobifunctional and / or heterobifunctional crosslinking agent reagent may be introduced into these building blocks.

On the other hand, as a technique for covalently binding a nucleic acid and a peptide, a method using an enzyme called terminal deoxynucleotidyl transferase (hereinafter abbreviated as TdT) has also been reported (see, for example, JP-A-2004-331574). This enzyme is an enzyme capable of incorporating triphosphate into the 3 ′ end of DNA. Using this enzyme, a modified nucleic acid can be introduced at the 3 ′ end and covalently bound to the peptide. For example, as a modified nucleic acid, a nucleotide having a maleimide group or a thiol group bonded to a nucleobase via a linker can be reacted with a thiol group present in the peptide to form a covalent bond. Further, the thiol group in the modified nucleic acid can react with a maleimide group introduced into the peptide, or a compound having a disulfide bond or a halogenated methyl group to form a covalent bond. Furthermore, as a method for carrying out without an protecting group in an efficient and mild aqueous medium, a method of chemoselectively fusing a peptide and a nucleic acid (oligonucleotide) using an oxime or thiazolidine formation reaction has been reported ( Forget D, Boturyn D, Defrancq E, Lhomme J, Dumy P., Highly efficient synthesis of peptide-oligonucleotide conjugates: chemoselective oxime and thiazolidine formation. Chemistry. 2001, 7 (18): 3976-84. In the oxime formation reaction, a peptide containing an oxyamine group (—ONH ₂ ) and a nucleic acid containing an aldehyde group (—CHO) or the reverse reaction product is used. The thiazolidine forming reaction is a reaction in which a thiazolidine ring is formed using a peptide containing a cysteine residue (—CH (NH ₂ ) CH ₂ SH) and a nucleic acid containing an aldehyde group (—CHO).

In addition, there are a method of binding a peptide and a nucleic acid on a ribosome using puromycin, a method of conjugating a photosensitive cross-linking reagent to the end of the nucleic acid, and a method of binding with affinity between streptavidin and biotin. Although it is not strictly a covalent fusion, a peptide nucleic acid (PNA) and a normal nucleic acid are annealed to form a non-covalent but firm duplex, or a nucleic acid is attached to the end of a protein. It can also be attached to form a duplex.
Hereinafter, a membrane-permeable cyclic peptide will be described as a typical example of such a fusion molecule of a medium molecule and a membrane permeation promoting peptide.

[Membrane-permeable cyclic peptide]
In the present specification, the “membrane-permeable cyclic peptide” refers to the following formula (II):
Cyclo [(X) _n -Leu-Leu-Val-Bmt-Abu]
... (II)
(In the formula, n X's are natural or non-natural amino acids or derivatives thereof independently of each other, n is an integer of 5 to 50, and Bmt is (4R) -4-[(E ) -2-butenyl] -4-methyl-L-threonine; Abu represents L-2-aminobutyric acid, and the amino group forming the peptide bond may be N-alkylated.)
It consists of the amino acid sequence represented by these, or its homologous amino acid residue. However, cyclosporin A is excluded from the membrane-permeable cyclic peptide of this embodiment.

In the membrane-permeable cyclic peptide represented by the above formula (II), the sequence represented by (X) _n corresponds to the middle molecule in the present invention, and Leu-Leu-Val-Bmt-Abu (SEQ ID NO: 1) The sequence represented is the membrane permeation promoting peptide sequence in the present invention. Therefore, the membrane-permeable cyclic peptide represented by the formula (II) is a peptide including a cyclic structure composed of 8 or more amino acids as a whole. From the viewpoint of stabilizing the cyclic structure, it is preferably composed of 9 or 10 or more amino acids. The upper limit of the total number of amino acid residues is preferably 25 or less, more preferably 13 or less, from the viewpoint of stabilizing the cyclic structure. Therefore, a cyclic peptide consisting of 11 amino acid residues is most preferable. In the case of a cyclic peptide consisting of 11 amino acid residues, there are about 47,000 kinds (6 6) of hydrophobic nonpolar natural amino acids (Ile, Leu, Val, Ala, Gly, Phe) except for Pro. Including both L and D fields, there are about 3 million species (12 to the sixth power). And of course, the addition of special amino acids and amino acid derivatives can give a greater variety (infinite) diversity.

In the cyclic structure, two amino acids are disulfide bond, peptide bond, alkyl bond, alkenyl bond, ester bond, thioester bond, ether bond, thioether bond, phosphonate ether bond, azo bond, CSC bond, CN -C bond, C = NC bond, amide bond, lactam bridge, carbamoyl bond, urea bond, thiourea bond, amine bond, thioamide bond, etc. Not. By cyclizing the peptide, the structure of the peptide may be stabilized and the affinity for the target may be increased.

Cyclization is not limited to the bond between the N-terminal and C-terminal amino acids of the peptide, and may be due to the bond between the terminal amino acid and an amino acid other than the terminal, or the bond between amino acids other than the terminal. In the cyclic peptide, when one of the amino acids bonded for ring formation is a terminal amino acid and the other is a non-terminal amino acid, the cyclic peptide has a structure in which a linear peptide is attached to the cyclic structure like a tail. In the present specification, such a structure may be referred to as a “lasso type”.

In the present specification, “amino acid or derivative thereof” is used in its broadest sense, and includes artificial amino acid variants and derivatives in addition to natural amino acids. Amino acids may be shown in conventional one-letter code or three-letter code. In the present specification, examples of amino acids or derivatives thereof include natural proteinaceous L-amino acids; unnatural amino acids; chemically synthesized compounds having characteristics known in the art that are characteristic of amino acids. Examples of non-natural amino acids include α, α-disubstituted amino acids (such as α-methylalanine), N-alkyl-α-amino acids, D-amino acids, β-amino acids, α-, whose main chain structure is different from that of the natural type. Hydroxy acids, amino acids whose side chain structure is different from the natural type (norleucine, homohistidine, etc.), amino acids with extra methylene in the side chain (such as “homo” amino acids, homophenylalanine, homohistidine, etc.), extra in the side chain Amino acids having halogen (F, Cl, Br, I), extra amine (N) or amino group in the side chain, amino acid having extra oxine (O) or carboxy group in the side chain, extra sulfone (in the side chain) S), amino acids having a thio group, and amino acids (such as cysteic acid) in which the carboxylic acid functional group in the side chain is substituted with a sulfonic acid group, but are not limited thereto. .

Amino acids include proteinogenic amino acids and non-proteinogenic amino acids. In the present specification, the “protein amino acid” means an amino acid (Arg, His, Lys, Asp, Glu, Ser, Thr, Asn, Gln, Cys, Gly, Pro, Ala, Ile, Leu, Met, Phe constituting a protein. , Trp, Tyr, and Val).
As used herein, “non-protein amino acid” means a natural or non-natural amino acid other than protein amino acids.

[Method for producing peptide]
The method for preparing the peptide according to the present invention is not particularly limited. Peptides or cyclic peptides according to the present invention are well known in liquid phase methods, solid phase methods, chemical synthesis methods such as hybrid methods combining liquid phase methods and solid phase methods, gene recombination methods, translation synthesis using cell-free translation systems, etc. Or a method analogous thereto.

1. Synthesis by Solid Phase Method The peptide or cyclic peptide according to the present invention can be prepared by solid phase synthesis in addition to the method shown in Scheme 1 described below, but is not limited thereto.
In the solid phase method, for example, a hydroxyl group of a resin having a hydroxyl group is esterified with a carboxy group of a first amino acid (usually a C-terminal amino acid of a target peptide) in which an α-amino group is protected with a protecting group. . As the esterification catalyst, known dehydration condensing agents such as 1-mesitylenesulfonyl-3-nitro-1,2,4-triazole (MSNT), dicyclohexylcarbodiimide (DCC), diisopropylcarbodiimide (DIPCDI) and the like can be used.

Next, the α-amino group protecting group of the first amino acid is removed, and a second amino acid in which all functional groups other than the carboxy group of the main chain are protected is added to activate the carboxy group. , Combining the first and second amino acids. Further, the α-amino group of the second amino acid is deprotected, a third amino acid in which all functional groups other than the carboxy group of the main chain are protected is added, the carboxy group is activated, and the second and Bind a third amino acid. This is repeated, and when a peptide having a desired length is synthesized, all functional groups are deprotected.

Protecting groups for α-amino groups include benzyloxycarbonyl (Cbz or Z) group, tert-butoxycarbonyl (Boc) group, fluorenylmethoxycarbonyl (Fmoc) group, benzyl group, allyl group, allyloxycarbonyl (Alloc). ) Group and the like. The Cbz group can be deprotected by hydrofluoric acid, hydrogenation, etc., the Boc group can be deprotected by trifluoroacetic acid (TFA), and the Fmoc group can be deprotected by treatment with piperidine.

For protecting the α-carboxy group, methyl ester, ethyl ester, benzyl ester, tert-butyl ester, cyclohexyl ester or the like can be used. As other functional groups of amino acids, the hydroxy group of serine or threonine can be protected with a benzyl group or a tert-butyl group, and the hydroxy group of tyrosine is protected with a 2-bromobenzyloxycarbonyl dilute or tert-butyl group. The amino group of the lysine side chain and the carboxy group of glutamic acid or aspartic acid can be protected in the same manner as the α-amino group and α-carboxy group.

The activation of the carboxy group can be performed using a condensing agent. Examples of the condensing agent include dicyclohexylcarbodiimide (DCC), diisopropylcarbodiimide (DIPCDI), 1-ethyl-3- (3-dimethylaminopropyl) carbodiimide (EDC or WSC), (1H-benzotriazol-1-yloxy) tris (Dimethylamino) phosphonium hexafluorophosphate (BOP), 1- [bis (dimethylamino) methyl] -1H-benzotriazolium-3-oxide hexafluorophosphate (HBTU) and the like. The peptide chain can be cleaved from the resin by treatment with an acid such as TFA or hydrogen fluoride (HF).

2. Cyclization method The linear peptide biologically or chemically synthesized by the above-described method can be cyclized by various methods other than the method shown in Scheme 1 described later. For example, there is a technique for generating a disulfide bond via a cysteine residue introduced by chemical modification or genetic engineering modification (see, for example, US Pat. No. 4,033,940 and US Pat. No. 4,102,877). Another technique is to generate linear precursors in the cell and then add exogenous agents such as proteases or nucleophiles to chemically cyclize these linear precursors (eg, Julio A. Camarero and Tom W. Muir, J. Am. Chem. Soc., 1999, 121 (23), pp 5597-5598). In another technique, a cyclized peptide is generated in a cell by genetic engineering modification. For example, utilizing the trans-splicing ability of a split intein, the precursor peptide interposed between the two parts of the split intein is cyclized. Here, an intein (sometimes referred to as a protein intron) is a part of a protein molecule that is automatically excised, and the remaining part (extein) is rebound by a peptide bond (“protein splicing”). Say things.

Also, in another method, a cyclized peptide is generated through chemical modification by introducing an unnatural amino acid or a cross-linking agent molecule. For example, a technique is known in which (S ′)-α- (2′-pentenyl) alanine is cross-linked using an olefin metathesis reaction to cyclize the peptide within the molecule (see Japanese Patent Application Publication No. 2008-501623). ).

Furthermore, a method using the S _N 2 reaction of a chloroacetyl group and a thiol group of cysteine (Goto, Y., et al., Reprogramming the translation initiation for the synthesis of physiologically stable cyclic peptides. ACS Chem. Biol. 2008; 3 (2): 120-9) is more stable under reducing conditions in cells than the disulfide bond between cysteine residues found in natural peptides and proteins. Click chemistry using the Huisgen reaction between an azide group and an alkynyl group, for example, forms a triazole ring by reacting two functional groups by allowing a monovalent copper ion to act on azidohomoalanine and propargylglycine. However, since these two functional groups do not react with any proteinaceous amino acid under these conditions, a cyclic structure can be selectively formed at the target position (Sako, Y., et al., Ribosomal synthesis of bicyclic peptides via two orthogonal inter-side-chain reactions. J. Am. Chem. Soc. 2008; 130 (23): 7232-4).

[Cyclic peptide library and screening method]
The present invention also provides a library of cyclic peptides as described above and a screening method using the same. As used herein, “cyclic peptide library” refers to the following formula (II):
Cyclo [(X) _n -Leu-Leu-Val-Bmt-Abu]
... (II)
(In the formula, n X's are each independently any natural or non-natural amino acid or derivative thereof, n is an integer of 5 to 50, and Bmt is (4R) -4-[(E ) -2-butenyl] -4-methyl-L-threonine; Abu represents L-2-aminobutyric acid, and the amino group forming the peptide bond may be N-alkylated.)
A random amino acid sequence in a portion other than the amino acid residue that is a binding site to PPI, preferably a binding site to PPIA: MeLeu-MeLeu-MeVal-MeBmt-Abu (SEQ ID NO: 3) An aggregate of peptides containing a plurality of cyclic peptides. As for the amino acid residues other than the binding site with PPI, all amino acids may have a random sequence, or a part of the amino acid residues may have a random sequence. Preferably contains 2 million or more cyclic peptides.

As a method other than the method for creating a peptide library described in Example 3, there is also a method called a split-mix method based on a peptide solid phase synthesis method, for example. In this method, resin beads on which different amino acids (for example, 20 types) are immobilized are prepared and mixed together. Thereafter, the resin beads are equally divided (for example, 20 equal parts), and 20 different amino acids are coupled to each. After the resin beads that have been reacted individually are mixed together again, and then divided again and coupled to different amino acids many times, different peptide sequences are finally immobilized on each resin bead. The completed peptide library is completed.

On the other hand, by using a translation system, which is a polypeptide synthesis system using ribosomes, peptides having different sequences can be synthesized simply by changing the mRNA sequence. Since it is easy to prepare mRNA having a random sequence, it is possible to easily construct a highly diverse peptide library by using a translation system.

The screening method using the cyclic peptide library includes a step of bringing the cyclic peptide library into contact with the target molecule and incubating. In the present specification, the target molecule is not particularly limited, and may be a low molecular compound, a high molecular compound, a nucleic acid, a peptide, a protein, a sugar, a lipid, or the like, but is typically a protein. In particular, since the cyclic peptide according to the present invention has high in vivo stability and excellent cell membrane permeability, it is preferable to target intracellular proteins. Since the cyclic peptide according to the present invention is also excellent in protease resistance, it can be screened for target molecules having protease activity.

The target molecule can be contacted with a cyclic peptide library, for example, immobilized on a solid support. In the present specification, the “solid phase carrier” is not particularly limited as long as it can immobilize a target molecule, and is made of a microtiter plate made of glass, metal, resin, etc., substrate, beads, nitrocellulose membrane, nylon A membrane, a PVDF membrane, etc. are mentioned. The target molecule can be immobilized on these solid phase carriers by a known method.
The target molecule and the library are brought into contact with each other in an appropriately selected buffer, and the pH, temperature, time and the like are adjusted to interact with each other.

The screening method according to the present invention next includes a step of selecting a cyclic peptide bound to the target molecule. For binding to the target molecule, for example, the peptide is detectably labeled by a known method, and after the incubation, the surface of the solid phase carrier is washed with a buffer, and the compound bound to the target molecule is analyzed by mass spectrometry or the like. This can be done by identification using NMR methods. Examples of detectable labels include enzymes such as peroxidase and alkaline phosphatase, radioisotopes such as ¹²⁵ I, ¹³¹ I, ³⁵ S, and ³ H, fluorescein isothiocyanate, rhodamine, dansyl chloride, phycoerythrin, tetramethylrhodamine isothiocyanate, Examples thereof include fluorescent materials such as infrared fluorescent materials, luminescent materials such as luciferase, luciferin, and aequorin, and nanoparticles such as gold colloids and quantum dots. In the case of an enzyme, an enzyme substrate may be added to cause color development and detection. Detection can also be performed by binding biotin to the peptide and binding avidin or streptavidin labeled with an enzyme or the like.

In the above step, not only the presence / absence or degree of binding is detected / measured, but also the increase or inhibition of the activity of the target molecule is measured, and the cyclic peptide having such enhancement or inhibition activity is subjected to mass spectrometry or NMR. And can be identified. By such a method, a cyclic peptide having physiological activity and useful as a pharmaceutical can be obtained.

The cyclic peptide that binds to the target molecule obtained by screening the cyclic peptide library according to the present invention may be further modified and optimized by a known method or a method analogous thereto.

Hereinafter, the present invention will be specifically described with reference to examples. However, the present invention is not limited to these examples.

[Example 1] Analysis of interaction between PPIA and micelle-mixed CsA Cyclosporin A (CsA; see Fig. 1 and SEQ ID NO: 2) is highly hydrophobic because of its structure, and is considered to easily migrate to the membrane. However, the solubility in an aqueous solution is very low, and it is inherently difficult to escape from the membrane and bind to intracellular targets (Cn). Accordingly, the present inventors believe that CsA is prevented from being trapped in the cell membrane by the PPI abundantly present in the cell pulling out CsA from the membrane, and efficient intracellular transfer of CsA is realized. It was. Therefore, in order to verify this, first, a sample in which CsA was incorporated into a DPC (dodecylphosphocholine) micelle was prepared (upper left in FIG. 5A), and PPIA (upper right in FIG. 5A) labeled with ¹⁵ N was stable. When CsA was added, it was analyzed whether CsA bound to PPIA (FIG. 5A bottom).

Figure ^{5B, 15} N stable isotope labeled PPIA (200 [mu] M) shows a ¹ H ¹⁵ N HSQC measurement results of the spectrum before and after the addition of 20mMDPC containing therein the CsA of 200 [mu] M with respect to (before addition: black, added After: gray). ^{In the 1} H ¹⁵ N HSQC spectrum, each signal corresponds to one amino acid residue, and the position of the signal in the spectrum changes when the chemical shift changes due to changes in the surrounding chemical environment such as binding. Therefore, in this case, the residue of PPIA whose signal position is changed depending on the presence or absence of DPC containing CsA can be identified as the CsA binding site. At this time, since only the labeled molecule gives a signal, only the signal of PPIA stably labeled with ¹⁵ N is observed at this time. FIG. 6 (A) shows the chemical shift change of each residue due to the combination of CsA and PPIA added in the state of being encapsulated in DPC micelles (the ones greatly changed by 0.8 ppm or more are plotted at 0.8 ppm uniformly). FIG. 6 (B) shows the mapping onto the PPIA / CsA complex (in the absence of micelles) structure. Furthermore, FIG. 7 shows ¹ H ¹⁵ N HSQC spectra (before addition: black, after addition: gray) before and after the addition of 20 mM DPC not containing CsA to ¹⁵ N stable isotope-labeled PPIA (200 μM). ).

According to FIG. 5, since the signal of PPIA bound to CsA in the presence of DPC was detected, it was revealed that PPIA can extract CsA from DPC micelles and bind to it. Moreover, since the chemical shift change accompanying the binding was consistent with the PPIA / CsA complex structure determined in the absence of micelles, the binding mode of CsA to PPIA was the same regardless of the presence or absence of DPC micelles. (FIG. 6 (A)). Furthermore, when a ¹ H ¹⁵ N HSQC spectrum before and after addition of 20 mM DPC without CsA was added to PPIA (200 μM) labeled with ¹⁵ N as a stable isotope, the residue changed in chemical shift only with DPC. Most of the residues were residues involved in CsA binding (FIG. 7, signal surrounded by a dotted line). Therefore, it was strongly suggested that PPIA efficiently draws a membrane-permeable medium molecule such as CsA from the membrane by previously directing the CsA binding surface to the membrane (center of FIG. 4, left of FIG. 7) (FIG. 4). Center to right).

[Example 2] Preparation and NMR measurement of CsA-containing lipid bilayer membrane (liposome) In the case of an experiment using DPC micelles, considering the molecular weights of CsA (1.2 kDa) and DPC (0.35 kDa), Since the weight ratio of CsA: DMPC in this case is 3:97 and 3% is calculated by CsA, it is considered that the PPI is likely to be pulled out. Therefore, the extraction of CsA by PPIA in the same lipid bilayer membrane (liposome) as the biological membrane was examined. After drying 60 mg of DMPC (dimyristoylphosphatidylcholine) dissolved in chloroform under a nitrogen stream, it was placed under vacuum for 2 hours to create a completely dry lipid membrane film, to which 1.0 mL of NMR buffer was added, Liposomes were prepared by vortexing. The prepared liposomes were further homogenized by sonication. The final concentration of DMPC in this liposome solution is 89 mM. CsA was added to the final concentration of 200 μM and vortexed to incorporate CsA into the lipid bilayer membrane. The final molar ratio of CsA: DMPC is 1: 445, and the weight ratio of CsA: DMPC is 0.4: 99.6. The situation is closer to a biological membrane, and CsA is thin and PPIA is difficult to pull out. become. First, DMPC liposomes containing no CsA were added to PPIA (200 μM) labeled with stable isotope with ¹⁵ N, and when ¹ H ¹⁵ N HSQC spectra were compared before and after, changes in chemical shift were observed in some residues. (FIG. 8). Residues that have changed chemical shifts are residues that are also involved in CsA binding, and PPIA makes a membrane-permeable medium molecule such as CsA from the membrane by directing the CsA binding surface to the membrane in advance (center of FIG. 4). Efficient pulling was further supported. Next, when a CsA-containing liposome was added to ¹³ C stable isotope-labeled PPIA and the signal change of PPIA was observed, it was observed that CsA was extracted from the DMPC liposome to PPIA in a time-dependent manner (Fig. 9 and 10). Therefore, PPI is not only an intracellular presentation protein responsible for the activity of presenting medium molecules to the target, but is thought to improve membrane permeability by efficiently extracting medium molecules in the membrane. .

[Example 3] Preparation of peptide library CsA is an 11-residue cyclic peptide, but the motif that binds to PPIA is 5 residues of MeLeu-MeLeu-MeVal-MeBmt-Abu (FIG. 1 and SEQ ID NO: 3). Therefore, an 11-residue cyclic peptide group was created by creating a library by randomly changing the sequence of other sites of CsA while retaining this peptide sequence.

The library is created by a synthesis scheme of Non-Patent Document 1 (Wu et al, Total Synthesis of Cyclospholine: Access to N-methylated Peptide via Isonitrile Coupling Reactions, JACS, 2010, 132, 4098-) Was carried out with some modifications. According to this document, CsA can be roughly synthesized into three blocks (Scheme 1). In the following scheme 1, in each block, the site where the side chain structure can be modified is shown in black, and the site which is necessary for the binding activity with PPIA and cannot be changed is shown in gray. Block A is unchanged. Block B can modify the side chain at a site corresponding to 1 amino acid and block C corresponding to a maximum of 5 amino acids.

[Scheme 1] Total synthesis scheme of CsA

Block A is a known compound that is known to be obtained by the method of Aebi et al. (Evans DA, Weber AE. J Am Chem Soc. 1986; 108: 6757-6761.). Aebi JD, Dhaon MK, Rich DH. J Org Chem. 1987; 52: 2881-2886).

The block B was totally synthesized by the following scheme according to Non-Patent Document 1, but at that time, the side chain (R ₁ ) of B3 can be freely modified within a range in which no side reaction occurs. Therefore, even if R ₁ is limited to Gly, Ala, Val, Leu, Ile, and Phe side chains with low reactivity, the D-form and L-form should have 6 × 2 = 12 variations. Is possible.

[Synthesis scheme of block B]

The block C was completely synthesized by the following scheme according to Non-Patent Document 1, and in this case, the side chains of C1, C2, C3, C4, and C6 (R ₂ , R ₃ , R ₄ , R ₅ , R, respectively) ₆ ) can be freely modified within a range in which no side reaction occurs. Therefore, less reactive Gly, Ala, giving Val, Leu, Ile, be limited to the side chain of Phe, together D-L body ^{(6 × 2) 5 = 12} 5 kinds of variations It is possible. Therefore, it is possible to create a maximum of 12 ⁶ = 2989854 types of libraries by the synthesis scheme shown here together with R ₁ .

[Synthesis scheme of block C]

Peptides with enhanced membrane permeability by the method of the present invention, in particular, cyclic peptides obtained by screening from the cyclic peptide library of the present invention are expected to be developed as medium molecular drugs that can bind to desired target molecules in cells. Useful in the pharmaceutical industry.

Claims (13)

1. A medium-molecule transmembrane method,
   The medium molecule has a molecular weight of 500 to 5000,
   The method comprises the step of obtaining a membrane-permeable fusion by fusing a peptide sequence that binds to the medium molecule with a protein that is present in a target cell and has an activity of pulling out the medium molecule from a lipid bilayer; and Contacting the cell with a membrane permeable fusion.
2. The method according to claim 1, wherein the peptide sequence consists of a peptide having 5 or more residues including a natural or non-natural amino acid having no side chain charge or a derivative thereof.
3. The method according to claim 1 or 2, wherein the medium molecule includes a peptide, a macrolide compound, a nucleic acid, or a derivative thereof.
4. The method according to any one of claims 1 to 3, wherein the target cell is a prokaryotic cell, an animal cell or a plant cell, and the medium molecule is introduced into these cells.
5. The method according to any one of claims 1 to 4, wherein the protein present in the target cell and having an activity of extracting a middle molecule from a lipid bilayer is a protein having a peptidylprolyl isomerase activity.
6. The method according to claim 5, wherein the protein having peptidylprolyl isomerase activity is peptidylprolyl isomerase A (cyclophilin A).
7. The peptide sequence is an amino acid sequence represented by the following formula (I):
   Leu-Leu-Val-Bmt-Abu (I)
   Wherein Bmt represents (4R) -4-[(E) -2-butenyl] -4-methyl-L-threonine; Abu represents L-2-aminobutyric acid and forms a peptide bond The method according to any one of claims 1 to 6, wherein the amino group may be N-alkylated) or a homologous amino acid residue thereof.
8. A medium molecule having a molecular weight of 500 to 5000 is prepared, and the medium molecule is fused with a peptide sequence that binds to a protein that is present in the target cell and has an activity of pulling out the medium molecule from the lipid bilayer. A method for producing a membrane-permeable fusion.
9. The method according to claim 8, wherein the protein present in the target cell and having an activity of extracting a middle molecule from a lipid bilayer is a protein having a peptidylprolyl isomerase activity.
10. The method according to claim 9, wherein the protein having peptidylprolyl isomerase activity is peptidylprolyl isomerase A (cyclophilin A).
11. Formula (II) below:
    Cyclo [(X) _n -Leu-Leu-Val-Bmt-Abu]
    ... (II)
    (In the formula, n X's are natural or non-natural amino acids or derivatives thereof independently of each other, n is an integer of 5 to 50, and Bmt is (4R) -4-[(E ) -2-butenyl] -4-methyl-L-threonine; Abu represents L-2-aminobutyric acid, and the amino group forming the peptide bond may be N-alkylated.)
    A membrane-permeable cyclic peptide consisting of the amino acid sequence represented by (except for cyclosporin A).
12. A cyclic peptide library comprising 2 million or more kinds of membrane-permeable cyclic peptides according to claim 11.
13. A step of constructing a cyclic peptide library in which random amino acids are introduced into n Xs represented by (X) _{n according} to claim 11,
    Contacting the cyclic peptide library with a target molecule;
    Selecting a membrane permeable cyclic peptide bound to the target molecule;
    A method for screening a membrane-permeable cyclic peptide having a binding ability to a target molecule.

Images