JP3256307B2 - Method for constructing molecular structure of biologically active ligand - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for constructing a molecular structure of a compound having a physiological activity,
The present invention relates to a method for automatically generating a molecular structure of a novel compound, which can be used for designing a molecular structure of a drug, a pesticide or other compound having a physiological activity.

[0002]

BACKGROUND OF THE INVENTION There are two conceptual steps in drug design. A lead creation process for finding a compound (referred to as a lead) that serves as a starting point for subsequent drug development, and a lead optimization process for finding a compound with higher activity and less toxicity and side effects starting from the lead. A lead is a compound known to have a certain desired physiological activity, and is also called a lead compound. Most commonly performed drug design is this second step, often performed by synthesizing a series of substituent derivatives from known active compounds and comparing their activities. The first step is extremely difficult to perform artificially, and it is extremely difficult to find out from natural substances, from the side effects of known drugs, to modify the structure from biologically active substances (neurotransmitters, hormones, vitamins, etc.), Or, as soon as an available compound is available, discovery by random screening using a prepared evaluation system,
It depends heavily on chance, luck, and trial and error. Starting with an appropriate lead will determine the success or failure of drug development in the future, and there are many diseases for which good drugs cannot be obtained due to the lack of an appropriate lead. It is widely desired. What pharmacological activity a compound exhibits depends on which receptor (biopolymer) it binds to, but one compound does not always bind to only one receptor. Therefore, a slight difference in the structure may reverse the strength of the main activity and the side activity (side effect) or cause the activity to disappear. Also,
Even if the main activity is not changed, behavior in the body such as easiness of metabolism and interaction with a membrane greatly changes, often affecting the properties as a drug. Even more, drugs from new leads having different molecular skeletons are expected to show completely different behaviors even though they have the same main activity.

In order for a drug to exhibit physiological activity, it must reach a certain site of a receptor in the body and specifically interact with the receptor (usually a strong non-covalent association).
In recent years, the technology for isolating and purifying receptors has been advanced, and the number of enzymes and nucleic acids whose three-dimensional structures (three-dimensional coordinates of constituent atoms) are clarified by X-ray crystal analysis has increased. In particular, analysis of the complex with the ligand (in the case of an enzyme, an inhibitor) shows that a strong van der Waals interaction between the ligand molecule and the receptor, which shows strong binding, due to the good fit of the molecular surface. It has been found that hydrophilic functions such as hydrogen bonding and electrostatic interaction contribute to the stabilization of the complex. Despite the fact that the apparent structure is completely different from the active substance or enzyme substrate that originally works in vivo, binding to the same receptor has been confirmed, and many drugs have the same action or inhibitor. This indicates that what is required for strong binding between the receptor and the drug is the three-dimensional shape of both molecules and the complementarity of these properties. Therefore, if humans can design a molecular structure having a three-dimensional structure that satisfies these requirements, it should be possible to artificially create leads. However, humans are apt to be obscured by the structure of known active compounds and are not good at three-dimensional insights, making it difficult to create leads even when the three-dimensional structure of the receptor has been elucidated by crystallographic analysis. Therefore, an approach in which such a structure is presented to a computer and then selected by a human is considered effective for finding a new active structure and using it as a lead. Computers are not bound by prejudice, they cover possibilities and give results based on quantitative evidence.

[0004] There are two approaches to finding a structure that meets the conditions required for leads.
A method of searching for a structure that meets the conditions from a structure database of known compounds, and a method of newly constructing a structure that meets such conditions. If the three-dimensional structure of the receptor is available, the following reports are available on both approaches.

[0005] DesJarlais, Kuntz et al. Described the shape of the inner hole of the drug-binding site by several spheres that fit well, and used a crystal database (published by Cambridge University, UK: more than 100 scientific studies around the world). Register the analysis reported in the journal)
A method (DOCK) for judging the similarity of a receptor to a set of spheres by expressing it with several spheres including it (R.
L. DesJarlais, RPSheridan, GLSeibel, JSDixo
n, ID Kuntz and R. Venkataraghavan, J. Med. Che
m. 31, 722-729 (1988); RL DesJarlais, GL Seibe
l, ID Kuntz, PS Furth and JC Alvarez, Proc. N
atl. Acad. Sci. USA, 87, 6644-6648 (1990)). This technique is used for the protease of AIDS virus (HIV protea).
se: Registered in the Protein Data Bank), and as a result, the crystal structure of a compound called haloperidol, known as a beta-blocker and antipsychotic, was selected as a good fit for the drug-binding site of this enzyme, and minor structural modifications were made. When they were synthesized and tested for their activity, they inhibited this enzyme and proved to be effective in treating AIDS. However, this method judges only the shape of the molecule, does not consider hydrogen bonds and other interactions, and considers only the crystal structure and molecular conformation,
Leakage and inefficiency of promising candidate compounds are problems.

Also, Lewis et al. Assumed a diamond lattice having a side length of a carbon-carbon single bond at a drug binding site, and newly provided a molecular structure by appropriately setting carbon atoms on lattice points. Announced how to build (RA Lewis, J.
Comp.-Aided Molecular Design, 4, 205-210 (199
0)). However, in this method, atoms other than sp ³ hybrids cannot be set, and no specific interaction with the receptor is taken into consideration at all, which is almost impractical.

Nishibata and Itai et al. Have published a method for assembling a structure that fits well into the drug-binding site of the receptor and that has a stable interaction, and a program LEGEND for executing this method (Y. Nishibata and A. Itai, Tetrahedron, 4
7, 8986-8990 (1991)). In this method, atoms or atomic groups are added one by one based on random numbers and a force field so that van der Waals interaction, hydrogen bonding, and electrostatic interaction with the receptor become advantageous. . It is good at suggesting a number of possible structures, but has proven to be effective, but no actual success has yet been identified.

[0008] Bohm has developed a method of constructing a novel structure by inserting a suitable one of a large number of fragment structures prepared in a program into a drug-binding site of a receptor, and linking them well. (H.-J. Bohm, J. of Co
mp.-Aided Molecular Design, 6, 61-78 (1992)). However, there are no results yet to prove the effectiveness of this method.

If the three-dimensional structure of the receptor is not known, the above two approaches cannot be used.
Another approach is needed. The number of biopolymers whose three-dimensional structure has been published is about a thousand including those derived from bacteria, and it is necessary to determine the structure of human receptors, which are said to be tens of thousands in the body, one by one. Is expected to take a long time. Therefore, an approach for constructing a structure of a ligand molecule which is active when there is no information on the three-dimensional structure of the receptor is now particularly required.

When there is no information on the three-dimensional structure of the receptor, the most effective method for finding the relationship between the structure and the activity from the side of the ligand molecule such as a drug, a physiologically active substance, or a biologically active substance is the superposition of molecules. It is a match. Multiple compounds that exhibit similar physiological actions by binding to the same site of the same receptor (although it is not necessary to overlap when the molecular structures are similar)
Must have common features in shape and intermolecular interaction, and by overlapping those molecules three-dimensionally, the common features can be extracted. In addition, the determination of the molecular conformation (called active conformation or active conformation) when a molecule having a rotatable single bond in a molecule expresses an activity by binding to a receptor may be determined by superposition of molecules in some cases. Made possible by Estimating the environment and size of the drug binding site of the receptor from the ligand side is called receptor mapping, and the one estimated as such is called a receptor model. This receptor mapping can be effectively performed based on the superposition of molecules.

Although there are various methods for superimposing molecules, the results of superposition differ greatly depending on the method, and different structure-activity relationship models or receptor models (estimation of functional groups considered necessary, shape of drug binding site of receptor, etc.) And estimation of properties and environment). In order to arrive at the correct model as much as possible, the use of logical superposition is essential.

[0012] As a method of superimposing molecules that has been generally performed so far, three or more corresponding pairs of atoms are designated based on the similarity of the positions of molecular skeletons and heteroatoms.
Overlapping has been performed by least squares calculation so that the coincidence of the positions of the molecules is the best, or by appropriately judging visually on a computer graphics screen.

However, when superimposing molecules of a plurality of drugs or physiologically active substances that bind to the same receptor, Itai and Kato et al.
Do not need to match, the shape of the molecule, hydrogen bonding,
A method of superposing molecules such that electrostatic properties and other physicochemical properties match between the superimposed molecules (RECEP
S) was devised and announced. (Y. Kato, A. Itai and Y.
Iitaka, Tetrahedron, 43, 5229-5236 (1987); A. Itai,
Y. Kato, N. Tomioka, Y. Iitaka, Y. Endo, M. Haseg
awa, K. Shudo, H. Fujiki and S. Sakai, Proc. Natl.
Acad. Sci. USA, 85, 3688-3692 (1988); Y. Kato, A. I.
noue, M. Yamada, N. Tomioka and A. Itai, J. Comp.-
Aided Molecular Design, 6 in press (1992); A. Ita
i, N. Tomioka, Y. Kato, Y. Nishibata and S. Saito,
in 'Medicinal Chemistry in21st Century', edited by
CG Wermuth, Blackwell Scientific Publishings, Lo
ndon, (1992) in press; A. Itai, N. Tomioka and Y. K
ato, in 'QSAR: Newdevelopments and Applications',
edited by T. Fujita, Elsevier Science Publishers,
Amsterdam, (1992), in press.) Using this RECEPS, it is possible to interactively move and rotate molecules on a computer graphics screen using the agreement between molecules of these properties as an index, and change the conformation. It became possible to obtain superposition. This enables, for the first time, the superposition of molecules having completely different apparent chemical structures.

Then, in a system in which hydrogen bonding plays an important role in RECEPS, a computer automatically creates a superimposed structure in which the positions of hydrogen bonding functional groups expected from a plurality of molecules on the receptor side are matched. Added features that cover and rank. With this improvement,
You can now get results that include objective and correct overlays, covering the various possibilities derived from overlaying and conformation.

Using the above-described molecular superposition method,
When the three-dimensional structure of the receptor is not known, it is possible to extract features common to a plurality of ligand molecules having completely different structures. The method of designing a new ligand molecule based on this information and creating a lead has not yet been established.

[0016]

Accordingly, an object of the present invention is to provide a method for constructing a novel active ligand molecule structure based on the structure of a known active ligand molecule. Another object of the present invention is to provide a method for constructing a novel active ligand molecule structure based on a known receptor molecular structure.

[0017]

The present inventors have made intensive efforts to maintain the position and orientation of a specific functional group present in a ligand molecule in a space considered to be acceptable for the ligand molecule. While developing a method to automatically construct the molecular structure of a new ligand by comprehensively searching for the binding pathway by appropriately arranging atoms in space,
We succeeded in solving the above problems.

That is, the present invention provides a method for constructing a molecular structure of a ligand having a physiological activity, comprising the steps of (i) extracting two or more functional groups presumed to be essential for the physiological activity of the ligand; First to specify the position and direction of the functional group
A step, (ii) a second step of setting a connection method between the functional groups extracted in the first step, (iii) a third step of designating the number, type, and existence region of atoms connecting the two functional groups, (iv) placing a probe atom in the inner pore of the receptor for the ligand,
When the position of the probe atom satisfies a predetermined condition,
A fourth step, which is adopted as the position of an atom connecting the two functional groups, and (v) creating a bonding path between the two functional groups by repeating the fourth step for all the atoms specified in the third step. A fifth step, and (vi) for all combinations of the two functional groups set to be connected in the second step, by combining the binding paths obtained by repeating the third to fifth steps. ,
A sixth step of forming a skeleton of the ligand molecule.
The above method is provided.

The fourth step may be performed by arranging probe atoms on a three-dimensional lattice point, or at a point on the circumference where a bond atom can be calculated from the bond distance and bond angle. It may be performed by arranging atoms. Further, the present invention provides a method for constructing a molecular structure of a ligand having a physiological activity, comprising: (i) obtaining three-dimensional lattice point information based on the molecular structure of one or more known ligands; (Ii) extracting two or more functional groups presumed to be essential for biological activity from the molecular structure of the known ligand, Second to specify direction
(Iii) a third step of setting the way of connecting the functional groups extracted in the second step, (iv) a fourth step of specifying the number, type, and existence region of the atoms connecting the two functional groups, (v) a fifth step of arranging a probe atom in the inner hole of the receptor and adopting the position of the probe atom as the position of an atom connecting two functional groups when the position of the probe atom satisfies a predetermined condition; A) a sixth step of creating a bond path between two functional groups by repeating the fifth step for all the atoms specified in the fourth step, and (vii) setting to connect in the third step. A method comprising the step of forming a skeleton of the ligand molecule by combining the binding pathways obtained by repeating the fourth to sixth steps for all combinations of the two functional groups described above. Is provided.

Further, the positions and directions of the functional groups designated in the second step are sequentially designated again in positions and directions which are presumed to have the same effect, and the third to seventh steps are repeated. Is also good. Further, the functional groups extracted in the second step may be sequentially replaced with other functional groups having the same physicochemical properties, and the third to seventh steps may be repeated.
Furthermore, in the present invention, in addition to the first step to the seventh step, (viii) determining whether or not there is an atom whose valence is not satisfied in the molecular skeleton of the ligand, and If present, an eighth step of adding a hydrogen atom to the atom; (ix) determining, under predetermined conditions, whether it is necessary to form a ring structure in the molecular structure of the ligand; The present invention provides a method for constructing a molecular structure of a biologically active ligand, comprising a ninth step of forming a ring structure and (x) a tenth step of optimizing the molecular structure of the ligand.

According to the method of the present invention, a new ligand molecular structure can be constructed even when a known ligand molecule exists but the receptor structure is not known. Further, the present invention provides a method of constructing a molecular structure of a ligand having a physiological activity, wherein (i) a three-dimensional lattice point created in an inner hole of a receptor based on a known receptor structure of the ligand. A first step of obtaining information; (ii) a second step of extracting two or more functional groups that interact with the receptor from functional groups in a known ligand molecule, and specifying the position and direction of the functional groups; (iii) a third step of setting the way of connecting the functional groups extracted in the second step, (iv) a fourth step of specifying the number, type, and existence region of the atoms connecting the two functional groups,
(v) a fifth step of disposing a probe atom in the inner hole of the receptor and, when the position of the probe atom satisfies a predetermined condition, adopting a position of an atom connecting two functional groups,
i) a sixth step of creating a bond path between two functional groups by repeating the fifth step for all the atoms specified in the fourth step, and (vii) connecting in the third step. A seventh step of forming a skeleton of the ligand molecule by combining the binding pathways obtained by repeating the fourth to sixth steps for all combinations of the two set functional groups. It provides a method. According to the method of the present invention, a novel molecular structure of a ligand molecule can be constructed even when the structure of a complex between a receptor and a ligand molecule has been elucidated.

Furthermore, the present invention relates to a method for constructing a molecular structure of a ligand having a physiological activity, comprising: (i) a tertiary molecule formed in an inner hole of a receptor based on a known receptor structure of the ligand. The first step of obtaining information on the original grid points, (i
i) a second step of selecting two or more functional groups capable of interacting with the receptor and designating the position and direction of the functional groups, (iii)
A third step of setting a connection method between the functional groups selected in the second step, (iv) a fourth step of designating the number, type, and existence region of atoms connecting the two functional groups; A fifth step in which a probe atom is arranged in the inner hole of the body, and the position of the probe atom satisfies a predetermined condition, and is adopted as a position of an atom connecting the two functional groups; (vi) the fourth step By repeating the fifth step for all the atoms specified in the above, a bonding route between the two functional groups is created.
And (vii) combining the binding pathways obtained by repeating the fourth to sixth steps for all combinations of the two functional groups set to be connected in the third step, thereby obtaining the ligand It is an object of the present invention to provide the above method, comprising a seventh step of forming a molecular skeleton. According to the method of the present invention, even when the structure of the drug binding site is clarified by the structural analysis of the receptor alone, or when the structure of the complex of the receptor and the ligand molecule is clarified, the novel ligand Can be constructed.

The present invention also relates to a method for constructing a molecular structure of a ligand having a physiological activity, comprising the steps of: (i) preparing the atomic coordinates of one or more known ligand molecules in the inner hole of a receptor of the ligand; And (ii) extracting two or more functional groups presumed to be essential for physiological activity from the structure of the known ligand molecule, and The second step of specifying the position and direction of (i)
ii) a third step of setting the connection between the functional groups extracted in the second step, (iv) the number and type of atoms connecting the two functional groups,
And a fourth step of designating the existence region, (v) placing a probe atom in the inner hole of the receptor, and if the position of the probe atom satisfies a predetermined condition, the atom of the atom connecting the two functional groups A fifth step adopted as a position, (vi) a sixth step of creating a bonding route between two functional groups by repeating the fifth step for all atoms designated in the fourth step, and (vii ) For all combinations of the two functional groups set to be connected in the third step,
The present invention provides the above method, comprising a seventh step of forming a skeleton of the ligand molecule by combining the binding pathways obtained by repeating the steps from the sixth step to the sixth step. As the atomic coordinates of the ligand molecule input in the first step, it is possible to use the atomic coordinates of a known single recombinant molecule or the atomic coordinates of a ligand molecule superimposed on the assumption of interaction with the same receptor. it can.

In the present invention, the term "receptor" refers to any biopolymer that can be a target of a drug, and includes proteins, nucleic acids, polysaccharides, and complexes thereof. A ligand refers to a low-molecular-weight (about 1,000 or less) compound that strongly binds (interacts) to a specific site of a specific receptor, and is a biologically active substance such as a neurotransmitter, a hormone, or a vitamin. And synthetic physiologically active substances, drugs, enzyme substrates and the like.

The term "functional group" refers to an atom considered to have a specific function in a chemical reaction or molecular recognition or an atomic group (atomic group) that always acts as a group. For example, an amino group, a nitro group, a benzene ring and the like can be exemplified. The direction of the functional group refers to the direction in which it can interact with the receptor. For a functional group that forms a hydrogen bond, the direction of a lone electron pair or hydrogen is used. For an aromatic ring such as a benzene ring, the direction of a π electron perpendicular to the ring is used as the direction of the functional group.

The term "probe atom" refers to a temporary atom to be placed in the pore of the receptor in order to determine whether a certain position in the pore of the receptor is appropriate as the position of the atom constituting the ligand molecule. And Three-dimensional grid information is used to represent a receptor model spatially and position-dependently.
The grid spacing is about 0.2 to 0.5 angstroms, and refers to data obtained by calculating and storing information on the environment of the receptor at each position on each grid point. The three-dimensional grid information is
For example, for each lattice point, an address indicating a spatial position, occupancy information of each ligand molecule (up to 6 molecules), electrostatic potential, whether or not a hydrogen bonding functional group is expected, and if so, the property is hydrogen It includes data such as whether it is donating or accepting and whether the nature of the aromatics is a problem. The information occupied by the ligand molecule indicates whether or not the three-dimensional lattice point is within the region occupied by the ligand molecule, that is, whether or not the three-dimensional lattice point is within the region included in the van der Waals radius of the atoms constituting the ligand molecule. It is information for each. This three-dimensional lattice point information can be used instead of the three-dimensional structure of the receptor, and the molecular structure of the ligand can be constructed while determining at a high speed whether or not the three-dimensional lattice point fits.

[0027]

According to the method of the present invention, a novel ligand compound having a physiological activity can be predicted with a high probability. In addition, the method of the present invention enables high-speed molecular design of a novel ligand compound having a physiological activity.

[0028]

FIG. 1 shows the concept of the method of the present invention. In the method of the present invention, first, receptor mapping is performed based on the molecular structure of a known ligand by, for example, RECEPS technique, three-dimensional lattice information is obtained, and an inner hole model of a receptor is constructed. For example, according to the LINKOR method, a functional group of a known ligand molecule, which is important for the interaction with the receptor, is extracted, the position and the direction of the functional group are designated, and the position and the direction of the functional group are stored. The molecular structure of a novel ligand can be constructed. O in FIG.
The -H, NH, and benzene rings are functional groups extracted from the molecular structure of known ligands that are important for interaction with the receptor. In addition, a schematically shows an inner pore model of a receptor constructed by receptor mapping using RECEPS based on the molecular structure of a known ligand.

FIG. 2 shows a method of bonding between two functional groups which can be used in the method of the present invention. The method of the invention comprises two coupling methods, which can be selected at the time of use. FIG. 2 shows a method of connecting a functional group A and a functional group B with three atoms a, b, and c. In bonding method 1, atoms a, b, and c are taken on a three-dimensional lattice point. In FIG. 2, dotted lines derived from A and B, and a and b indicate the bond distances, and arrows derived from A and B indicate the directions of the functional groups A and B.

In the bonding method 2, the atoms a, b and c are set not on the three-dimensional lattice points but on the circumference where the bond atoms calculated from the bond distance and bond angle can exist. Yen (a),
(b) and (c) show candidates for the positions of the atoms a, b and c, respectively. FIG. 3 shows a method for forming a ring structure in the present invention. FIG. 3A shows that one carbon atom is placed at the center of two hydrogen atoms in step L30 of the flowchart of FIG.
This carbon atom is bonded to a carbon atom bonded to each of the above hydrogen atoms to evaluate a bond distance, a bond angle, and a dihedral angle, and if this is within an allowable range, a method of forming a ring structure is schematically illustrated. Is shown. FIG. 3 (b) shows FIG.
In step L33 of the flowchart of FIG. 7, a portion where a ring structure is more stable for interatomic contact including a hydrogen atom is searched, and a hydrogen atom is converted into a carbon atom while considering a bond length, a bond angle, and a torsion angle. The method of forming a new ring structure after substitution with an atom is schematically shown.

FIGS. 4 and 5 illustrate the use of LINKOR.
2 shows a flowchart of one embodiment of the method of the present invention. Details will be described later. FIG. 6 shows a flowchart of a molecular superposition method using RECEPS and receptor mapping. FIG. 7 shows the concept of the molecular superposition method using RECEPS and receptor mapping. RECEP
There are two methods for molecular superposition of S, an interactive superposition method and an automatic superposition method. By superimposing the molecular structures of known ligands using an interactive superposition method or an automatic superposition method and optimizing the superposition model in consideration of other properties, three-dimensional lattice point information (shape, Receptor mapping represented by information such as hydrogen bonding, electrostatic potential, and aromaticity) is performed. RE
The CEPS method is characterized by performing molecular superposition in consideration of interaction with a receptor, and performing receptor mapping using three-dimensional lattice information.

FIG. 8 shows the chemical structure of morphine (a), the crystal structure (b), and the position and direction (c) of the extracted functional group (benzene ring and protonated nitrogen). Also, a
Indicates the wall of the receptor. FIG. 9A shows a morphine receptor model represented using three-dimensional lattice points. The space surrounded by the basket represented by the diagram shows the inner hole of the receptor. Further, FIG. 9 (b) shows the position of an anionic or hydrogen-bonding heteroatom expected on the receptor side.

FIG. 10 shows the results obtained by the method of the present invention.
2 shows a part of the molecular structure of a morphine receptor ligand. The molecular structure surrounded by a frame is the same as the molecular skeleton of a compound already known to have analgesic activity. FIG.
The molecular structure of a group of compounds that have the same molecular skeleton as that of the ligand obtained by the method of the present invention and are already known to have analgesic activity equal to or higher than that of morphine is shown.

FIG. 12 shows the structure of a known inhibitor analyzed by X-ray crystallography together with the bacterial protease lysopas pepsin. Pro for proline, Phe for phenylalanine, His for histidine, Val for valine, Ty
r indicates tyrosine, and Ψ [CH ₂ -NH] indicates that the peptide bond C0NH in this portion is CH ₂ NH. In the upper part, the structure of the lysopass pepsin inhibitor is represented by an amino acid sequence. Further, the lower part shows the chemical structure and the three-dimensional structure of the portion surrounded by the frame in the amino acid sequence of the lysopathepsin inhibitor shown in the upper part. In the chemical structure of the inhibitor, five functional groups that directly interact with the substrate binding site of lysopas pepsin are circled and numbered.

FIG. 13 shows the results obtained by the method of the present invention.
The molecular structure of the ligand of lysopass pepsin is shown (output 1).
~ Output 11). FIG. 14 shows a state in which the molecular structure of the ligand of lysopass pepsin (shown as output 9 in FIG. 13) obtained by the method of the present invention is placed in the inner hole containing the substrate binding site of lysopass pepsin. Is shown. In FIG. 14, the thick line represents the molecular structure of the ligand of lysopass pepsin obtained by the method of the present invention, and the thin line represents the structure of the substrate binding site of lysopass pepsin. In addition, ASP79 is 79
ASP218 means aspartic acid at position 218.

The first embodiment of the present invention will be described with reference to the flowcharts of FIGS. FIG.
FIG. 4 shows a flow from inputting coordinates of a known ligand molecule to obtaining three-dimensional lattice point information by the PS method.
And 5 show the flow from the input of the atomic coordinates of the ligand molecule by the LINKOR method and the three-dimensional lattice point information obtained by the RECEPS method to the output of the new ligand molecular structure. 4 and 6, R and L indicate each step.

First, in R1 of FIG. 6, the atomic coordinates of one kind of ligand molecule known to have a physiological activity or two or more kinds of ligand molecules known to have the same physiological activity are input. I do. As the structure of the ligand molecule, in addition to the structure determined by crystal analysis, a similar structure in a crystal database or a structure obtained by molecular modeling using energy calculation can be used.

Next, in R2, for each ligand molecule, the charge of each atom is calculated by molecular orbital calculation,
Further, a hydrogen bonding functional group is given a type number for each functional group. Here, the term “hydrogen-bonding functional group” refers to a functional group containing hydrogen for forming a hydrogen bond or a hetero atom having an isolated atom pair.

Next, in R3, it is determined whether or not there is only one kind of ligand molecule. When there is only one kind of ligand molecule, the step of R11 is performed without performing the steps of R4 to R10 described below. When there are two or more types of ligand molecules in R3, two types of molecules are selected from the ligand molecules in R4, one of which is a template molecule and the other is a trial molecule.

Next, in R5, it is selected whether to perform automatic superposition of the ligand molecules or to perform interactive superposition. Here, superposition of molecules refers to placing the three-dimensional structures of multiple active molecules in corresponding orientations in the same space in order to three-dimensionally search for and compare homology and differences in molecular structures and properties. Say. Focus on what properties and physical quantities (such as the shape of the surface of the molecule, the position of the contained atoms,
The problem is how to evaluate good overlap. For example, in RECEPS, superposition can be performed with emphasis on matching molecular properties such as hydrogen bonding, ionic bonding, and hydrophobic interaction for interacting with a receptor. Molecular superposition may be performed with a molecular model, but it is easy and quantification is possible.
It is preferable to use computer graphics because the coordinates can be stored as they are. The term “automatic superposition of molecules” refers to a method in which a computer automatically finds a superposition method while covering the positional relationship and molecular conformation of superposition. For example, in RECEPS, a special least squares calculation is performed so that the hydrogen bond atoms on the receptor side coincide with each other, assuming that the same receptor binds to the same site. The term "molecule interactive superposition" refers to superposition of molecules on a computer graphics display while rotating, translating, or bonding and rotating while referring to the index of overlap.

In the step of R5, it is generally preferable to select automatic superposition. However, when there are few hydrogen bonding sites in the molecule and it is difficult to use the automatic superposition method, interactive superposition is used. Good to choose. R5
When the automatic superimposition of the ligand molecules is selected in, the automatic superimposition using dummy atoms is performed in R6. A dummy atom is defined as the position of a group (eg, a hydrogen donating group) involved in the interaction of one molecule when there are two molecules forming a hydrogen bond (eg, an acceptor and a ligand). , A virtual atom arranged at a position presumed to be appropriate as a position of a group (for example, a hydrogen accepting group) involved in the interaction of the other molecule. For example, in a receptor-ligand system in which two or more hydrogen bonds are presumed to be involved in the activity of the ligand, they are arranged on the receptor side based on the position and properties of the hydrogen bonding functional group of the template molecule. The mutual relationship between the template molecule and the trial molecule and the relationship between the dummy molecule and the template molecule are set so that the dummy atom matches the dummy atom located on the receptor side at the maximum based on the position and properties of the hydrogen bonding functional group of the trial molecule. A model in which the template molecule and the trial molecule are superimposed can be created while changing the conformation of the trial molecule. All ligand molecules are superimposed by repeating the automatic superimposition with the template molecule in place of the trial molecule (Y. Kat
o, A. Inoue, M. Yamada, N. Tomioka, and A. Itai, Compu
ter-Aided Molecular Design 6, pp.475-486 (1992))
.

If the interactive superposition is selected in R5, then, in R7, a region for forming a three-dimensional lattice point is set for the template molecule. Next, at R8, three-dimensional lattice point information (shape, hydrogen bonding, electrostatic potential, aromaticity, etc.) about the template molecule is calculated and stored. Next, in R9, the trial molecule is superimposed on the template molecule interactively with reference to the index of the coincidence of the physicochemical properties calculated in real time using the three-dimensional lattice point information. By repeating the superposition with the template molecule in place of the trial molecule, all the ligand molecules are superimposed.

Next, in R10, an automatic superposition model of the template molecule and the trial molecule obtained in R6, or R9
Optimize the interactive superposition model of the template molecule and the trial molecule obtained in the above. All of the superimposition models with good superimposition indices obtained in R6 or R9 are re-evaluated by indices (hydrogen bonding, electrostatic potential, shape, aromaticity, etc.) using three-dimensional lattice points, and Get a promising superposition model.

Alternatively, using the result of the superposition obtained in R6 or R9 as an initial structure, the molecular structure is interactively manipulated to improve the superposition model, or the superposition is performed by the Simplex method using three-dimensional lattice data. By optimizing the superposition model, an appropriate superposition model is obtained. next,
In R11, a region for creating a three-dimensional lattice point is set based on the superimposed structure of the ligand molecules obtained in steps R2 to R10 or the structure of a single ligand molecule input in R1.

Next, in R12, three-dimensional lattice point information on all molecules is calculated in the region set by R11. Receptor mapping can be performed based on this three-dimensional grid point information. Here, the receptor mapping means that when there is no experimental information on the three-dimensional structure of the receptor, the environment of the receptor is estimated and expressed from the information of the ligand molecule. Receptor mapping by RECEPS uses three-dimensional lattice points inside and around the molecule, and determines the size, shape, and shape of the inner hole of the receptor (receptor) model based on the structure of the ligand molecule or a plurality of superimposed ligand molecules. Expected sites, electrostatic properties, aromaticity, etc. of the hydrogen bonding functional group are expressed. The space occupied by the three-dimensional lattice points occupied by one or more molecules is assumed to be an inner hole of the receptor. As for the electrostatic potential in the area outside the area, an average value that is weighted by activity at each three-dimensional lattice point is adopted.

Next, at R13, a hydrogen bonding site is set. When there are a plurality of ligand molecules, a hydrogen bonding functional group having high commonality between the superimposed molecules is set as a hydrogen bonding site. When one type of the ligand molecule is used, it is considered important by the user's judgment. The hydrogen bonding functional group is set as a hydrogen bonding site.

Next, in R14, the three-dimensional grid point information is output to a file. Next, in R15, the atomic coordinates of the superposition model of the ligand molecule are output. Next, in L1 of FIG. 4, the atomic coordinates of the known single ligand molecule or the atomic coordinates of the superimposed model of the ligand molecule output in R15 are input.

Next, in L2, the charge of each atom is calculated in advance for each ligand molecule by molecular orbital calculation, and the type number of each functional group is determined by the user for the hydrogen bonding functional group. Give it. Next, in L3, the three-dimensional lattice point information calculated in R12 is input. Next, in L4, a functional group presumed to be essential for activity is extracted from the above-mentioned known ligand molecule or ligand molecule group that interacts with the receptor.

Further, by preparing a plurality of functional groups having the same interaction as fragments of these functional groups, and sequentially substituting these, a new ligand molecule structure can be constructed. Examples of the above-mentioned substitutable functional groups are shown in Table 1 below.

[0050]

[Table 1] Table 1 No. Functional group 1 N (protonated) of N 2 SP ³ of SP ^{2 3} N 4 of SP ³ Aromatic N (protonated) 5 O 6 of aromatic N 6 hydroxy group O 8 of ether O 8 carbonyl group O 9 O 10 phenyl group of carboxylate anion Next, in L5, the position and direction of the functional group extracted in the above step L4 are set.

When the ligand molecule is a single molecule, the position of the functional group in the molecule is used as a site. When the ligand molecule is a plurality of molecules, the position of the corresponding functional group in each of the superimposed molecules is determined. Using the position of the center of gravity as the site,
By setting the fragment structure coordinates of each functional group, the position and direction of the extracted functional group are specified. In addition, if there is a position and a direction in which an equivalent effect is expected, the position and the direction can be sequentially replaced with those.

Next, at L6, the distance between the sites is calculated. Next, in L7, the way of connecting the sites is set. In the case of two sites, only the route connecting the sites may be considered, but in the case of three or more sites, the generated structure differs depending on which sites are connected and in what order. There are three connection methods: a method of connecting linearly, a method of connecting all sites with a core at the center, and a method of connecting all sites in a ring. Although it can be set so as to cover all types of connection, it is usually preferable to select a method of connecting linearly.

For example, when there are three sites A, B, and C, ABC, ABCB, CAB
There are six types of linear connections. These can be tried sequentially, but it is usually preferable to perform only one method of connecting the sites in ascending order of distance. Also, when there are many sites, it is possible to specify that some of the sites are to be connected without connecting all of them.

Next, in L8, one of the ways of connecting the sites
Choose one. For example, a connection method of ABC is selected. Next, in L9, the number of bonds between the sites is set. For example, when the connection method of ABC is selected, the number of inter-site bonds is two, AB and BC.

Next, at L10, two sites to be combined are set. For example, two sites A and B are selected. Next, in L11, an upper limit and a lower limit of the number of atoms involved in the bonding between the sites (hereinafter, referred to as “the number of bonding atoms”) are set. For example, when two sites A and B are selected, the appropriate number of atoms involved in the bond between A and B can be calculated from the distance between A and B according to the following formula.

Lower bound atom number = Truncation after decimal point ｛(A
The shortest distance between -B) / (2.735 / 2)｝-1 The upper limit of the number of bonded atoms = rounded up after the decimal point ｛(the shortest distance between A and B) / (2.006 / 2)｝ The constant 2.735 in the above equation is − C1 {C2-C3}
Is the distance between C1 and C3, and the constant 2.006 is-
It is the distance between C1 and C3 when C1 = C2-C3 =.

Next, in L12, the number of bonding atoms is set. For example, if the upper limit of the number of bond atoms is calculated to be 5 atoms and the lower limit is calculated to be 2 atoms in the above calculation, first, a bond path for connecting the sites with the lower limit bond atom number of 2 atoms is searched, and then 3 The number of bonding atoms is sequentially increased, for example, searching for a bonding path connecting the sites with atoms, and searching is performed up to the upper limit of 5 bonding atoms.

Next, at L13, a bonding atom type is set. As the bonding atom species, combinations of probe atoms shown in Table 2 can be prepared and used sequentially.
If necessary, it is also possible to limit only to carbon atoms or use fragments such as amide and benzene rings.

[0059]

Type [Table 2] Table 2 Probe atom ^{C: SP 3 1 C: SP} 2 2 C: group 3 C: Aromatic ^{49 # 1N: SP 2 9 #} 2N: SP 3 ( protonated) 39 # 3N: SP ³ 8 # 4N: SP ² (aromatic / protonation) 9 (50) # 5N: SP 2 ( aromatic) 9 (50) # 6O: hydroxy groups 6 #. 7O: ether group 6 #. 8O: group 7 # . 9O: carboxylate anion 47 # 10C: phenyl group 51 H: bound to C: SP ² 5 H: bound to C: SP ³ 5 H: N binding: SP ³ 28 H: bonded to C: aromatic 5 H: Bond to O: hydroxy group 21 Next, at L14, the probe atom is determined based on the bond distance and bond angle with the atom in the site (hereinafter referred to as “key atom”) that bonds to the probe atom or another probe atom. Territory where there can exist Set the range roughly (see FIG. 2).

That is, in the bonding method 1, the upper limit and the lower limit of the address of the three-dimensional lattice point where the bonding atom can exist are calculated. In the bonding method 2, a circle in which a bonding atom can exist is calculated. Next, at L15,
The probe atoms are arranged in the region set by L14.
In the combining method 1, one point of the three-dimensional lattice point existing in the region set by L14 is
Probe atoms are generated at regular intervals on the circumference.

Next, in L16, an appropriate one is selected from the probe atoms. Whether or not the probe atom is in the inner hole of the acceptor is determined based on the lattice point information of the nearest three-dimensional lattice point, and those in the spatial region in the inner hole of the acceptor are selected. Furthermore, it is determined whether or not the selected probe atom has a bond distance, bond angle, and torsion angle within a permissible range from an ideal value (standard value) with a key atom or another probe atom in the site, If it is within this certain range, the three-dimensional lattice point or a point on the circumference is adopted as the position of the bonding atom.

Next, in L17, it is determined whether all three-dimensional grid points or points on the circumference in the area set in L14 have been selected. If all three-dimensional lattice points or points on the circumference in the area set in L14 have not been selected, the process returns to L15. The steps from L15 to L16 are repeated for all the three-dimensional lattice points or points on the circumference in the region where the probe atoms can exist, and the adopted all three-dimensional lattice points or points on the circumference are registered. .

Next, in L18, it is determined whether or not all of the bonding atom types prepared in L13 have been selected. If not all have been selected, the process returns to L13, and the steps of L14 to L17 are repeated while changing the bonding atom type. Next, in L19, it is determined whether or not all the bonding atom numbers have been selected in L12. If all the bond atoms are not selected, the process returns to L12, and the steps of L13 to L18 are repeated while changing the number of bond atoms.

Next, at L20 in FIG. 5, a connection route between the sites is generated. For example, by combining method 1, a
When site A and site B are connected by two atoms of and A, a set of three-dimensional lattice points registered for site A
Set b of 3D grid points registered for ¹ and site B
^From all the combinations with ¹ , only the combination in which the coupling distance, the coupling angle, and the torsion angle are appropriate is selected to complete the coupling path.

According to the bonding method 1, a, b and c
If at three atoms connecting site A and site B, for each one of the three-dimensional grid points of the set a ¹ in the three-dimensional grid points that are employed as described above as the position of atom that binds to site A , L14 to L17, a
A set c of three-dimensional lattice points adopted as the positions of the atoms bonded to ¹ is obtained. 3D lattice point included in c and site B
All out of the combinations, bond length and set b ¹ of a three-dimensional lattice points registered as the position of the binding atoms,
The combination path and the torsion angle are selected only in a reasonable combination to complete the connection path.

The same applies to the case where sites A and B are connected by four or more atoms. For example, when site A and site B are connected by 2n atoms, a set of three-dimensional lattice points after the obtained from a ¹ in order to a ^n, a set of three-dimensional grid points from the site B side from b ¹ to b ⁿ in order, a ⁿ
Bond length ^{-b n, a n-1 -a} n -b n and a ⁿ -b
bond angle of ⁿ -b ^n-1 is employed only as a coupling path reasonable.

For example, when site A and site B are connected by two atoms a and b according to the bonding method 2, a
Must be on the circumference of a circle (a) that is inclined by an angle corresponding to the atomic type of A and maintains the bond distance between A-a. All points on the circumference of the circle (a) are candidates for the position of a, as long as they do not overlap with the region where the receptor exists, and probe atoms are assigned to points taken on this circumference every 5 to 10 °. I will put it. The same applies to b coupled to site B. The combination path is completed by selecting only a combination in which the distance between a and b and the angles of A and b and a and b and B are appropriate.

When the site A and the site B are connected by three atoms a, b, and c according to the bonding method 2, the circle (a) and the circle
In the same way that (b) was calculated, from one point a on the circle (a), set the atoms so that the angle of Aac and the distance ac are valid for the types of atoms a and c. Circle where c can exist (c)
Is calculated. The probe atom c on the circle (c) is a reasonable distance c-b from the probe atom b on the circle (b), and a-c-
If we can have an angle of b and an angle of cbb,
Register as a connection path between A and B. The same applies when sites A and B are connected by four or more atoms.

Next, in L21, least-squares superposition between connection paths is performed, and those having a least-square residual of a fixed value or less are removed as similar structures. Since a large number of connection paths searched as described above are searched using three-dimensional lattice points or points at fixed intervals on the circumference, those belonging to the same structure are registered as different paths. Can happen. Therefore, the first binding path obtained from the above steps is used as a template as a reference for judging the similarity between the binding paths, and the second and subsequent binding paths are sequentially superimposed by least squares superimposition. If the square residual is less than a certain value, it is removed, and if it is more than a certain value, it is registered as a template to create a template group. The binding pathways that make up this template group are dissimilar to each other.

Next, at L22, it is determined whether all the two sites to be combined have been selected at L10. If not all are selected, the process returns to L10, another site pair is set, and the steps of L11 to L21 are repeated to register the adopted connection route. Then, L2
In 3, form a new ligand molecule backbone by combining all binding pathways between each site.

Next, in L24, it is determined whether or not all the ways of connecting the sites have been selected in L8. If all of the sites are not selected, the process returns to L8, and another order (for example, ACB) for connecting the sites is selected.
Steps 9 to L23 are repeated. Next, in L25, it is determined whether or not the designation of the position and direction of the functional group performed in L5 is changed to another position and direction in which an equivalent effect is expected without changing the interaction site on the receptor side. When the designation of the position and direction of the functional group is changed, returning to L5, the designation of the position and direction of the functional group set in L5 without changing the interaction site on the receptor side is expected. And the steps of L6 to L24 are repeated. For example, L1
If you enter the atomic coordinates of multiple ligand molecules in,
The position and direction of the functional group of each ligand molecule may be specified.

Next, in L26, it is determined whether or not an operation for changing the type of the functional group extracted in L4 to another functional group capable of interacting with the receptor is performed. When performing this operation, returning to L4, changing the type of the functional group extracted in L4 to another functional group capable of interacting with the receptor, and repeating L5 to L25. Next, at L27,
Output the entire ligand molecule skeleton constructed by the above steps.

Further, at L28, one of the whole ligand molecule skeletons is selected. Next, at L29,
A hydrogen atom is added to the unsatisfied valence atoms in the ligand molecule skeleton by calculating the bond distance and angle. If there are no unsatisfied atoms, this step is skipped. Next, in L30, it is determined under predetermined conditions whether or not it is necessary to form a ring structure in the molecular structure of the ligand, and if necessary, a ring structure is formed.

For example, if the distance between two hydrogen atoms is unusually close (0.6 angstroms), place one carbon atom at the center of the two atoms and replace this carbon atom with each of the hydrogen atoms When the valence angle is within the allowable range, the carbon atom is adopted when the valence angle is within the allowable range, and two hydrogen atoms are removed to form a ring structure (see FIG. 3A). See).

Next, in L31, a structure having a similar ring structure formed from another bonding route is removed. When the ligand molecule forms a ring structure as described above, a substance belonging to essentially the same ring structure may be formed from a different binding path, and may be registered as a different binding path. The structure having a similar ring structure formed from is removed. This step can be performed in the same manner as in L21.

Next, in L32, the degree of freedom is limited so that the position and direction of the functional group extracted in L4 do not change, and the structure optimization calculation of the ligand molecule is performed to remove the molecular structure having a very high energy. I do. Empirical energy calculations can be used to optimize the ligand molecular structure,
For example, the molecular force field calculation program MM2 (NL Alling
er, RA Kok and MRImam, J. Comp. Chem. 9 , 591
-595, (1987)). In order to minimize the energy, the functional group can be fixed by using a conjugate gradient method and excluding the atoms in the functional group set in L4 from the parameters of the energy minimization.

Further, in L33, a portion where a ring structure is more stable for interatomic contact including a hydrogen atom is searched for, and removal of an atom is performed while considering a bond length, a bond angle, and a torsion angle. And / or form a new ring structure by substitution (see FIG. 3 (b)). This process is repeated until no new ring structures are formed. Then, at L34, a similar structure formed from another binding pathway and finally settling on the same ring structure is removed.

This step can be performed in the same manner as in L21. Next, in L35, the obtained ligand molecular structure is narrowed down according to predetermined conditions. In the case of a rigid structure in which the molecular structure of the obtained ligand does not have a single bond that can be rotated, unconstrained structure optimization is performed,
It is possible to remove the deviation of the position and the direction between the initially designated functional groups. Alternatively, both unconstrained and constrained structure optimization calculations may be performed, and those having an energy difference larger than a certain value may be removed.

If the molecular structure of the obtained ligand is not a rigid structure, it is possible to carry out a constrained structure optimization calculation to remove any displacement or misalignment between the functional groups extracted in L4. it can. Next, at L36,
Output the atomic coordinates of the structure of the obtained ligand molecule. Next, in L37, it is determined whether or not all ligand molecule skeletons have been selected in L28. If all the ligand molecule skeletons have not been selected, the process returns to L28, the next ligand molecule skeleton is selected, and the steps from L29 to L36 are repeated.

The method of the present invention was applied to morphine, which has no conformational freedom and also has a functional group essential for activity. Morphine is a natural compound exhibiting strong analgesic activity isolated from poppy flowers, and many analogous compounds have been designed on the basis of its structure and chemically synthesized to examine its analgesic activity. If the molecular structure of those known analgesic active compounds is constructed by the method of the present invention, the effectiveness of the method of the present invention is demonstrated.

FIG. 8 (a) shows the chemical structure of morphine, and FIG. 8 (b) shows the crystal structure of bromate. As a result of various modifications of the molecular structure and investigation of its activity, the essential components of this molecular structure are the benzene ring and the nitrogen atom, both of which have a specific positional relationship in morphine. It turned out to be important. Although it is not clearly known what state of this nitrogen is essential, it is assumed that the protonated state of the tertiary amine is active, and the direction of H bonded to N in the morphine crystal is changed to the direction of the lone pair. The position of the center and the direction of the normal vector are presumed to be essential for the activity of the ligand molecule, assuming that the position of the constituent atoms does not matter as long as the position of the ring current of the π electron coincides with the benzene ring The position and direction of the functional group to be
(C)). On the other hand, by RECEPS, receptor mapping was performed from a single molecule of morphine, and a receptor model was constructed on a three-dimensional lattice point (see FIG. 9).
The molecular structure of the ligand was constructed by OR. The interval between the three-dimensional lattice points was 0.35A. Table 3 shows the calculation conditions used.

[0082]

[Table 3]Table 3 Calculation condition Description ARECEP = RECEPS Creates three-dimensional lattice information from ligand molecules. GRDNAM = morph.crg File name of three-dimensional lattice information NUTMPM = 1 Number of molecules used as template ATMPNM = morph2.d Coordinate file of template molecules NFRAG = 2 Number of functional groups IFRAG = 1 Quaternary amine (functional group 1) Is the atom in the coordinate file. IFRAG = 38 What atom is the benzene ring atom (functional group 2) in the coordinate file? LKORDER = 1 2 Connect functional groups 1 and 2. RSLMT = 0.12 If the least squares residual is less than this, a similar structure is assumed. ANGFIL = angl.parm A file of bond angle parameters for geometry optimization. DISFIL = dist.parm File of bond distance parameters for structural optimization. LNKDIS = 0.38 Permissible range from standard value of bond distance (angstrom) LNKANG = 6.5 Permissible range from standard value of bond angle (unit °) LNKTOT = 8.0 Permissible range from standard value of torsion angle (unit °) RNGCON = 0.7 If the sum is less than 0.7 times the sum of the van der Waals radii, it is determined whether a ring structure can be obtained. USRFLG = 1 10 Only the phenyl group of No. 10 in Table 1 is selected as fragment 1. USRFLG = 2 2 Select only No. 2 quaternary amine in Table 1 as fragment 2 You.

However, in the construction of the molecular structure of the ligand, in the step of L20, a bonding route between sites is generated by the bonding method 1, and the position and direction of the functional group are not reset in the step of L25. No substitution with an equivalent functional group was performed in the step of L26. Furthermore, L3
In step 5, optimization without limitation is performed by using the MNDO method in the MOPAC semi-empirical molecular orbital calculation program, and those in which the positional relationship between the functional groups has largely changed from the set state are excluded. And the molecular structure of the ligand was narrowed down. As a result, about 50 molecular structures were obtained. FIG. 10 shows the molecular structure of a part thereof. These molecular structures include a number of molecular skeletons of compounds known to exhibit the same or higher activity as morphine. FIG. 11 shows the structures of known active compounds having these molecular skeletons. Therefore, it is highly probable that other compounds and derivatives thereof show analgesic activity equal to or higher than that of morphine.

Hereinafter, a second embodiment of the present invention will be described. When the structure of the complex between the receptor and the ligand had been elucidated, the molecular structure of the ligand could be constructed as follows. Instead of the steps of R1 to R15 and L1 to L4 of the first embodiment, based on the known structure of the receptor, the information of the three-dimensional lattice points created in the inner hole of the receptor is calculated, and The same processing steps as in the first embodiment are performed, except that the functional group of the ligand molecule directly interacting with the receptor is extracted and the structural coordinates of the functional group are set.

The above-described method was carried out using the bacterial protease lysopas pepsin and its peptide-like inhibitor (C-terminal) Pro-
Phe-His-Phe Ψ [CH ₂ NH] -Phe-Val-Tyr (N-terminus) (wherein Ψ [CH ₂ NH] means that the peptide bond CONH-
₂ NH is shown. ). First, five functional groups that directly interact with the substrate binding site of the enzyme were extracted from the structure of the inhibitor in the enzyme-substrate complex crystal, and fixed at that position (see FIG. 12). 5 above
It was specified that any three of the extracted functional groups were connected. GREEN (Journal of Computer Ages Molec
ular Design, vol.1, pp.197-210 (1987), Nobuo Tomio
Using ka, Akiko Itai, and Youichi Iitaka), the region where the center of the atom in the ligand molecule enters was represented as three-dimensional lattice point information. When a molecular structure of a novel ligand was constructed in this region, a molecular structure as shown in FIG. 13 was obtained. These molecules are non-peptides,
As shown in FIG. 14, it can be stably located at the same position in the inner pore of lysopas pepsin, as with the peptide-like inhibitors described above. Furthermore, by calculating the strength of the binding of these molecules to the lysopass pepsin enzyme by energy calculation, it was found that they exhibited almost the same interaction (complex stability) as the above peptide-like inhibitors. Was. Therefore,
Compounds having these molecular structures and derivatives thereof are
It is highly likely that it has a strong lysopas pepsin inhibitory activity.

Hereinafter, a third embodiment of the present invention will be described. When the three-dimensional structure of the receptor alone is elucidated and the environment of the drug binding site is known, R1 to R in the first example are used.
15, and calculating the information of the three-dimensional lattice points created in the inner hole of the receptor in place of the steps of L1 to L4, and
The molecular structure of the ligand molecule is constructed by performing the same processing steps as in the first embodiment, except that the user sets the types and coordinates of two or more functional groups capable of interacting with the functional groups present in the receptor. It is possible.

[Brief description of the drawings]

FIG. 1 illustrates the concept of the method of the present invention.

FIG. 2 shows a method of bonding between functional groups in the method of the present invention.

FIG. 3 shows a method for forming a ring structure in the present invention.

FIG. 4 shows a part of a flow chart of one embodiment of the method of the present invention using LINKOR.

FIG. 5 shows a part of a flowchart of one embodiment of the method of the present invention using LINKOR.

FIG. 6 shows a flowchart of a molecular superposition method using RECEPS and receptor mapping.

FIG. 7 shows the concept of a molecular superposition method using RECEPS and receptor mapping.

FIG. 8 shows the molecular structure of morphine, (a) shows the chemical structure of morphine molecule, (b) shows the crystal structure of morphine bromate molecule, and (c) shows the molecular structure of morphine. The position and direction of the extracted functional group are shown.

FIG. 9 shows a morphine receptor model represented using three-dimensional grid points. The space enclosed by the basket represented by the diagram (a) indicates the space assumed to be allowed for the ligand molecule. (B) further indicates the position of the expected anionic or hydrogen-bonded heteroatom on the receptor side.

FIG. 10 shows a part of the molecular structure of a morphine receptor ligand constructed by the method of the present invention.

FIG. 11 shows the molecular structure of a group of compounds known to have analgesic activity equivalent to morphine.

FIG. 12 shows the structure of an inhibitor of lysopas pepsin.

FIG. 13 shows the molecular structure of a lysopas pepsin ligand constructed by the method of the present invention.

FIG. 14 shows a state in which a ligand molecule having a molecular structure constructed according to the method of the present invention is arranged in the inner hole of lysopas pepsin.

Continuation of the front page (56) References 14th Information Chemistry Symposium, 19th Structural Activity Correlation Symposium Abstracts, pp. 316-319 Atsushi Inoue et al. Proceedings of the 15th Symposium on Information Chemistry and the 20th Symposium on Structure-Activity Correlation Pages 258-261 Takaki Kanazawa et al. `` Method for constructing new active structures based on molecular superposition '' (58) ⁷ , DB name) G06F 17/50 638 JICST file (JOIS)

Claims (12)

(57) [Claims] 1. A method for constructing a molecular structure of a ligand having a physiological activity, comprising the steps of: (i) obtaining three-dimensional lattice point information based on the molecular structure of one or more known ligands; A first step of constructing a model of an inner hole, (ii) extracting two or more functional groups presumed to be essential for biological activity from the molecular structure of the known ligand, and extracting the positions and directions of the functional groups. A second step of specifying (iii)
A third step of setting the way of connecting the functional groups extracted in the second step, (iv) a fourth step of specifying the number, type, and existence area of the atoms connecting the two functional groups, (v) receiving A fifth step in which a probe atom is arranged in the inner hole of the body, and the position of the probe atom satisfies a predetermined condition, and is adopted as a position of an atom connecting the two functional groups; (vi) the fourth step By repeating the fifth step for all the atoms specified in the above, a bonding route between the two functional groups is created.
Step (vii) For all combinations of the two functional groups set to be connected in the third step, the fourth step to the sixth step
A seventh step of forming a skeleton of the ligand molecule by combining the binding pathways obtained by repeating the steps, (viii) whether or not there is an atom having an unsatisfied valence in the molecular skeleton of the ligand; Judge
When the atom is present, an eighth step of adding a hydrogen atom to the atom, (ix) under predetermined conditions, determine whether it is necessary to form a ring structure in the molecular structure of the ligand, The above method, comprising a ninth step of forming a ring structure, if necessary, and (x) a tenth step of optimizing the molecular structure of the ligand. Further, the positions and directions of the functional groups specified in the second step are sequentially specified again in positions and directions estimated to have the same effect, and the third step to the tenth step are performed. A method for constructing the molecular structure of the ligand having biological activity according to claim 1, which is repeated. 3. The method according to claim 1, further comprising sequentially substituting the functional groups extracted in the second step with other functional groups having the same physicochemical properties, and repeating the third to tenth steps. A method for constructing a molecular structure of a ligand having a physiological activity. 4. A method for constructing a molecular structure of a ligand having a physiological activity, comprising: (i) information on a three-dimensional lattice point created in an inner hole of a receptor based on a known receptor structure of the ligand; (Ii) extracting two or more functional groups that interact with the receptor from functional groups in a known ligand molecule, and designating the position and direction of the functional groups, iii) a third step of setting the connection between the functional groups extracted in the second step, (iv) the number of atoms connecting the two functional groups,
A fourth step of designating the type and presence area, (v) arranging a probe atom in the inner hole of the receptor and connecting two functional groups when the position of the probe atom satisfies a predetermined condition; A fifth step adopted as the position of the atom, (vi) the fourth step
A sixth step of creating a bond path between the two functional groups by repeating the fifth step for all the atoms specified in the step, (vii) two functional groups set to be connected in the third step For all combinations of
A seventh step of forming a skeleton of the ligand molecule by combining the binding pathways obtained by repeating the steps from the sixth step to the (viii) atom having an unfilled valence in the molecular skeleton of the ligand; It is determined whether or not it is present, and if the atom is present, an eighth step of adding a hydrogen atom to the atom, (ix) it is necessary to form a ring structure in the molecular structure of the ligand under predetermined conditions The method according to claim 1, further comprising: a ninth step of determining whether or not there is, and forming a ring structure if necessary, and (x) a tenth step of optimizing the molecular structure of the ligand. 5. The method further comprises sequentially specifying the position and direction of the functional group specified in the second step in the position and direction estimated to have the same effect, and performing the third step to the tenth step. A method for constructing the molecular structure of the ligand having physiological activity according to claim 4, which is repeated. 6. The method according to claim 4, further comprising sequentially substituting the functional groups extracted in the second step with other functional groups having the same physicochemical properties, and repeating the third to tenth steps. A method for constructing a molecular structure of a ligand having the above-mentioned physiological activity. 7. A method for constructing a molecular structure of a ligand having a physiological activity, comprising: (i) information on a three-dimensional lattice point created in an inner hole of a receptor based on a known receptor structure of the ligand; A second step of selecting two or more functional groups capable of interacting with the receptor, and specifying the position and direction of the functional group, and (iii) selecting the second step. A third step of setting how to connect the functional groups, (iv) a fourth step of specifying the number, type, and existence area of the atoms connecting the two functional groups, and (v) a probe in the inner hole of the receptor. When the atoms are arranged and the position of the probe atom satisfies a predetermined condition, the fifth step is adopted as the position of the atom connecting the two functional groups, (vi) for all the atoms specified in the fourth step Creating a binding pathway between the two functional groups by repeating the fifth step 6 process, (vii) the third
A seventh step of forming a skeleton of the ligand molecule by combining the binding pathways obtained by repeating the fourth to sixth steps for all combinations of two functional groups set to be connected in the step (Viii) determining whether there is no unsatisfied atom in the molecular skeleton of the ligand, and if the atom is present, an eighth step of adding a hydrogen atom to the atom; (ix) according to predetermined conditions, it is determined whether or not it is necessary to form a ring structure in the molecular structure of the ligand, and if necessary, a ninth step of forming a ring structure, and (x) Such a method, comprising a tenth step of optimizing the molecular structure. 8. Further, the positions and directions of the functional groups specified in the second step are sequentially specified again in positions and directions estimated to have the same effect, and the third to tenth steps are performed. A method for constructing the molecular structure of the ligand having biological activity according to claim 7, which is repeated. 9. The method according to claim 7, further comprising sequentially substituting the functional groups extracted in the second step with other functional groups having the same physicochemical properties, and repeating the third to tenth steps. A method for constructing a molecular structure of a ligand having the above-mentioned physiological activity. 10. A method for constructing a molecular structure of a ligand having a biological activity, comprising: (i) the atomic coordinates of one or more known ligand molecules; and a three-dimensional lattice formed in an inner hole of a receptor of the ligand. A first step of inputting point information;
(ii) a second step of extracting two or more functional groups presumed to be essential for physiological activity from the structure of the known ligand molecule, and specifying the position and direction of the functional group, (iii) Second
A third step of setting a connection method between the functional groups extracted in the step,
(iv) a fourth step of designating the number, type, and region of the atoms connecting the two functional groups, (v) arranging a probe atom in the inner hole of the receptor, and the position of the probe atom being a predetermined position. When the condition is satisfied, the fifth step is adopted as the position of the atom connecting the two functional groups. (Vi) By repeating the fifth step for all the atoms specified in the fourth step, The sixth step of creating a binding pathway between groups, (v
ii) For all combinations of the two functional groups set to be connected in the third step, by combining the binding paths obtained by repeating the fourth to sixth steps, the skeleton of the ligand molecule is changed. A seventh step of forming, (viii) determining whether there is an atom whose valence is not satisfied in the molecular skeleton of the ligand, and adding a hydrogen atom to the atom if the atom exists. (Ix) determining, under predetermined conditions, whether it is necessary to form a ring structure in the molecular structure of the ligand, and, if necessary, forming a ring structure, and (ix) a ninth step, x) The method as described above, comprising a tenth step of optimizing the molecular structure of the ligand. 11. Further, the positions and directions of the functional groups designated in the second step are sequentially designated again in positions and directions estimated to have the same effect, and the third to tenth steps are performed. A method for constructing a molecular structure of a ligand having a physiological activity according to claim 10, which is repeated. 12. The method according to claim 10, further comprising sequentially substituting the functional groups extracted in the second step with other functional groups having the same physicochemical properties, and repeating the third to tenth steps. A method for constructing a molecular structure of a ligand having the above-mentioned physiological activity.