JP2001521042A - Interaction of alpha-conotoxin peptide with neuronal nicotinic acetylcholine receptor - Google Patents

Abstract

  (57) [Summary] The present invention relates to derivatives of conopeptides (α-4 / 7 conotoxin peptides) with amino acid substitutions as described herein, while maintaining the basic activity of MII. The present invention also relates to the discovery of the three-dimensional structure of MII and its relationship to its specificity for the α3β2 subunit of the neuronal nicotinic acetylcholine receptor (nAChR). The invention also relates to a computer-based program for the representation of the three-dimensional structure of MI and its peptide analogs, peptidomimetics or non-peptidomimetics. Structural features can be correlated with biological activity to facilitate the design of peptidomimetics that exhibit the same specificity for α-4 / 7 conotoxin peptide analogs and neuronal nAChRs. Such analogs and mimetics are useful as cardiovascular agents and for the treatment and detection of small cell lung cancer (SCLC).

Description

DETAILED DESCRIPTION OF THE INVENTION

The present invention is a grant number GM invested by the National Institute of Health, Bethesda, MD.
-54710, PO1 48677 and MH-53631 with government support. The United States Government has certain rights in the invention.

BACKGROUND OF THE INVENTION [0002] The present invention relates to an alpha conopeptide MII (α-4 / 7 conotoxin peptide) wherein the amino acid sequence is substituted as described herein, while maintaining activity substantially similar to MII. )). The invention is further directed to the discovery of the three-dimensional structure of MII and its relationship to its specificity for the α3β2 subtype of the neuronal nicotinic acetylcholine receptor (nAChR). The discovery of the three-dimensional structure allows for the design of peptidomimetics that exhibit the same specificity for α-4 / 7 conotoxin peptide analogs and neuronal nAChRs. Such analogs and mimetics are useful as cardiovascular agents and for the treatment and detection of small cell lung cancer (SCLC). The present invention relates to a computer-based program for displaying the three-dimensional structure of MII and peptide analogs, peptide mimetics or their non-peptide mimetics.

The publications and other materials used herein to exemplify the background of the invention and, in special cases, to provide further details regarding the practice, are incorporated herein by reference. For convenience, they are referenced numerically in the text below and are categorized in the attached source list.

The α-conopeptide MII has the amino acid sequence: Gly-Cys-Cys-Ser-Asn-Pro-Val-Cys-H
It has is-Leu-Glu-His-Ser-Asn-Leu-Cys (SEQ ID NO: 1). This peptide is
It has two disulfide bonds between the first and third cysteine residues and between the second and fourth cysteine residues. The C-terminus may contain a carboxyl or amide group, preferably an amide group. The amino acid at position 6 may be proline or 4-trans-hydroxyproline, preferably proline. The identification of MII is described in U.S. Patent No. 5,514,774, which is hereby incorporated by reference.

[0005] A major sign of neurobiology lies in elucidating the function of several forms of nicotinic receptors and ion channels that have recently been discovered in the nervous system by molecular techniques. The use of gene knock-out organisms is one way to study such functions. Complementary approa
ch) is to use a ligand that specifically inhibits a particular molecular form of the receptor or ion channel. Such ligands should be able to distinguish closely related members of the receptor lineage.

[0006] Nicotinic acetylcholine receptor (nAChR) is a ligand-gated ion channel that is an important component of the nervous system. The typical role of these receptors is defined at the neuromuscular junction, and nicotinic receptors enriched in muscle endplates are important high-level detectors for detecting neurotransmitter release from presynaptic terminals of motor axons. Work as a molecule. However, in addition to these skeletal muscle nicotinic receptors, there are many other molecular forms of nicotinic receptors; these are generally referred to as neuronal nicotinic receptors.
In the central nervous system (CNS), nAChRs play an important role in regulating the release of neurotransmitters, including dopamine, norepinephrine, acetylcholine, GABA and glutamate.

[0007] Fish-hunting cone snail, Con
The small peptide toxin (conotoxin) from the venom of C. us magus is the natural source of ligands that distinguish between closely related molecular forms of a single receptor or ion channel series. Alpha conopeptide MII is a small peptide that selectively acts on neuronal nicotinic receptors. In the autonomic nervous system, MII has recently been used to help pharmacologically scrutinized nAChRs that mediate synaptic transmission in parasympathetic ciliary ganglia. In the frog sympathetic ganglia, MII distinguished nAChRs in B vs. C nerves. In the CNS, the specificity of MII allows the identification of nAChR subunits that regulate nicotine-stimulated dopamine release. In the retina, a selective block of MII was used to confirm the role of nAChR in visual circuit development. The detection of the presence and location of small cell lung cancer (SCLC) tumors, treatment of patients with SCLC and the use of MII to inhibit the growth of SCLS tumors is described in US Pat. No. 5,595,972, Explicitly included herein. The use of MII to treat disorders caused by nicotine-stimulated dopamine release is described in US Pat. No. 5,780,433, which is hereby incorporated by reference. The use of MII as a cardiovascular agent was described on Nov. 18, 1996.
Co-pending Provisional Application No. 60 / 031,141, filed on the date of which is hereby incorporated by reference. To date, neither the report of the three-dimensional structure of MII nor the systematic study of the relationship between the three-dimensional structure and function of the molecule has been performed, nor has the study been essential for the systematic design of MII analogs and mimetics.

[0008] It is desirable to identify derivatives of MII, peptide analogs and peptidomimetics of α-conotoxin, which are selective for the specific subtype of nAChRs and described herein. Having use.

SUMMARY OF THE INVENTION [0009] The present invention relates to an alpha conopeptide MII (α-4 / 7 conotoxin peptide) having the amino acid sequence substituted as described herein, while maintaining activity substantially similar to MII. )). More specifically, the present invention provides a compound represented by the general formula (SEQ ID NO: 2):

[0010]

Embedded image Xaa-Cys-Cys-Xaa-Xaa ₁ -Xaa ₂ -Xaa-Cys-Xaa ₃ -Xaa-Xaa ₄ -Xaa ₅ -Xaa-Xaa-Xaa-Xaa-Cys

Wherein Xaa represents an amino acid selected from the group consisting of natural, modified or unnatural amino acids. The modification may be by the addition, substitution or removal of one or more amino acids. The modification may be due to the addition or substitution of an analog of the amino acid itself, such as a peptidomimetic, an amino acid with an altered side chain, or a mimetic of an organic molecule having the same spatial / structural / functional relationship. You may. Amino either individual Xaa _1, the preferred properly be Asn or His; Xaa ₂ is any of one amino acid, preferably be Pro or hydroxy -Pro; Xaa ₃ is any amino acid, preferably His or Asn Xaa ₄ is any amino acid, preferably Glu; and Xaa ₅ is any amino acid, preferably His or Asn. Most preferably, Xaa ₁ is Asn; Xaa ₂ is Pro or hydroxy-Pro; Xaa ₅ is His. The C-terminus can include a carboxyl or amide group, preferably an amide group.

A second aspect of the invention is directed to the discovery of the three-dimensional structure of MII and its relationship to its specificity for the α3β2 subtype of neuronal nicotinic acetylcholine receptor (nAChR). Based on the correlation of the structure to the biological activity, the discovery of the three-dimensional structure was based on the α-4 / 7 conotoxin peptide analog, neuronal nAC
It allows the design of peptidomimetics and non-peptidomimetics that exhibit specificity for various subtypes of hR. In one embodiment, compounds are developed that exhibit higher specificity for the α3β2 subtype than for the α3β4 subtype of neuronal nAChRs, as evidenced by MII. In a second embodiment, compounds are developed that exhibit greater specificity for the α3β4 subtype than for the α3β2 subtype of neuronal nAChRs, as evidenced by some of the MII derivatives described herein. . The derivatives, peptide analogs, peptidomimetics and non-peptidomimetics of the present invention are useful as cardiovascular agents, gastric motility agents, urinary incontinence agents, smoking suppressants, and for the treatment and detection of small cell lung cancer (SCLC). .

[0013] A third aspect of the invention is directed to the synthesis of mimetics, such as organic molecules based on the three-dimensional structure of MII that exhibit the same or different specificities for neuronal nAChRs.

A fourth aspect of the present invention relates to the identification of a peptide fragment that defines the activity and specificity of said conopeptide MII and a peptide analog with altered nicotinic AChR subtype specificity based on its three-dimensional structure and Directed to the synthesis of peptide mimetics.

[0015] A fifth aspect of the invention is directed to the synthesis of mimetics, such as organic molecules based on overall structure activity information on MII.

A sixth aspect of the present invention relates to a computer program for expressing the three-dimensional structure (such as a visual display) of MII or a peptide analog or peptidomimetic thereof, and furthermore, the identity of each structure of MII and its structure Also provided is a computer program for displaying the position in the overall structure of the device down to the atomic level. There are many currently available computer programs for displaying the three-dimensional structure of a molecule. Generally, these programs include means for displaying coordinates (such as a visual display), means for changing the coordinates, and an image of the molecule having the changed coordinates in order to input the coordinates of the three-dimensional structure of the molecule. Provide a means to display. In a further aspect, as described herein, MII in three-dimensional space
The NMR coordinates of the atomic positions of the molecule are programmed into the program, wherein the coordinates have been obtained from NMR analysis of the MII molecule and relative protein modeling of MII or portions thereof can be performed. Accordingly, there is also provided a computer program for displaying the three-dimensional structure of a peptide analog or peptidomimetic of MII. Preference is given to Molecula using the coordinates described in FIG.
r Computer program C AVEAT available from Simulations, Inc. (Waltham, MA).

Outline of Sequence Listing SEQ ID NO: 1 is the amino acid sequence of α-CT _x MII peptide. SEQ ID NO: 2 is the amino acid sequence of alpha conopeptide MII. SEQ ID NO: 3 is the amino acid sequence of the MII FANT chimera.

DETAILED DESCRIPTION OF THE INVENTION Derivatives of MII The present invention relates to derivatives of MII, α-4 / 7 conotoxin peptides, wherein the amino acid residues are substituted as described herein, while maintaining the basic activity of MII. More specifically, the present invention provides a compound represented by the general formula (SEQ ID NO: 2):

[0019]

Embedded image Xaa-Cys-Cys-Xaa-Xaa ₁ -Xaa ₂ -Xaa-Cys-Xaa ₃ -Xaa-Xaa ₄ -Xaa ₅ -Xaa-Xaa-Xaa-Cys

Wherein Xaa represents an amino acid selected from the group consisting of natural, modified or unnatural amino acids. The modification may be by the addition, substitution or removal of one or more amino acids. The modification is a peptidomimetic,
It may be due to the addition or substitution of an amino acid with an altered side chain, or an analog of the amino acid itself, such as, for example, a mimetic of an organic molecule having the same space / structure / function relationship. Xaa ₁ is any amino acid, preferably Asn or His; Xaa ₂ is any amino acid, preferably Pro or hydroxy-Pro; Xaa ₃ is any amino acid, preferably His or Asn. Xaa ₄ is any amino acid, preferably Glu; and Xaa ₅ is any amino acid, preferably His or Asn. Most preferably, Xaa ₁ is Asn; Xaa ₂ is Pro or hydroxy-Pro; Xaa ₅ is His. The derivative contains two disulfide bonds between the first and third cysteine residues and between the second and fourth cysteine residues. The C-terminus may include a carboxyl or amide group, preferably an amide group. The amino acids are α-amino acids, which include natural amino acids, including unusual amino acids such as γ-carboxyglutamic acid, as well as modified or unnatural amino acids, such as those described in [Robert et al. (1983)]. .

The derivatives of the present invention are highly selective for any of the α3β2 and α3β4 subtypes of neuronal nAChRs present in the autonomic and central nervous systems. Thus, they are useful as cardiovascular agents, gastric motility agents, urinary incontinence agents and smoking suppressants, and are useful against SCLC. The conotoxins can be made by recombinant DNA techniques well known in the art. Also, these conopeptide derivatives are small enough to be chemically synthesized. The general chemical syntheses for preparing the conopeptide derivatives are described below, along with specific chemical syntheses of the conopeptide derivatives and an indication of the biological activity of these compounds. The conopeptides of the invention can be synthesized and / or can be substantially pure. "Substantially pure" means that the peptide is present in the substantial absence of other biological molecules of the same type; preferably in an amount of at least about 85% pure, and preferably at least about 85% pure. It is meant to be present in an amount of 95%.

II. Preparation of Recombinants and Chemically Synthetic Derivatives The conopeptides of the invention can be made by recombinant DNA techniques well known in the art, such as, for example, those described in [Sambrook el al. (1979)]. The peptide produced by this method is isolated, reduced if necessary, and oxidized, if present in the final molecule, to form the correct disulfide bond.

One method of forming disulfide bonds in the conopeptides of the present invention is to air oxidize the linear peptide for extended periods at cold or room temperature. This method produces substantial amounts of bioactive disulfide-linked peptides. The oxidized peptide is fractionated using reverse-phase high performance liquid chromatography (HPLC) or the like to separate peptides having different linkage configurations. Then, by comparing these fractions in the effluent of the natural material, or by using simple assays, the specific fraction with the correct binding with the greatest biological potency is easily determined.
Since it has been shown that cross-linking and / or rearrangement occurring in vivo produces biologically potent conopeptide molecules, the linear peptide or oxidation product having one or more fractions may be converted to It can also be seen that it can often be used for in vivo administration. However, somewhat higher doses may be required due to dilution due to the presence of other fractions of lower biopotency.

Many, or even all, of these peptides could be prepared using recombinant DNA technology. However, if the peptides are not prepared as well, they may be chemically synthesized by any suitable method, such as by a full solid phase method, a partial solid phase method, a fragment condensation method, or a typical solution combination. In conventional liquid phase peptide synthesis, the peptide chain can be prepared by a series of coupling reactions that add constituent amino acids to a growing peptide chain of the desired sequence. Various coupling agents (e.g., dicyclohexylcarbodiimide or diisopropylcarbonyldiimidazole), various active esters (e.g., N-hydroxyphthalimide or N-hydroxysuccinimide), and the use of various cleavage agents, are well-known typical peptide methodology. A typical solution synthesis is
Dissertation [Methoden der Organischen Chemie (Houben-Weyl): Synthese von Peptid
en, (1974)]. The technology for complete solid phase synthesis is described in the textbook [Solid-Phase
Peptide Synthesis, (Stewart and Young, 1969)] and exemplified by the disclosure of U.S. Patent No. 4,105,603 (Vale et al., 1978). Synthetic fragment condensation methods are exemplified in US Pat. No. 3,972,859 (1976). Other possible syntheses are described in U.S. Pat. Nos. 3,842,067 (1974) and 3,862,925 (1975).
). The synthesis of peptides containing γ-carboxyglutamic acid residues is exemplified by Rivier et al. (1987), Nishiuchi et al. (1993) and Zhou et al. (1996). What is common to such chemical syntheses is the protection of labile side groups at various amino acid sites that prevent chemical reactions at the site from occurring with appropriate protecting groups until the protecting groups are completely removed. is there. Normally, the a-amino group on an amino acid or fragment is protected while the entity is reacting at the carboxyl terminus, and then the a-amino group is selectively removed to allow the next reaction to proceed at that position. Is to make it happen. Thus, as a step of such synthesis, each of the amino acid residues located in the desired sequence of the peptide chain with the appropriate side-chain protecting group attached to a variety of residues having an labile side chain. It is common to produce an intermediate compound containing. As long as the choice of the side chain amino protecting group is taken into account, one generally chooses one that is not removed during the deprotection of the a-amino group during the synthesis. However, for some amino acids (eg, His), protection is not required. In selecting a particular side-chain protecting group to be used in the synthesis of the peptide, the following general rules: (a)
The protecting group retains its protective properties and does not detach under coupling conditions, (b
) The protecting group is stable under the reaction conditions selected for removal of the a-amino protecting group at each step of the synthesis, and (c) the side chain protecting group is undesired. Must be removable upon completion of the synthesis containing the desired amino acid sequence under reaction conditions that do not otherwise alter the peptide chain.

[0025] While liquid phase synthesis can be used, peptides can be used in Merrifield.
) Although preferably prepared using solid phase synthesis, other equivalent chemical syntheses known in the art may also be used as described above. Solid phase synthesis begins at the C-terminus of the peptide by coupling the protected a-amino acid to a suitable resin. Such starting materials can be converted to a-amino protected amino acids by ester linkage to chloroemethylated resin or hydroxyl resin, or by amide linkage to benzhydrylamine (BHA) resin or paramethylbenzhydrylamine (MBHA) resin. It can be prepared by contact. The preparation of the hydroxymethyl resin is described by Bodansky et al. (1966). Chloromethylated resin is available from Bio Rad Labora
Commercially available from tories (Richmond, CA) and Lab. Systems, Inc. The preparation of such resins is described by Stewart and Yound (1969). BHA and MBHA carriers are commercially available and are commonly used when the desired polypeptide to be synthesized has an unsubstituted amide at the C-terminus. Either comb, solid resin support has the formula -O-CH ₂ - resin support, -NH BHA resin support, or
Any of those known in the art such as those possessed by -NH MBHA resin carriers may be used. If an unsubstituted amide is desired, the use of a BHA or MBHA resin is preferred since cleavage gives the amide directly. If N-methylamide is desired, it can be formed from N-methyl BHA resin. If other substituted amides are desired, the teachings of U.S. Pat. No. 4,569,967 (Kornreich et al., 1986) may be used, and if further groups other than the free acid are desired at the C-terminus, Houben-Weyl Textbook (197
It may be preferable to synthesize the peptide using the typical method described in 4). C-protected by Boc or Fmoc or by a side-chain protecting group
If the terminal amino acid is to be synthesized with a peptide having a free acid at the C-terminus, if appropriate, stir at 60 ° C. for 24 hours with KF in DMF, Horiki et al. (1978)
First, coupling to a chloromethylated resin is performed according to the procedure described in (1). Following coupling of the BOC-protected amino acid to the resin support, the a-amino protected amino acid is removed by using trifluoroacetic acid (TFA) in methylene chloride (CH ₂ Cl ₂ ) or TFA alone. I do. The deprotection is performed at a temperature between about 0 ° C. and room temperature. Other standard cleavage reagents, such as HCl in dioxane, specific a
-Amino protecting group removal conditions can be used as described in Schroder and Lubke (1965).

After removal of the a-amino protecting group, the remaining a-amino- and side-chain protected amino acids are coupled stepwise in the desired order to give an intermediate compound as defined above, or
As an alternative to adding each amino acid separately to the synthesis, some of them can be coupled to each other prior to addition to the solid phase reactor. Selection of the appropriate coupling reagent is within the skill of the art. Particularly suitable as coupling reagents are N, N'-dicyclohexylcarbodiimide (DCC, DIC, HBYU, HATU, TBTU in the presence of HoBt or HoAt).

The activating reagent used for the solid phase synthesis of the peptide is well known in the peptide field. Examples of suitable activating reagents are carbodiimides such as N, N'-diisopropylcarbodiimide and N-ethyl-N '-(3-dimethylaminopropyl) carbodiimide. Other activating reagents and their use in peptide coupling are described in Schroder and Lubke (1965), and Kapoor (1970).

[0028] Each protected amino acid or amino acid sequence is introduced into the solid-phase reactor in an excess of about 2-fold or more, and is introduced in a dimethylformamide (DMF): CH ₂ Cl ₂ (1: 1) medium, or in a single DMF or The coupling can be performed in CH ₂ Cl ₂ . If an intermediate coupling occurs, the coupling procedure is repeated prior to removal of the a-amino protecting group prior to coupling of the next amino acid. The success of the coupling reaction at each stage of the synthesis, if performed manually, is preferably monitored by the ninhydrin reaction, as described in Kaiser et al. (1970). The coupling reaction can be performed automatically, as with a Beckman 990 automated synthesizer, using a program such as that reported by Rivier et al. (1978).

After completion of the desired amino acid sequence, the intermediate peptide can be removed from the resin carrier by treatment with a reagent such as liquid hydrogen fluoride or (if using Fmoc chemistry) or TFA, The treatment not only cleaves the peptide from the resin, but also removes any remaining side-chain protecting groups and the N-terminal a- The amino protecting group is also cleaved. M
If et is present in the sequence, preferably, the peptide is excised from the resin with HF to eliminate possible S-alkylation with trifluoroacetic acid (
The Boc protecting group is first removed using TFA) / ethanedithiol. If hydrogen fluoride or TFA was used for cleavage, one or more scavengers such as anisole, cresol, dimethyl sulfide and methyl ethyl sulfide are included in the reactor.

Preferably, cyclization of a linear peptide is effected to create a bond between Cys residues, as opposed to cyclizing the peptide while still being part of the peptide-resin. To affect such disulfide bonds, the fully protected peptide is cleaved from the hydroxymethylated resin or chloromethylated resin carrier by ammonia hydrolysis, as is well known in the art, to form a fully protected amide intermediate. Which can then be cyclized appropriately and deprotected. Alternatively, deprotection and cleavage of the peptide from the above resin, or benzhydrylamine resin (BHA) or methylbenzhydrylamine (MBHA) resin, can be performed with hydrofluoric acid (HF) or TFA at 0 ° C. The oxidation can then be performed as described above.

The conopeptide derivatives of the present invention may possess a different biological activity than MII, or may block the native α3β2-containing nAChRs and, with low potency, α3β4-containing nA
The same or a different degree of the same biological activity as MII, which is known to block ChR (US Pat. No. 5,780,433, incorporated herein by reference);
As well as other activities described herein

III. Three-dimensional structure of MII Overview Another aspect of the present invention is directed to the discovery of the three-dimensional structure of MII and the relationship between its specificity for the α3β2 subtype of the neuronal nicotinic acetylcholine receptor (nAChR) and its structure. The discovery of the three-dimensional structure, neuron nA
It allows the design of α-conotoxin peptide analogs and peptidomimetics that will show the same specificity for ChRs. Such analogs and peptidomimetics are useful as cardiovascular agents and for small cell lung cancer (SCLC)
Useful for the treatment or detection of

MII has the following amino acid sequence (SEQ ID NO: 1):

[0034]

Embedded image Gly-Cys-Cys-Ser-Asn-Pro-Val-Cys-His-Leu-Glu-His-Ser-Asn-Leu-Cys

Has the following formula: The C-terminus may contain a carboxyl or amide group, preferably an amide group. The amino acid at position 6 may be proline or 4-trans-hydroxyproline, preferably proline.

The identification of MII is described in US Pat. No. 5,514,774 and US Pat. No. 5,780,43.
No. 3 and incorporated herein by reference. (A) Small cell lung cancer (SC
Treating patients with (LS), (b) inhibiting SCLC proliferation, (c)
) Detecting the presence of SCLC tumors, and (d) the use of MII to detect the location of the SCLC tumors are described in US Pat. include. The use of MII to treat the CNS is described in US Pat. No. 5,780,433, which is hereby incorporated by reference. The use of MII as a cardiovascular agent is described in U.S. application Ser.
No. 031,141, incorporated herein by reference. B. Determination of peptide three-dimensional structure using NMR

Different techniques give different, complementary information about the protein structure. Direct determination of amino acids from the protein or the corresponding gene or c
Primary structures are obtained by biochemical methods, either from the nucleic acid sequence of DNA. NMR methods can be used to obtain secondary and tertiary structures that require detailed information about the arrangement of atoms inside a protein. NMR utilizes the magnetic properties of nuclei. Certain nuclei, such as ¹ H, ¹³ C, ¹⁵ N and ³¹ P, have a magnetic moment or spin. The chemical environment of the nucleus can be probed by nuclear magnetic resonance (NMR), and this technique can be used to obtain information about interatomic distances in the molecule. These distances can then be used to derive a three-dimensional model of the molecule. Most protein molecular structure determinations by NMR have used ¹ H spins. This is because hydrogen atoms are abundant in proteins.

When protein molecules are placed in a strong magnetic field, the spins of their hydrogen atoms align along the magnetic field. By applying a radio frequency (RF) pulse to the sample, this equilibrium alignment can be changed to an excited state. When the nuclei of the protein molecules return to their equilibrium, they emit measurable RF radiation. The exact frequency of radiation emitted from each nucleus depends on the molecular environment of the nucleus, and will differ for each atom unless they are chemically equivalent and have the same molecular environment. These various frequencies are obtained in comparison to a reference signal and are called chemical shifts. The nature, duration and combination of the applied RF pulse can vary greatly, and various molecular properties of the sample can be probed by selecting the appropriate combination of pulses.

In principle, a unique signal (eg, a proton on the CH ₃ side chain of an alanine residue) for each hydrogen atom in a protein molecule except for those that are chemically equivalent (eg,
Chemical shift). In practice, however, such one-dimensional NMR spectra of protein molecules contain duplicate signals from many hydrogen atoms. This is because the differences in chemical shifts are often smaller than the resolution capabilities of the device. This problem has been solved by designing experimental conditions for obtaining a two-dimensional NMR spectrum, and the results are plotted in a figure. The diagonal lines in the figure correspond to normal one-dimensional NMR spectra. Peaks off the diagonal are due to interactions between hydrogen atoms that are spatially close together. By varying the nature of the applied RF pulse, these off-diagonal peaks may reveal different types of interactions.

A two-dimensional experiment is a separate component: a preparation period; an evolution period that “labels” spins when processing in the xy plane according to their chemical shifts; during which correlations with other spins are created And a detection period during which free induction decay is recorded. The experiments are distinguished by the nature of the correlation probed during the mixing period.
COSY (correlation spectroscopy) experiments, for example, include nitrogen and C within the same amino acid residue.
It gives peaks between hydrogen atoms covalently bonded to one or more atoms, such as hydrogen atoms bonded to α atoms. For amino acids with a single ^Hβ proton, standard double quantum filtered COSY (DOQ-COSY) recorded in D ₂ O can be used. However, for amino acids having a β-methylene group, E. FIG. Improved COSY, such as COSY, is more appropriate. On the other hand, nuclear Overhauser effect spectroscopy (NOESY) gives peaks between pairs of hydrogen atoms that are spatially close together, even for amino acid residues that are quite far apart in the primary sequence. In total correlation spectroscopy (TOCSY), one observes whether or not all protons sharing a coupling partner with each other are directly linked to each other.

The two-dimensional NMR spectrum of a protein is interpreted by the method of chain assignment. Clearly, a two-dimensional NOE spectrum contains three-dimensional information about the protein molecule by specifying which groups are spatially close to each other. However, it is by no means trivial to assign the observed peak in the spectrum to a hydrogen atom at a particular residue along the polypeptide chain. This is because the order of the peaks along the diagonal is not a simple relationship to the order of the amino acids along the polypeptide chain. The problem is that the chain assignment method was invented by ETH, Zurich
Kurt Wuthrich's lab has been settled in principle. Chain assignment is based on the number of hydrogen atoms at different amino acid residues and their covalent differences. Each type of amino acid has a specific combination of cross-peaks, COSY
It has a specific set of covalently bonded hydrogen atoms that will give a "fingerprint" in the spectrum. From the COSY spectrum, it is therefore possible to identify the H atom belonging to each amino acid residue and, in addition, determine the nature of the side chain of that residue. However, the order of these "fingerprints" along the diagonal is not related to the amino acid sequence of the protein.

However, linkage specificity assignments can consist of NOE spectra that record signals from H atoms that are spatially close together. In addition to interactions between distant H atoms in the sequence, these spectra also record interactions between H atoms from sequencely adjacent residues. Therefore, these signals in the NOE spectrum are, in principle,
It is possible to determine which fingerprints in the COSY spectrum are attributable to residues adjacent to previously identified residues. In practice, it is difficult to make unique assignment decisions for fragments longer than di- or tri-peptides. This is because NOE signals occur not only between residues that are spatially close to each other but also those that are far apart in the sequence. Thus, a peptide segment uniquely identified by NMR usually matches the corresponding segment in the independently determined amino acid sequence of the protein. NMR spectroscopy identifies H atoms in a protein that are spatially close together, and then uses this information to indirectly derive a three-dimensional model of the protein.

[0043] Distance constraints can be used to derive possible structures of protein molecules. The result of the sequence specificity assignment of the NMR signal is preferably an interactive computer graphic, which is a list of distance constraints from a particular hydrogen atom in one residue to a hydrogen atom in another residue. The list contains such a majority of distances and is usually divided into three distances in the range of 1.8 ° to 5 ° depending on the NOE peak intensity. This list immediately identifies the secondary structural elements of the peptide or protein molecule. This is because both the α-helix and β-sheet have a very specific set of interactions between their hydrogen atoms of less than 5 °. It is also possible to derive a model of the three-dimensional structure of the molecule. To obtain a more qualitative picture of the conformation of the peptide and to generate a three-dimensional structure that is consistent with the NMR data, the principles of simulated annealing [Nilges, M. (1988)] can be employed. The simulated annealing calculation is performed using predetermined experimental data (for example, NOE-
Perform a global search for all possible three-dimensional conformations that satisfy (lead distance and dihedral angle) and calculate the root mean square (RMS) of the deviations (described herein).
) Is used as a measure of the degree of convergence of those calculated structures.

C. I also Ah small peptides consisting described example alpha-CT _x MII slightly 16 amino acid residues of MII structure, it has a well defined three-dimensional solution structure. In addition to the two disulfide bridges that form a hydrophobic Cys bond, there are three helical regions in the structure that contribute to the formation of a very rigid three-dimensional conformation. The presence of such stable secondary structures and disulfide bridges makes multidimensional NMR effective in obtaining very high resolution of the peptide structure. The amino-terminal region has an almost complete turn of the α-helix (Cys ^{2 to} Ser ⁴ ), followed by φ = −89 ° (measured from the minimum energy structure) and ψ = + 132.
Followed by Asn ⁵ with a skeletal dihedral angle of ゜, thereby essentially forming a 90 ° fold. Helix substantially with two folding (Pro ⁶ ~Glu ¹¹⁾ is a major secondary structure composition of the peptide. This helix is His ¹² (φ = −
136 °, ψ = + 77 °; measured with minimum energy structure. End with), it is coordinated to the remaining C- terminal twisting 3 ₁₀ helix and (Ser ¹³ ~Cys ¹⁶⁾ in the N- terminal direction. These two folding related to Asn ⁵ and His ¹³ are probably an important residues for the peptide entire folding, the presence of such folding are shown in FIG. 4 (others have a negative shift Is a positive shift)
Further supported by the C _α proton shift data and the ³ J _{NH-C α} coupling constant of FIG. 2 (> 8.0 Hz coupling constant, while others have a coupling constant of <5.0 Hz). You. Possible functions of Asn ⁵ is that induces folding between N- terminal segment and the major alpha-helix; This feature is from various globular protein structure 45 215 alpha-into This is consistent with the research by Richardson and Richardson [Richardson, JS (1988)] on Rix. They found that for the α-helix, 3.5 per Asn at the N-cap position.
A prominent preference of 2.6: 1 was reported for Pro at 1: 1 and N-cap + 1 positions (Helix Initiator). Regarding the second turn around His ¹³
The function of this would be to carry Cys ³ and Cys ¹⁶ close together and form a second disulfide bridge.

The space-filling model of α-CTx MII shown in FIG. 7B is aimed at investigating the surface distribution of the hydrophobic and hydrophilic side groups of the molecule. The hydrophobic residues are colored purple to distinguish them from yellow polar residues or red (positive) and blue (negative) charged residues. The flat surface located on top of the purple molecule is a distinct structural feature that represents a cluster of hydrophobic residues exposed to the solvent. This hydrophobic surface has Gly ¹ , C (excluding the N-terminal amino group).
ys ^2, is largely formed by a resign Fido bridge between Cys ^3, Leu ^15, Cys ¹⁶ and Cys ³ and Cys ^16. This flat surface may be important for ligand-receptor binding by hydrophobic interactions. There is another very distinct surface consisting of hydrophilic residues that are completely both polar and charged. Left of the model, red, blue and yellow clusters represent folded area of ^{His 2 (Glu 11 ~Asn 14)} . This highly charged surface, almost perpendicular to the hydrophobic surface, will be responsible for its initial recognition by the nAChR based on long-range electrostatic interactions. This potential receptor-binding interface consists of consecutive residues Glu ¹¹ , His ¹² , Ser ¹³ and A
consisting of sn ^14.

In view of the above analysis and the following examples, the following conceptually important structural features unique to nCAhRs that target the α-4 / 7 conotoxin are summarized below: (1) First and third There are two disulfide bridges between cysteines and between the second and fourth cysteines. Several isoforms generated at various disulfide pairs were tested for their activity on nAChRs and showed that they did not retain the native activity. (2) exactly at position i + 3 (i is the second cysteine residue),
There is a central α-helix consisting of 6 residues, ending at position (j is the third cysteine residue). This α-helix is essential for the whole backbone conformation, in particular its length (6 amino acids of MII) and helical pitch are nAChR
Is believed to be an important feature for targeting NMR data for the α-4 / 7 conotoxin MII mutant indicate that the mutation further extends the central helix beyond the j + 3 position to the C-terminus, thereby altering its helical pitch. . This structural change is associated with a loss of activity on the order of three orders of magnitude. (3) The two folds at both the N and C termini (N- and C-cap) of the central helix are also unique structural features found in this conotoxin.
They are closely linked to the central helix and write down the three-dimensional folding motif, their position is i + 2 for N-cap, where i is the second cysteine residue and C-cap Is j + 4, where j is the third cysteine residue. Based on comparison of the primary sequence among the sequenced α-4 / 7 conotoxins, there appears to be some preference at both the N- and C-termini.
There is asparagine, a good hydrogen bonding partner, and histidine and tyrosine, sterically bulky aromatic residues. For example, Richardson, JS et al. (19
From the 215 α-helix subjects of the study according to 88), there is a 3.5: 1 preference for Asn at the N-cap position. Mutations at these positions in conotoxins are 3
It has been found that orders of magnitude cause a loss of their function. (4) The proline residue at position i + 3 (i is the second cysteine residue) is conserved in all α-4 / 7 conotoxins. Its structural role is, generally speaking,
Due to its sterically hindered ring structure, it appears to be in preventing the continuation of the central helix towards the amino terminus. (5) Analysis of MII binding kinetics suggests that this peptide has two distinct binding surfaces that interact with the α and β subunits of nAChR. NMR analysis is consistent with this, where one MII surface is essentially hydrophobic, with residues i-2, i-1, i (i is the second cysteine residue), j + 7 (j is the third Cysteine) and a fourth cysteine, and a disulfide bridge between the second and fourth cysteine residues. The other face is hydrophilic in nature, and residues j + 3, j
+4, j + 5 and j + 6, where j is the third cysteine. Mutation analysis suggests that the hydrophilic surface is particularly important for peptide-on rate (peptide docking) and the hydrophobic surface is especially important for peptide-off rate (peptide locking). The use of two physically distinct peptide surfaces to interact with adjacent α and β receptor subunits allows the specific design of ligands that are selective for various combinations of α and β subunits. Represent the plan.

IV. Interaction of MII with nAChR MII achieves its remarkable subtype specificity by a double-faceted interaction with the receptor. The potency ranking of the MII rank indicates that MII interacts with both the α and non-α subunits of the nAChR. Blocking by MII is consistent with competitive inhibition, which is voltage-independent of MII activity and dihydro-, a competitive antagonist.
β-erythroidin is evidenced by its ability to affect the ability of MII to block the receptor. The shape of the time-course recovery curve after blocking the α3β2 and α3β4 receptors by high concentrations of MII indicates that each of these receptors has two binding sites for MII, and occupation of either site by toxin is Suggests blocking body function. There is seen a general pattern of two ligand binding sites for the α / non-α subunit combination of neuronal nAChRs. This is in contrast to α7 homomeric receptors, which may have up to five Ach binding sites. The rate of recovery after blocking by MII is significantly longer for α3-containing receptors than for receptors without the α3 subunit. However, the IC ₅₀ for the MII block of the α3β2 receptor is about four orders of magnitude lower than that of the α3β4 receptor. Surprisingly, the off rate of MII is essentially the same for both receptor subtypes, and the IC ₅₀ differences are almost entirely due to the on rate of MII. Taken together, the data show that the on-rate of MII is largely controlled by its interaction with the β subunit, while the interaction with the α subunit is primarily M
II causes off speed. The kinetics of the synthetic chimera of MII supports this conclusion (Example 10). The selectivity effect of MII is due to the "dock-and-lock" model (Example 12).

[0048] The three-dimensional solution structure of MII, together with the data discussed in detail herein, is linked to a two-step dock-and-lock mechanism that accounts for the pronounced subtype specificity of MII. The NMR structure suggests that MII is a roughly wedge-shaped peptide with a hydrophilic edge and a hydrophobic edge. All four amino acids of α-conotoxin MII (HLEH) substituted for FATN are located on the hydrophilic edge of the peptide. The kinetic results obtained with the FATN analog and the broad type of toxin acting on the α3β2 nicotinic receptor thus indicate that the docking surface of the toxin is located on the hydrophilic side of the wedge, Interacts with the β2 subunit at a fast on-time. Another aspect of the wedge, characterized by solvent exposed hydrophobic residues, is the α3 of the nAChR target of this peptide.
It may be a place to attract the locking surface of α-conotoxin MII, which is assumed to interact with sites on the subunit with high affinity. α3β2 and α3β4
Α-conotoxin MII and FATN-α-conotoxin MI interacting with the receptor
FIG. 14 shows the manga notation of I analog.

Using differential on-and-off rates for MII, assays to screen compounds for nAChR antagonist activity and receptor subtype specificity can be developed. M
The compounds to be screened for on-and-off rates for α-II binding and α3β2 and α3β4 are determined and compared. For the α3β2 subtype, the on rate of the compound is greater than or equal to the MII on rate, for the α3β4 subtype, the compound on rate is less than or equal to the MII on rate, and the compound has an off rate of MII. If above the off-rate, the compound is a nAChR antagonist and is specific for the α3β2 subtype. Regarding the α3β2 subtype,
If the on-rate of the compound is less than or equal to the on-rate of MII, and for the α3β4 subtype, the on-rate of the compound is greater than or equal to the on-rate of MII and the off-rate of the compound is greater than or equal to the off-rate of MII If the compound is a nAChR antagonist and the α
Specific to the 3β4 subtype. Furthermore, it modulates the activity of neuronal nAChRs with α3β2 and α3β4 subtypes based on the three-dimensional structure of MII as disclosed herein, structural / biological activity of MII and molecular modeling analysis disclosed herein. Can be developed. Thus, the present invention develops compounds that modulate the activity of neuronal nAChRs and have subtype specificity for either the α3β2 or α3β4.

V. Rational Drug Design The goal of rational drug design is to generate structural analogs of biologically active polypeptides or compounds with which they interact (agonists, antagonists, inhibitors, binding partners, etc.). By producing such analogs, they will be more active or stable than natural molecules, have different sensitivities to denaturation, or affect the function of various other molecules It is possible to create a drug. In one approach, as described herein,
Alternatively, a three-dimensional structure of the MII will be created by computer modeling or by a combination of both approaches. Another approach, "alanine scanning," involves the random substitution of alanine for residues throughout the molecule and measures the resulting effect on function.

It is also possible to isolate the MII-specific antibody selected by the functional assay and then to solve its crystal structure. In principle, this approach yields a pharmacore upon which the next drug design can be based. It completely bypasses protein crystallography by generating anti-idiotypic antibodies to functional, pharmacologically active antibodies. As a mirror image of a mirror image, the binding site of the anti-idiotype would be expected to be an analog of the original antigen. The peptide could then be identified and isolated from a bank of chemically or biologically generated peptides using the anti-idiotype. The selected peptide will then function as the pharmacore. Anti-idiotypes can be generated using the methods described herein for making antibodies using the antibodies as antigens.

Thus, drugs can be designed that have improved MII activity or that function as enhancers, inhibitors, agonists, antagonists or molecules affected by MII or MII function. Further, knowing the polypeptide sequence,
Allows prediction of structure-function relationships using a computer. Substances that have been identified as modulators of polypeptide function are essentially
It may be a peptide or a non-peptide. Non-peptide "small molecules" are often preferred for many in vivo pharmaceutical uses. Thus, a mimetic or mimetic of the substance (especially if a peptide) can be designed for pharmaceutical use.

Designing a mimetic for a known pharmaceutically active compound is a known approach to the development of medicaments based on “lead” compounds. If the active compound is difficult or expensive to synthesize, or if it is inappropriate for the particular method of administration, for example, it will be immediately degraded by proteases in the gastrointestinal tract In particular, if pure peptides are unsuitable active agents for oral compositions, this may be desirable. Generally, mimetic design, synthesis and testing are performed to avoid randomly screening the majority of molecules for target properties.

There are several steps commonly taken in the design of a mimetic from a compound having a given target property. First, certain portions of the compound that are significant and / or important in determining the target properties are determined. In the case of peptides, this may be achieved by systematically changing the amino acid residues in the peptide, for example by substituting each residue in turn. Generally, such peptide motifs are purified using an alanine scan. These parts or residues that make up the active region of the compound are known as "pharmacophores". Once a pharmacophore has been found, using data from a set of sources such as, for example, spectroscopic techniques, X-ray diffraction data and NMR, for example, stereochemistry,
Model according to physical properties such as binding, size and / or charge. Computer analysis, similarity mapping (which models the charge and / or volume of the pharmacophore rather than the bonds between atoms) and other techniques can be used for this modeling method.

In a variant of this method, the three-dimensional structure of the ligand and its binding partners are modeled. This is particularly useful if the ligand and / or binding partner changes conformation upon binding, allowing the model to take this into account in the design of the mimetic. A template molecule is then selected that is capable of implanting a chemical group that mimics the pharmacophore. The mimetic can be easily synthesized while retaining the biological activity of the lead compound, suitable for pharmacologically acceptable, and conveniently, so as not to degrade in vivo. Select the implanted chemical group. Alternatively, if the mimetic is peptide-based, additional stability may be achieved by cyclizing the peptide to increase its rigidity. The mimetics or mimetics found by this approach are then screened to see if they have target properties or to what extent they exert it.
Then, further optimizations and modifications are made to arrive at one or more final mimetics for in vivo or clinical trials.

VI. Design of biologically active molecules based on the structure of MII Overview In general, biologically active molecules exert their effects by binding to receptors. The ability of a given molecule to bind to the receptor is determined by the presence of certain key functional groups and the three-dimensional (3D) representation of these features by the molecule. The goal of molecular modeling in the context of designing bioactive molecules is to understand the determinants of receptor binding and to carry out this finding in the design or discovery of new molecules with the desired activity. The availability of structure-activity data or the structure of the receptor allows the proposition of pharmacophore patterns which, in the best case, will include the functionality required for activity and the relative orientation of the functional groups (Gu
nd, 1979). Recent advances in technology for generating 3D molecular structures of typical drug-sized organic molecules and searching databases of 3D structures combined with the 3D pharmacophore hypothesis now support the discovery of new active molecules New tools are given to medicinal chemists. If the pharmacophore hypothesis is correct, the 3D pharmacophore can be used as a database query to search molecular databases for molecules expected to have the desired activity. If a receptor structure or receptor site model is available, another type of 3D search can be used. In this regard, a molecule is selected based on its steric or chemical reconciliation to the receptor structure without having to suggest which interactions may be most important. The methods are complementary and both support the discovery of active molecules.

The importance of the HLEH fragment of MII has been investigated and its biological activity and receptor binding have been correlated to its structural, chemical and physical properties.
New compounds with potentially different affinities for nicotinic nAChRs, such as higher affinities for the α3β2 subtype, to help understand the range of biological activities found in a series of compounds These features are used to help guide the design of compounds with or higher affinity for the α3β4 subtype.

A wide range of experimental and theoretical data is routinely used to reveal patterns of such features. This process, commonly referred to as pharmacophore mapping, involves three main aspects: discovering the required features for biological activity; the required molecular conformation (ie, the “biologically active” conformation). And clarifying the superposition or arrangement rules for a series of compounds.
The main information used for pharmacophore mapping is derived solely from a series of synthesized compounds and their measured biological activities. From this, a structure-activity relationship emerges and one can begin to formulate a rudimentary pharmacophore hypothesis. If the structure of the macromolecular target or target-ligand complex is known, it can be determined experimentally,
This information is clearly very useful to the pharmacophore-mapping process, either as it is or computer modelled. A variety of molecular modeling and computational chemical techniques can then be applied in conjunction with the experimental data to develop pharmacophore models. These techniques are based on simple, qualitative molecular graphic comparisons of a series of several models to create pharmacophore mappings and to provide precise quantitative measurements to determine their ability to reproduce experimental results. To the method.

[0059] Pharmacophore mapping is primarily a qualitative application. Familiar with pharmacophore mapping is the field of three-dimensional quantitative structure-activity relationship (3D-QSAR). The 3D-QSAR technique has been tried to derive and examine quantitative models of biological activity; models that can be used to predict the potency of compounds outside the experimental set, fitting into the potency of the experimental compound. . In fact, many 3D-QSAR applications require the proposed pharmacophore model.

Pharmacophore mapping has its roots in traditional medicinal chemistry and structure-activity relationships derived from the synthesis and biological testing of a range of compounds. From the relative measured potency of the individual members of the series, the concept of a type of functionality, as well as its spatial relationships, which are probably important for activity, begin to emerge. A more accurate model can then be formulated using more detailed molecular modeling experiments, but without it one can begin to materialize the proposed pharmacophore hypothesis.

Many ligand-design experiments utilize molecular modeling as well as chemical synthesis to assist in elucidating pharmacophore patterns. The structure-activity data and the results of the conformational analysis are used to propose a bioactive conformation and overlay for the series. This can be done by using molecular graphics.

[0062] Another common approach in pharmacophore mapping is the use of structurally rigid molecules to capture receptor binding requirements. One of the advantages of using less conformally resilient molecules is that less resilient molecules can more accurately define the bioactive conformation, but the selected rigid molecules have better affinity. It is to show. The use of rigidizability to better define pharmacophore space has been used successfully in making active non-peptidic compounds from peptide leads. Recent work with non-peptide compounds based on cyclic RGD peptide analogs by Ku et al. (1995) and coworkers is a good example of this.

[0063] Mapping receptor sites involves a variety of computer techniques to identify energetically favorable binding sites on macromolecules. The most direct approach involves, depending on the empirically determined physical properties, such as electrostatic or lipophilic potential, degree of curvature, and hydrogen bonding properties, the solvent accessible surface of the macromolecular target ( Or “painting” casts generated in other ways). Such a method for characterizing the surface of macromolecules is described in Grasp (Columbia Universi).
ty), DelPhi (Biosym Technologies), MOLCAD (Tripos, Inc.), and Hint (Virginia Commonwealth University). Subsequent molecular design involves the identification or design of ligands that have characteristics complementary to the identified surface properties. More advanced algorithms include the actual calculation of the enthalpy of interaction between the target and a potential ligand or fragment. In practice, the coordinates of the protein or protein fragment of interest (which can be rotated or otherwise transformed) remove any undesired ligands (or portions thereof) and / or any undesired solvent molecules. The coordinates are then processed to link the molecular mechanics parameters to the positions of the atoms to obtain a processing target for mapping. The target may be split into separate binding sites. The target or its split site is determined by MCSS (Molecular Simulations, I
nc.), a regular grid of site points that are either overfilled with a given functional group fragment, which allows subsequent relaxation to the desired position, or where a single fragment probe is located sequentially. Examples of programs for developing the site-grid algorithm include Grin / Grid (Molecular Discovery, Ltd.), Ludi (Biosym Technologies), Leapfrog (Tripos, Inc.) and L
egend (University of Tokyo). In both techniques, the enthalpy contribution of binding affinity is estimated using molecular mechanics force fields to systematically determine the appropriate position of the selected functional group.

In the site-grid approach, a box is defined around a desired site of the target in a defined grid. The grid division, ie the distance between grid points, can be determined by the practitioner or set by a computer program. Similarly,
Other parameters of points in the grid, such as hydrophobicity or other properties, can be determined as well. Probes (i.e., computer models) of one or more selected groups, functional groups, molecules or molecular fragments are located at the grid points and the probe-target pair interaction energies are determined for each grid point. Data for each selected group, functional group, etc. can be collected, recovered as a data set, and visualized or printed on a computer monitor in various text or graphic formats.

As an alternative to locating the groups of each set of grid points, as described above, by overlapping multiple copies near the protein target, the selected fragment,
Multiple copies of the group, molecule, etc. can satisfy the target (determined by coordinates as described above). The model is then subjected to group minimization (ie, molecular mechanics minimization) calculations to identify points or regions of favorable interaction. The data can be treated as in the grid approach.

Receptor site mapping allows seeds for ligand growth via database search as described above, and Grow / l for de novo design of new chemical groups.
Seeds for the ink method are provided. A program for ligand growth first starts with an extensible fragment dictionary (dict) to place the appropriate functional groups at site points.
ionary). The general algorithm or subgraph isomorphic protocol is then invoked to combine the small aliphatic chains or rings with the fragments. Stochastic increases can be introduced by modification of the internal degrees of freedom and translation and rotation of the candidate model in the coupling cavity. The resulting set of molecules is scored and filtered by a function that takes into account the steric constraints of the binding site, the complementarity of electrostatic and hydrophobic interactions, and the solvation estimates. Programs of this type that could be applied include Ludi (Biosym
Technologies), Leapfrog (Tripos, Inc.), Legend (University of Tokyo), G
row (made by Upjohn), Builder / Delegate (University of California, San Fran
cisco) and Sprout (University of Leeds). Click detection offers another strategy for site mapping and ligand growth. DOCK (University of Ca
lifornia, San Francisco) and similar programs fill a given binding site with a minimal set of possible atom-size spheres; then attempt a database search to place atoms in the center of the site-filling range (or " Orient the ligand so that it overlaps the nucleus "). The shape complementarity increased by the scoring functions to include the steric requirements of the cavity and the potential energy function.

Optimization of ligand (from any source) can be enhanced using the three-dimensional structure of MII. The receptor site map or hydrophilicity profile of the MII can be used to identify preferred positions for the functional group components of the ligand, or otherwise typically controlled by minimal steric considerations of the ligand structure itself. The conformational search for possible ligand structures may be filtered or constrained. The availability of well-defined binding sites determines the mode of ligand binding to the target protein via simulated annealing, distance geometry, metropolis-Monte Carlo or methods that utilize force-field directly in the stochastic search of binding modes You can also. Examples of programs that can be applied to rationally explain ligand binding include Autodock (
Scripps Clinic), DGEOM (QCPE # 590), Sculpt (Interactive Simulation
s, Inc.), or any of the molecular dynamics programs described above. Once a manageable set of possible binding orientations has been obtained, the appropriate mode of binding can be easily identified via modifications in the test ligand designed to alter the binding affinity of each model under consideration in a predictable manner. can do. For example, a ligand may be modified to include a functional group at a position that does not match one binding model and still matches another model. Then, using the combined data,
You can remove the model. Once a suitable model has been created by repeated removal of unlikely binding modalities, the potential for improved leads is readily apparent from the local protein environment.

Another protocol for ligand optimization involves a 3D database search for binding site findings. Modeling may reveal multiple candidates for the bioactive conformation of a given ligand. Probes for the correct conformation may include 3D searches to identify several bound mimetics of each possible adaptation. The structure-activity relationship of the unhindered ligand will indicate which functional groups should be retained in the hindered mimetic. Finally, the steric and electrostatic requirements of the binding site could constitute a filter that prioritizes the possibilities obtained.

The structure of MII allows for accurate model building of cognate peptides and its subsequent use in drug design. The MII structure can easily be used to develop a reliable model either by way of an informed template construction method or by minimizing the mechanical structure of local molecules following repeated point mutations. Sy
Examples of programs that can be applied to the development of k-NC models include Composer (Birbeck Col.
lege), Modeler (MSI) and Homology (Biosym Technology). The resulting model can then be subjected to any of the CADD techniques described above.

B. Use of the Structure of α-Conotoxin MII in Computer-Assisted Drug Design The availability of the three-dimensional structure of α-conotoxin MII enables a structure-based drug discovery approach. Structure-based approaches include de novo molecular design, computer-assisted optimization of lead molecules, and computer-based selection of candidate drug structures based on structural criteria. New pepidomimetic modules can be developed directly from peptide ligand structures by conformationally restricted designs for peptide replacement or database searches. Alternatively, structure-based lead discovery can be accomplished using a target protein structure, eg, a receptor with its ligand removed.

Multiple uncomplexed states of MII can be generated by several methods to obtain additional target conformations. Experimental equivalents and the resulting uncomplexed model can be subjected to techniques such as receptor site mapping to target protein structure and potential ligands or chemical groups ("fragments" or "seed").
The site of favorable interaction energy with the structure of ")" can be identified. Such evaluation can be based on techniques such as fragment seed binding and growth. Fragment seed binding is "bound" Seed ": a method for designing a structure comprising two or more mapped chemical groups suitably spaced to reach a corresponding site of favorable interaction. Growth refers to growth. Refers to the design of a structure that extends a given molecule or group based on receptor site mapping or to fill available space.Database of chemical structure based on data from the receptor site mapping. Potential or suboptimal ligands of any origin can be identified using a receptor site map. As revealed by this, multiple ligand conformations and orientations can be filtered according to energy preferences.The structure of MII allows the high homology of MII by either informed cognate template methods or minimization following repeated orientation mutations. A quality model can be created, and the created structure of MII can then be processed as a further protein target by the methods outlined above.
These methods and their application to MII are described in subsequent sections.

The amino acid sequence of α-conotoxin MII has been determined (Cartier et al., 1996), and determinants (specific amino acid residues) of MII specificity on α3β2 neuronal nicotinic receptors have been identified ( Harvey et al., 1997, which is hereby incorporated by reference. Knowledge of the determinants and bond orientation of the peptide may suggest ways to achieve conformational restriction and peptide bond replacement. For a less biased approach, a database of three-dimensional structures may be searched to replace one or more of the peptide ligands, preferably non-peptide substituents,
A computer algorithm for identifying is included. By this method, compounds can be made that abundantly place a bioactive conformation, ie, compounds that are based on a particularly biologically relevant conformation of the peptide ligand. Algorithms for this purpose are Cast-3D (ChemicalAbstracts Service), 3DB
Unity (Tripos, Inc.) and MACCS / ISIS-3D (Molecular Design Limite)
This is executed by a program such as d company. These geometry searches can be augmented by three-dimensional searches that remove hits with impeding dimensions using joint site size and shape requirements.

Programs that can be used to synchronize geometry and conformational requirements in searches that apply to MII include CAVEAT (University of California, Berkel)
ey). By way of example, a non-limiting list of computer programs for performing rigid three-dimensional searches includes: 3Dsearch (Seridan, rp et al., J. Chen. Inf. Comput. Sci. 29: 255.1989). Aladdin (Van Drie, JH et al., J. Comput. Aided Mol. Design 3: 255. 1989) UNITY (Tripos, Inc.) MACCS-3D (MDL) CATALYST (Biosyn / MSI, Inc.) All of these search protocols can be used with existing corporate databases, the Cambridge Structural Database, or with chemical databases available from chemical suppliers.

As will be appreciated by practitioners in the art, a variety of computer analyzes can be used to provide a given peptide with MII (or a portion or complex thereof) or other α-conotoxins as described above. The degree of similarity of the three-dimensional structure between the peptide or a portion or complex thereof can be determined. Such an analysis is based on CAVEAT (Molecular Simulations, I) described by Lauri and Bartlott (1994).
nc., Waltham, Mass.) using commercially available software applications.

CAVEAT is a program that supports the design of new molecules using database search. It is a special purpose program written to assist in the design of small molecules that can mimic protein loops important in protein-protein interactions. The basic assumption of the program is that protein-protein recognition is based on specific positioning of amino acid side chains, and that the scaffold is not critical and can be replaced. The CAVEAT program has been tried to find ring systems in which side chains can exist in the desired orientation. The CAVEAT database is
Vector distance and angular ring forms, as well as ring
Consists of a 3D structure and back-to-back indexes. A typical question can determine the Cα-Cβ vector relationship that will require that the required side chain analog be present.
Such a vector search could be performed by the most common database search programs discussed in this section; however, CAVEAT was optimized for this task. Although the early CAVEAT application was for protein mimicry, the program could be useful in any situation that required a new scaffold to have a set of functional groups present in a defined orientation. Recent CAVEAT suites of the program have interactions that gather the hits based on similarity in size or basis, rank the hits by the size and number of ring-fused atoms, and lower the designed molecule Or not. Two new databases, CAVEAT:
Triad is available for use with Iliad, a computer-generated collection of all tricyclic hydrocarbons, and a collection of acyclic molecules. Cambidg software
It can be used to create a CAVEAT database from e Structural Database.

In addition to retaining potential pharmacophore factors present in the peptide explicitly, the introduction of groups that donate or accept hydrogen bonds that can displace aligned water molecules into the ligand structure Provides a significant entropic increase, which typically leads to the favorable free energy of the bond. Such aligned waters can be identified from the structure, and other aligned waters can be located during computer simulation of the fully solvated structure.

The structural coordinates of the peptides of the invention may be machine-read to represent them in a three-dimensional form or for computer-aided manipulation of the determined structural coordinates or three-dimensional structure, or other applications involving calculations based thereon. It can be stored in a machine-readable form on a possible data storage medium, such as a computer hard drive, diskette, DAT tape, etc. In order to use the structural adjustments made for the MII illustrated in FIG. 8, it is often necessary to represent or convert to a three-dimensional form, or otherwise manipulate them. This is typically done by using commercially available software, such as a program that can create a three-dimensional graphical representation of a molecule or a portion thereof from a set of structural coordinates. By way of example, a non-limiting list of computer programs for examining or otherwise manipulating protein or peptide structures includes the following: Midas (University of California, San Francisco) MidasPlus (University of California, San Francisco) MOIL (University of Illinois) Yummie (Yale University) Sybyl (Tripos, Inc.) Insight / Discovery (Biosym Technologies) MacroModel (Columbia University) Quanta (Molecular Simulations, Inc.) Cerius (Molecular Simulations, Inc.) Alcherny (Tripos, Inc.) Lab Vision (Tripos, Inc.) Rasmol (Glaxo Research and Development) Ribbon (University of Alabama) NAOMI (Oxford University) Explorer Eyechem (Silicon) Graphics, Inc.) Univision (Cray Research) Molscript (Uppsala University) Chem-3D (Cambridge Scientific) Chain (Baylor College of Medicine) O (Uppsala University) GRASP (Columbia University) X-Plor (Molecular) Simulations, I nc .: Yale University) Spartan (Wavefunction, Inc.) Catalyst (Molecular Simulations, Inc.) Molcadd (Tripos, Inc.) VMD (University of Illinois / Beckman Institute) Sculpt (Interactive Simulations, Inc.) Procheck (Brookhaven National Laboratory) DGEOM (QCPE) RE VIEW (Brunel University) Modeller (Birbeck College, University of London) Xmol (Minnesota Supercomputing Centers) Protein Expert (Cambridge Scientific) HyperChem (Hypercube) MD Display (University of Washington) PKD (National Center for Biotechnology Information, NIH)

For example, data defining the three-dimensional structure of an α-conotoxin peptide, or a portion of MII or a structurally similar homolog or analog, can be stored in a machine-readable data storage medium, typically Using a computer capable of reading data from the storage medium, it can be displayed as a graphic three-dimensional image of the peptide structure and can be programmed with instructions for creating an image from such data. Thus, the present invention provides a memory containing data for imaging the structural coordinates of a composition of the present invention, such as those described in Example 8, along with further optional data and instructions for manipulating such data. Equipment such as a computer.
Such data can be used for various purposes, such as elucidation of other relevant structures and drug discovery (WO 97/08300).

In comparative protein modeling, a first data set is converted to a second data set using an instrument programmed with instructions for using a first set of such machine readable data with a second set of machine readable data. Combined with the data set, at least a portion of the coordinates corresponding to the second set of machine readable data may be determined. For example, the first set of data may include a Fourier transform of at least a portion of the MII coordinates illustrated in FIG. 8, while the second data set may include MII potential analog coordinates. Thus, using molecular replacement, a set of coordinates as illustrated in FIG. 8 can be developed to determine the effect of such replacement on the structure.

Thus, another object of the present invention is to provide a method for determining the three-dimensional structure of a peptide analog or peptidomimetic of MII using comparative protein modeling techniques and structural coordinates for the compositions of the present invention. It is. Comparative protein modeling involves building a model of the unknown structure using the structural coordinates of one or more related peptides. Comparative protein modeling can be performed by fitting a common or similar part of a protein or peptide whose three-dimensional structure is to be elucidated to the three-dimensional structure of a known similar structural element.
Comparative protein modeling can include reconstructing some or all of the three-dimensional structure with replacement of amino acids (or other components) by amino acids of the relevant structure to be elucidated. For example, using the MII structural coordinates, comparative protein modeling can be used to determine the three-dimensional structure of the MII peptide analog. These coordinates can be stored, displayed, manipulated, and otherwise used in a manner similar to the MII coordinates of Example 8.

The present invention further identifies reactive amino acids, such as cysteine residues, in the three-dimensional structure, preferably at or near the receptor binding site: a water accessible surface or a space filling of all atoms. Create and visualize molecular surfaces, such as surfaces, including van der Waals surfaces; calculate and visualize the size and shape of peptide surface features; within a three-dimensional structure, preferably at or near the receptor binding site The potential H
A peptide of the invention or a peptide thereof for locating the binding donor and acceptor; and calculating the hydrophobic and hydrophilic regions in the three-dimensional structure, preferably in or near the acceptor binding site. Provide for use of partial structural coordinates. Using the above approach to characterize the peptide and its interaction with the receptor,
One can pre-select the set, design or select compounds with properties complementary to the surface features of the peptide (eg, size, shape, charge, hydrophobic / hydrophilic, receptor specificity, etc.). The structure coordinates are used to predict or calculate the orientation, binding constant or relative affinity of a peptide for a given receptor subtype in the bound state, and to design or select compounds with improved or altered affinity. You can use that information.

In such a case, the structural coordinates of the α-conotoxin peptide, or a portion thereof, are subjected to the desired operation in a machine-readable form, and any necessary additional data, such as the structure of the potential analog and / or Alternatively, the functional properties are determined and input to a programmed instrument with instructions for determining various amino acids and other molecular properties. Compounds of the structure selected or designed according to any of the above methods may comprise nAC
One may test for its ability to bind to the hR, inhibit the binding of the nAChR to its natural or artificial ligand, and / or inhibit the biological function mediated by the nAChR.

The present invention also provides peptidomimetic and mimetic methods for designing compounds that can bind to nAChRs. One such method includes nACh
Graphically displaying a three-dimensional image based on coordinates defining the three-dimensional structure of the MII coupled to R or a portion thereof. Characterize the interaction between a portion of MII and the receptor to identify candidate groups for replacement. One or more portions of MII that interact with the receptor may be replaced by a substituent selected from the knowledge base of one or more candidate substituents, and / or the group may be added to MII to further interact with the receptor. Interaction is acceptable.

The computer-based approaches and structural insights disclosed herein provide nAC
It also allows for the design or identification of molecules with reduced or substantially impossible ability to bind to the hR subtype. For example, the same modeling and computational methods can be applied to the data described herein, but for the opposite purpose, i.e., to design or design compounds that lack substantial binding affinity for one or more nAChR subtypes. It can also be applied to identify. Such information is provided by nAC
It may be useful in research efforts aimed at identifying antagonists of a single subtype of hR. Compounds first identified by any such method are also encompassed by the present invention.

A graphical three-dimensional image of any of the molecules or molecular complexes described herein can be displayed for storage, transfer and use of the structural coordinates of the crystalline material of the present invention in such programs. When using a device programmed with instructions for using data coded with machine readable data, for example, loading a computer with one or more programs of the type defined above;
A machine-readable data storage medium is provided that includes a data storage material encoded with the data. Machine-readable data storage media, including data storage materials, include conventional computer hard devices, floppy disks, DAT tapes, CD-ROMs, and other magnetic, magneto-optical, Optical, optical floppy and other media are included.

Even more preferred is the structural coordinate of MII or a portion thereof, in particular, the root mean square deviation from the skeletal atoms of the amino acids of such a protein having a structural coordinate of MII shown in FIG. A machine readable data storage medium capable of displaying a graphical three-dimensional image of a defined molecule or molecular complex. An exemplary embodiment of this aspect of the invention is a conventional 3.5 "diskette, DAT tape or hard drive coded with a data set that preferably includes the coordinates of FIG. 8 in PDB format.

In another embodiment, the machine readable data storage medium is annex I,
Instructions for using and using a data storage material (or, again, a derivative thereof) coded with a first set of machine readable data including a Fourier transform of the structural coordinates described in Attachment II or Attachment III When using an instrument programmed in the second step, it is used to generate a second pattern containing the X-ray diffraction pattern of the molecule or molecular complex.
Combined with the set of machine-readable data, at least a portion of the structural coordinates corresponding to the second set of machine-readable data may be determined. Examples of systems useful for this aspect of the invention are provided in WO 97/08300.

VII. Definitions The present invention uses the following definitions. "Biological activity" is used herein to broadly describe function. "Derivative" refers to an amino acid sequence in which one or more residues of the native sequence have been replaced. The "features" of a compound include any combination of the structural, physical or chemical attributes of the compound that are necessary to elicit a particular biological activity. A peptide “fragment”, “portion” or “segment” refers to at least about 2 contiguous amino acids, typically at least about 3 contiguous amino acids,
And most preferably a stretch of amino acid residues of at least about 4 or more contiguous amino acids. "Neuronal nAChR" refers to the natural nicotinic acetylcholine receptor found in the CNS. "Peptide analog" refers to the amino acid sequence of α-conotoxin which is
Refers to molecules having fewer, different or modified residues. The modification is
It may be present by the addition, substitution or deletion of one or more amino acid residues.
Such modifications may include the addition or substitution of analogs of the amino acid itself, such as peptidomimetics or amino acids having a modifying group, such as a modified side chain.
The analog may have a different function than the native α-conotoxin molecule, or may exhibit the same function or varying degrees of the same function. For example, the analogs can be designed to have higher or lower biological activity, have a longer shelf life or increased stability, or be easier to formulate. In this specification, for the sake of simplicity, the analogs of the present invention are sometimes referred to as proteins or peptides, but other types of molecules, such as peptidomimetics or chemically modified peptides, are contemplated. In this specification, for the sake of simplicity, peptide analogs can be referred to as peptides.

“Peptidomimetic or mimetic” is intended to refer to a substance that has the essential biological activity of an α-conotoxin peptide. Peptidomimetics or mimetics can be peptide-containing molecules that mimic elements of peptide secondary structure (Johnson et al., 1993). The rationale behind the use of peptide mimetics or mimetics is that, like those of ligands and receptors, the peptide backbone exists primarily to orient amino acid side chains to facilitate molecular interactions. . Peptidomimetics are designed to allow similar molecular interactions as the natural molecule. The mimetic need not be a peptide, but it will retain the essential biological activity of the native conotoxin peptide. "Related composition" refers to a composition comprising a conotoxin peptide analog or peptidomimetic as an active ingredient. "Simulated annealing" refers to a strategy for superimposing elastic molecules and examining potential alignments. "Substantially pure" refers to a peptide that is present in the substantial absence of other biomolecules of the same type; it is preferably at least about 85% pure, and most preferably at least about 95% pure. Present in quantity.

“Substantially similar activity” refers to the activity of a modified peptide with respect to the native α-conotoxin peptide. The modified peptides will have substantially similar activity. The modified peptide can have an altered amino acid sequence and / or can include a modified amino acid. In addition to activity similarity, modified polypeptides may have other useful properties, such as longer shelf life. Alternatively, the similarity of activity of the modified peptide may be higher than that of the native α-conopeptide. The modified peptides are synthesized using conventional techniques, or are encoded by modified nucleic acids and are produced using conventional techniques. Modified nucleic acids are prepared by conventional techniques. The modified peptide can be produced using a nucleic acid having an activity substantially similar to the function of the native α-conotoxin gene. “NAChR subtype” refers to nicotinic acetylcholine receptors in different molecular forms.

VIII. Preparation of Peptide Analogs and Mimetics Peptide analogs and peptidomimetics of the invention that are specific for the α3β2 or α3β4 subtype of nAChRs are described herein using conventional techniques such as drug modeling, drug design and combinatorial chemistry. It is prepared based on the method described in (1). Suitable techniques include, but are not limited to, US Pat. No. 5,571,698, US Pat. No. 5,511, all of which are hereby incorporated by reference.
No. 4,774, WO 95/21193 (Ecker et al., (1995); Persidis (1997); Johnson et al. (
1993); Sun et al. (1993)) and the references cited herein. The development of peptide analogs and peptidomimetics
Prepare using commercially available drug design software, including those described in rsidis et al. (1997). These peptide analogs and peptidomimetics have substantially similar activity to the α-conotoxins described herein and in the published publications.

IX. Preparation of Pharmaceutical Composition Containing Peptide Analogs and Mimetics A pharmaceutical composition containing the compound of the present invention as an active ingredient is prepared by
It can be prepared according to conventional pharmaceutical composition techniques, such as those in ceutical Sciences (1990). Typically, an antagonist amount of the active ingredient will be mixed with a pharmaceutically acceptable carrier. The carrier may take a wide variety of forms depending on the form of preparation desired for administration, eg, intravenous, oral or parenteral.

For oral administration, the compounds can be formulated in solid or liquid preparations, such as capsules, pills, tablets, troches, fluxes, powders, suspensions or emulsions.
In preparing the compositions in oral dosage form, for example, in the case of oral liquid preparations (such as suspensions, elixirs and solutions), for example, water, glycols, oils, alcohols, flavoring agents, storage Excipients, colorants, suspending agents and the like; or, in the case of oral solid preparations (eg, powders, capsules and tablets), starch, sugar, diluents, granulating agents, lubricants, Any of the usual pharmaceutical media can be used, such as carriers, such as binders, disintegrants, and the like. Because of their ease in administration, tablets and capsules represent the most advantageous oral dosage unit form, in which case solid pharmaceutical carriers are obviously employed. If desired, tablets may be sugar-coated or enteric-coated by standard techniques.

For parenteral administration, the compounds can be dissolved in a pharmaceutical carrier and administered as either a solution or a suspension. Examples of suitable carriers include water, saline, dextrose,
Fructose solution, ethanol, or animal, vegetable or synthetic oils. The carrier may also contain other components such as preservatives, suspending agents, solubilizing agents, buffers and the like. Where the compound is administered intrathecally, it can also be dissolved in cerebrospinal fluid.

The active ingredient, which is a peptide, can also be administered in a cell-based delivery system that introduces a DNA sequence encoding an effective agent into a patient's body, particularly into the spinal cord, into cells designed to be implanted. A suitable delivery system is described in U.S. Pat.
5,550,050 and international publication numbers WO 92/19195, WO 94/25503, WO 95/01203, WO 95/05452, WO 96/02286, WO 96/02646, WO 96/40871, WO 96/40959 and WO 97/12635. Suitable DNA sequences can be prepared synthetically for each active ingredient based on the disclosed sequences and the known genetic code.

The active ingredients of the present invention are administered in an amount sufficient to produce the desired effect. The dosage range over which these agents show this effect can vary widely, depending on the degree of patient deficiency, the patient, the route of administration, and other disease conditions present in the patient. Typically, the active ingredient will comprise about 0.05 mg / kg to about 250 mg / kg, preferably about 0.1 mg / kg.
It shows its therapeutic effect in the dosage range of mg / kg to about 100 mg / kg of the active ingredient. Suitable doses may be administered in multiple divided doses per day. Typically,
Dosages or divided doses may contain from about 0.1 mg to about 500 mg of active ingredient per unit dosage form. A more preferred dose will contain from about 0.5m to about 100mg of active ingredient per unit dosage form. The dosage generally starts at a lower level and is increased until the desired effect is achieved.

EXAMPLES The invention is further described in the following examples, which are provided for illustrative purposes and are not intended to limit the invention in any way. Utilizing standard techniques well known in the art of the techniques described in detail below.

Example 1 Synthesis of α-Conotoxin MII and MII Derivatives MII originally isolated from Conus magus was chemically synthesized on an ABI model 431 peptide synthesizer using the standard Fmoc method and subjected to two-step oxidation. Protocol (
Monje et al., 1993) were used to properly fold into its biologically active conformation. Once purified, the identity of the peptide to native MII can be determined by Cartier et al., 1
Confirmed as described by 996.

The MII derivative was similarly chemically synthesized and folded. The first derivative was prepared with alanine in place of each non-cysteine amino acid residue (ie, alanine-walk) to determine the effect of each residue substitution. Additional derivatives or analogs were similarly prepared by replacing the amino acid residue of MII with any amino acid except for the cysteine and His ₁₂ amino acid residues.

Example 2 Biological Activity of α-Conotoxin MII and MII Derivatives, Peptide Analogs and Peptidomimetics Each MII derivative, peptide analog and peptidomimetic was analyzed for the α3 of nAChR as described by Cartier et al. (1996). Were tested for activity on neuronal nAChRs in Xenopus laevis oocytes containing and β2 subunits. Briefly, oocytes (1-2 days after harvest) were treated with rat nAC.
cRNA encoding the α3 and β2 subunits of hR were injected and incubated at 25 ° C. for 1-4 days before use. Electrophysiological currents were measured by Cartier (199
Measured using conventional techniques as described in 6). As controls, measurements were taken on oocytes perfused with acetylcholine and oocytes incubated with either MII or the analogs prepared in Example 1.

FIG. 1 shows the effect of alanine substitution in conopeptide MII on the ability of an analog to block acetylcholinergic currents in voltage-clamped Xenopus oocytes expressing cloned rat α3 and β2 subunits. Most substitutions reduced the potency of MII by less than 4-fold. Notable exceptions were: Glu ₁₁ : 8-fold reduction; His ₉ : 25-fold reduction; His ₁₂ :> 10 ⁴ reduction. In addition, Asn ₅ (which can be replaced by His) and Pro ₆ along with His ₁₂ (which can be replaced by Asn) have been shown to be essential for high affinity binding to α3β2 nAChRs. Similar effects were shown when various residues of MII were substituted with other amino acids. The more conservative the substitution of amino acid residues in the analog, the more similar the activity exhibited for the synthetic analog.
The sequences of some α-conotoxins and their biological activities are given in Shon et al. (1997).

Example 3 NMR Spectroscopy NMR Spectroscopy 500 μL of 5 mM sodium phosphate made from a sample containing 8 mg of peptide from 90% H ₂ O and 10% D ₂ O (Cambridge Isotope Laboratory) Dissolved in buffer. The final pH of the peptide solution was 3.3 and the peptide concentration was 9.4 mM. Hydrogen Rabiru amide for "100%" experiments requiring be replaced by deuterons, the sample was lyophilized and dissolved in D ₂ O (99.96% isotope purity). After standing at room temperature overnight, the sample was freeze-dried again and dissolved in D ₂ O (99.996% isotopic purity).

All NMR data are in pulse-field-gradient (z) units
Obtained using a Varian 600 MHz Unity Plus spectrometer equipped with A series of 2-D¹H N
MR experiments were performed at 275 K; these were 75, 150, 250 and 350 ms.
NOESY (Jeener, J. (1979); Kumer, A. (1980); Macura,
S. (1981)), TOCSY using a 64 ms mixing time (Braunschweiler, L. (1983); Davis, D.G. (1985)), DQF-COSY (Piantini, U. (1982); Rance, M. (1).
983)) and PE-COSY (Muller, L. (1987)). All spectra are 4K data, except for a spectral width of 6600 Hz and PE-COSY (8K data points).
And recorded in phase sensitive mode (States, DJ (1982)). 8 or more
32 scans were signal flattened for each free induction decay (FID) with a 2 s relaxation delay.
Average; each 2-D experiment was performed with a 400-512 FID. In NOESY experiments, solvent signals were suppressed by gradient echo coupled to a WATERGATE sequence (Piotto, M. (1992)). "Flip-back"
Ruth was appropriately inserted after the mixing time to restore the residual magnetization along the z-axis.
In TOCSY, coupled to scalar¹Coherence along H spin
7.7kHz spin lock field required for transfer
(Rucker, S. (1989)) and set the residual water signal during the relaxation delay
WATERGA with minimum presaturation
Water suppression was obtained using a TE sequence. In PE-COSY, the sample is 100% D _Two No water suppression was required because it was prepared in O. A 30 ° mixed pulse was used for PE-COSY, and each FID was recorded with 32 scans to compensate for the loss of sensitivity due to the small mixed pulse.

The 2-D NMR data was transferred to an SGI workstation (Indigo ² ) and VNMR 5
Except for PE-COSY processed using .10 (Varian), processing was performed using Felix 95.0 (MSI, San Diego). The FID uses a window function (sine b shifted 90 ° and 135 °) in both dimensions before the Fourier transform.
ell)). Each Fourier transform FID was baseline corrected by fitting a third order polynomial, and each dimension was referenced to a 4.76 ppm chemical shift value of the residual water signal.

Except for the Cβ proton of Ser ¹³ (not observed), complete resonance assignment was easily obtained, but the 1-D spectrum of the peptide in the amide proton region had several resonance overlaps. As identified on the basis of the spin-type, it begins at the unique residues Val ^7, to the carboxyl-terminal residue Cys ^16, and traces a continuous d _{alpha N} NOE binding (connectivity). Along the trace, the resonance assignment of each residue was confirmed based on its spin type from TOCSY experiments. The remaining N- terminal residues, were assigned based spin system continuously d _{alpha N} NOE bond starting from Asn ⁵ identified from NOE cross peaks between the amide protons and β proton and its side chain γNH2 protons Was. Additionally, proximal amide protons, continuous NOE resulting from d _NN confirmed the resonance attribution of the. FIG. 2 shows continuous N extending from Cys ² to Asn ⁵ and from Val ⁷ to Cys ^16.
2 illustrates two loci of OE d _{α N} bond.

Example 4 Generation of Dihedral and Distance Constraints NOE cross-peak volumes were determined using NOESY with four different mixing times using FELIX 95.0.
Measured from data. The NOE buildup curve was then fitted to a two-dimensional polynomial, and exact distances were generated for all assigned NOE cross peaks. 1.8Å
Was used as a non-overlapping geminal Cβ proton cross-peak, as well as a suitable reference for lower distance binding. Several non-overlapping geminal Cβ proton cross-peaks were averaged and used to calibrate the measured NOE volume. A 1Å pseudoatom correlation (Wuthrich, K. (1983)) was added to the upper limit of their NOE cross-peaks, including spectroscopically altered methyl and methylene protons. As a two-sided constraint, the ³ J _{NH - H2} coupling constant was measured from the 1-D ¹ H spectrum recorded at 32K data points, then -120 ° (± 30 °) for ³ J _{HN - H2} > 8.0 Hz. °) and ² J _{NH - H2} <-60 ° for 5.0 Hz (. was converted to φ dihedral angle centered on ± 30 °) (Pardi, a et al. (1984)). ^The _3Jαβ coupling constant was also determined from PE-COSY recorded from a sample in which the amide proton was completely exchanged for deuteron in “100%” D ₂ O. Measurement ³ J _.alpha..beta coupling constants derived from a residue having a geminal β protons not AMX spin system Contact and overlap, combined with chi ₁ dihedral angle between successive d _{NH-H alpha} and d _.alpha..beta NOE cross peak intensity (Hyberts, S. (1987); Wagner, G. (1987)
).

First, a 99 residue NOE cross-peak with a high level of confidence in its assignment was selected and corrected to a NOE cross-peak containing at least one pseudoatom. The chirality constraints of 15 and 11 φ and 5 χ ₁ dihedral angles were also
The first set of 10 structures using the DGII (Havel, T. (1991)) module of the program (MSI, San Diego) was included in the restriction file to generate. DGII
All the structures generated by the calculations have a similar overall skeletal folding pattern, so the lowest energy structure with the lowest number of NOE violations was selected and subjected to the iterative relaxation matrix approach (IRMA). This approach improves the accuracy and precision of interproton NOE-derived distance constraints by evaluating the complete relaxation network of spins in the molecule (Boelens, R. (1988)). In this stage, in addition to 16 dihedral angles and 15 chirality constraints, a total of 158 NOE-derived inter-proton distances (including inter-residue NOEs with pseudo-atom correction) were determined by IRMA and bound molecular dynamics (RMD). ) Followed by an energy minimization step. None of the distances from the NOE containing at least one pseudoatom have been processed by IRMA, and thus they have not been revealed by the process. Each IRMA cycle employed experimental NOE intensities measured as a function of mixing time and merged them with calculated theoretical NOE intensity values for the model structure. A new set of distance constraints was then estimated from this mixed NOE intensity matrix. Rotational correlation time of 1.5 ns estimated from NOE collection curve of β proton where one pair of atoms is bonded to one non-overlapping atom, and diagonal leakage rate of 1.0 s Was used in each IRMA cycle. After each cycle of the IRMA, the structure was subjected to an RMD and energy minimization step using the Lennard-Jones potential at a 12.0 ° cutoff. 1000 fs RMD 1000 for 5ps
The five RMD cycles of the step were calculated at 700K, followed by 5ps at 500k and 5s at an additional 300K. The structure was minimized using a combined gradient minimization of 1500 steps following the steepest descent of 100 steps. It performed 4 IRMA / RMD cycles, 0.007 for 0.407 and R ₆ for R ₁ in the final R- factor reaches (NOE strength and distance, reflecting the relationship 1 / r ⁶ between r) Focusing was achieved (Gonzales, C. (1991)). This IRMA / RMD protocol is described in detail in the instructions provided by the Insight II software to determine the distance converted from the experimentally measured NOE cross peak.

[0108] that the majority of peptides are helically short from recorded ³ J _NH-C ^α binding constants and NOESY experiment from a high resolution 1-D spectrum - from both the NOE cross peaks range - and medium It became clear. 16 9 of the residues is determined to have a ³ J _{NH-C α} binding constant of less than 5.0 Hz, among residues of these 9, we, alpha
- very typical of the helix, short - and medium - range NOE cross peaks, d _NN, d _{NN (1. 1 + 2),} was identified d _{alpha N} and d _{alpha N} a _{(. 1 1 + 2)} ( Wuthrich, K. et al. (1986))
. FIG. 2 shows the NOE cross-peaks consistent with their intensity, along with the ³ J _{NH- Cα} binding constants used to generate the dihedral angle constraints. In addition to the observed NOE cross-peaks, the Cα proton chemical shift appropriate for random-coil values is a good tool for identifying local helical components in peptides and proteins (Wishart, DS et al. (1991); Rothemund, S. et al. (1996)). Plot shown in Figure 4, peptide, suggesting strongly that it comprises a helical conformation of large parts throughout the sequence, except for residues Asn ⁵ and His ^12. The backbone conformation of these residues could be derived from the helical conformation based on its positive shift in the plot.

Example 5 Molecular Modeling. Distance Geometry and Simulated Annealing Computer models of α-CT _X MII were constructed and manipulated using the Insight II and Discover programs (MSI, San Diego) from Silicon Graphics Work Station. Simulated annealing (SA) calculations are based on the San Diego Supercomputer Cent
The test was performed with a Cray C-90 supercomputer. Harger and co-workers' consistent valence force field
(Dauber-Osguthorpe, P. (1988)) was used for both DGII and simulated annealing calculations. The molecule extended with two disulfide bridges is built up as a triple charged cation, with all positive charges on the N-terminal nitrogen and the imidazole side chains of His ⁹ and His ¹² .

The expected metrication and majorization with the constant weighting scheme following the triangular unequal joint smoothing and four-dimensional inset procedure is 750 kcal / mol of initial energy and 0. Except for 20000 steps of simulated internal organ annealing at 3ps step size,
Used by default for DGII calculations.

To develop a more qualitative image of the α-CT _X MII conformation and to create a three-dimensional structure that fits the NMR data, a simulated annealing, such as that developed by Colle and Gronenborn or co-workers. The principle (Nilges, M (19
88)) was used. The calculation incorporated two significant new developments from the previous work. First, the interproton distance was introduced into a two-phase method during high temperature molecular dynamics (MD); initially, only the backbone protons were included, but subsequently they contained side chain protons. Second, the transition from the quartic equation to the Lennard-Jones non-binding potential was achieved in the final kinetic phase with the final maturation of the non-binding energy factor. Extract a conformational space as large as the actual given computer processor and disk limit, performing a total of 50 rounds of SA. Final minimization was performed with the convergence criterion for large derivatives not exceeding 0.01 kcal mol ^{-1 -1 -1} .

A total of 50 structures from the SA calculation are 154 distances (within 85 residues, 42 contiguous, 22 middle and 5 long ranges), 14 pairs Described using face constraints and 15 chiral constraints. Using a group clustering scheme based on RMS deviations, these structures were classified into 30 groups over an energy range of 396-5360 kcal / mol, with two groups containing 18 out of 50 structures. It was the lowest energy. Total minimum energy structure (MES)
The structural statistics of these five families with MES energies within 22 kcal / mol of are given in Table 1. The second lowest energy group (Group 3) was 79 kcal / mol for all MES. Through interactive graphic scrutiny, we find that many representatives of the high energy family (Groups 3-30) have unusually high internal energy terms or unusual features (eg, including broken bonds, backbones, and cysteine bridges). Knots). Abnormality on interesting structures, L-alpha -Cys ⁸ is observed in the structure 8 of group 1 which has been reversed D-α-Cys ⁸ (PDB accepted code 1 m 2c). This structure meets the criteria used to classify Family 1, but does not affect the final model and emphasizes the importance of maintaining proper chirality constraints during SA.

In FIG. 5, superposition of family 1 structures shows how well a given set of experimentally determined and converged in structural calculations with refined constraints. . As shown in Table 1, the paired skeleton and heavy atom RMS deviations between 14 structures (Gly ¹ was excluded due to its mobility) in Group 1 calculated over residues Cys ² -Cys ¹⁶ were respectively , 0.76 ± 0.31 and 1.35 ± 0.34 °
It is. In the SA approach, no specific constraints were used to bring the backbone amide bond into the trans configuration. In this family, all amide bonds at all residues of all structures are trans, and all 14 structures in family 1 are represented in FIGS.
High skeletal angle order parameter (Hyberts, S. et al. (1992); Palalaghy, P.
As it is seen from. Et al. (1993)), having a dihedral angle nearly identical backbone except for Gly ^1. However, this consistency backbone dihedral angles are not manually maintained at a level of chi ₁ side chain dihedral. High angular order parameters, such as those seen in FIG. 6C for residues 2, 5, 9 and 10, indicate that the entire structure of family 1 has an equivalent side chain rotational isomer state. All other residues, gauche ^- has a plurality of side chains rotamer state are generally classified as gauche + or trans classification. Table 1 shows the structure of group 1.
Assembles well, as determined by the range and magnitude of the energy contribution detailed in. Clearly, significant energy contributions to the maintenance of these structures are found in the non-bonded terms (both van der Waals and Coulombs).
These modeling tests are performed in vacuum, and any modulation of the effective dielectric constant, either by empirical solutions or by explicit solution inclusion, reduces the net electrostatic effect for this simple limiting peptide Just work.

[0114]

[Table 1]

[0115] I also Ah small peptides consisting of Example 6 three-dimensional solution structure example alpha-CT _x MII slightly 16 amino acid residues, it has a well defined three-dimensional solution structure. In addition to the two disulfide bridges that form a hydrophobic Cys bond, there are three helical regions in the structure that contribute to the formation of a very rigid three-dimensional conformation. The presence of such stable secondary structures and disulfide bridges makes multidimensional NMR effective in obtaining very high resolution of the peptide structure. The amino-terminal region has an almost complete turn of the α-helix (Cys ^{2 to} Ser ⁴ ), followed by φ = −89 ° (measured from the minimum energy structure) and ψ = + 132.
Followed by Asn ⁵ with a skeletal dihedral angle of ゜, thereby essentially forming a 90 ° fold. Helix substantially with two folding (Pro ⁶ ~Glu ¹¹⁾ is a major secondary structure composition of the peptide. This helix is His ¹² (φ = −
136 °, ψ = + 77 °; measured with minimum energy structure. End with), it is coordinated to the remaining C- terminal twisting 3 ₁₀ helix and (Ser ¹³ ~Cys ¹⁶⁾ in the N- terminal direction. These two folding related to Asn ⁵ and His ¹³ are probably an important residues for the peptide entire folding, the presence of such folding are shown in FIG. 4 (others have a negative shift Is a positive shift)
Further supported by the C _α proton shift data and the ³ J _{NH-C α} coupling constant of FIG. 2 (> 8.0 Hz coupling constant, while others have a coupling constant of <5.0 Hz). You. Possible functions of Asn ⁵ is that induces folding between N- terminal segment and the major alpha-helix; This feature is from various globular protein structure 45 215 alpha-into This is consistent with the research by Richardson and Richardson [Richardson, JS (1988)] on Rix. They found that for the α-helix, 3.5 per Asn at the N-cap position.
A prominent preference of 2.6: 1 was reported for Pro at 1: 1 and N-cap + 1 positions (Helix Initiator). Regarding the second turn around His ¹³
The function of this would be to carry Cys ³ and Cys ¹⁶ close together and form a second disulfide bridge.

The space-filling model of α-CTx MII shown in FIG. 7B is aimed at investigating the surface distribution of hydrophobic and hydrophilic side groups of the molecule. The hydrophobic residues are colored purple to distinguish them from yellow polar residues or red (positive) and blue (negative) charged residues. The flat surface located on top of the purple molecule is a distinct structural feature that represents a cluster of hydrophobic residues exposed to the solvent. This hydrophobic surface has Gly ¹ , C (excluding the N-terminal amino group).
ys ^2, is largely formed by a resign Fido bridge between Cys ^3, Leu ^15, Cys ¹⁶ and Cys ³ and Cys ^16. This flat surface may be important for ligand-receptor binding by hydrophobic interactions. There is another very distinct surface consisting of hydrophilic residues that are completely both polar and charged. Left of the model, red, blue and yellow clusters represent folded area of ^{His 2 (Glu 11 ~Asn 14)} . This highly charged surface, almost perpendicular to the hydrophobic surface, will be responsible for its initial recognition by the nAChR based on long-range electrostatic interactions. This potential receptor-binding interface consists of consecutive residues Glu ¹¹ , His ¹² , Ser ¹³ and A
consisting of sn ^14.

Example 7 Electrophysiology A cDNA clone encoding the nAChR subunit was obtained from S. Heinemann and
Supplied by D. Johnson (Salk Institute, San Diego, CA). cRNA was transcribed using a RiboMAX ^™ large scale RNA production system (Promega, Madison, WI). Diguanosine triphosphate (Sigma, St. Louis, MO) was used for the synthesis of capped transcripts according to the manufacturer's protocol. The plasmid structures of the mouse and rat nAChR subunits were described conventionally: α1, β1, γ
Α2; α3; α4; α7; α9; β2; β4.

The cRNA was injected with a Drummond 10 μl microdispenser (Drummond Scientific, Broomall, PA) using conventional techniques. It was fitted into a micropipette drawn from a glass capillary provided for a microdispenser. The pipette tip was crushed to an outer diameter of 22-25 μm and back filled with paraffin before attaching to a microdispenser. The cRNA was aspirated into a micropipette and 50 nl containing 5 ng of cRNA subunit was injected into each oocyte. In the case of muscle subunits, 0.5-5 ng of each subunit was injected.

Oocytes were harvested from Xenopus frog, cut into clumps of 20-50 oocytes, and OR-2 (82.5 mM NaCl, 2.0 mM KCl, 1.0 mM MgCl ₂ and 5 mM
580 U / ml type 1 collagenase (Worthington Bi. M HEPES, pH about 7.3)
ochemical, Freehold, NJ) in a 50 ml polyethylene tube (Sarst
edt). The tubes were incubated for 1-2 hours on a rotary shaker rotated at 50 rpm. At the midpoint of the incubation, the solution was replaced with a fresh collagenase solution. The oocytes are then washed 6-8 times with about 50 ml volume of OR-2 and ND-96 (96.0 mM NaCl, 2.0 mM KCl, 1.8 mM).
CaCl ₂ , 1.0 mM MgCl ₂ , 5 mM HEPES, pH = 7.1-7.5) / Pen / Strep / Gen (1
00U / ml penicillin G (Sigma), 100 μg / ml streptomycin (Sigma)
) And 100 μg / ml gentamicin (Gibco BRL, Grand Island, NY). Oocytes were visually inspected and only healthy oocytes were transferred to a second dish containing ND96 and antibiotics. Oocytes were injected 1-2 days after collection and recordings were made 1-7 days after injection.

The injected oocytes were made from Sylgard, placed in a recording chamber of about 30 μl consisting of cylindrical wells (about 4 mm diameter × 2 mm depth), and at a rate of about 1 ml / min, ND96 or ND96 containing 1 μM atropine (ND96A)
Perfused by gravity. The total toxin solution also contained 0.1 mg / ml bovine serum albumin (BSA) to reduce non-specific adsorption of peptides. The perfusion medium was supplied by a distribution valve (SmartValve, Cavro Scientific Instruments, Sunn
yvale, CA) and a series of three-way solenoid valves (Model 161T031, Neptun
e Research, Northboro, Mass.) could be switched to containing peptide or acetylcholine (ACh). ACh-gate current is a two-electrode voltage-clamp amplifier for "fast" clamping (model OC-725B, Warner Inst
rument Corp., Hamden, CT) set and at maximum (× 2000) clamp gain. Glass microelectrodes drawn from fiber-filled borosilicate capillaries (1 mm OD x 0.75 mm ID, WPI Inc., Sarasota, FL) and filled with 3M KCl served as voltage and current electrodes. The resistance is 0.5 to 5 MΩ for voltage,
The value for the current was 0.5 to 2 MΩ. The membrane potential is clamped at -70 mV,
The current signal recorded through the virtual earth was low-pass filtered (5 Hz cut-off) and digitized at a sampling frequency of 20 Hz. The solenoid perfusion value is
Solid state relay (Model ODC5, Opto 22 in PB16HC digital I / O backplane)
, Temecula, CA). A / D conversion and digital control of the solenoid valves were performed on a Lab-LC or Lab-NB board (National Instruments, Austin, TX) in a Macintosh (Quadra 630 or IIcx) computer. The computer communicated with the distribution valve through its serial port. Data acquisition and activation of dispensing and solenoid valves are performed using the graphical programming language La
It was controlled automatically by a home-made virtual memory device composed of bVIEW (National Instruments, Austin, TX).

To apply a pulse of ACh to the oocytes, the perfusate was changed to one containing ACh for 1 second. This was done automatically at 1-5 minute intervals. The shortest time interval was chosen such that the regeneration control response was obtained without observable degradation. This time interval was dependent on the nAChR subtype to be tested. The concentration of ACh was 1 μM for oocytes expressing α1βγδ, 1 mM for α7 and 300 μM for all others. ACh was diluted with ND96A for all but α7; in the case of α7, the diluent was ND96. In the control response, the ACh pulse was ND96 (for α7) or (for all others)
Perfusion with ND96A preceded. Atropine was not used in α7-expressing oocytes because it has been shown to be an antagonist of these receptors (Gerzanich, 1994). For the response to the toxin (test response), the perfusion solution was switched to that containing the toxin while maintaining an interpulse of ACh. Toxin was continuously perfused until equilibrium was achieved. Total A is assumed because less, if any, bound toxin is washed away in a short time (<2 seconds) in response to the peak.
The Ch pulse contained no toxin. The peak amplitude of the ACh-gate current response was measured by virtual storage. All recordings were made at room temperature (about 22 ° C).

The injected oocytes were voltage clamped at a membrane potential of -70 mV and ACh
The response to was measured every 5 minutes. After measuring some control responses at -70 mV, the membrane potential was stepped to a 1 minute test potential, starting 10 seconds before application of ACh. The membrane potential returned to -70 mV between each test. The test potential is sequentially from -90 mV to -1 in 10 mV steps.
Changes were made until the final test at 0 mV followed by a membrane potential of +10 mV. After the test series, the perfusion solution was switched to that containing the toxin. After reaching equilibrium, the test series was repeated.

All data analysis was performed on an Ap containing a 486 Intel processor and Windows 3.1 ^™.
Prism software running on the ple Power Macintosh 6100/66 device (Graph
hPad Software Inc., San Diego, CA). The test response was normalized using the average of the last three control responses to give a "% response" or "% block". Each data point in the dose response curve represents the mean ± SEM of at least three oocytes. The dose response curve was fitted to the equation:% response = 100 / (1+ (toxin) / IC ₅₀ ) ^nH where n _H is the Hill coefficient.

The kinetic trace shows data from only one experiment. All experiments were repeated with at least three oocytes. The kinetic parameters reported are the average from all experiments performed on that subtype. The data describing toxin inhibition over time fit the equation described in the form of a simple exponential association: Y = (1-e- ^kt ). Data describing the recovery over time from toxin inhibition at low toxin concentrations fit the equation described for a simple exponential dissociation form: Y = e- ^kt . The data describing the recovery over time from toxin inhibition at high toxin concentrations fit the equation described for the complex exponential dissociation form: Y = (1- (1-e- ^kt ) ² ).

Example 8 MII Subtype Specificity Using the nAChR subunit expressed in Xenopus oocytes, the efficacy of MII in various nAChR combinations was quantified. Each subtype was inhibited by MII in a dose-dependent manner, but had widely different potencies (FIG. 9). α-Conotoxin MII most potently inhibited the α3β2 combination with less than 200-fold to> 10000-fold potency in other subtypes (Table 2).

[0126]

[Table 2]

Many small molecules block nACh by opening open channels.
Acts as a non-competitive inhibitor of R. The open channel block of nAChR generally depends on the membrane potential. The inventors investigated whether the MII block of α3β2 nAChR was voltage-dependent. To accomplish this, the acetylcholine response of α3β2 nAChR to acetylcholine was measured over a range of membrane potential both in the presence and absence of MII at its approximate IC ₅₀ (500 pM). As shown in FIG. 2, blocking with α-conotoxin MII showed no voltage dependence over the range tested (-100 mV to +10 mV).

To further investigate the mechanism of action of MII, different off-rates of dihydro-β-erythroidin (DHβE) and α-conotoxin MII were utilized. DHβE is nA
It has been shown to be a competitive antagonist of ChR. Blocking of α3β2 nAChRs expressed in oocytes by DHβE is well reversed after 4 min efflux (FIG. 11A). In contrast, α3β2 nAChRs recover relatively slowly from the MII block following toxin shedding (FIG. 11C). To test whether α-conotoxin MII acts at the same site as DHβE, we pre-applied DHβE to α3β2 nAChR. Then we applied both DHβE and α-conotoxin MII. If α-conotoxin MII and DHβE act at the same site, pre-application of DHβE will protect the receptor from blocking by α-conotoxin MII. In this case, recovery from the block would be followed by a faster time-course block with DHβE alone. If pre-application of DhβE does not prevent binding of α-conotoxin MII to the receptor, then recovery from the block by applying both DhβE and MII will result in α-conotoxin MII
It would be expected to follow a slower temporal block by itself. As shown in FIG. 11B, the recovery over time following the shedding of DHβE and α-conotoxin MII was similar to that of shedding of DHβE alone, indicating that DHβE prevents MII binding.

At low concentrations (eg, around the IC ₅₀ ) of α-conotoxin MII, recovery from the block following toxin shedding is time-course well fitted by a single exponential extinction function consistent with simple bimolecular dissociation (FIG. 12). However, as the concentration of α-conotoxin MII approaches saturation, there is an earlier lag in recovery after toxin shedding. This lag is greatest at sufficiently saturating concentrations of α-conotoxin MII (data not shown). Recovery over time from a block at saturating concentrations of MII contradicts simple bimolecular dissociation (see dashed line in FIG. 13). Instead, the time course is that each receptor is AC
It fits well with a model hypothesized to have two binding sites that can be occupied by either h or α-conotoxin MII. Also, in this model, occupancy of any binding site by α-conotoxin MII is sufficient to block receptor-induced opening. Thus, at saturating concentrations, both binding sites are occupied by MII. Upon toxin shedding, each binding site becomes unoccupied following simple bimolecular dissociation over time. Because both binding sites must be unoccupied by toxin for agonist activation of the receptor to occur, an early lag in functional recovery following a saturated toxin block is observed.
The value of K _off calculated using this model under saturating toxin concentration was compatible with that calculated under non-saturating conditions making the model valid (Table 3).

[0130]

[Table 3]

Example 9 Kinetics of Interaction with Receptor Subtypes The affinity of a ligand is determined by its minimal rate of association (k _on ) and dissociation (k _off ). When MII blocks the current of the α3β2 receptor, as determined by measurements on voltage-clamped Xenopus oocytes expressing nAChR, its on-rate is relatively fast (k _on = 4 × 10 ⁸ min). ^-1 M ^-1 ); its _off speed is slow (k _off = 0.13 min ^-1 ). This results in a high affinity interaction between α-conotoxin MII and α3β2 nAChR (K _d = 3.3 × 10 ⁻¹⁰ M).
. If either the α3 or β2 subunit is replaced with a different subunit, a substantially lower affinity is observed. It shows that when the α3 subunit is exchanged for α4, the resulting α4β2 receptor is blocked by MII with an 800-fold higher _Kd . If the β2 subunit is switched (resulting in α3β4), a 8000-fold reduction in MII efficacy is seen. Furthermore, these two
The most striking results are obtained when the M _on and k _{of f} values for the two types of receptors (α3β4 and α3β2) are measured. Despite a nearly four order of magnitude reduction in potency, α-conotoxin was found to be α3β4-versus-α3β.
When blocking 2, k _off hardly changes even if it _exists . Thus, differences in potency is essentially described by a decrease corresponding in toxin k _on. Conversely, when the α subunit is exchanged (ie, to obtain α2β2 or α4β2), there is a significant increase in k _off for MII, the quantification of which currently exceeds our assay capacity. These results, k _on the toxin is controlled largely by the determinant on β2 subunit, whereas, .alpha.3 subunit suggests that it is also possible to have a major determinant for controlling the k _off.

Example 10 Chimera Affinity for α3 / β2 Subtype To further investigate the kinetic interaction of the toxin with the receptor, a chimera of α-conotoxin MII was constructed. A series of peptides from the venom of Conus aulicus, different from α-conotoxin MII, have been isolated, which prefer the α3β2 interface over the α3 / β4 interface. One of these peptides is α-conotoxin AuIA. 12
The presence of either histidine or asparagine at the position is highly conserved in all conopeptides. The conserved sequence of each auricus peptide is used in the design of the MII chimera shown in Table 4, wherein the four amino acid (9-12) strings of MII are replaced by FATN. In general, the substitution of these four amino acids results in altered selectivity and affinity for the nicotine receptor. Compared to MII, the affinity of the chimera for the α3 / β2 interface was 100-fold lower (K _d varied from 330 pM for wild type to 500 nM for chimera), but k _off varied by factors less than 2. did. Thus, four amino acid substitutions do not affect the toxin residues that determine the off-rate (otherwise must be for the peptide), but do affect the residues that determine the on-rate. Amino acid fragment 9
Another chimera in which １２12 was substituted for one peptide for another was synthesized.

Example 11 Cyclic Peptide Based on the MII Motif To test the importance of the HLEH fragment of MII (amino acid residues 9-12 of SEQ ID NO: 1), several cyclic peptides based on this structure were prepared using the standard Constructed using FMOC chemistry. These are cyclic cys-HLEH-cys (SEQ ID NO: 4),
It includes cyclic A-HLEH-A linked via an amide bond (SEQ ID NO: 5) and cyclic HLEH linked via an amide bond (amino acid residues 9-12 of SEQ ID NO: 1). These small peptides are screened for biological activity according to the procedure of Example 8.

[0134]

[Table 4]

Example 12 Dock-and-Lock Model A comparison of data on the interaction between MII and nAChRs (Examples 9 and 10) demonstrates a two-step “dock” to explain the significant subtype specificity of MII. -Emerges as an "and-lock" mechanism. The data is the “face” of the two MII estimates
Interact with each of its binding sites on the receptor, and suggest that a double interaction can provide a general mechanism for the observed high receptor subtype specificity and affinity. Thus, it has been proposed that α-conotoxin MII has two interacting surfaces, which are referred to as the “docking surface” and the “locking surface”. The docking surface interacts relatively quickly with sites on the β2 subunit,
That site is referred to as the "docking site". The second phase of the toxin interaction involves a different surface of the toxin, the "locking surface". This locking surface binds to a site on the α3 subunit, which is referred to as a “locking site”. The absence of the docking interaction, - k _{o n} for the "locking surface locking site" interaction, is quite slow. However, if the peptide docks (via the β2 docking site), then the locking phase proceeds very quickly (FIG. 14).
A and 7B).

The above data indicate that a switch in the receptor subunit can result in orders of magnitude reduction in the affinity of MII without affecting its k _off (this indicates that the locking site is still intact) , Meaning that the change in affinity is primarily due to a change in docking interactions). Thus, the results with the FATN-αMII analog (Example 11; SEQ ID NO: 3) and the kinetics obtained with the wild-type toxin acting on the α3β4 nicotine receptor show that the docking surface of the toxin has a wedge ) Is located on the hydrophilic side, implying that this hydrophilic surface interacts with the β2 subunit at an early on-time. On the other side of the wedge, characterized by hydrophobic residues exposed to the solvent, is the attractive locus for the locking surface of α-conotoxin MII, which is located at a site on the α3 subunit of the neuronal nicotinic receptor target of this peptide. It is postulated to interact with high affinity. α-conotoxin MII interacting with α3β2 and α3β4 receptors, and FATN-α-
A schematic diagram of the conotoxin MII analog (SEQ ID NO: 3) is shown in FIG.

The existence of two distinct interacting surfaces that lead to distinct docking-and-locking interactions as described herein is a feature of the general design of many Conus peptides, A novel paradigm for achieving subtype-selective interaction with unit receptors can be shown. The term "Janus-ligand" is indicated to refer to a binding molecule with two distinct interacting faces (from the two facing Roman gods of the beginning, Janus). α-Conotoxin MII may be the original Janus ligand.

Example 13 Synthesis of Organic Molecules as Ligand The HLEH fragment of MII appears to be essential in determining the activity and selectivity of MII conopeptide. Computer evaluation of HLEH fragments
Attempts to develop an organic framework for use in developing small molecules. The spatial alignment of the HLEH fragments is studied computationally using vector analysis. The database is then searched in the database to find organic fragments with the appropriate spatial requirements and geometry to be used as the backbone from which the organic model is synthesized. The presence of two physically distinct peptide surfaces (hydrophilic and hydrophobic) on the MII conopeptide and their interaction with the α and β receptor subunits is due to the receptor in different combinations of α and β subunits. 2 shows a specific strategy for designing ligands selective for.

[0139] It will be appreciated that the methods and compositions of the present invention may be incorporated in various illustrative forms, only a few of which are disclosed herein. It will be apparent to one skilled in the art that other embodiments do not depart from the scope of the invention. Thus, the described embodiments are illustrative and should not be construed as limiting.

Reference list Badansky et al. (1966). Chem. Lnd. 38: 1597-98. Boelens, R. (1988). Mol. Struct. 173: 299. Braunschweiler, L et al. (1983). J. Magn. Reson. 53: 521-528. Cartier, GE et al. (1996). J. Biol. Chem. 271: 7522-7528. Dauber-Osguthorpe. P. et al. (1988). Proteins 43-47-Davis, DQ et al. 1985), J. Am. Chem. Soc 107: 2820-2821. Davis, J. et al. (1993). Biochemistry 32: 7396-7405. Deslarlais, RL (1997) "Generation and use of Three-Dimensional D.
atabases for Drug Discovery "in Pracrical Applicarion of Computer-Aided
Drug Design, Charifson, eds. Marcel Dekker, New York. Ecker et al. (1995) Bio / Technology 13: 351-360. Fainzilber, M. et al. (1994) Biochemistry 33: 9523-9529. Gerzanich, V. et al. (1994). Mol. Phormacol. 45: 212-220. Gonzales, C et al. (1991). J. Mogn. Reson. 91: 659-664. Gray, WR et al. (1981). J. Biol. Chem. 256: 4734-4740. Gray, WR et al. (1983). J. Biol. Chem. 258: 12247-12251. Guddat, LW et al. (L996). Biochemistry 33: 11329-11335. Gund, P., Pharmacophoric pattern searching and receptor mapping, Ann. Re
Chem. 14: 288 (1979) Hannr, RM et al. (1994). Biochemistry 33: 14058-14063. Harvey, SC et al. (1997). Mol. Pharmacol. 51: 336-342. Havel. 1991). Prog. Mol. Biol. Biophys. 56: 43-78.

Horiki, K. et al. (1978). Chemistry Letters 165-68. Hu, SH et al. (1996). Structure 4: 417-423. Hyberts, S. et al. (1987). Eur. J. Biochem. 164; 625-635. Hyberts, S et al. (1992). Protein Sci. 1: 736-751. Ieener, J. et al. (1979) J. Chem. Phys 71: 4546-4553. Johnson et al. (1993). "Peptide Turn Mimetics. "in Biotechnology and Phar
Ed. macy, Pezutto et al., Chapman and Hall. New York. Kaiser et al. (1970). Anal. Biochem. 34: 595. Kapoor (1970). J. Pharm. Sci. 59: 1-27. (1989). Biochemistry 28: 4853-4860. Kohno, T. et al. (1995). Biochemistry 34: 10256-10265. Kornreich, WD et al. (1986). U.S. Patent No. 4,569,967. Kreienkamp, HJ et al. (1994). J. Biol. Chem. 269: 8108-811 4. Ku, TW et al. (1995). J. Med. Chem. 38: 9-12. Kumar, A. et al. (1980). Biochem., Biophys, Res. Commun. 95: 1-6 Kutak, JM et al. (1997).,] Neurosci. 17: 5263--5470. Lauri, G. and PA Bartlett (1994). J. Comp. Aid. Mol. Des. 8: 5.
1-66. Macura, S. et al. (1981). A. Magn. Reson. 43: 959-281. Martinez, JS et al. (1995). Biochemistry 34: 14519-14526. McIntcesh. JM et al. (1982). Arch. Biochem. Biophys 218: 329-3
34. Mclntosh, JM et al. (1994). J. Biol. Chem. 269: 16733-167.
39. Methoden der Organischen Chemie (Houben-Weyl): Synthese von Peptiden,
E. Wunsch (Ed.), Georg Thieme Verlag, Stuttgart, Ger. (1974). Monje, VD et al. (1993). Neuropharmcology 32: 1141-1149. Muller, L. (1987). J. Magn. Reson. 72 : 191 196. Myers, RA et al. (1991). Biochemistry 30: 9370-9377. Nilges, M. et al. (1988). FEBS Lett. 239: 129-136. Nishiuchi Y. et al. (1993). acid-con
taining peptides by the Boc strategy. Int. J. Protein Res. 42: 533-
538. Pallaghy, P. et al. (1993). J. Mol. Biol. 234: 405-420. Pardi, A. et al. (1984). J. Mol. Biol. 180: 741-751. Pardi, A. et al. 1989). Biochemistry 28: 5494-5501. Persidis et al. (1997). Bio / Technology 15: 1035-1036.

Piantini, U. et al. (1982), J. Am. Chem. Soc. 104: 6800-6801. Piotto, M. et al. (1992). J. Biomol. NMR 2: 661-665. Rance, M. (1983). Biochem. Biophys. Res. Commun. 117; 479-485. Remington's Pharmaceutical Sciences), 18th edition, Mack Publishing Co., Easton.
Richardson. JS et al. (1988). Science 240: 1648-1652. Rivier, JR et al. (1978). Biopolymers 17: 1927-38. Rivier, JR et al. (1987). Total synthesis and further characterizatio.
n of the gamma-carboxyglutamate-containing 'sleeper' peptide from Conu
26: 8508-8512 Roberts et al. (1983). The Peptides 5: 342-429. Rothemund. S et al. (1996). Biopolymers 39: 207-219. Ruckcer. S. (1989) Mol Phys. 68. : 509-517. Sambrook, J et al. (1979). MolecuSar Cloning: A Laboratory Manual, 2nd edition,
Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Sargent, PB (1993). Ann. Rev. Neurosci. 16; 403-443. Schroder & Lubke (1965). The Peptides 1: 72-75, Academic Press.
(1997). Biochemistry 36: 15693-15700. Sine. SM et al. (1995). Neuron 15: 205-21. Skalicky, J et al. (1993). Protein Sci. 2: 1591-1603. Statrs, DJ et al. (1982) J. Magn. Reson. 48: 286-292. Stewarto and Young, Solid-Phase Peptide Synthesis, Freeman & Co. San.
Francisco, CA (1969). Sun et al. (1993). Gene 137: 127-132. Tegenfeldt, J. et al. (1979). J. Magn. Reson. 36: 453-457. Ullian, EM et al. (1997). J. Neurosci. (In press). Utkin, YN et al. (1994). Toxicon. 32: 1153-1157. Valc et al. (1978). U.S. Patent No. 4,105,603. Wagner, G. et al. (1987). J. Mol. Biol. 196: 611-639. Wilshart7 DS et al. (1991). J. Mol. Biol. 222: 311-333. Wuthrich, K. et al. (1983). J. Mol. Biol. 169: 949-. 961. Wuthrich, K. et al. (1986). NMR of Proteins and Nucleic Acids. Wiley-In.
terscience, New York. Zafaralla, GC et al. (1988). Biochemisuy 27: 7102-7105. Zhou LM et al. (1996a). Synthetic Analogues of Contryphan-G: NMDA Ant.
agonists Acting Through a Novel Polyamine-Coupled Site. J. Neurochem. 66: 620-628.

US Pat. No. 3,972,859 (1976). US Pat. No. 3,842,067 (1974). US Pat. No. 3,862,925 (1975). US Pat. No. 4,105,603. No. 5,514,774 (1996). US Pat. No. 5,550,050 (1996). US Pat. No. 5,571.698 (1996). US Pat. No. 5,595,972. No. 5,780,433 (1998). PCT application number WO 92/19195. PCT application number WO 94/25503. PCT application number WO 95/01203. PCT application number WO 95/05452. PCT PCT application number WO 96/02286. PCT application number WO 96/40871. PCT application number WO 96/40959. PCT application number WO 97/12635. PCT application Number WO 97/0 300.

[Sequence list]

[Brief description of the drawings]

The file of this patent contains at least one drawing printed in color. Copies of this patent color drawing will be provided by the Patent and Trademark Office at the necessary fee.

FIG. 1 shows acetylcholinergic currents of alanine substitutions in the conopeptide MII in voltage-clamped Xenopus oocytes whose analogs express cloned rat α3 and β2 subunits. Indicates the effect on the ability to block.

FIG. 2 shows a NOESY spectrum (250 ms mixing time) of α-CT _x MII in H ₂ O, illustrating a continuous NOE bond between the adjacent α and the amide proton. d _{alpha N} NOE cross peaks are connected by a line, it is labeled with residue number. Two traces between residues Cys ² ~Asn ⁵ and Val ⁷ ~Cys ¹⁶ is indicated by binding continuous thereof cut by the P ro ^6.

Figure 3, alpha-CT the _x MII amino acid sequence, disulfide bridges and indicating the disulfide cross-linking patterns, ³ J _NH-C α _H as measured in the short to medium range NOE cross peaks as well as experimental Shown with coupling constants. Down arrow <is of 5.0 Hz, upward they ³ J constant> ³ J constant 8.0H
z. Filled bars represent NOE crossing peaks, thicknesses classify strong (thick) to weak (thin) NOE intensities, and widths indicate short to medium distances along the linear array. *** represents overlapping NOE cross peaks.

FIG. 4 shows the difference in the chemical shift of the α proton (ppm) between the observed value from α-CT _x MII in H ₂ O and the value of the random coil, plotted against residues. did.

FIGS. 5A-B show the superposition of the 14 lowest energy states of α-CT _x MII calculated from simulated annealing.

FIGS. 6A-C show angle order parameters for scaffolds obtained from the 14 lowest energy structures, plotted against residue number. 6A shows φ, FIG. 6B shows Δ, and FIG.

FIG. 7 shows the surface (A, B) and backbone (C, D, E) notation of α-CT _x MII.PnIA and GI. The X-ray crystal structure of α-CT _x PnIA (A) and the lowest energy structure of α-CT _x MII (B) are as follows: the hydrophobic residue is purple, the polar side chain is yellow, and the charged side chain is red (positive). ) And blue (negative). The skeleton structures of α-CT _x Pn IA (C), α-CT _x MII (D), and α-CT _x GI (E)
The ribbon and the dots having the same color code are displayed with the surface distribution to illustrate similarity or difference. Arrows point to terminal residues, blue for carboxyl groups and yellow for amino groups.

FIG. 8 lists the coordinates of α-conotoxin MII. These coordinates are under the access code 1m2c, Brookhaven Protein Data Bank, Upton, NY
Registered in 19973.

FIG. 9 shows the selective block of α3β2 nAChR by α-conotoxin MII. Oocytes expressing different nAChR subunit combinations were voltage clamped and the response to ACh was measured at different α-conotoxin MII concentrations. Data are means ± SEM for at least 3 oocytes at each concentration
Represents

FIG. 10 shows that blocking of α3β2nAChR by α-conotoxin MII is independent of membrane potential. Oocytes expressing α3β2nAChR were voltage clamped and the response to ACh was measured over a range of membrane potentials. The peak intensities in the absence (open circles) and presence (closed triangles) of 500 pM α-conotoxin MII are plotted against membrane potential. The scaled toxin response (reverse black triangle) is also shown.

FIGS. 11A-C show that a competitive antagonist, dihydro-β-erythroidin (DHβE), protects α3β2nAChR from blocking by α-conotoxin MII. Oocytes expressing α3β2nAChR were voltage clamped and the response to ACh was measured. In FIG. 11A, 500 μM DHβ
E was added for 5 minutes. The oocytes were then continuously perfused with buffer without DHβE. In FIG. 11B, 500 μM DHβE was added for 5 minutes, followed by
500 μM DHβE and 20 nM α-conotoxin MII were co-added for 2 minutes. The oocytes were then continuously perfused with a buffer without toxin.

FIGS. 12A and B show the kinetics of α3β2 and α3β4 nAChR blocking by α-conotoxin MII. FIG. 12A shows 500
To the ACh of voltage clamped oocytes expressing α3β2nAChR, measured during the addition of pM α-conotoxin MII (first graph) and subsequent washing from toxin (second graph) Show the reaction. FIG.
Response of voltage clamped oocytes expressing α3β4nAChR to ACh measured during the addition of M α-conotoxin MII (first graph) and subsequent washing from toxin (second graph) Is shown.

FIGS. 13A and B show the kinetics of recovery of α3β2 and α3β4 nAChR from block by saturating concentrations of α-conotoxin MII.
FIG. 13A shows voltage clamped oocytes expressing α3β2nAChR and their response to ACh. 2 μM α-conotoxin MII was added for 5 minutes. The oocytes were then continuously perfused with a buffer without toxin. FIG.
3B shows voltage clamped oocytes expressing α3β4nAChR and their response to ACh. 2.5 mM α-conotoxin MII was added for 5 minutes. The oocytes were then continuously perfused with a buffer without toxin. The data was fitted with a two-site model (solid line). The dashed line represents the expected one-index recovery kinetics.

FIGS. 14A-C show the ligand / receptor interaction MII “Dock & Lock” model. In this figure, MII binds to the boundary between α and β subunits. MII interacts with the β subunit at very fast k _on and k _off , and thus with moderate affinity. Conversely, MII interacts with the α subunit at a very slow k _on and k _off and thus with a mild affinity. As shown in FIGS. 14A and 14B, MII approaches the receptor and binds to the β subunit very quickly (docking interaction). However, this rapid binding to the β subunit, which is close to the α subunit binding site, results in an accelerated rate of MII interaction with the α subunit and subsequent blocking of the receptor. FIG. 14C shows that once MII binds to the α subunit, it is a very slow k _off (rocking interaction). Thus, by targeting the subunit boundaries of the nicotinic receptor, MII
Converts two relatively mild binding interactions into a very strong and specific interaction in one molecular form of the neuronal nAChR.

──────────────────────────────────────────────────続 き Continuation of front page (81) Designated country EP (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE ), OA (BF, BJ, CF, CG, CI, CM, GA, GN, GW, ML, MR, NE, SN, TD, TG), AP (GH, GM, KE, LS, MW, SD, SZ, UG, ZW), EA (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), AL, AM, AT, AU, AZ, BA, BB, BG, BR, BY, CA, CH, CN, CU, CZ, DE, DK, EE, ES, FI, GB, GE, GH, GM, HR, HU, ID, IL, IS, JP, KE, KG, KP , KR, KZ, LC, LK, LR, LS, LT, LU, LV, MD, MG, MK, MN, MW, MX, NO, NZ, PL, PT, RO, RU, SD, SE, SG, SI, SK, SL, TJ, TM, TR, TT, UA, UG, UZ, VN, YU, ZW (71) Applicant Salk Institute North Holland, North Holland Pines Law, La Hora, California 92037 USA No. 10010 (71) Applicant Cognitives Inc.84108 U.S.A., Salt Lake City, Utah, No. 421, Suite 201 (72) Inventor Key June, United States 44120 Sheikh Heights, Ohio No. 3136, Huntington Road (72) Inventor Bardello Eme Olivera, Salt Lake City, Utah 84108 U.S.A. Ian Avenue 1370 (72) Inventor Jean E. Revier 9637 Black Gold Road, La Hora, California 92037 United States 9274 Inventor Stephen Sea Cober, United States 921024 Nerd Road, Encinitas, California 92024 No. (72) Inventor Gregory S. Shen, United States 84109 Salt Lake City, Utah, Lakeline Circle 2218 No. (72) Inventor Jay Michael Macintosh United States 84108 Utah, Salt Lake City, South・ 2000 East 1151 (72) Inventor G. Edward Cartier United States 84109 Utah Salt Lake City, East Stringham Avenue 2629, Apartment Number B-216 (72) Inventor Doju・ Yoshikami America Country 84102 Utah Salt Ray click City, East 600 South 1074 No. F-term (reference) 4B024 AA01 AA12 BA63 CA04 DA02 GA18 HA01 4H045 AA10 BA17 DA50 EA28 EA50 FA74 5B049 BB00 EE08 5B075 ND20 QM08 UU19 UU26

Claims (32)

[Claims] (Claim 1) Xaa-Cys-Cys-Xaa-Xaa ₁ -Xaa ₂ -Xaa-Cys-Xaa ₃ -Xaa-Xaa ₄ -Xaa ₅ -Xaa-Xaa-Xaa-Cys
Wherein, Xaa, each of Xaa _1, Xaa _2, Xaa ₃ and Xaa ₄ are, naturally, be an amino acid selected from the group consisting of modified or unnatural amino acids, Xaa ₅ is His or Asn] a A substantially pure peptide having. 2. The peptide according to claim 1, wherein the C-terminal is amidated. 3. The peptide according to claim 1, wherein Xaa ₁ is Asn or His. 4. A peptide according to claim 1, wherein wherein Xaa ₂ is Pro or hydroxy -Pro. 5. The peptide according to claim 1, wherein Xaa ₃ is His. 6. The peptide according to claim 1, wherein Xaa ₄ is Glu. 7. Xaa ₁ is Asn or His and Xaa ₂ is Pro or hydroxy-
The peptide according to claim 1, which is Pro. 8. The peptide of claim 7 wherein Xaa ₁ is Asn. 9. The peptide according to claim 7, wherein Xaa ₄ is Glu and Xaa ₅ is His. A is is 10. Xaa _3, peptides of claim 7, wherein Xaa ₄ is Glu. 11. The peptide according to claim 1, wherein the biological activity of said peptide is substantially the same as the biological activity of α-conotoxin MII. 12. The peptide according to claim 7, wherein the biological activity of said peptide is substantially the same as the biological activity of α-conotoxin MII. 13. The peptide according to claim 10, wherein the biological activity of said peptide is substantially the same as the biological activity of α-conotoxin MII. 14. A peptide analog, peptidomimetic or mimetic of α-conotoxin MII that is specific for one or more subtypes of neuronal nicotinic acetylcholine receptor (neuronal nAChR). 15. The peptide analog, peptide mimetic or mimetic according to claim 14, wherein said specificity is for the α3β2 subtype of neuronal nAChR. 16. The peptide analog, peptide mimetic or mimetic according to claim 14, wherein said specificity is for the α3β4 subtype of neuronal nAChR. 17. The peptide analog, peptide mimetic or mimetic according to claim 14, wherein said specificity is for the α2β2 subtype of neuronal nAChR. 18. A method for screening for neuronal nicotinic acetylcholine (nAChR) receptor antagonist activity and receptor subtypes, comprising binding of α-conotoxin MII and the compound to be screened to various subunits of nAChR. And determine the on-and-off speed for
The method comprising comparing and-off rates. 19. The method of claim 18, wherein the nAChR subunit is α3β2 or α3β4. 20. A derivative, peptide analog, peptidomimetic or mimetic of α-conotoxin MII that selectively modulates biological activity at nAChR. 21. A method for determining the three-dimensional structure of an α-conotoxin, analog, comprising: (a) obtaining NMR spectroscopic data for the peptide; (b) providing three-dimensional structural coordinates for the composition of MII; (C) determining the three-dimensional structure of alpha-conotoxin by analyzing the data with reference to previous structural coordinates using molecular replacement. 22. The following: (a) binds to neuronal nicotinic acetylcholine receptor (nAChR); (b) inhibits binding of ligand to neuronal nAChR; and / or (c) is mediated by the natural ligand of nAChR. 22. The method of claim 21, further comprising testing the compound so identified for its ability to inhibit a given biological function. 23. A neuronal nicotinic acetylcholine receptor (nAChR)
A) providing a coordinate defining the three-dimensional structure of MII or a part thereof; (b) with respect to the advantage of interacting with one or more selected functional groups Characterizing the points associated with the three-dimensional structure; (c) providing a database of one or more dandiate compounds; and (d) from the database, the points that best interact with the three-dimensional structure. The method comprising identifying those compounds having a fitting structure. 24. The following: (a) binds to neuron nAChR; (b) inhibits binding of MII to their natural or unnatural ligand, thereby; and / or (c) is mediated by MII 24. The method of claim 23, further comprising testing the compound so identified for its ability to inhibit a biological function. 25. A machine readable data storage medium that, when using a machine programmed with instructions for using the data, is capable of displaying a graphic three-dimensional representation of a molecule containing an MII peptide or a portion thereof. The storage medium comprising data storage material encrypted with possible data. 26. A machine-readable data storage medium, wherein the stored protein is based on the coordinates of FIG. 8 or less than 1.5 ° when using a machine programmed with instructions for using the data. The storage medium comprising a data storage material encrypted with machine readable data capable of displaying a graphic three-dimensional representation of α-conotoxin or a portion thereof based on coordinates having root mean square deviation therefrom with respect to the skeletal atoms. 27. A machine-readable data storage medium, comprising: a machine programmed with instructions for using a first set of machine-readable data and a second set of machine-readable data. Data storage material encrypted with a first set of machine readable data that, when combined with a second set of readable data, may determine at least a portion of coordinates corresponding to the second set of machine readable data. The storage medium comprising: 28. A method for displaying a three-dimensional representation of the composition of MII, comprising: (a) reading data stored on the machine-readable medium of claim 25, wherein the data is defined by the data. 26. Provide a machine programmed with instructions for using said data to display a graphic three-dimensional representation of a protein or protein-ligand complex or portion thereof, and mounted with a machine-readable storage medium of claim 25; (B) the method comprising the machine reading the data and displaying the three-dimensional representation. 29. The neuronal nicotinic acetylcholine receptor (nAChR)
And (b) graphically displaying a three-dimensional notation based on coordinates defining the three-dimensional structure of MII or a portion thereof; and (b) binding to the protein. (C) providing a knowledge base of one or more candidate replacement sites; and (d) from the knowledge base, The method comprising identifying one or more substitution sites that can be used to replace one or more selected moieties and retain at least a portion of a ligand binding affinity for the protein. 30. The method of claim 29, further comprising testing said compound for its modulatory activity of said neuronal nAChR. 31. A neuronal nicotinic acetylcholine receptor (nAChR)
A) providing a coordinate defining the three-dimensional structure of MII or a portion thereof; and (b) characterizing a point associated with the three-dimensional structure, wherein the protein is characterized by: (C) identifying one or more portions of a ligand known to bind to a protein proximate to the characterized point, with respect to the advantage of one or more functional group interactions with (D) providing a knowledge base of one or more molecular fragments or molecules; (e) from the knowledge base one or more that permit linking of the preferred points identified in (b) to a portion of the ligand. Identifying a fragment or molecule; and (f) one or more fragments so identified in an orientation and position selected to allow the modified ligand to bind to the protein. The method others that involves modifying the ligand structure by covalent binding to it of the molecule. 32. The method of claim 31, further comprising testing said compound for its modulatory activity of said neuronal nAChR.