The x-ray crystal structure of human BACE or BACE-like proteins is useful for solving the structure of other molecules or molecular complexes, and identifying and/or designing potential modifiers of human BACE activity.

Description of the Invention

SUMMARY OF THE INVENTION

In one aspect, the present invention provides a crystal of beta secretase. In one embodiment, the crystal has tetragonal space group symmetry P4.sub.32.sub.12, and preferably, the beta secretase includes glycosylated beta secretase. In another embodiment, the crystal of beta secretase includes a unit cell defined by the dimensions a, b, c, .alpha., .beta., and .gamma., wherein a is about 94 .ANG. to about 134 .ANG., b is about 94 .ANG. to about 134 .ANG., c is about 170 .ANG. to about 210 .ANG., and .alpha.=.beta.=.gamma.=90.degree.. Preferably, the crystal has tetragonal space group symmetry P4.sub.32.sub.12. Methods of using the crystals in drug screening assays are also provided.

In another aspect, the present invention provides a method for crystallizing a human beta secretase molecule or molecular complex. The method includes preparing purified human beta secretase in the presence of a potential modifier and crystallizing the human beta secretase from a solution having a pH of about 3.5 to about 5.6.

In another aspect, the present invention provides a method for modifying a human beta secretase including contacting the human beta secretase with an HIV protease under conditions sufficient to produce a proteolytically cleaved human beta secretase, wherein the N-terminus of the proteolitically cleaved human beta secretase differs from the N-terminus of the non-cleaved human beta secretase. Methods for crystallizing the modified human beta secretase molecule or molecular complex, and crystals prepared thereby, are also provided.

In another aspect, the present invention provides a molecule or molecular complex that forms a crystal having tetragonal space group symmetry P4.sub.32.sub.12 and includes at least a portion of a human beta secretase or beta secretase-like binding pocket, wherein the binding pocket includes the amino acids listed in Table 3 (see Original Patent) and the binding pocket is defined by a set of points having a root mean square deviation of less than about 0.45 .ANG. from points representing the backbone atoms of said amino acids as represented by the structure coordinates listed in Table 1 (see Original Patent).

In another aspect, the present invention provides a scalable three-dimensional configuration of points, at least a portion of said points being derived from structure coordinates as listed in Table 1 of at least a portion of a human beta secretase molecule or molecular complex that forms a crystal having tetragonal space group symmetry P4.sub.32.sub.12 and that includes a human beta secretase or beta secretase-like binding pocket. Preferably the scalable three-dimensional configuration of points are displayed as a holographic image, a stereodiagram, a model, or a computer-displayed image.

In another aspect, the present invention provides a machine-readable data storage medium including a data storage material encoded with machine readable data which, when using a machine programmed with instructions for using said data, displays a graphical three-dimensional representation of a molecule or molecular complex. Preferably the molecule or molecular complex includes at least a portion of a human beta secretase or beta secretase-like binding pocket including the amino acids listed in Table 3, the binding pocket being defined by a set of points having a root mean square deviation of less than about 0.45 .ANG. from points representing the backbone atoms of said amino acids as represented by structure coordinates listed in Table 1 for a beta secretase or beta secretase-like molecule or molecular complex that forms a crystal having a unit cell defined by the dimensions a, b, c, .alpha., .beta., and .gamma., wherein a is about 94 .ANG. to about 134 .ANG., b is about 94 .ANG. to about 134 .ANG., c is about 170 .ANG. to about 210 .ANG., and .alpha.=.beta.=.gamma.=90.degree..

In another aspect, the present invention provides a method for obtaining structural information about a molecule or a molecular complex of unknown structure. In one embodiment, the method includes crystallizing the molecule or molecular complex; generating an x-ray diffraction pattern from the crystallized molecule or molecular complex; and applying to the x-ray diffraction pattern at least a portion of the structure coordinates as set forth in Table 1 for human beta secretase that forms a crystal having tetragonal space group symmetry P4.sub.32.sub.12, to generate a three-dimensional electron density map of at least a portion of the molecule or molecular complex whose structure is unknown.

In another aspect, the present invention provides a method for homology modeling a human beta secretase homolog. In one embodiment, the method includes aligning the amino acid sequence of a human beta secretase homolog with an amino acid sequence of human beta secretase and incorporating the sequence of the human beta secretase homolog into a model of human beta secretase formed from structure coordinates as set forth in Table 1 for human beta secretase that forms a crystal having a unit cell defined by the dimensions a, b, c, .alpha., .beta., and .gamma., wherein a is about 94 .ANG. to about 134 .ANG., b is about 94 .ANG. to about 134 .ANG., c is about 170 .ANG. to about 210 .ANG., and .alpha.=.beta.=.gamma.=90.degree., to yield a preliminary model of the human beta secretase homolog; subjecting the preliminary model to energy minimization to yield an energy minimized model; and remodeling regions of the energy minimized model where stereochemistry restraints are violated to yield a final model of the human beta secretase homolog.

In another aspect, the present invention provides computer-assisted methods for identifying, designing, or making a potential modifier of human beta secretase activity. Preferably the methods include screening a library of chemical entities.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

Tables 1 and 2 (see Original Patent) list atomic structure coordinates derived by x-ray diffraction of a crystal of human beta secretase expressed in CHO cells with the N-terminus proteolytically cleaved by HIV protease (Table 1, P4.sub.32.sub.12 crystal form) and without the proteolytic cleavage (Table 2, P3.sub.221 crystal form). Column 2 lists a number for the atom in the structure. Column 3 lists the element whose coordinates are measured. The first letter in the column defines the element. Column 4 lists the type of amino acid. Column 5 lists a number for the amino acid in the structure. Columns 6-8 list the crystallographic coordinates X, Y, and Z respectively. The crystallographic coordinates define the atomic position of the element measured. Column 9 lists an occupancy factor that refers to the fraction of the molecules in which each atom occupies the position specified by the coordinates. A value of "1" indicates that each atom has the same conformation, i.e., the same position, in all molecules of the crystal. Column 10 lists a thermal factor "B" that measures movement of the atom around its atomic center. Column 11 lists the chain id (AA for molecule A in the asymmetric unit, BB for molecule B in the asymmetric unit, CC for molecule C in the asymmetric unit, WW for water molecules, and LL for inhibitor molecules). Column 12 lists the element whose coordinates are measured.

Crystalline Form(s) and Method of Making

The three-dimensional structure of human beta secretase was solved using x-ray crystallography to 2.9 .ANG. resolution. Accordingly, the invention includes a human beta secretase crystal and/or a crystal with human beta secretase co-crystallized with a ligand. As used herein, "ligand" refers to a chemical entity that can form a reversible complex with the protein and that could function as a drug candidate (e.g., modifiers and inhibitors). Thus, the term "ligand" as used herein does not include chemical entities that could not function as a drug candidate (e.g., water, metal ions, and solvents). Preferably, the crystal has tetragonal space group symmetry P4.sub.32.sub.12. Preferably, the crystal includes tetragonal shaped unit cells, each unit cell having dimensions a=114.0.+-.20 .ANG., b=114.+-.20 .ANG., c=190.+-.20 .ANG., and .alpha.=.beta.=.gamma.=90.degree.. More preferably, the crystal includes tetragonal shaped unit cells, each unit cell having dimensions a=114.0.+-.5 .ANG., b=114.+-.5 .ANG., c=190.+-.5 .ANG., and .alpha.=.beta.=.gamma.=90.degree.. Most preferably, the crystal includes tetragonal shaped unit cells, each unit cell having dimensions a=114.0 .ANG., b=114 .ANG., c=190 .ANG., and .alpha.=.beta.=.gamma.=90.degree.. Preferably, the crystallized enzyme is a monomer with two monomers in the asymmetric unit.

According to the present invention, human beta secretase can be isolated from a variety of cell lines, for example, the mammalian cell line CHO-K1.

As used herein, a "molecular complex" means a protein in covalent or non-covalent association with a chemical entity (e.g., a ligand). In a preferred embodiment, molecular complexes of purified human beta secretase at a concentration of about 1 mg/ml to about 80 mg/ml may be crystallized in the presence of a modifier at a concentration of about 0.1 to about 10 mM, for example, by using a streak seeding procedure from a solution. Preferably, the solution includes about 5% by weight to about 50% by weight PEG, PEG-MME (PEG monomethyl ether), PEG-DME (PEG dimethyl ether), or polyoxyalkylenepolyamines that preferably have a number average molecular weight of about 200 to about 20,000 (e.g., materials available under the trade designation JEFFAMINE from Huntsman Corp., Salt Lake City, Utah). Optionally, the solution includes a salt. When the solution includes a salt, preferably the solution includes at most about 0.5 M salt. Exemplary salts include sodium chloride, ammonium sulfate, magnesium sulfate, lithium sulfate, or combinations thereof. Optionally, the solution includes about 0% by weight to about 40% by weight organic solvent (e.g., DMSO or MPD), preferably at most about 20% by weight organic solvent. Preferably, the solution is buffered to a pH of about 3.5 to about 5.6 and more preferably about 4.0 to about 4.7. A buffer having a pK.sub.a of about 3 to about 6 is preferred for use in the crystallization method. A particularly preferred buffer is about 10 mM to about 200 mM sodium acetate. Variation in buffer and buffer pH as well as other additives such as PEG, PEG-MME, PEG-DME, or polyoxyalkylenepolyamines is apparent to those skilled in the art and may result in similar crystals.

The invention further includes a human beta secretase crystal that is isomorphous with a human beta secretase crystal having a unit cell defined by the dimensions of a, b, c, .alpha., .beta., and .gamma., wherein a is about 94 .ANG. to about 134 .ANG., b is about 94 .ANG. to about 134 .ANG., c is about 170 .ANG. to about 210 .ANG.; and .alpha.=.beta.=.gamma.=90.degree..

X-Ray Crystallographic Analysis

Each of the constituent amino acids of human beta secretase is defined by a set of structure coordinates as set forth in Table 1. The term "structure coordinates" refers to Cartesian coordinates derived from mathematical equations related to the patterns obtained on diffraction of a monochromatic beam of x-rays by the atoms (scattering centers) of a human beta secretase complex in crystal form. The diffraction data are used to calculate an electron density map of the repeating unit of the crystal. The electron density maps are then used to establish the positions of the individual atoms of the human beta secretase protein or protein/ligand complex.

Slight variations in structure coordinates can be generated by mathematically manipulating the human beta secretase or human beta secretase/ligand structure coordinates. For example, the structure coordinates set forth in Table 1 could be manipulated by crystallographic permutations of the structure coordinates, fractionalization of the structure coordinates, integer additions or subtractions to sets of the structure coordinates, inversion of the structure coordinates or any combination of the above. Alternatively, modifications in the crystal structure due to mutations, additions, substitutions, and/or deletions of amino acids, or other changes in any of the components that make up the crystal, could also yield variations in structure coordinates. Such slight variations in the individual coordinates will have little effect on overall shape. If such variations are within an acceptable standard error as compared to the original coordinates, the resulting three-dimensional shape is considered to be structurally equivalent. Structural equivalence is described in more detail below.

It should be noted that slight variations in individual structure coordinates of the human beta secretase would not be expected to significantly alter the nature of chemical entities such as ligands that could associate with the binding pockets. In this context, the phrase "associating with" refers to a condition of proximity between a chemical entity, or portions thereof, and a human beta secretase molecule or portions thereof. The association may be non-covalent, wherein the juxtaposition is energetically favored by hydrogen bonding, van der Waals forces, or electrostatic interactions, or it may be covalent. Thus, for example, a ligand that bound to a binding pocket of human beta secretase would also be expected to bind to or interfere with a structurally equivalent binding pocket.

For the purpose of this invention, any molecule or molecular complex or binding pocket thereof, or any portion thereof, that has a root mean square deviation of conserved residue backbone atoms (N, C.alpha., C, O) of less than about 0.45 .ANG., when superimposed on the relevant backbone atoms described by the reference structure coordinates listed in Table 1, is considered "structurally equivalent" to the reference molecule. That is to say, the crystal structures of those portions of the two molecules are substantially identical, within acceptable error. As used herein, "residue" refers to one or more atoms. Particularly preferred structurally equivalent molecules or molecular complexes are those that are defined by the entire set of structure coordinates listed in Table 1 .+-.a root mean square deviation from the conserved backbone atoms of those amino acids of less than about 0.45 .ANG.. More preferably, the root mean square deviation is at most about 0.35 .ANG., and most preferably at most about 0.2 .ANG.. Other embodiments of this invention include a molecular complex defined by the structure coordinates listed in Table 1 for those amino acids listed in Table 3, Table 4, or Table 5 (see Original Patent), .+-.a root mean square deviation from the conserved backbone atoms of those amino acids of less than about 0.45 .ANG., preferably at most about 0.35 .ANG., and more preferably at most about 0.2 .ANG..

The term "root mean square deviation" means the square root of the arithmetic mean of the squares of the deviations. It is a way to express the deviation or variation from a trend or object. For purposes of this invention, the "root mean square deviation" defines the variation in the backbone of a protein from the backbone of human beta secretase or a binding pocket portion thereof, as defined by the structure coordinates of human beta secretase described herein.

It will be readily apparent to those of skill in the art that the numbering of amino acids in other isoforms of human beta secretase may be different than that of human beta secretase expressed in CHO.

Active Site and Other Structural Features

Applicants' invention provides information about the shape and structure of the binding pocket of human beta secretase in the presence of a potential modifier. The secondary structure of the human beta secretase monomer includes two domains consistent with a typical aspartic protease fold.

Binding pockets are of significant utility in fields such as drug discovery. The association of natural ligands or substrates with the binding pockets of their corresponding receptors or enzymes is the basis of many biological mechanisms of action. Similarly, many drugs exert their biological effects through association with the binding pockets of receptors and enzymes. Such associations may occur with all or any parts of the binding pocket. An understanding of such associations helps lead to the design of drugs having more favorable associations with their target, and thus improved biological effects. Therefore, this information is valuable in designing potential modifiers of beta secretase-like binding pockets, as discussed in more detail below.

The term "binding pocket," as used herein, refers to a region of a molecule or molecular complex, that, as a result of its shape, favorably associates with another chemical entity. Thus, a binding pocket may include or consist of features such as cavities, surfaces, or interfaces between domains. Chemical entities that may associate with a binding pocket include, but are not limited to, cofactors, substrates, modifiers, agonists, and antagonists.

The amino acid constituents of a human beta secretase binding pocket as defined herein are positioned in three dimensions in accordance with the structure coordinates listed in Table 1 (see Original Patent). In one aspect, the structure coordinates defining a binding pocket of human beta secretase include structure coordinates of all atoms in the constituent amino acids; in another aspect, the structure coordinates of a binding pocket include structure coordinates of just the backbone atoms of the constituent amino acids.

The binding pocket of human beta secretase preferably includes the amino acids listed in Table 3 (see Original Patent), more preferably the amino acids listed in Table 4, and most preferably the amino acids listed in Table 5, as represented by the structure coordinates listed in Table 1. Alternatively, the binding pocket of human beta secretase may be defined by those amino acids whose backbone atoms are situated within about 4 .ANG., more preferably within about 7 .ANG., most preferably within about 10 .ANG., of one or more constituent atoms of a bound substrate or modifier. In yet another alternative, the binding pocket may be defined by those amino acids whose backbone atoms are situated within a sphere centered on the coordinates representing the alpha carbon atom of residue Thr 231, the sphere having a radius of about 15 .ANG., preferably about 20 .ANG., and more preferably about 25 .ANG..

The term "beta secretase-like binding pocket" refers to a portion of a molecule or molecular complex whose shape is sufficiently similar to at least a portion of a binding pocket of human beta secretase as to be expected to bind related structural analogues. As used herein, "at least a portion" means that at least about 50% of the amino acids are included, preferably at least about 70% of the amino acids are included, more preferably at least about 90% of the amino acids are included, and most preferably all the amino acids are included. A structurally equivalent binding pocket is defined by a root mean square deviation from the structure coordinates of the backbone atoms of the amino acids that make up binding pockets in human beta secretase (as set forth in Table 1) of less than about 0.45 .ANG., preferably at most about 0.35 .ANG., and more preferably at most about 0.2 .ANG.. How this calculation is obtained is described below.

Accordingly, the invention provides molecules or molecular complexes including a human beta secretase binding pocket or beta secretase-like binding pocket, as defined by the sets of structure coordinates described above.

Three-Dimensional Configurations

X-ray structure coordinates define a unique configuration of points in space. Those of skill in the art understand that a set of structure coordinates for protein or an protein/ligand complex, or a portion thereof, define a relative set of points that, in turn, define a configuration in three dimensions. A similar or identical configuration can be defined by an entirely different set of coordinates, provided the distances and angles between coordinates remain essentially the same. In addition, a scalable configuration of points can be defined by increasing or decreasing the distances between coordinates by a scalar factor while keeping the angles essentially the same.

The present invention thus includes the scalable three-dimensional configuration of points derived from the structure coordinates of at least a portion of a human beta secretase molecule or molecular complex, as listed in Table 1, as well as structurally equivalent configurations, as described below. Preferably, the scalable three-dimensional configuration includes points derived from structure coordinates representing the locations of a plurality of the amino acids defining a human beta secretase binding pocket.

In one embodiment, the scalable three-dimensional configuration includes points derived from structure coordinates representing the locations the backbone atoms of a plurality of amino acids defining the human beta secretase binding pocket, preferably the amino acids listed in Table 3, more preferably the amino acids listed in Table 4, and most preferably the amino acids listed in Table 5. Alternatively, the scalable three-dimensional configuration includes points derived from structure coordinates representing the locations of the side chain and the backbone atoms (other than hydrogens) of a plurality of the amino acids defining the human beta secretase binding pocket, preferably the amino acids listed in Table 3, more preferably the amino acids listed in Table 4, and most preferably the amino acids listed in Table 5.

Likewise, the invention also includes the scalable three-dimensional configuration of points derived from structure coordinates of molecules or molecular complexes that are structurally homologous to beta secretase, as well as structurally equivalent configurations. Structurally homologous molecules or molecular complexes are defined below. Advantageously, structurally homologous molecules can be identified using the structure coordinates of human beta secretase according to a method of the invention.

The configurations of points in space derived from structure coordinates according to the invention can be visualized as, for example, a holographic image, a stereodiagram, a model, or a computer-displayed image, and the invention thus includes such images, diagrams or models.

Structurally Equivalent Crystal Structures

Various computational analyses can be used to determine whether a molecule or a binding pocket portion thereof is "structurally equivalent," defined in terms of its three-dimensional structure, to all or part of human beta secretase or its binding pockets. Such analyses may be carried out in current software applications, such as the Molecular Similarity application of QUANTA (Molecular Simulations Inc., San Diego, Calif.) version 4.1, and as described in the accompanying User's Guide.

The Molecular Similarity application permits comparisons between different structures, different conformations of the same structure, and different parts of the same structure. The procedure used in Molecular Similarity to compare structures is divided into four steps: (1) load the structures to be compared; (2) define the atom equivalences in these structures; (3) perform a fitting operation; and (4) analyze the results.

Each structure is identified by a name. One structure is identified as the target (i.e., the fixed structure); all remaining structures are working structures (i.e., moving structures). Since atom equivalency within QUANTA is defined by user input, for the purpose of this invention equivalent atoms are defined as protein backbone atoms (N, C.alpha., C, and O) for all conserved residues between the two structures being compared. A conserved residue is defined as a residue which is structurally or functionally equivalent. Only rigid fitting operations are considered.

When a rigid fitting method is used, the working structure is translated and rotated to obtain an optimum fit with the target structure. The fitting operation uses an algorithm that computes the optimum translation and rotation to be applied to the moving structure, such that the root mean square difference of the fit over the specified pairs of equivalent atom is an absolute minimum. This number, given in angstroms, is reported by QUANTA.

Machine Readable Storage Media

Transformation of the structure coordinates for all or a portion of human beta secretase or the human beta secretase/ligand complex or one of its binding pockets, for structurally homologous molecules as defined below, or for the structural equivalents of any of these molecules or molecular complexes as defined above, into three-dimensional graphical representations of the molecule or complex can be conveniently achieved through the use of commercially-available software.

The invention thus further provides a machine-readable storage medium including a data storage material encoded with machine readable data which, when using a machine programmed with instructions for using said data, displays a graphical three-dimensional representation of any of the molecule or molecular complexes of this invention that have been described above. In a preferred embodiment, the machine-readable data storage medium includes a data storage material encoded with machine readable data which, when using a machine programmed with instructions for using said data, displays a graphical three-dimensional representation of a molecule or molecular complex including all or any parts of a human beta secretase binding pocket or an beta secretase-like binding pocket, as defined above. In another preferred embodiment, the machine-readable data storage medium includes a data storage material encoded with machine readable data which, when using a machine programmed with instructions for using said data, displays a graphical three-dimensional representation of a molecule or molecular complex defined by the structure coordinates of all of the amino acids listed in Table 1, .+-. a root mean square deviation from the backbone atoms of said amino acids of less than about 0.45 .ANG..

In an alternative embodiment, the machine-readable data storage medium includes a data storage material encoded with a first set of machine readable data which includes the Fourier transform of the structure coordinates set forth in Table 1, and which, when using a machine programmed with instructions for using said data, can be combined with a second set of machine readable data including the x-ray diffraction pattern of a molecule or molecular complex to determine at least a portion of the structure coordinates corresponding to the second set of machine readable data.

For example, a system for reading a data storage medium may include a computer including a central processing unit ("CPU"), a working memory which may be, e.g., RAM (random access memory) or "core" memory, mass storage memory (such as one or more disk drives or CD-ROM drives), one or more display devices (e.g., cathode-ray tube ("CRT") displays, light emitting diode ("LED") displays, liquid crystal displays ("LCDs"), electroluminescent displays, vacuum fluorescent displays, field emission displays ("FEDs"), plasma displays, projection panels, etc.), one or more user input devices (e.g., keyboards, microphones, mice, track balls, touch pads, etc.), one or more input lines, and one or more output lines, all of which are interconnected by a conventional bidirectional system bus. The system may be a stand-alone computer, or may be networked (e.g., through local area networks, wide area networks, intranets, extranets, or the internet) to other systems (e.g., computers, hosts, servers, etc.). The system may also include additional computer controlled devices such as consumer electronics and appliances.

Input hardware may be coupled to the computer by input lines and may be implemented in a variety of ways. Machine-readable data of this invention may be inputted via the use of a modem or modems connected by a telephone line or dedicated data line. Alternatively or additionally, the input hardware may include CD-ROM drives or disk drives. In conjunction with a display terminal, a keyboard may also be used as an input device.

Output hardware may be coupled to the computer by output lines and may similarly be implemented by conventional devices. By way of example, the output hardware may include a display device for displaying a graphical representation of a binding pocket of this invention using a program such as QUANTA as described herein. Output hardware might also include a printer, so that hard copy output may be produced, or a disk drive, to store system output for later use.

In operation, a CPU coordinates the use of the various input and output devices, coordinates data accesses from mass storage devices, accesses to and from working memory, and determines the sequence of data processing steps. A number of programs may be used to process the machine-readable data of this invention. Such programs are discussed in reference to the computational methods of drug discovery as described herein. References to components of the hardware system are included as appropriate throughout the following description of the data storage medium.

Machine-readable storage devices useful in the present invention include, but are not limited to, magnetic devices, electrical devices, optical devices, and combinations thereof. Examples of such data storage devices include, but are not limited to, hard disk devices, CD devices, digital video disk devices, floppy disk devices, removable hard disk devices, magneto-optic disk devices, magnetic tape devices, flash memory devices, bubble memory devices, holographic storage devices, and any other mass storage peripheral device. It should be understood that these storage devices include necessary hardware (e.g., drives, controllers, power supplies, etc.) as well as any necessary media (e.g., disks, flash cards, etc.) to enable the storage of data.

Structurally Homologous Molecules, Molecular Complexes, and Crystal Structures

The structure coordinates set forth in Table 1 can be used to aid in obtaining structural information about another crystallized molecule or molecular complex. The method of the invention allows determination of at least a portion of the three-dimensional structure of molecules or molecular complexes which contain one or more structural features that are similar to structural features of human beta secretase. These molecules are referred to herein as "structurally homologous" to human beta secretase. Similar structural features can include, for example, regions of amino acid identity, conserved active site or binding site motifs, and similarly arranged secondary structural elements (e.g., .alpha. helices and .beta. sheets). Optionally, structural homology is determined by aligning the residues of the two amino acid sequences to optimize the number of identical amino acids along the lengths of their sequences; gaps in either or both sequences are permitted in making the alignment in order to optimize the number of identical amino acids, although the amino acids in each sequence must nonetheless remain in their proper order. Preferably, two amino acid sequences are compared using the Blastp program, version 2.0.9, of the BLAST 2 search algorithm, as described by Tatusova et al., FEMS Microbiol Lett., 174:247-50 (1999). Preferably, the default values for all BLAST 2 search parameters are used, including matrix=BLOSUM62; open gap penalty=11, extension gap penalty=1, gap x_dropoff=50, expect=10, wordsize=3, and filter on. In the comparison of two amino acid sequences using the BLAST search algorithm, structural similarity is referred to as "identity." Preferably, a structurally homologous molecule is a protein that has an amino acid sequence sharing at least 65% identity with a native or recombinant amino acid sequence of human beta secretase (for example, SEQ ID NO:1). More preferably, a protein that is structurally homologous to human beta secretase includes a contiguous stretch of at least 50 amino acids that shares at least 80% amino acid sequence identity with the analogous portion of the native or recombinant human beta secretase (for example, SEQ ID NO:1). Methods for generating structural information about the structurally homologous molecule or molecular complex are well-known and include, for example, molecular replacement techniques.

Therefore, in another embodiment this invention provides a method of utilizing molecular replacement to obtain structural information about a molecule or molecular complex whose structure is unknown including the steps of:

(a) crystallizing the molecule or molecular complex of unknown structure;

(b) generating an x-ray diffraction pattern from said crystallized molecule or molecular complex; and

(c) applying at least a portion of the structure coordinates set forth in Table 1 to the x-ray diffraction pattern to generate a three-dimensional electron density map of the molecule or molecular complex whose structure is unknown.

By using molecular replacement, all or part of the structure coordinates of human beta secretase or the human beta secretase/ligand complex as provided by this invention can be used to determine the structure of a crystallized molecule or molecular complex whose structure is unknown more quickly and efficiently than attempting to determine such information ab initio.

Molecular replacement provides an accurate estimation of the phases for an unknown structure. Phases are a factor in equations used to solve crystal structures that cannot be determined directly. Obtaining accurate values for the phases, by methods other than molecular replacement, is a time-consuming process that involves iterative cycles of approximations and refinements and greatly hinders the solution of crystal structures. However, when the crystal structure of a protein containing at least a structurally homologous portion has been solved, the phases from the known structure provide a satisfactory estimate of the phases for the unknown structure.

Thus, this method involves generating a preliminary model of a molecule or molecular complex whose structure coordinates are unknown, by orienting and positioning the relevant portion of human beta secretase or the human beta secretase/modifier complex within the unit cell of the crystal of the unknown molecule or molecular complex so as best to account for the observed x-ray diffraction pattern of the crystal of the molecule or molecular complex whose structure is unknown. Phases can then be calculated from this model and combined with the observed x-ray diffraction pattern amplitudes to generate an electron density map of the structure whose coordinates are unknown. This, in turn, can be subjected to any well-known model building and structure refinement techniques to provide a final, accurate structure of the unknown crystallized molecule or molecular complex (Lattman, Meth. Enzymol., 115, 55-77 (1985); M. G. Rossman, ed., "The Molecular Replacement Method," Int. Sci. Rev. Ser., No. 13, Gordon & Breach, New York (1972)).

Structural information about a portion of any crystallized molecule or molecular complex that is sufficiently structurally homologous to a portion of human beta secretase can be resolved by this method. In addition to a molecule that shares one or more structural features with human beta secretase as described above, a molecule that has similar bioactivity, such as the same catalytic activity, substrate specificity or ligand binding activity as human beta secretase, may also be sufficiently structurally homologous to human beta secretase to permit use of the structure coordinates of human beta secretase to solve its crystal structure.

In a preferred embodiment, the method of molecular replacement is utilized to obtain structural information about a molecule or molecular complex, wherein the molecule or molecular complex includes a human beta secretase subunit or homolog. A "subunit" of human beta secretase is a human beta secretase molecule that has been truncated at the N-terminus or the C-terminus, or both. In the context of the present invention, a "homolog" of human beta secretase is a protein that contains one or more amino acid substitutions, deletions, additions, or rearrangements with respect to the amino acid sequence of human beta secretase (SEQ ID NO:1), but that, when folded into its native conformation, exhibits or is reasonably expected to exhibit at least a portion of the tertiary (three-dimensional) structure of human beta secretase. For example, structurally homologous molecules can contain deletions or additions of one or more contiguous or noncontiguous amino acids, such as a loop or a domain. Structurally homologous molecules also include "modified" human beta secretase molecules that have been chemically or enzymatically derivatized at one or more constituent amino acid, including side chain modifications, backbone modifications, and N- and C-terminal modifications including acetylation, hydroxylation, methylation, amidation, and the attachment of carbohydrate or lipid moieties, cofactors, and the like.

A heavy atom derivative of human beta secretase is also included as a human beta secretase homolog. The term "heavy atom derivative" refers to derivatives of human beta secretase produced by chemically modifying a crystal of human beta secretase. In practice, a crystal is soaked in a solution containing heavy metal atom salts, or organometallic compounds, e.g., lead chloride, gold thiomalate, thiomersal or uranyl acetate, which can diffuse through the crystal and bind to the surface of the protein. The location(s) of the bound heavy metal atom(s) can be determined by x-ray diffraction analysis of the soaked crystal. This information, in turn, is used to generate the phase information used to construct three-dimensional structure of the protein (T. L. Blundell and N. L. Johnson, Protein Crystallography, Academic Press (1976)).

Because human beta secretase can crystallize in more than one crystal form, the structure coordinates of human beta secretase as provided by this invention are particularly useful in solving the structure of other crystal forms of human beta secretase or human beta secretase complexes.

The structure coordinates of human beta secretase as provided by this invention are particularly useful in solving the structure of human beta secretase mutants. Mutants may be prepared, for example, by expression of human beta secretase cDNA previously altered in its coding sequence by oligonucleotide-directed mutagenesis. Mutants may also be generated by site-specific incorporation of unnatural amino acids into beta secretase proteins using the general biosynthetic method of Noren et al., Science, 244:182-88 (1989). In this method, the codon encoding the amino acid of interest in wild-type human beta secretase is replaced by a "blank" nonsense codon, TAG, using oligonucleotide-directed mutagenesis. A suppressor tRNA directed against this codon is then chemically aminoacylated in vitro with the desired unnatural amino acid. The aminoacylated tRNA is then added to an in vitro translation system to yield a mutant human beta secretase with the site-specific incorporated unnatural amino acid.

Selenocysteine or selenomethionine may be incorporated into wild-type or mutant human beta secretase by expression of human beta secretase-encoding cDNAs in auxotrophic E. coli strains (Hendrickson et al., EMBO J., 9:1665-72 (1990)). In this method, the wild-type or mutagenized human beta secretase cDNA may be expressed in a host organism on a growth medium depleted of either natural cysteine or methionine (or both) but enriched in selenocysteine or selenomethionine (or both). Alternatively, selenomethionine analogues may be prepared by down regulation methionine biosynthesis. (Benson et al., Nat. Struct. Biol., 2:644-53 (1995); Van Duyne et al., J. Mol. Biol., 229:105-24 (1993)).

The structure coordinates of human beta secretase listed in Table 1 are also particularly useful to solve the structure of crystals of human beta secretase, human beta secretase mutants or human beta secretase homologs co-complexed with a variety of chemical entities. This approach enables the determination of the optimal sites for interaction between chemical entities, including candidate human beta secretase modifiers and human beta secretase. Potential sites for modification within the various binding sites of the molecule can also be identified. This information provides an additional tool for determining the most efficient binding interactions, for example, increased hydrophobic interactions, between human beta secretase and a chemical entity. For example, high resolution x-ray diffraction data collected from crystals exposed to different types of solvent allows the determination of where each type of solvent molecule resides. Small molecules that bind tightly to those sites can then be designed and synthesized and tested for their potential human beta secretase inhibition activity.

All of the complexes referred to above may be studied using well-known x-ray diffraction techniques and may be refined versus 1.5-3.5 .ANG. resolution x-ray data to an R value of about 0.30 or less using computer software, such as X-PLOR (Yale University, 81992, distributed by Molecular Simulations, Inc.; see, e.g., Blundell & Johnson, supra; Meth. Enzymol., Vol. 114 & 115, H. W. Wyckoff et al., eds., Academic Press (1985)). This information may thus be used to optimize known human beta secretase modifiers, and more importantly, to design new human beta secretase modifiers.

The invention also includes the unique three-dimensional configuration defined by a set of points defined by the structure coordinates for a molecule or molecular complex structurally homologous to human beta secretase as determined using the method of the present invention, structurally equivalent configurations, and magnetic storage media including such set of structure coordinates.

Further, the invention includes structurally homologous molecules as identified using the method of the invention.

Homology Modeling

Using homology modeling, a computer model of a human beta secretase homolog can be built or refined without crystallizing the homolog. First, a preliminary model of the human beta secretase homolog is created by sequence alignment with human beta secretase, secondary structure prediction, the screening of structural libraries, or any combination of those techniques. Computational software may be used to carry out the sequence alignments and the secondary structure predictions. Structural incoherences, e.g., structural fragments around insertions and deletions, can be modeled by screening a structural library for peptides of the desired length and with a suitable conformation. For prediction of the side chain conformation, a side chain rotamer library may be employed. If the human beta secretase homolog has been crystallized, the final homology model can be used to solve the crystal structure of the homolog by molecular replacement, as described above. Next, the preliminary model is subjected to energy minimization to yield an energy minimized model. The energy minimized model may contain regions where stereochemistry restraints are violated, in which case such regions are remodeled to obtain a final homology model. The homology model is positioned according to the results of molecular replacement, and subjected to further refinement including molecular dynamics calculations.

Rational Drug Design

Computational techniques can be used to screen, identify, select and/or design chemical entities capable of associating with human beta secretase or structurally homologous molecules. Knowledge of the structure coordinates for human beta secretase permits the design and/or identification of synthetic compounds and/or other molecules which have a shape complementary to the conformation of the human beta secretase binding site. In particular, computational techniques can be used to identify or design chemical entities, such as modifiers, agonists and antagonists, that associate with a human beta secretase binding pocket or an beta secretase-like binding pocket. Potential modifiers may bind to or interfere with all or a portion of an active site of human beta secretase, and can be competitive, non-competitive, or uncompetitive inhibitors; or interfere with dimerization by binding at the interface between the two monomers. Once identified and screened for biological activity, these inhibitors/agonists/antagonists may be used therapeutically or prophylactically to block human beta secretase activity and, thus, prevent the onset and/or further progression of Alzheimer's disease. Structure-activity data for analogues of ligands that bind to or interfere with human beta secretase or beta secretase-like binding pockets can also be obtained computationally.

The term "chemical entity," as used herein, refers to chemical compounds, complexes of two or more chemical compounds, and fragments of such compounds or complexes. Chemical entities that are determined to associate with human beta secretase are potential drug candidates. Data stored in a machine-readable storage medium that displays a graphical three-dimensional representation of the structure of human beta secretase or a structurally homologous molecule, as identified herein, or portions thereof may thus be advantageously used for drug discovery. The structure coordinates of the chemical entity are used to generate a three-dimensional image that can be computationally fit to the three-dimensional image of human beta secretase or a structurally homologous molecule. The three-dimensional molecular structure encoded by the data in the data storage medium can then be computationally evaluated for its ability to associate with chemical entities. When the molecular structures encoded by the data is displayed in a graphical three-dimensional representation on a computer screen, the protein structure can also be visually inspected for potential association with chemical entities.

One embodiment of the method of drug design involves evaluating the potential association of a known chemical entity with human beta secretase or a structurally homologous molecule, particularly with a human beta secretase binding pocket or beta secretase-like binding pocket. The method of drug design thus includes computationally evaluating the potential of a selected chemical entity to associate with any of the molecules or molecular complexes set forth above. This method includes the steps of: (a) employing computational means to perform a fitting operation between the selected chemical entity and a binding pocket or a pocket nearby the binding pocket of the molecule or molecular complex; and (b) analyzing the results of said fitting operation to quantify the association between the chemical entity and the binding pocket.

In another embodiment, the method of drug design involves computer-assisted design of chemical entities that associate with human beta secretase, its homologs, or portions thereof. Chemical entities can be designed in a step-wise fashion, one fragment at a time, or may be designed as a whole or "de novo."

To be a viable drug candidate, the chemical entity identified or designed according to the method must be capable of structurally associating with at least part of a human beta secretase or beta secretase-like binding pockets, and must be able, sterically and energetically, to assume a conformation that allows it to associate with the human beta secretase or beta secretase-like binding pocket. Non-covalent molecular interactions important in this association include hydrogen bonding, van der Waals interactions, hydrophobic interactions, and electrostatic interactions. Conformational considerations include the overall three-dimensional structure and orientation of the chemical entity in relation to the binding pocket, and the spacing between various functional groups of an entity that directly interact with the beta secretase-like binding pocket or homologs thereof.

Optionally, the potential binding of a chemical entity to a human beta secretase or beta secretase-like binding pocket is analyzed using computer modeling techniques prior to the actual synthesis and testing of the chemical entity. If these computational experiments suggest insufficient interaction and association between it and the human beta secretase or beta secretase-like binding pocket, testing of the entity is obviated. However, if computer modeling indicates a strong interaction, the molecule may then be synthesized and tested for its ability to bind to or interfere with a human beta secretase or beta secretase-like binding pocket. Binding assays to determine if a compound (e.g., an inhibitor) actually interferes with human beta secretase can also be performed and are well known in the art. Binding assays may employ kinetic or thermodynamic methodology using a wide variety of techniques including, but not limited to, microcalorimetry, circular dichroism, capillary zone electrophoresis, nuclear magnetic resonance spectroscopy, fluorescence spectroscopy, and combinations thereof.

One method for determining whether a modifier binds to a protein is isothermal denaturation. This method includes taking a sample of a protein (in the presence or absence of substrates) at a fixed elevated temperature where denaturation of the protein occurs in a given time frame, adding the chemical entity to the protein, and monitoring the rate of denaturation. If the chemical entity does bind to the protein, it is expected that the rate of denaturation would be slower in the presence of the chemical entity than in the absence of the chemical entity. For example, this method has been described in Epps et al., Anal. Biochem., 292:40-50 (2001).

One skilled in the art may use one of several methods to screen chemical entities or fragments for their ability to associate with a human beta secretase or beta secretase-like binding pocket. This process may begin by visual inspection of, for example, a human beta secretase or beta secretase-like binding pocket on the computer screen based on the human beta secretase structure coordinates listed in Table 1 or other coordinates which define a similar shape generated from the machine-readable storage medium. Selected fragments or chemical entities may then be positioned in a variety of orientations, or docked, within the binding pocket. Docking may be accomplished using software such as QUANTA and SYBYL, followed by energy minimization and molecular dynamics with standard molecular mechanics forcefields, such as CHARMM and AMBER.

Specialized computer programs may also assist in the process of selecting fragments or chemical entities. Examples include GRID (Goodford, J. Med. Chem., 28:849-57 (1985); available from Oxford University, Oxford, UK); MCSS (Miranker et al., Proteins: Struct. Funct. Gen., 11:29-34 (1991); available from Molecular Simulations, San Diego, Calif.); AUTODOCK (Goodsell et al., Proteins: Struct. Funct. Genet., 8:195-202 (1990); available from Scripps Research Institute, La Jolla, Calif.); and DOCK (Kuntz et al., J. Mol. Biol., 161:269-88 (1982); available from University of California, San Francisco, Calif.).

Once suitable chemical entities or fragments have been selected, they can be assembled into a single compound or complex. Assembly may be preceded by visual inspection of the relationship of the fragments to each other on the three-dimensional image displayed on a computer screen in relation to the structure coordinates of human beta secretase. This would be followed by manual model building using software such as QUANTA or SYBYL (Tripos Associates, St. Louis, Mo.).

Useful programs to aid one of skill in the art in connecting the individual chemical entities or fragments include, without limitation, CAVEAT (P. A. Bartlett et al., in "Molecular Recognition in Chemical and Biological Problems," Special Publ., Royal Chem. Soc., 78:182-96 (1989); Lauri et al., J. Comput. Aided Mol. Des., 8:51-66 (1994); available from the University of California, Berkeley, Calif.); 3D database systems such as ISIS (available from MDL Information Systems, San Leandro, Calif.; reviewed in Martin, J. Med. Chem., 35:2145-54 (1992)); and HOOK (Eisen et al., Proteins: Struc., Funct., Genet., 19:199-221 (1994); available from Molecular Simulations, San Diego, Calif.).

Human beta secretase binding compounds may be designed "de novo" using either an empty binding site or optionally including some portion(s) of a known modifier(s). There are many de novo ligand design methods including, without limitation, LUDI (Bohm, J. Comp. Aid. Molec. Design., 6:61-78 (1992); available from Molecular Simulations Inc., San Diego, Calif.); LEGEND (Nishibata et al., Tetrahedron, 47:8985 (1991); available from Molecular Simulations Inc., San Diego, Calif.); LeapFrog (available from Tripos Associates, St. Louis, Mo.); and SPROUT (Gillet et al., J. Comput. Aided Mol. Design, 7:127-53 (1993); available from the University of Leeds, UK).

Once a compound has been designed or selected by the above methods, the efficiency with which that entity may bind to or interfere with a human beta secretase or beta secretase-like binding pocket may be tested and optimized by computational evaluation. For example, an effective human beta secretase or beta secretase-like binding pocket modifier must preferably demonstrate a relatively small difference in energy between its bound and free states (i.e., a small deformation energy of binding). Thus, the most efficient human beta secretase or beta secretase-like binding pocket modifiers should preferably be designed with a deformation energy of binding of at most about 10 kcal/mole; more preferably, at most 7 kcal/mole. Human beta secretase or beta secretase-like binding pocket modifiers may interact with the binding pocket in more than one conformation that is similar in overall binding energy. In those cases, the deformation energy of binding is taken to be the difference between the energy of the free entity and the average energy of the conformations observed when the modifier binds to the protein.

An entity designed or selected as binding to or interfering with a human beta secretase or beta secretase-like binding pocket may be further computationally optimized so that in its bound state it would preferably lack repulsive electrostatic interaction with the target enzyme and with the surrounding water molecules. Such non-complementary electrostatic interactions include repulsive charge-charge, dipole-dipole, and charge-dipole interactions.

Specific computer software is available in the art to evaluate compound deformation energy and electrostatic interactions. Examples of programs designed for such uses include: Gaussian 94, revision C (M. J. Frisch, Gaussian, Inc., Pittsburgh, Pa. 81995); AMBER, version 4.1 (P. A. Kollman, University of California at San Francisco, 81995); QUANTA/CHARMM (Molecular Simulations, Inc., San Diego, Calif. 81995); Insight II/Discover (Molecular Simulations, Inc., San Diego, Calif. 81995); DelPhi (Molecular Simulations, Inc., San Diego, Calif. 81995); and AMSOL (Quantum Chemistry Program Exchange, Indiana University). These programs may be implemented, for instance, using a Silicon Graphics workstation such as an Indigo.sup.2 with "IMPACT" graphics. Other hardware systems and software packages will be known to those skilled in the art.

Another approach encompassed by this invention is the computational screening of small molecule databases for chemical entities or compounds that can bind in whole, or in part, to a human beta secretase or beta secretase-like binding pocket. In this screening, the quality of fit of such entities to the binding site may be judged either by shape complementarity or by estimated interaction energy (Meng et al., J. Comp. Chem., 13:505-24 (1992)).

This invention also enables the development of chemical entities that can isomerize to short-lived reaction intermediates in the chemical reaction of a substrate or other compound that interferes with or with human beta secretase. Time-dependent analysis of structural changes in human beta secretase during its interaction with other molecules is carried out. The reaction intermediates of human beta secretase can also be deduced from the reaction product in co-complex with human beta secretase. Such information is useful to design improved analogues of known human beta secretase modifiers or to design novel classes of potential modifiers based on the reaction intermediates of the human beta secretase and modifier co-complex. This provides a novel route for designing human beta secretase modifiers with both high specificity and stability.

Yet another approach to rational drug design involves probing the human beta secretase crystal of the invention with molecules including a variety of different functional groups to determine optimal sites for interaction between candidate human beta secretase modifiers and the protein. For example, high resolution x-ray diffraction data collected from crystals soaked in or co-crystallized with other molecules allows the determination of where each type of solvent molecule sticks. Molecules that bind tightly to those sites can then be further modified and synthesized and tested for their beta secretase modifier activity (Travis, Science, 262:1374 (1993)).

In a related approach, iterative drug design is used to identify modifiers of human beta secretase. Iterative drug design is a method for optimizing associations between a protein and a compound by determining and evaluating the three-dimensional structures of successive sets of protein/compound complexes. In iterative drug design, crystals of a series of protein/compound complexes are obtained and then the three-dimensional structures of each complex is solved. Such an approach provides insight into the association between the proteins and compounds of each complex. This is accomplished by selecting compounds with inhibitory activity, obtaining crystals of this new protein/compound complex, solving the three-dimensional structure of the complex, and comparing the associations between the new protein/compound complex and previously solved protein/compound complexes. By observing how changes in the compound affected the protein/compound associations, these associations may be optimized.

A compound that is identified or designed as a result of any of these methods can be obtained (or synthesized) and tested for its biological activity, e.g., inhibition of beta secretase activity.

Pharmaceutical Compositions (Modifiers)

Pharmaceutical compositions of this invention include a potential modifier of human beta secretase activity identified according to the invention, or a pharmaceutically acceptable salt thereof, and a pharmaceutically acceptable carrier, adjuvant, or vehicle. The term "pharmaceutically acceptable carrier" refers to a carrier(s) that is "acceptable" in the sense of being compatible with the other ingredients of a composition and not deleterious to the recipient thereof. Optionally, the pH of the formulation is adjusted with pharmaceutically acceptable acids, bases, or buffers to enhance the stability of the formulated compound or its delivery form.

Methods of making and using such pharmaceutical compositions are also included in the invention. The pharmaceutical compositions of the invention can be administered orally, parenterally, by inhalation spray, topically, rectally, nasally, buccally, vaginally, or via an implanted reservoir. Oral administration or administration by injection is preferred. The term parenteral as used herein includes subcutaneous, intracutaneous, intravenous, intramuscular, intra-articular, intrasynovial, intrasternal, intrathecal, intralesional, and intracranial injection or infusion techniques.

Dosage levels of about 0.01 to about 100 mg/kg body weight per day, preferably of about 0.5 to about 75 mg/kg body weight per day of the human beta secretase inhibitory compounds described herein are useful for the prevention and treatment of human beta secretase mediated disease. Typically, the pharmaceutical compositions of this invention will be administered about 1 to about 5 times per day or alternatively, as a continuous infusion. Such administration can be used as a chronic or acute therapy. The amount of active ingredient that may be combined with the carrier materials to produce a single dosage form will vary depending upon the host treated and the particular mode of administration. A typical preparation will contain about 5% to about 95% active compound (w/w). Preferably, such preparations contain about 20% to about 80% active compound.
 

Claim 1 of 1 Claim

1. A crystal of beta secretase co-crystallized with a ligand, wherein the amino acid sequence of said beta secretase consists of SEQ ID NO:2, said crystal has tetragonal space group symmetry P4.sub.32.sub.12, a unit cell defined by the dimensions a, b, c, .alpha., .beta., and .gamma. wherein a is 114 .ANG., b is 114 .ANG., c is 190 .ANG., and .alpha.=.beta.=.gamma.=90 degrees, and wherein said ligand is N1-((2S,3R)-4-(3-iodobenzylamino)-1-(3,5-difluorophenyl)-3-hydroxybutan-2- -yl)-5-methyl-N3,N3-dipropylbenzene-1,3-diamide.

____________________________________________
If you want to learn more about this patent, please go directly to the U.S. Patent and Trademark Office Web site to access the full patent.