MXPA01008063A - &bgr;-SECRETASE ENZYME COMPOSITIONS AND METHODS - Google Patents

Abstract

Disclosed are various forms of an active, isolated&bgr;-secretase enzyme in purified and recombinant form. This enzyme is implicated in the production of amyloid plaque components which accumulate in the brains of individuals afflicted with Alzheimer's disease. Recombinant cells that produce this enzyme either alone or in combination with some of its natural substrates (&bgr;-APPwt and&bgr;-APPsw) are also disclosed, as are antibodies directed to such proteins. These compositions are useful for use in methods of selecting compounds that modulate&bgr;-secretase. Inhibitors of&bgr;-secretase are implicated as therapeutics in the treatment of neurodegenerative diseases, such as Alzheimer's disease.

Description

COMPOSITIONS AND METHODS OF ENZYME ß-SECRETASA Field of the Invention The invention relates to the discovery of several active forms of β-secretase, an enzyme that unfolds the β-amyloid precursor protein (APP) at one of the two cleavage sites necessary to produce the β-amyloid peptide ( Aß). The invention also relates to inhibitors of this enzyme, which are considered candidates for therapeutics in the treatment of amyloidogenic diseases such as Alzheimer's disease. Additional aspects of the present invention include selection methods, assays and kits for discovering such therapeutic inhibitors, as well as diagnostic methods for determining whether an individual carries a mutant form of the enzyme. Background of the Invention Alzheimer's disease is characterized by the presence of numerous amyloid plaques and neurofibrillary tangles present in the brain, particularly in those regions of the brain that involve memory and knowledge. The β-amyloid peptide (Aβ) is a peptide of 39-43 amino acids that is the main component of amyloid plaques and is produced by the unfolding of a large protein known as amyloid precursor protein (APP) at a site (s) specific (s) dentreo of the N-terminal region of the protein. The normal processing of APP includes the unfolding of the protein at the C-terminal point of amino acids 16-17 towards the N-terminal of the ß-AP region, releasing a secreted hectodomain, a-sAPP, thus preventing the production of ß-AP. The cleavage by the β-secretase enzyme of APP between Met671 and Asp672 and subsequent processes at the C-terminal end of APP produce the Aβ peptide, which is highly implicated in the etiology of Alzheimer's disease (Seubert et al. ., in Pharmacological Treatment of Alzheimer's disease, Wiley-Liss, Inc., pp 345-366, 1997; Zhao, J., et al., J. Biol. Chem. 271: 31407-31411, 1996). It is unclear if the level of β-secretase enzyme and / or activity is inherently higher than normal in Alzheimer's patients; however, it is clear that its cleavage product, the Aβ peptide, is abnormally concentrated in the amyloid plaques present in their brains. Therefore, it would be desirable to isolate, purify and characterize the enzyme responsible for the pathogenic cleavage of APP in order to help answer this and other questions surrounding the etiology of the disease. In particular, it is also desirable to use the isolated enzyme or active fragments thereof, in methods for screening candidate drugs for their ability to inhibit β-secretase activity. Drugs that exhibit inhibitory effects on β-secretase activity are expected to be useful therapeutics in the treatment of Alzheimer's disease and other amyloidogenic disorders characterized by the deposition of the Aβ peptide which contains fibrils. U.S. Patent 5,744,346 (Chrysler, et al.,) Describes the initial isolation and partial purification of the β-secretase enzyme characterized by its size (apparent molecular weight in the range of 260 to 300 kilodaltons when measured by gel exclusion chromatography). gel) and enzymatic activity (ability to unfold the 695 amino acid isotype of the β-amyloid precursor protein between amino acids 596 and 597). The present invention provides a significant improvement in the purity of the β-secretase enzyme, by providing a purified β-secretase enzyme that is at least 200 times purer than that described above. Such a purified protein has utility in several applications, including crystallization for structure determination. The invention also provides methods for producing recombinant forms of the β-secretase enzymes that have the same size and enzymatic profiles as the naturally occurring forms. It is a further discovery of the present invention, that human β-secretase is a so-called "aspartyl" (or "aspartic") protease.

Summary of the Invention The invention relates to the β-secretase protein which has now been purified for apparent homogeneity and in particular to a purified protein characterized by a specific activity of at least about 0.2 x 105 and preferably at least 1.0 x 105 nM / h / μg proteins in a representative β-secretase assay, the substrate assay MBP-C125sw. The resulting enzyme having a characteristic activity in the cleavage of the 695 amino acid isotype of the β-amyloid precursor protein (β-APP) between amino acids 596 and 597 thereof, is at least 10,000 times, preferably at least 20,000 times and more preferably more than 200,000 times greater specific activity than the activity exhibited by a solubilized but unenriched membrane fraction from human 293 cells, as previously characterized. In one embodiment, the purified enzyme is less than 450 amino acids in length, comprising a polypeptide having the amino acid sequence SEQ ID NO: 70 [63-452]. In the preferred embodiments, there are purified proteins in a variety of "truncated forms" relative to the proenzymes referred to herein as SEQ ID NO: 2 [1-501], such as the forms having the amino acid sequences SEQ ID NO. : 70 [63-452], SEQ ID NO: 69 [63-501], SEQ ID NO: 67 [58-501], SEQ ID NO: 68 [58-452], SEQ ID NO: 58 [46-452 ], SEQ ID NO: 74 [22-452], SEQ ID NO: 58 [46-452]. More generally, it has been found that particularly useful forms of the enzymes, particularly with respect to the crystallization studies described herein, are characterized by an N-terminus at position 46 with respect to SEQ ID NO: 2 and a C-terminus between positions 452 and 470 with respect to SEQ ID NO: 2 and more particularly by an N-terminus at position 22 with respect to SEQ ID NO: 2 and a C-terminus between positions 452 and 470 with respect to SEQ ID NO: 2. These forms are considered to unfold in the "anchor" domain of the transmembrane. Other particularly useful purified forms of the enzyme include: SEQ ID NO: 43 [46-501], SEQ ID NO: 66 [22-501] and SEQ ID NO: 2 [1-501]. More generally, it is appreciated that, useful forms of the enzyme have an N-terminal residue corresponding to a residue selected from the group consisting of 22, 46, 58 and 63 residues with respect to SEQ ID NO: 2 and a C -terminal selected from a residue between positions 452 and 501 with respect to SEQ ID NO: 2 or a C-terminal between residue positions 452 and 470 with respect to SEQ ID NO: 2. They are also described in FIG. present the enzyme forms isolated from a mouse, exemplified by SEQ ID NO: 65.

This invention is further directed to a crystalline protein composition formed from a purified β-secretase protein, such as the various protein compositions described above. According to one embodiment, the purified protein is characterized by its ability to bind to the inhibitor substrate of the β-secretase P10-P4'sta D- »V which is at least equal to the capacity exhibited by a protein having the sequence of amino acids SEQ ID NO: 70 [46-419], when the proteins are tested for binding to said substrate under the same conditions. According to another embodiment, the purified protein that forms the crystallization composition is characterized by its binding affinity for the β-secretase inhibitor substrate SEQ ID NO: 72 (P10-P4'sta D? V) which is at least 1/100 of the affinity exhibited by a protein having the amino acid sequence SEQ ID NO: 43 [46-501]. When said protein is tested for binding to the substrate under the same conditions. The proteins that form the crystalline composition can be glycosylated or deglycosylated. The invention also includes a crystalline protein composition containing a β-secretase substrate or inhibitor molecule, examples of which are provided herein, particularly exemplified by inhibitors derived from peptides such as SEQ ID NO: 78, SEQ ID NO: 72, SEQ ID NO: 81 and derivatives thereof. Inhibitors generally useful in this regard will have a Ki of no more than about 50 μM to 0.5 μM. Another aspect of the invention is directed to an isolated protein, comprising a polypeptide that (i) is less than about 450 amino acid residues in length (ii) includes an amino acid sequence that is at least 90% identical to SEQ ID NO : 75 [63-423] including conservative substitutions thereof and (iii) exhibits β-secretase activity, as evidenced by the ability to unfold a substrate selected from the group consisting of the 695 amino acid isotype of the beta amyloid precursor protein ( ßAPP) between amino acids 596 and 597 thereof, MBP-C125wt and MBP-C125sw. Peptides that meet this criterion include, but are not limited to polypeptides that include the sequence SEQ ID NO: 75 [63-423] such as SEQ ID NO: 58 [46-452], SEQ ID NO: 58 [ 46-452], SEQ ID NO: 58 [46-452], SEQ ID NO: 74 [22-452] and may also include conservative substitutions within such sequences. According to a further embodiment, the invention includes isolated protein compositions, such as those described above, in combination with a β-secretase inhibitor substrate or molecule, such as MBP-C125wt, MBP-C125sw, APP, APPsw and unfolding fragments β-secretase thereof. Additional β-secretase cleavable fragments useful in this regard are described in the specification thereof. Particularly useful inhibitors include peptides derived from or including SEQ ID NO: 78, SEQ ID NO: 81 and SEQ ID NO: 72. In general such inhibitors will have Kls of less than about 1 μM. Such inhibitors can be labeled with a detectable reporter molecule. Such labeled molecules are particularly useful, for example, in ligand binding assays. According to a further aspect, the invention includes protein compositions, such as those described above, expressed by a heterologous cell. According to a further embodiment, such cells can co-express a substrate or protein or peptide that is β-secretase-dehydrating. One or both of the expressed molecules can be heterologous to the cell. In a related embodiment, the invention includes antibodies that specifically bind to a β-secretase protein comprising a polypeptide that includes an amino acid sequence that is at least 90% identical to SEQ ID NO: 75 [63-423], including conservative substitutions thereof, but which lacks significant immunoreactivity with a protein a sequence selected from the group consisting of SEQ ID N0: 2 [1-501] and SEQ ID NO: 43 [46-501]. In a related additional embodiment, the invention includes isolated nucleic acids comprising a sequence of nucleotides encoding a β-secretase protein that is at least 95% identical to a protein selected from the group consisting of SEQ ID NO: 66 [22- 501], SEQ ID NO: 43 [46-501], SEQ ID NO: 57 [1-419], SEQ ID NO: 74 [22-452], SEQ ID NO: 58 [46-452], SEQ ID NO : 59 [1-452], SEQ ID NO: 60 [1-420], SEQ ID NO: 67 [58-501], SEQ ID NO: 68 [58-452], SEQ ID NO: 69 [63-501 ], SEQ ID NO: 70 [63-452], SEQ ID NO: 75 [63-423] and SEQ ID NO: 71 [46-419] or a sequence complementary to any of such nucleotides. Specifically excluded from this nucleotide is a nucleic acid encoding a protein having the sequence SEQ ID NO: 2 [1-501]. Additionally, the invention includes an expression vector comprising such isolated nucleic acids operably linked to the nucleic acid with regulatory sequences effective for expression of the nucleic acid in a selected host cell, for heterologous expression. The host cells can be a eukaryotic cell, a bacterial cell, an insect cell or a yeast cell. Such cells can be used, for example in a method for producing a recombinant β-secretase enzyme, wherein the method further includes subjecting a cultured medium or extract of said cell to an affinity matrix, such as a matrix formed from a molecule or β-secretase inhibitor antibody as described above. The invention is also directed to a screening method for compounds that inhibit the production of Aβ, which comprises contacting a β-secretase polypeptide, such as those of full length or truncated form described above, with (i) a compound of test and (ii) a β-secretase substrate and select the test compound as capable of inhibiting the production of Aβ if the β-secretase peptide exhibits less β-secretase activity in the presence than in the absence of the test compound . Such an assay can be cell-based, with one or both of the enzymes and the substrate produced by the cell, such as the co-expression cell referred to above. The equipment incorporating such selection methods is also part of the invention. The selection method may further include administering a test compound to a mammalian subject having Alzheimer's disease or Alzheimer's disease-like pathologies and selecting the compound as a therapeutic agent candidate if, following such administration, the subject maintains or improves their cognitive ability or the subject shows reduced plate loading.

Preferably, such a subject is one comprising a transgene for the human β-amyloid precursor protein (β-APP), such as a mouse carrying a transgene encoding a human β-APP, including mutant variants thereof, as exemplified in the specification. In a related embodiment, the invention includes the β-secretase inhibitor compound selected according to the methods described above. Such compounds can be selected, for example, from a selection system that displays the phage ("file"), as is known in the art. According to another aspect, such files can be "derived" from the sequence peptide SEQ ID NO: 97 [P10-P4 'D->. V]. Other inhibitors include or may be derived from peptide inhibitors identified herein, such as the inhibitors SEQ ID NO: 78, SEQ ID NO: 72, SEQ ID NO: 78 and SEQ ID NO: 81. Mice also form part of the invention knock out characterized by the inactivation or cancellation of an endogenous β-secretase gene, such as the genes encoding a protein having at least 90% sequence identity to the sequence SEQ ID NO: 65. Cancellation or inactivation can be induced, such as by insertion of a Cre-lox expression system into the mouse genome. According to a further related aspect, the invention includes a method of selecting drugs effective in the treatment of Alzheimer's disease or other cerebrovascular amyloidosis characterized by the deposition of Aβ. According to this aspect of the invention, a mammalian subject characterized by overexpression of β-APP and / or Aβ deposition is given a test compound selected for its ability to inhibit the activity of β-secretase, a β-protein. -secretase according to claim 37. The compound is selected as a potential therapeutic drug compound, if it reduces the amount of Aβ deposition in the subject or if it maintains or improves the cognitive capacity in the subject. According to a preferred embodiment, the mammalian subject is a transgenic mouse carrying a transgene encoding a human β-APP or a mutant thereof. The invention also includes a method for treating a patient suffering from or having a propensity for Alzheimer's disease or other cerebrovascular amyloidoses. According to this aspect, the enzymatic hydrolysis of APP to Aβ is blocked by administering to the patient a pharmaceutically effective dose of a compound effective to inhibit one or more of the various forms of the enzyme described herein. According to another feature, the therapeutic compound is derived from a peptide selected from the group consisting of SEQ ID NO: 72, SEQ ID NO: 78, SEQ ID NO: 81 and SEQ ID NO: 97. Such derivation can be affected by various phage selection systems described herein, in conjunction with the selection methods of the invention or other methods. Alternatively or in addition, derivation can be achieved through rational chemical approaches, including molecular modeling, known in the medicinal chemistry art. Such compounds are preferably rather potent inhibitors of the β-secretase enzyme activity, evidenced by a Ki of less than about 1-50 μM in a MBP-C125sw assay. Such a compound also forms the basis for the therapeutic drug compounds according to the present invention, which may also include a pharmaceutically effective excipient. According to yet another related aspect, the invention includes a method for diagnosing the presence of or a predisposition for Alzheimer's disease in a patient. This method includes detecting the level of expression of a gene comprising a nucleic acid encoding the β-secretase in a cell sample from a patient and diagnosing the patient as having or having a propensity to Alzheimer's disease, if the level of expression is significantly greater than a predetermined control expression level. Detectable nucleic acids and primers useful in such detection, are described in detail herein. Such nucleic acids can exclude a nucleic acid encoding preproenzyme [1-501]. The invention is further directed to the method for diagnosing the presence of or a propensfor Alzheimer's disease in a patient, which comprises measuring the enzymatic activof the β-secretase in a cellular sample of a patient and diagnosing the patient as having or it is prone to Alzheimer's disease, if the level of enzymatic activis significantly higher than a predetermined level of control activ The diagnostic methods can be carried out in a whole cell assay and / or a nucleic acid derivative of a patient's cell sample. The invention also includes a method for purifying a β-secretase protein enzyme molecule. According to this aspect, an impure sample contains activof the β-secretase enzyme with an affinmatrix that includes a β-secretase inhibitor, such as the various inhibitory molecules described herein. These and other objects and features of the invention will become fully apparent when the following detailed description of the invention is read in conjunction with the accompanying drawings. Brief Description of the Figures Figure 1A shows the sequence of a polynucleotide (SEQ ID NO: 1) encoding the human β-secretase translation product shown in Figure 2A.

Figure IB shows the polynucleotide of Figure 1A, including the putative 5 'and 3' untreated regions (SEQ ID NO: 44). Figure 2A shows the amino acid sequence (SEQ ID NO: 2) [1-501] of the predicted translation product of the open reading structure of the polynucleotide sequence shown in Figures 1A and IB. Figure 2B shows the amino acid sequence of an active fragment of human β-secretase (SEQ ID NO: 43) [46-501]. Figure 3A shows the translation product encoding an active fragment of human β-secretase, 452stop, (amino acids 1-452 with reference to SEQ ID NO: 2; SEQ ID NO: 59) including an epitope-FLAG marker (underlined; SEQ ID NO: 45) at the C-terminus. Figure 3B shows the amino acid sequence of a fragment of human β-secretase (amino acids 46-452 (SEQ ID NO: 58) with reference to SEQ ID NO: 2; including an epitope-FLAG marker (underlined; SEQ ID NO: 45) at the C-terminus. Figure 4 shows an elution profile of recombinant β-secretase eluted from a gel filtration column. Figure 5 shows the full-length amino acid sequence of β-secretase 1-501 (SEQ ID NO: 2) including the ORF encoding it (SEQ ID NO: 1), with certain characteristics indicated, such as sites " D-active "indicating active catalytic sites of aspartic acid, a transmembrane region starting at position 453, as well as the leader sequence (" Signal ") (residues 1-22; SEQ ID NO: 46) and the putative region (residues 23-45; SEQ ID NO: 47) and where the polynucleotide region corresponding to the porenzyme region corresponding to amino acids 46-501 is shown (SEQ ID NO: 43) (nt 135-1503) as SEQ ID NO: 44. Figures 6A and 6B show images of SDS-PAGE gels stained with silver in which fractions containing purified β-secretase were run under conditions of reduction (6A) and not reduction (6B). Figure 7 shows a silver-stained SDS-PAGE of the purified β-secretase from 293T heterologous cells expressing the recombinant enzyme. Figure 8 shows a silver-stained SDS-PAGE of the purified β-secretase from Cos A2 homologous cells expressing the recombinant enzyme. Figure 9 shows a scheme in which the primers derived from the polypeptide (SEQ ID NO: 76) encoding the N-terminus of the naturally occurring purified β-secretase (SEQ ID NO: 77) were used for the portions of additional PCR-clone molecules, such as fragment SEQ ID NO: 79 which is encoded by the nucleic acid SEQ ID NO: 78, as illustrated. Figure 10 shows an alignment of the amino acid sequence of human β-secretase ("Seq Imapain Humana" 1-501, SEQ ID NO: 2) compared to ("pBS / mlmpain H # 3 cons") consensus sequence of mouse: SEQ ID NO: 65. Figure HA shows the nucleotide sequence (SEQ ID NO: 80) of an insert used in the pCF preparation vector. Figure 11B shows a linear scheme of pCEK. Figure 12 shows a scheme of clone pCEK 27 used for mammalian cells transfected with β-secretase. Figure 13 (A-E) shows the nucleotide sequence of clone pCEK 27 (SEQ ID NO: 48) with the OFR indicated by the amino acid sequence SEQ ID NO: 2. Figure 14A shows the nucleotide sequence inserted in the original vector pCDNA3. Figure 14B shows a graph of β-secretase activity in lysed cells of COS transfected cells with vectors derived from clones encoding β-secretase. Figure 15A shows an image of an SDS gel PAGE loaded with triplicate samples of the lysates made up of heterologous cells transfected with mutant APP (751wt) and β-galactosidase as control (d bands) and of cells transfected with mutant APP (751wt) and β-secretase (f-bands) where the bands a, b and c show lysates of untreated cells, cells transfected only with β-galactosidase and cells transfected only with β-secretase, respectively, and the band and indicates the markers. Figure 15B shows an image of an SDS PAGE gel loaded with triplicate samples of the lysates made up of heterologous cells transfected with mutant APP (Swedish mutation) and ß-galactosidase as control (c bands) and of cells transfected with mutant APP (Swedish mutation ) and β-secretase (e bands) where bands a and b show lysates of cells transfected only with β-galactosidase and cells transfected only with β-secretase and band d indicates the markers. Figures 16A and 16B show the Western blots of the cell supernatants tested for the presence or increase in soluble APP (sAPP). Figures 17A and 17B show the Western blots of the a-unfolding APP substrate in co-expressing cells. Figure 18 shows the production of Aβ (x-40) in 293T cells co-transfected with APP and β-secretase. Figure 19A shows a schematic of a substrate fragment of APP and its use in relation to antibodies SW192 and 8E-192 in the assay. Figure 19B shows the cleavage sites of β-secretase in the wild type sequence and APP Swedish. Figure 20 shows a schematic of a second substrate fragment of APP derived from APP 638 and its use in conjunction with antibodies SW192 and 8E-192 in the assay. Figure 21 shows a scheme of the pohCK751 vector. Brief Description of the Sequences This section briefly identifies the sequence identification numbers referred to herein. The number ranges shown here in square brackets and throughout the specification refer to the amino acid sequence SEQ ID NO: 2, using the conventional order N? C terminal. SEQ ID NO: 1 is a nucleic acid sequence encoding human β-secretase, including an active fragment, as exemplified herein. SEQ ID NO: 2 is the predicted translation product of SEQ ID NO: l [1-501]. SEQ ID NOS: 3-21 are degenerate oligonucleotide primers described in Example 1 (Table 4), designed from the regions of SEQ ID NO: 2. SEQ ID NOS: 22-41 are additional oligonucleotide primers. used in the PCR cloning methods described herein, shown in Table 5. SEQ ID NO: 42 is a polynucleotide sequence encoding the active β-secretase enzyme shown as SEQ ID NO: 43. SEQ ID NO : 43 is the sequence of a portion of the active enzyme of the human β-secretase, the N-terminal which corresponds to the N-terminal of the predominant form of the protein isolated from natural sources [46-501]. SEQ ID NO: 44 is a polynucleotide encoding SEQ ID NO: 2, including the 5 'and 3' untranslated regions. SEQ ID NO: 45 is the FLAG sequence used in connection with certain polynucleotides. SEQ ID NO: 46 is the putative guide region of β-secretase [1-22]. SEQ ID NO: 47 is the pre-pro putative region of the β-secretase [23-45]. SEQ ID NO: 48 is the sequence of clone pCEK C1.27 (Figures 13A-E). SEQ ID NO: 49 is a nucleotide sequence of a fragment of the gene encoding human β-secretase. SEQ ID NO: 50 is the predicted translation product of SEQ ID NO: 49. SEQ ID NO: 51 is the predicted internal amino acid sequence of a portion of human β-secretase.

SEQ ID NOS: 52 and 53 are peptide substrates suitable for use in β-secretase assays used in the present invention. SEQ ID NO: 54 is a peptide sequence cleavage site recognized by human β-secretase. SEQ ID NO: 55 are amino acids 46-69 of SEQ ID NO: 2. SEQ ID NO: 56 is an internal peptide only from the N-terminus for the transmembrane domain of β-secretase. SEQ ID NO: 57 is the β-secretase [1-419]. SEQ ID NO: 58 is the β-secretase [46-452]. SEQ ID NO: 59 is the β-secretase [1-452]. SEQ ID NO: 60 is the β-secretase [1-420]. SEQ ID NO: 61 is EVM [hydroxyethylene] AEF. SEQ ID NO: 62 is the amino acid sequence of the transmembrane domain of the β-secretase shown in (Figure 5). SEQ ID NO: 63 is P26-P4 'of APPwt. SEQ ID NO: 64 is P26-P1 'of APPwt. SEQ ID NO: 65 is the mouse β-secretase (Figure , lower sequence). SEQ ID NO: 66 is the β-secretase [22-501]. SEQ ID NO: 67 is the β-secretase [58-501]. SEQ ID NO: 68 is the β-secretase [58-452]. SEQ ID NO: 69 is the β-secretase [63-501].

SEQ ID NO: 70 is the β-secretase [63-452]. SEQ ID NO: 71 is the β-secretase [46 -419]. SEQ ID NO: 72 is P10-P4 'staD? V. SEQ ID NO: 73 is P4 - P4 'staD? V (KTEEISEVN [sta] VAEF) SEQ ID NO: 74 is the β-secretase [22-452]. SEQ ID NO: 75 is the β-secretase [63-423]. SEQ ID NO: 76 is the nucleic acid encoding the N-terminal of the β-secretase that occurs naturally. SEQ ID NO: 77 is the peptide fragment in the N-terminal of the β-secretase that occurs naturally. SEQ ID NO: 78 is a P3-P4'XD? V (VMXVAEF, where X is hydroxyethylene or statin). SEQ ID NO: 79 is a peptide fragment of the β-secretase that occurs naturally. SEQ ID NO: 80 is a nucleotide inserted into the pCF vector used therein. SEQ ID NO: 81 is P4-P4'XD? V (EVMXVAEF, where X is hydroxyethylene or statin SEQ ID NO: 82 is an APP SEVKMDAEF fragment (P5- P4'wt) SEQ ID NO: 83 is an APP SEVNLDAEF fragment (P5- P4 'sw) SEQ ID NO: 84 is a SEVKLDAEF APP fragment, SEQ ID NO: 85 is an APP SEVKFDAEF fragment.

SEQ ID NO: 86 is an APP SEVNFDAEF fragment. SEQ ID NO: 87 is a SEVKMAAEF APP fragment. SEQ ID NO: 88 is a SEVNLAAEF APP fragment. SEQ ID NO: 89 is a SEVKLAAEF APP fragment. SEQ ID NO: 90 is a SEVKMLAEF APP fragment. SEQ ID NO: 91 is a SEVNLLAEF APP fragment. SEQ ID NO: 92 is a SEVKLLAEF APP fragment. SEQ ID NO: 93 is a SEVKFAAEF APP fragment. SEQ ID NO: 94 is an APP SEVNFAAEF fragment. SEQ ID NO: 95 is a SEVKFLAEF APP fragment. SEQ ID NO: 96 is a SEVNFLAEF APP fragment. SEQ ID NO: 97 is a fragment derived from APP P10-P4 '(D? V): KTEEISEVNLVAEF. Detailed Description of the Invention 1. Definitions Unless otherwise indicated, all terms used herein have the same meaning as they would have for the person skilled in the art of the present invention. The practitioners are particularly directed to Sambrook, et al., (1989) Molecular Cloning: A Laboratory Manual (Second Edition), Cold Spring Harbor Press, Plainview, N.Y. , and Ausubel, F.M., et al. , (1998) Current Protocols in Molecular Biology, John Wiley & Sons, New York, N.Y. , for definitions, technical terms and standard methods known in the molecular biology art, particularly as regards the cloning protocols described herein. It is understood that this invention is not limited to the particular methodology, protocols, and reagents described, since they may vary to produce the same result. The terms "polynucleotide" and "nucleic acid" are used interchangeably herein and refer to a polymeric molecule having a structure that supports bases capable of hydrogen bonding to typical polynucleotides, wherein the polymer structure presents the bases in a manner which allows such hydrogen bonding in a specific sequence manner between the polymer molecule and a typical polynucleotide (e.g., Simple braid DNA). Such bases are typically inosine, adenosine, guanosine, cytosine, uracil and thymidine. The polymeric molecules include double and single stranded DNA and RNA, and structural modifications thereof, for example, methylphosphonate linkages. The term "vector" refers to a polynucleotide having a nucleotide sequence that can assimilate new nucleic acids, and propagate those new sequences in an appropriate host. Vectors include, but are not limited to plasmids and recombinant viruses. The vector (for example, plasmid or recombinant virus) comprising the nucleic acid of the invention can be in a carrier, for example, a plasmid in complex with the protein, a plasmid in complex with lipid-based nucleic acid transduction systems , or other non-viral vehicle systems. The term "polypeptide" as used herein refers to a compound made from a single chain of amino acid residues linked by peptide bonds. The term "protein" may be synonymous with the term "polypeptide" or may refer to a compound of two or more polypeptides. The term "modified", when referring to a polypeptide of the invention, means a polypeptide that is modified either by natural processes, such as processing or other post-translational modifications, or by chemical modification techniques that are well known in the art. . Among the many known modifications that may be present include, but are not limited to, acetylation, acylation, amidation, ADP-ribosylation, glycosylation, GPI anchor formation, covalent attachment of a lipid or a lipid derivative, methylation, myeliation, pegylation , prenylation, phosphorylation, ubiquination, or any similar process. The term "β-secretase" is defined here in Section III. The term "biologically active" used in conjunction with the term β-secretase refers to the possession of a β-secretase enzyme activity, such as the ability to unfold the β-amyloid precursor protein (APP) to produce the β-peptide. -aminoid (Aß). The term "fragment", when referring to the β-secretase of the invention, means a polypeptide having an amino acid sequence that is the same as a part but not the entire amino acid sequence of the full-length β-secretase polypeptide . In the context of the present invention, the full length β-secretase is generally identified as SEQ ID NO: 2, the ORF of the full length nucleotide; however, according to a discovery of the invention, the active form that occurs naturally is probably one or more N-terminal truncated versions, such as amino acids 46-501, 22-501, 58-501 or 63-501; other active forms are C-terminal truncated forms ending approximately between amino acids 450 and 452. The numbering system used throughout the document is based on the numbering of the sequence SEQ ID NO: 2. An "active fragment" is a fragment of β-secretase that retains at least one of the β-secretase functions or activities, including but not limited to the above-treated β-secretase enzyme activity and / or the ability to bind to the inhibitor substrate described herein as P10- P4 'staD- »V. The contemplated fragments include, but are not limited to, a β-secretase fragment that retains the ability to unfold the β-amyloid precursor protein to produce the β-amyloid peptide. Such a fragment preferably includes at least 350, and more preferably at least 400, contiguous amino acids or conservative substitutions thereof of β-secretase, as described herein. More preferably, the fragment includes active residues of aspartyl acid in the structural proximities identified and defined by the primary structure of the polypeptide shown as SEQ ID NO: 2 and denoted here also as "D-active" sites. A "conservative substitution" refers to the substitution of an amino acid in a class by an amino acid in the same class, where a class is defined by common physicochemical properties of amino acid side chain and high substitution frequencies in homologous proteins found in nature (as determined, eg, by a standard Dayhoff frequency-exchange matrix or a BLOSUM matrix). The six general classes of amino acid side chain, categorized as described above, include: Class I (Cis); Class II (Ser, Tre, Pro, Ala, Gli); Class III (Asn, Asp, Gln, Glu); Class IV (His, Arg, Lis); Class V (He, Leu, Val, Met); and Class VI (Fen, Tir, Tri). For example, it is considered that the substitution of an Asp for another class III residue such as an Asn, Gln, or Glu, is a conservative substitution. "Optimal alignment" is defined as an alignment that gives the highest percentage score of identity. Such alignment can be carried out using a variety of commercially available sequence analysis programs, such as the local alignment program LALIGN using a ktup of 1, default parameters and omitting PAM. A preferred alignment is the pairwise alignment using the CLUSTAL-W program in MacVector, operated with default parameters, including an open interval penalty of 10.0, an extended interval penalty of 0.1, and a BLOSUM30 matrix of similarity. The "percent sequence identity", with respect to two amino acid or polynucleotide sequences, refers to the percentage of residues that are identical in the two sequences when the sequences are optimally aligned. Thus, 80% amino acid sequence identity means that 80% of the amino acids in two or more optimally aligned polypeptide sequences are identical. If it is necessary to insert a range within a first sequence to optimally align it with a second sequence, the percent identity is calculated using only the residues that are in pairs with a corresponding amino acid residue (ie, the calculation does not consider the residues in the second sequences that are in the "range" of the first sequence A first polypeptide region is said to be "corresponds" to a second polypeptide region when the regions are essentially co-extensive when the sequences containing the regions are aligned using a sequence alignment program, as in the above. The corresponding polypeptide regions typically contain a similar, if not identical, number of residues. It will be understood, however, that the corresponding regions may contain insertions or deletions of residues one with respect to the other, as well as some differences in their sequences. A first polynucleotide region is said to "correspond" to a second polynucleotide region when the regions are essentially co-extensive when the sequences containing the regions are aligned using a sequence alignment program, as in the above. The corresponding polynucleotide regions typically contain a similar, if not identical, number of residues. It will be understood, however, that the corresponding regions may contain insertions or deletions of bases with respect to each other, as well as some differences in their sequences. The term "sequence identity" means the identity of nucleic acid or amino acid sequence in two or more aligned sequences, aligned as defined above. The "sequence similarity" between two polypeptides is determined by comparing the amino acid sequence and its conserved amino acid substitutes of a polypeptide to the sequence of a second polypeptide. Thus, 80% protein sequence similarity means that 80% of the amino acid residues in two or more aligned protein sequences are conserved as amino acid residues, i.e., they are conservative substitutions. "Hybridization" includes any process by which a strand of nucleic acid binds with a complementary strand of nucleic acid through base pairs. Thus, strictly speaking, the term refers to the ability of the complement of the target sequence to join the test sequence, or vice versa. The "hybridization conditions" are based in part on the melting temperature (Tm) of the complex or nucleic acid binding test and are typically classified by the degree of "stiffness" of the conditions under which the hybridization is measured. The specific conditions that define the various degrees of stiffness (ie, high, medium, low) depend on the nature of the polynucleotide whose hybridization is desired, particularly at its GC percentage content, and can be determined empirically according to the methods known in the art. Functionally, maximum stiffness conditions can be used to identify nucleic acid sequences that have strict identity or near-strict identity with the hybridization test; while high stiffness conditions are used to identify nucleic acid sequences that have approximately 80% or more sequence identity with the test. The term "gene" as used herein means the segment of DNA involved to produce a polypeptide chain; may include regions that precede and follow the coding region, e.g., 5 'untranslated sequences (5'UTR) or "guide" and sequences 3'UTR or "tracker", as well as sequences (introns) that intervene between the individual segments of coding (exons). The term "isolated" means that the material is removed from its original environment (e.g., the original environment if it occurs naturally). For example, a naturally occurring polynucleotide or polypeptide present in a living animal is not isolated, but the same polynucleotide or polypeptide, separated from any or all of the coexisting materials in the natural system, is isolated. Such isolated polynucleotides can be part of a vector and / or such polynucleotides or polypeptides can be part of a composition, such as a recombinantly produced cell (heterologous cell) expressing the polypeptide, and even being isolating them in such a vector or composition is not part of their natural environment. An "isolated polynucleotide having a sequence encoding β-secretase" is a polynucleotide containing the β-secretase coding sequence, or an active fragment thereof, (i) alone, (ii) in combination with additional sequences of coding, such as fusion protein or signal peptide, in which the β-secretase coding sequence is the dominant coding sequence, (iii) in combination with non-coding sequences, such as introns and control elements, such as the promoter and terminator elements or 5 'and / or 3' untranslated regions, effective for the expression of the coding sequence in a suitable host, and / or (iv) in a vector or host environment in which the coding sequence β-secretase is a heterologous gene. The terms "heterologous DNA", "heterologous RNA", "heterologous nucleic acid", "heterologous gene", and "heterologous polynucleotide" refer to nucleotides that are not endogenous to the cell or part of the genome in which they are present; generally such nucleotides have been added to the cell, by transfection, micro injection, electroporation, or the like. Such nucleotides generally include at least one coding sequence, but this coding sequence need not be expressed. The term "heterologous cell" refers to a recombinantly produced cell that contains at least one heterologous DNA molecule. A "recombinant protein" is an isolated, purified, or identified protein by virtue of expression in a heterologous cell, this cell having been transduced or transfected, either transiently or stably, with a recombinant expression vector designed to drive the expression of the protein in the host cell. The term "expression" means that a protein is produced by a cell, commonly as a result of transfecting the cell with a heterologous nucleic acid. "Co-expression" is a process by which two or more proteins or RNA species of interest are expressed in a single cell. The co-expression of the two or more proteins is typically achieved by transfecting the cell with one or more recombinant expression vectors carrying coding sequences for the proteins. In the context of the present invention, for example, it can be said that a cell "co-expresses" two proteins, if one or both proteins are heterologous to the cell. The term "expression vector" refers to vectors that have the ability to incorporate and express heterologous DNA fragments in an external cell. Many prokaryotic and eukaryotic expression vectors are commercially available. The selection of appropriate expression vectors is within the knowledge of those skilled in the art. The terms "purified" or "substantially purified" refer to molecules, either polynucleotides or polypeptides, that have been removed from their natural environment, isolated or separated, and are at least 90% and more preferably at least one -99% free of other components with which they are associated naturally. Notwithstanding the foregoing, such a descriptor does not prevent the presence in the same sample of splicing or other protein variants (glycosylation variants) in the same sample, otherwise homogeneous. It is generally considered that a protein or polypeptide is "purified to apparent homogeneity" if a sample containing it shows a single band of protein on an electrophoretic polyacrylamide silver staining gel.

The term "crystallized protein" means a protein that is co-precipitated out of the solution into pure crystals consisting only of the crystal, but possibly including other components that are tightly bound to the protein. A "variant" polynucleotide sequence can encode an altered "variant" amino acid sequence by one or more amino acids from the reference polypeptide sequence. The variant polynucleotide sequence can encode a variant amino acid sequence, which contains "conservative" substitutions, wherein the substituted amino acid has structural or chemical properties similar to the amino acid it replaces. Additionally, or alternatively, the variant polynucleotide sequence may encode a variant amino acid sequence, which contains "non-conservative" substitutions, wherein the substituted amino acid has structural or chemical properties dissimilar to the amino acid it replaces. The variant polynucleotides may also encode variant amino acid sequences, which contain amino acid insertions or deletions, or both. In addition, a variant polynucleotide can encode the same polypeptide as the reference polynucleotide sequence but, due to the degeneracy of the genetic code, it has a polynucleotide sequence altered by one or more bases from the reference polynucleotide sequence. An "allelic variant" is an alternate form of a polynucleotide sequence, which may have a substitution, deletion or addition of one or more nucleotides that does not substantially alter the function of the encoded polypeptide. "Alternative splicing" is a process by which multiple iso-forms of polypeptide are generated from a single gene, and involves splicing together non-consecutive exons during the processing of some, but not all, transcripts of the gene. In this way, a particular exon can be connected to any of several alternative exons to form messenger RNAs. The alternatively spliced mRNAs produce polypeptides ("splice variants" in which some parts are common while other parts are different.) The "splice variants" of the β-secretase, when referenced in the context of a mRNA transcript, are mRNA produced by the alternative splicing of coding regions, ie, exons, from the β-secretase gene.Referring to the "splice variants" of β-secretase in the context of the protein itself, they are products of β-secretase translation that are encoded by alternatively spliced β-secretase mRNA transcripts.

A "mutant" amino acid or polynucleotide sequence is a variant amino acid sequence, or a variant polynucleotide sequence, that encodes a variant amino acid sequence that has significantly altered biological activity or function from that of the naturally occurring protein. A "substitution" results from the replacement of one or more nucleotides or amino acids by different nucleotides or amino acids, respectively. The term "modulated" as used herein refers to the change in activity of the polypeptide of the invention. The modulation may refer to an increase or decrease in biological activity, binding characteristics, or any other biological, functional or immunological property of the molecule. The terms "antagonist" and "inhibitor" are used interchangeably herein and refer to a molecule which, when bound to the polypeptide of the present invention, modulates the activity of the enzyme by blocking, decreasing or shortening the duration of biological activity . An antagonist as used herein may also be referred to as a "β-secretase inhibitor" or a "β-secretase blocker". The antagonists may themselves be polypeptides, nucleic acids, carbohydrates, lipids, small molecules (commonly less than 1000 kD), or derivatives thereof., or any other ligand that binds and modulates the activity of the enzyme. ß-secretase compositions The present invention provides an isolated human β-secretase active enzyme, which is further characterized as a protease or aspartyl (aspartic) proteinase, optionally in purified form. As defined more fully in the following sections, the β-secretase exhibits a proteolytic activity involved in the generation of the β-amyloid peptide from the β-amyloid precursor protein (APP), as described in US Pat. 5,744,346, incorporated herein by reference. Alternatively, or additionally, the β-secretase is characterized by its ability to bind, with a moderately high affinity, to an inhibitory substrate described herein as P10-P4 'staD → V (SEQ ID NO: 72). According to an important characteristic of the present invention, a human form of β-secretase has been isolated, and its natural form has been characterized, purified and sequenced. According to another aspect of the invention, nucleotide sequences encoding the enzyme have been identified. In addition, the enzyme has been further modified for expression in altered forms, such as truncated forms, which have a protease activity similar to that of the naturally occurring or full-length recombinant enzyme.

Using the information provided herein, practitioners can isolate DNA that encodes various active forms of the protein from available sources and can express the protein recombinantly in a convenient expression system. Alternatively and additionally, professionals can purify the enzyme from natural or recombinant sources and use it in purified form to further characterize its structure and function. According to a further feature of the invention, the polynucleotides and proteins of the invention are particularly useful in a variety of classification analysis formats, including cell-based classification for drugs that inhibit the enzyme. Examples of uses of such analyzes are provided, as well as additional utilities for the compositions in Section IV, below. The β-secretase is of particular interest due to its activity and implication to generate fibril peptide components that are the major components of the amyloid plaques in the central nervous system (CNS), as seen in Alzheimer's disease, Down syndrome and other CNS disorders. Accordingly, a useful feature of the present invention includes an isolated form of the enzyme that can be used, for example, for the classification of inhibitory substances that are candidates as therapeutics for such disorders. A. Isolation of Polynucleotides that encode Human β-secretase. Polynucleotides encoding human β-secretase were obtained by PCR cloning and hybridization techniques as detailed in Examples 1-3 and described below. Figure 1A shows the sequence of a polynucleotide (SEQ ID NO: 1) encoding a form of human β-secretase (SEQ ID NO: 2 [1-501].) Polynucleotides encoding human β-secretase are conveniently isolated from from any of several human tissues, preferably tissues of neuronal origin, including but not limited to neuronal cell lines such as the commercially available human neuroblastoma cell line IMR-32 available from American Type Culture Collection (Manassas, VA; ATTC CCL 127) and human fetal brain, such as a human fetal brain cDNA file available from OriGene Technologies Inc. (Rockville, MD). Briefly, the regions encoding the human β-secretase were isolated by methods well known in the art, using hybridization tests derived from the coding sequence provided as SEQ ID NO: 1. Such tests can be designed and produced by well-known methods in The technique. Exemplary tests, including degeneration tests, are described in Example 1. Alternatively, a cDNA file is classified by PCR, using, for example, the primers and conditions described herein in Example 2. Such methods are discussed in more detail below in Part B. cDNA files were also classified using a 3 '-RACE protocol (Rapid Amplification of cDNA Terminals) according to methods well known in the art (White, BA, PCR Cloning Protocols edition, Humana Press, Totowa, NJ, 1997, schematically shown in FIG 9). Here the primers derived from the 5 'portion of SEQ ID NO: 1 are added to the partial cDNA substrate clone found by classifying a fetal brain cDNA file as described above. A representative 3 'RACE reaction used to determine the major sequence is detailed in Example 3 and is described in more detail below in Part B. The human β-secretase, as well as additional members of the neuronal aspartyl protease family described herein can be identified by the use of degenerate random primers designed according to any portion of the polypeptide sequence shown as SEQ ID NO: 2. For example, in experiments carried out in support of the present invention, and detailed here in Example 1, eight groups of degenerate initiator, each degenerated 8 times, were designed based on a single 22 amino acid peptide region selected from SEQ ID: 2. Such techniques can be used to identify similar additional sequences from other species and / or representing other members of this protease family. Preparation of polynucleotides The polynucleotides described herein can be obtained by classifying cDNA files using oligonucleotide tests, which can be hybridized and / or polynucleotides that can be amplified by PCR encoding human β-secretase as described above. CDNA files prepared from a variety of tissues are commercially available, and methods for classifying and isolating cDNAs are well known to those skilled in the art. Similarly genomic files can be classified to obtain genomic sequences including regulatory regions and introns. Such techniques are described, for example, in Sambrook et al. , (1989) Molecular Cloning: A Laboratory Manual (2nd edition), Cold Spring Harbor Press, Plainview, N.Y. and Ausubel, FM et al. , (1998) Current Protocols in Molecular Biology, John Wiley & amp;; Sons, New York, N.Y. The polynucleotides can be extended to obtain upstream and downstream sequences such as promoters, regulatory elements and 5 'and 3' untranslated regions (UTRs). The extent of the available transcribed sequence can be carried out by numerous methods known to those skilled in the art, such as PCR or primer extension (Sambrook et al.,, Supra) or by the RACE method using, for example, the kit MARATHON RACE (Cat. Number K1802-1; Clontech Palo Alto CA). Alternatively, the "restriction site" PCR technique (Gobinda et al., (1993) PCR Methods Applic. 2: 318-22), which utilizes universal primers for the recovery of flanking sequence adjacent to a known location, can used to generate additional coding regions. The first genomic DNA is amplified in the presence of an initiator to a binding sequence and a specific primer to the known region. The amplified sequences are subjected to a second PCR step with the same linker and another internal specific primer to the first. The products of each PCR step are transcribed with an appropriate RNA polymerase and sequenced using reverse transcriptase. Reverse PCR can be used to amplify or extend sequences using divergent primers based on a known region (Triglia T et al., (1998) Nucleic Acids Res 16: 8186). The primers can be designed using OLIGO (R) 4.06 Primer Analysis Software (1992); National Biosciences Inc., Plymouth, Minn.), Or other suitable program, to be 22 to 30 nucleotides in length, to have a GC content of 50% or more, and to anneal the target sequence at temperatures of approximately 68-72 °. C. the method uses several restriction enzymes to generate a suitable fragment in the known region of a gene. The fragment is then circulated by intramolecular ligation and used as a PCR model. The PCR capture (Lagerstrom M et al., (1991) PCR Methods Applic 1: 111-19) is a method for the PCR amplification of DNA fragments adjacent to a known sequence in the artificial chromosome of human DNA and yeast . PCR capture also requires digestions and multiple restriction enzyme ligations to place a designed double stranded sequence within a flank part of the DNA molecule prior to PCR. Another method that can be used to recover flank sequences is that of Parker, JD et al. , (1991; Nucleic Acids Res 19: 3055-60). Additionally, PCR, nested primers and PromoterFinder (TM) files "to carry out" genomic DNA (Clontech, Palo Alto, CA) can be used. This process avoids the need to classify files and is useful for finding intron / exon junctions. Preferred files for full-length cDNA classification are those that have been selected by their dimension to include larger cDNAs. Randomly initiated files in which more sequences are contained than those containing the 5 'regions and upstream of genes are also preferred. A randomly initiated file can be particularly useful if an oligo d (T) file does not produce a full-length cDNA. Genomic files are useful by extension in the 5 'untranslated regulatory region. The polynucleotides and oligonucleotides of the invention can also be prepared by solid phase methods according to known synthetic methods. Typically, fragments of more than about 100 bases are individually synthesized, then joined to form continuous sequences up to several hundred bases. B. Isolation of β-secretase The amino acid sequence for a full-length human β-secretase translation product is shown as SEQ ID NO: 2 in Figure 2A. According to the discovery of the present invention, this sequence represents a "prepro" form of the enzyme deduced from the nucleotide sequence information described in the previous section in conjunction with the methods described below. The comparison of this sequence with the sequences determined from the biologically active form of the purified enzyme from natural sources, as described in Part 4 below, indicates that it is likely that an active and predominant form of the enzyme is represented by the sequence shown in Figure 2B (SEQ ID NO: 43), in which the first 45 amino acids of the deduced sequence of open reading structure have been removed. This suggests that the enzyme may be modified post-translationally by proteolytic activity, which may be naturally autocatalytic. Further analysis, illustrated by the schemes shown here in Figure 5, indicates that the enzyme contains a putative hydrophobic transmembrane region near its C-terminus. As described below, a further discovery of the present invention is that the enzyme can be truncated prior to this transmembrane region and still retain the β-secretase activity. 1. Purification of β-secretase from Natural and Recombinant Sources According to an important feature of the present invention, the β-secretase has now been purified from natural and recombinant sources. The U.S. Patent 5,744,346, incorporated herein by reference, describes the isolation of β-secretase in a single peak having an apparent molecular weight of 260-300,000 (Daltons) by gel exclusion chromatography. It is a discovery of the present invention that the native enzyme can be purified to apparent homogeneity by column affinity chromatography. The methods disclosed herein have been used on preparations of brain tissue as well as preparations of 293T and recombinant cells; according to this, it is believed that these methods are generally applicable on a variety of tissue sources. The practitioner will realize that certain stages of preparation, particularly the initial stages, may require modification to suit a particular tissue source and will adapt such procedures according to methods known in the art. Methods for purifying β-secretase from human brain as well as from cells are detailed in Example 5. Briefly, cell membranes or brain tissue are homogenized, fractionated, and subjected to various types of column chromatographic matrices , which include agglutinin-wheat germ agarose (wga), anion exchange chromatography and dimension exclusion. The activity of the fractions can be measured using any analysis appropriate for the β-secretase activity, such as the MBP-C125 cleavage analysis detailed in Example 4. The fractions containing β-secretase activity elute from this column in a volume of peak elusion corresponding to a dimension of approximately 260-300 kilodaltons. The above purification scheme, which produces approximately a 1,500 fold purification, is similar to that described in detail in the US patent. 5,744,346, incorporated herein by reference. In accordance with the present invention, further purification can be achieved by applying the cation exchange through-flow material to an affinity column employing as an affinity matrix a specific inhibitor of β-secretase, called "P10-P4 'staD- > V "(NH2-KTEEISEVN [sta] VAEF-C02H; SEQ ID NO: 72). This inhibitor and the methods for producing an affinity column of Sepharose incorporating it, are described in Example 7. After washing the column, the β-secretase and a limited number of contaminating proteins were eluted with borate buffer pH 9.5 . The eluate was then fractionated by anion exchange HPLC, using a Mini-Q column. Fractions containing the activity peak were pooled to give the final β-secretase preparation. The results of an exemplary process using this purification scheme are summarized in Table 1. Figure 6A shows an image of a SDS PAGE gel process of silver staining under reduced conditions, in which the β-secretase is processed as a 70 kilodalton band. The same fractions produced under non-reductive conditions (Figure 6B) provide evidence for cross-disulfide oligomers. When the fractions of the anion exchange group 18-21 (see Figure 6B) were treated with dithiothreitol (DTT) and rechromatographed on a Mini-Q column, then subjected to SDS-PAGE under non-reductive conditions, an execution of single band at approximately 70 kilodaltons. Surprisingly, the purity of this preparation is at least about 200 times greater than the previously purified material described in US Pat. 5,744,346. By way of comparison, the purest fraction described there exhibited a specific activity of approximately 253 nM / h / μg of protein, taking into consideration the MW of the substrate MBP-C26sw (45 kilodaltons). Accordingly, the present method provides a preparation that is at least about 1000 times greater in purity (affinity elution) and as high as approximately 6000 times greater in purity than that preparation, which represented at least 5 to 100 times greater in purity that the enzyme is present in a solubilized but unenriched membrane fraction of human 293 cells. Table 1 Preparation of secretase-ß from Human Brain a Activity display in MBP-C125 sw b Specific activity = (Product Conc. nM) (Dilution factor) (Sun volume on top) (incubation time h) (Conc. above μg / vol ) Example 5 also describes purification schemes used to purify recombinant materials from heterologous cells transfected with the β-secretase coding sequence. The results of these purifications are illustrated in Figures 7 and 8. Further experiments carried out in support of the present invention showed that the recombinant material has an apparent molecular weight in the range of 260,000 to 300,000 daltons when measured by exclusion chromatography. of gel. Figure 4 shows an activity profile of this preparation run on a gel exclusion chromatography column, such as a column Superdex 200 (26/60), according to the methods described in the US patent. 5,744,346, incorporated herein by reference. 1. Sequence of the β-secretase Protein Figure 9 shows a schematic view summarizing the methods and results for determining the sequence of the cDNA encoding the N-terminal peptide sequence determined from the purified β-secretase. The N-terminal sequence of purified β-secretase protein isolated from natural sources produced a 21 residue peptide sequence, SEQ ID NO: 77, as described above. This peptide sequence, and its inverse fully degenerate translated nucleotide sequence, SEQ ID NO: 76, is shown in the upper portion of Figure 4. Two sets of partially degenerate primers used for the RT-PCR amplification of a fragment of CDNA encoding this peptide in Figure 4. Primer set 1 consists of DNA nucleotide primers # 3427-3434 shown in Table 3 (Example 3). The RT-PCR matrix that uses primer combinations from this set with reverse transcribed cDNA from primary human neuronal cultures as models produced the predicted 54 bp cDNA product with primers # 3428-3433, also described in Table 3. In further experiments carried out in support of the present invention, it was found that the oligonucleotides of primer sets 1 and 2 can also be used to amplify cDNA fragments of predicted size from mouse brain mRNA. The DNA sequence demonstrated that such primers can also be used to clone murine homologs and other homologous species of human β-secretase and / or additional members of the aspartyl protease family described herein by standard RACE-PCR technology. The sequence of a murine homolog is presented in Figure 10 (bottom sequence; "pBS / MuImPain H # 3 cons"); SEC ID NO: 65 The murine polypeptide sequence is approximately 95% identical to the human polypeptide sequence. 2. 5 'and 3'RACE-PCR for Sequence, Cloning, and Additional Analysis of mRNA The unambiguous internal nucleotide sequence from the amplified fragment provided information that facilitated the design of internal primers that are matched with the upper strand (coding) for 3 'RACE, and with the lower strand (uncoded) for 5' RACE (Frohman, MA, MK Dush and GR Martin (1988). "Rapid production of full-length cDNA from rare transcripts: amplification using a specific oligonucleotide primer single gene "PROC. Nati. Acad. Sci. USA 85 (23): 8998-9002.). The DNA primers used for this experiment (# 3459 and # 3460) are illustrated schematically in Figure 9, and the exact sequence of these primers is presented in Table 4 of Example 3. Initiators # 3459 and # 3476 (Table 5 ) were used for the initial 3 'RACE amplification of downstream sequences from the IMR-32 cDNA file in the pLPCXlox vector. The file was previously subdivided into 100 groups of 5,000 per group, and the plasmid DNA was isolated from each group. An account of the 100 groups with the initiators described in Part 2 above identified individual groups containing ß-secretase deposits from the file. Such analyzes can be used for the RACE-PCR analysis.

A PCR fragment of approximately 1.8 Kb was observed by fractionation of agarose gel from the reaction products. The PCR product was purified from the gel and subjected to DNA sequence analysis using primer # 3459 (Table 5). The sequence of the resulting clone, designated as 23A, was determined. Six of the first seven amino acids deduced from one of the 23A reading structures were an exact match with the last 7 amino acids of the N-terminal sequence (SEQ ID NO: 77) determined from the purified protein isolated from sources natural in other experiments carried out in support of this invention. This observation provided internal validation of the sequences, and defined the appropriate reading structure downstream. In addition, this DNA sequence facilitated the design of additional primers to further extend the downstream sequence, verifying by sequencing the opposite strand in the upstream direction, and further facilitating the isolation of the cDNA clone. A human ß-secretase DNA sequence is illustrated as SEQ ID NO: 42 corresponding to SEQ ID NO: 1 including the 5 'and 3' untranslated regions. This sequence was determined from the partial cDNA clone (9C7e.35) isolated from a commercially available human fetal brain cDNA file purchased from OriGene ™, the 23A product from 3 'RACE, and additional ctions - a total of 12 cDNAs independent to determine the sequence of the compound. The compound sequence was installed by stretches of DNA sequence overlap from both strands of the clone or PCR fragment. The predicted total length translation product is shown as SEQ ID NO: 2 in Figure IB. 4. Distribution of ß-secretase tissue and Related Transcripts Oligonucleotide primer # 3460 (SEQ ID NO: 39, Table 5) was used as an extremely labeled test in Northern blots to determine the size of the transcript encoding β-secretase and to examine its expression in IMR-32 cells. Additional primers were used to isolate the mouse cDNA and to characterize mouse tissues, using prepared cDNA preparations Marathon RACE (Clontech, Palo Alto, CA). Table 2 summarizes the results of experiments in which various human and murine tissues were tested for the presence of transcripts encoding β-secretase by PCR or Northern bloting. For example, the oligonucleotide test 3460 (SEQ ID NO: 39) hybridized to a 2 Kb transcript in IMR-32 cells, indicating that the mRNA that encodes the β-secretase enzyme has 2 Kb in size in this tissue. Northern blot analysis of total RNA isolated from human Jurkat T cell line, and human myelomonocyte line Thp 1 with test 3460 of oligonucleotide 3460, also revealed the presence of a 2 kb transcript in these cells. The oligonucleotide test # 3460 also hybridizes to a transcription of ~ 2 kb in Northern blots containing RNA from all human organs examined to date, from adult and fetal tissue. The organs inspected include heart, brain, liver, pancreas, placenta, lung, muscle, uterus, bladder, kidney, spleen, skin, and small intestine. In addition, certain tissues, eg, pancreas, liver, brain, muscle, uterus, bladder, kidney, spleen and lung, show expression of transcripts greater than 4.5 kb, 5 kb, and 6.5 kb that hybridize with the oligonucleotide test # 3460 . In subsequent experiments carried out in support of the present invention, Northern blot results were obtained with the oligonucleotide test # 3460 using a riboprueb derived from SEQ ID NO: 1, comprising nucleotides # 155-1014. This clone provides a 860 bp riboprobe, comprising the catalytic portion encoding the β-secretase domain, for high-rigidity hybridization. This test hybridized with high specificity to the exact equivalent of the mRNA expressed in the samples that were examined. Northern blots of the mRNA were isolated from IMR-32 and the 1st HNC tested with this riboprueba revealed the presence of the 2 kb transcript previously detected with oligonucleotide # 3460, as well as a new transcription MW greater than -5 kb . Hybridization of adult and fetal human tissue RNA with this 860 nt riboprueba also confirmed the results obtained with oligonucleotide test # 3460. The mRNA encoding the β-secretase was expressed in all the tissues examined, predominantly as a ~ 5 kb transcript. In adults, its expression appeared lowest in the brain, placenta, and lung, intermediate in the uterus, and vejiga, and highest in the heart, liver, pancreas, muscle, kidney, spleen, and lung. In fetal tissue, the message is expressed uniformly in all the tissues examined. Table 2 Distribution of Human Tissue and transcripts of murine β-secretase Size of the Brain Region human message Clontech found: (Kb) Tissue / Human Organ Weaved Mouse / Human Organ Heart 3.5, 3.8, 5 and 7 Cerebellum 2Kb, 4Kb, 6Kb Brain 2, 3, 4, and 7 3.5, 3.8, 5 and 7 Cerebral Cortex 2Kb, 4Kb, 6Kb Liver 2, 3, 4, and 7 3.5, 3.8, 5 and 7 Medulla 2Kb, 4Kb, 6Kb Pancreas 2, 3, 4, and 7 ndd Spinal Cord 2Kb, 4Kb, 6Kb Placenta 2a, 4, and 7b nd Occipital Pole 2Kb, 4Kb, 6Kb Lung 2a, 4, and 7 ° 3.5, 3.8, 5 and 7 Frontal Lobe 2Kb , 4Kb, 6Kb Muscle 2a and 7b 3.5, 3.8, 5 and 7 Amygdala 2Kb, 4Kb, 6Kb Uterus 2a, 4, and 7 nd Caudate Nucleus 2Kb, 4Kb, 6Kb Bladder 2a, 3, 4, and 7 nd Callous Body 2Kb, 4Kb, 6Kb Kidney 2a, 3, 4, and 7 3.5, 3.8, 5 and 7 Hippocampus 2Kb, 4Kb, 6Kb Spleen 2a, 3, 4, and 7 nd Testicle nd 4.5Kb, 2kb Substance black 2Kb, 4Kb, 6Kb Stomach nd 5a Thalamus 2Kb, 4Kb, 6Kb Small bowel nd 3.5, 3.8, 5 and 7 Cerebro0 f 2a, 3, 4, and 7 nd Liver f 2a, 3, 4, and 7 nd Lung f 2a, 3, 4, and 7 nd Muscle f 2a, 3, 4, and 7 nd Heart f 2a, 3, 4, and 7 nd Kidney f 2a, 3, 4, and 7 nd Skin f 2a, 3, 4, and 7 nd Small intestine f 2a, 3, 4, and 7 nd Human Cell Line Mouse IMR32 2a 5 and 7 U937 2a THP1 2a Jurkat 2a HL60 none A293 5 and 7 NALM6 5 and 7 A549 5 and 7 Hela 2, 4, 5, and 7 PC12 2 and 5 J774 5Kb, 2Kb P388D1 cc146 5 Kb (very small) 2Kb P19 5Kb, 2Kb RBL 5Kb, 2Kb EL4 5Kb, 2Kb 3 for oligos 3460 only from test ° 1 = fetal b weak dnd = not determined . Active Forms of ß-secretase a. N-terminal The open reading structure of full length (ORF) of human ß-secretase was described in the above and its sequence is shown in Figure 2A as SEQ ID NO: 2.

However, as mentioned above, a later discovery of the present invention indicates that the predominant form of the naturally occurring active molecule is truncated at the N-terminus by approximately 45 amino acids. That is, the protein purified from natural sources was sequenced at the N-terminus according to methods known in the art (Argo Bioanalytica, Morris Plains, NJ). The N-terminus produced the following sequence: ETDEEPEEPGRRGSFVEMVDNLRG ... (SEQ ID NO: 55). This corresponds to amino acids 46-69 of the putative sequence derived from ORF. Based on this observation and others described below, the N-terminus of a predominantly active form of human brain that occurs naturally from the enzyme is amino acid 46, with respect to SEQ ID NO: 2. The subsequent processing of the purified protein provided the sequence of an internal peptide: ISFAVSACHVHDEFR (SEQ ID NO: 56), which is an amino terminus of the putative transmembrane domain, as defined by the ORF. These peptides were used to validate and provide the reading structure information for the isolates described elsewhere in this application. In further studies carried out in support of the present invention, the N-terminal sequence of the β-secretase isolated from additional cell types was revealed to be N-terminal amino acids numbers 46, 22, 58, or 63 with respect to the ORF sequence shown in Figure 2A, depending on the tissue from which the protein is isolated, having the form as its 46 N-terminal amino acid predominating in the tested tissues. That is, in the experiments carried out in support of the present invention, the full-length β-secretase construct (ie, encoding SEQ ID NO: 2) was transfected into the 293T cells and COS A2 cells, using the Fugene technique described in Example 6. The β-secretase was isolated from the cells by preparing a crude particulate fraction of the cell pellet, as described in Example 5, followed by extraction with Triton X-100 at 0.2 % containing regulator. The Triton extract was diluted with buffer at pH 5.0 and passed through a SP Sepharose column, essentially according to the methods described in Example 5A. This stage removed most of the proteins contained. After adjusting the pH to 4.5, the β-secretase was further purified and concentrated in P10-P4'staD- >V Sepharose, as described in Examples 5 and 7. Fractions were analyzed for the N-terminal sequence, according to standard methods known in the art. The results are summarized in Table 3, below. The N-terminal main sequence of the protein derived from the 293T cell was the same as that obtained from the brain. In addition, minor amounts of protein were also observed starting just after the signal sequence (in Tre-22) and the start of the homology domain of aspartyl protease Met-63). An additional principal form found in Cos A2 cells resulted from the cleavage of Gli-58. Table 3 N-terminal Sequences and Quantities of the β-secretase Forms in Various Types of Cells b. C-terminal Subsequent experiments carried out in support of the present invention revealed that the C-terminal amino acid sequence of full length presented as SEQ ID N0: 2, can also be truncated, while still retaining the β-secretase activity of the molecule. More specifically, as described in more detail below in Part D, the truncated C-terminal forms of the enzyme that terminate just before the putative transmembrane region, i.e., at or about 10 C-terminal amino acids toward amino acid 452 with respect to SEQ ID NO: 2, they exhibited β-secretase activity, as evidenced by the ability to unfold APP to the appropriate cleavage site and / or the ability to bind SEQ ID NO: 72. In this way, using the reference amino acid positions provided by SEQ ID NO: 2, a form of the β-secretase extends from position 46 to position 501 (β-secretase 46-501, SEQ ID NO: 43). Another form extends from position 46 to any position including and beyond position 452, (β-secretase 4-452 +), with a preferred form being β-secretase 46-452 (SEQ ID NO: 58) . More generally, another preferred form extends from position 1 to any position including and beyond position 452, but not including position 501. Other active forms of the β-secretase protein start at amino acid 22, 58, or 63 and may extend to any point including and beyond the cysteine at position 420, and more preferably, including and beyond position 452, while still retaining the enzymatic activity (ie, β-secretase 22-452 +; β-secretase 58-452 +; β-secretase 63-452 +). As described below in Part D, those forms truncated at a C-terminal position at or before position 452, or even several amino acids later, are particularly useful in crystallization studies, since they lack all or a significant portion of the transmembrane region, which can interfere with the crystallization of the protein. The recombinant protein extending from position 1 to 452 has been purified by affinity using the methods described herein. C. Crystallization of β-secretase According to a further aspect, the present invention also includes purified β-secretase in crystallized form, in the absence or presence of binding substrates, such as a peptide, a modified peptide, or small inhibitors. molecules This section describes methods and utilities of such compositions. 1. Crystallization of the Protein The purified β-secretase as described above can be used as a starting material to determine a crystallographic and coordinate structure for the enzyme. Such structural determinations are particularly useful for defining the conformation and dimension of the binding site of the substrate. This information can be used in the design and modeling of the substrate inhibitors of the enzyme. As discussed herein, such inhibitors are candidate molecules for therapeutics for the treatment of Alzheimer's disease and other amyloid diseases characterized by amyloid deposits of Aβ peptide. The crystallographic structure of the β-secretase is determined first by crystallizing the purified protein. Methods for crystallizing proteins, and in particular proteases, are now well known in the art. The practitioner refers to Principies of Protein X-ray Crystallography (J. Drenth, Springer Verlag, NY, 1999) for the general principles of crystallography. Additionally, the equipment for generating protein crystals are generally available from commercial suppliers, such as Hampton Research (Laguna Niguel, CA). Additional advice can be obtained from numerous research articles that have been written in the area of crystallography of protease inhibitors., especially with respect to the HIV-1 and HIV-2 proteases, which are proteases of aspartic acid. Although any of the various forms of β-secretase described herein can be used for crystallization studies, particularly preferred forms lack the first 45 amino acids of the full length sequence shown as SEQ ID NO: 2, since this appears to be the predominant form that occurs naturally in the human brain. It is thought that some form of post translational modification, possibly autocatalysis, serves to remove the first 45 amino acids in a favorably fast order, since, to date, virtually no naturally occurring enzyme has been isolated with all the first 45 amino acids intact. . In addition, it is considered preferable to remove the transmembrane putative region from the molecule prior to crystallization, since this region is not necessary for catalysis and could potentially make the molecule more difficult to crystallize.

Thus, a good candidate for crystallization is β-secretase 46-452 (SEQ ID NO: 58), since this is a form of the enzyme that (a) provides the predominant N-terminus that occurs naturally, and (b) it lacks the "sticky" transmembrane region, while (c) retains β-secretase activity. Alternatively, enzyme forms that have extensions that stretch part of the way can also be used. (approximately 10-15 amino acids) within the transmembrane domain. In general, any form of the enzyme that either (i) exhibits β-secretase activity, and / or (ii) binds to a known inhibitor can be used to determine the X-ray crystallographic coordinates of the ligand binding site. , such as the inhibitor ligand P10-P4 'staD- > V, with a binding affinity that is at least 1/1000 of the binding affinity of β-secretase [46-501] (SEQ ID NO: 43) to P10-P4 'staD- > V. Accordingly, a number of additional truncated forms of the enzyme can be used in these studies. The suitability of any particular form can be achieved by contacting it with the affinity matrix P10-P4'staD- > V described above. The truncated forms of the enzyme that bind to the matrix are suitable for such subsequent analyzes. Thus, in addition to 46-452, discussed above, experiments in support of the present invention revealed that a truncated form terminating at residue 419, most likely 46-419, also binds to the affinity matrix and is Consequently, an alternative candidate protein composition for the X-ray crystallographic analysis of β-secretase. More generally, any form of the enzyme that ends before the transmembrane domain, particularly those that end between approximately residue 419 and 452 are suitable in this regard. At the N-terminus, as described above, generally the first 45 amino acids will be removed during cellular processing. Other suitable forms that are presented or expressed naturally are listed in Table 3 above. These include, for example, a protein that starts at a residue 22, one that starts at residue 58 and one that starts at residue 63. However, the analysis of the whole enzyme, which starts at residue 1, can also provide information about the enzyme. Other forms, such as 1-420 (SEQ ID NO: 60) to 1-452 (SEQ ID NO: 59), including intermediate forms, for example, 1-440, may be useful in this regard. In general, it will also be useful to obtain the structure on any subdomain of the active enzyme. Methods for purifying the protein, including the active forms, were described above. In addition, since the protein is apparently glycosylated in its naturally occurring forms (and recombinants expressed in mammals), it may be desirable to express the protein and purify it from bacterial sources, which do not glycosize mammalian proteins, or express it in sources , such as insect cells, which provide uniform patterns of glycosylation, to obtain a homogeneous composition. Appropriate vectors and codon optimization procedures to complete this are known in the art. Following expression and purification, the protein is adjusted to a concentration of approximately 1-20 mg / ml. According to the methods that have worked for other crystallized proteins, the regulator and salt concentrations present in the initial protein solution are reduced to as low a level as possible. This can be achieved by dialyzing the sample against a starter regulator using microdialysis techniques known in the art. Regulators and crystallization conditions will vary from protein to protein, and possibly from fragment to fragment of the active β-secretase molecule, but can be determined empirically using, for example, matrix methods to determine optimal crystallization conditions. (Drentz, J., supra; Ducruix, A., et al. , edic. Crys tall izat ion of Nucleic Acids and Proteins: A Practical Approach, Oxford University Press, New York, 1992). After the dialysis, the conditions for the crystallization of the protein are optimized. Generally, methods for optimization may include producing a "reticulum" of 1 μl drops of the protein solution, mixed with 1 μl of the reservoir solution, which is a variable pH regulator and ionic strength. These drops are placed in individual sealed containers, typically in a "hanging drip" configuration, for example in commercially available containers (Hampton Research, Laguna Niguel, CA). The precipitation / crystallization generally takes place between 2 days and 2 weeks. The deposits are checked for evidence of precipitation or crystallization, and the conditions for forming crystals are optimized. The optimized crystals are not judged by dimension or morphology, but by the diffraction quality of the crystals, which must provide more than 3 Á of resolution.

Typical precipitating agents include ammonium sulfate (NH4SO4), polyethylene glycol (PEG) and methyl pentane diol (MPD). All chemicals used should be of the highest possible degree (e.g., ACS) and can also be re-purified by standard methods known in the art, prior to use. The regulators and exemplary precipitants that form an empirical network to determine crystallization conditions are commercially available. For example,. the "Crystal Screen" (Hampton Research) team provides a scattered matrix method of test conditions that is derived and selected from known crystallization conditions for macromolecules. This provides a "reticle" to quickly test a wide range of pH, salts, and precipitants using a very small sample (50 to 100 microliters) of macromolecule. In such studies, a solution of 1 μl of regulator / precipitant (s) is added to an equal volume of dialyzed protein solution, and the mixtures are allowed to stand for at least two days to two weeks, with careful monitoring of crystallization. Chemicals can be obtained from common commercial supplies; however, it is preferable to use adequate purity grades for crystallization studies, such as those provided by Hampton Research (Laguna Niguel, CA). Common regulators include Citrate, TEA, CHES, Acetate, ADA and the like (to provide an optimum pH range), typically at a concentration of approximately 100 mM. Typical precipitants include (NH4) 2S04, MgSO4, NaCl, MPD, Ethanol, polyethylene glycol of various dimensions, Isopropanol, KCl; and the similar (Ducruix). Various additives can be used to help improve the character of the crystals, including substrate analogs, ligands, or inhibitors, as discussed below in Part 2, as well as certain additives, including, but not limited to: 5% Jeffamine 5 % Polypropylene glycol P400 5% Polyethylene glycol 400 5% Ethylene glycol 5% 2-methyl-2,4-pentanediol 5% Glycerol 5% Dioxane 5% Dimethyl sulfoxide 5% N-Octanol 100 M (NH4) 2S04 100 mM 100 mM CsCl 100 mM CoS04 100 mM MnCl2 100 mM 100 mM ZnSO4 100 mM LIC12 100 mM 100 mM MgCl2 100 mM 100 mM Glucose 1, 6-Hexanodiol 100 mM Dextran Sulfate 100mM 100mM 6-amino caproic acid 100mM 1.6 hexane diamine 100mM 1,8 diamino octane 100mM 100mM Spermidine 0.17mM n-dodecyl-β-D-maltoside NP 40 20mM n-octyl-β-D- glucopyranoside According to a discovery of the present invention, the full-length β-secretase enzyme contains at least one transmembrane domain, and its purification is aided by the use of a detergent (Triton X-100). The membrane proteins may crystallize intact, but may require specialized conditions, such as the addition of a non-ionic detergent, such as C8G (8-alkyl-β-glucoside) or an n-alkyl-maltoside (CpM). The detergent selection is somewhat empirical, but certain detergents are commonly used. A number of membrane proteins have been "salted" successfully by adding high concentrations of salt to the mixture. PEG has also been successfully used to precipitate a number of membrane proteins (Ducruix, et al., Supra). Alternatively, as discussed above, a truncated C-terminal form of the protein binding the inhibitor is crystallized but lacks the transmembrane domain, such as β-secretase 46-452. After crystallization the conditions are determined, the crystallization of a greater amount of protein can be achieved by methods known in the art, such as vapor diffusion or equilibrium dialysis. In vapor diffusion, a drop of the protein solution is balanced against a larger reservoir of solution containing precipitant or other dehydrating agent. After sealing the solution is balanced to achieve supersaturation concentrations of proteins and thereby induce crystallization in the drop. The equilibrium dialysis can be used for the crystallization of proteins at low ionic strength. Under these conditions, a phenomenon known as "salting" occurs, through which protein molecules achieve the balance of electrostatic charges through interactions with other protein molecules. This method is particularly effective when the solubility of the protein is low at the lowest ionic strength. Various apparatuses and methods are used, including microdiffusion cells in which a dialysis membrane is attached to the bottom of a capillary tube, which can be curved in its lower portion. The final crystallization condition is achieved by slowly changing the composition of the external solution. A variation of these methods uses a dialysis setting of concentration gradient equilibrium. Microdiffusion cells are available from commercial providers such as Hampton Research (Laguna Niguel, CA). Once crystallization is achieved, the crystals are characterized by purity (e.g., SDS-PAGE) and biological activity. The largest crystals (>0.2 mm) are preferred to increase the resolution of X-ray diffraction, which is preferred in the order of 10-1.5 Angstroms. The selected crystals are subjected to X-ray diffraction, using a strong monochromatic X-ray source, such as a Synchrotron source or a rotating anode generator, and the resulting X-ray diffraction patterns are analyzed, using methods known in the art. technique. In one application, the amino acid sequence of β-secretase and / or the X-ray diffraction data are recorded on a computer-readable medium, which means any means that can be read and accessed directly by a computer. These data can be used to model the enzyme, a subdomain thereof, or a ligand thereof. The computer algorithms useful for this application are publicly and commercially available. 2. Crystallization of more Inhibitor Protein As mentioned above, it is advantageous to co-crystallize the protein in the presence of a binding ligand, such as an inhibitor. Generally, the process of optimizing the crystallization of the protein is followed with the addition of a concentration greater than 1 mM of the inhibitory ligand during the precipitation phase. These crystals are also compared to crystals formed in the absence of the ligand, so that measurements of the ligand binding site can be made. Alternatively, 1-2 μl of 1-25 mM of an inhibitor compound is added to the drop containing crystals developed in the absence of the inhibitor in a process known as "impregnation". Based on the coordinates of the binding site, additional optimization of the inhibitor is achieved. Such methods have been used advantageously to find new, more potent inhibitors for HIV proteases (See, eg, Viswanadhan, VN et al., J. Med. Chem. 39: 705-712, 1996; Muegge, I. et al., J Med. Chem. 42: 791-804, 1999). An inhibitory ligand used in these co-crystallization and impregnation experiments is P10-P4'staD- > V (SEQ ID NO: 72), a statin peptide inhibitor described above. The methods for producing the molecule are described herein. The inhibitor is mixed with βsecretase and the mixture is subjected to the same optimization tests described above, concentrating on the conditions set for the enzyme alone. The coordinates are determined and comparisons made between the free ligand and the bound enzyme, according to methods well known in the art. Additional comparisons can be made by comparing the inhibitory concentrations of the enzyme with the coordinates, as described by Viswanadhan, et al. , supra. The analysis of these comparisons provides the guide for the design of additional inhibitors, using this method.

D. Biological activity of β-secretase 1. ß-secretase occurring naturally In studies carried out in support of the present invention, isolated, purified forms of β-secretase were tested for enzymatic activity using one or more native substrates or synthetic. For example, as discussed above, when β-secretase was prepared from human brain and purified to homogeneity using the methods described in Example 5A, a single band was observed by silver stain after electrophoresis of fractions of sample of the anion exchange chromatography (last step) on an SDS-polyacrylamide gel under reducing conditions (+ β-mercaptoethanol). As summarized in Table 1, above, this fraction produced a specific activity of approximately 1.5 x 109 nM / h / mg protein, where the activity was measured by hydrolysis of MBP-C125S. 2. Recombinant Isolated β-secretase Several recombinant forms of the enzyme were produced and purified from transfected cells. Since these cells were produced to over-produce the enzyme, it was found that the purification scheme described with respect to the naturally occurring forms of the enzyme (e.g., Example 5A) could be shortened with positive results. For example, as detailed in Example 6, 293T cells were transfected with clone 27 of pCEK (Figure 12 and Figure 13A-E) and Cos A2 cells were transfected with pCFßA2 using "FUGENE" 6 Transfection Reagent (Roche Molecular Biochemicals Research, Indianapolis, IN). The pCF vector was structured from the vector of origin pCDNA3, commercially available from Invitrogen, by inserting SEQ ID NO: 80 (Figure HA) between the HindIII and EcoRI sites. This sequence contains the major tripartite guiding sequence of the last promoter of adenovirus and hybrid junction created from the first exon and intron of the last major region of adenovirus and a synthetic region-generated IgG variable region splice acceptor. PCDNA3 was cut with the restriction endonucleases HindIII and EcoRI, then blunted by filling the ends with Klenow fragment of DNA polymerase I. The cut and blunted vector was gel purified, and ligated with the isolated fragment of pED.GI. The pED fragment was prepared by digesting it with PvuII and Smal, followed by gel purification of the resulting base-pair fragment 419, which was subsequently classified for orientation and confirmed by sequencing. To create the pCEK expression vector, the pCF expression cassette was transferred into the EBV expression vector pCEP4 (Invitrogen, Carlsbad, CA). PCEP4 was cut with BglII and Xbal, filled, and the larger 9.15 kb fragment containing pBR, hygromycin, and EBV sequences was ligated to the 1.9 kb NruI to the Xmnl fragment of pCF containing the expression cassette (CMV splice, TPL / MLP / Igg, flank region Sp6, SVpoliA, M13). PCFßA2 (clone A2) contains full-length β-secretase in the pCF vector. The pCF vector is replicated in COS and 293T cells. In each case, the cells were pelletized and a crude particulate fraction was prepared from the pellet. This fraction was extracted with a regulator containing 0.2% Triton X-100. The Triton extract was diluted with a pH 5.0 regulator and passed through a SP Sepharose column. After adjusting the pH to 4.5, the fractions containing β-secretase activity were concentrated, with some further purification on the P10-P4 '(statin) D->. V Sepharose, as described for the brain enzyme. The silver staining of the fractions revealed co-purified bands on the gel. Fractions corresponding to these bands were subjected to N-terminal amino acid determination. The results of these experiments revealed some heterogeneity of the β-secretase species within the fractions. These species represent various forms of the enzyme, for example, from 293T cells, the main N-terminus is the same found in the brain, where (with respect to SEQ ID NO: 2) amino acid 46 is found in the N-terminus. Minor amounts of protein initiated just after the signal sequence (at residue 23) and at the beginning of the aspartyl protease homology domain (Met-63) were also observed. A larger additional form of protein was found in Cos A2 cells, resulting from cleavage in Gli 58. These results are summarized in Table 3, above. 2. Comparison of Isolated β-secretase Presented Naturally with Recombinant β-secretase As described above, the naturally occurring β-secretase derived from the human brain as well as the recombinant forms of the enzyme exhibited activity to unfold APP, particularly as it is evidenced by the activity in the MBP-C125 analysis. In addition, the key peptide sequences from the naturally occurring form of the enzyme equal portions of the deduced sequence derived from the cloning of the enzyme. Additional confirmation can be taken that the two enzymes act identically from subsequent experiments in which several inhibitors were found to have very similar affinities for each enzyme, as estimated by a comparison of the IC50 values measured for each enzyme under conditions of similar analyzes. These inhibitors were discovered in accordance with a further aspect of the invention, which is described below. Significantly, the inhibitors produce almost identical IC 50 values and orders of potency range in enzyme preparations derived from brain and recombinants, when compared in the same analysis. In subsequent studies, comparisons were made between the full-length recombinant enzyme having a C-terminal sequence of indicator "FLp501" (SEQ ID NO: 2, + SEQ ID NO: 45) and a recombinant enzyme truncated at position 452"452Stop" (SEQ ID NO: 58 or SEQ ID NO: 59). Both enzymes exhibited activity to unfold β-secretase substrates such as MBP-C125, as described above. The C-terminal truncated form of the enzyme exhibited activity to unfold the substrate MBP-C125sw as well as the substrate P26-P4 ', with an order of similar range of potency for the various inhibitor drugs tested. In addition, the absolute IC50s were comparable for the two enzymes tested with the same inhibitor. All IC50s were less than 10 μM. 1. Cellular β-secretase Subsequent experiments carried out in support of the present invention revealed that the isolated β-secretase polynucleotide sequences described herein encode β-secretase or β-secretase fragments that are active in the cells. This section describes experiments carried out in support of the present invention, the cells were transfected with β-secretase alone encoding DNA, or co-transfected with secretase encoding DNA and wild type APP encoding DNA as detailed in Example 8. a. Transfection with β-secretase In experiments carried out in support of the present invention, transcripts containing genes expressing the full-length polypeptide (SEQ ID NO: 2) were transfected into COS cells (Fugene and Effectene methods). Whole cell lysates were prepared and various amounts of lysate were tested for β-secretase activity according to standard methods known in the art or described herein in Example 4. Figure 14B shows the results of these experiments. As shown, lysates prepared from transfected cells, but not from mock or control cells, exhibited considerable enzymatic activity in MPB-C125sw analysis, indicating "overexpression" of β-secretase by these cells. b. Co-transfection of cells with β-secretase and APP In subsequent experiments, 293T cells were co-transfected with clone 27 pCEK, Figures 12 and 13 or the poCK vector containing the full-length β-secretase molecule (1-501; SEQ ID NO: 2) and with a plasmid containing either the wild-type or Swedish APP construction pohCK751, as described in Example 8. The β-specific cleavage was analyzed by Western and ELISA analyzes to confirm that the correct site of unfolding. Briefly, 293T cells were co-transfected with equivalent amounts of plasmids encoding ßAPPsw or wt and β-secretase or β-galactosidase (β-gal) control cDNAs according to standard methods. The ßAPP and ß-secretase cDNAs were delivered by the pohCK or pCEK vectors, which do not replicate in 293T cells (clone 27 of pCEK, Figures 12 and 13, pohCK751 expressing ßAPP 751, Figure 21). Conditioned medium and cell lysates were harvested 48 hours after transfection. Western analysis was carried out on the conditioned medium and cell lysates. ELISAs were carried out for the detection of Aβ peptide on the conditioned medium to analyze several cleavage APP products. Western Blot Results It is known that β-secretase is split specifically in Met-Asp in APPwt and Leu-Asp in APPsw to produce the Aβ peptide, starting in position 1 and releasing soluble APP (sAPPβ). Immunological reagents, specifically antibody 92 and 92sw (or 192sw), respectively, specifically detecting cleavage at this position in substrates APPwt and APPsw, have been developed, as described in US Pat. 5,721,130, incorporated herein by reference. Western blot analyzes were performed on gels in which cell lysates were separated. These analyzes were performed using methods well known in the art, using as Ab 92 or Ab92S primary antibody reagents, which are specific for the C-terminus of the N-terminal fragment of APP derived from APPwt and APPsw, respectively. In addition, ELISA format analyzes were performed using antibodies specific to the N-terminal amino acid of the C-terminal fragment. Monoclonal antibody 13G8 (specific for the C-terminal APP-epitope at positions 675-695 of APP695) was used in a Western blot format to determine whether transfected cells express APP. Figure 15A shows that reproducible transfection was obtained with expression levels of APP in vast excess over endogenous levels (triplicate deposits are indicated as 1,2,3 in Figure 15A). Three forms of APP - mature, immature and endogenous - can be seen in cells transfected with APPwt and APPsw. When ß-secretase was co-transfected with APP, small C-terminal fragments appeared in triplicate deposit lines from co-transfected cells (Western blot Figure 15A, mostly straight line group). In parallel experiments, where the cells were co-transfected with β-secretase and APPsw substrate, literally all mature APPs were unfolded (a group of mainly straight lines labeled "1,2,3" of Figure 15B). This suggests that there is an extensive cleavage by the mature APP β-secretase (upper band), which results in C-terminal fragments of an expected dimension in the lysate for cleavage at the β-secretase site. The co-transfection with the Swedish substrate also resulted in an increase in two CTF fragments of different dimension (indicated by a star). In conjunction with the additional Western and ELISA results described below, these results are consistent with a second splitting that occurs on the APPsw substrate after initial cleavage at the β-secretase site. The conditioned medium from the cells was analyzed for reactivity with the 192sw antibody, which is specific for B-s-APPsw. Analyzes using this antibody indicated a dramatic increase in soluble split APP of β-secretase. This is observed in the gel illustrated in Figure 16B by comparing the dark bands present in the "APPsw ßsec" sam with the bands present in the "APPsw ßgal" sam. The specific antibody for β-s-APPwt also indicates an increase in the cleaved β-secretase material, as illustrated in Figure 16A. Since the antibodies used in these experiments are specific for the cleaved β-secretase site, the previous results show that the β-secretase p501 unfolds the APP at this site, and the overexpression of this recombinant clone leads to a dramatic increase in the ß-secretase that is processed in the correct β-secretase site in whole cells. This process works on the wild type APP substrate and substantially increases on the Swedish APP substrate. Since approximately 20% of the APP secreted in 293T cells is β-sAPP, with the increase observed below for APPsw, it is likely that almost all the sAPP is β-sAPP. This observation was subsequently confirmed by independent Western analyzes in which alpha and total sAPP were measured. Monoclonal antibody 1736 is specific for the cleaved β-APP of exposed α-secretase (Selkoe, et al.). When Western blots were carried out using this antibody as the primary antibody, a slight but reproducible decrease in a-unfolded APPwt was observed (Figure 17A), and a dramatic decrease was observed in the a-unfolded APPsw material (note the almost absence of reactivity in Figure 17B in the lines labeled "APPsw ßsec." These results suggest that the recombinant overexpressed p501-secretase unfolds APPsw so efficiently or extensively that little or no substrate remains for the a-secretase to unfold. This further indicates that all the sAPP in the APPsw ßsec sam (illustrated in Figure 16B) are ß-sAPP.Aβ ELISA results The conditioned medium from the recombinant cells was harvested, diluted as necessary and tested for the production of the Aβ peptide by ELISA or microtiter plates coated with the monoclonal antibody 2G3, which is specific for the recognition of the C-terminus of Aß (l-40), with the biotinylated detector reagent mAb 3D6, which measures Aβ (x-40) (i.e., all truncated N-terminal forms of the Aβ peptide). Overexpression of β-secretase with APPwt results in an approximately 8-fold increase in Aβ production (x-40), with 1-40 representing a small percentage of the total. This was also a substantial increase in the production of Aßl-40 (Figure 18). With the APPsw there was an approximate increase of 2 times in Aß (x-40). Without adhering to any particular underlying theory, it is considered that the less dramatic increase in Aβ (x-40) ß-sec / APPsw cells compared to ß-sec / APPwt cells is due in part to the fact that the processing of the APPsw substrate is much more efficient than the APPwt substrate. That is, a significant amount of APPsw is processed by the endogenous β-secretase, thus further increases in the transfection of β-secretase are limited. These data indicate that the expression of the recombinant β-secretase increases the production of Aβ and that the β-seretase is the rate that limits the production of Aβ in the cells. This means that the activity of the β-secretase enzyme is the rate that limits the production of aβ in the cell, and therefore provides a good therapeutic goal. IV. Utility A. Vectors and Expression Cells Expressing the β-secretase The invention includes cloning and further expression of the members of the aspartyl protease family described above, for example, by inserting the polynucleotides encoding the proteins into the standard expression vectors and that they transfer appropriate host cells according to the standard methods previously treated. Such expression vectors and cells expressing, for example, the human β-secretase enzyme described herein, have utility, for example in production components (purified enzyme or transfecated cells) for the selection assay treated in Part B, later. Such a purified enzyme also has utility in providing starting materials for the crystallization of the enzyme, as described in section III, above. In particular, the truncated form (s) of the enzyme, such as 1-452 (SEQ ID NO: 59) and 46-452 (SEQ ID NO: 58), and the deglycosylated forms of the enzyme described herein are considered useful in this regard, as are other forms of truncated trajectory in the transmembrane region, for example 1-460 or 46-458. In accordance with the present invention, the sequences of polynucleotides encoding human β-secretase, splice variants, protein fragments, fusion proteins, or functional equivalents thereof, collectively referred to herein as "β-secretase" , can be used in recombinant DNA molecules that direct the expression of β-secretase in appropriate host cells. Due to the inherent degeneracy of the genetic code, other nucleic acid sequences that encode substantially the same or a functionally equivalent amino acid sequence can be used to clone and express β-secretase. Such variants will be easily verified by persons skilled in the art. The polynucleotide sequence of the present invention can be designed to alter a β-secretase encoding the sequence for a variety of reasons, including but not limited to, alterations that modify the cloning, process and / or expression of the gene product. For example, alterations can be introduced using techniques that are well known in the art, eg, site-directed mutagenesis, to insert new restriction sites, to alter glycosylation patterns to change the codon preference, to produce splice variants, etc. For example, it may be advantageous to produce nucleotide sequences encoding the β-secretase which possess codons that do not occur naturally. The codons preferred by a particular prokaryotic or eukaryotic host (Murray E. et al., (1989) Nuc Acids Res 17: 477-508) can be selected, for example, to increase the expression rate of the β-secretase polypeptide or to produce recombinant RNA transcripts that have desirable properties, such as a longer half-life, that the transcripts produced from the sequence occur naturally. This may be particularly useful for producing recombinant enzymes in non-mammalian cells, such as bacteria, yeast, or insect cells. The present invention also includes recombinant constructs comprising one or more of the sequences as described above. The constructs comprise a vector, such as a plasmid or viral vector, in which a sequence of the invention has been inserted, in a forward or inverse orientation. In a preferred aspect of this embodiment, the construct further comprises regulatory sequences, including, for example, a promoter, operably linked to the sequence. A large number of suitable vectors and promoters are known to those skilled in the art and are commercially available. The proper cloning and expression of vectors for use with prokaryotic and eukaryotic hosts is also described in Sambrook et al. ,. { supra). The present invention also relates to host cells that are genetically engineered with vectors of the invention and to the production of proteins and polypeptides of the invention by recombinant techniques. The host cells are genetically engineered (i.e. transduced, transformed or transfected) with the vector of this invention which can be, for example, a cloning vector or an expression vector. The vector can be, for example, in the form of a plasmid, a viral particle, a phage, etc. The engineered host cells can be cultured in conventional nutrient media modified as appropriate to activate the promoters, select the transformants or amplify the β-secretase gene. The culture conditions, such as temperature, pH and the like, are those previously used with the host cell selected for expression and will be apparent to those skilled in the art. Exemplary methods for the transfection of various cell types are provided in Example 6, herein. As described above, according to a preferred embodiment of the invention, the host cells can be co-transfected with an enzyme substrate, such as with the APP (such as the Swedish or wild-type mutation form), in order to measure the activity in a cellular environment. Such host cells are of particular utility in the screening assays of the present invention, particularly for the selection of therapeutic agents that are capable of traversing cell membranes. The polynucleotides of the present invention can be included in any of a variety of suitable expression vectors to express a polypeptide. Such vectors include chromosomal, non-chromosomal and synthetic DNA sequences, e.g., derived from SV40; bacterial plasmids, phage DNA; baculovirus; yeast plasmids; vectors derived from combinations of plasmids and phage DNA, viral DNA such as vaccinia, adenovirus, fowl pox virus and pseudorabies. However, any other vector can be used as long as it is replicable and viable in the host. The appropriate DNA sequence can be inserted into the vector by a variety of methods. In general, the DNA sequence is inserted into an appropriate restriction endonuclease site (s) by procedures known in the art. Such procedures and related subcloning procedures are considered within the scope of those skilled in the art. The DNA sequence in the expression vector is operably linked to an appropriate transcription control sequence (promoter) to direct mRNA synthesis. Examples of such promoters include: CMV, LTR or SV40 promoters, the lac or trp promoter of E. coli, the lambda PL phage promoter and other known promoters for controlling the expression of genes in prokaryotic or eukaryotic cells or their viruses. The expression vector also contains a ribosome binding site for the start of translation and a transcription terminator. The vector may also include appropriate sequences to amplify the expression. In addition, the expression vectors preferably contain one or more selectable marker genes to provide a phenotypic characteristic for the selection of transformed host cells such as resistance to dihydrofolate reductase or neomycin for eukaryotic cell cultures or such as resistance to tetracycline or ampicillin in E. coli. The vector containing the appropriate DNA sequence as described above, as well as an appropriate promoter or control sequence, can be employed to transform an appropriate host to allow the host to express the protein. Examples of appropriate expression hosts include: bacterial cells, such as E. coli, Streptomyces and Salmonella typhimurium; fungal cells such as yeast; insect cells such as Drosophila and Spodoptera Sf9; mammalian cells such as CHO, COS, BHK, HEK 293 or Bowes melanoma; adenovirus; plant cells, etc. It is understood that not all cells or cell lines will be capable of producing fully functional β-secretase; for example, it is likely that human β-secretase is highly glycosylated in natural form, and such glycosylation may be necessary for activity. In this case, eukaryotic host cells may be preferred. The selection of an appropriate host is considered to be within the scope of those skilled in the art from the teachings herein. The invention is not limited by the host cells used. In bacterial systems, several expression vectors may be selected depending on the proposed use for β-secretase. For example, when large amounts of β-secretase or fragments thereof are necessary for the induction of antibodies, vectors, which direct high levels of expression of the fusion proteins that are easily purified, may be desirable. Such vectors include, but are not limited to, cloning and expression vectors of multifunctional E. coli such as Bluescript (R) (Stratagene, La Jolla, CA), in which the coding sequence of β-secretase can be linked in the vector in structure with the sequence for the amino-terminal Met and the subsequent 7 residues of beta-galactosidase in order to produce a hybrid protein; pIN vectors (Van Heeke &Schuster (1989) J Biol Chem 264: 5503-5509); pET vectors (Novagen, Madison Wl); and the similar. Several vectors containing constitutive or inducible promoters such as alpha factor, alcohol oxidase and PGH can be used in the yeast Saccharomyces cerevisiae. For reviews, see Ausubel et al. (supra) and Grant et al. (1987); Methods in Enzymology 153: 516-544). In cases where plant expression vectors are used, the expression of a sequence encoding the β-secretase can be directed by any of several promoters. For example, viral promoters such as the 35S and 19S promoters of CaMV (Brisson et al (1984) Nature 310: 511-514) can be used alone or in combination with the omega leader sequence of TMV (Takamatsu et al. ) EMBO J 6: 307-311). Plant promoters such as the small RUBISCO subunit (Coruzzi et al (1984) EMBO J 3: 1671-1680; Broglie et al (1984) Science 224: 838-843) may alternatively be used; or heat shock promoters (Winter J and Sinibaldi RM (1991) Results Probl. Cell Differ. 17: 85-105). These constructs can be introduced into plant cells by direct DNA transformation or pathogen-mediated transfection. For the review of such techniques, see Hobbs S or Murry LE (1992) in McGraw Hill Yearbook of Science and Technology, McGraw Hill, New York, NY. , pp 191-196; or Weissbach and Weissbach (1988) Methods for Plant Molecular Biology, Academic Press, New York, NY., pp 421-463. The β-secretase can also be expressed in an insect system. In such a system, Autographa calif omica nuclear polyhedrosis virus (AcNPV) is used as a vector to express foreign genes in Sf9 cells of Spodoptera frugiperda or Trichoplusia larvae. The sequence encoding the β-secretase is cloned into a non-essential region of the virus, such as the polyhedrin gene and placed under the control of the polyhedrin promoter. Successful insertions of the sequence encoding Kv-SL will render the polyhedrin gene inactive and produce the protein layer of the layer lacking the recombinant virus. The recombinant viruses are then used to infect S. frugiperda cells or Trichoplusia larvae in which β-secretase is expressed (Smith et al (1983) J Virol 46: 584; Engelhard EK et al. (1994) Proc Nat Acad Sci 91: 3224-3227). In mammalian host cells, various virus-based expression systems can be used. In cases where an adenovirus is used as an expression vector, a sequence encoding the β-secretase can be ligated into an adenovirus transcription / translation complex consisting of the latter promoter and the tripartite leader sequence. Insertion into a non-essential E3 region of the viral genome will result in a viable virus capable of expressing the enzyme in the infected host cells (Logan and Shenk (1984) Proc Nati Acad Sci 81: 3655-3659). In addition, transcription enhancers, such as the sarcoma rous virus (RSV) enhancer (bird sarcoma), can be used to increase expression in mammalian host cells. Specific start signals may also be required for efficient translation of a sequence encoding β-secretase. Those signals include the ATG start codon and adjacent sequences. In cases where the sequence encoding the β-secretase, its start codon and the upstream sequences are inserted into the appropriate expression vector, additional translation control signals may not be necessary. However, in cases where only the coding sequence or a portion thereof is inserted, exogenous transcriptional control signals including the ATG start codon can be provided. In addition, the start codon must be the correct reading structure to ensure transcription of the complete insert. The exogenous transcriptional elements and the start codons can be of various origins, both natural and synthetic. The efficiency of expression can be improved by the inclusion of appropriate enhancers for the cellular system in use (Scharf D et al. (1994) Results Probl Cell Differ 20: 125-62; Bittner et al. (1987) Methods in Enzymol 153: 516-544). In a further embodiment, the present invention relates to host cells containing the constructions described above. The host cell can be a larger eukaryotic cell, such as a mammalian cell or a lower eukaryotic cell, such as a yeast cell or the host cell can be a prokaryotic cell, such as a bacterial cell. The introduction of the construction into the host cell can be effected by calcium phosphate transfection, DEAE-Dextran-mediated transfection or electroporation (Davis, L., Dibner, M., and Battey, I. (1986) Basic Methods in Molecular Biology) or newer methods, including lipid transfection with "FUNGENE" (Roche Molecular Biochemicls, Indianapolis, IN) or "EFECTENE" (Quiagen, Valencia, CA) or other DNA carrying molecules. Cell-free translation systems can also be used to produce polypeptides using RNAs derived from the DNA constructs of the present invention. The host cell strain can be selected for its ability to modulate the expression of the inserted sequences or to process the expressed protein in the desired form. Such modifications of the protein include, but are not limited to, acetylation, carboxylation, glycosylation, phosphorylation, lipidation and acylation. Post-translation processing that unfolds a "prepro" form of the protein may be important to correct insertion, folding and / or function. For example, in the case of β-secretase, it is likely that the N-terminus of SEQ ID NO: 2 will be truncated, in order for the protein to start at amino acid 22, 46 or 57-58 of SEQ ID NO: 2. Different host cells such as CHO, HeLa, BHK, MDCK, 293, WI38, etc. they have specific cellular machinery and characteristic mechanisms for such post-translational activities and can be selected to ensure correct modification and processing of the introduced foreign protein. The long-term high yield production of recombinant, stable expression proteins may be preferred. For example, cell lines stably expressing β-secretase can be transformed using expression vectors containing viral origins of replication or endogenous expression elements and a selectable marker gene. Following the introduction of the vector, the cells can be allowed to grow for 1-2 days in an enriched medium before being switched to a selective medium. The purpose of the selectable marker is to confer resistance to the selection and its presence allows the growth and recovery of the cells that successfully express the introduced sequence. Resistant clusters of stably transformed cells can be proliferated using tissue culture techniques appropriate for the cell type. For example, in experiments carried out in support of the present invention, over-expression of the "452stop" form of the enzyme has been achieved. Host cells transformed with a nucleotide sequence encoding the β-secretase can be cultured under conditions suitable for the expression and recovery of the encoded protein from the cell culture. The protein or fragment thereof produced by a recombinant cell can be secreted, bound to the membrane or contained intracellularly, depending on the sequence and / or the vector used. As will be understood by those skilled in the art, expression vectors containing the polynucleotides encoding β-secretase can be designed with signal sequences that direct the secretion of the β-secretase polynucleotide through a prokaryotic or eukaryotic cell membrane. . The β-secretase can also be expressed as a recombinant protein with one or more additional polypeptide domains added to facilitate the purification of the protein. Such domains that facilitate purification include, but are not limited to metal chelation peptides such as histidine-tryptophan modules that allow purification in immobilized metals, protein A domains that allow purification in immobilized immunoglobulins and the domain used in extension / affinity purification systems of FLAGS (Immunex Corp. Seattle, Wash.). The inclusion of a protease cleavable polypeptide linker sequence between the purification domain and the β-secretase is used to facilitate purification. One such expression vector provides expression of a fusion protein comprising β-secretase (e.g., a fragment of soluble β-secretase) fused to a polyhistidine region separated by an enterokinase cleavage site. Histidine residues facilitate purification in IMIAC (immobilized metal ion affinity chromatography, as described in Porath et al (1992) Protein Expression and Purification 3: 263-281) while the enterokinase cleavage site provides a means for isolate the β-secretase from the fusion protein. PGEX vectors (Promega, Madison, Wis.) Can also be used to express foreign polypeptides as fusion proteins with glutathione S-transferase (GST). In general, the fusion proteins are soluble and can be easily purified from lysed cells by adsorption to ligand-agarose beads (e.g., glutathione agarose in the case of GST fusions) followed by elution in the presence of the free ligand. Following transformation of a suitable host strain and growth of the host strain to an appropriate cell density, the selected promoter is induced by appropriate means (e.g., temperature change or chemical induction) and the cells are cultured for an additional period. The cells are typically collected by centrifugation, fractionated by physical or chemical means and the resulting crude extract is retained for further purification. The microbial cells employed in the expression of proteins can be fractionated by any convenient method including freeze / thaw cycles, sonication, mechanical fractionation or use of cell-using agents or other methods that are well known to those skilled in the art. The β-secretase can be recovered and purified from recombinant cell cultures by any of several methods well known in the art or preferably by the purification scheme described herein. Re-doubling steps of protein may be used, as necessary, upon completion of the mature protein configuration. Examples of methods for purifying naturally occurring forms of β-secretase as well as purified forms are provided in the Examples. B. Methods for Selecting β-secretase Inhibitors The present invention also includes methods for identifying molecules, such as synthetic drugs, antibodies, peptides or other molecules, which have an inhibitory effect on the β-secretase activity described herein. , generally referred to as inhibitors, antagonists or blockers of the enzyme. Such assay includes the steps of providing a human β-secretase such as the β-secretase comprising SEQ ID NO: 2, SEQ ID NO: 43 or more, particularly with reference to the present invention, an isolated protein, approximately 450 amino acid residues in length, which includes an amino acid sequence that is at least 90% identical to SEQ ID NO: 75 [63-423] including conservative substitutions thereof, which exhibits β-secretase activity, as described herein. The β-secretase enzyme is contacted with a test compound to determine if it has a modulating effect on the activity of the enzyme, as will be discussed below and is selected from the test compounds capable of modulating the activity of the β -secretase. In particular, the inhibitory compounds (antagonists) are useful in the treatment of disease conditions associated with amyloid deposition, particularly Alzheimer's disease. Those skilled in the art will understand that such tests can be conveniently transformed into equipment. Particularly useful screening assays employ cells that express both β-secretase and APP. Such cells can be made recombinantly by co-transfection of the cells with polynucleotides encoding the proteins, as described in Section III, above or they can be made by transfecting a cell which naturally contains one of the proteins with the second protein. In a particular embodiment, such cells grow in multi-well culture dishes and are exposed to various concentrations of a test compound or compounds for a predetermined period of time, which can be determined empirically. All cell lysates, culture media or cell membranes are tested for β-secretase activity. Test compounds that significantly inhibit activity compared to the control (as discussed above) are considered therapeutic candidates. The isolated β-secretase, its binding ligand, catalytic or immunogenic fragments or oligopeptides thereof, can be used to select the therapeutic compounds in any of a variety of drug selection techniques. The protein used in such a test can be bound to the membrane, free in solution, fixed to a solid support, supported on a cell surface or localized intracellularly. The formation of binding complexes between the β-secretase and the agent to be tested can be measured. Compounds that inhibit the binding between β-secretase and its substrates, such as APP or fragments of APP, can be detected in one assay. Preferably, the enzymatic activity will be monitored and the candidate compounds will be selected on the basis of their ability to inhibit such activity. More specifically, a test compound will be considered as an inhibitor of β-secretase if the activity of the β-secretase measured is significantly lower than the activity of the β-secretase measured in the absence of the test compound. In this context, the term "significantly lower" means that in the presence of the test compound the enzyme exhibits an enzymatic activity, which, when compared to the enzymatic activity measured in the absence of the test compound, is measurably lower, within the safety limits of the test method. Such measurements can be assessed by a change in the end point analysis of a single K ^ and / or Vmax assay or any other standard method in the art. In Example 4 of the present, exemplary methods for testing β-secretase are provided. For example, in studies carried out in support of the present invention, the compounds were selected based on their ability to inhibit β-secretase activity in the MBP-C125 assay. Compounds that inhibited the activity of the enzyme at a concentration less than about 50 μM were selected for further selection. The groups of compounds that are most likely candidates for inhibiting activity comprise a further aspect of the present invention. Based on the studies carried out in support of the invention, it has been determined that the peptide compound described here as P10-P4 'staD- »V (I KNOW THAT ID NO: 72) is a reasonably potent inhibitor of the enzyme. Additional studies based on this sequence and peptide mimetics of portions of this sequence have revealed several small molecule inhibitors. Random files of peptides or other compounds can also be classified by their suitability as inhibitors of β-secretase. Combination files can be produced for many types of compounds that can be synthesized step by step. Such compounds include polypeptides, mimetics that change to beta, polysaccharides, phospholipids, hormones, prostaglandins, steroids, aromatics, heterocyclic compounds, benzodiazepines, N-substituted oligomeric glycines and oligocarbamates. Large combinational files of the compounds can be constructed by synthetic encoded file (ESL) methods described in Affymax, WO 95/12608 Affymax, WO 93/06121, Columbia University, WO 94/08051, Pharmacopeia WO 95/35503 and Scripps, WO 95/30642 (each of which is incorporated for reference for all purposes). A preferred source of test compounds for use in screening for therapeutic or therapeutic guidelines is a file that shows a phage. See e.g. Devlin, WO 91/18980; Key, B.K. et al., eds., Phage Display of Peptides and Proteins, A Laboratory Manual, Academic Press, San Diego, CA, 1996. Phage display is a powerful technology that allows phage genetics to be used to select and amplify peptides or proteins from desired characteristics from files containing different sequences 108-109. The files can be designed for selected variegation of amino acid sequence in desired positions that allow the derivation of the file towards desired characteristics. The files are designed so that the peptides are expressed fused to proteins that are deployed on the surface of the bacteriophage. The phage displaying peptides of the desired characteristics is selected and can be re-developed for expansion. Since the peptides are amplified by the propagation of the phage, the selected phage DNA can be easily sequenced facilitating rapid analysis of the selected peptides. The phage encoding the peptide inhibitors can be selected by selecting the phage that binds specifically to the β-secretase protein. The files are generated fused to proteins such as gene II that is expressed on the surface of the phage. The files can be composed of peptides of various lengths, linear or constructed by the inclusion of two Cis amino acids, fused to the phage protein or they can also be fused to additional proteins like a goop. It can be started with files composed of random amino acids or with files that are derived from sequences in the ßAPP substrate that surrounds the cleavage site of the β-secretase or preferably, to the substituted site D- »V exemplified in SEQ ID NO: 72 Derivative files can also be designed towards the peptide inhibitors and substrates described herein or deviated to peptide sequences obtained from the selection of the binding phage from the initial files by providing additional test inhibitor compounds. The β-secretase is immobilized and the phage binds specifically to the selected β-secretase. Limitations, such as a requirement that the phage not bind in the presence of a known active site inhibitor of β-secretase (eg the inhibitor described herein), serve to further direct the active site-specific compounds of selection of the phage This can be complicated by a differential selection format. The highly purified β-secretase derived from the brain or preferably from recombinant cells, can be immobilized in 96-well plastic dishes using standard techniques (reference to the phage book). The recombinant β-secretase designed to fuse to a binding peptide (eg, strepavidin binding motifs, His, FLAG or myc markers) can also be used to another immobilized protein (such as streptavidin or appropriate antibodies) in the petri dishes of plastic. The phage is incubated with the bound β-secretase and the unbound phage is removed by washing. The phage is chosen and this selection is repeated until a phage-to-secretase binding population is recovered. The union and elusion are carried out using standard techniques. Alternatively, the β-secretase can "bind" when expressed in Cos or other mammalian cells that develop in petri dishes. In this case the phage that binds to the cells expressing the β-secretase would be selected, and would be selected against the phage that binds to the control cells, which do not express the β-secretase. Phage display technology can also be used to select the preferred substrates of the β-secretase and incorporate the identified characteristics of the preferred substrate peptides obtained by the phage display in the inhibitors. In the case of β-secretase, rection of the sequence of amino acids surrounding the cleavage site of the APP and the cleavage site of the APPsw has provided information for the purpose of establishing the phage display selection file for identify the preferred substrates of the β-secretase. As mentioned above the rection of the sequence of a particularly good peptide inhibitor, P10-P4staD- >; V, as described herein, provides information for establishing a "derived" file towards this sequence. For example, the peptide substrate file containing 108 different sequences is fused to a protein (such as a gene III protein) expressed on the surface of the phage and to a sequence that can be used to bind streptavidin, or another protein such as the His marker and the antibody for His. The phage is digested with protease, and the undigested phage is removed by binding to an appropriate immobilized binding protein, such as streptavidin. This selection is repeated until a population of phage encoding the substrate peptide sequence is recovered.

The DNA in the phage is sequenced to produce the substrate sequences. These sub-strata are then used for the further development of peptidomimetics, particularly peptidomimetics which have inhibitory properties. The combinational files and other compounds are initially selected for suitability by determining their ability to bind or preferentially, to inhibit the activity of the β-secretase in any of the assays described herein or otherwise known in the art. The compounds identified by such selections are then further analyzed for potency in such assays. The inhibitory compounds can then be tested for prophylactic and therapeutic efficacy in transgenic animals predisposed to an amyloidogenic disease, such as several rodents carrying a transgene containing human APP, e.g. mouse carrying a 717 mutation of the APP described by Games et al. , Nature 373: 523-527, 1995 and Wadsworth et al. , (US 5,811,633, US 5,604,131, US 5,720,936), and mouse carrying a Swedish APP mutation such as that described by McConlogue et al. , (US 5,612,486) and Hsiao et al. , (U.S. 5,877,399); Staufenbiel et al. , Proc. Nati Acad. Sci. USA 94, 13287-13292 (1997), Sturchler-Pierrat et al. , Proc. Nati Acad. Sci. USA 94,13287-13292 (1997); Borchelt et al. , Neuron 19,939-945 (1997), all of which are incorporated herein by reference. The compounds or agents found to be effective and safe in such animal models will be further tested in standard toxicological tests. Compounds that exhibit appropriate toxicology and pharmacokinetic profiles will move toward human clinical trials for the treatment of Alzheimer's disease and related diseases. The same selection approach can be used in other potential agents such as the peptidomimetics described above. In general, in therapeutic selection compounds based on the above assays, it is useful to determine whether the test compound has an acceptable toxicity profile, e.g. in a variety of cells in vi tro and animal model (s). It can also be useful to look for the The compound (s) tested and identified against databases of existing compounds to determine whether the compound or analogs thereof have been previously employed for pharmaceutical purposes, and if so, the optimal routes of administration. and dose ranges. Alternatively routes of administration and dose ranges can be determined empirically, using methods well known in the art (see, eg, Benet, LZ, et al., Pharmacokinetics in Goodman &Gilman's The Pharmacological Basis of Therapeutics, Novena Editing, Hardman, JG et al., Eds. McGraw-Hill, New York, 1966) applied to standard animal models, such as a transgenic PDAPP animal model (eg Games, D., et al., Nature 373: 523-527 , 1995, Johnson-Wood, K., et al., Proc. Nati, Acad. SCI, USA 94: 1550-1555, 1997). To optimize the activity and / or specificity of the compound, it may be desirable to construct a neighbor-neighbor analog file for the search for analogues with greater specificity and activity. Methods for synthesizing neighbor-close and / or objectified compound files are well known in the field of combinational archives. C. Inhibitors and Therapeutics Part B above describes the classification method for compounds that have the inhibitory activity of β-secretase. To summarize, a guide is provided for specific classification methods for potent and selective inhibitors of the β-secretase enzyme. Significantly, the practitioner is directed to specific peptide substrate / inhibitor sequences, such as P10-P4 'staD- > V, on which the design of the drug can be based and on additional sources such as phage display files derived, which should provide additional guide compounds. The practitioner is also provided with abundant advice for further refinement of the enzyme binding site, for example, by crystallizing the purified enzyme according to the methods provided herein. Note the success in this area that has been obtained in the area of HIV protease inhibitor development, it is contemplated that such efforts will lead to further optimization of the test compounds described herein. With the optimized compounds available, it is possible to define a pharmacophore compound, and also search existing pharmacophore databases, eg, as provided by Tripos, to identify other compounds that may differ in the 2-D structural formulas with the compounds originally discovered but that share a structure and common pharmacopoeial activity. The test compounds are tested in any of the inhibitor assays described herein, in various stages of development. Therefore the present invention includes β-secretase inhibitory agents disclosed by any of the methods described herein, particularly the inhibitor assays and the crystallization / optimization protocols. Such inhibitory agents are therapeutic candidates for treating Alzheimer's disease, as well as other amyloidoses characterized by deposition of the Aβ peptide. Considerations concerning the therapeutic index (toxicology) bioavailability and doses treated in Part B above are also important to be considered with respect to those therapeutic candidates D. Diagnostic Methods The present invention also provides methods for diagnosing individuals carrying mutations that provide activity improved β-secretase. For example, there are forms of Alzheimer's disease from families in which the underlying genetic disorder has already been recognized. Members of families that possess this genetic predisposition can be monitored for alterations in the nucleotide sequence encoding the β-secretase and / or the promoter regions thereof, since it is apparent, in view of the techniques herein, that individuals who overexpress the enzyme or catalytically possess more efficient forms of the enzyme would probably produce relatively more Aβ peptide. The support for this assumption is provided by the observation reported herein, that the amount of β-secretase enzyme is the rate that limits the production of Aβ in cells. More specifically, people suspected of having a predisposition to develop or who already have the disease, as well as members of the general population, can be classified by obtaining a sample of their cells, which may be for example blood cells or fibroblasts, and when testing the samples for the presence of genetic mutations in the β-secretase gene, for example in comparison to SEQ ID NO: 1 described herein. Alternatively or in addition, the cells of such individuals can be tested for β-secretase activity. According to this embodiment, a particular enzyme preparation could be tested for increased affinity and / or Vmax with respect to a substrate of the β-secretase such as MBP-C125, as described herein, with the comparisons made to the range normal of the measured values in the general population. Individuals whose ß-secretase activity is increased compared to normal values are susceptible to developing Alzheimer's disease or other amyloidogenic diseases that include the deposition of the Aβ peptide. E. Therapeutic Animal Models An additional utility of the present invention is found in the creation of certain transgenic and / or knockout animals that are also useful in the classification assays described herein. Of particular use is a transgenic animal that overexpresses the enzyme β-secretase, such as adding an additional copy of the mouse enzyme or adding the human enzyme. Such an animal can be made according to methods well known in the art (e.g. Cordel1, U.S. Patent 5,387,742; Wadsworth et al. , US 5,811,633, US 5,604,131, US 5,720,936; McConlogue et al. , US 5,612,486; Hsiao et al. , US 5,877,399; and "Mouse Embryo Manipulation, A Laboratory Manual", B. Hogan, F. Costantini and E. Lacy, Cold Spring Harbor Press, 1986), which replaces one or more of the constructions described with respect to the ß -secretase, herein, for the APP constructions described in the above references, all of which are incorporated for reference. A transgenic mouse that overexpresses β-secretase will make higher levels of Aβ and sβAPP from APP substrates than a mouse that expresses endogenous β-secretase. This would facilitate the analysis of APP processing and the inhibition of that processing by candidate therapeutic agents. The improved production of the Aβ peptide in transgenic mouse for the β-secretase would allow the acceleration of the pathology similar to AD observed in transgenic APP mice. This result can be achieved by crossing either the mouse that expresses the β-secretase with a mouse that displays pathology similar to AD (such as the PDAPP mouse or Hsiao) or by creating a transgenic mouse that expresses both the β-secretase and the transgene APP. Such transgenic animals are used to classify the β-secretase inhibitors, with the advantage that they will test the ability of such inhibitors to gain entry to the brain and effect inhibition in vivo. Another animal model contemplated by the present invention is the so-called "knockout mouse" in which the enzyme endogenase is found either permanently (as described in US Patent Nos. 5,464,764, 5,627,059 and 5,631,153, which are incorporated in their entirety for reference) or with inducible deletion (as described in U.S. Patent No. 4,959,317, which is incorporated in its entirety for reference), or which is inactive, as described in greater detail below. Such mice serve as controls for β-secretase activity and / or can be crossed with APP mutant mice, to provide validation of the pathological sequelae. Such mice can also provide a classification for other drug targets, such as drugs specifically targeted in cases of Aβ deposition. The β-secretase knockout mice provide a model of the potential effects of the β-secretase inhibitors in vivo. The comparison of the effects of β-secretase inhibitors in vivo for the β-secretase knockout phenotype can help guide the development of the drug. For example, the phenotype may or may not include pathologies observed during the β-secretase inhibitor test. If the knockout does not show pathologies observed in the mice treated with drugs, it could be inferred that the drug is interacting not specifically with another objective in addition to the β-secretase target. The tissues of the knockout can be used to establish drug binding assays or to carry out the expression of cloning to find the targets that are responsible for these toxic effects. Such information can be used to design additional drugs that do not interact with these undesirable objectives. Knockout mice will facilitate the analysis of potential toxicities that are inherent to the inhibition of β-secretase. Knowledge of potential toxicities will help guide the design of designer drugs or drug delivery systems to reduce such toxicities. Inducible knockout mice are particularly useful for distinguishing toxicity in an adult animal from effects observed in standard knockouts. If knockouts confer lethal fetal effects, inducible knockouts will be advantageous. The methods and technology for developing knockout mice have matured to the point that several of the commercial enterprises generate such mice on the basis of a contract (e.g. Lexicon Genetics, Woodland TX; Cell and Molecular Technologies, Lavallete, NJ; Crysalis, DNX Transgenic Sciences, Princeton, NJ). The methodologies are also available in the art. (See Galli-Taliadoros, L.A. et al., J. Immunol., Meth. 181: 1-15, 1995). In summary, a genomic clone of the enzyme of interest is required. Where, as in the present invention, the exons encoding the regions of the protein have been defined, it is possible to achieve inactivation of the gene without further knowledge of the regulatory sequences that control transcription. Specifically, a genomic 129 file of mouse strain can be classified by hybridization or PCR, using the sequence information provided herein, according to methods well known in the art. (Ausubel, Sambrook) The genomic clone thus selected is then subjected to restriction delineation and exonic partial sequencing for confirmation of the homologous mouse and to obtain information for the construction of the knock-out vector. The appropriate regions are then sub-cloned into a "knock-out" vector carrying a selectable marker, such as the vector carrying a neor cassette that renders the cells resistant to aminoglycoside antibiotics such as gentamicin. The construct is further designed for the fractionation of the gene of interest, such as by insertion of a sequence replacement vector, into which a selectable marker is inserted into an exon of the gene, where it serves as a mutagen, which interrupts the Coordinated transcription of the gene. The vectors are then designed by transfection in embryonic stem (ES) cells, and the appropriate colonies are isolated. Clones of positive ES cells are microinjected into blastocysts isolated from the host to generate chimeric animals, which are then bred and sorted by transmission of the germ line of the mutant allele. According to a preferred additional embodiment, the β-secretase knockout mice can be generated so that the mutation is inducible, such as by inserting a lox region in the knockout mouse flanking the β-secretase gene region. Such mice are then crossed with mice carrying a "Cre" gene under an inducible promoter, which results in at least some progeny carrying both the "Cre" and the lox constructs. When the expression of Cre is induced, it serves to fraction the gene flanked by the lox constructions. Such a "Crelox" mouse is particularly useful when it is suspected that the knock-out mutation can be lethal, and also provides the opportunity to knock out the genes in the selected tissue, such as the brain. Cre-lox are provided by US Patent 4,959,317 incorporated herein by reference, and are made on the basis of a contract by Lexicon Genetics, Woodlands TX, etc. The following examples illustrate, but in no way attempt to limit the present invention Example 1 Isolation of the Coding Sequences for Human β-secretase A. PCR Cloning The poly A + RNA of human neuroblastoma cells IMR were reverse transcribed using the Perkin-Elmer equipment. one degraded eight times, coding the terminal portions N and C of the amino acid sequence obtained from the purified protein (shown in Ta bla 4; oligos from 3407 to 3422). The PCR reactions consisted of lOng RNA cDNA, 1.5mM MgCl2, 0.125μl AmpliTaq® Gold, 160μM of each dNTP (plus an additional 2OμM of the reverse transcriptase reaction) Perkin-Elmer TAQ regulator (AmpliTaq® Gold equipment, Perkin-Elmer, Foster City, CA), in a reaction volume of 25μl. Each of the oligonucleotide primers 3407 to 3414 was used in combination with each of oligos 3415 to 3422 for a total of 64 reactions. The reactions were run on the Perkin-Elmer 7700 Sequence Detection machine under the following conditions: 10 minutes at 95 ° C, 4 cycles of 45 ° C tempered for 15 seconds, extension at 72 ° C for 45 seconds and denaturation at 95 ° C for 15 seconds followed by 35 cycles under the same conditions with the exception that the tempering temperature was raised to 55 ° C. (The above conditions are referred to herein as "Reaction Conditions 1") the PCR products were visualized on 4% agarose gel (Northern blots) and a prominent band of the expected size (68 bp) was observed in the reactions, particularly with primers 3515-3518. The 68 kb band and the internal region encoded by the expected amino acid sequence were sequenced. This gave the exact DNA sequence for 22 bp of the inner region of this fragment. The additional sequence was deduced from the efficiency of several groups of initiators of the discontinuous sequence when generating this PCR product. Initiating groups 3419 to 3422 gave a very poor or no product, considering that groups 3415 to 3418 gave a robust signal. The difference between these groups is a CTC (3415 to 3418) versus TTC (3419 to 3422) in the most extreme 3 'of the groups. Since the CTCs started more efficiently, we can conclude that the GAG inverse complement is the correct codon. Since the Met coding is unique it was concluded that the next codon is ATG. Thus, the exact sequence of DNA obtained is: CCC. GGC CGG .AGG. GGC AGC. TTT. GTG GAG .ATG. GT (SEQ ID NO: 49) encoding the amino acid sequence P G R R G S F V E M V (SEQ ID NO: 50). This sequence can be used to design precise oligonucleotides for RACE PCR 3 and 5 'r in any cDNA or files or to design specific hybridization tests to be used to select files. Since the degenated PCR products were found to be quite robust, this reaction can also be used as a diagnostic for the presence of genes containing this sequence. File groups can be selected using this PCR product to indicate the presence of a clone in the group. The group can be decomposed for individual identification. The known complexity and / or size classification groups can provide information on the abundance of this clone in a file or source and can approximate the size of the clone or full length message. For the generation of a test, PCR reactions using oligonucleotides 3458 and 3469 or 3458 (SEQ ID NO: 19) and 3468 (SEQ ID NO: 20) (Table 4) can be carried out using the product RACE 23, clone 9C7E.35 (30 ng, clone 9C7E.35 that was isolated from the source file, see Example 2) or cDNA generated from the brain, using the standard PCR conditions (Perkin-Elmer, rtPCR and AmpliTaq® Gold equipment) with the following: 25 μl reaction volume 1.5 mM MgC12, 0.125 μl AmpliTaq® Gold (Perkin-Elmer), initial 95 ° for 10 min. to activate the AmpliTaq® Gold, 36 cycles of 65 ° 15 sec, 72 ° 45 sec, 95 ° for 15 sec, followed by 3 min at 72 °. The product was purified in a Quiagen purification equipment by PCR and used as a substrate for random priming to generate a radiolabelled test (Sambrook et al., Supra, Amersham RediPrime® equipment). This test was used to isolate clone 27 of full-length closed pCEK shown in Figures 12 and 13 (A-E). Derivation of pCEK clone 27 from the full-length clone A human primary neuronal cell file in the mammalian expression vector of the pCEK2 vector was generated using cDNA of selected size and clusters of genes generated from differently sized grafts. The cDNA file for the selection of β-secretase was made with poly (A) + RNA isolated from primary human neuronal cells. The cloning vector was pCEK2 (Figure 12). pCEK2 The double-stranded cDNA inserts were synthesized using the Stratagene cDNA Synthesis Kit with some modifications. The inserts were then fractioned according to their sizes. A total of five fractions were individually ligated with double cut pCEK2 (Notl and Xhol) and subsequently transformed into the E. coli strain XL-10 Gold which was designed to accept very large plasmids. Transformed E. coli fractions were plated onto Terrific Broth agar plates containing ampicillin and allowed to develop for 18 hours. Each fraction produced approximately 200,000 colonies to give a total of one million colonies. The colonies were then scraped off the plates and the plasmids were isolated from them in groups of approximately 70,000 clones / group. 70,000 kidneys from each group of the file were selected for the presence of the putative β-secretase gene using the PCR reaction diagnosis (the degenerate primers 3411 and 3417 shown above). The samples from the 1.5 kb group were classified using a radiolabelled test generated from a PCR product of 390 b.p. generated from clone 9C7E.35. For the generation of a test, the PCR product was generated using 3458 and 3468 as primers and the clone 9C7E.35 (30 ng) as substrate. The PCR product was used as a substrate for random priming to generate a radiolabelled test. A total of 180,000 cites were classified from the 1.5 kb group (70,000 original cessions in this group) by hybridization with the PCR test and 9 positive samples were identified. Four of these isolates were isolated and by restriction delineation it appeared that they encode two independent sequences from 4 to 5 kb (clone 27) and 6 to 7 kb (clone 53) in length. Sequencing of clone 27 verified that it contained a coding region of 1.5 kb. Figure 13 (AE) shows the sequence of clone 7 of pCEK (clone27) Table 4 Example 2 Classification of the Human Fetal Brain cDNA File The Human Fetal Brain Rapid-Screen ™ cDNA File Panel is provided as an array of 96-cavity format consisting of 5000 kidnaps (plasmid DNA) by cavities of a human fetal brain file. The subplates are available for each cavity consisting of 96 cavities of 50 percent each in E. coli. This is an oligo-dT primer file, of selected size and inserted unidirectionally in the vector pCMV-XL3. 96 cavities were classified from the main plate using PCR. Reaction Conditions 1 described in Example 1, above, were continued using only primers 3407 and 3416 with 30ng of plasmid DNA from each well. Two groups showed the positive band 70bp. The same primers and conditions were used to classify lμl of E. coli from each cavity of one of the subplates. E. coli from the single positive cavity was then plated on LB / amp plates and single colonies were classified using the same PCR conditions. The positive clone, approximately 1 Kb in size was labeled 9C7E.35. It contained the sequence of the original peptide as well as the 5 'sequence that included a methionine. The 3 'sequence did not contain a stop codon, suggesting that this was not a full length clone, consistent with Northern blot data. Example 3 PCR Cloning Methods 3 'RACE was used in experiments carried out in support of the present invention to elucidate the polynucleotide coding of human β-secretase.

Appropriate methods and conditions for the repetition of the experiments described herein and / or the determination of the polynucleotide sequences encoding additional members of the new family of aspartyl proteases described herein, can be found, for example, in White, BA, ed., PCR Cloning Protocols; Humana Press, Totowa, NJ, 1997 or Ausubel, supra, both of which are incorporated herein by reference. RT-PCR For the reverse transcription polymerase chain reaction (RT-PCR), two sets of partially degenerate primers used for the RT-PCR amplification of a cDNA fragment encode this peptide. The primer set 1 consisted of DNA's # 3427-3434, the sequences of which are shown in Table 5, below. Matrix RT-PCR using primer combinations from this set with reverse transcribed cDNA from primary human neuronal cultures as the model yielded the 54 bp cDNA product predicted with primers # 3428 + 3433. All RT-reactions PCR used 10-50 ng of poly-A + RNA entry equivalents per reaction and was carried out for 35 cycles using cycle-by-stage conditions with a denaturation of 95 ° C for 1 minute, 50 ° C for 30 sec and 72 ° C extension for 30 sec. The degeneracy of primers # 3428 + 3433 was further decomposed, resulting in primer set 2, comprising DNAs # 3448-3455 (Table 5). Matrix RT-PCR was repeated using primer set 2 and reverse transcribed cDNA from poly-A + RNA from human neuroblastoma cells IMR-32 (American Type Culture Collection, Manassas, VA), as well as human neuronal cultures primary, as a model for amplification. Initiators # 3450 and 3454 from set 2 more efficiently amplified a cDNA fragment of the predicted size (72 bp), although primers 3450 + 3453 and 3450 + 3455 also amplified the same product, albeit at lower efficiency. A PCR product of 72 bp was obtained by amplifying cDNA from IMR-32 cells and primary human neuronal cultures with primers 3450 and 3454. RACE-PCR 5 'and 3' Internal primers that match the upper braid ( coding) for the 3 'Extreme Amplification of 5' End (RACE) PCR and lower braid (without coding) for 5 'RACE PCR were designed and elaborated according to methods known in the art (eg, Forman, MA, MK Dush and GR Martin (1988). "Rapid reproduction of full-length cDNA from rare transcripts: amplification using a single oligonucleotide-specific gene primer." Proc. Nati .Acad.Sci. US .A. 85 (23): 8998 -9002). The DNA primers used for this experiment (# 3459 and # 3460) are presented in Table 4. These primers can be used in standard RACE-PCR methodology using commercially available models (eg Ready Marathon cDNA®, Clontech Labs) or models of CDNA designs specifically prepared from RNAs of interest as described by Frohman et al. , (ibid). In experiments carried out in support of the present invention, a variation of RACE was employed to exploit an IMR-32 cDNA file cloned into the retrovirus expression vector pLPCXlox a derivative of pLNCX. Since the vector junctions provide unique anchor sequences that contact the cDNA inserts in this file, they serve the purpose of the 5 'and 3' anchor primers in the RACE methodology. The sequences of the specific 5 'and 3' anchor primers that we used to amplify the cDNAs of the β-secretase from the file, primers # 3475 and # 3476, were derived from the DNA sequence of the vector provided by Clontech Labs , Inc. and are shown in Table 3. Primers # 3459 and # 3476 were used for the 3 'RACE amplification of the current sequence below our IMR-32 cDNA file in the pLPCXlox vector. The file has been previously subdivided into 100 groups of 5, 000 cgs per group and the plasmid DNA was isolated from each group. A survey of the 100 groups with the initiators identified as diagnostic for the presence of the β-secretase clone, according to the methods described in Example 1, above, provided individual groups from the file for RACE-PCR. 100 ng of the model plasmid from group 23 was used for PCR amplification with primers 3459 + 3476. The amplification was carried out for 40 cycles using ampli-Taq Gold®, under the following conditions: denaturation at 95 ° C for 1 min, annealing at 65 ° C for 45 sec and extension at 72 ° C for 2 min. The reaction products were fractionated by agarose gel chromatography, according to methods known in the art (Ausubel; Sambrook). A fragment of approximately 1.8 kb PCR was revealed by fractionation of agarose gel from the reaction products. The PCR product was purified from the gel and subjected to DNA sequence analysis using primer # 3459. The resulting sequence, designated 23A and the predicted amino acid sequence deduced from the DNA sequence are shown in Figure 5. Six of the first seven amino acids deduced from one of the 23A reading structures were an exact match with the last 7 amino acids of the N-terminal sequence determined from the purified protein and sequenced in further experiments carried out in support of the present invention, from natural sources. Table 5 Example 4 Tests of the β-secretase Inhibitor Tests to measure the β-secretase activity are well known in the art. Particularly useful assays, summarized below, are detailed in the U.S. Patent. 5,744,346 incorporated herein by preference. A. Preparation of MBP-C125sw 1. Cell preparation Two 250 ml cell culture flasks containing 50 ml of LBamp 100 per flask were seeded with a colony per E. coli flask pMAL-C125SW cl.2. { E. coli expressing the fusion protein MBP-C125sw). The cells were allowed to develop overnight at 37 ° C. The aliquots (25 ml) were seeded in 50 ml per LBamp 100 flask in 2 liter flasks, which were then allowed to develop at 30 °. The optical density was measured at 600 nm (OD60o) vs LB broth; 1.5 ml of 100 mM IPTG were added when the OD was "0.5" At this point, an aliquot of pre-incubation was removed by SDS-PAGE ("-I".) 0.5 ml was centrifuged from this aliquot for 1 min in a Beckman microcentrifuge and the resulting pellet was dlved in 0.5 ml lxLSB.

The cells were incubated / induced for 5-6 hours at 30 C, after which the aliquot was removed after incubation.

("+ I"). The cells were then centrifuged at 9,000 rpm in a KA9.1 rotor for 10 minutes at 4 ° C. The pellets were retained and stored at -20C. 2. Pellet extraction of bacterial cells Pellets of frozen cells were resuspended in 50 ml of 0.2 M NaCl, 50 mM Tris, pH 7.5, then sonicated in rosette containers for 5 x 20 seconds of explosions, with 1 min rest between Explosions The extract was centrifuged at 16,500 rpm in a KA18.5 rotor for 30 min (39,000 x g). Using pipettes as a pistil, the pellets were suspended in 50 ml of urea extraction buffer (7.6 M urea, 50 mM Tris pH 7.5, 1 mM EDTA, 0.5% TX-100). The total volume was approximately 25 ml per flask. The suspension was then sonified 6 x 20 sec, with one minute of rest between explosions. The suspension was then centrifuged again at 16,500 rpm for 30 min in the KA18.5 rotor. The resulting supernatant was added to 1.5 L of regulator consisting of 0.2 M Tris-regulating NaCl 50, pH 7.5, with 1% Triton X-100 (0.2M NaCl-Tris-1% Tx) and stirred gently at 4 degrees C for 1 hour, followed by centrifugation at 9,000 rpm in KA9.1 for 30 min at 4 ° C. The supernatant was loaded onto a column of washed amylose (100 ml of 50% mixture; New England BioLabs). The column was washed with 0.2 M NaCl-Tris-1% TX until the baseline (+ 10 column volumes) was then reduced with 2 column volumes of 0.2 M NaCl-Tris 1% Triton X-100. The protein was then eluted with 10 mM of maltose in the same regulator. An equal volume of 6 M guanidine HCl / 0.5% TX-100 was added to each fraction. The peak fractions were pooled and diluted to a final concentration of approximately 2 mg / ml. The fractions were stored at -40 degrees C before dilution (20 times, at 0.1 mg / ml in 0.15% Triton X-100). The diluted aliquots were also stored at -40 C. B. Antibody-based assays The assays described in this section are based on the ability of certain antibodies, hitherto "cleavage site antibodies" to distinguish the cleavage of APP by β-secretase, based on the single cleavage site and the consequent exposure of a specific C-terminal formed by the splitting. The recognized sequence is a sequence of approximately the usual 3-5 residues which is located immediately at the amino terminus of the β-amyloid peptide (βAP) produced by the cleavage of the β-APP-β-secretase, such as Val-Lis- Met in wild type or Val-Asn-Leu- in the variant form of double mutation Swedish of APP. The recombinantly expressed proteins, described below, were used as substrates for β-secretase. MBP-C125 assay: MBP-C125 susbstartos were expressed in E. coli as a fusion protein of the last 125 amino acids of APP fused to the carboxyl terminus of the maltose binding protein (MBP), using the commercially available vectors of New England Biolabs. The cleavage site β was thus 26 amino acids downstream of the beginning of the C-125 region. This latter site is recognized by the monoclonal antibody SW192. The recombinant proteins were generated with both of the wild-type APP sequence (MBP-C125wt) at the cleavage site (..Val-Lis-Met-Asp-Ala ..) or the double mutation "Swedish" (MBP-C125sw) (..Val-Asn-Leu-Asp-Ala ..). As shown schematically in Figure 19A, cleavage of the intact MBP fusion protein results in the generation of a truncated amino-terminal fragment, with the new positive epitope Ab SW-192 not covered at the carboxyl terminus. This amino-terminal fragment can be recognized in Western blots with the same Ab, or quantitatively, using a biotinylated anti-MBP reporter SW-192 sandwich format, as shown in Figure 19A. Polyclonal anti-MBP antibodies were raised in rabbits (Josman Labs, Berkeley) by immunization with recombinantly expressed purified MBP (New England Biolabs). The antiserum was purified by affinity on a column of immobilized MBP. Substrates MBP-125 SW and WT were expressed in E. coli, then purified as described above. The 96-well microtiter plates were covered with purified anti-MBP antibody (at a concentration of 5-10 μg / ml), followed by blocking with 2.5 g / liter of human serum albumin in 1 g / liter of sodium phosphate monobasic, 10.8 g / liter of dibasic sodium phosphate, 25 g / liter of sucrose, 0.5 g / liter of sodium azide, pH 7.4. The appropriately diluted β-secretase enzyme (5μl) was mixed with 2.5μl of 2.2μM MBP-125sw substrate base, in a 50 μl reaction mixture with a final regulator concentration of 20 mM acetate buffer, pH 4.8 , 0.06% Triton X-100, in individual cavities of a 96-well microtiter plate and incubated for one hour at 37 ° C. The samples were then diluted 5 times with Specimen Diluent (Sample Diluent) (0.2 g / 1 of monobasic sodium phosphate, 2.15 g / 1 of dibasic sodium phosphate, 0.5 g / 1 of sodium azide, 8.5 g / 1 of sodium chloride, 0.05% Triton X-405, 6 g / 1 BSA), additionally diluted 5-10 times in Specimen Diluent in plates covered with anti-MBP and incubated for 2 hours at room temperature. After incubations with samples or antibodies, the plates were washed at least four times in TTBS (0.15 M NaCl, 50 mM Tris, pH 6.5, 0.05% Tween-20). Biotinylated SW192 antibodies were used as the reporter. The SW192 polyclonal antibodies were biotinylated using biotin-NHS (Pierce), following the manufacturer's instructions. Usually, the biotinylated antibodies were used at approximately 240 ng / ml, the exact concentration varied with the batch of antibodies used. After incubation of the plates with the reporter, ELISA was developed using alkaline phosphatase labeled streptavidin (Boeringer-Mannheim) and 4-methyl umbelliferyl phosphate as a fluorescent substrate. The plates were read in a Cytofluor 2350 Fluorescent Measurement System. The recombinantly generated MBP-26SW (analogous product) was used as a standard to generate a standard curve, which allowed the conversion of fluorescent units, in quantities of product generated. This test protocol was used to select inhibitory structures using "files" of compounds assembled in 96-well microtiter plates. The compounds were added, in a final concentration of 20μg / ml in 2% DMSO, in the assay format described above, and the degree of product generated was compared with the incubations of control β-secretase (2% DMSO only) , to calculate the "% inhibition". The "hits" were defined as compounds that result in >35% inhibition of enzyme activity in the test concentration. This assay was then used to provide IC50 values for the inhibitors by varying the concentration of the test compound through a range to calculate from a dose response curve, the concentration required to inhibit the enzyme activity by 50%. In general, inhibition was considered significant as compared to the activity of the control in this assay if it results in activity that is at least 1 standard deviation and preferably 2 standard deviations lower than a value of average activity determined across a range of samples. In addition, a reduction in activity that is greater than about 25%, and preferably greater than about 35%, of the activity of the control can also be considered significant. Using the above test system, 24"hits" (> 30% inhibition at 50μm concentration) of the first 6336 compounds tested (0.4% hit rate) were identified. Of these 12 compounds had IC50s less than 50 μM, including reclassification in the P26-P4'sw assay, below. Test P26-P4'sw. The substance P26-P4'sw is a biotin-linked peptide of the sequence (biotin) CGGADRGLTTRPGSGLTNIKTEEISEVNLDAEF (SEQ ID NO: 63). Standard P26-P1 has the sequence (biotin) CGGADRGLTTRPGSGLTNIKTEEISEVNL (SEQ ID NO: 64), where the N-terminal "CGG" serves as a link between the biotin and the substrate in both cases. Peptides were prepared by Anaspec, Inc. (San Jose, CA) using solid phase synthesis with boc amino acids. Biotin was coupled to terminal sulfhydryl cysteine by Anaspec, Inc,. after peptide synthesis, using EZ-Iodoacetyl-LC-Biotin link (Pierce). The peptides were stored as stocks 0.8-1.0 Mm in 5 mM Tris, with the pH adjusted around neutral (pH 6.5-7.5) with sodium hydroxide. For the enzyme assay, the substrate concentration can vary from 0-200 μM. Specifically for test compounds, for the inhibitory activity, the substrate concentration is 1.0 μM. The compounds to be tested were added in DMSO, with a final concentration of DMSO of 5%; in such experiments, the controls also received 5% DMSO. The concentration of the enzyme was varied, to give product concentrations within the linear range of the ELISA assay (125-2000 pM, after dilution). These compounds were incubated in 20 mM sodium acetate, pH 4.5, 0.06% Triton X-100, at 37 ° C for 1 to 3 hours. Samples were diluted 5-fold in sample diluent (145.4 mM sodium chloride, 9.51 mM sodium phosphate, 7.7 mM sodium azide, 0.05% Triton X-405, 6 gm / liter bovine serum albumin, pH 7.4 ) to quench the reaction, then it was further diluted for the ELISA as necessary. For the ELISA, Costar High Binding 96-well assay plates (Corning, Inc., Corning, NY) were coated with monoclonal antibody SW 192 of clone 16A7, or a similar affinity clone. The biotin-P26-P4 'standards were diluted in the sample diluent to a final concentration of 0 to 2 nM. Diluted samples and standards (100 μl) were incubated in SW192 plates at 4 ° C for 24 hours. Plates were washed 4 times in TTBS buffer (150 mM sodium chloride, 25 mM Tris, 0.05% Tween 20, pH 7.5), then incubated with 0.1 ml / cavity of estrepatavidin - alkaline phosphates (Roche Molecular Biochemicals, Indianapolis , IN) was diluted 1: 3000 in the sample diluent. After incubation for one hour at room temperature, the plate was washed 4 times in TTBS, as described in the previous section, and incubated with fluorescent substrate solution A (31.2 gm / liter 2-amino-2-methyl- 1-propanol, 30 mg / liter, adjusted to pH 9.5 with HCl). The fluorescence values were read after 30 minutes. C. Assays using Synthetic Oligopeptide Substrates This assay format is particularly useful for measuring the activity of partially purified β-secretase preparations. Synthetic oligopeptides incorporating the known cleavage site of the β-secretase were prepared, and optional detectable markers, such as fluorescent or chromogenic residues. Examples of such peptides, as well as their methods of production and detection are described in the U.S. Patent. 5,942,400 issued, incorporated herein by reference. The cleavage products can be detected using high performance liquid chromatography, or fluorescent or chromogenic detection methods suitable for the peptide to be detected, according to methods well known in the art. By way of example, one such peptide has the sequence SEVNL DAEF (SEQ ID NO: 52), and the cleavage site is between residues 5 and 6. Another preferred substrate has the sequence ADRGLTTRPGSGLTNIKTEEISEVNLDAEF (SEQ ID NO: 53) , and the cleavage site is between residues 26 and 27. D. Ens-Seretase Assays of Cell Extracts or Raw Tissues The cells or tissues were extracted in the extraction buffer (20 mM HEPES, pH 7.5, 2 mM EDTA, 0.2% Triton X-100, 1 mM PMSF, 20 μg / ml pepstatin, 10 μg / ml E-64). The volume of the extraction regulator will vary between samples, but will be at least 200μl per 106 cells. The cells can be suspended by grinding with a micropipette, although the tissue may require homogenization. The suspended samples were incubated for 30 minutes on ice. If necessary pipetting is allowed, the non-solubilized material was removed by centrifugation at 4 degrees C, 16,000 x g (14,000 rpm in a Beckman microcentrifuge) for 30 minutes. The supernatant was analyzed by dilution in the final test solution. The dilution of the extract will vary, but should be sufficient so that the protein concentration in the assay is not greater than 60 μg / ml. The assay reaction also contained 20 mM sodium acetate, pH 4.8, and 0.06% Triton X-100 (including Triton contributed by the extract and the substrate), and 220-110 nM MBP-C125 (a dilution of 1 : 10 or 1:20 of the 0.1 mg / ml stock described in the protocol for the preparation of the substrate). The reactions were incubated for 1-3 hours at 37 degrees C before being quenched with at least 5 times of dilution in the sample diluent and assayed using the standard protocol. Example 5 Purification of β-secretase A. Purification of the naturally occurring β-secretase 293 human cells were obtained and processed as described in the U.S. Patent. 5,744,346 incorporated herein by reference. (293 cells were available from the American Type Culture Collection, Manassas, VA). The frozen tissue (cell paste 293 or human brain) was cut into pieces and combined with five volumes of homogenized buffer (20 mM Hepes, pH 7.5, 0.25 M sucrose, 2 mM EDTA). The suspension was homogenized using a mixer and centrifuged at 16,000 x g for 30 minutes at 4 ° C. The supernatant was discarded and the pellets were suspended in extraction buffer (20 mM MES, pH 6.0, 0.5% Triton X-100, 150 mM NaCl, 2 mM EDTA, 5 μg / ml leupeptin, 5 μg / ml E64, 1 μg / ml pepstatin, 0.2 mM PMSF) in the original volume. After swirling, the extraction was completed by shaking the tubes at 4 ° C for a period of one hour. The mixture was centrifuged as above at 16,000 x g, and the supernatants were pooled. The pH of the extract was adjusted to 7.5 by adding ~ 1% (v / v) of 1M Tris base (not neutralized). The neutralized extract was loaded onto a column of wheat germ agglutinin agarose (WGA-agarose) pre-equilibrated with 10 column volumes of 20 mM Tris, pH 7.5, 0.5% Triton X-100, 150 mM NaCl, 2 mM EDTA, at 4 ° C. One milliliter of the agarose resin was used for each 1 g of the original tissue used. The WGA column was washed with 1 column volume of the equilibrium regulator, then 10 volumes of 20 mM Tris, pH 7.5, 100 mM NaCl, 2 mM NaCl, 2 mM EDTA, 0.2% Triton X-100 and then eluted as follow. Three quarters of the column volume of 10% hydrolyzed chitin in 20 mM Tris, pH 7.5, 0.5%, 150 mM NaCl, 0.5% Triton X-100, 2 mM EDTA were passed through the column after which the flow was stopped for fifteen minutes. An additional volume of five columns of 10% hydrolyzed chitin solution was then used to elute the column. All the previous elusions were combined (eluted WGA grouped) The pooled WGA eluate was diluted 1: 4 with 20 mM NaOAc, pH 5.0, 0.5% Triton X-100, 2 mM EDTA. The pH of the diluted solution was adjusted to 5.0 by adding a few drops of glacial acetic acid while monitoring the pH. This "SP" charge was passed through a 5 ml column of Pharmacia HiTrap SP equilibrated with 20 mM NaOAc, pH 5.0, 0.5% Triton X-100, 2 mM EDTA, in 4 ml / min at 4 ° C. The above methods provided peak activity having a specific activity of more than 253 nM product / ml / h / μg protein in the MBP-C125-SW assay, where the specific activity was determined as described below, with approximately purification 1,500 times of the protein. The specific activity of the purified β-secretase was measured as follows. MBP substrate C125-SW was combined at approximately 220 nM in 20 mM sodium acetate, pH 4.8, with 0.06% Triton X-100. The amount of product generated was measured by the β-secretase assay, also described below. The specific activity was calculated as: Specific Activity = (Conc. Of product nMHfactor dilution) (Vol. Sol. Enzyme) (Time Incub h) (Conc Enzyme mg / vol) The Specific Activity is expressed as pmoles of product produced per μg of β-secretase per hour. The additional puffification of human brain enzyme was achieved by loading the SP flow through the fraction in the affinity column P10-P4 'staD- »V, according to the general methods described below. The results of this purification step are summarized in Table 1, above. B. Purification of β-secretase from Recombinant Cells The recombinant cells produced by the methods described herein were generally made to over-express the enzyme; that is, they produced dramatically more enzyme per cell than was found endogenously produced by cells or by most tissues. It was found that some of the steps described above can be omitted from the preparation of the purified enzyme under these circumstances, with the results being achieved even at higher levels of purification. CosA2 or 293T cells transfected with the construction of the β-secretase gene (see Example 6) were pelleted, frozen and stored at -80 degrees until use. The cell pellets were resuspended by homogenization for 30 seconds using a hand homogenizer (0.5 ml / pellet of approximately 106 cells in extraction buffer consisting of 20 mM TRIS buffer, pH 7.5, 2 mM EDTA, 0.2% Triton X-100, plus the protease inhibitors: 5 μg / ml E-64, 10 μg / ml pepstatin, 1 mM PMSF), centrifuged at maximum speed in a microcentrifuge (40 minutes at 4 degrees C). The pellets were suspended in the original volume of the extraction regulator, then stirred 1 hour at 4 degrees C with rotation, and centrifuged again in a microcentrifuge at maximum speed for 40 minutes. The resulting supernatant was saved as the "extract". The extract was then diluted with 20 mM sodium acetate, pH 5.0, 2 mM EDTA and 0.2% Triton X-100 (SP regulator A), and 5 M was added.

NaCl at a final concentration of 60mM NaCl. The pH of the solution was then adjusted to pH 5.0 with glacial acetic acid diluted 1:10 in water. The aliquots were saved ("SP load"). The SP charge was passed through a 1 ml SP HiTrap column which was pre-washed with 5 ml SP buffer A, 5 ml SP buffer B (SP buffer A with 1 M NaCl) and 10 ml of regulator A of SP. An additional 2 ml of 5% SP buffer B was passed through the column to displace any remaining sample from the column. The pH of the through flow SP was adjusted to pH 4.5 with 10X diluted acetic acid. This flow was then applied to a Sepharose Affinity column P10-P4 'staD? V, as described below. The column (250 μl bed size) was pre-equilibrated with at least 20 column volumes of equilibrium buffer (25 mM NaCl, 0.2% Triton X-100, 0.1 mM EDTA, 25 mM sodium acetate, pH 4.5), then loaded with the diluted supernatant. After loading, subsequent steps were carried out at room temperature. The column was washed with wash buffer (125 mM NaCl, 0.2% Triton X-100, 25 mM sodium acetate, pH 4.5) before the addition of 0.6 column bed volumes of borate elution buffer (200 mM NaCl , 0.2% reduced Triton X-100, 40 mM sodium borate, pH 9.5). The column was then capped and an additional 0.2 ml of elution buffer was added. The column was allowed to stand for 30 minutes. Two volumes of elution buffer bed were added, and column fractions (250 μl) were collected. The protein peaks were eluted in two fractions. 0.5 ml of pepstatin 10 mg / ml per milliliter of collected fractions was added. Cell extracts made from cells transfected with full-length clone 27 (which encodes SEQ ID NO: 2; 1-501), stop 419 (SEQ ID NO: 57) and stop 452 (SEQ ID NO: 59) were detected by Western blot analysis using antibody 264A (polyclonal antibody directed to amino acids 46-67 of β-secretase with reference to SEQ ID NO: 2). Example 6 Preparation of Heterologous Cells Expressing Recombinant β-Secretase Two separate assays (clone 27 of pCEK and clone 53 of pCEK) were transfected into 293T or C0S (A2) cells using Fugene and Effectene methods known in the art. 293T cells were obtained from Edge Biosystems (Gaithersburg, MD). They are KEL293 cells transfected with the large SV40 antigen. COSA2 are a sub-clone of COSÍ cells; subcloned in soft agar. FuGENE method: 293T cells were seeded at 2xl05 cells per cavity of a 6-well culture plate. After growth overnight, the cells were at approximately 40-50 of confluence. The medium was changed a few hours before transfection (2 ml / well). 3 μl of FuGENE 6 transfection reagent (Roche Molecular Biochemicals, Indianapolis, IN) was diluted for each sample in 0.1 ml of serum-free culture medium (DME with 10 mM Hepes) and incubated at room temperature for 5 minutes. . One microgram of DNA was added to each sample (0.5-2 mg / ml) to a separate tube. The diluted FuGENE reagent was added dropwise to the concentrated DNA. After gently covering to mix, the mixture was incubated at room temperature for 15 minutes. The mixture was added dropwise onto the cells and gently stirred until mixed. The cells were then incubated at 37 degrees C in an atmosphere of 7.5% C02. The conditioned medium and the cells were harvested after 48 hours. The conditioned medium was collected, centrifuged and isolated from the pellet. Protease inhibitors (5 μg / ml E64, 2 μg / ml peptsatin, 0.2 mM PMSF) were added before freezing. The cell monolayer was rinsed once with PBS, then 0.5 ml of lysis buffer (1 mM HIPIS, pH 7.5, 1 mM EDTA, 0.5% Triton X-100, 1 mM PMSF, 10 μg / ml E64) was added. The lysate was frozen and thawed, mixed by vortex, then centrifuged and the supernatant was frozen until analysis. Effective Method DNA (0.6 μg) was added with "EFFECTENE" reagent (Qiagen, Valencia, CA) in a 6-well culture plate using a standard transfection protocol according to the manufacturer's instructions. The cells were harvested 3 days after transfection and the cell pellets were frozen rapidly. The whole cell lysates were prepared and various amounts of lysate were tested for the β-secretase activity using the MBP-C125 substrate. Figure 14B shows the results of these experiments, in which the picomoles of the formed product were plotted against the micrograms of the lysate of COS cells added to the reaction. The legend in the figures describes the enzyme source, where the activity of the cells transfected with DNA from clone 27 of pCEK and clone 53 of pCEK (grades 27 and 53) is shown using Effective as closed diamonds and solid frames, respectively, the activity of the cells transfected with DNA from clone 27 prepared with FuGENE are shown as open triangles, and the transfected imitation and the control charts show no activity (closed triangles and "X" markers). The product with values greater than 700 pM is outside the linear range of the analysis. Example 7 Preparation of the Sepharose Affinity Matrix P10-P4'sta (D? V) A. Preparation of the inhibitor peptide P10-P4'sta (D? V) The P10-P4 'sta (D? V) has the sequence NH2-KTEEISEVN [sta] VAEF-COOH (SEQ ID NO: 72), wherein "sta" represents a statin residue. The synthetic peptide was synthesized on a peptide synthesizer using boc-protected amino acids for chain installation. All chemicals, reagents and boc amino acids were purchased from Applied Biosystems (ABI; Foster City, CA) with the exception of dichloromethane and N, N-dimethylformamide that were from Burdick and Jackson. The starting resin, the boc-Phe-0CH2-Pam resin was also purchased from ABI. All amino acids were coupled following the pre-activation for the corresponding HOBT ester, using the equivalent of 1.0 of 1-hydroxybenzotriazole (HOBT), and the equivalent of 1.0 N, N-dicyclohexylcarbodiimide (DCC) in dimethylformamide. The protection group boc in the amino acid a-amine was removed with 50% trifluoroacetic acid in dichloromethane after each coupling step and before the cleavage of Hydrogen Fluoride. The amino acid side chain protection was as follows: Glu (Bzl), Lis (Cl-CBZ), Ser (OBzl), Tre (OBzl). All other amino acids were used without additional side chain protection including boc-Statin. [(Bzl) benzyl, (CBZ) carbobenzoxy, (Cl-CBZ) chlorocarbobenzoxy, (OBzl) O-benzyl] The side chain protected peptide resin was deprotected and cleaved from the resin by reaction with anhydrous hydrogen fluoride ( HF) at 0 ° C for one hour. This generates the crude peptide totally deprotected as a C-terminal carboxylic acid. Following the HF treatment, the peptide was extracted from the resin in acetic acid and lyophilized. The crude peptide was then purified using preparative reverse phase HPLC on a Vydac C4, 330A, 10 μm 2.2 cm I.D. column. x 25 cm in length. The solvent system used with this column was 0.1% TFA / H20 (regulator [A]) and 0.1% TFA / CH3CN (regulator [B]) as the mobile phase. Typically the peptide was loaded onto the column in 2% [B] at 8-10 ml / min. and a linear gradient of 2% [B] to 60% [B] was eluted in 174 minutes. The purified peptide was subjected to mass spectrometry, and analytical reverse phase HPLC to confirm its composition and purity. B. Incorporation in the Affinity Matrix All manipulations were carried out at room temperature. 12.5 ml of 80% of the NHS-Sepharose mixture (ie 10 ml of the packed volume, Pharmacia, Piscataway, NJ) was drained in a Bio-Rad EconoColumn (BioRad, Richmond, CA) and washed with 165 ml of HCl 1.0 mM cold as ice. When the bed was drained completely, the bottom of the column was closed and 5.0 ml of 7.0 mg / ml peptide P10-P4'sta (D? V) (dissolved in 0.1 M HEPES, pH 8.0) was added. The column was covered and incubated with rotation for 24 hours. After incubation, the column was allowed to drain, then washed with 8 ml of 1.0 M ethanolamine, pH 8.2. An additional 10 ml of the ethanolamine solution was added, and the column was covered again and incubated overnight with rotation. The bed of the columan was washed with 20 ml of 1.5 M sodium chloride, 0.5 M Tris, pH 7.5, followed by a series of regulators containing 0.1 mM EDTA, 0.2% Triton X-100 and the following components; 20 mM sodium acetate, pH 4.5 (100 ml); 20 mM sodium acetate pH 4.5, 1.0 M sodium chloride (100 ml); 20 mM sodium borate, pH 9.5, 1.0 M sodium chloride (200 ml); 20 mM sodium borate, pH 9.5, (100 ml). Finally, the column bed was washed with 15 ml of 2 mM Tris, 0.01% sodium azide (without Triton or EDTA), and stored in the regulator at 4 ° C. EXAMPLE 8 Co-Transfection of Cells with β-secretase and APP 293T cells were co-transfected with equivalent amounts of plasmids encoding the cDNA of APPsw or wt and β-secretase or β-galalactoside control (β-gal) using FuGENE Reagent 6 , as described in Example 4, above. We used clone 27 of pCEK or pohCJ containing full length β-secretase for the expression of β-secretase. The construction of plasmid pohCK751 used for the expression of APP in these transfections was derived as described in Dugan et al. , JBC, 270 (18) 10982-10989 (1995) and schematically shown in Figure 21. The β-gal control plasmid was added so that the total amount of transfected plasmid was the same for each condition. The pCEK and pohCK vectors expressing ß-gal do not replicate in 293T or COS cells. Triplicated cavities of cells were transfected with the plasmid, according to standard methods described above, then incubated for 48 hours, before collection of the conditioned medium and cells. The whole cell lysates were prepared and tested for the enzymatic activity of the β-secretase. The amount of β-secretase activity expressed by the transfected 293T cells was comparable to or greater than that expressed by the CosA2 cells used in the single transfection studies. Western blot analyzes were carried out in conditioned media and cell lysates, using the 13G8 antibody, and Aβ ELISAs carried out in the conditioned medium to analyze the various APP cleavage products. Although the invention has been decribed with reference to specific methods and modalities, it will be appreciated that various modifications and changes may be made without departing from the invention. All references to patents and literature referred to in the present are incorporated therein for reference.

Claims (115)

1. CLAIMS 1. A purified β-secretase enzyme protein for apparent homogeneity.
2. 2. The purified β-secretase enzyme protein of claim 1, wherein the enzyme has been sufficiently purified so that its activity by unfolding the 695 amino acid isotype of the β-amyloid precursor protein (β-APP) between the amino acids 596 and 597 thereof is at least 10,000 times greater than an activity exhibited by a solubilized but unenriched membrane fraction of human 293 cells.
3. 3. The purified β-secretase enzyme protein of claim 1, characterized by a specific activity of at least about 0.2xl05 nM / h / μg protein in a MBP-C125sw substrate assay.
4. 4. The purified β-secretase enzyme protein of claim 3, wherein the specific activity is at least 1.0x1O5 nM / h / μg protein.
5. 5. The purified β-secretase enzyme protein of claim 1, wherein the protein has less than 450 amino acids in length, comprising a polypeptide having the amino acid sequence SEQ ID NO: 70 [63-452].
6. 6. The purified protein of any of claims 1-5, wherein the protein consists of a polypeptide having the amino acid sequence SEQ ID NO: 70 [63-452].
7. 7. The purified protein of any of claims 1-5, wherein the protein consists of a polypeptide having the amino acid sequence SEQ ID NO: 69 [63-501].
8. 8. The purified protein of any of claims 1-5, wherein the protein consists of a polypeptide having the amino acid sequence SEQ ID NO: 67 [58-501].
9. 9. The purified protein of any of claims 1-5, wherein the protein consists of a polypeptide having the amino acid sequence SEQ ID NO: 68 [58-452].
10. 10. The purified protein of any of claims 1-5, wherein the protein comprises a polypeptide having the amino acid sequence SEQ ID NO: 58 [46-452]. The purified protein of claim 10, wherein the protein consists of a polypeptide having the amino acid sequence SEQ ID NO: 74 [22-452]. 12. The purified protein of claim 10, wherein the protein consists of a polypeptide having the amino acid sequence SEQ ID NO: 58 [46-452]. The purified protein of claim 10, wherein the protein is characterized by an N-terminus at position 46 with respect to SEQ ID NO: 2 and a C-terminus between positions 452 and 470 with respect to SEQ. ID NO: 2. The purified protein of claim 10, wherein the protein is characterized by an N-terminus at position 22 with respect to SEQ ID NO: 2 and a C-terminus between positions 452 and 470 with respect to SEQ ID NO: 2. 15. The purified protein of any of claims 1-5, wherein the protein consists of a polypeptide having the amino acid sequence SEQ ID NO: 43 [46-501]. 16. The purified protein of any of claims 1-5, wherein the protein consists of a polypeptide having the amino acid sequence SEQ ID NO: 66 [22-501]. 17. The purified protein of any of claims 1-5, wherein the protein consists of a polypeptide having the amino acid sequence SEQ ID NO: 2 [1-501]. 18. The purified protein of any of claims 1-5, wherein the protein has an N-terminal residue corresponding to a residue selected from the group consisting of residues 22, 46, 58 and 63 with respect to SEQ ID NO: 2 and a C-terminus selected from a residue between positions 452 and 501 with respect to SEQ ID NO: 2. 19. The purified protein of claim 18, wherein the C-terminus is between the positions of residue 452 and 470 with respect to SEQ ID NO: 2. The purified protein of claim 1, wherein the protein is isolated from a mouse. 21. The protein of claim 20 wherein the polypeptide has the sequence SEQ ID NO: 65. 22. The purified protein of any of claims 1-21 wherein the protein is produced by means of a heterologous cell. 23. A crystalline protein composition formed from a purified β-secretase protein. 24. The crystalline protein composition of claim 23, wherein the purified protein is characterized by a binding affinity for the β-secretase inhibitor substrate P10-P4 'staD- »V which is at least 1/100 of an affinity exhibited by a protein having the amino acid sequence SEQ ID NO: 43 [46-501], when the proteins are tested for binding to the substrate under the same conditions. 25. The crystalline protein composition of any of claims 23-24, wherein the composition is formed of a protein having a sequence selected from the group consisting of SEQ ID NO: 66 [22-501], SEQ ID NO. : 43 [46-501], SEQ ID NO: 74 [22-452], SEQ ID NO: 43 [46-452] and SEQ ID NO: 71 [46-419]. 26. The crystalline protein composition of any of claims 23-24, wherein the composition is formed of a protein having a sequence selected from the group consisting of SEQ ID NO: 2 [1-501], SEQ ID NO. : 59 [1-452], and SEQ ID NO: 60 [1-420]. 27. The crystalline protein composition of any of claims 23-24, wherein the composition is formed of a protein having an N-terminal residue corresponding to a residue selected from the group consisting of residues 22, 46, 58 and 63 with respect to SEQ ID NO: 2 and a C-terminal selected from a residue between positions 452 and 501 with respect to SEQ ID NO: 2. 28. The crystal protein of claim 27, wherein the C -terminification is between the positions of residues 452 and 470 with respect to SEQ ID NO: 2. 29. The crystalline protein composition of any of claims 23-28 wherein the protein is glycosylated. 30. The crystalline protein composition of any of claims 23-28 wherein the protein is deglycosylated. 31. The crystalline protein composition of any of claims 23-30 wherein the composition further includes an inhibitory or substrate molecule of β-secretase. 32. The crystalline protein composition of claim 31, wherein the β-secretase inhibitor is a peptide having less than about 15 amino acids and comprises the sequence SEQ ID NO: 78 (VMXVAEF; P3-P4'X D? V), including conservative substitutions thereof. 33. The crystalline protein composition of claim 31, wherein the β-secretase inhibitor has SEQ ID NO: 72 [P10-P4 'staD → V], including conservative substitutions thereof. 34. The crystalline protein composition of any of claims 31-35, wherein the β-secretase inhibitor has the sequence SEQ ID NO: 81 [EVMXVAEF], wherein X is hydroxyethylene or statin. 35. The crystalline protein composition of claim 31, wherein the β-secretase inhibitor is characterized by a Ki of no more than about 0.5mM 36. The crystalline protein composition of claim 31, wherein the β-secretase inhibitor it is characterized by a Ki of no more than about 5OμM. 37. An isolated protein comprising a polypeptide that (i) is less than about 450 amino acid residues in length, (ii) includes an amino acid sequence that is at least 90% identical to SEQ ID NO: 75 [63-423], including conservative substitutions thereof, and (iii) exhibits β-secretase activity, as evidenced by the ability to unfold a substrate selected from the group consisting of the 695 amino acid isotype of the beta amyloid precursor protein (βAPAP) between amino acids 596 and 597 of it, MBP-C125wt and MBP-C125sw. 38. The protein of claim 37, wherein the polypeptide includes the amino acid sequence of SEQ ID NO: 75 [63-423]. 39. The protein of claim 37, wherein the polypeptide has the sequence SEQ ID NO: 75 [63-423]. 40. The protein of claim 37, wherein the amino acid sequence is at least 95% identical to SEQ. ID NO: 58 [46-452]. 41. The protein of claim 40, wherein the polypeptide has the sequence SEQ ID NO: 58 [46-452]. 42. The protein of claim 37, wherein the protein consists of a polypeptide having the sequence SEQ ID NO: 58 [46-452]. 43. The protein of claim 37, wherein the protein consists of a polypeptide having the sequence SEQ ID NO: 74 [22-452]. 44. The protein of any of claims 37-43, wherein the protein is expressed by a heterologous cell. 45. A composition comprising the protein of any of claims 37-44 and a β-secretase substrate or inhibitory molecule. The composition of claim 45, wherein the β-secretase substrate is selected from the group consisting of MBP-C125wt, MBP-C125sw, APP, APPsw, and β-secretase cleavable fragments thereof. 47. The composition of claim 46 wherein the β-secretase cleavable fragments are selected from the group consisting of SEVKMDAEF (P5-P4 'wt), SEVNLDAEF (sw), SEVKLDAEF, SEVKFDAEF, SEVNFDAEF, SEVKMAAEF, SEVNLAAEF, SEVKLAAEF; SEVKMLAEF, SEVNLLAEF, SEVKLLAEF, SEVKFAAEF, SEVNFAAEF, SEVKFLAEF and SEVNFLAEF. 48. The composition of claim 45 wherein the β-secretase inhibitor in a peptide having less than about 15 amino acids and comprises the sequence SEQ ID NO: 78 (VM [X] VAEF, where X is hydroxyethylene or statin), including conservative substitutions of it. 49. The composition of claim 48 wherein the β-secretase inhibitor has the sequence SEQ ID NO: 81 (VM [X] VAEF, where X is hydroxyethylene or statin). 50. The composition of claim 45 wherein the β-secretase inhibitor has the sequence SEQ ID NO: 72 (P10-P4 'staD → V), including conservative substitutions thereof. 51. The composition of any of claims 45 and 48-50, wherein the β-secretase inhibitor has a Ki of no more than about 1 μM. 52. The composition of any of claims 45 and 48-50, wherein the β-secretase inhibitor is labeled with a detectable reporter molecule. 53. An enzyme of the β-secretase protein isolated from mouse having the sequence SEQ ID NO: 65. An antibody that specifically binds to a purified β-secretase protein comprising a polypeptide that includes an amino acid sequence that is less 90% identical to SEQ ID N0: 75 [63-423], including conservative substitutions thereof, wherein the antibody further lacks significant immunoreactivity with a protein of a sequence selected from the group consisting of SEQ ID N0: 2 [1-501] SEQ ID NO: 43 [46-501]. 55. The antibody of claim 54 wherein the antibody reacts with a protein selected from the group consisting of SEQ ID NO: 66 [22-501], SEQ ID NO: 67 [58-501], SEQ ID NO: 69 [ 63-501], SEQ ID NO: 59 [1-452], SEQ ID NO: 74 [22-452], SEQ ID NO: 58 [46-452], SEQ ID NO: 68 [58-452], and SEQ ID NO: 70 [63-452]. 56. An isolated nucleic acid comprising a nucleotide sequence encoding a β-secretase protein that is at least 95% identical to a protein selected from the group consisting of SEQ ID NO: 66 [22-501], SEQ ID NO: 43 [46-501], SEQ ID NO: 57 [1-419], SEQ ID NO: 74 [22-452], SEQ ID NO: 58 [46-452], SEQ ID NO: 59 [1-452], SEQ ID NO: 60 [1-420], SEQ ID NO: 67 [58-501], SEQ ID NO: 68 [58-452], SEQ ID NO: 69 [63-501], SEQ ID NO: 70 [63-452], SEQ ID N0: 75 [63-423], and SEQ ID NO: 71 [46-419], or a complemantary sequence of any such nucleotide and specifically excluding a nucleic acid which encodes a protein having the sequence SEQ ID NO: 2 [1-501]. 57. The isolated nucleic acid of claim 56, wherein the nucleotide sequence encodes a protease having an amino acid sequence SEQ ID NO: 58 [46-452]. 58. The isolated nucleic acid of claim 56, wherein the nucleotide sequence encodes a protease having a sequence SEQ ID NO: 43 [46-501]. 59. The isolated nucleic acid of claim 56, wherein the nucleotide sequence encodes a protease having a sequence SEQ ID NO: 66 [22-501]. 60. The isolated nucleic acid of claim 56, wherein the nucleotide sequence encodes a protease having a sequence SEQ ID NO: 74 [22-452]. 61. An expression vector, comprising the nucleic acid isolated from any of claims 56-60, and operably linked to the nucleic acid, the regulatory sequence is effective for the expression of the nucleic acid in a selected host cell. 62. The recombinant expression vector of claim 61, wherein the vector is suitable for transfection of a bacterial cell. 63. A heterologous cell transfected with the vector of any of claims 61-62, wherein the cells express a biologically active β-secretase. 64. The cell of claim 63, wherein the cell is a eukaryotic cell. 65. The cell of claim 63, wherein the cell is a bacterial cell. 66. The cell of claim 63, wherein the cell is an insect cell. 67. The cell of claim 63, wherein the cell is a yeast cell. 68. A method for producing a recombinant β-secretase enzyme, comprising culturing a cell according to any of claims 63-67 under conditions to promote cell growth, and subjecting an extract or culture medium of the cell to an affinity matrix. 69. The method of claim 68, wherein the affinity matrix contains a β-secretase inhibitory molecule. 70. The method of claim 69, wherein the inhibitory molecule is SEQ ID NO: 72 [P10-P4 'staD? V]. 71. The method of claim 68, wherein the matrix contains an antibody characterized by a binding capacity to β-secretase. 72. The method of claim 71, wherein the antibody is according to any of claims 54-55. 73. A heterologous cell comprising (i) a nucleic acid molecule encoding a β-secretase protein according to any of claims 37-43; (ii) a nucleic acid molecule encoding a substrate molecule of β-secretase; and (iii) a regulatory sequence operably linked to (i) and (ii), effective for the expression of the nucleic acid molecules in the cell. 74. The cell of claim 73, wherein the nucleic acid encoding the β-secretase protein is heterologous to the cell. 75. The cell of claim 73, wherein both nucleic acids encoding the β-secretase protein and the one encoding the β-secretase substrate molecule, are heterologous to the cell. 76. The cell of claim 73, wherein the β-secretase substrate molecule is selected from the group consisting of MBP-C125wt, MBP-C125sw, APPwt, APPsw and cleavable β-secretase fragments thereof. 77. The cell of claim 76 wherein the cleavable β-secretase fragments are selected from the group consisting of SEVKMDAEF (P5-P4 'wt), SEVNLDAEF (sw), SEVKLDAEF, SEVKFDAEF, SEVNFDAEF, SEVKMAAEF, SEVNLAAEF, SEVKLAAEF; SEVKMLAEF, SEVNLLAEF, SEVKLLAEF, SEVKFAAEF, SEVNFAAEF, SEVKFLAEF and SEVNFLAEF. 78. A classification method for compounds that inhibit Aß production, which comprises contacting an α-secretase-ailated polypeptide in accordance with claim 37 with (i) a test compound, and (ii) a β-substrate. secretase and selecting the test compound as capable of inhibiting Aß production if the β-secretase polypeptide exhibits less β-secretase activity in the presence of the compound than in the absence of the compound. 79. The method of claim 78, wherein the active β-secretase polypeptide has a sequence selected from the group consisting of SEQ ID NO: 43 [46-501] and SEQ ID NO: 58 [46-452]. 80. The method of claim 78, wherein the β-secretase polypeptide and the substrate are produced by a cell according to claim 73. The method of claim 78, further including supplying the test compound to a mammalian subject having Alzheimer's disease or pathology similar to Alzheimer's disease and selecting the compound as a therapeutic agent candidate if, after such administration, the subject maintains or improves the cognitive ability or the subject shows reduced plaque burden. 82. The method of claim 81, wherein the subject is a mammalian species comprising a transgene. 83. The method of claim 81, wherein the subject is a mouse carrying a transgene encoding a human β-amyloid precursor protein (β-APP), including a mutant variant thereof. 84. The method of any of claims 78-83, wherein the β-secretase substrate is selected from the group consisting of MBP-C125wt, MBP-C125sw, APP, APPsw, and cleavable β-secretase fragments thereof. . 85. The method of claim 78 wherein the divisible fragment of β-secretase is selected from the group consisting of SEVKMDAEF (P5-P4 'wt), SEVNLDAEF (sw), SEVKLDAEF, SEVKFDAEF, SEVNFDAEF, SEVKMAAEF, SEVNLAAEF, SEVKLAAEF; SEVKMLAEF, SEVNLLAEF, SEVKLLAEF, SEVKFAAEF, SEVNFAAEF, SEVKFLAEF and SEVNFLAEF. 86. A method of classifying compounds that inhibit the production of Aβ, which comprises measuring the binding of a purified β-secretase polypeptide according to any of claims 1-22 or 37-43, with a β-inhibitor compound. secretase in the presence of a test compound and selecting the test compound as the active site binding compound of β-secretase, if the binding of the inhibitor in the presence of the test compound is less than the binding of the inhibitor in the absence of the test compound. 87. The method of claim 86, wherein the inhibitor compound is labeled with a detectable label. 88. The method of claim 86, wherein the β-secretase inhibitor is a peptide having less than about 15 amino acids and comprises the sequence SEQ ID NO: 78 (VM [X] VAEF, wherein X is hydroxyethylene or statin ), including conservative substitutions thereof. 89. The method of claim 86, wherein the β-secretase inhibitor has the sequence SEQ ID NO: 72 (P10-P4 'staD- »V), including conservative substitutions thereof. 90. The method of claim 86, wherein the β-secretase inhibitor has a Ki relative to the β-secretase of less than about 50 μM. 91. A β-secretase inhibitor compound selected according to the method of any of claims 78-80. 92. The inhibitor of claim 91, wherein the compound is selected from a phage display selection system. 93. The compound of claim 92, wherein the phage display selection system is derived from the sequence SEQ ID NO: 72 [P10-P4 'D? V]. A β-secretase inhibitor compound selected according to the method of any of claims 81-90. 95. The inhibitor of claim 94, wherein the compound is selected from a phage display selection system. 96. The compound of claim 95, wherein the phage display selection system is derived from the sequence SEQ ID NO: 72 [P10-P4 'D- »V]. 97. A β-secretase inhibitor compound selected according to the method of any of claims 86-90. 98. The inhibitor of claim 97, wherein the compound is selected from a phage display selection system. 99. The compound of claim 98, wherein the phage display selection system is derived from the sequence SEQ ID NO: 72 [P10-P4 'D? V]. 100. A β-secretase inhibitor, comprising a peptide containing the sequence SEQ ID NO: 78 (VM [X] VAEF, wherein X is hydroxyethylene or statin), including conservative substitutions thereof. 101. The β-secretase inhibitor of claim 100, having the sequence SEQ ID NO: 72 (P10-P4'staD? V). 102. The β-secretase inhibitor of claim 100, having the sequence SEQ ID NO: 78. 103. The β-secretase inhibitor of claim 100, having the sequence SEQ ID NO: 81. 104. A kit of classification, comprising an isolated β-secretase protein according to any of claims 1-22 or 37-43, a divisible β-secretase substrate, and means for detecting the cleavage of the substrate by β-secretase. 105. The sorting kit of claim 104, wherein the β-secretase protein is present in a heterologous cell. 106. The sorting kit of claim 104, wherein the substrate molecule of β-secretase is selected from the group consisting of MBP-C125wt, MBP-C125sw, APPwt, APPsw, and cleavable fragments of β-secretase from the same. 107. The sorting kit of claim 106 wherein the cleavable fragment of β-secretase is selected from the group consisting of SEVKMDAEF (P5-P4 'wt), SEVNLDAEF (sw), SEVKLDAEF, SEVKFDAEF, SEVNFDAEF, SEVKMAAEF, SEVNLAAEF, SEVKLAAEF; SEVKMLAEF, SEVNLLAEF, SEVKLLAEF, SEVKFAAEF, SEVNFAAEF, SEVKFLAEF and SEVNFLAEF. 108. A knock-out mouse, characterized by the inactivation or cancellation of an endogenous β-secretase gene. 109. The knock-out mouse of claim 108, wherein the β-secretase gene encodes a protein having at least 90% sequence identity to the sequence SEQ ID NO: 65. 110. The knock-out mouse of Claim 108, wherein the cancellation is inducible. 111. The knock-out mouse of claim 110, wherein the inducible expression is effected by a Cre-lox expression system inserted into the mouse genome. 112. A classification method for drugs effective in the treatment of Alzheimer's disease or other cerebrovascular amyloidoses characterized by Aβ deposition, which comprises supplying a mammalian subject characterized by over-expression of β-APP and / or deposition of Aβ a test compound selected for its ability to inhibit the β-secretase activity of a β-secretase protein according to any of claims 37-43, and to select the compound as a potential therapeutic drug compound, if reduced the amount of Aß deposition in the subject or whether it maintains or improves the cognitive capacity in the subject. 113. The method of claim 112, wherein the mammalian subject is a transgenic mouse carrying a transgene encoding a human β-APP or a mutant thereof. 114. A method for treating a patient affected or having a propensity for Alzheimer's disease or other cerebrovascular amyloidosis, which comprises blocking the enzymatic hydrolysis of APP to Aβ in the patient by administering to the patient a pharmaceutically effective dose of an effective compound for inhibiting a protein of the β-secretase enzyme according to any of claims 37-43. 115. The method of claim 114, wherein the compound is derived from a peptide selected from the group consisting of SEQ ID NO: 72, SEQ ID NO: 78, SEQ ID NO: 81 and SEQ ID NO: 97.