WO2002101232A2

WO2002101232A2 - Bace interacting proteins

Info

Publication number: WO2002101232A2
Application number: PCT/EP2002/006667
Authority: WO
Inventors: Els Marjaux; Diana Ines Dominguez-Tinland; Wim Annaert; Bart De Strooper
Original assignee: Vlaams Interuniversitair Instituut Voor Biotechnologie Vzw
Priority date: 2001-06-12
Filing date: 2002-06-07
Publication date: 2002-12-19
Also published as: AU2002345051A1; WO2002101232A3

Abstract

The present invention relates to the field of Alzheimer's disease. The invention describes the identification, isolation, sequencing of several genes which encode for proteins that interact with b-secretase (BACE). These BACE-interacting proteins may be involved in pathways that lead to the development of Alzheimer's disease. Nucleic acids and proteins comprising or derived from these BACE-interacting proteins are useful in screening and diagnosing Alzheimer's disease, in identifying and developing therapeutics for the treatment of Alzheimer's disease, and in producing cell lines and transgenic animals useful as models for Alzheimer's disease.

Description

BACE interacting proteins

Field of the invention

The present invention relates to the field of Alzheimer's disease. The invention describes the identification, isolation, sequencing of several genes that code for proteins that interact with β-secretase (BACE). These BACE-interacting proteins are involved in pathways that lead to the development of Alzheimer's disease. Nucleic acids and proteins comprising or derived from these BACE-interacting proteins are useful in screening and diagnosing Alzheimer's disease, in identifying and developing therapeutics for the treatment of Alzheimer's disease, and in producing cell lines and transgenic animals useful as models for Alzheimer's disease.

Background of the invention Alzheimer's disease (AD) is characterized by a variety of neuropathological features, such as synaptic loss and neuronal cell death. However, the histopathological hallmark of AD is the presence of neurofibrillary tangles and amyloid plaques throughout the hippocampus and cerebral neocortex. The amyloid plaques are complex cellular lesions with a protein core that consists mainly of the 40-42 amino acid residue containing amyloid β peptide (Aβ). The formation of Aβ requires two consecutive cleavages of a type I transmembrane glycoprotein, the amyloid β precursor protein (APP). Cleavage at the aminoterminus of Aβ is performed by β-secretase. β-secretase then cleaves within the transmembrane domain of the C-terminal membrane bound APP fragment to release Aβ. Recently a new aspartic protease named BACE (β-site APP cleaving enzyme) was identified as the major β-secretase. BACE is a type I integral membrane protein that contains the two active site motifs characteristic of aspartic proteases. BACE fulfils most of the criteria expected for a true β-secretase candidate. It has a broad tissue distribution and furthermore it is enriched in the brain. In the cell small amounts can be detected in the ER and in lysosomes, but BACE is mainly localized in Golgi and endosomes. The Golgi apparatus and endosomes offer BACE the slightly acidic environment necessary to obtain its optimal activity. Overexpression of BACE in cells leads to an increased cleavage of APP at the β-secretase site. Cells treated with antisense oligonucleotides complementary to BACE mRNA have decreased Aβ production. KO mice deficient in BACE were previously generated. These mice do not exhibit any obvious physical or behavioural difference compared to wild type mice but their β-secretase activity is abolished. The results suggest that BACE has no vital function for which no redundancy exists. It is currently not understood how BACE is transported to its intracellular localisation and how the activity of BACE is modulated in the cell. Possibly the transport and activity of BACE is regulated by proteins that interact with BACE. However, in the art there are no reports of such proteins. The present invention provides a solution to this problem by the identification and characterization of several proteins that interact with BACE. Said proteins are modulators of BACE activity and are therapeutic targets for the treatment of AD. Furthermore mutations in said proteins are causative of Alzheimer's disease.

Aims and detailed description of the invention Alzheimer's disease is a degenerative disorder of the human central nervous system characterized by progressive memory impairment and cognitive and intellectual decline during mid to late adult life. The disease is accompanied by a constellation of neuro- pathological features principal amongst which are the presence of extracellular amyloid or senile plaques, and neurofibrillary tangles in neurons. Recently BACE has been suggested to be an important target for interfering with the formation of extracellular amyloid. In the present invention it is shown that in non-neuronal cells a BACE derivative that lacks the cytoplasmic domains is as active as wild type BACE in cleaving APP. However, surprisingly, in neurons the same BACE mutant showed almost no beta-secretase activity, suggesting that the cytoplasmic domain of BACE is required to get full enzymatic activity in neurons. Full BACE activity in neurons can therefore only be obtained through certain proteins that interact with the cytoplasmic domain of BACE. The present invention provides isolated nucleic acid sequences encoding domains or functional fragment thereof that are capable to interact with the transmembrane or cytoplasmic domain of BACE in vivo. Said peptides or protein fragments are further defined herein as BACE-interacting proteins while said nucleic sequences encoding BACE-interacting proteins are further defined as BACE- interacting genes. Thus, in the present invention a number of murine brain proteins that interact with murine BACE have been identified by using a yeast two-hybrid system. Several binders were isolated out of a murine hippocampus cDNA library that specifically interact with the transmembrane domain and cytoplasmic domain of murine BACE. Thus in a first embodiment the invention provides a BACE-interacting polypeptide comprising a BACE interacting domain capable of interacting with the transmembrane and cytoplasmic domain of BACE. A 'domain' is defined as a specific isolated region of a protein, here a region of a BACE-interacting protein that interacts with BACE. By employing other methods of identifying BACE-interacting proteins, as disclosed below and known in the art, or employing cDNA libraries from other tissues or species, it is clear that one can now identify and isolate a variety of nucleic acids, for example from other species, encoding BACE-interacting proteins. Once identified, these sequences can be used to clone larger cDNAs or genomic fragments (including entire genes which include BACE-interacting functional domains) or can be used to identify smaller, minimally active fragments that retain BACE-interacting activity (e.g., by iteratively deleting residues from the ends of BACE-interacting peptides and testing for retention of activity). In addition, BACE-interacting peptides or proteins can be identified which interact with specific functional domains of BACE or which interact specifically with mutant or normal forms. In another embodiment the invention provides BACE-interacting proteins selected from the group consisting of sequences comprising SEQ ID NO: 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38 or 40 and homologues thereof. By 'homologues' it is meant proteins or protein fragments that have a homology of at least 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95% or even more homology with sequences comprising SEQ ID NO: 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38 or 40. Homology is typically measured using sequence analysis software (e.g., Sequence Analysis Software Package of the Genetics Computer Group, University of Wisconsin Biotechnology Center, 1710 University Avenue, Madison, Wis. 53705). Such software matches similar sequences by assigning degrees of homology to various insertions, deletions, substitutions, and other modifications. Conservative substitutions typically include substitutions within the following groups: glycine, alanine; valine, isoleucine, leucine; aspartic acid, glutamic acid, asparagine, glutamine; serine, threonine; lysine, arginine; and phenylalanine, tyrosine. In what follows, a short description of the isolated BACE-interacting proteins of the invention is presented. SEQ ID NO: 2 represents a domain of the BAT3 protein that interacts with BACE. The accession numbers in GenBank for BAT3 are AF109719 (mouse) and XM_004176 (human). The identity between the isolated domain of SEQ ID NO: 2 and its human counterpart is 100%. SEQ ID NO: 4 represents a domain of the NIX (Bnip3l) that interacts with BACE. NIX (Bnip3l) is a mitochondrial protein being a member of a subfamily of pro-apoptotic proteins. The accession numbers in GenBank for NIX (Bnip3l) are AF067395, NM_009761 (mouse) and AF067396, AF255051 (human). The identity between the isolated domain of SEQ ID NO: 4 and its human counterpart is 98%. SEQ ID NO: 6 represents a domain of the neutral sphingomyelinase (nSmase) that interacts with BACE. nSMase is a protein localized in the endoplasmic reticulum. It has been shown that this enzyme has lyso-platelet activating factor (PAF) phospholipase C activity. The accession numbers in GenBank for nSMase are NM_009213, AJ222800 (mouse) and AJ222801 , NM_003080, XM_004489, BC000038 (human). The identity between the isolated domain of SEQ ID NO: 6 and its human counterpart is 87%. SEQ ID NO: 8 represents a domain of the arsenite translocating ATPase that interacts with BACE. Arsenite translocating ATPase is an efflux pump that can confer resistance to arsenite and antimonite by their extrusion from the cells. The accession numbers for the complete proteins in GenBank for arsenite transporter are NM_019652, AF039405 (mouse) and AF047469, BC002651 (human). The identity between the isolated domain of SEQ ID NO: 8 and its human counterpart is 85%. SEQ ID NO: 10 represents a domain of the A1U that interacts with BACE. A1 U is a protein that interacts with ataxin-1 , the latter known to be involved in the pathology of spinocerebellar ataxia type 1. A1U is widely expressed in Purkinje cells. The accession numbers for the complete proteins in GenBank for A1 U are AB040050 (mouse) and XM 002073, NM_020131 , AF188240 (human). The identity between the isolated domain of SEQ ID NO: 10 and its human counterpart is 100%. SEQ ID NO: 12 represents a domain of the cytochrome oxidase subunit 2 that interacts with BACE. Cytochrome oxidase is an inner mitochondrial protein involved in ATP synthesis. The accession numbers for the complete proteins in GenBank for cytochrome oxidase are AB042432 (mouse) and AAB58946 (human). The identity between the isolated domain of SEQ ID NO: 12 and its human counterpart is 67%. SEQ ID NO: 14 represents a domain of the GL004 that interacts with BACE. GL004 is probably a membrane bound protein. The accession numbers for the complete proteins in GenBank for GL004 are AK017226 (mouse) and AF226049 (human). The identity between the isolated domain of SEQ ID NO: 14 and its human counterpart is 94%. SEQ ID NO: 16 represents a domain of the Gpr 37-like protein 1 that interacts with BACE. Gpr 37-like protein 1 is probably a multi-transmembrane protein. The accession numbers for the complete proteins in GenBank for Gpr 37-like protein 1 are AJ306537 (mouse) and NM_004767, Y16280 (human). The identity between the isolated domain of SEQ ID NO: 16 and its human counterpart is 100%. SEQ ID NO: 18 represents a domain of the lifeguard protein (NMP35/LFG) that interacts with BACE. Lifeguard was isolated as an anti-apoptotic gene that provides protection from fas- mediated cell death. Lifeguard is probably also involved in neuronal development. The accession numbers for the complete proteins in GenBank for NMP35/LFG are AK013476 (mouse) and AF190461 , XM_012187 (human). The identity between the isolated domain of SEQ ID NO: 18 and its human counterpart is 88%. SEQ ID NO: 20 represents a domain of the Otx-2 that interacts with BACE. Otx-2 is a homeobox transcription factor that is involved in anterior embryonic pattern formation. The accession numbers for the complete proteins in GenBank for Otx-2 are X68884 (mouse) and XM_012334, NM_021728 (human). The identity between the isolated domain of SEQ ID NO: 20 and its human counterpart is 93%. SEQ ID NO: 22 represents a domain of the Tip20p that interacts with BACE. Tip20p is a protein involved in the transport between endoplasmic reticulum and Golgi. The accession numbers for the complete proteins in GenBank for Tip20p are unknown (mouse) and AF000560 (human). The identity between the isolated domain of SEQ ID NO: 22 and its human counterpart is 96%. SEQ ID NO: 24 represents a domain of the s-rex/ NSP- C /Reticulon-1 that interacts with BACE. s-rex/ NSP-C /Reticulon-1 is found in neural and neuroendocrine cells and is associated with the endoplasmic reticulum. Its expression is found to be correlated with the degree of neuronal differentiation. Strikingly, the expression of NSP-C is reduced in the brains of patients with Down syndrome and Alzheimer's disease. The accession numbers for the complete proteins in GenBank for s-rex/ NSP-C /Reticulon-1 are unknown (mouse) and Q16799 (human). The identity between the isolated domain of SEQ ID NO: 24 and its human counterpart is 61%. SEQ ID NO: 26 represents a domain of the NMDAR1 subunit isoform 4beta that interacts with BACE. NMDAR1 subunit isoform 4beta is a protein involved in neuronal development and plasticity. The accession numbers for the complete proteins in GenBank for NMDAR1 subunit isoform 4beta are unknown (mouse) and AF015731 (human). The identity between the isolated domain of SEQ ID NO: 26 and its human counterpart is 100%. SEQ ID NO: 28 represents a domain of the neuron specific enolase 2 that interacts with BACE. Neuron specific enolase 2 is a protein involved in neuronal differentiation. The accession numbers for the complete proteins in GenBank for neuron specific enolase 2 are NM_013509 (mouse) and XM_006974 (human). The identity between the isolated domain of SEQ ID NO: 28 and its human counterpart is 99.1%. SEQ ID NO: 30 represents a domain of the hypothetical protein FLJ22056 that interacts with BACE. The accession numbers for the complete proteins in GenBank for hypothetical protein FLJ22056 are AK011884 (mouse) and XM_015185 (human). The identity between the isolated domain of SEQ ID NO: 30 and its human counterpart is 61 %. SEQ ID NO: 32 represents a domain of the FKBP38 that interacts with BACE. FKBP38 is a protein involved in protein folding and stabilization of multiprotein complexes. The accession numbers for the complete proteins in GenBank for FKBP38 are BC003739 (mouse) and L37033 (human). The identity between the isolated domain of SEQ ID NO: 32 and its human counterpart is 91%. SEQ ID NO: 34 represents a domain of the Bcl-rambo that interacts with BACE. Bcl-rambo is a pro- apoptotic protein localized in the outer mitochondrial membrane. The accession numbers for the complete proteins in GenBank for Bcl-rambo are not known (mouse) and AF325209 (human). The identity between the isolated domain of SEQ ID NO: 34 and its human counterpart is 97%. SEQ ID NO: 36 represents a domain of the Phogrin/ PTP-NP/ IA2beta/PTP-IAR that interacts with BACE. Phogrin is an integral membrane protein localized to dense-core secretory granules of neuroendocrine cells (mainly pancreatic beta-cells) and is probably involved in signal transduction. The accession numbers for the complete proteins in GenBank for Phogrin are U57345 (mouse) and AF007555 (human). The identity between the isolated domain of SEQ ID NO: 36 and its human counterpart is 56%. SEQ ID NO: 38 represents a domain of the PIK4Kbeta that interacts with BACE. PIK4Kbeta is a protein involved in signalling and membrane traffic. The accession numbers for the complete proteins in GenBank for PIK4Kbeta are not known (mouse) and BC000029 (human). The identity between the isolated domain of SEQ ID NO: 38 and its human counterpart is 100%. SEQ ID NO: 40 represents a domain of the hypothetical protein FLJ20445 that interacts with BACE. The accession numbers for the complete proteins in GenBank for FLJ20445 are AK009364 (mouse) and AK000452 (human). The identity between the isolated domain of SEQ ID NO: 40 and its human counterpart is 59%.

The terms 'identical' or percent 'identity' in the context of two or more nucleic acids or polypeptide sequences, refer to two or more sequences or subsequences that are the same or have a specified percentage of amino acid residues or nucleotides that are the same (i.e. 70% identity over a specified region of a particular BACE-interacting protein), when compared and aligned for maximum correspondence over a comparison window, or designated region as measured using sequence comparison algorithms or by manual alignment and visual inspection. Preferably, the identity exists over a region that is at least about 25 amino acids or nucleotides in length, or more preferably over a region that is 50-100 amino acids or nucleotides or even more in length. Examples of useful algorithms are PILEUP (Higgins & Sharp, CABIOS 5:151 (1989), BLAST and BLAST 2.0 (Altschul et al. J. Mol. Biol. 215: 403 (1990). Software for performing BLAST analyses is publicly available through the National Center for Biotechnology Information (http://www/ncbi.nlm.nih.gov/).

In another embodiment the invention provides an isolated nucleic acid encoding for a BACE-interacting protein comprising a BACE interacting domain. In another embodiment the invention provides an isolated nucleic acid encoding for a BACE-interacting protein (polypeptide) comprising a BACE interacting domain selected from the group consisting of sequences comprising SEQ ID NO: 1 , 3, 5, 7, 9, 11 , 13, 15, 17, 19, 21 , 23, 25, 27, 29, 31 , 33, 35, 37, 39 and homologues thereof. By 'homologues' it is meant nucleic acid sequences that have a homology of at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95% or even more homology with SEQ ID NO: 1 , 3, 5, 7, 9, 11 , 13, 15, 17, 19, 21 , 23, 25, 27, 29, 31 , 33, 35, 37 or 39.

Thus the present invention provides isolated nucleic acids corresponding to, or relating to, nucleic acid sequences encoding BACE-interacting proteins disclosed herein. These sequences include normal BACE-interacting gene sequences from humans and other mammalian species, mutant BACE interacting gene sequences from humans and other mammalian species, homologous sequences from non-mammalian species such as for example Drosophila and C. elegans, subsets of these sequences useful as probes and PCR primers, subsets of these sequences encoding fragments of BACE interacting proteins or corresponding to particular structural domains or polymorphic regions, complementary or antisense sequences corresponding to fragments of the BACE interacting genes, sequences in which the BACE interacting coding regions have been operably joined to exogenous regulatory regions, and sequences encoding fusion proteins of the portions of the BACE interacting proteins fused to other proteins useful as markers of expression, as "tags" for purification, or in screens and assays for proteins interacting with BACE. In addition to the disclosed BACE-interacting gene sequences, one of ordinary skill in the art can now identify and isolate nucleic acids representing BACE-interacting genes or cDNAs which are allelic to the disclosed sequences or which are heterospecific (BACE-interacting genes derived from other species) homologues. Thus, the present invention provides isolated nucleic acids corresponding to these alleles and homologues by means that are well known in the art. Briefly, one of ordinary skill in the art can now screen preparations of genomic or cDNA, including samples prepared from individual organisms (e.g., human AD patients or their family members) as well as bacterial, viral, yeast or other libraries of genomic or cDNA, using probes or PCR primers to identify allelic or homologous sequences. Because it is desirable to identify BACE-interacting gene mutations which can contribute to the development of AD or other disorders, because it is desirable to identify BACE-interacting polymorphisms which are not pathogenic, and because it is also desired to create a variety of animal models which can be used to study AD and screen for potential therapeutics, it is particularly contemplated that BACE-interacting sequences will be isolated from other preparations or libraries of human nucleic acids and from preparations or libraries from animals including rats, hamsters, guinea pigs, rabbits, dogs, cats, goats, sheep, pigs, and non-human primates. Furthermore, BACE- interacting gene homologues from for example invertebrate species, including C. elegans and other nematodes, as well as Drosophila and other insects, can have particular utility for drug screening. For example, invertebrates bearing mutant BACE- interacting gene homologues which cause a rapidly occurring and easily scored phenotype (e.g., abnormal development of specific organs (e.g. wings, eyes, vulva) after several days) can be used as screens for drugs which block the effect of the mutant gene. Such invertebrates can prove far more rapid and efficient for mass screenings than larger vertebrate animals. Once lead compounds are found through such screens, they can be tested in higher animals. Standard hybridization screening or PCR techniques can be employed to identify and/or isolate such allelic and homologous sequences using relatively short BACE-interacting gene sequences. The sequences may include 8 or fewer nucleotides depending upon the nature of the target sequences, the method employed, and the specificity required. Future technological developments can allow the advantageous use of even shorter sequences. With current technology, sequences of 9-50 nucleotides, and preferably about 18-24 are preferred. These sequences can be chosen from those disclosed herein, or can be derived from other allelic or heterospecific homologues. When probing mRNA or screening cDNA libraries, probes and primers from coding sequences (rather than introns) are preferably employed, and sequences which are omitted in alternative splice variants typically are avoided unless it is specifically desired to identify those variants. Allelic variants of the BACE-interacting genes may be expected to hybridize to the disclosed sequences under stringent hybridization conditions, as defined herein, whereas lower stringency may be employed to identify heterospecific homologues. 'Mutant' as used herein refers to a gene that encodes a mutant protein. With respect to proteins, the term "mutant" means a protein which does not perform its usual or normal physiological role and which is associated with, or causative of, a pathogenic condition or state. Therefore, as used herein, the term "mutant" is essentially synonymous with the terms "dysfunctional," "pathogenic," "disease-causing," and "deleterious." With respect to the genes encoding BACE and BACE-interacting proteins of the present invention, the term "mutant" refers to genes encoding BACE and BACE-interacting proteins, bearing one or more nucleotide/amino acid substitutions, insertions and/or deletions which for example can lead to the development of the symptoms of Alzheimer's when expressed in humans. This definition is understood to include the various mutations that naturally exist, including but not limited to those disclosed herein, as well as synthetic or recombinant mutations produced by human intervention. The term "mutant," as applied to genes encoding BACE and BACE-interacting proteins, is not intended to embrace sequence variants which, due to the degeneracy of the genetic code, encode proteins identical to the normal sequences disclosed or otherwise presented herein; nor is it intended to embrace sequence variants which, although they encode different proteins, encode proteins which are functionally equivalent to normal BACE and BACE-interacting proteins.

In another embodiment the isolated nucleic acids of the present invention include any of the above described sequences or fragments thereof when included in vectors. Appropriate vectors include cloning vectors and expression vectors of all types, including plasmids, phagemids, cosmids, episomes, and the like, as well as integration vectors. The vectors may also include various marker genes (e.g., antibiotic resistance or susceptibility genes) that are useful in identifying cells successfully transformed therewith. In addition, the vectors may include regulatory sequences to which the nucleic acids of the invention are operably joined, and/or may also include coding regions such that the nucleic acids of the invention, when appropriately ligated into the vector, are expressed as fusion proteins. Such vectors may also include vectors for use in yeast "two hybrid," baculovirus, and phage-display systems. The vectors may be chosen to be useful for prokaryotic, eukaryotic or viral expression, as needed or desired for the particular application. For example, vaccinia virus vectors or simian virus vectors with the SV40 promoter (e.g., pSV2), or Herpes simplex virus or adeno- associated virus may be useful for transfection of mammalian cells including neurons in culture or in vivo, and the baculovirus vectors may be used in transfecting insect cells (e.g., butterfly cells). A great variety of different vectors are now commercially available and otherwise known in the art, and the choice of an appropriate vector is within the ability and discretion of one of ordinary skill in the art.

In another embodiment, the present invention provides assays for identifying compounds that are capable of inducing or inhibiting the expression of the BACE interacting genes and proteins. A method for identifying compounds which can modulate the expression of a gene encoding a BACE-interacting protein comprises the steps of (1) contacting a cell with at least one compound wherein said cell includes a regulatory region of a gene encoding a BACE-interacting protein operably joined to a coding region, and (2) detecting a change in expression of said region. 'Compound' means any anorganic or organic compound, including simple or complex inorganic or organic molecules, peptides, peptido-mimetics, proteins, antibodies, carbohydrates, nucleic acids or derivatives thereof. Candidate/test compounds such as small molecules, e.g. small organic molecules, and other drug candidates can be obtained, for example, from combinatorial and natural product libraries as described above. "Inhibitors," "activators" of BACE-interacting proteins refer to inhibitory or activating molecules identified using in vitro and in vivo assays for BACE-interacting protein function. As such a "modulator" can be an inhibitor or an activator. To examine the extent of inhibition, samples or assays comprising said BACE-interacting proteins are treated with a potential inhibitor and are compared to control samples without the inhibitor. Control samples (untreated with inhibitors) are assigned a relative BACE- interacting protein value of 100%. Inhibition of BACE-interacting proteins is achieved when said BACE-interacting protein activity value relative to the control is about 90%, preferably 50%, more preferably 25-0%. Activation of BACE-interacting proteins is achieved when the BACE-interacting protein activity value relative to the control is 110%, more preferably 150%, most preferably at least 200-500% higher or 1000% or even higher. The wording Operably joined' can be explained as follows. A coding sequence and a regulatory region are said to be operably joined when they are covalently linked in such a way as to place the expression or transcription of the coding sequence under the influence or control of the regulatory region. If it is desired that the coding sequences be translated into a functional protein, two DNA sequences are said to be operably joined if induction of promoter function results in the transcription of the coding sequence and if the nature of the linkage between the two DNA sequences does not (1) result in the introduction of a frame-shift mutation, (2) interfere with the ability of the regulatory region to direct the transcription of the coding sequences, or (3) interfere with the ability of the corresponding RNA transcript to be translated into a protein. Thus, a regulatory region would be operably joined to a coding sequence if the regulatory region were capable of effecting transcription of that DNA sequence such that the resulting transcript might be translated into the desired protein or polypeptide. the method, also called assay, may be performed in vitro using nontransformed cells, immortalized cell lines, or recombinant cell lines, or in vivo using transgenic animal models. In particular, the assays may detect the presence of increased or decreased expression of BACE interacting genes or proteins on the basis of increased or decreased mRNA expression (using, e.g., the nucleic acid probes disclosed herein), increased or decreased levels of BACE interacting protein products or increased or decreased levels of expression of a marker gene (e.g., beta-galactosidase or luciferase) operably joined to a 5' regulatory region of a BACE interacting gene in a recombinant construct. Thus, for example, one may culture cells known to express a particular BACE interacting protein and add to the culture medium one or more test compounds. After allowing a sufficient period of time (e.g., 0-72 hours) for the compound to induce or inhibit the expression of the BACE interacting gene, any change in levels of expression from an established baseline may be detected using any of the techniques described above and well known in the art. In particularly preferred embodiments, the cells are from an immortalized cell line such as a human neuroblastoma, glioblastoma or a hybridoma cell line. Using the nucleic acid probes disclosed herein, detection of changes in the expression of a BACE interacting gene, and thus identification of the compound as an inducer or repressor of BACE interacting gene expression, requires only routine experimentation. In a preferred embodiment, a recombinant assay is employed in which a reporter gene such as for example a beta- galactosidase, green fluorescent protein, alkaline phosphatase, or luciferase is operably joined to the 5' regulatory regions of a BACE interacting gene. The human BACE interacting gene regulatory regions may be easily isolated and cloned by one of ordinary skill in the art. The reporter gene and regulatory regions are joined in-frame (or in each of the three possible reading frames) so that transcription and translation of the reporter gene may proceed under the control of the BACE interacting gene regulatory elements. The recombinant construct may then be introduced into any appropriate cell type. The transformed cells may be grown in culture and, after establishing the baseline level of expression of the reporter gene, test compounds may be added to the medium. The ease of detection of the expression of the reporter gene provides for a rapid, high through-put assay for the identification of inducers and repressors of the BACE interacting gene. Compounds identified by this method will have potential utility in modifying the expression of the BACE interacting genes in vivo. These compounds may be further tested in Alzheimer animal models known in the art to identify those compounds having the most potent in vivo effects. In addition, as further described herein, with respect to small molecules having binding activity on BACE interacting proteins or can modulate the interaction between BACE and a BACE-interacting protein, these molecules may serve as "lead compounds" for the further development of pharmaceuticals by, for example, subjecting the compounds to sequential modifications, molecular modelling, and other routine procedures employed in rational drug design. In another embodiment the invention provides a method or an assay for the identification of compounds which can modulate the activity of a BACE-interacting protein comprising the steps of (1) providing a cell expressing a normal or mutant gene encoding a BACE-interacting protein and (2) contacting said cell with at least one candidate compound, and detecting a change in a marker of said activity. In light of the present disclosure, one of ordinary skill in the art can practice new screening methodologies which will be useful in the identification of proteins and other molecules which bind to, or otherwise directly interact with, BACE interacting proteins. The proteins and molecules will include endogenous cellular components which interact with BACE interacting proteins in vivo and which, therefore, provide new targets for pharmaceutical and therapeutic interventions, as well as recombinant, synthetic and otherwise exogenous compounds which may have BACE interacting protein binding capacity and, therefore, may be candidates for pharmaceutical agents. Thus, in one series of embodiments, cell lysates or tissue homogenates (e.g., human brain homogenates, lymphocyte lysates) may be screened for proteins or other molecules which bind to one of the normal or mutant BACE-interacting proteins. Alternatively, any of a variety of exogenous compounds, both naturally occurring and/or synthetic (e.g., libraries of small molecules or peptides), may be screened for BACE-interacting protein binding capacity. Small molecules are particularly preferred in this context because they are more readily absorbed after oral administration, have fewer potential antigenic determinants, and/or are more likely to cross the blood brain barrier than larger molecules such as nucleic acids or proteins. The effect of agents which bind to a BACE-interacting protein (a normal or a mutant form) can be monitored either by the direct monitoring of this binding using instruments (e.g., BIAcore, LKB Pharmacia, Sweden) to detect this binding by, for example, a change in fluorescence, molecular weight, or concentration of either the binding agent or BACE-interacting protein component, either in a soluble phase or in a substrate-bound phase. Once identified by these methods described above, the candidate compounds may then be produced in quantities sufficient for pharmaceutical administration or testing (e.g. mg or greater quantities), and formulated in a pharmaceutically acceptable carrier (see, e.g., Remington's Pharmaceutical Sciences, Gennaro, A., ed., Mack Pub., 1990). These candidate compounds may then be administered to for example transformed cells or to transgenic animal models, to cell lines derived from the animal models or from human patients, or eventually to Alzheimer's patients. In addition, once identified by the methods described above, the candidate compounds may also serve as "lead compounds" in the design and development of new pharmaceuticals. For example, as is well known in the art, sequential modification of small molecules (e.g., amino acid residue replacement with peptides; functional group replacement with peptide or non-peptide molecules) is a standard approach in the pharmaceutical industry for the development of new pharmaceuticals. Such development generally proceeds from a "lead compound" which is shown to have at least some of the activity (e.g., BACE-interacting protein binding - promoting or blocking activity) of the desired pharmaceutical. In particular, when one or more compounds having at least some activity of interest are identified, structural comparison of the molecules can greatly inform the skilled practitioner by suggesting portions of the lead compounds which should be conserved and portions which may be varied in the design of new candidate compounds. Thus, the present invention also provides a means of identifying lead compounds which may be sequentially modified to produce new candidate compounds for use in the treatment of Alzheimer's Disease. These new compounds then may be tested both for BACE-interacting protein-binding - promoting or blocking activity (e.g., in the binding assays described above) and for therapeutic efficacy (e.g., in known animal models). This procedure may be iterated until compounds having the desired therapeutic activity and/or efficacy are identified. In each of the present series of embodiments, an assay or method is conducted to detect binding between a "BACE-interacting protein component" and some other moiety. Of particular utility will be sequential assays in which compounds are tested for the ability to bind to only the normal or only the mutant forms of the BACE-interacting protein functional domains using mutant and normal BACE-interacting protein components in the binding assays. Such compounds are expected to have the greatest therapeutic utilities, as described more fully below. The " BACE-interacting protein component" in these assays may be a complete normal or mutant form of a BACE-interacting protein but need not be. Rather, particular functional domains of the BACE-interacting protein, as disclosed in this patent application, may be employed either as separate molecules or as part of a fusion protein. For example, to isolate proteins or molecules that interact with these functional domains, screening may be carried out using fusion constructs and/or synthetic; peptides corresponding to these regions. Obviously, various combinations of fusion proteins and BACE-interacting protein functional domains are possible. In addition, the functional domains may be altered so as to aid in the assay by, for example, introducing into the functional domain a reactive group or amino acid residue (e.g., cysteine) which will facilitate immobilization of the domain on a substrate (e.g., using sulfhydryl reactions). Similarly, other functional domain or antigenic fragments may be created with modified residues. The proteins or other molecules identified by these methods may be purified and characterized by any of the standard methods known in the art. Proteins may, for example, be purified and separated using electrophoretic (e.g., SDS-PAGE, 2D PAGE) or chromatographic (e.g., HPLC) techniques and may then be microsequenced. For proteins with a blocked N-terminus, cleavage (e.g., by CNBr and/or trypsin) of the particular binding protein is used to release peptide fragments. Further purification/characterization by HPLC and microsequencing and/or mass spectrometry by conventional methods provides internal sequence data on such blocked proteins. For non-protein molecules, standard organic chemical analysis techniques (e.g., IR, NMR and mass spectrometry; functional group analysis; X-ray crystallography) may be employed to determine their structure and identity. Methods for screening cellular lysates, tissue homogenates, or small molecule libraries for candidate BACE-interacting protein-binding molecules are well known in the art and, in light of the present disclosure, may now be employed to identify compounds which bind to normal or mutant BACE-interacting protein components or which modulate BACE-interacting protein activity as defined by non-specific measures (e.g., changes, in intracellular Ca²⁺, GTP/GDP ratio) or by specific measures (e.g., changes in amyloid-beta peptide production or changes in the expression of other downstream genes which can be monitored by differential display, 2D gel electrophoresis, differential hybridization, or SAGE methods). The preferred methods involve variations on the following techniques: (1) direct extraction by affinity chromatography; (2) co-isolation of BACE-interacting protein components and bound proteins or other molecules by immunoprecipitation; (3) the Biomolecular Interaction Assay (BIAcore); and (4) the yeast two-hybrid systems. These and others are discussed briefly below. (1) Affinity Chromatography: in light of the present disclosure, a variety of affinity binding techniques well known in the art may be employed to isolate proteins or other molecules which bind to the BACE-interacting protein disclosed herein. In general, a BACE-interacting protein component may be immobilized on a substrate (e.g., a column or filter) and a solution including the test compound(s) is contacted with the BACE-interacting protein, fusion or fragment under conditions which are permissive for binding. The substrate is then washed with a solution to remove unbound or weakly bound molecules. A second wash may then elute those compounds which strongly bound to the immobilized normal or mutant BACE- interacting protein component. Alternatively, the test compounds may be immobilized and a solution containing one or more BACE-interacting protein components may be contacted with the column, filter or other substrate. The ability of the BACE-interacting protein component to bind to the test compounds may be determined as above or a labeled form of the BACE-interacting protein component (e.g., a radio-labeled or chemiluminescent functional domain) may be used to more rapidly assess binding to the substrate-immobilized compound(s). In addition, for membrane associated proteins, it may be preferred that a particular BACE-interacting protein, fusion or fragment be incorporated into lipid bilayers (e.g., Iiposomes) to promote their proper folding. This is particularly true when a BACE-interacting protein component including at least one transmembrane domain is employed (e.g. arsenite translocating ATPase, Gpr 37-like protein 1). Such BACE-interacting protein-liposomes may be immobilized on substrates (either directly or by means of another element in the liposome membrane), passed over substrates with immobilized test compounds, or used in any of a variety of other well known binding assays for membrane proteins. Alternatively, the BACE-interacting protein component may be isolated in a membrane fraction from cells producing the component, and this membrane fraction may be used in the binding assay. (2) Co-lmmunoprecipitation: another well characterized technique for the isolation of the BACE-interacting protein components and their associated proteins or other molecules is direct immunoprecipitation with antibodies. This procedure has been successfully used, for example, to isolate many of the synaptic vesicle associated proteins (Phizicky and Fields (1994) Microbiol. Reviews 59:94-123). Thus, either normal or mutant, free or membrane-bound BACE-interacting protein components may be mixed in a solution with the candidate compound(s) under conditions which are permissive for binding, and the BACE-interacting protein component(s) may be immunoprecipitated. Proteins or other molecules which co- immunoprecipitate with the BACE-interacting protein component may then be identified by standard techniques as described above. General techniques for immunoprecipitation may be found in, for example, Harlow and Lane, (1988) Antibodies: A Laboratory Manual, Cold Spring Harbor Press, Cold Spring Harbor, N.Y. (3) The Biomolecular Interaction Assay: another useful method for the detection and isolation of binding proteins is the Biomolecular Interaction Assay or "BIAcore" system developed by Pharmacia Biosensor and described in the manufacturer's protocol (LKB Pharmacia, Sweden). In light of the present disclosure, one of ordinary skill in the art is now able to employ this system, or a substantial equivalent, to identify proteins or other molecules having BACE-interacting protein binding capacity. The BIAcore system uses an affinity purified anti-GST antibody (commercially available from Amersham- Pharmacia-Biotech) to immobilize GST-fusion proteins onto a sensor chip. Obviously, other fusion proteins and corresponding antibodies may be substituted. The sensor utilizes surface plasmon resonance which is an optical phenomenon that detects changes in refractive indices. A homogenate of a tissue of interest is passed over the immobilized fusion protein and protein-protein interactions are registered as changes in the refractive index. This system can be used to determine the kinetics of binding and to assess whether any observed binding is of physiological relevance. (4) The Yeast Two-Hybrid System: The yeast "two-hybrid" system takes advantage of transcriptional factors that are composed of two physically separable, functional domains (Phizicky and Fields, 1994). The most commonly used is the yeast GAL4 transcriptional activator consisting of a DNA binding domain and a transcriptional activation domain. Two different cloning vectors are used to generate separate fusions of the GAL4 domains to genes encoding potential binding proteins. The fusion proteins are co-expressed, targeted to the nucleus and, if interactions occur, activation of a reporter gene (e.g., lacZ) produces a detectable phenotype. For example, the Clontech Matchmaker System-2 may be used with the Clontech brain cDNA GAL4 activation domain fusion library with BACE-interacting protein-GAL4 binding domain fusion clones (Clontech, Palo Alto, Calif.). In light of the disclosures herein, one of ordinary skill in the art is now able to produce a variety of BACE-interacting protein-fusions, including fusions including either normal or mutant functional domains of the BACE- interacting proteins, and to screen such fusion libraries in order to identify BACE- interacting proteins binding proteins. (5) Other Methods: the nucleotide sequences and protein products, including both mutant and normal forms of BACE-interacting genes and proteins can be used with the above techniques to isolate other interacting proteins, and to identify other genes whose expression is altered by the over- expression of normal BACE-interacting gene sequences, by the under-expression of normal BACE-interacting gene sequences, or by the expression of mutant BACE- interacting gene sequences. Identification of these interacting proteins, as well as the identification of other genes whose expression levels are altered in the face of mutant BACE-interacting gene sequences (for instance) will identify other gene targets which have direct relevance to the pathogenesis of for example AD in its clinical or pathological forms. Specifically, other genes will be identified which may themselves be the site of other mutations causing Alzheimer's Disease, or which can themselves be targeted therapeutically (e.g., to reduce their expression levels to normal or to pharmacologically block the effects of their over-expression) as a potential treatment for this disease. Specifically, these techniques rely on PCR-based and/or hybridization- based methods to identify genes which are differentially expressed between two conditions (a cell line expressing normal BACE-interacting genes compared to the same cell type expressing a mutant BACE-interacting proteins gene sequence). These techniques include differential display, serial analysis of gene expression (SAGE), and mass-spectrometry of protein 2D-gels and subtractive hybridization (reviewed in Nowak (1995) Science 270:368-371 and Kahn (1995) Science 270:369-370). As will be obvious to one of ordinary skill in the art, there are numerous other methods of screening individual proteins or other molecules, as well as large libraries of proteins or other molecules (e.g., phage display libraries and cloning systems from Stratagene, La Jolla, Calif.) to identify molecules which bind to normal or mutant BACE-interacting protein components. All of these methods comprise the step of mixing a normal or mutant BACE-interacting protein, fusion, or fragment with test compounds, allowing for binding (if any), and assaying for bound complexes. In another embodiment the invention provides a method of identifying compounds which can modulate the interaction between a BACE-interacting protein or a functional fragment thereof and BACE comprising the steps of (1) providing a cell expressing a normal or mutant gene encoding a BACE-interacting protein and a normal or mutant gene encoding BACE, and (2) contacting said cell with at least one candidate compound, and detecting a change in the interaction. A 'functional fragment thereof means a contiguous polypeptide fragment derived from a BACE-interacting protein that is still capable of binding to BACE with the same binding specificity as the entire protein. The ability to disrupt specific BACE interactions with a particular or more BACE-interacting proteins is of great therapeutic value, and will be important in understanding the etiology of AD and in identifying additional targets for therapy. It should be clear that the methods used to identify compounds which disrupt BACE interactions may be applied equally well to interactions involving either normal or mutant BACE and either normal or mutant BACE interacting proteins. Assays for compounds which can disrupt BACE interactions may be performed by any of a variety of methods well known in the art. In essence, such assays will parallel those assays for identifying proteins and molecules interacting with BACE-interacting proteins as described herein before. Thus, once a protein or compound that interacts with a BACE-interacting protein is identified by any method, that method or an equivalent method may be performed in the presence of said candidate molecules or proteins which disrupt the interaction between BACE and a BACE-interacting protein. Thus, for example, the assay may employ methods including (1) affinity chromatography; (2) immunoprecipitation; (3) the Biomolecular Interaction Assay (BIAcore); or (4) the yeast two-hybrid systems. Such assays can be developed using either normal or mutant purified BACE proteins, and/or either normal or mutant and purified BACE-interacting proteins. For affinity methods, either the BACE or the BACE-interacting protein may be affixed to a matrix, for example in a column, and the counterpart protein (the BACE- interacting protein if BACE is affixed to the matrix, or BACE if the BACE-interacting protein is affixed to the matrix) is then exposed to the affixed protein either before or after adding the candidate compound(s). In the absence of a disruptive effect by the candidate compound(s), the interaction between BACE and BACE-interacting protein will cause the counterpart protein to bind to the affixed protein. Any compound which disrupts the interaction will cause release of the counterpart protein from the matrix. Release of the counterpart protein from the matrix can be measured using methods known in the art. For BACE-BACE-interacting protein interactions which are detectable by yeast two-hybrid systems, these assays may also be employed to identify compounds which disrupt the interaction. Briefly, the BACE and BACE-interacting proteins (or appropriate structural domains of each) are employed in the fusion proteins of the system and the cells may be exposed to candidate compounds to determine their effect upon the expression of the reporter gene. By appropriate choice of a reporter gene, such a system can be readily adapted for high through-put screening of large libraries of compounds by, for example, using a reporter gene which confers resistance to an antibiotic which is present in the medium, or which rescues an auxotrophic strain grown in minimal medium. These assays may be used to screen many different types of molecules for their disruptive effect on the interactions of BACE. For example, the molecules may belong to a library of synthetic molecules, or be specifically designed to disrupt the interaction. The compounds may also be peptides corresponding to the interacting domain of either protein. This type of assay can be used to identify compounds that disrupt a specific interaction between a given BACE-interacting protein variant and a given interacting protein. In addition, compounds that disrupt all interactions with BACE may be identified. For example, a compound that specifically disrupts the folding of BACE or BACE-interacting proteins is expected to disrupt all interactions between BACE and other proteins. Alternatively, this type of disruption assay can be used to identify compounds which disrupt only a range of different BACE interactions, or only a single BACE interaction. To explain the wording 'a change in a marker of said activity' and 'detecting a change in the interaction' a non-limited list of different monitoring systems or read-out systems are presented below which can be used to measure the effect of a compound or a protein on the modulation of the activity of a BACE-interacting protein or BACE- interacting gene expression or the modulation between BACE and a particular or several BACE-interacting proteins. Specific measures of BACE-interacting gene expression can be employed to screen candidate compounds for their ability to affect BACE-interacting gene or protein activity. Thus, using BACE-interacting nucleic acids and antibodies against them one may use mRNA levels or protein levels as a marker for the ability of a candidate compound to modulate BACE-interacting gene or protein activity. The use of such probes and antibodies to measure gene and protein expression is well known in the art. Also the monitoring of the effect of compounds may be screened for their ability to modulate the activity of the BACE-interacting proteins based upon their effects on the trafficking and intracellular localization of the BACE-interacting proteins. Since BACE has been observed immunocytochemically to be localized in membrane structures associated with mainly the endosomes, and some of the BACE-interacting proteins disclosed herein are responsible for the transport of BACE to its normal intracellular localisation, it is clear that a change in the activity of one particular or even more BACE-interacting genes or proteins has an effect on the normal intracellular localisation of BACE. Differences in localization of mutant and normal BACE-interacting proteins or differences in localisation of BACE itself can, therefore, contribute to the etiology of BACE-interacting protein-related diseases such as Alzheimer's disease. Compounds which can affect the localization of the BACE- interacting proteins or BACE can, therefore, be identified as potential therapeutics. Standard techniques known in the art can be employed to detect the localization of the BACE-interacting proteins or BACE. Generally, these techniques will employ antibodies and in particular antibodies which selectively bind to one or more mutant BACE-interacting proteins but not to normal BACE-interacting proteins. As is well known in the art, such antibodies may be labeled by any of a variety of techniques (e.g., fluorescent or radioactive tags, labeled secondary antibodies, avidin-biotin, etc.) to aid in visualizing the intracellular location of a BACE-interacting protein. Also Western blots of purified fractions from cell lysates enriched for different intracellular membrane bound organelles (e.g., lysosomes, synaptosomes, endosomes, Golgi) may also be employed. The monitoring of a specific compounds or compounds can also be observed for their ability to modulate the activity of the specific BACE-interacting proteins based upon measures in intracellular Ca²⁺, Na⁺ or K⁺ levels or metabolism. Indeed, some of the isolated BACE-interacting proteins are membrane associated proteins which may serve as, or interact with, ion receptors or ion channels. Thus, compounds may be screened for their ability to modulate BACE-interacting protein- related calcium or other ion metabolism either in vivo or in vitro by measurements of ion channel fluxes and/or transmembrane voltage or current fluxes using patch clamp, voltage clamp and fluorescent dyes sensitive to intracellular calcium or transmembrane voltage. Ion channel or receptor function can also be assayed by measurements of activation of second messengers such as cyclic AMP, cGMP tyrosine kinases, phosphates, increases in intracellular Ca²⁺ levels, etc. Recombinantly made proteins may also be reconstructed in artificial membrane systems to study ion channel conductance and, therefore, the "cell" employed in such assays may comprise an artificial membrane or cell. Assays for changes in ion regulation or metabolism can be performed on cultured cells expressing endogenous normal or mutant BACE- interacting proteins. Such studies also can be performed on cells transfected with vectors capable of expressing one of the BACE-interacting proteins, or functional domains of one of the BACE-interacting proteins, in normal or mutant form. In addition, to enhance the signal measured in such assays, cells may be co-transfected with genes encoding ion channel proteins. Compounds may be screened for their ability to modulate the activity of the BACE-interacting proteins or one or more interactions between BACE and a BACE-interacting protein based upon their effects on apoptosis or cell death. Indeed, in the current invention pro-apoptotic (e.g. bcl-rambo) and anti- apoptotic (e.g. lifeguard) proteins have been isolated that specifically interact with BACE and compounds can modulate the activity of such pro-apoptotic proteins or prevent the binding of said pro-apoptotic proteins with BACE and disturb the delicate equilibrium between life and death in the cell. Thus, for example, baseline rates of apoptosis or cell death can be established for cells in culture, or the baseline degree of neuronal loss at a particular age may be established post-mortem for animal models or human subjects, and the ability of a candidate compound to suppress or inhibit apoptosis or cell death may be measured. Cell death may be measured by standard microscopic techniques (e.g., light microscopy) or apoptosis may be measured more specifically by characteristic nuclear morphologies or DNA fragmentation patterns which create nucleosomal ladders (see, e.g., Gavrieli et al. (1992) J. Cell Biol. 119:493-501 ; Jacobson et al. (1993) Nature 361 :365; Vito et al. (1996) Science 271 :521-525). TUNEL may also be employed to evaluate cell death in brain (see, e.g., Lassmann et al., 1995). In another monitoring method the quality and the quantity of amyloid beta production can be easily measured by methods known in the art and thus used to monitor the effect of a compound on a BACE-interacting protein or the modulation of the interaction between a BACE-interacting protein and BACE. Another monitoring method makes use of the possibility of effects of compounds, that modulate the activity of a BACE-interacting protein or modulate the interaction between a BACE- interacting protein and BACE, on the levels of phosphorylation of microtubule associated proteins (MAPs) such as Tau. The abnormal phosphorylation of Tau and other MAPs in the brains of victims of Alzheimer's Disease is well known in the art. Thus, compounds which prevent or inhibit the abnormal phosphorylation of MAPs may have utility in treating BACE-interacting protein associated diseases such as AD. As above, cells from normal or mutant animals or subjects, or the transformed cell lines and existing animal models may be employed.

In another embodiment the invention provides a diagnostic method for determining if a subject bears a mutant gene encoding a BACE-interacting protein comprising the steps of (1) providing a biological sample of said subject, and (2) detecting in said sample a mutant nucleic acid encoding a BACE-interacting protein, a mutant BACE- interacting protein, or a mutant BACE-interacting protein activity. The BACE interacting genes and gene products, as well as other products derived thereof (e.g., probes, antibodies), can be useful in the diagnosis of Alzheimer's Disease, but also possibly in the diagnosis of presenile and senile dementias, and probably also in neurologic diseases such as stroke and cerebral hemorrhage - all of which are seen to a greater or lesser extent in symptomatic subjects bearing mutations in currently known genes that are involved in AD disease. Diagnosis of inherited cases of these diseases can be accomplished by methods based upon the nucleic acids (including genomic and mRNA/cDNA sequences), proteins, and/or antibodies. Preferably, the methods and products are based upon the human BACE interacting genes, proteins or antibodies against BACE interacting proteins. As will be obvious to one of ordinary skill in the art, however, the significant evolutionary conservation of large portions of the BACE interacting nucleotide and amino acid sequences, even in species as diverse as humans and C. elegans and Drosophila, allow the skilled artisan to make use of such non-human BACE interacting-homologue nucleic acids, proteins and antibodies even for applications directed toward human or other mammalian subjects. Thus, for brevity of exposition, but without limiting the scope of the invention, the following description will focus upon uses of the human homologues of BACE interacting genes and proteins. It will be understood, however, that homologous sequences from other species, including those disclosed herein, will be equivalent for many purposes. As will be appreciated by one of ordinary skill in the art, the choice of diagnostic methods of the present invention will be influenced by the nature of the available biological samples to be tested and the nature of the information required. BACE and BACE- interacting genes, for example, are highly expressed in brain tissue but brain biopsies are invasive and expensive procedures, particularly for routine screening. Other tissues which express BACE interacting genes at significant levels, however, may demonstrate alternative splicing (e.g., white blood cells) and, therefore mRNA derived from BACE interacting genes or proteins from such cells may be less informative. Thus, assays based upon a subject's genomic DNA may be the preferred methods for diagnostics of BACE interacting genes as no information will be lost due to alternative splicing and because essentially any nucleate cells may provide a usable sample. When a diagnostic assay is to be based upon BACE interacting proteins, a variety of approaches are possible. For example, diagnosis can be achieved by monitoring differences in the electrophoretic mobility of normal and mutant proteins. Such an approach will be particularly useful in identifying mutants in which charge substitutions are present, or in which insertions, deletions or substitutions have resulted in a significant change in the molecular mass of the resultant protein. Alternatively, diagnosis may be based upon differences in the proteolytic cleavage patterns of normal and mutant proteins, differences in molar ratios of the various amino acid residues, or by functional assays demonstrating altered function of the gene products. In some preferred embodiments, protein-based diagnostics will employ differences in the ability of antibodies to bind to normal and mutant BACE interacting proteins. Such diagnostic tests may employ antibodies which bind to the normal proteins but not to mutant proteins, or vice versa. In particular, an assay in which a plurality of monoclonal antibodies, each capable of binding to a mutant epitope, may be employed. The levels of anti-mutant examples binding in a sample obtained from a test subject (visualized by, for example, radiolabelling, ELISA or chemiluminescence) may be compared to the levels of binding to a control sample. Such antibody diagnostics may be used for in situ immunohistochemistry using biopsy samples of CNS tissues obtained antemortem or postmortem, including neuropathological structures associated with these diseases such as neurofibrillary tangles and amyloid plaques, or may be used with fluid samples such a cerebrospinal fluid or with peripheral tissues such as white blood cells. When the diagnostic assay is to be based upon nucleic acids from a sample, either mRNA or genomic DNA may be used. When mRNA is used from a sample, many of the same considerations apply with respect to source tissues and the possibility of alternative splicing. That is there may be little or no expression of transcripts unless appropriate tissue sources are chosen or available, and alternative splicing may result in the loss of some information. With either mRNA or DNA, standard methods well known in the art may be used to detect the presence of a particular sequence either in situ or in vitro (see, e.g. Sambrook et al., eds. (1989) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Press, Cold Spring Harbor, N.Y.). For in situ detection of a mutant nucleic acid sequence of a BACE interacting protein, a sample of tissue may be prepared by standard techniques and then contacted with a probe, preferably one which is labelled to facilitate detection, and an assay for nucleic acid hybridization is conducted under stringent conditions which permit hybridization only between the probe and highly or perfectly complementary sequences. Because many mutations in genes that cause diseases detected to date consist of a single nucleotide substitution, high stringency hybridization conditions will be required to distinguish normal sequences from most mutant sequences. A significant advantage of the use of either DNA or mRNA is the ability to amplify the amount of genetic material using the polymerase chain reaction (PCR), either alone (with genomic DNA) or in combination with reverse transcription (with mRNA to produce cDNA). Other nucleotide sequence amplification techniques may be used, such as ligation-mediated PCR, anchored PCR and enzymatic amplification as will be understood by those skilled in the art. Sequence alterations may also generate fortuitous restriction enzyme recognition sites which are revealed by the use of appropriate enzyme digestion followed by gel-blot hybridization. DNA fragments carrying the site (normal or mutant) are detected by their increase or reduction in size, or by the increase or decrease of corresponding restriction fragment numbers. Genomic DNA samples may also be amplified by PCR prior to treatment with the appropriate restriction enzyme and the fragments of different sizes are visualized, for example under UV light in the presence of ethidium bromide, after gel electrophoresis. Genetic testing based on DNA sequence differences may be achieved by detection of alteration in electrophoretic mobility of DNA fragments in gels. Small sequence deletions and insertions can be visualized by high resolution gel electrophoresis of single stranded DNA, or as changes in the migration pattern of DNA heteroduplexes in non-denaturing gel electrophoresis. Alternatively, a single base substitution mutation may be detected based on differential PCR product length in PCR. The PCR products of the normal and mutant gene may be differentially detected in acrylamide gels. Nuclease protection assays (S1 or ligase-mediated) also reveal sequence changes at specific locations. Alternatively, to confirm or detect a polymorphism resulting in restriction mapping changes. Ligated PCR, ASO, REF- SSCP chemical cleavage, endonuclease cleavage at mismatch sites or SSCP may be used. Both REF-SSCP and SSCP are mobility shift assays which are based upon the change in conformation due to mutations. DNA fragments may also be visualized by methods in which the individual DNA samples are not immobilized on membranes. The probe and target sequences may be in solution or the probe sequence may be immobilized. Autoradiography, radioactive decay, spectrophotometry and fluorometry may also be used to identify specific individual genotypes. Mutations can also be detected by direct nucleotide sequencing. The wording 'stringent hybridisation conditions' is a term of art understood by those of ordinary skill in the art. For any given nucleic acid sequence, stringent hybridisation conditions are those conditions of temperature, chaotrophic salts, pH and ionic strength which will permit hybridisation or that nucleic acid sequence to its complementary sequence and not to substantially different sequences. The exact conditions which constitute "stringent" conditions, depend upon the nature of the nucleic acid sequence, the length of the sequence, and the frequency of occurrence of subsets of that sequence within other non-identical sequences. By varying hybridisation conditions from a level of stringency at which non-specific hybridisation conditions occurs to a level at which only specific hybridisation is observed, one of ordinary skill in the art can, without undue experimentation, can determine conditions which will allow a given sequence to hybridise only with complementary sequences. Hybridisation conditions, depending upon the length and commonality of a sequence, may include temperatures of 20°C-65°C and ionic strengths from 5x to 0.1x SSC. Highly stringent hybridisation conditions may include temperatures as low as 40-42°C (when denaturants such as formamide are included) or up to 60-65°C in ionic strengths as low as O.lxSSC. These ranges, however, are only illustrative and, depending upon the nature of the target sequence, and possible future technological developments, may be more stringent than necessary. Less than stringent conditions are employed to isolate nucleic acid sequences which are substantially similar, allelic or homologous to a given sequence.

In another embodiment the isolated nucleotide sequences that comprise a nucleotide sequence encoding a BACE-interacting protein or a functional fragment thereof can be used as a medicament. In yet another embodiment the isolated nucleotide sequences that comprise a nucleotide sequence encoding a BACE-interacting protein or a functional fragment thereof can be used for the manufacture of a medicament to treat Alzheimer's disease. The term 'to manufacture of a medicament' relates here to the application of gene therapy to deliver at least one BACE-interacting gene to a cell of a patient in need thereof. In other words the present invention provides the nucleic acids of BACE-interacting genes for the transfection of cells in vitro and in vivo. These nucleic acids can be inserted into any of a number of well-known vectors for the transfection of target cells and organisms as described below. The nucleic acids are transfected into cells, ex vivo or in vivo, through the interaction of the vector and the target cell. The nucleic acids encoding BACE-interacting proteins, under the control of a promoter, then expresses a particular BACE-interacting protein of the present invention, thereby mitigating the effects of absent, partial inactivation, or abnormal expression of the said BACE-interacting gene. Such gene therapy procedures have been used to correct acquired and inherited genetic defects, cancer, and viral infection in a number of contexts. The ability to express artificial genes in humans facilitates the prevention and/or cure of many important human diseases, including many diseases which are not amenable to treatment by other therapies (for a review of gene therapy procedures, Nabel & Feigner, TIBTECH 11 :211-217 (1993); Mintani & Caskey, TIBTECH 11 :162-166 (1993); Mulligan, Science 926-932 (1993); Dillon, TIBTECH 11 :167-175 (1993); Van Brunt, Biotechnology 6(10): 1149-1154 (1998); Vigne, Restorative Neurology and Neuroscience 8:35-36 (1995); Kremer & Perricaudet, British Medical Bulletin 51(1); 31-44 (1995); Haddada et al., in Current Topics in Microbiology and Immunology (Doerfler & Bόhm eds., 1995); and Yu et al., Gene Therapy 1:13-26 (1994)). Delivery of the gene or genetic material into the cell is the first critical step in gene therapy treatment of disease. A large number of delivery methods are well known to those of skill in the art. Preferably, the nucleic acids are administered for in vivo or ex vivo gene therapy uses. Non-viral vector delivery systems include DNA plasmids, naked nucleic acid, and nucleic acid complexed with a delivery vehicle such as a liposome. Viral vector delivery systems include DNA and RNA viruses, which have either episomal or integrated genomes after delivery to the cell. Methods of non-viral delivery of nucleic acids include lipofection, microinjection, biolistics, virosomes, Iiposomes, immunoliposomes, polycation or lipid: nucleic acid conjugates, naked DNA, artificial virions, and agent-enhanced uptake of DNA. Lipofection is described in, e.g., US Pat. No. 5,049,386, US Pat No. 4,946,787; and US Pat. No. 4,897,355 and lipofection reagents are sold commercially (e.g., Transfectam™ and Lipofectin™). Cationic and neutral lipids that are suitable for efficient receptor-recognition lipofection of polynucleotides include those of Flegner, WO 91/17424, WO 91/16024. Delivery can be to cells (ex vivo administration) or target tissues (in vivo administration). The preparation of lipid: nucleic acid complexes, including targeted Iiposomes such as immunolipid complexes, is well known to one of skill in the art (see, e.g., Crystal, Science 270:404-410 (1995); Blaese et al., Cancer Gene Ther. 2:291-297 (1995); Behr et al., Bioconjugate Chem. 5:382-389 (1994); Remy et al., Bioconjugate Chem. 5:647-654 (1994); Gao et al., Gene Therapy 2:710- 722 (1995); U.S. Pat. Nos. 4,186,183, 4,217,344, 4,235,871 , 4,261,975, 4,485,054, 4,501 ,728, 4,774,085, 4,837,028, and 4,946,787). The use of RNA or DNA viral based systems for the delivery of nucleic acids take advantage of highly evolved processes for targeting a virus to specific cells in the body and trafficking the viral payload to the nucleus. Viral vectors can be administered directly to patients (in vivo) or they can be used to treat cells in vitro and the modified cells are administered to patients (ex vivo). Conventional viral based systems for the delivery of nucleic acids could include retroviral, lentivirus, adenoviral, adeno-associated and herpes simplex virus vectors for gene transfer. Viral vectors are currently the most efficient and versatile method of gene transfer in target cells and tissues. Integration in the host genome is possible with the retrovirus, lentivirus, and adeno-associated virus gene transfer methods, often resulting in long-term expression of the inserted transgene. Additionally, high transduction efficiencies have been observed in many different cell types and target tissues. The tropism of a retrovirus can be altered by incorporating foreign envelope proteins, expanding the potential target population of target cells. Lentiviral vectors are retroviral vector that are able to transduce or infect non-dividing cells and typically produce high viral titers. Selection of a retroviral gene transfer system would therefore depend on the target tissue. Retroviral vectors are comprised on c/^'s-acting long terminal repeats with packaging capacity for up to 6-10 kb of foreign sequence. The minimum c/s-acting LTRs are sufficient for replication and packaging of the vectors, which are then used to integrate the therapeutic gene into the target cell to provide permanent transgene expression. Widely used retroviral vectors include those based upon murine leukemia virus (MuLV), gibbon ape leukemia virus (GaLV), simian immunodeficiency virus (SIV), human immunodeficiency virus (HIV), and combinations thereof (see, e.g., Buchscher et al., J. Virol. 66:2731-2739 (1992); PCT/US94/05700. In applications where transient expression of the nucleic acid is preferred, adenoviral based systems are typically used. Adenoviral based vectors are capable of very high transduction efficiency in many cell types and do not require cell division. With such vectors, high titter and levels of expression have been obtained. This vector can be produced in large quantities in a relatively simple system. Adeno-associated virus ("AAV") vectors are also used to transduce cells with target nucleic acids, e.g., in the in vitro production of nucleic acids and peptides, and for in vivo and ex vivo gene therapy procedures (see, e.g., U.S. Patent No. 4,797,368; WO 93/24641 ; Kotin, Human Gene Therapy 5:793-801 (1994); Muzyczka. Construction of recombinant AAV vectors are described in a number of publications, including U.S. Pat. No. 5,173,414; Hermonat & Muzyczka, Proc. Natl. Acad. Sci. U.S.A. 81 :6466-6470 (1984); and Samulski ef al., J.Virol. 63:03822-3828 (1989). In particular, at least six viral vector approaches are currently available for gene transfer in clinical trials, with retroviral vectors by far the most frequently used system. All of these viral vectors utilize approaches that involve complementation of defective vectors by genes inserted into helper cell lines to generate the transducing agent. pLASN and MFG-S are examples are retroviral vectors that have been used in clinical trials (Dunbar et al., Blood 85:3048-305 (1995); Kohn ef al., Nat. Med. 1 :1017-102 (1995); Malech et al., Proc. Natl. Acad. Sci. U.S.A. 94/22 12133-12138 (1997)); Pa317/pLASN was the first therapeutic vector used in a gene therapy trials. (Blaese ef al., Science 270:475-480 (1995)). Transduction efficiencies of 50% greater have been observed for MFG-S packaged vectors (Ellem et al. Immunol Immunother. 44(1): 10-20 (1997); Dranoff ef a/., Hum. Gene Ther. 1 :111-2 (1997)). Recombinant adeno-associated virus vectors (rAAV) are a promising alternative gene delivery systems based on the defective and non-pathogenic parvovirus adeno-associated type 2 virus. All vectors are derived from a plasmid that retains only the AAV 145 bp inverted terminal repeats flanking the transgene expression cassette. Efficient gene transfer and stable transgene delivery due to integration into the genomes of the transduced cell are key features for this vector system (Wagner et al., Lancet 351 :9117 1702-3 (1998). Replication-deficient recombinant adenoviral vectors (Ad) are predominantly used transient expression gene therapy, because they can be produced at high titter and they readily infect a number of different cell types. Most adenovirus vectors are engineered such that a transgene replaced the Ad E1a, E1b, and E3 genes; subsequently the replication defector vector is propagated in human 293 cells that supply deleted gene function in trans. Ad vectors can transduce multiple types of tissues in vivo, including nondividing, differentiated cells such as those found in the liver, kidney and muscle system tissues. Conventional Ad vectors have a large carrying capacity. An example of the use of an Ad vector in a clinical trial involved polynucleotide therapy for antitumor immunization with intramuscular injection (Sterman ef al., Hum. Gene Ther. 7:1083-9 (1998)). Additional examples of the use of adenovirus vectors for gene transfer in clinical trials include Sterman ef al., Hum. Gene Ther. 9:7 1083-1089 (1998); Alvarez ef al., Hum. Gene Ther. 5:597-613 (1997); Topf et al., Gene Ther. 5:507-513 (1998)). Packaging cells are used to form virus particles that are capable of infecting a host cell. Such cells include 293 cells, which package adenovirus, and ψ2 cells or PA317 cells, which package retrovirus. Viral vectors used in gene therapy are usually generated by producer cell line that packages a nucleic acid vector into a viral particle. The vectors typically contain the minimal viral sequences required for packaging and subsequent integration into a host, other viral sequences being replaced by an expression cassette for the protein to be expressed. The missing viral functions are supplied in trans by the packaging cell line. For example, AAV vectors used in gene therapy typically only possess ITR sequences from the AAV genome which are required for packaging and integration into the host genome. Viral DNA is packaged in a cell line, which contains a helper plasmid encoding the other AAV genes, namely rep and cap, but lacking ITR sequences. The cell line is also infected with adenovirus as a helper. The helper virus promotes replication of the AAV vector and expression of AAV genes from the helper plasmid. The helper plasmid is not packaged in significant amounts due to a lack of ITR sequences. Contamination with adenovirus can be reduced by, e.g., heat treatment to which adenovirus is more sensitive than AAV. In many gene therapy applications, it is desirable that the gene therapy vector be delivered with a high degree of specificity to a particular tissue type. A viral vector is typically modified to have specificity for a given cell type by expressing a ligand as a fusion protein with a viral coat protein on the viruses outer surface. The ligand is chosen to have affinity for a receptor known to be present on the cell type of interest. For example, Han et al., Proc. Natl. Acad. Sci. U.S.A. 92/9747-9751 (1995), reported that Moloney murine leukemia virus can be modified to express human heregulin fused to gp70, and the recombinant virus infects certain human breast cancer cells expressing human epidermal growth factor receptor. This principle can be extended to other pairs of virus expressing a ligand fusion protein and target cell expressing a receptor. For example, filamentous phage can be engineered to display antibody fragments (e.g., FAB or Fv) having specific binding affinity for virtually any chosen cellular receptor. Although the above description applies primarily to viral vectors, the same principles can be applied to nonviral vectors. Such vectors can be engineered to contain specific uptake sequences thought to favor uptake by specific target cells. Gene therapy vectors can be delivered in vivo by administration to an individual patient, typically by systemic administration (e.g., intravenous, intraperitoneal, intramuscular, subdermal, or intracranial infusion) or topical application, as described below. Alternatively, vectors can be delivered to cells ex vivo, such as cells explanted from an individual patient (e.g., lymphocytes, bone marrow aspirates, tissue biopsy) or universal donor hematopoietic stem cells, followed by reimplantation of the cells into a patient, usually after selection for cells which have incorporated the vector. Ex vivo cell transfection for diagnostics, research, or for gene therapy (e.g., via re-infusion of the transfected cells into the host organism) is well known to those of skill in the art. In a preferred embodiment, cells are isolated from the subject organism, transfected with a nucleic acid (gene or cDNA), and re-infused back into the subject organism (e.g., patient). Various cell types suitable for ex vivo transfection are well known to those of skill in the art (see, e.g., Freshney ef al., Culture of Animal Cells, A Manual of Basic Technique (3^rd ed. 1994)) and the references cited therein for a discussion of how to isolate and culture cells from patients). In a specific embodiment, stem cells are used in ex vivo procedures for cell transfection and gene therapy. The advantage to using stem cells is that they can be differentiated into other cell types in vitro, or can be introduced into a mammal (such as the donor of the cells) where they will engraft in the bone marrow. Methods for differentiating CD34+ cells in vitro into clinically important immune cell types using cytokines such a GM-CSF, IFN-ψand TNF-ψ are known (see Inaba ef al., J. Exp. Med. 176: 1693-1702 (1992)). Stem cells are isolated for transduction and differentiation using known methods. For example, stem cells are isolated from bone marrow cells by panning the bone marrow cells with antibodies which bind unwanted cells, such as CD4+ and CD8+ (T cells), CD45+ (panB cells), GR-1 (granulocytes), and lad (differentiated antigen presenting cells) (see Inaba et al., J. Exp. Med. 176:1693-1702 (1992)). Vectors (e.g., retroviruses, adenoviruses, Iiposomes, etc.) containing therapeutic nucleic acids can be also administered directly to the organism for transduction of cells in vivo. Alternatively, naked DNA can be administered. Administration is by any of the routes normally used for introducing a molecule into ultimate contact with blood or tissue cells. Suitable methods of administering such nucleic acids are available and well known to those of skill in the art, and, although more than one route can be used to administer a particular composition, a particular route can often provide a more immediate and more effective reaction than another route. Examples 1) APP processing in COS cells by a BACE derivative that lacks the cytoplasmic domain

Several reports indicated that the cytoplasmic domain of BACE contains determinants important for BACE transport through the secretory pathway and BACE localization.

To determine whether deletion of the cytoplasmic domain of BACE affects its β-secretase activity, we compared APP processing in COS cells overexpressing mouse BACE or a mouse BACE derivative lacking the cytoplasmic tail (SEQ ID NO

41 ). COS cells were transiently co-transfected with a plasmid encoding human APP (pSG5^**-hAPP) and either an empty pSGδ^" vector, a plasmid encoding mouse BACE wild type (pSG5^**-mBACE) or mBACEΔcyt (pSG5^**-mBACEΔcyt). Two days after transfection, COS cells were metabolically labeled with [³⁵S] methionine. Four hours after labeling the conditioned medium as well as the cells were analyzed for APP processing. Equal amounts of media were subjected to SDS-PAGE and Western blotting to detect sAPPβ, the secreted ectodomain of APP generated after β-secretase cleavage at Asp1 (Aβ numbering). Radiolabeled APP-C-Terminal Fragments (CTFs or stubs) were purified from cell extracts by immunoprecipitation using antibody B11.4

(De Strooper et al., EMBO journal 20, 4932-4938, 1995, antibody B11/4 is similar to antibody B12/4 and was generated by us against a peptide comprising the last twenty amino acid residues of the APP protein, and separated by SDS-PAGE. CTFs were subsequently detected by exposing the gel to a Phosphor Imaging screen. APP processing by α- and β-secretase leads to the generation of 3 different CTFs. The CTF generated by α-secretase cleavage of APP consists of 83 amino acids (C83). C99 and

C89 are the CTFs that originate from β-secretase cleavage of APP at the Asp1 and Glu11 site (Aβ numbering), respectively.

Expression of APP alone in COS cells resulted in APP-processing mainly at the α-secretase site, as observed by the accumulation of C83 stubs. Endogenous β-secretase activity leading to C99 and C89 production could not be detected, as expected, since β-secretase activity is known to be very low in this type of non- neuronal cells.

Co-expression of BACE wild type and APP induced APP processing at the two β-secretase sites, Asp1 and Glu11 , leading to the generation of C99 and C89 respectively. Surprisingly C89 stubs accumulated much more composed to C99 stubs, although the cleavage of Asp1 is the preferred cleavage site of BACE. Most probably this is caused by the further processing of C99 at the Glu11 site (Creemers et al. J. Biol. Chem. 276, 4211-4217,2001 ). This makes C89 more abundant in conditions when BACE is overexpressed. Concomitantly, there was a decrease in the amount of C83. This is in agreement with the competition between α- and β-secretase that has been reported previously. When BACEΔcyt was co-expressed with APP, similar results as for co-expression of BACE and APP were obtained, i.e. a clear increase in the accumulation of C99 and C89, with relatively more C89 and a decrease in C83. These results were corroborated by analyzing sAPPβ in the conditioned medium by Western blotting. Proteins secreted into the conditioned medium of transfected cells were seperated by SDS-PAGE, transferred to a nitrocellulose membrane and sAPPβ was specifically detected by antibody 53/4 (kindly provided by Dr. M. Savage, Cephalon, USA). A clear increase in sAPPβ was observed when APP was co-transfected with either BACE or BACEΔcyt.

Together, these results suggest that BACEΔcyt is as active as wild type BACE in cleaving APP when overexpressed in COS cells.

Next, and to control protein expression, BACE and BACEΔcyt were immunoprecipitated from COS cell lysates with antibody B45.1 (this antibody is described in Creemers et al., J. Biol Chem, 276, 4211-4217, 2001 ), that recognizes an epitope at the N-terminus of mature BACE.

Although BACE protein was immunoprecipitated from transfected COS cells, no BACEΔcyt protein could be detected after immunoprecipitation. However, according to the results obtained before, there was a clear β-secretase activity in COS cells transfected with BACEΔcyt, similar to that seen in BACE-transfected COS cells.

To test whether polyclonal antibody B45.1 was unable to detect BACEΔcyt, we tried to immunoprecipitate BACEΔcyt from transfected COS cells with an anti-myc antibody clone 9E10 provided by Dr. J. Creemers (CME, Leuven), since the protein encoded by pSG5^**-mBACEΔcyt contains a C-terminal myc tag. Still, BACEΔcyt could not be detected. This suggests that the lack of detection of BACEΔcyt in the immunoprecipitate was not due to the antibody. Moreover, antibody B45.1 was shown to detect BACEΔcyt in neurons. We also failed to detect BACEΔcyt in the conditioned medium of transfected cells. It is therefore clear that the levels of expression of BACEΔcyt in COS cells are below the detection limit of our assay, but are nevertheless sufficient to maintain sufficient BACE like proteolytic activity at least in regard to APP processing. At the moment we cannot explain the discrepancy between BACEΔcyt activity and the low levels of expression of the truncated protein. A possibility that has to be further investigated is that BACEΔcyt is rapidly degraded with no detrimental effects on the detected activity. 2) APP processing in neurons by a BACE derivative that lacks the cytoplasmic domain The effect of deleting the cytoplasmic domain on BACE activity was also analyzed in neurons. Cortical neurons were co-transduced with Semliki Forest Virus (SFV) encoding human APP and either mouse BACE or mouse BACEΔcyt. Transduced neurons were metabolically labeled with [³⁵S] methionine and APP processing was analyzed by immunoprecipitation of cell-associated CTFs and Western blotting of sAPPβ, as before.

Expression of APP alone in neurons resulted in APP processing mainly at the α-secretase site, since C83 was the major CTF immunoprecipitated. In contrast to COS, detectable levels of C99 were immunoprecipitated, in agreement with the higher endogenous BACE levels known to exist in neurons. When BACE was co-expressed in neurons, an increase in the amount of APP processed at the 2 β-secretase sites was observed. Concomitantly, high levels of sAPPβ were recovered from the conditioned medium.

In contrast to COS cells, in BACEΔcyt-expressing neurons APP was mainly processed by the non-amyloidogenic pathway, similar to control neurons expressing APP alone. Very little increase (if at all) in sAPPβ was observed upon co-expression of BACEΔcyt. The same is true when immunoprecipitated CTFs were analyzed. In this particular experiment more APP is expressed in BACEΔcyt-expressing neurons what results in slightly higher C99 and C89. However the relative abundance of the fragments resembles more APP processing in the absence of BACE protein. Similar results were obtained in three independent experiments. It is possible that neuron specific factors are required for the activation of BACE most probably through an interaction with the cytoplasmic domain of BACE. BACE and BACEΔcyt were immunoprecipitated from neuronal extracts with antibody B45.1 , that recognizes an epitope at the N-terminus of mature BACE. Surprisingly, in neurons both BACE and BACEΔcyt proteins could be detected by immunoprecipitation using antibody B45.1 , whereas BACEΔcyt could not be detected in COS cells. It is possible that the truncated BACE derivative is rapidly degraded in COS cells in a cell-type specific manner. 3) Identification of proteins interacting with BACEt/c

We have convincingly shown that deleting the cytoplasmic domain of BACE affects APP processing and that the effects are opposed in neuronal versus non-neuronal cells. Other studies have demonstrated that the cytoplasmic domain is important for targeting of BACE to the correct subcellular compartment. It is very likely that the cytoplasmic domain is required for specific protein-protein interactions needed for protein trafficking and activity. To identify such candidate proteins we have set up a yeast two-hybrid screening assay using the cDNA encoding the cytoplasmic domain of BACE as a bait.

A bait plasmid encoding the DNA binding domain of Gal4 (Gal4 BD) connected to the

DNA sequence encoding the cytoplasmic domain of BACE (pBD-GAL4-Bcyt) was constructed.

In preliminary experiments we transformed this plasmid in yeast. It turned out that the Gal4-Bcyt chimeric protein causes auto-activation, i.e. the fusion protein is capable of inducing expression of the HIS3 and lacZ reporter genes, even in the absence of the Gal4 activation domain. The pBD-GAL4-Bcyt phagemid vector was thus unsuitable for detecting protein-protein interactions in this yeast two-hybrid vector system. Therefore another bait plasmid encoding a chimeric protein, consisting of the Gal4 BD and the transmembrane and cytoplasmic domains of BACE (pBD-GAL4-Bt/c) was constructed. The transmembrane and cytoplasmic domains of BACE are depicted in SEQ ID NO 42.

When transformed into yeast cells, pBD-GAL4-Bt/c did not cause auto-activation. With this bait construct we screened a mouse hippocampal cDNA library. Mouse hippocampal mRNA was purified (Triazol) and cloned after reverse transcription into the lamba library HybriZAP, Stratagene. Mass excision in vivo was used to generate the pAD-GAL4 plasmid library, in accordance to the instructions of the manufacturer (Stratagene). This library was chosen because BACE is highly expressed in neurons and the deletion of BACE cytoplasmic domain affects BACE activity in neurons. Moreover, the hippocampus is a region in the brain that is most affected in patients with Alzheimer's disease. Finally, and in contrast to commercially available libraries, this home-made library was unamplified, increasing thereby the chance to pick up a wider range of neuron specific interacting proteins. In a pilot transformation reaction, pAD-GAL4-prey and pBD-GAL4-Bt/c were cotransformed and coexpressed in the YRG-2 yeast host strain. To estimate the amount of individual cDNAs that were screened each time, a small aliquot of the transformants was used to make a serial dilution. Because the pBD-GAL4-Bt/c and pAD-GAL4-cDNA plasmids contain the TRP1 gene and LEU2 gene respectively, double transformants can be selected on medium lacking tryptophane and leucine (SD-L-W). It was calculated that -13% of the mouse hippocampal library was screened with this pilot transformation.

The remaining of the transformants was plated on histidine-negative medium to select for yeast colonies expressing the HIS3 reporter gene. The HIS3 gene should only be expressed when a prey protein interacts with the Gal4 BD-Bt/c fusion protein. The transformants were incubated for 14 days at 30°C and histidine positive colonies were picked and grown on new plates. Colonies were considered as histidine positive when they had a diameter of at least 1 ,5mm after 14 days of growth. Colonies with a diameter smaller than 1 ,5mm were considered as background, a result of the leaky expression of the HIS3 reporter gene.

From this pilot transformation 101 histidine positive colonies were picked and were subsequently tested for the expression of the second reporter gene lacZ with a Filter Lift assay or β-galactosidase assay. As a positive control, a plasmid encoding full- length Gal4 protein was used. As negative control, the pBD-GAL4-Bt/c alone was used.

Of the 101 histidine positive colonies only 30 turned out to be β-galactosidase positive. DNA of this lacZ positive yeast colonies was isolated and used to transform bacteria. The transformed bacteria were grown on ampicilline media. In this way, only bacteria transformed with the pAD-GAL4-prey plasmid (that contains the ampicillin-resistance β-lactamase gene), but not pBD-GAL4-Bt/c (encoding chloramphenicol resistance) were selected. Plasmid DNA prepared from isolated bacterial colonies, was subsequently used for retransformation of YRG-2 yeast cells together with the bait plasmid. All the retransformants turned out to be HIS3 and lacZ positive. To identify the proteins encoded by the prey plasmid, the nucleotide sequence of the DNA was determined and subjected to similarity searches (NCBI BLAST sequence similarity searching) to identify related or homologous sequences. Surprisingly, of 30 HIS3 and lacZ positive yeast transformants, 14 were transformed with a prey plasmid of which the 5' nucleotide sequence consisted of a repeat of the dinucleotides guanine-adenine, that extended for at least 250 nucleotides. These plasmids thus encoded a protein consisting of 2 alternating amino acids: arginine (R) and glutamate (E) at the N-terminus. In the BLAST similarity searching database, no such protein, consisting of a RE-repeat of at least 83 amino acids, could be detected. Positive clones consisting of repeated GA nucleotides were considered as artefacts and therefore were not further investigated.

After the pilot transformation reaction, screening of the hippocampal library was continued. Another 304 HIS3 positive colonies were obtained, from which only 90 turned out to be lacZ positive. In total -50% of the library was screened, yielding 405 HIS3 positive transformants, from which 120 (30%) exhibited β-galactosidase activity.

To avoid sequencing prey plasmids containing the GA-repeat, a colony hybridization was performed on the transformed bacteria using a 5' [³²P] end-labeled probe, complementary to a GA-repeat. From the pilot transformation 3 GA-repeat positive colonies and 3 GA-repeat negative colonies were included as positive and negative controls, respectively. From 90 clones, 27 gave a positive signal, indicating that the isolated pAD-GAL4-prey plasmids contained a GA-repeat.

Therefore, from the 120 HIS3 and lacZ positive transformants a total of 41 (34%) contained a GA repeat and were discarded. From all the sequenced plasmids four corresponded to 3' untranslated regions of known genes, two plasmids were in antisense orientation to described genes and ten were in a different reading frame, although sequencing errors cannot be definitively excluded. The resulting prey proteins vary in length from 9 to 64 amino acids. We performed a comparison of these sequences to search for common motifs and we observed that, in general, most of these proteins are proline-rich (from 15-20% proline).

The remaining 50 sequences corresponded to cDNAs present in the databases. The results are summarized in Table 1. 4) GST pull down assay

Next we confirmed the interaction detected in the yeast two-hybrid system assay by an independent method using GST fusion proteins.

First, a prokaryotic expression vector encoding the cytoplasmic domain of BACE C-terminally fused to Glutathione S-Transferase (pGEX-4T-1-GST-Bcyt) was constructed.

From the different pAD-GAL4-constructs, radiolabeled recombinant proteins were generated using TNT coupled transcription/translation (Promega) in the presence of [³⁵S] methionine. Translated products were incubated with equal amounts of GST-Bcyt immobilized on glutathione beads. After washing, bound proteins were separated by SDS-PAGE and visualized by autoradiography. As a negative control, the proteins were incubated with GST immobilized on glutathione beads. 4.1 ) FKBP38 We have shown that FKBP38 Interacts with the cytoplasmic domain of BACE1 (B1cyt) in a GST-pull down assay. Furthermore FKBP38 also co-immunoprecipitates with full- length BACE1 when both proteins are overexpressed in COS cells. It also co- immunoprecipitates with BACE2, but not with Nicastrin, when both are overexpressed in COS cells. Briefly, FKBP38 was expressed using in vitro translation and then bound to GST or GST fused to either the BACE1 cytoplasmic tail (GST-Bcyt, last 24 amino acids), the transmembrane domain + the cytoplasmic tail of BACE1 (GST-B1TM+cyt) or the cytoplasmic tail of BACE with the leu-leu-lys motif (last 3 amino acids) deleted (GST-BcytdeltaLLK). Binding is observed with the three constructs, in contrast with the negative control. FuGene co-transfected COS cells (candidate protein + BACE1 or BACE2 or Nicastrin cDNAs) and cells transfected with empty vector, are lysed in buffers containing detergents: either directly in DIP buffer or first in 5 mM Tris pH7.4, 1 mM EGTA, 250 mM glucose, 1 % TritonX-100 and then diluted at least 1 :8 in DIP buffer. Cell lysate is centrifuged 20 minutes at -18000 x g to get rid of unbroken cells and cell debris. Part of the lysate is then either immunoprecipitated with antibodies anti-BACE1 , BACE2 or nicastrin or with preimmune serum (pi) in the presence of protein G-Sepharose ON at 4°C with rotation. Beads are washed 5 times with DIP buffer and once with 0.3 x TBS and bound proteins are separated by SDS-PAGE electrophoresis, transferred to a nitrocellulose membrane and probed with anti-HA antibody. 4.2) Arsenite transporter (AT)

We have shown that AT Interacts with B1cyt in GST-pull down assays. AT co- immunoprecipitates with full-length BACE1 and probably BACE2 (negative control in this case is also positive) when both proteins are overexpressed in COS cells. No co- immunoprecipitation with overexpressed Nicastrin (a membrane protein) is detected

4.3) Neutral Sphingomyelinase (nSMase)

Co-immunoprecipitates with full-length BACE1 but not with Nicastrin when both proteins are expressed in COS cells.

4.4) Neural membrane protein 35/Lifegard (NMP35) NMP35 co-localizes with BACE1 when both proteins are expressed in COS cells.

Materials and methods This invention relies on routine techniques in the field of recombinant genetics. Basic texts disclosing the general methods of use in this invention include Sambrook et al., Molecular Cloning, A Laboratory Manual (2^nd ed. 1989); Kriegler, Gene Transfer and Expression: A Laboratory Manual (1990); and Current Protocols in Molecular Biology (Ausubel et al., eds. 1994)). For nucleic acids, sizes are given in either kilobases (kb) or base paires (bp). These are estimates derived from agarose or acrylamide gel electrophoresis, from sequenced nucleic acids, or from published DNA sequences. For proteins, sizes are given in kilodaltons (kDa) or amino acid residue numbers. Protein sizes are estimated from gel electrophoresis, from sequences of proteins, from derived amino acid sequences, or from published protein sequences. Oligonucleotides that are not commercially available can be chemically synthesized according to the solid phase phosphoramidite triester method first described by Beaucage & Caruthers, Tetrahedron Letts. 22: 1859-1862 (1981 ), using an automated synthesizer, as described in Van Devanter el al., Nucleic Acids Res. 12: 6159 (1984). Purification of oligonucleotides is by either native acrylamide gel electrophoresis or by anion- exchange HPLC as described in Pearson & Reanier, J. Chrom. 255: 137 (1983). The sequence of the cloned genes and synthetic oligonucleotides can be verified after cloning using, e.g. the chain termination method for sequencing double stranded templates of Wallace et al., Gene 16: 21 (1981 ). 1) Cell culture

COS and BHK cells were cultured in DMEM:F12 (GibcoBRL) supplemented with 10% fetal calf serum (FCS). Mouse cortical neurons were cultured in neurobasal medium (GibcoBRL) supplemented with 250μl 200mM L-glutamine (GibcoBRL) and 2ml B27 Serum-free Supplement (GibcoBRL) per 100ml neurobasal medium.

2) Transient transfection of COS cells Foreign DNA can be transiently introduced into cultured COS cells using the FuGENE 6 Transfection Reagent (Roche), a lipid-based transfection reagent that complexes with and transports DNA into the cell during transfection.

A transfection mix with a FuGENE:DNA ratio of 3:1 (in μl and μg respectively) was added to a culture dish containing a monolayer of COS cells that was 50-80% confluent. This cell density was obtained by plating -700.000 cells in 4ml 10% FCS- containing DMEM :F12 medium (GibcoBRL) in a 60mm culture dish one day before transfection.

When COS cells were grown in 60mm culture dishes, 6μl of FuGENE 6 Reagent and 2μg DNA were used for transfection. The FuGENE was diluted in 94μl of serum-free DMEM :F12 medium in a sterile 1 ,5ml tube by directly pipetting into the medium, thereby preventing the reagent to contact the plastic tube wall, for this adversely affects the transfection efficiency. The contents were gently mixed by tapping. 2μg DNA were then added to the tube. For cotransfection with 2 plasmids, 1 μg of each plasmid was used. The DNA-FuGENE solution was gently mixed and incubated for -30 minutes at room temperature to allow the FuGENE 6 Reagent to complex with the DNA.

The medium of the COS cells was changed for 4ml fresh DMEM :F12 medium containing 10% FCS and the FuGENE-DNA mixture was added to the fresh medium. To ensure even dispersal the plates were gently swirled. The cells were kept in a 37°C incubator (0,5% C0₂) for 48 hours, refreshing once the medium 24 hours after transfection.

3) In vivo packaging of recombinant RNA into SFV particles

All handlings of SFV particles were performed under standard safety conditions in a L2 lab. High-level expression of recombinant proteins in non-dividing, postmitotic cells such as cortical neurons can be achieved by infecting the neurons with Semliki Forest Virus (SFV) particles that carry recombinant-derived RNA.

Recombinant SFV particles were generated by introducing 2 naked RNA molecules, obtained by in vitro transcription of a recombinant pSFV1 and a pSFV-HelpeM plasmid into BHK cells by electroporation. Recombinant pSFV1 plasmid encodes non-structural viral proteins required for SFV replication, but lacks the DNA region encoding structural proteins that is replaced by the heterologous insertion of foreign DNA. pSFV-Helperl contains the DNA region encoding SFV structural proteins, needed for assembly of viral particles. Translation of both recombinant SFV1 RNA and Helperl RNA within one BHK cell leads to new virus particles that carry recombinant-derived RNA that can subsequently be used to infect cortical neurons.

2 μg of a recombinant pSFV1 plasmid were linearized with 4U Spel, that cuts the vector at a unique site and the plasmid was subsequently purified. An in vitro transcription mix was prepared in an RNAse-free microcentrifuge tube and incubated at 37°C. In vitro transcription mix:

Component Amount per reaction

Plasmid 2μg

10x Transcription Buffer (Boehringer Mannheim) 1 x m⁷G(5')ppp(5')G 1 mM

DTT 5mM

RNasin (Boehringer Mannheim) 60U dNTP's 0,1 mM each dNTP

SP6 Polymerase (42,9U/μl) 30U

Total reaction volume 50μl

After 1 hour an additional 15U of SP6 DNA polymerase were added and the tube was incubated for another 30 minutes at 37°C. A 2μl aliquot was tested on an agarose gel and the rest was frozen at -70°C in 3 aliquots of 16μl each.

The required amount of BHK cells for infection was obtained by plating a 5 times dilution of a confluent monolayer one day before the infection in a 75cm² Falcon bottle. On the day of the infection, the cells were detached by adding 2,5ml trypsin-EDTA (GibcoBRL). After detaching, 7,5ml DMEM:F12 were added and the BHK cells were then harvested by centrifugation. After washing twice with 2ml PBS, the cells were resuspended in 700μl PBS and added to a mix of 16μl in vitro transcribed Helperl RNA and 16μl recombinant SFV RNA.

The RNA-BHK cell mix was transferred to an electroporation cuvette (EUROGENTEC) and electroporation was carried out twice (BIORAD gene pulser; 0,85V, 0,3-0,4s). The electroporated cells were transferred to 20ml 10%FCS- containing DMEM:F12 and incubated at 37°C for 4 hours.

In a L2 lab, the medium was changed for 7,5ml 10%FCS containing DMEM:F12 and incubated overnight at 37°C. The medium was centrifuged to get a cell-free virus containing supernatant, aliquoted in 500μl fractions and immediately frozen in liquid nitrogen and stored at -70°C. Reagents:

* 10x Transcription Buffer

0,4M Tris-HCI, pH8.0 60mM MgCI₂

100mM DTT 20mM spermidine

4) SFV infection of neuronal cells

125μl of supernatant, containing the recombinant virus particles were added to 60mm culture dishes containing cortical neurons and 1 ,25 ml of the neurobasal medium. Infection proceeded for 1 hour at 37°C. After 1 hour the medium was changed for fresh 1 ,2ml neurobasal medium and the neurons were incubated for 2 hours at 37°C.

5) Metabolic labeling Labeling of COS cells: 48 hours post-transfection the cells were washed with 3ml methionine-free medium, starved for methionine for 30 minutes in 1 ,2ml of this medium and then labeled for 4 hours at 37°C by adding 12μl [³⁵S]-methionine (100μCi/ml) to the medium. Labelling of neurons: Three hours post-infection, the medium was changed for 1 ,2ml methionine-free medium, containing 12μl [³⁵S]-methionine (100μCi/ml). The neurons were labeled for 4 hours at 37°C. Reagents:

* Methionine-free Medium (1 liter)

9,954g Dulbecco's MEM without methionine (GibcoBRL) 3,43g TES

3,57g HEPES 13,4ml HC0₃ ^" 10ml 100x L-Glutamine 6) Immunoprecipitation

To prepare cellular extracts from COS cells, the cells were first washed with 3ml PBS and then scraped in 1 ml PBS containing a protease inhibitor cocktail (1 mM EDTA,

14μg/ml aprotinine, 2μg/ml pepstatin). Cells were harvested at 1500xg for 10 minutes and lysed by incubating them for 20 minutes on ice in 200μl immunoprecipitation buffer containing protease inhibitors and 1 % Triton-X-100, to solubilize membrane proteins.

Cellular debris were removed after centrifugation at 18000xg for 15 minutes and 400μl

DIP buffer were added to this supernatant.

To make neuron extracts, neurons were washed in 1 ml PBS, directly scraped in 600μl DIP buffer and incubated for 10 minutes on ice. Cellular debris were removed by centrifugation.

To immunoprecipitate proteins, 25μl Protein G Sepharose 4 Fast Flow (Pharmacia

Biotech) and the appropriate antibody were added to the cell lysate and incubated overnight at 4°C with rotation. The sepharose beads were washed 5 times with 1 ml PBS and once with 0,3x TBS.

Reagents:

* Immunoprecipitation Buffer, pH7.4

5mM Tris-HCI, pH7.4 250mM sucrose I mM EGTA

* 1x DIP Buffer, pH7.4 (1 liter)

100ml 10x TBS 10ml Triton-X-100 10g sodium deoxycholate 1g SDS Ox TBS, pH 7.5 1500mM NaCI 200mM Tris-HCI

Antibodies

Rabbit polyclonal B11.4 has been raised against the last 20 amino acids of APP Rabbit polyclonal B45.1 has been raised against amino acids 46-61 of mouse BACE Qust C-terminal to the propeptide) 7) Yeast two-hybrid screening

The yeast two-hybrid system is an eukaryotic system used to detect protein-protein interactions in vivo. This system is based on the fact that many eukaryotic transcriptional activators are modular proteins composed of 2 separable functional domains: a DNA-binding domain (BD) and a transcriptional activation domain (AD). The DNA-binding domain binds a specific DNA sequence, the Gal4 Upstream Activating Sequence (UAS) and the activation domain interacts with components of the transcription machinery to initiate transcription of a gene downstream of the Gal4-UAS. In the HybriZAP™ two-hybrid vector system a protein of interest, called bait is expressed in yeast as a fusion to the DNA-binding domain of the Gal4 protein. Prey proteins are expressed as a fusion to the activation domain of the Gal4 protein. Neither of the BD nor AD can initiate transcription when separated and each domain continues to function when fused to other proteins. The pBD-GAL4-bait and pAD-GAL4-prey plasmids are transformed and co-expressed in a yeast host. When the bait and prey proteins interact, the Gal4 protein is reconstituted. The BD localizes the complex to a reporter gene Upstream Activating Sequence and the AD directs RNA polymerase II to transcribe this reporter gene. In the HybriZAP™ two-hybrid system the YRG-2 host strain is used. This host strain contains two reporter genes under the control of the Gal4-UAS: a nutritional HIS3 gene and an enzymatic reporter gene, lacZ, encoding β-galactosidase. To select for the expression of HIS3, the yeast transformants are plated on histidine-negative SD plates. Induction of HIS3 gene enables yeast transformants to grow on medium lacking histidine. Because the Gall promoter, which governs the expression of the HIS3 reporter gene, is a bit leaky, the expression of the second reporter gene lacZ, under the control of the iso-1 -cytochrome c promoter, is subsequently tested with a filter lift assay, to confirm the bait-prey protein interaction.

We used as bait plasmid pBD-GAL4-Bt/c, encoding the Gal4 protein fused to the transmembrane and cytoplasmic domains of BACE and as prey plasmid pAD-GAL4- mouse cDNA hippocampal library, encoding the Gal4 activation domain fused to proteins of a mouse hippocampal cDNA library.

8) Yeast transformation

Plasmid DNA was introduced in YRG-2 using a lithium acetate treatment.

A single YRG-2 yeast-colony (± 3mm diameter) was picked from a fresh YPAD-plate and grown in 2ml of YPAD medium at 30°C with shaking at 250rpm (Innova™ 4900, New Brunswick Scientific) for 8 hours. 25μl of this starter culture were transferred to 50ml of YPAD broth and grown at 30°C, until the culture reached an OD600 (GeneQuantpro, Pharmacia Biotech) of 0,8-1 ,0, which corresponds approximately to 2.10⁷ cells per ml (± 18 hours). The yeast cells were then harvested by centrifuging at 800xg for 5 minutes at room temperature, washed with 25ml sterile water, washed with 1 ml 0,1 M LiAc and resuspended in 0,1 M LiAc (YEASTMAKER™ Yeast Transformation System, Clontech) to a final volume of 500μl.

To increase transformation efficiency, the cells were divided into 10 microcentrifuge tubes, with each tube containing about 50μl of the cell suspension. After pelleting the yeast at 800xg for 5 minutes at room temperature, the LiAc solution was aspirated. Competent YRG-2 cells were then transformed as follows: 240μl 50% PEG (Clontech), 36μl 1 M LiAc and 25μl 2mg/ml denatured carrier DNA (Salmon Sperm DNA, Sigma) were added in this order. The PEG solution was added first, because it shields the cells from the detrimental effects of the concentrated LiAc. The carrier DNA was denatured by boiling at 99 °C (Thermomixer compact, Eppendorf) for 10 minutes and immediately chilled on ice. Finally, 50μl of a 2μg/μl plasmid mix, containing pBD-GAL4- Bt/c and pAD-GAL4-hippocampal cDNA library in a molar ratio of 2 to 1 , were added. The tubes were thoroughly vortexed (Labinco) for 1 minute. The transformation mix was incubated at 30°C for 30 minutes and then heat shocked at 42°C for 30 minutes, gently mixing the cells by inversion every 5 minutes. The cells were collected by centrifugation at 800xg for 5 minutes at room temperature and gently resuspended in 8ml sterile water. 12μl were removed for estimation of the total amount of cDNA clones screened and the rest was plated onto δ plates (24,3cm x 24,3cm) containing 250ml SD-L-W-H. The pBD pBD-GAL4-Bt c vector and pAD-GAL4-cDNA library vector contain the TRP1 and LEU2 gene, respectively, to select for double transformant on SD media lacking W and L. Histidine was omitted to select for yeast transformants that express the HIS3 reporter gene and therefore encode candidate BACEt/c interacting proteins. The plates were incubated for a maximum of 14 days at 30°C.

Reagents:

* YRG-2 yeast host strain

Matα ura3-52 his3-200 ade2-101 Iys2-801 trp1-901 leu2-3 112 gal4-542 gal80- 538 LYS2: : UAS_GALI-TATAGALI-HIS3 U RA3: :UAS_GAL4 i7mers(x3)-TATA_Cγcr/acZ * YPAD agar/broth, pH5.δ (per liter)

20g Difco Tryptone Peptone (Becton Dickinson) 10g Difco Yeast Extract (Becton Dickinson) 15-20g Difco Agar Technical (Becton Dickinson) deionized water to a final volume of 950ml

40mg adenine sulfate (Sigma) After autoclaving 50ml of a filter sterilized 40% glucose solution were added.

^* SD agar/broth, pH5.δ (per liter)

6,7g Difco Yeast nitrogen base without amino acids (Becton Dickinson) 1δ2,2g D-sorbitol (Sigma)

15-20g Difco Agar Technical (Becton Dickinson) deionized water to a final volume of δ50ml

After autoclaving ampicillin (400μg/ml broth, 2mg/ml agar), 100ml 10x Triple

Dropout Solution and 50ml of a filter sterilized 40% glucose solution were added.

^* 10x Triple Dropout Solution (-L-W-H), (per liter)

300mg L-isoleucine 1500mg L-valine

200mg L-Adenine hemisulfate salt 200mg L-Arginine HCI

300mg L-Lysine HCI 200mg L-Methionine 500mg L-Phenylalanine 2000mg L-Threonine 300mg L-Tyrosine

200mg L-Uracil 1000mg L-Glutamic acid 1000mg L-Aspartic acid 400mg L-Serine All amino acids were bought from Sigma.

9) Estimation of the total number of library cDNAs screened

The 12μl aliquot of the transformants were diluted in water as follows:

10μl transformants + 90μl water (1/10)

1 μl transformants + 99μl water (1/100) 1μl transformants + 999μl water (1/1000)

100μl of each of these dilutions were plated on 30ml SD-L-W-H plates, on which 60μl 500x histidine were added to obtain SD-L-W plates.

The colonies on the 3 plates were counted and the mean amount of colonies per μl transformants mix was calculated to estimate the total amount of transformants of the transformation reaction and the percentage of mouse cDNA hippocampal library screened. 10) Filter lift assay

14 days after the transformation, histidine positive yeast colonies were transferred to SD-L-W-H plates and grown for 3-7 days. A replica of the colonies was made on a filter paper (Hybond™ - C Extra, Amersham Pharmacia). For this, orientation-marked filter paper was placed carefully on the surface of the SD plate, thereby avoiding to make air bubbles. After 2-3 minutes the filter paper was lift from the plate using forceps and dipped colony side up in liquid nitrogen for 10 seconds to permeabilize the yeast cells. The filter paper was thawn on whatman paper and the permeabilization procedure was repeated once. The filter was then placed colony side-up onto a whatman paper soaked in 1 ,5ml Z-buffer containing X-gal (Stratagene) in a 90mm petri dish. X-gal (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside) is a chromogenic substrate of β-galactosidase which produces a dark blue precipitate on enzymatic hydrolysis. The petri dish was sealed with parafilm and incubated colony side up for up to 24 hours at 30°C. Reagents: * Z Buffer, pH7.0 (per liter)

10,7g Na₂HPO₄2H₂O 6,12g NaH₂P0₄2H₂0

0,75g KCl

0,246g MgS0₄7H₂0 deionized water to a volume of 1 liter autoclaved

Z Buffer with X-gal (100ml) 9δml Z Buffer

0,27ml 700mM β-mercaptoethanol 1 ,67ml X-gal (20mg/ml) * 700mM β-mercaptoethanol

5,55μl 12,6M β-mercaptoethanol 95μl of DE PC-treated water

11) Isolation of DNA from yeast cells Yeast cells that were positive for HIS3 and lacZ gene expression, were picked from a plate with a pipette tip and resuspended in 200μl extraction buffer in screw capped microcentrifuge tubes. After adding 400μl acid washed glass beads and 200μl phenol- chloroform-isoamylalcohol (25:24:1 ) the tubes were vigorously vortexed for 7 minutes and centrifuged at maximum speed for 10 minutes at room temperature. To precipitate the DNA, 140μl of the upper aqueous layer containing the DNA, were mixed with 500μl of an ethanol:7,5M NH₄Ac (6:1 ) solution, incubated at -70°C for 20 minutes and centrifuged for 15 minutes at 17500xg. The DNA pellet was washed with 100μl 70% ethanol, air-dried and resuspended in 5 μl H₂0. The DNA was subsequently used for transformation of MAX Efficiency^® DH5α™ Competent Cells (Life Technologies). Reagents:

* Extraction Buffer

10mM Tris-HCI, pHδ.O l OOmM NaCI 1 mM EDTA

2% Triton-X-100 1 % SDS

12) GST pull down assay

12.1 In vitro transcription/translation The DNA insert of a pAD-GAL4-prey plasmid encoding a putative BACEt/c-interacting protein was amplified using a 5' primer complementary to the GAL4-AD 3' end, that contained an in frame ATG codon and just upstream a T3 promoter. As 3' primer an oligonucleotide complementary to the 3'end of the multiple cloning site of pAD-GAL4- prey plasmid and containing stop codons in the three reading frames was used. The PCR reaction mixture was set up and the reaction proceeded with following cycling parameters:

Temperature (°C) Duration (s) Number of cycles

95 300 1

95 30 65 30 72 240

95 30 62 30 72 240

95 30 60 30 72 240

95 30 56 30 30 72 240

72 420 1

The amplified PCR fragments were purified and directly used in the TNT Coupled Reticulocyte Lysate System (Promega). For this, the DNA template and the reaction components were assembled in an RNase-free microcentrifuge tube, gently mixed by pipetting and the reaction proceeded for 90 minutes at 30°C. Reaction mixture:

Component Amount per reaction (μl)

TNT^® Rabbit Reticulocyte Lysate 12,5

TNT^® Reaction Buffer 1

TNT^® T3 RNA Polymerase 0,5

Amino Acid Mixture, Minus methionine, 1 mM 0,5

³⁵S Methionine (100μCi/ml) 1

RNasin^® Ribonuclease Inhibitor (40U/μl) 0,5

DNA template 1

Nuclease-free water To a final volume of 25 A 2μl aliquot of the translation reaction was removed and loaded on a 10% MOPS NuPAGE gel (see 3.2.7) to test the efficiency of the translation. The gel was fixed, dried and exposed to a Phosphor Imaging screen. Reagents: ^* Primers

5' primer: 5' AAGCTCGAAATTAACCCTCACTAAAGGGAAGTTTAAGTTTAATA CCACTACAATGGATGATG 3'

3' primer: 5' CTAACTAACTAGAGTCGACCCGGGGTCG 3' 12.2 Isolation of GST fusion proteins Glutathione S-Transferase (GST) is a protein that strongly interacts with reduced glutathione, even when it is N-terminally fused to a protein. When a GST-fusion protein is incubated with glutathione molecules, that are immobilized on sepharose beads, the chimeric protein will bind to glutathione and can subsequently be purified. Proteins that interact with the GST-fusion protein can be pulled down, when incubated with GST- fusion protein bound to glutathione sepharose beads.

Plasmids encoding GST fusion proteins were introduced in BL21 competent cells (Merck Eurolab). Expression of the GST-fusion proteins can be induced by IPTG, a lactose analogue. A colony of the transformed BL21 cells was picked and grown in 20ml LB-ampicillin overnight at 30°C with shaking at 250rpm (Innova™ 4900, New Brunswick Scientific). This starter culture was transferred to 500ml LB-ampicillin. The culture was incubated at 37°C until the OD600 (GeneQuant pro, Pharmacia Biotech) reached 0,4-0,5. To induce expression of GST fusion proteins, IPTG (Promega) was added to the culture to a final concentration of 0,5mM. After 2 hours incubation at 30°C, the culture was centrifuged for 15 minutes at 2500xg. The bacterial pellet was resuspended in 10ml TS buffer, containing a protease inhibitor cocktail (1 mM EDTA, 14μg/ml aprotinine, 2μg/ml pepstatin). After breaking the bacterial cells by sonication, Triton-X-100 was added to 1 % final concentration and the cell lysate was incubated for 15 minutes at 4°C with rotation. The insoluble debris was removed by centrifugating twice at 12500rpm for 20 minutes (Beckman J2-21 M/E).

In a pilot linkage reaction, 20μl glutathione sepharose beads 4B (Amersham Pharmacia) were incubated with 100μl bacterial extract for 30 minutes at 4°C with rotation. The beads were washed twice with TBS containing 1 % Triton-X-100 and subsequently washed twice with TBS. Bound proteins were eluted in 1x NuPAGE LDS sample buffer, denatured for 10 minutes at 70°C and loaded on a 4-12% MES NuPAGE gel. After electrophoresis, the gel was incubated for 20 minutes in Coomassie Brilliant Blue R-250 (Serva) and destained overnight in 30% methanol-10% CH₃COOH. From this pilot reaction, the volume of supernatant needed to obtain 5-1 Oμg of GST- fusion protein bound to beads was estimated. 12.3 GST pull down assay

10μl of the TNT reaction were incubated with 5-1 Oμg of either GST alone or GST- fusion proteins bound to glutathione sepharose beads. The mixture was incubated overnight at 4°C with rotation. The beads were washed twice with TBS containing 1% Triton-X-100 and subsequently washed twice with TBS. Bound proteins were separated by SDS-PAGE. The gel was fixed for 1 hour, dried and exposed to a Phospor Imaging screen (Molecular Dynamics). 13. Description of the vectors * pSG5^**-mBACE pSG5^**-mBACE contains the full length mBACE cDNA inserted into the Sstll-Sstl sites of the pSG5^**vector. pSG5^** derives from the commercially available pSG5 (Stratagen), but the MCS was modified to: EcoRI, Spel, Sstll, Hindlll, Notl, Xhol, Smal, Sstl, BamHI, Bglll This was done by cloning synthetical synthesized oligonucleotides.

* pGEM-T-mBΔcyt

The sequence encoding mouse BACE ectodomain and transmembrane domain was PCR amplified using pSG5^**-mBACE as template and primers 27 (sense) and 132 (antisense). The PCR fragment was directly ligated to pGEM-T vector (PROMEGA). PGEM-T multiple cloning site: Apal, Aatll, Sphl, Ncol, Sacll, Spel, Notl, Pstl, Sail,

Ndel, Sad. primer 27: 5' GGATTCATGGCCCCAGCGCTGCACTGGCT 3' primer 132: 5' GCGAGCTCCTAAGCTGCCAAGTCCTCTTCAGAAATGAGCTTTTG CTCTACCATGAGGCAGAGTGGCAACATG 3' * pSG5^**-mBΔcyt

The Ncol-Spel fragment of pGEM-T-mBΔcyt (after Klenow treatment) was cloned into the Smal site of the pSGδ^" vector.

* pSFV1-hAPP (De Strooper et al., EMBO J.14, 4932-4936, 1995). * pSFV1 -mBACE

The Spel-BamHI fragment of pSG5^**mBACE (after Klenow treatment) was cloned into the Smal site of the pSG5^" vector. * pSFV1-mBΔcyt

The Ncol-Spel fragment of pGEM-T-mBΔcyt (after Klenow treatment) was cloned into the Smal site of pSFV1 vector.

* pBD-Gal4-Bcyt

Annealed oligonucleotides (encoding nucleotides 1432 to 1506 of mBACE cDNA) were inserted into the EcoRI-Sall site of the pBD-GAL4-Cam vector. sense oligo153: 5' AATTCTGTCAGTGGCGCTGCCTGCGTTGCCTGCGCCACCAG

CACGATGACTTTGCTGATGACATCTCCCTGCTCAAGTAAG 3' antisense oligo 154: 5' TCGACTTACTTGAGCAGGGAGATGTCATCAGCAAAGTC

ATCGTGCTGGTGGCGCAGGCAACGCAGGCAGCGCCACT GACAG 3'

* pBD-GAL4-Bt/c

A PCR fragment encoding BACE transmembrane and cytoplasmic domains (from nucleotides 1381 to 1506 of mBACE cDNA) was amplified using pSG5"mBACE as template and primers 199 and 200. sense primer 199: 5' CGGAATTCGTCATGGCGGCCATCTGCGCC 3' antisense primer 200: 5' TCCCCCGGGTTACTTGAGCAGGGAGATGTCATC 3' The PCR fragment was digested with EcoRI and Smal and ligated to an EcoRI- Smal digested pBD-GAL4-Cam vector.

* pGEX-4T-1 -GST-Bcyt Annealed oligonucleotides 153 and 154 (from nucleotides 1432 to 1506 of mBACE cDNA) were inserted into the EcoRI-Sall sites of pGEX-4T-1 vector

Claims

1. A BACE-interacting polypeptide comprising a BACE interacting domain capable of interacting with the transmembrane and cytoplasmic region of BACE.

2. A BACE-interacting polypeptide according to claim 1 selected from the group essentially consisting of sequences comprising SEQ ID NO: 2, 4, 6, 8, 10, 12, 14,

16, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38 or 40 and homologues thereof.

3. An isolated nucleic acid encoding for a BACE-interacting polypeptide comprising a BACE interacting domain.

4. An isolated nucleic acid according to claim 3 selected from the group essentially consisting of sequences comprising SEQ ID NO: 1 , 3, 5, 7, 9, 11 , 13, 15, 17, 19,

21 , 23, 25, 27, 29, 31 , 33, 35, 37 or 39 and homologues thereof.

5. A method for identifying compounds which can modulate the expression of a gene encoding a BACE-interacting protein according to claims 3-4 comprising the steps of: - contacting a cell with at least one compound wherein said cell includes a regulatory region of a gene encoding a BACE-interacting protein operably joined to a coding region, and detecting a change in expression of said region.

6. A method of identifying compounds which can modulate the activity of a BACE- interacting protein comprising the steps of: providing a cell expressing a normal or mutant gene encoding a BACE- interacting protein according to claims 3-4, and contacting said cell with at least one candidate compound, and detecting a change in a marker of said activity.

7. A method of identifying compounds which can modulate the interaction between a BACE-interacting protein or a functional fragment thereof and BACE comprising the steps of: providing a cell expressing a normal or mutant gene encoding a BACE- interacting protein and a normal or mutant gene encoding BACE, and - contacting said cell with at least one candidate compound, and detecting a change in the interaction.

8. A diagnostic method for determining if a subject bears a mutant gene encoding a BACE-interacting protein comprising the steps of: providing a biological sample of said subject, and detecting in said sample a mutant nucleic acid encoding a BACE-interacting protein, a mutant BACE-interacting protein, or a mutant BACE-interacting protein activity.

9. The nucleotide sequences encoding BACE-interacting proteins of claims 3 and 4 and the BACE-interacting proteins of claims 1 and 2 as a medicament.

10. Use of the nucleotide sequences of claim 9 for the manufacture of a medicament to treat Alzheimer's disease.