EP1095279A1 - Method for identifying a presenilinase inhibitor - Google Patents

Abstract

The present invention pertains to methods for identifying a substance capable of reducing or eliminating the activity of the presenilinase wherein a cell or a cell line is cultivated expressing said presenilinase activity and a fusion-protein comprising the full-length presenilin 1 or presenilin 2 and a reporter is measured. The invention is furthermore concerned with substances identifiable with said methods, pharmaceutical compositions comprising said substances and the use of said substances in the manufacture of a medicament for the treatment of neurodegenerative diseases or Alzheimer's disease.

Description

Method for identifying a presenilinase inhibitor

Technical Field of the Invention

The present invention belongs to the field of presenilins and neurodegenerative diseases. More particularly, the present invention provides methods for the identification of presenilinase inhibitors, substances identifiable with said methods, their use in the manufacture of a medicament for the treatment of neurodegenerative diseases and pharmaceutical compositions comprising said substances.

Background Art

Neurodegenerative diseases are characterized by neuronal and synaptic cell loss. Neuronal cell loss is caused at least in part by apoptotic cell death. Neurodegenerative diseases include the chronic forms as Alzheimer's disease (AD), Parkinson's disease, Huntington's chorea and acute forms as stroke. The majority of Alzheimer's disease cases are late in onset so far lacking an obvious genetic linkage and are characterized as sporadic, whereas a small percentage (approximately 10%) of cases belonging to the subgroup of familiar Alzheimer's disease (FAD) are earlier in onset and segregate strongly within families suggesting a genetic etiology. AD is a neurodegenerative disorder marked by the gradual formation of extracellular neuritic plaques in the brain, particularly in the hippocampus and the adjoining cortex. In AD research, one small peptide has long claimed a large share of attention. Known as β amyloid (Aβ), it is the major constituent of the abnormal structures called 'amyloid plaques' that stud the brain of AD patients. Mutations in the gene that encodes the amyloid precursor protein (APP), which is cleaved by two putative secretases (β- and γ-secretase) to release Aβ, account for some inherited cases of the disease (Chartier-Harlin et al., 1991; Goate et al., 1991; Murrell et al., 1991; Hendriks et al., 1992; Mullan et al., 1992).

The discovery in 1995 that a new gene family - the presenilins - is responsible for the majority of early-onset autosomal dominant cases of familial AD has led to the expectation that a fundamental understanding of the disease mechanism may not be far off (Levy-Lahad et al., 1995; Rogaev et al., 1995; Sherrington et al., 1995). It has been shown, both in vivo, in fibroblasts and plasma of FAD patients as well as in transgenic animals and cell lines that the presenilin (PS) mutations cause a specific increase in the production of extracellular Aβ42, the long form of Aβ ending at residue 42 (Borchelt et al , 1996, 1997, Duff et al , 1996, Scheuner et al , 1996) Aβ42 was shown to be deposited early and selectively in the disease process and to be more fibrillogenic in vitro than the more prevalent species of Aβ ending at residue 40, termed Aβ40 (Jarret et al , 1993, Mann et al , 1996)

Several reports described the proapoptotic behaviour of PS based on data observed in cells transiently or stably overexpressing PS (Deng et al , 1996, Vito et al , 1996, Wolozin et al , 1996, Janicki et al , 1997, Kim et al , Science 1997, 277 373-376, Loetscher et al , 1997) Overexpression of PS2 increases the susceptibility of neurons to apoptotic stimuli and thus lead to neuronal death (Kim et al , Science 1997, 277 373-376)

The presenilins undergo regulated proteolytic cleavage into the normal N-terminal (NTF) and C- terminal fragments (CTF) These fragments are approximately 21-28 kDa (PS1 NTF), 28-30 kDa (PS2 NTF), 16-24 kDa (PS1 CTF) and approximately 20-25 kDa (PS2 CTF) respectively, in size (Okochi et al , 1997, Haas et al , 1998, Kim et al , Science, 1997, Thinakaran et al 1996, Podlisny et al , 1997)

Furthermore, the PS proteins constitute substrates of a member of the caspase 3 protease family (CPP32) after its activation late in the course of apoptosis and are cleaved into alternative fragments (Kim et al , Science 1997, 277 373-376, Loetscher et al , 1997) According to Kim et al (Science 1997, 277 373-376), these alternative fragments of PS2 are a 20 kDa CTF and a 34 kDa NTF In our hands, this alternative CTF fragment of PS2 has the molecular weight of 16 kDa and thus is subsequently termed CTF^g The cleavage region of normal proteolytic PS2 cleavage is located within the distal region of exon 10 (originally called exon 9) between Met²⁹⁸ and Ala³⁰⁵, whereas the suspected region of alternative cleavage is located within the PS2 loop encoded by exon 1 1 after Asp³²⁶ or Asp³²⁹ (Kim et al , Science 1997, 277 373-376) According to WO 98/47917 and Kim et al (Science 1997, 277 373-376), cleavage of the PS1 polypeptide in cells overexpressing PS 1 (molecular weight of 45 kDa) yields a 23 kDa normal, endogenous CTF and a 14 kDa alternative CTF

To date, neither the enzymatic activity leading to cleavage of the presenilins, the activity of the so-called presenilinase nor the enzymatic activity leading to the cleavage of APP C-terminal fragments C99 and C83 into Aβ, the activity of the γ-secretase, has been identified The presenilins may be cofactors for the activity of the γ-secretase or even γ-secretases itself by being autoactivated aspartvl proteases with γ-secretase activity (Wolfe et al , 1999) Alternatively, presenilins may be trafficking proteins, and transport the substrates APP-C99 and C83 into cellular compartments where they get cleaved by the putative γ-secretase.

There is evidence that the proteolytic fragments are the biologically active form of the presenilins (Thinakaran et al., 1996, Podlisny et al., 1997). Thus, if said proteolytic fragments constitute the active form of presenilins, they also may be either directly or indirectly involved in γ-secretase activity and therefore in the generation of Aβ.

The endogenous presenilin protein level only consists of these proteolytic fragments. In the "steady state" the PS-protein does not exist in its full length indicating that immediately after expression the full-length presenilins get cleaved into stable fragments. Due to the overexpression of the presenilins the level of proteolytical fragments is increased until saturation level. When this saturation level is exceeded, no more proteolytical cleavage takes place and the full-length protein is detectable, i. e. in cells overexpressing presenilin, fragments are detected as well as, according to the degree of overexpression, the full-length protein. An increase in full-length PS may however, lead to increased degradation of said full-length PS (e.g. in the proteasomal pathway) and finally result in a reduction of the quantity of full-length PS (see also Steiner et. al, 1998). A deletion of the PS1 gene in mice leads to a significantly reduced Aβ secretion in neurons derived from PS1 1 knock-out mouse embryos (De Strooper, B. et al., 1998). Currently, there exists only symptomatic treatment of neurodegenerative diseases and in particular of AD. However, there is no disease-modifying treatment to cope with the pathology of said diseases. At present, there is no therapeutic way of preventing the pathology due to the deposition of Aβ and the subsequent formation of 'amyloid plaques'. Also, there is no therapeutic way of preventing the neuronal cell death due to normal or alternative PS fragments which then may cause apoptosis. The suppression of the generation of said PS fragments which may be the biologically active form of the presenilins and subsequently preventing the deposition of Aβ may be achieved by reducing or eliminating the activity of the presenilinase. Reducing or eliminating the activity of the presenilinase should go along with a concomitant increase in the full-length protein, a decrease in the biologically active fragments and thus may lead to the reduction of Aβ and/or may also prevent apoptotic cell death. Therefore, said presenilinase constitutes an excellent target in the treatment of neurodegenerative diseases, in particular AD. It is thus necessary to develop methods to identify highly specific substances capable of reducing or eliminating said presenilinase activity Additionally, it is important to use said methods to find substances specifically capable of inhibiting the presenilinase activity and employ these substances in the treatment of neurodegenerative diseases, in particularly AD.

WO 97/41443 discloses the screening for the identification of compounds which inhibit or otherwise modulate the cleavage of PS1 comprising the measurement of PS1 into 18 kDa and 28 kDa species.

WO 98/47917 discloses a 20 kDa PS2-CTF fragment resulting from alternative cleavage, an antibody specific for said fragment and a method for screening compounds that inhibit proteolytic processing of PS2 comprising providing a compound to a cell proteolytically processing PS2 and the measurement of said fragment in said cell.

WO 97/27296 describes the presenilin-interacting proteins S5a, GT24, p0071, Rabl l, retionoid X receptor-β, cytoplasmic chaperonin, Y2H35, Y2H171 and Y2H41.

Thus, in the prior art only the measurement of specific PS fragments is disclosed, but not the measurement of full-length presenilin fused to a reporter. Furthermore, no substances capable of reducing or eliminating the activity of the presenilinase are disclosed.

The problem underlying the present invention therefore is to provide methods for identifying substances capable of reducing or eliminating the activity of the presenilinase comprising the measurement of full-length presenilin fused to a reporter and to provide said substances.

Summary of the invention

The above-captioned technical problem is solved by the embodiments characterized in the claims and the description. The before-mentioned disadvantages in the art are overcome by the claims and the description of the present invention.

The present invention pertains to methods for identifying a substance capable of reducing or eliminating the activity of the presenilinase wherein a cell or a cell line is cultivated expressing said presenilinase activity and a fusion-protein comprising the full-length presenilin 1 or presenilin 2 and a reporter is measured. The invention is furthermore concerned with substances identifiable with said methods, pharmaceutical compositions comprising said substances and the use of said substances in the manufacture of a medicament for the treatment of neurodegenerative diseases, preferably Alzheimer's disease. Brief description of the figures

Figure 1 a) Sequence and structure of a general hammerhead ribozyme-target RNA complex. The hammerhead ribozyme contains two domains, the substrate binding domain through which it recognizes and binds its target RNA via base pairing (marked by asterisks), and its catalytic domain that possesses the catalytic activity to cleave its target RNA at the 3 'end of the trinucleotide GUX [X=C, A, U] b) Secondary structure of PS2 mRNA. The secondary structure of a part of the PS2 mRNA, starting with nucleotide 1 in the 5' untranslated region and ending at position nt 1236 in the translated region, was predicted by "mfold" (described infra in example 1) Open loops that are good candidate regions suitable for targeting ribozymes are shown as black circles. The cleavage sites at the target trinucleotides are indicated with arrows. The numbering of the nucleotides corresponds to the sequence of human PS2 in the EMBL Data Bank, Accession No. L43964. The prediction for the secondary structure of the remaining part of the PS2 mRNA (nts 1001-2236) did not yield suitable open loop regions (data not shown). c) Location of ribozymes and the corresponding substrate RNAs. Three different trinucleotides were chosen and the appropriate synthetic ribozymes designed for in vitro ribozyme cleavage studies and the exogenous use in cell culture experiments A ribozyme targeted to a trinucleotide was designed with flanking substrate binding domains of various lengths, i.e 5 ribozymes (rzl 173/13 3, rzl 173/12, etc ) targeted to the GUCτ 73 trinucleotide (nucleotide numbering according to EMBL Data Bank, Accession No L43964). As substrate RNAs (target sites) for in vitro cleavage studies we routinely used short synthetic, 5' [32p] -labeled (indicated by asterisks), RNAs (shown as black bars with an arrow directing to the right site indicating sense RNA) In addition, as target site GUC1 173 we generated a larger substrate RNA (367 bp), in vitro transcribed from plasmid pBSK+/PS2 Ncol with T7 polymerase (Fig 4b) and [^^P-CTP]- labeled (indicated by asterisks) The only ribozyme we used endogenously for further functional analyses was ribozyme rzl 173 (indicated with "+") In order to screen positive rzl 173 overexpressing cell clones for PS2 mRNA levels, we used an antisense RNA containing the ribozyme target site of PS2 (shown as black bars with an arrow directing to the left site indicating antisense RNA) as RNA probe for the RNase protection assay This probe was generated by in vitro transcription from plasmid pBSK+/PS2 Ncol with T3 polymerase (Fig 4b) and [32p-CTP]- labeling Figure 2

In vitro cleavage studies of different synthetic ribozymes targeted to various regions in the

PS2 mRNA. a) Three synthetic ribozymes (rzl 173, rz232, rz308, nucleotide numbering according to EMBL Data Bank, Accession No. L43964) were targeted to various regions in the PS2 mRNA and analyzed for their cleavage capacity in vitro with synthetic, 5' [32p]-labeled RNA substrates containing the specific target trinucleotide. Each ribozyme was used with different lengths of the flanking substrate binding region (i.e. rzl 173/13.3, 12, 9, etc.). For rzl 173 and rz232 so-called 'antisense ribozymes', as- 12 and as- 15.1, respectively, were generated (described infra in example 1). The ribozyme cleavage reaction was carried out under standard conditions and ribozyme: target molar ratios were used as indicated. As controls substrate RNAs were used without ribozyme treatment (lane"-"). Reactions were stopped and loaded onto a 20 % SDS- polyacrylamide/ 6 M urea gel, dried onto filter paper and exposed on X-Omat AR films (Kodak). b) Three different ribozymes, rzl 173, rz232 and rz308, with a substrate binding domain in between 15-16 b were used for further detailed analyses concerning the required ribozyme:target molar ratio for efficient cleavage in vitro. The ribozyme cleavage reaction was carried out under standard conditions (described infra in example 1) with ribozyme:target molar ratios as indicated. As controls substrate RNAs were used without ribozyme treatment (lane "-"). Reactions were stopped and loaded onto a 20 % SDS-polyacrylamide/ 6 M urea gel, dried onto filter paper and exposed on X-Omat AR films (Kodak).

Figure 3 a) In vitro efficiency of ribozyme rzl 173 with substrate binding domains varying in length at different ribozyme:target molar ratios. Ribozyme rzl 173 was used with flanking substrate binding domains in between 9-16 b in length (rzl 173/13.3, 12, 9). The corresponding synthetic RNA substrate was 5' [ 2p]-labeled. The ribozyme cleavage reaction was carried out under standard conditions (described infra in example 1) with ribozyme:target molar ratios as indicated. Reactions were stopped and loaded onto a 20 % SDS-polyacrylamide/ 6 M urea gel, dried onto filter paper and exposed on X-Omat AR films (Kodak). Cleavage efficiencies were calculated with the Phospor Imaging System (BioRad). b) Kinetics of the in vitro ribozyme cleavage of synthetic rzl 173/13.3. The ribozyme cleavage reaction was carried out under standard conditions (described infra in example 1) with a concentration of ribozyme target RNA of 100 1 As RNA substrate a 5' [32p]-l_abeled synthetic RNA was used containing the target trinucleotide GUC i \ 73 Aliquots were taken at the indicated time points and loaded onto a 20 % SDS-polyacrylamide/ 6 M urea gel The cleavage kinetic of ribozyme rzl 173/13 3 is shown in the upper figure Ribozyme cleavage in percentage was calculated with the Phosphor Imaging System (BioRad) and is shown in the lower figure

Figure 4 a) Sequence and structure of the ribozyme rzll73/13.3auto - PS2 mRNA complex. The binding of hammerhead ribozyme rzl 173/13 3auto to a specific sequence of the PS2 mRNA is shown (nucleotide numbering according to EMBL Data Bank, Accession No L43964) Base pairing between the flanking regions of the substrate binding domain of rzl 173 and the surrounding nucleotides of GUC1 173 in the PS2 mRNA is indicated by asterisks, the Wobble base pair "G-U" is marked by points The ribozyme rzl 173/13 3auto is an example for a fusion ribozyme comprising the PS2-specific ribozyme rzl 173/13 3 and the autocatalytical hammerhead- ribozyme directly fused with its 5' end to the 3' end of the PS2- specific ribozyme rzl 173/13.3 b) Vector constructs for in vitro and in vivo expression of rzl 173/13.3 and the corresponding substrates. In vitro transcription from the plasmid pBSK+/PS2 rzl 173 13 3 yielded the biosynthetic ribozyme rzl 173 that was then tested for in vitro cleavage activity in comparison to the synthetic ribozyme Plasmid pBSK+/PS2 rzl 173 13 3auto encodes the ribozyme construct box that was finally cloned into the response plasmid (pUHD 10-3) of the tetracycline-sensitive gene expression system (H Bujard, Heidelberg) used for the PS2 'knockdown' in HeLa cells This ribozyme construct box contained the PS2 specific ribozyme rzl 173/13 3 and an autocatalytic ribozyme (see a) Transcription from the T3 promotor of construct pBSK+/PS2 Ncol generated the sense PS2/NcoI fragment containing the ribozyme rzl 173 target sequence of PS2 that was used in in vitro ribozyme cleavage reactions Transcription from the T7 promotor yielded the antisense PS2/NcoI fragment that was used as probe in the RNasel protection assay for quantification of PS2 mRNA levels in rzl 173/13 3 expressing HeLa cell clones

Figure 5

In vitro transcribed ribozyme rzl 173/13.3 cleaved 367 b long PS2 transcript. The in vitro transcribed ribozyme rzl 173 was incubated in increasing amounts (0, 1 , 0,3, 0,5, 1, 3, 5 μl of the total in vitro transcription reaction) together with the 367 b long, [³²P]-labeled PS2 transcript under standard conditions (described infra in example 1 ) The first lane shows the substrate RNA without treatment In lane "-" substrate RNA was incubated under standard conditions without ribozyme Marker RNA of known size was loaded onto the polyacrylamide gel for comparison

Figure 6 a) PS2 mRNA levels of various cell clones inducibly expressing rzll73/13.3. 49 clones that were stably transfected with the construct pUHD 10-3/PS2-rzl 173.13 3auto were tested for PS2 expression after omission of doxycycline with the RNase protection assay (RPA) The first two lanes are control reactions, in which tRNA is used for hybridization with the [32p]-labeled antisense RNA probe of PS2 These hybridization reactions were carried out in the absence (-) or presence (+) of RNases mRNA from the control cell line HtTA was used in the RPA as standard and indicated the endogenous PS2 mRNA level Cell line HtTA/PS2 rzl 173 40 was selected for further detailed analyses on the protein level b) PS2 protein levels in the selected cell line HtTA/PS2 rzl 173.40. Extracts were made from the PS2 'knock-down' cell line at different time points after omission of doxycycline. For comparison with the endogenous PS2 protein level in HeLa cells, extracts were prepared from cells growing in standard medium supplemented with doxycycline at day 0 and 14 Proteins were immunoprecipitated using antibody 3711, separated on SDS/ polyacrylamide gels, blotted onto PVDF membranes and hybridized with the monoclonal antibody BI.HF5C (1 2000 dilution) Both antibodies recognize the hydrophilic loop of PS2

Figure 7

The PS2 'knock-down' HeLa cell line was less sensitive against an apoptotic stimulus as calculated by ethidiumbromide/ acridine orange staining. Three HeLa cell lines (the PS2 'knock-down' cells [PS2 k d ] and cell lines overexpressing wildtype [PS2 wt] or mutant PS2 [PS2 mut]) were used for determination of their apoptotic sensitivity to staurosporine (a) Overexpression of wildtype or mutant PS2 was demonstrated by immunofluorescence with antibody 2972 (1 300 dilution), that recognizes the N-terminus of PS2, in the presence (+Dox) or the absence (-Dox) of doxycycline (b) After treatment with staurosporine in concentrations as indicated for 18 h, the cells were fixed and incubated with ethidiumbromide and acridine orange as described infra in example 1 Figure 8

The PS2 'knock-down' caused an inhibition of apoptosis.

The three transfected HeLa cell lines (the PS2 'knock-down' cells [PS2 k.d.] and two cell lines overexpressing wildtype [PS2 wt] or mutant PS2 [PS2 mut]) as well as the original HeLa cell line were treated with indicated concentrations of staurosporine for 18 h under standard conditions (described infra in example 1). As a control, cells were not treated with staurosporine (lane "0"). (a) Apoptosis sensitivity. Analyses were carried out using a cell death detection ELISA (Boehringer Mannheim; described infra in example 1). The degree of apoptosis was expressed directly as the absorbance at 405-490 nm. (b) Alomar Blue reduction assay. Cell viability was measured using the Alamar Blue reduction (see example 1) and given in percentage of the control, (c) LDH release. LDH release was determined (Boehringer Mannheim; described infra in example 1) and given as optical densities.

Figure 9

The PS2 'knock-down' seemed to have no influence on the caspase 3 (CPP32) activation following an apoptotic stimulus. The three HeLa cell lines (the PS2 'knock-down' cells [PS2 k.d.] and cell lines overexpressing wildtype [PS2 wt] or mutant PS2 [PS2 mut]) were treated with 1 μM staurosporine under standard conditions (described infra in example 1). As a control, extracts were made of cells not treated with staurosporine (lane "c"). (a) Extracts were prepared at the indicated time points and identical protein amounts were loaded onto a 12 % SDS/ polyacrylamide gel. After blotting onto PVDF membranes, hybridization with a CPP32-specific antibody (Transduction Laboratories, 1:500 dilution) was carried out. This antibody recognizes the CPP32 holoenzyme and the 17 kDa active N-terminal fragment (shown in a as an example) that is generated upon proteolytic cleavage, (b) The CPP32 holoenzyme is indicated by an arrow.

Figure 10

The PS2 'knock-down' seemed to have no influence on PARP cleavage following an apoptotic stimulus. The three HeLa cell lines (the PS2 'knock-down' cells [PS2 k.d ] and cell lines overexpressing wildtype [PS2 wt] or mutant PS2 [PS2 mut]) were treated with 1 μM staurosporine under standard conditions (described infra in example 1). As a control, extracts were made of cells not treated with staurosporine (lane "c"). (a) Extracts were prepared at the indicated time points and identical protein amounts were loaded onto a 12 % SDS/ polyacrylamide gel. After blotting onto PVDF membranes hybridization with a PARP-specific antibody (Boehringer Mannheim, 1 :2000 dilution) was carried out. This antibody recognizes the PARP holoenzyme and the proteolytic fragments (indicated by arrows), (b) PARP holoenzyme and the 85 kDa fragment are marked by arrows.

Figure 11

Inhibition of apoptosis by PS2 'knock-down' - time-course experiments. The three HeLa cell lines (the 'PS2 knock down' cells [PS2 k.d.] and cell lines overexpressing wildtype [PS2 wt] or mutant PS2 [PS2 mut]) were treated with 1 μM staurosporine under standard conditions. As a control, extracts were made of cells not treated with staurosporine (lane "c"). a) No CTF16 could be detected during apoptosis in the PS2 'knock-down' cell line. Extracts were prepared at the indicated time points and identical protein amounts were used for immunoprecipitation with the polyclonal PS2/loop-specific antibody 371 1 After 12 % SDS/ polyacrylamide gel electrophoresis and blotting onto PVDF membranes, hybridization with the monoclonal antibody BI.HF5C (1 :2000 dilution), that was also raised against the loop region, was carried out. b) The PS2 'knock-down' showed an inhibitory effect on apoptosis compared with the overexpression of wildtype or mutant PS2. In parallel to extract preparations (a) the cells were analyzed for apoptosis using the cell death detection ELISA (Boehringer Mannheim). The degree of apoptosis was expressed directly as the absorbance at 405-490 nm.

Figure 12

No CTF_jg generation occured at subtoxic staurosporine concentration after 18 h. HeLa cells overexpressing wildtype (PS2 wt) or mutant PS2 (PS2 mut) were treated with staurosporine concentrations as indicated under standard conditions. As a control, cells were not treated with staurosporine (lane „c"). Extracts were prepared after 18 h and identical protein amounts were used for immunoprecipitation with the polyclonal PS2/loop-specific antibody 371 1. After 12 % SDS/ polyacrylamide gel electrophoresis and blotting onto PVDF membranes, hybridization with the monoclonal antibody BI.HF5C (1 :2000 dilution), that was also raised against the loop region, was carried out. Figure 13

Generation of a dominant negative PS2 mutation. a) Full-length PS2 accumulates in independent cell lines stably expressing PS2 Asp366Ala Individual clones were metabolically labeled with ³⁵S-methionine and cell lysates were immunoprecipitated with the PS2 specific antibody 3711. b) Detection of PS2 CTFs Cell lysates from HEK293 cells or HEK293 cells stably expressing the PS2 Asp366Ala mutation (clone 11) were immunoprecipitated with antibody 3711 and PS2 CTFs were detected with the monoclonal antibody BI.HF5C. Robust levels of endogenous PS2 CTFs are detected in untransfected HEK293 cells, whereas generation of the PS2 CTF is almost completely inhibited in cells stably expressing PS2 Asp366Ala c) Stable expression of PS2 Asp366Ala inhibits accumulation of endogenous PS1 fragments Cell lysates from HEK293 cells or HEK293 cells stably expressing the PS2 Asp366Ala mutation (clone 11) were immunoprecipitated with the PS1 specific antibody 3027 and PS1 CTFs were detected with the monoclonal antibody BI 3D7 Robust levels of encogenous PS1 CTFs are detected in untransfected HEK293 cells, whereas generation of the PS1 CTF is almost completely inhibited in cell overexpressing PS2 Asp366Ala

Figure 14

PS2 Asp366Ala affects processing of βAPP a) Expression of PS2 Asp366Ala results in a marked reduction of Aβ production Pooled clones stably expressing PS2 Asp366Ala or a subcloned cell line (clone 11) were metabolically labeled with ³⁵S-methionine for 2 h followed by a cold chase for additional 2 h Conditioned media were immunoprecipitated with antibody 3926 to synthetic Aβ Cells stably exspressing the PS2 Asp366Ala produce significantly reduced amounts of Aβ In addition reduced production of p3, which results from the combined action of α- and γ-secretase (Haass and Selkoe, 1993) is observed The effect of the dominant negative PS2 mutation on amyloidogenesis appears to be enhanced in the subcloned cell line b) Expression of PS2 Asp366Ala results in the accumulation of C-terminal proteolytic fragments of βAPP Cell lysates from pooled clones stably expressing PS2 Asp366Ala or from a subcloned cell line (clone 1 1) were immunoblotted with antibody 5818 to detect the C-terminal fragments of βAPP Figure 15

Upper part: The presenilin-luciferase fusion-protein.

The presenilin-luciferase fusion-protein is schematically depicted. The presenilin proteins probably contain eight transmembrane domains (TM), with both the N- and C-terminus, as well as the large hydrophilic loop between TM6 and TM7 oriented toward the cytoplasm (Haass, 1997). The putative presenilinase cleaves in the loop region The gray box indicates the published cleavage site (Podlisny et al, 1997; Shirotani et al., 1997; Shirotani et al., 1999; Wisniewski et al, 1997). In this figure, the luciferase fused into the N-terminus of presenilin with or without amino acid deletions in the presenilin protein is shown.

(A) Stably transfected cell line expressing the presenilin (PS1 or PS2)-luciferase fusion-protein. The fusion-protein is cleaved due to the activity of the presenilinase into fragments of regulated proteolytic cleavage The NTF is fused to the reporter luciferase.

(B) The method for identifying a substance capable of reducing or eliminating the activity of the presenilinase according to the present invention is shown (details see in the description and in example 2) Said substance is termed "inhibitor" in this figure

Only if said substance according to the present invention is actually capable of reducing or inhibiting the activity of the presenilinase (see fig 15B + inhibitor, right side), the luciferase fused to the N-terminus is bound to the microtiter plates and gives rise to a signal due to the presence of full-length presenilin protein

If no substance capable of reducing or inhibiting the activity of the presenilinase is present (see fig 15B - inhibitor, left side), the presenilinase-luciferase fusion-protein is cleaved by the endogenous presenilinase, the C-terminus is bound to the microtiter plates via antibody-antigen interaction and the luciferase fused to the N-terminus is washed away during the washing steps In this case, no signal is detected

Detailed description of the preferred embodiments of the invention

The invention pertains to a method for identifying a substance capable of reducing or eliminating the activity of the presenilinase A method according to the present invention comprises cultivating a cell or a cell line to express said presenilinase activity and a fusion-protein, said fusion-protein comprising the full-length presenilin 1 or presenilin 2 and a reporter, incubating said cell or cell line with a test substance, measuring the quantity of the full-length fusion-protein employing the reporter and comparing the quantity of full-length fusion-protein obtained by the before-mentioned step to the quantity of full-length fusion-protein measured for a control. Thus, in the absence of a substance capable of reducing or eliminating the activity of the presenilinase, i.e. in case the test substance does not have the desired effect, the presenilin is cleaved into fragments and thus no full-length protein coupled to the reporter can be detected (see figure 15). Two particular examples of the method according to the present invention which should not be construed as limiting the present invention are disclosed in examples 2 and 3, infra. The activity of the presenilinase as used herein may be the activity of any enzyme (e.g. a protease) or of a protein or of a chemical substance capable of cleaving said presenilin. Said activity of the presenilinase is thus capable of generating fragments of regulated or alternative proteolytic cleavage. Preferably, said presenilinase activity leads to cleavage of presenilin into fragments of regulated proteolytic cleavage, e.g. cleavage of PS2 is within the distal region of exon 10 (originally called exon 9) between Met²⁹⁸ and Ala³⁰⁵. Preferably, said activity of the presenilinase may also be the autocatalytical cleavage activity of the presenilin wherein the presenilin itself leads to the generation of fragments without the activity of another enzyme or substance. Preferably, as there is evidence that presenilin may have the activity of the γ-secretase, said activity of the presenilinase may also be the activity of the γ-secretase.

A suitable cell or cell line, preferably an eukaryotic cell or a cell line, to be transformed with nucleic acid constructs to express said fusion-protein may be any cell or cell line known to the expert in the field, in particulary cells or cell lines used in neurological and neurobiology research. Examples of such cells or cell lines useful for producing the transformed cell lines of the invention include mammalian cells or cell lines (e.g., the cell lines H4, U373, NT2, human embryonic kidney (HEK) 293, PC12, COS, CHO, fibroblasts, myelomas, neuroblastomas, hybridomas, oocytes, embryonic stem cells), insect cells lines (e.g., using baculovirus vectors such as pPbac or pMbac (Stratagene, La Jolla, CA)), yeast (e.g., using yeast expression vectors such as pYESHIS (Invitrogen, CA)), and fungi.

To accomplish expression in eukaryotic cells, a wide variety of vectors have been developed and are widely available which allow inducible (e.g., LacSwitch expression vectors, Stratagene, La Jolla, or the tTA-response plasmid pUHD 10-3, Gossen and Bujard, 1992) or cognate (e.g., pcDNA3 vectors, Invitrogen, Chatsworth, CA) expression of presenilin nucleotide sequences under the regulation of an artificial promoter element. Such promoter elements are often derived from CMV or SV40 viral genes, although other strong promoter elements which are active in eukaryotic cells can also be employed to induce transcription of presenilin nucleotide sequences. Typically, these vectors also contain an artificial polyadenylation sequence and 3' UTR (untranslated region) which can also be derived from exogenous viral gene sequences or from other eukaryotic genes. Furthermore, in some constructs, artificial, non-coding, spliceable introns and exons are included in the vector to enhance expression of the nucleotide sequence of interest (in this case, presenilin sequences). These expression systems are commonly available from commercial sources and are typified by vectors such as pCDNA3 and pZeoSV (Invitrogen, San Diego, CA). Both of the latter vectors have been successfully used to cause expression of presenilin proteins in transfected COS, CHO, and PC12 cells (Levesque et el. 1996). Innumerable commercially-available as well as custom-designed expression vectors are available from commercial sources to allow expression of any desired presenilin transcript in more or less any desired cell type, either constitutively or after exposure to a certain exogenous stimulus (e.g., withdrawal of tetracycline or exposure to IPTG).

Vectors may be introduced into the recipient or "host" cells by various methods well known in the art including, but not limited to, calcium phosphate transfection, strontium phosphate transfection, DEAE dextran transfection, electroporation, lipofection (e.g., Dosper Liposomal transfection reagent, Boehringer Mannheim, Germany), microinjection, ballistic insertion on micro-beads, protoplast fusion or, for viral or phage vectors, by infection with the recombinant virus or phage. A fusion-protein according to the present invention is a protein encoded by the nucleic acid encoding the full-length presenilin 1 (PS1) or presenilin 2 (PS2) linked to the nucleic acid encoding the reporter (described infra) and is expressed in a cell line as described supra according to standard methods known to the expert in the art (see also e.g. Sambrook et al., 1989). "Full- length" as used herein relates to the entire presenilin and the respective gene as described infra and comprises also mutations of said gene. Thus, "full-length" also relates to presenilin or the respective gene with N-terminal deletions of e.g. 1-100 amino acids or 1-300 nucleotids, respectively, which occur e.g. during the cloning of the fusion construct as described infi-a. "Full- length" as understood in the present invention does not relate to fragments of regulated or alternative cleavage. Said fusion-protein may also comprise one or several linker or spacer molecules e.g. located between the presenilin (PS) and the reporter and /or at the 5' or 3' end of the construct. Said linker or spacer molecules are peptides, preferably 2-50 amino acids long, or chemical substances capable of linking the presenilin to the reporter or capable of maintaining a certain space between presenilin and reporter to enable the proper function of both molecules and to avoid steric hindrance Preferably, the reporter is fused to the N-terminus of the PS in one of the following ways

(i) direct fusion to the N-terminus of PS with or without amino acid deletions in the presenilin protein or (ii) fusion to the N-terminus via a spacer (2-50 amino acids) between the reporter protein and the presenilin N-terminus with or without amino acid deletions in the presenilin protein or (iii) fusion into the N-terminus of presenilin, as shown in figure 15 (upper section), with or without amino acid deletions in the presenilin protein Said fusion-proteins according to the present invention are preferably encoded by one of the following constructs Thus in a preferred embodiment, the DNA encoding the reporter is fused to the N-terminus of the DNA encoding the full-length PS in one of the following ways' (i) direct fusion to the N-terminus of PS with or without nucleic acid deletions in the presenilin gene or (ii) fusion to the N-terminus via a spacer (6-150 base pairs) between the reporter gene and the presenilin N-terminus with or without nucleic acid deletions in the presenilin gene or (iii) fusion into the N-terminus of presenilin, as shown in figure 15 (upper section) for the expressed protein, with or without nucleic acid deletions in the presenilin gene. The genes encoding the transcripts of PS1 and PS2 which map to human chromosome 14 (PS1) and chromosome 1 (PS2) as well as the gene products PS1 and PS2 are known in the art As used herein, the term "presenilin 1 gene" or "PS 1 gene" means the mammalian gene first disclosed and described in Sherrington et al (1995) The term "presenilin 2 gene" or "PS2 gene" means the mammalian gene first disclosed and described in US 5840540 A, and later described in Rogaev et al (1995) and Levy-Lahad et al (1995), and WO 96/34099 Al (all herein incorporated by reference) including any allelic variant and heterospecific mammalian homologues Additional human splice variants as described in WO 96/34099 Al have been found in which a single codon or a region encoding thirty-three residues may be spliced-out in some transcripts The term "presenilin-2 gene" or "PS2 gene" primarily relates to a coding sequence, but can also include some or all of the flanking regulatory regions and/or introns The term PS2 gene specifically includes artificial or recombinant genes created from cDNA or genomic DNA, including recombinant genes based upon splice variants The presenilin 2 gene has also been referred to as the E5-1 gene (e g US 5840540 A) or the STM2 gene (e g , Levy-Lahad et al , 1995) A control according to the present invention may be an experimental setup employing the method according to the present invention wherein said cell a cell line as described infra is not incubated with said test substance. A control may also be the quantity of reporter bound to presenilin fragments measured in the supernatant. Said control may then be compared to the quantity of full- length fusion-protein of presenilin and reporter bound to a solid support. A non-limiting example of such a control is disclosed in example 2, infra. Said control is used for calibration of the method of the present invention. Further methods to provide controls or standards for the method of the present invention are known to the expert in the field and are embraced by the present invention.

Any of the well-known reporter genes can be operatively linked to the full-length presenilin gene and may be expressed in the cell line according to the present invention.

Examples of suitable reporter genes include, but are not limited to E. coli β-galactosidase (β-gal, Luban and Goff, 1995), xanthine-guanine phosphoribosyl transferase (Chu and Berg, 1985), galactokinase (Schumperli et al, 1982), interleukin-2 (Cullen, 1986), thymidine kinase (Searle et al., 1985), alkaline phosphatase (Toh et al., 1989; Henthorn et al., 1988), secretory alkaline phosphatase (SEAP) or secreted placental alkaline phosphatase (Berger et al., 1988) and chloramphenicol-acetyltransferase (CAT, Alton and Vapnek, 1979, Gorman et al., 1982; Tsang et al., 1988) green fluorescent protein (GFP) produced by the bioluminescent jellyfish (Chalfie et al., 1994), derivatives therof such as blue fluorescent protein (BFP), firefly luciferase (deWet et al., 1987, Engebrecht and Silverman, 1984). Many of these and other useful reporter genes are available from commercial sources.

Expression products of the reporter genes such as reporter enzymes can be measured using standard methods. For example, bioassays can be carried out for biologically active proteins such as interleukin-2. Enzyme assays can be performed when the reporter gene product is a reporter enzyme such as alkaline phosphatase or β-galactosidase. Alternatively, various types of immunoassays such as competitive immunoassays, direct immunoassays and indirect immunoassays may be used. Such immunoassays involve the formation of immune complexes containing the reporter gene product and a measurable "reporter" or a "label". As used herein, the term "reporter" includes moieties that can be detected directly, such as fluorochromes and radiolabels, and moieties such as enzymes that must be reacted or derivatized to be detected. Thus, the term "employing the reporter" or "detection of the reporter" as used herein relates to direct or indirect detection of said expression products of the reporter genes with standard methods known in the art. Examples for said detection of the reporter include, but are not limited to the measurement of the substrate or the substrate reaction product of a reporter enzyme or the detection of light emitted by the reporter gene product such as luminescence, the detection of a coloured substrate or substrate reaction product which is due to the activity of a reporter enzyme and the detection of radioactivity due to the activity of the reporter gene product. In competitive immunoassays, samples from induced cultures (following cell disruption if the reporter gene product is not secreted) are incubated with an antibody against the reporter gene product and a known amount of labeled reporter gene product. Any unlabeled product produced by the cells competes with the labeled material for binding to the antibody. The resulting immune complexes are separated and the amount of labeled complex is determined. The reporter gene product produced by the cells can be quantified by comparing observed measurements to results obtained from standard curves. Direct immunoassays involve incubating culture samples with a labeled antibody against the reporter gene product and separating any immune complexes that form. The amount of label in the complexes is determined and can be quantified by comparison to standard curves. Enzyme-linked immunosorbant assays (ELISAs) can also be carried out by well known methods, e.g., as described in U.S. Pat. No. 4665018.

The particular reporter used will depend upon the type of immunoassay used. Examples of reporters that can be used include, e.g., radiolabels such as ³²P, ¹²⁵I, ³H and ¹⁴C; fluorescent reporters such as fluorescein and its derivatives, rhodamine and its derivatives, dansyl and umbelliferone; chemiluminescers such as the various luciferin compounds; and enzymes such as horseradish peroxidase, alkaline phosphatase, lysozyme and glucose-6-phosphate dehydrogenase. The antibody or reporter gene product, as the case may be, can be tagged with such labels by known methods. For example, coupling agents such as aldehydes, carbodiimides, dimaleimide, imidates, succinimides, bisdiazotized benzadine and the like may be used to tag the antibodies with fluorescent, chemiluminescent or enzyme labels.

The genetic control elements used in this invention can be inserted into many reporter gene- containing vectors, including but not limited to plasmids pSV2Apap, pMAMneo-CAT, pMAMneo-LUC, pSVOCAT, pBCO, pBLCAT2, pBLCAT3, pONl, pCHHO, pCH126 and various plasmids described by De Wet et al, 1987.

In a preferred embodiment, the invention pertains to a method as described wherein the presenilinase is specific for presenilin 1.

In another preferred embodiment, the invention pertains to a method as described wherein the presenilinase is specific for presenilin 2.

In another preferred embodiment the invention pertains to a method as described wherein the substance capable of reducing or eliminating the activity of the presenilinase reduces or eliminates the autoproteolytical cleavage of the presenilin. Thus, in another preferred embodiment, the invention pertains to a method as described wherein the substance capable of reducing or eliminating the activity of the presenilinase reduces or eliminates the activity of γ-secretase.

The preferred substances to be found with said method according to the present invention are described in more detail infra.

In another preferred embodiment, the invention pertains to a method as described wherein the substance prevents cleavage of the presenilin 1 into fragments of regulated endoproteolytic cleavage wherein the N-terminal fragment (NTF) is approximately 21-28 kDa in size and the C- terminal fragment (CTF) is approximately 16-24 kDa in size (Haas et al., 1998; Okochi et al.,

1997; Thinakaran et al., 1996).

In another preferred embodiment, the invention pertains to a method as described wherein the substance prevents cleavage of the presenilin 2 into fragments of regulated endoproteolytic cleavage wherein the N-terminal fragment (NTF) is approximately 28-30 kDa (NTF) kDa in size and the C-terminal fragment (CTF) is approximately 20-25 kDa in size (Haas et al., 1998; Kim et al., J Biol Chem 1997, 272, 11006-11010; Podlisny et al., 1997).

The sizes of said fragments can not be exactly determined, as known to the skilled artisan, and vary slightly (1 up to 4 kDa) according to the method used (e.g. SDS polyacrylamide gels and marker proteins used thereon). Thus, only approximate sizes are indicated.

As understood herein, fragments of regulated proteolytic cleavage are products of presenilin cleavage by the presenilinase (see definition infra), not by a member of the caspase 3 protease family (CPP32). According to the present invention, the NTF fragment of regulated or normal proteolytic cleavage has a smaller molecular weight than the NTF of alternative cleavage (e.g. for

PS2 approximately 30 kDa (regulated) versus 34 kDa (alternative) according to WO 98/47917) and the CTF of regulated or normal cleavage has a higher molecular weight.

Even more particularly, the invention pertains to a method as described wherein the reporter is fused to the N-terminus of the presenilin.

In a further aspect, the present invention is concerned with a method as described supra wherein

"measuring the quantity of the full-length protein employing the reporter" comprises immobilizing antibodies specific for the C-terminal portion, preferably the loop region of the C-terminal portion, of presenilin 1 or presenilin 2 to a solid surface, extracting the protein from said cell or cell line after cultivation, incubating said protein extracts with said antibodies, and measuring the amount of full-length fusion-protein bound by said antibodies by detection of the reporter Even more particularly, the invention pertains to a method as described wherein any unbound fusion-protein fragments are measured by detection of the reporter after incubating said protein extract according to the above-mentioned step

More particularly, the invention pertains to a method as described wherein the cell or cell line is expressing said fusion-protein at such a level that little or no full-length fusion-protein is detected when no substance capable of reducing or eliminating the activity of the presenilinase is present More particularly, the invention pertains to a method as described wherein the reporter is luciferase

Most particularly, the invention pertains to a method as described wherein the method is a high throughput screening assay (HTS) HTS relates to an experimental setup wherein a large number of substances is tested simultaneously Preferably, said HTS setup may be carried out in microplates, may be partially or fully automated and may be linked to electronic devices such as computers for data storage, analysis, and interpretation using bioinformatics. Preferably, said automation may involve robots capable of handling large numbers of microplates and capable of carrying out several thousand tests per day Preferably, a test substance which shows a desired inhibitory function in a cell-free system will also be tested in a cell-based system The term HTS also comprises ultra high throughput screening formats (UHTS) Preferably, said UHTS formats may be carried out using 384 or 1536 well microplates, sub-microliter or sub-nanoliter pipettors, improved plate readers and procedures to deal with evaporation HTS methods are described e g in US 5876946 A or US 5902732 A The expert in the field can adapt the above-described method to a HTS or UHTS format without the need of carrying out an inventive step According to the present invention, said method may be an immunological or a molecular biology or biochemical method Immunological methods are known to the expert in the field and include, but are not limited to ELISAs (enzyme-linked immuno-sorbent assay) or Sandwich-E IS As, dot- blots, immunoblots, radioimmunoassays (Radioimmunoassay RIA), diffusion-based Ouchterlony tests, rocket lmmunofluorescent assays or Western-blots Examples for immunological methods are e g described in An Introduction to Radioimmunoassay and Related Techniques, Elsevier Science Publishers, Amsterdam The Netherlands (1986), Bullock et al, Techniques in Immunocytochemistry, Academic Press, Orlando, FL Vol 1 (1982), Vol 2 (1983), Vol 3 (1985), Tijssen, Practice and Theory of Enzyme Immunoassays: Laboratory Techniques in Biochemistry and Molecular Biology, Elsevier Science Publishers, Amsterdam, The Netherlands (1985) Molecular biology or biochemical methods are known to the expert in the field and include, but are not limited to reporter gene assays such as β-gal-, CAT-, SEAP- GFP-, BFP- or luciferase- assays, polymerase-chain reaction (PCR), RT-PCR, Northern- or Southern-blots which are published e.g. in: Sambrook et al.(1989) Molecular Cloning: A Laboratory Manual, 2^nd ed., Cold

Spring Harbor Laboratory Press, Cold Spring Harbor, New York und Bertram, S. und Gassen,

H.G. Gentechnische Methoden, G. Fischer Verlag, Stuttgart, New York, 1991).

The invention is further concerned with a substance capable of reducing or eliminating the activity of the presenilinase, identifiable with a method as described above.

The term "substance" as used herein means a chemical, pharmaceutical, biochemical or biotechnological substance.

Preferably, the invention is further concerned with a substance identifiable with said methods, wherein the substance is selected from the group consisting of: a) a substance capable of reducing or eliminating the enzymatic activity of the presenilinase or b) a substance capable of reducing or eliminating the expression of the presenilinase at the translational or transcriptional level or c) a substance capable of reducing or eliminating the expression of the substrate of the presenilinase or d) a substance capable of reducing or eliminating the formation of complexes between presenilin fragments.

As used herein, a substance according to a) identifiable with said methods is directly inhibiting the presenilinase by blocking the enzymatic activity of the presenilinase. Such a substance may be another enzyme capable of cleaving the presenilinase e.g. a protease or a so-called anti-enzyme or antizyme. Such a substance further includes a biochemical substance, e.g. a polypeptide capable of blocking the substrate-binding site of the presenilinase such as a substrate analogue. Also included are chemical substances, preferably substances which chemically modify the presenilinase in a way that the presenilinase cannot carry out its enzymatic activity any more, such as derivatizing agents. Other substances known in the art capable of carrying a similar action are also included in the present invention.

A substance according to b) identifiable with said methods may be any inhibitor of the presenilinase translation or transcription. Examples include, but are not limited to transcription terminators or repressors or translation inhibitors such as ribozymes, rifampicin or chloroamphenicol .

A substance according to c) identifiable with said methods may be a substance blocking the translation or transcription of the substrate of the presenilinase. Such substances include, but are not limited to substances inhibiting e.g. PS 1 or PS2 at the RNA or DNA level, e.g. a ribozyme as disclosed in example 1 (RNA level, see also e.g. figure 1, figure 4) or a mutation of PS1 or PS2 as disclosed in example 4 (DNA level, see also figures 13 and 14).

A substance according to d) identifiable with said methods may be a substance capable of reducing or eliminating the formation of complexes between presenilin fragments. As disclosed in Steiner et al. (1998), the proteolytic N- and C-terminal fragments generated by the presenilinase may associate to form stable complexes. Fragments not stably associated are degraded. Therefore, a substance capable of reducing or eliminating the activity of the presenilinase according to the present invention may be a substance capable of reducing or completely blocking the formation of said complexes between presenilin CTF and NTF, thus leading to increased degradation of said fragments and consequently to the reduction of Aβ deposition and amyloid plaques. As described supra, there is no therapeutic way of preventing the neuronal cell death due to normal or alternative PS fragments which then may cause apoptosis. The suppression of the generation of said PS fragments which may be the biologically active form of the presenilins and subsequently preventing the deposition of Aβ may be achieved by reducing or eliminating the activity of the presenilinase. Reducing or eliminating the activity of the presenilinase should go along with a concomitant increase in the full-length protein, a decrease in the biologically active fragments and thus may lead to the reduction of Aβ and/or also may prevent apoptotic cell death. Thus, in another preferred embodiment, the invention is concerned with a substance as described, wherein the substance is capable of reducing the amount of presenilin fragments and preventing neuronal cell death and/or capable of reducing the deposition of Aβ and/or reducing the formation of amyloid plaques.

In another preferred embodiment, the invention is concerned with a substance identifiable with said methods, wherein the presenilinase is specific for presenilin 1.

In another preferred embodiment, the invention is concerned with a substance identifiable with said methods, wherein the presenilinase is specific for presenilin 2.

In another preferred embodiment, the invention is concerned with a substance as described, wherein the substance reduces or eliminates the autoproteolytical cleavage of the presenilin. It was surprisingly found that presenilin itself may have autoproteolytic activity. It was further found that presenilin 1 and 2 itself may be capable of cleaving the C-terminal fragments of APP leading to the deposition of Aβ and therefore may be a cofactor for the γ-secretase or the γ-secretase itself. Thus, a substance capable of reducing or eliminating the autoproteolytic cleavage of the presenilin can prevent neuronal cell death due to increased apoptosis because of the presence of presenilin fragments and also reduce the deposition of Aβ and subsequently reduce the formation of amyloid plaques.

In another preferred embodiment, the invention is concerned with a substance as described, wherein the substance reduces or eliminates the activity of γ-secretase.

In another preferred variant of this embodiment, the invention is concerned with a substance as described, wherein the substance prevents cleavage of presenilin 1 into fragments of regulated cleavage wherein the N-terminal fragment (NTF) is approximately 21-28 kDa in size and the C- terminal fragment (CTF) is approximately 16-24 kDa in size.

In another preferred variant of this embodiment, the invention is concerned with a substance as described, wherein the substance prevents cleavage of presenilin 2 into fragments of regulated endoproteolytic cleavage wherein the N-terminal fragment (NTF) is approximately 28-30 kDa

(NTF) kDa in size and the C-terminal fragment (CTF) is approximately 20-25 kDa in size.

Preferably, the invention comprises a substance identifiable with a method as described, wherein the substance is an antisense-oligonucleotide or a ribozyme.

Antisense oligonucleotides are DNA or RNA molecules that are complementary to at least a portion of a specific mRNA molecule (Weintraub, 1990). In the cell, antisense nucleic acids hybridize to the corresponding mRNA, forming a double-stranded molecule. The antisense nucleic acids interfere with the translation of the mRNA, since the cell will not translate a mRNA that is double-stranded. The use of antisense methods to inhibit the in vitro or in vivo (also in the animal model) translation of genes is well known in the art (e.g. Marcus-Sekura, 1988). An antisense core nucleic acid may at least contain 10 nucleotides complementary to the target message. Said antisense oligonucleotides also comprise peptide nucleic acids, phosphodiester antisense oligonucleotides and phosphorothioate oligonucleotides (Boado RJ et al, 1998).

Antisense nucleic acids have been described in the art to inhibit the expression of proteins associated with toxicity or gene products introduced into the cell, such as those introduced by an infectious agent (e. g. a virus). They furthermore are useful to block expression of a mutant protein or a dominantly active gene product such as amyloid precursor protein in AD as described in WO 981881 1 Al . Similarly, the antisense oligonucleotide of the present invention may be used to block the presenilinase or presenilin expression in neurodegenerative diseases or preferably in

AD or FAD

The term "ribozyme" used in the present invention relates to an RNA capable of specifically interacting with a target RNA and of irreversibly cleaving it at a defined site. Preferably, the ribozyme has a central sequence not complementary to the target RNA that is responsible for its catalytic activity (catalytic domain or region (a)), and two flanking sequences essentially complementary to two neighboring sequences of the target RNA (substrate binding domain or hybridization region (b)) so as to allow binding of the ribozyme via base-pairing and thus selective cleavage of the target RNA

Thus, according to the present invention, said ribozyme may comprise a catalytic region (a) and at least one hybridization region (b), with the hybridization region (b) essentially being complementary to a region of the mRNA that is transcribed from the presenilinase or presenilin gene

The ribozyme according to the invention is preferably characterized in that the hybridization region (b) consists of two domains flanking the catalytic region (a) and being essentially complementary to the target nucleic acid region so as to be capable of selectively binding to all mRNAs that are transcribed by the presenilinase or presenilin gene in order to selectively cleave these RNAs (see also figure 1 for a hammerhead ribozyme).

The term "essentially complementary" as used in the present invention is to be understood such that the complementarity between ribozyme and target nucleic acid region is so high that it allows the specific binding of the ribozyme via hybridization and selective cleavage of the target nucleic acid region under those conditions under which the ribozyme is used The ribozymes are preferably completely complementary to the target nucleic acid region

The term "selective cleavage" as used in the invention is to be understood such that the expression of the target gene, e g. the presenilinase or presenilin gene is suppressed to such an extent that the desired therapeutical effect is achieved

The selective inhibition of the gene expression in cells by the ribozyme according to the invention does therefore not mean that the target gene will be irreversibly damaged or eliminated Rather, the use of the ribozymes advantageously only leads to the selective inhibition of the translation of said gene The property of ribozymes to specifically bind target RNA and to inactivate them by cleavage has been successfully demonstrated several times for the case of specific inhibition of

HIV-RNA (Lisziewicz et al , 1993, Yu et al 1993, Morgan and Anderson, 1993, Yamada et al ,

1994)

Said ribozyme can also be presented by the following general formula

(b) (a) (b)

5' [N3.20] [CUGANGARNc oSGAAA] [N3.20] 3', wherein N is G, C, A or U, R is a purine, and S is a pyrimidine, and wherein the central region N₀. ₃0 of sequence (a) can be replaced by a linker which is different from nucleic acid, e.g., a hydrocarbon chain (Thomson et al, 1993).

The conserved nucleotides within the catalytic region are essential for the catalytic effect but can be optionally modified by the person skilled in the art with the below-mentioned method (Joyce, 1992; Yuan and Altman, 1994) such that ribozyme effectivity and selectivity is favorably influenced. The length of the hybridization region (b) (N_3-2o) depends on many factors and is selected such that a sufficient hybridization to the RNA to be cleaved is achieved under the selected conditions (such as temperature, ion environment) in order to allow efficient cleavage, but, if the difference between the target RNA and non-target RNA does not comprise the cleavage motif per se, there is no sufficient hybridization to the non-target RNA. The choice of the length of the hybridization region thus depends on, e.g., the GC content of the RNAs and the number of nucleotides differing between target RNA and non-target RNA. Preferably, the lengths of the 5' hybridization region and the 3' hybridization region are equal, but they can be asymmetrical, e.g., a combination of three and 20 nucleotides. The overall length of the hybridization region (b) is 12 to 30 nucleotides.

The ribozyme according to the invention can be a hammerhead, hairpin or axehead ribozyme. The structure of hammerhead ribozyme in general is known to the person skilled in the art and is described in, e.g., Symons (1992), and Rossi (1993). As outlined below, the skilled practitioner may modify the catalytic structure such that it yields optimum results for the projected use in terms of effectivity and substrate specificity.

Hairpin ribozymes were originally identified to be part of the minus strand of the TRSV (tobacco ringspot virus) satellite RNA. In the meantime, it has been shown that these ribozymes can effectively cleave target RNAs in trans, the mechanism of action being similar to that of the hammerhead ribozymes. The regions being responsible for substrate binding and catalytic effect were determined and the invariable structure or sequence motifs characterized. The cleavage motif of the target RNA is N'GNPy (N is G, C, U or A, Py is C or U) (see, e.g., Rossi, 1993, and Hampel et al., 1990). On the basis of the requirements with respect to the structure and sequence of the hairpin ribozyme necessary for an effective cleavage and with respect to the cleavage motif on the target RNA explained in the art, the skilled practitioner can construct a ribozyme using standard techniques that possesses the desired properties.

Axehead ribozymes were originally defined to be part of the genomic and antigenomic RNA of the hepatitis delta virus. Here, too, it was possible to determine the minimum sequence and/or structure necessary for a cleavage in trans, and, as described above for hammerhead and hairpin ribozymes, the person skilled in the art can construct axehead ribozymes on the basis of the data described in the art which exhibit the properties required for the purpose according to the invention (see, e.g. Been, 1994, and Wu et al., 1993).

Said ribozyme may also be a fusion-ribozyme comprising a presenilinase-or presenilin-specific ribozyme and an autocatalytical hammerhead-ribozyme fused with its 5' end to the 3' end of the presenilinase-or presenilin-specific ribozyme.

Determined target sequences with the pertaining, highly specific ribozyme were observed to allow a considerably increased catalytic activity by adapting the catalytic region (Koizumi et al., 1989, Koizumi and Ohtsuka, 1992). If the kinetic data show a too low ribozyme efficiency, the person skilled in the art can optimize the ribozyme structure by well-established in vitro evolutionary processes (Joyce, 1992; Yuan and Altman, 1994).

According to the present invention, the ribozyme may be modified such that resistance to nucleases is achieved, increasing the retention time and thus the effectivity of the ribozyme at the target site, e.g., in certain cells of a patient. Furthermore, the amount of ribozyme to be applied and, if any, related side-effects can be reduced.

Examples of such modifications are the substitution of the 2'-OH groups of the ribose by 2'-H, 2'-O-methyl, 2'-O-allyl, 2'-fluoro or 2'-amino groups (Paolella et al., 1992, and Pieken et al., 1991) or the modification of phosphodiester compounds, by, e.g., replacing one or two oxygen atoms by sulphur atoms (phosphorous thioate or phosphorous dithioate; Eckstein, 1985, and Beaton et al., in: Eckstein, F. (ed.) Oligonucleotides and analogues - A practical approach - Oxford, JRL Press (1991), 109-135 or by a methyl group (methyl phosphonate; Miller, loc. cit., 137-154). Further modifications include conjugation of the RNA with poly-L-lysine, polalkyl derivatives, cholesterol or PEG. Preferably, the ribozymes according to the invention contain at least one of the above-described phosphate modifications and/or at least one of the above- described ribose modifications.

Additionally, the invention pertains to a pharmaceutical composition comprising a substance as described and a pharmaceutically acceptable carrier therefor. The term "pharmaceutically acceptable carrier" as used herein refers to conventional pharmaceutic excipients or additives used in the pharmaceutical manufacturing art (see e.g. Remington's Pharmaceutical Sciences (1980)). Said pharmaceutical composition of the present invention may contain a vector comprising the substance of the present invention to be used for gene therapy and may contain a colloidal dispersion system or liposomes for targeted delivery of the pharmaceutical composition. Suitable vectors comprise plasmids, viruses (including phage) and integratable DNA fragments (i e , integratable into the host genome by recombination) All forms of vectors which serve an equivalent function and which are, or become, known in the art are suitable for use herein Suitable vectors will contain replicon and control sequences which are derived from species compatible with the intended expression host

One example of a targeted delivery system for antisense polynucleotides is said colloidal dispersion system Colloidal dispersion systems include macromolecule complexes, nanocapsules, microspheres, beads, and lipid-based systems including oil-in-water emulsions, micelles, mixed micelles, and liposomes or liposome formulations The preferred colloidal system of this invention is a liposome Liposomes are artificial membrane vesicles which are useful as delivery verhicles in vitro and in vivo These formulations may have net cationic, anionic or neutral charge characteristics are useful characteristics with in vitro, in vivo and ex vivo delivery methods. It has been shown that large unilamellar vesicles (LUV), which range in size from 0.2-4.0 μm can encapsulate a substantial percentage of an aqueous buffer containing large macromolecules. RNA, DNA and intact virions can be encapsulated within the aqueous interior and be delivered to cells in a biologically active form (Fraley, et al , 1981) In addition to mammalian cells, liposomes have been used for delivery of polynucleotides in plant, yeast and bacterial cells In order for a liposome to be an efficient gene transfer vehicle, the following characteristics should be present (1) encapsulation of the genes of interest at high efficiency while not compromising their biological activity, (2) preferential and substantial binding to a target cell in comparison to non- target cells, (3) delivery of the aqueous contents of the vesicle to the target cell cytoplasm at high efficiency; and (4) accurate and effective expression of genetic information (Mannino et al , 1988).

The composition of the liposome is usually a combination of phosphohpids, particularly high- phase-transition-temperature phosphohpids, usually in combination with steroids, especially cholesterol Other phosphohpids or other lipids may also be used The physical characteristics of liposomes depend on pH, ionic strength, and the presence of divalent cations The pharmaceutical composition of the present invention may contain said vector as a naked "gene expression vector" This means that the construct is not associated with a delivery vehicle (e g liposomes, colloidal particles and the like) One of the principal advantages of naked DNA vectors is the lack of a immune response stimulated by the vector itself

The invention also relates to the use of a substance as described in the manufacture of a medicament for the treatment of neurodegenerative diseases Preferably, the invention relates to the use of a substance as described in the manufacture of a medicament for the treatment of Alzheimer's disease.

Most preferable, the invention relates to the use of a substance as described in the manufacture of a medicament for the treatment of familiar Alzheimer's disease.

Neurodegenerative diseases include, but are not limited to, Alzheimer's disease (AD), Parkinson's disease, Huntigton's chorea and stroke.

The term "Alzheimer's disease (AD)" refers to a neurodegenerative disorder marked by the gradual formation of extracellular neuritic plaques in the brain, particularly in the hippocampus and the adjoining cortex. The majority of Alzheimer's disease cases are late in onset lacking an obvious genetic linkage and are characterized as sporadic. The term "familiar Alzheimer's disease

(FAD)" refers to a subgroup of AD comprising a small percentage (approximately 10%) of cases which are earlier in onset and segregate strongly within families suggesting a genetic etiology.

The following examples serve to further illustrate the present invention; but the same should not be construed as limiting the to the scope of the invention disclosed herein. Example 1 exemplifies a method of reducing or eliminating the presenilinase activity by reducing or eliminating the presenilinase substrate at the RNA level with ribozymes. Example 2 illustrates a method for identifying a substance or substances capable of reducing or eliminating the activity of the presenilinase. Example 3 illustrates a method for identifying a substance or substances capable of reducing or eliminating the activity of the presenilinase wherein said substance is capable of reducing or eliminating the formation of stable complexes between N-terminal and C-terminal PS- fragments. Example 4 demonstrates methods of reducing or eliminating the generation of presenilin fragments and Aβ by mutagenizing the substrate of the presenilinase.

Example 1 - Method of reducing or eliminating the presenilinase activity by reducing or eliminating the presenilinase substrate at the RNA level with ribozymes

This example illustrates the reduction or elimination of presenilinase activity at the RNA level with substances, in particular ribozymes which cleave the PS2-specific RNA Thus, said ribozymes reduce or eliminate the presenilinase substrate, the full-length PS2 and therefore reduce or eliminate the presenilinase activity and the occurrence of presenilin fragments which are linked to the pathology of neurodegenerative diseases, preferably AD

General Methods cDNA constructs

Human wildtype (wt) PS2 cDNA (Science 269 973-977, 1995) The N141V mutation in the human PS2 cDNA was generated by site-directed mutagenesis (Stratagene) Both full-length PS2 cDNA sequences were cloned into the EcoRI restriction site of the tT A-response plasmid pUHD 10-3 (Gossen and Bujard, 1992) to generate the tetracycline-regulated expression vectors pUHD10-3/PS2wt and pUHD10-3/PS2mut for inducible expression in cells A 297 bp PS2 Ncol cDNA fragment (nts 960-1257 according to the EMBL Data Bank, Accession No L43964) was cloned into the pBluescriptII/SK+ plasmid (pBSK+/PS2 Ncol) and used for in vitro transcription

Ohgoribonucleotide sequences

Ribozyme sequences rz 1173/13.3 5'-UUCUUUGGCUGAUGAGGCCGUGAGGCCGAAACACAGCG-3', rz 1173/12 5'-CUUUGGCUGAUGAGGCCGUGAGGCCGAAACACAG-3', rz 1173/9 5'-UUGGCUGAUGAGGCCGUGAGGCCGAAACAC-3', rz 1173/11.12 5'-CUUUGGCUGAUGAGGCCGUGAGGCCGAAACACAA-3', as 1173/12 5'-CUUUGGCUGAUUCGGCCGUGAGGCCGAUACACA-3', rz 232/15.1 5'-UGGUUUUUCUGAUGAGGCCGUUAGGCCGAAACACGUCG-3', rz 232/12 5'-GUUUUUCUGAUGAGGCCGUUAGGCCGAAACACGU-3', rz 232/10 5'-UUUUCUGAUGAGGCCGUUAGGCCGAAACACGU-3', as 232/15.1 5'-UGGUUUUUCUGAUUCGGCCGUUAGGCCGAUACACGUC-3', rz 308/15 5'-GAAUCCCGCUGAUGAGGCCGUUAGGCCGAAACCUUGG-3', rz 308/12 5'-GAUCCCGCUGAUGAGGCCGUUAGGCCGAAACCUUG-3' Autoribozv e sequence: 5'-GAUCCGUCGACGGACUCGAGUCCGUCCUGAUGAGUC

CGUGAGGACGAAACGGAUC-3 '

Fusion ribozymes comprising a PS2 specific ribozyme and the autoribozyme: rz 1173/13.3auto 5'-UUCUUUGGCUGAUGAGGCCGUGAGGCCGAAACACAGCGG

AUCCGUCGACGGACUCGAGUCCGUCCUGAUGAGUCCGUGAGGACGAAACGGAUC-

3'. rz 1173/12auto 5'-CUUUGGCUGAUGAGGCCGUGAGGCCGAAACACAGGAUCCG

UCGACGGACUCGAGUCCGUCCUGAUGAGUCCGUGAGGACGAAACGGAUC-3' rz 1173/9auto 5'-UUGGCUGAUGAGGCCGUGAGGCCGAAACACGAUCCGUCGA

CGGACUCGAGUCCGUCCUGAUGAGUCCGUGAGGACGAAACGGAUC-3'. rz 1173/11.125'-CUUUGGCUGAUGAGGCCGUGAGGCCGAAACACAAGAUCC

GUCGACGGACUCGAGUCCGUCCUGAUGAGUCCGUGAGGACGAAACGGAUC-3' rz 232/15.1 5'-UGGUUUUUCUGAUGAGGCCGUUAGGCCGAAACACGUCGGAUCC

GUCGACGGACUCGAGUCCGUCCUGAUGAGUCCGUGAGGACGAAACGGAUC-3' rz 232/125'-GUUUUUCUGAUGAGGCCGUUAGGCCGAAACACGUGAUCC

GUCGACGGACUCGAGUCCGUCCUGAUGAGUCCGUGAGGACGAAACGGAUC-3* rz 232/105'-UUUUCUGAUGAGGCCGUUAGGCCGAAACACGUGAUCC

GUCGACGGACUCGAGUCCGUCCUGAUGAGUCCGUGAGGACGAAACGGAUC-3' rz 308/155'-GAAUCCCGCUGAUGAGGCCGUUAGGCCGAAACCUUGGGAUCC

GUCGACGGACUCGAGUCCGUCCUGAUGAGUCCGUGAGGACGAAACGGAUC-3' rz 308/125'-GAUCCCGCUGAUGAGGCCGUUAGGCCGAAACCUUGGAUCCG

UCGACGGACUCGAGUCCGUCCUGAUGAGUCCGUGAGGACGAAACGGAUC-3'

RNA substrate sequences: 1173 5'-CGCUGUGUCCCAAAGAA-3', 232 5'-CGACGUGUUAAAAACCA-3', 308 5'-CCAAGGUCCGGGAUUC-3'

Ribozyme numbering corresponds to the nucleotide position in the PS2 mRNA of the guanidine in the target GUX (indicated in the RNA substrate sequences in bold), after which the phosphodiester bond is cleaved The numbering of the nucleotides correponds to the PS2 sequence in the EMBL Data Bank, Accession No L43964 The RNA substrates, which represent partial sequences of the PS2 mRNA are named accordingly The number of base pairs formed by hybridization of the substrate binding domain of the ribozyme to the target mRNA is indicated in numbers, wobble base pairs in numbers behind the point (i e rzl 173/13 3) Synthetic and in vitro transcribed ribozymes and RNA substrates were strictly handled under RNase free conditions DEPC (diethylpyrocarbonate) water or nuclease free water (Promega, Heidelberg) was used Oligoribonucleotide purification was done either by HPLC (reversed phase, trityl on) or on denaturing SDS-PAGE/ 8 M urea

Synthesis of synthetic ribozymes and RNA substrates Oligoribonucleotides were synthesized using standard phosphoramidite chemistry (Boehringer Ingelheim Pharma KG, Department of Chemical Reseach, Biberach, Interactiva Inc , Ulm) The ribozymes contained stabilizing 2'- methylnucleosides and 3 '-terminal modifications, namely 3 ',3 '-inverted termini and 2', 3'- dideoxynucleosides, making them suitable for the cell transfections 2'-0-methyl-ribonucleotides were introduced in order to prevent endonucleolytic degradation, while 3 '-terminal dideoxynucleotides (ddA and ddC) or 3 ',3 '-inverted dG residues protected the sequence from exonucleolytic degradation These modifications have been reported to increase stability several thousand fold over native ribozymes, while the catalytic activity is only minimally impaired The ribozymes were synthesized with flanking substrate binding regions of varying length (between 6- 8 nts) which hybridize to the PS2 mRNA by base pairing The catalytic domain of the designed hammerhead ribozymes contained the minimal set of conserved ribonucleotides Cloning of ribozyme DNA into plasmids and in vitro transcription. The DNA coding for the ribozyme rzl 173/13 3 was cloned into pBluescriptII/SK+ (Stratagene) The resulting plasmid pBSK+/PS2-rzl l73 13 3 (see Fig 4b) was transcribed in vitro using T7 polymerase according to manufacturer's instructions (Clontech) and the purity of the ribozyme RNA was controlled by OD260/280 measurement and gel electrophoresis (20 % SDS-PAGE/ 8 M urea) For the tetracycline-regulated expression in HeLa cells, the DNA encoding a self-splicing ribozyme was attached directly at the 3' end of rzl 173/13 3 cDNA to generate pBSK+/PS2-rz 1173 13 3 auto (see Figs 4a, b) The rzl 173/13 3auto DNA sequence was cloned into the tTA-responsive plasmid pUHD 10-3 to generate plasmid pUHD 10-3 /PS2-rz 1173 13 3auto (Fig 4b)

The prediction of secondary structure of the PS2 mRNA

The most probable secondary structure of the PS2 mRNA was determined by the method of Zuker et al (1989) by using the SQUIGGLES software included in the Wisconsin Sequence Analysis Package (Genetic Computer Group Inc ) [32 P] -labeling of substrate and ribozyme RNA

As RNA substrates for the in vitro cleavage reaction we used either short 16-17 base (b), 5'[32p]- labeled synthetic oligoribonucleotides or a 367 b long RNA substrate that was radioactively labeled upon in vitro transcription. The phosphorylation reaction was carried out in a total of 20 μl containing 20 pmol synthetic substrate RNA, 3 μl (lOμCi/μl) [γ-³²P]-ATP, 2 μl lOx phosphorylation buffer, 13 μl H2O and 10 U polynucleotide kinase (Boehringer Mannheim) by incubation at 37°C for 1 hour. For in vitro transcription, the plasmid pBSK+/PS2.NcoI (Fig. 4b) was linearized with Xhol, phenol chloroform-extracted, and ethanol-precipitated. In vitro transcription was carried out in 20 μl of a mixture containing lμl 10 mM GTP, lμl 10 mM ATP, lμl 10 mM UTP, 2 μl 10 x transcription buffer, 1 μl 0.2 M DTT, 1 μl RNase inhibitor (20 U), 5 μl -³²P-CTP (10 mCi/ml), 1 μl 0.1 mM CTP, 5 μl H₂O, 1 μl (10 U) T7 RNA Polymerase (in vitro transcription kit, Clontech). The reactions were incubated for 45 min at room temperature (RT). To degrade the template DNA, 1 μl RNase-free DNase I was added and incubated at 37°C for 30 min. After phenol/chloroform extraction, both, the 5'[³2p]-labeled synthetic as well as the in vitro transcribed, [³²P]-labeled RNA substrates were purified by 20% SDS-PAGE/ 6M urea, eluted from the gel, precipitated and resuspended in DEPC water. An aliquot was counted in a scintillation counter (Amersham) to determine the specific radioactivity.

In vitro ribozyme cleavage assay

[³2p]_iabeled substrate RNA (20.000cpm/reaction) and ribozyme RNA were incubated in 50 mM Tris-HCl (pH 7.5) and 10 mM MgCl₂ for 5 min at 95°C followed by a 60 min incubation at 37°C. Reactions were stopped by addition of formamide gel-loading buffer (80 % formamide, 10 mM EDTA, pH 8.0, 0,002 % bromphenol blue and xylene cyanol). Substrates and cleavage product(s) were separated by electrophoresis on a 20 % SDS-polyacrylamide/ 6 M urea denaturing gel and detected by autoradiography (X-OMAT AR films, Kodak).

Cell culture, cell lines and DNA transfection

Cell culture. Control cells and all transfected HeLa cell lines were cultured in Dulbecco's modified

Eagle's medium (DMEM) supplemented with 10 % heat-inactivated FCS, 100 units/ml penicillin,

(100 μg) streptomycin and 1 mM L-glutamine at 37°C in humidified air atmosphere with 5 %

CO₂.

Cell lines. The HtTa cell line, stably stably transfected with the pUHD 15-1/neo DNA plasmid encoding the tetracycline-sensitive transactivator (tTA) of the "Tet-off' expression system (Gossen, M and Bujard, H 1992) The HtTa cell line was transfected with the plasmids pUHD10-3/PS2 wt (wildtype PS2), pUHD10-3/PS2 mut (mutant (N141V) PS2) or pUHDlO- 3/PS2-rzl l73 13.3auto (ribozyme rzl 173/13 3) to give rise to the double stable cell lines HtTA/PS2-wt 13, HtTA/PS2-mut 5, and HtTA/PS2-rzl 173 40, respectively DNA transfection Stable DNA transfections were performed by the calcium-phosphate precipitation method with 50 μg of purified DNA (Qiagen, Hilden) and 5 μg of pCEP4 plasmid DNA for hygromycin selection Hygromycin-resistant clones were obtained after 2 weeks of selection in medium containing 200 μg/ml hygromycin Individual cell clones were isolated, expanded, and tested for either wildtype or mutant PS2 overexpression by Western blotting and immunocytochemistry using PS2-specific antibodies PS2 'knock-down' cell clones were identified by quantitation of the PS2 mRNA level (RNase protection assay, Boehringer Mannheim) and the PS2 protein level (immunoprecipitation/ Western blotting)

Antibodies, immunoprecipitation and Western blotting

Antibodies. To recognize PS2, the following three antibodies were used polyclonal antibodies 3711 and 2972 raised against PS2/loop- and PS2/N-terminus-GST fusion-proteins, respectively, monoclonal antibody BI.HF5C raised against the same PS2 loop - GST fusion-protein as 3711 To recognize the C-terminus of PS1, the polyclonal antibody 3875 raised against a PSl/loop - GST fusion-protein, and the monoclonal antibody BI 3D7 raised against the same fusion-protein were used The polyclonal antibodies were a kind gift of C Haass (Mannheim) For the analysis of poly(ADP-ribose) polymerase (PARP) cleavage and caspase 3 activation, we used a polyclonal antibody against recombinant full-length PARP (Boehringer Mannheim) and a monoclonal antibody against the N-terminus of CPP32 (Transduction Laboratories), respectively Immunoprecipitation. Cells that were grown in 6 cm²-dishes to confluence were lysed with a mixture of 50 mM Tris-HCl, pH 7 6, 150 mM NaCl, 2 mM EDTA, 0 2 % NP-40, 1 mM PMSF and 5 μg/ml leupeptin (buffer A) Immunoprecipitations were done with 3 μl of antibody 371 1 and 20 μl pre-washed protein A-Sepharose (Pharmacia) for 2 h at 4°C Immunoprecipitates were sequentially washed in buffer A, high salt buffer (buffer A with 500 mM NaCl) and in buffer A containing 0 1 % SDS The precipitates were solubilized with 2x SDS sample buffer and electrophoresed on 12 % SDS-PAGE/ 6 M urea The proteins were blotted onto PVDF membranes for 1 h at 400 mA, and membranes were treated with antibodies according to the Western blotting protocol Western blotting. Cells were grown in 6 cm^-dishes to confluence. Cell lysis were carried out in a buffer containing 150 mM NaCl, 50 mM Tris-HCl, pH 7.6, 2 mM EDTA, 0,2 % (v/v) NP40, 1 mM PMSF, and 5 μg/ml Leupeptine. Triton X-100 and Nonidet P-40 were added to a final concentration of 1 %. 30 μg of protein extract was loaded onto a 10-12 % SDS-PAGE and electrophoresed. After blotting of proteins onto PVDF membranes for 1 h at 400 mA, filters were blocked with 5% low-fat milk powder in 10 mM Tris-HCl, 170 mM NaCl, pH 8.0 / 0.1 % Tween (TBST) at 4°C overnight. After washing the filters with TBST, the membranes were probed with the primary antibodies in 5 % milk powder/ TBST at RT. Following a washing step with TBST, filters were incubated for 1 h at RT with the peroxidase-conjungated secondary antibodies (Amersham) in 5 % milk powder/ TBST. Chemiluminescence was detected using the ECL detection system (Amersham) and exposition to X-ray films (BioMax MR, Kodak).

RNA isolation and quantification by RNase protection assay

RNA isolation. mRNA isolation was carried out as described by the manufacturer's instructions (Boehringer Mannheim). Cells were grown in 75 cm^-culture flasks and washed twice with ice cold phosphate-buffered saline (PBS) (1,7 M KH₂PO₄, 5 mM Na₂HPO₄, 0,15 M NaCl, pH 7,4). Cells were trypsinized, pelleted by centrifugation, and lysed in 3 ml lysis buffer (0.1 M Tris-HCl, pH7.5, 0.3 M LiCl, 10 mM EDTA, 1 % lithium dodecylsulfate, 5 mM DTT). The DNA was mechanically sheared by passing the extracts six times through a 21 gauge needle. 1.5 ml of a biotin-labeled oligo (dT)₂rj probe was added and mixed with pre- washed 150 μl streptavidine magnetic particles. After separating and washing of the generated biotin-streptavidine complex, the poly (A+)-selected mRMA was eluated with 25 μl H₂O and its concentration measured. RNAse protection assay (RPA). For RPA the plasmid pBSK+/PS2.NcoI was linearized with Xhol and in vitro transcription was carried out as described before with the T3 polymerase. RPAs were performed according to manufacturer's instructions (Boehringer Mannheim). Radioactively labeled antisense RNA probes (5x10^ cpm) were coprecipitated with 1 μg isolated mRNA in 30 μl hybridization buffer (40 mM Pipes (1,4-Piperazindiethane-sulfoneacid), 400 mM NaCl, 1 mM EDTA, 80 % formamid, pH 6.4) and incubated at 45°C overnight. The same amount of mRNA was incubated with 1 μl yeast tRNA (5 μg/μl) as control reaction. After digestion with RNase A (5 μg/μl) and 2,5 μl RNase Tl (lOU/μL) for 30 min at 37°C, the protected RNA hybrid- fragments were extracted by phenol/chloroform/isoamyl-alcohol (25:24: 1). After ethanol precipitation the fragments were resolved on 5 % SDS-polyacrylamide/8 M urea gels and exposed overnight at -80°C to X-Omat AR films (Kodak). Induction and analysis of apoptosis

Induction of apoptosis. Apoptosis was induced in HeLa cells that were 80 % confluent Staurosporine was added at different concentrations for various periods After incubation, cells were tested for viability and apoptotic parameters

Cell viability After apoptosis induction, cell viability was determined using Alamar Blue reduction assay In addition, cell membrane integrity was determine by measuring the LDH- release (Boehringer Mannheim)

Qualitative analysis of morphological changes of apoptotic cells Moφhological changes of apoptotic cells were determined by labeling the cells with different dyes Cells were plated onto glass slides which were covered with poly-L-lysine (100 μg/ml, Sigma) and laminin (2 μl ml, Sigma) After incubation with staurosporine, cells were rinsed with PBS and stained with 20 μl of a mixture of 100 μg/ml acridine orange (Sigma) and 100 μg/ml ethidium bromide (Sigma) and viewed by fluorescence microscopy Apoptotic cells were scored based on characteristic changes of chromatin condensation and nuclear fragmentation Alternatively, staining with Hoechst 33258 (0 5 μg/ml) was carried out, and apoptotic cell nuclei were detected by fluorescence microscopy Quantitative analysis by ELISA. DNA fragmentation was measured by quantification of cytosolic oligonucleosome-bound DNA using a cell death detection ELISA kit (Boehringer Mannheim) following the manufacturer's instructions

Ribozyme strategy for the 'knock-down' ofPS2 mRNA

Requirement for cleavage activity of hammerhead ribozymes is the presence of a GUX

(X=C,A,U) triplet in the target RNA (Haseloff and Gerlach, 1988, Ruffner et al , 1990) The hammerhead ribozymes consist of two domains, the substrate binding domain (I e the hybridization region (b)) and the catalytic domain or region (a) (Fig la) (Haseloff and Gerlach,

1988) They represent an advanced class of antisense oligonucleotides since they combine the substrate specificity of complementary nucleic acids with the potential not only to hybridize to but also to degrade susceptible substrate RNAs catalytically Due to their catalytic activity less ribozyme RNA molecules need to be present in the cell than antisense RNA molecules for an efficient 'knock-down' of the target mRNA

The PS2 mRNA was searched for potential GUX consensus sites (Fig lb) Several factors were taken into consideration when designing the most suitable PS2-cleaving ribozymes (i) the accessibility of the mRNA target sites for the ribozyme, (ii) the strength of the ribozyme-target

RNA binding, and (in) the stability of the ribozyme (i) To select the most accessible target sites of the PS2 mRNA, the most probable mRNA secondary structures were calculated using the "mfold" software (described supra under general methods; Zuker et al., 1989) (Fig. lb). By that, three GUX triplets (GUU 3 , GUC308, and GUC1 173, numbering of nucleotides according to EMBL Data Bank, Accession No. L43964) were identified in open loop regions of the PS2 mRNA that should be accessible to ribozymes in vivo (Figs, lb, lc). One of these ribozymes was targeted to the coding region of PS2, whereas two ribozymes were directed to the 5' untranslated region of the mRNA (Fig. lc). For these three target sites, synthetic PS2-specific ribozymes (rz 232, 308, 1173) were designed for in vitro cleavage reactions, (ii) It is known that the length of the substrate binding domain of a hammerhead ribozyme effects both its specificity and its turnover number. However, strong interactions between a ribozyme and the substrate RNA can prevent rapid dissociation of the ribozyme following cleavage of the target that could significantly reduce the catalytic activity of the ribozyme. The optimum length of the substrate binding domain has been reported to be in the order of 12-16 nucleotides. In order to carefully select the most efficient ribozyme for cell culture experiments we designed ribozymes targeted to the same site in the PS2 mRNA, but with flanking regions of various lengths (Fig. lc). As a control for the specificity of the ribozyme reaction, we used a so-called 'antisense ribozyme' (i.e. rz232/as-15.1) that comprises the exact sequence of the respective ribozyme, but carries mutations of conserved bases in the catalytic domain, thus disabling this ribozyme to cleave its target RNA. Any effect on the PS2 protein level observed with the 'antisense-ribozyme' is related to an antisense effect rather than a ribozyme effect. As a further control, we used a ribozyme with a randomized substrate binding sequence to evaluat non-specific effects, (iii) To increase ribozyme stability in cell culture experiments, the synthetic ribozymes have been chemically modified (described supra under general methods).

Selection of the most effective synthetic ribozyme in vitro

To study the cleavage activity of the ribozymes and to compare their efficiencies we started to optimize the in vitro ribozyme cleavage reaction (see general methods). The magnesium dependence of the ribozyme reaction was determined (final concentration of 20 mM MgCl₂ data not shown) and was in concordance with previous observations that hammerhead ribozymes have an absolute requirement for divalent metal ions, preferentially Mg²⁺ or Mn²+ _ln order to fulfil their cleavage activity (Uhlenbeck, 1987).

In vitro cleavage reactions for the selection of the most efficient ribozyme were carried out using different partial sequences of the PS2 mRNA as substrates (Fig. lc), that were short 16-17 b in size, synthetic oligoribonucleotides. Each RNA substrate containing the GUC (rz308, rzl 173) or the GUU (rz232) trinucleotide was targeted by appropriate ribozymes varying in the length of the flanking substrate binding domain (Fig. la, c, 2a; length in bases indicated in numbers, i.e 13.3, 12, 9). In standard in vitro cleavage reactions, we could detect the expected 5', end-labeled cleavage products of the RNA substrates (GUC1 173, GUC₂3₂, GUU308) (Fig- 2a). This was true for all active ribozymes (rz232, rz308, rzl 173). The synthetic ribozymes with the longer substrate binding domains ranging up to 15-16 b exhibited a higher in vitro cleavage activity than those with shorter flanking arms (Fig. 2a). When the [³2p]-labeled RNA substrates were incubated with the corresponding 'antisense ribozymes', no cleavage product was detected (Fig. 2a). In addition, incubation of the RNA substrates without ribozyme did not give rise to a cleavage product (Fig. 2a, lanes "-"). We further analyzed the ribozyme:target ratios required for efficient in vitro RNA cleavage for the three selected ribozymes rzl 173/13.3, rz232/15.1 and rz308/15. As shown in figure 2b, both, rzl 173/13.3 and rz308/15, cleaved the target PS2 RNA substrate at a molar ratio of ribozyme to target of 1 :1 under standard in vitro cleavage conditions, though with a lower efficiency than at a greater molar excess of the ribozyme. In contrast, ribozyme rz232/15.1 cleavage products were only detectable, when the ribozyme was in a 50 fold molar excess over the target RNA (Fig. 2b).

Since ribozyme rzl 173 very effectively cleaved the PS2 mRNA in the coding region, we selected this ribozyme for the PS2 'knock-down' in cultured cells and investigated the optimal length of its substrate binding domain in more detail (described supra under general methods). Whereas rzl 173/13.3 (substrate binding domain: 13 bases and 3 bases forming wobble base pairs with the target mRNA) produced significant amounts of the expected cleavage product within 1 h at a molar ribozyme: target ratio of 1: 1, much higher molar ratios were required when the binding domain was shortened to 9 bases (Fig. 3a). Time course studies with ribozyme rzl 173/13.3 revealed that ribozyme-mediated target RNA cleavage could be detected already after 5 min of incubation (Fig. 3b). After 6 h the RNA substrate was almost completely degraded (Fig. 3b).

In vitro transcribed ribozyme rzl 173/13.3 also cleaves longer substrate RNAs To ensure that ribozyme rzl 173 also cleaves longer substrate RNA molecules that might already adopt a secondary structure, we in vitro transcribed plasmid pBSK+/PS2.NcoI into a 367 b RNA and studied the cleavage of this substrate RNA by the in vitro transcribed ribozyme rzl 173/13.3auto (Fig. 5). The autocatalytic ribozyme should be able to splice itself out of the initial transcript to generate ribozyme rzl 173 with a defined 3' end. This approach was reported to keep the ribozyme in the nucleus, because of an inhibition of the transport into the cytoplasm, where it can directly act on newly transcribed substrate RNA (Liu and Carmichael, 1994). Calculations with the "mfold" software (described supra under general methods) predicted the same secondary structure for the 367 b in vitro transcribed substrate RNA as for the same sequence stretch in the context of the full-length PS2 mRNA. Incubation of the substrate RNA (PS2.NcoI mRNA fragment) with increasing amounts of biosynthetic ribozyme rz 1173/13.3 resulted in a cleavage of the target RNA into the predicted fragments of 259 and 108 bases (Fig. 5). The cleavage of the longer RNA substrate was not as efficient as the cleavage of the shorter synthetic oligoribonucleotides (compare to Figs. 2 and 3).

'Knock-down' of endogenous PS2 in a ribozyme expressing HeLa cell line For the PS2 'knock-down' in cells we applied the inducible "Tet-off' expression system. The efficacy of ribozyme rzl 173/13.3 in cells was first tested in transient transfection experiments. HtTA cells transiently transfected with the plasmid pUHD10-3/PS2-rzl l73.13.3auto (Fig. 4b) resulted in an about 30% reduction of PS2 mRNA (data not shown). We then produced over 49 clonal cell lines and analyzed their PS2 mRNA level in RNase protection assays (RPA) using the [³²P]-labeled antisense PS2.NcoI mRNA fragment as probe. The PS2 mRNA levels of a number of cell clones are shown in figure 6a The PS2 RNA contents of these clones were analyzed after complete induction of ribozyme expression by omission of doxycycline for 3 days. The time course of the induction of ribozyme expression after the system was shut off with either 1 μg or 2 ng doxycycline/ml was experimentally determined in detail (data not shown). We selected clone 40 for further analyses and demonstrated the 'knock-down' of PS2 also on the protein level (Fig 6b). Expression of ribozyme rzl 173/13.3 resulted in an almost complete extinction of the PS2 protein 2 days after omission of doxycycline. Although cells were continously cultivated under doxycycline-free conditions, the PS2 protein, suφrisingly, returned to basal level within 2 weeks (Fig 6b) Therefore, further functional analyses in PS2 k.d cells were performed 2 days after doxycycline omission

Apoptosis sensitivity is decreased in PS 2 'knock-down' cells and increased in cells overexpressing wildtype or mutant PS2

It has been reported in the literature that wildtype and mutant PS2 have a proapoptotic potential, when overexpressed in various cell lines (Deng et al , 1996, Vito et al., 1996; Wolozin et al , 1996, Janicki et al , 1997) Furthermore, cells expressing mutant forms of PS2 show an increased sensitivity to apoptotic stimuli compared to wildtype PS2 expressing cells (Wolozin et al., 1996; Janicki et al., 1997). To address the question whether PS2 is actively involved in apoptosis, we studied the sensitivity of PS2 k.d. HeLa cells to staurosporine compared HeLa cells inducibly overexpressing wildtype and mutant PS2. The level of induced PS2 expression was determined by immunocytochemistry (Fig. 7a) and biochemical analyses (data not shown). Various methods were used to assess apoptosis, including fluorescence microscopy, cell death detection ELISA, caspase 3 activation, and proteolytic cleavage of PARP and PS2. For determination of cell viability, Alamar Blue reduction and LDH release was measured.

Apoptosis was induced by increasing concentrations of staurosporine. Cells were incubated with an ethidiumbromide/ acridine orange mixture that staines living cells green. Apoptotic cells showed a characteristic chromatin condensation, nuclear fragmentation and the generation of apoptotic bodies, and their chromatin was stained orange upon ethidiumbromide intercalation (Fig. 7b). In the absence of an apoptotic stimulus, no apoptotic cells could be detected in the PS2 'knock-down' cells (PS2 k.d.) and in the wildtype (PS2 wt) or mutant PS2 (PS2 mut) overexpressing cells (Fig. 7b, lane K). At very low concentrations of staurosporine (1 pM), apoptotic cells appeared in the PS2 wt and PS2 mut cultures, with a higher frequency of apoptotic cells in the PS2 mut cell line (Fig. 7b). No apoptotic cells were observed at the same staurosporine concentration in the PS2 k.d. cell line (Fig. 7b). Higher concentrations of staurosporine yielded a comparable number of apoptotic cells in the PS2 wt and PS2 mut cultures (Fig. 7b). In the ribozyme-mediated PS2 k.d. cells only the highest concentration of staurosporine (1 nM) resulted in the occurence of apoptotis. Thus, the sensitivity of HeLa cells to the apoptotic stimulus, staurosporine, is dependent on the PS2 expression level. The N141V PS2 mutation caused an earlier onset, rather than an increase in the extent of cell death. Interestingly, the PS2 'knock-down' resulted in a significant reduction of apoptosis sensitivity. Similar data were obtained when cells were stained with Hoechst 33258 to visualize apoptotic cells (data not shown).

Since visualization of apoptotic cells with ethidiumbromide/ acridine orange staining is not a quantitative method, we applied a cell death detection ELISA that determines mono- and oligonucleosomes in the cytoplasmic fraction of cell lysates (Fig. 8a). In addition, cell viability was assessed using an Alamar Blue reduction assays (Fig. 8b). To distinguish between apoptosis and necrosis, the release of LDH was measured (Fig. 8c). As shown in figure 8a, PS2 wt and PS2 mut cells exhibited a more pronounced response to subtoxic concentrations of staurosporine (1- 100 pM) than HeLa cells expressing endogenous PS2 levels. At higher concentrations of staurosporine (>lnM), secondary necrosis set in, and the difference between these cell lines became blurred. A lower degree of apoptosis (Fig. 8a) together with the increased LDH release (Fig. 8c) clearly indicated that at staurosporine concentrations >10 nM, necrosis instead of apoptosis was the prevailing mode of cell injury. The ELISA results reflected the marked resistance of PS2 k.d. cells to apoptosis stimulation by 1 pM - 1 nM staurosporine, compared to cells expressing normal levels of PS2 (Fig. 8a). In this concentration range, staurosporine had no significant effect on cell viability (Fig. 8b).

The PS2 expression level does not affect the kinetics of caspase 3 activation and PARP cleavage To establish whether PS2 'knock-down' or overexpression changes the kinetics of processes characteristic of the execution phase of apoptosis, we studied the time-course of caspase 3 activation and poly(ADP)ribose polymerase (PARP) cleavage following the induction of apoptosis with 1 μM staurosporine. It is known that the final step in the cascade of protease activation during apoptosis is the activation of caspase 3 that in turn leads to the cleavage of specific proteins that either are actively involved in the apoptosis or just 'innocent bystanders' (Martin and Green, 1995; Alnemri et al., 1996; Chinnaiyan and Dixit, 1996). Kim et al. (Science 1997, 277: 373-376) and Loetscher et al. (1997) reported that presenilin 1 and 2 are both cleaved during apoptosis by a protease that belongs to the caspase 3 protease family. The caspase 3, or CPP32, is activated by cleavage into two proteolytic fragments (17 and 10 kDa in size). The antibody used for immunoprecipitation of caspase 3 recognizes the uncleaved CPP32-holoenzyme and the 17 kDa fragment, but not the 10 kDa C-terminal fragment (example shown in Fig. 9a). No difference in caspase 3 activation following induction of apoptosis could be detected between PS2 wt, PS2 mut and PS2 k.d. cells (Fig. 9b). PARP constitutes one downstream target of activated caspase 3 and is cleaved into two proteolytic fragments, 85 and 27 kDa in size (Kim et al., Science 1997, 277: 373-376). Therefore, PARP is quite often used as marker for apoptosis (example shown in Fig. 10a). We then analyzed the kinetics of PARP cleavage in the three HeLa cell lines. Again, there was no significant difference in the time-course of the appearance of the proteolytic PARP fragments (Fig. 10).

Normal proteolytic, not alternative PS2 fragments seemed to be directly involved in apoptosis The 'knock-down' of endogenous PS2 resulted in a marked inhibition of apoptosis 18 h after the induction with subtoxic staurosporine concentrations (Figs. 7 and 8). On the other hand, no difference could be observed between the ribozyme-mediated PS2 k.d. cell line and control cells regarding the kinetics of caspase 3 activation and PARP cleavage using 1 μM staurosporine in time-course experiments (Figs. 9 and 10). The analysis of CTF generation in the course of cell death revealed that the PS2 k.d. reduced both, the normal proteolytic PS2 fragments and also the alternative PS2 fragments (Fig. 11a). The generation of CTF_] occured at earlier time points in the PS2 mut than in the PS2 wt cultures, pointing to an earlier onset of apoptosis caused by the N141V mutation in PS2 (compare to Fig. 7b). The presence or absence of the PS2 alternative fragments can account for the difference in the sensitivity to an apoptotic stimulus (Fig. 1 lb), arguing for a direct involvement of the alternative fragments in the execution of apoptosis. Another implication of our findings would be that the 'knock-down' of endogenous normal PS2 CTF₂ renders cells less vulnerable to apoptotic stimuli suggesting normal CTF₂₂ as active mediator of cell death. In order to address the question whether normal or alternative PS2 fragments are actively involved in programmed cell death, we analyzed fragment formation at subtoxic staurosporine concentrations (Fig. 12). Suφrisingly, at low concentrations of the apoptotic stimulus, at which cells clearly underwent apoptosis without loss of cell viability and at which PS2 'knock-down' exerted a strong inhibitory effect on apoptosis (see Fig. 8), no alternative CTF 15 generation could be observed. This finding speaks against an active role of the alternative PS2 fragments in the apoptotic cascade and rather suggests that the normal endoproteolytic cleavage products are active mediators of programmed cell death.

Example 2 - Method for identifying substances capable of reducing or eliminating the activity of the presenilinase

Example 2 is schematically depicted in figure 15.

The cell line used for this test is selected in a way that it is expressing the fusion-protein comprising PS1 and luciferase or PS2 and luciferase at a level that only little or no full-length protein is detected (i.e. slightly overexpressing said fusion-protein) which would otherwise increase the background of the test. This can be done by either selecting an appropriate cell clone or, alternatively, by applying the inducible expression systems, e.g. tTA-response plasmid pUHD 10-3 (Gossen and Bujard, 1992), with which it is possible to vary the expression level of the exogenous protein, as the presenilin-luciferase fusion-protein, by using a specific doxycyclin (tetracyclin) concentration in the culture medium. In this cell line the C-terminal PS-fragment and a fragment consisting of the PS-N-terminus and luciferase are detectable. Normally, these fragments are bound to each other or interact with each other in a stable complex, but they can be separated from each other by using either 0.5% SDS or 1% Triton in the extraction buffer (Capell et al., 1998). Since 0.5 % SDS would interfere with antibody binding to microtiter plates, 1% Triton is used for extraction. After extraction of these cells and incubation of these protein extracts to a microtiter plate which is pre-coated with an antibody specific for the PS C-terminus, the C-terminal fragment binds via antigen-antibody interactions to the microtiter plate. The fragment consisting of PS N-terminus and luciferase is removed by extensive wash steps. The addition of the luciferase substrate (Boehringer Mannheim GmbH, Mannheim, Germany) should lead, when measured, only to background signals. These background signals depend on the degree of expression of the fusion-protein and thereby the existence of the full-length fusion- protein. This full-length fusion-protein can also bind via its C-terminus to the coated plate and can, due to the presence of luciferase, give a signal after addition of the luciferase substrate. After incubation of this cell line with test substances a substance capable of reducing or eliminating the activity of the presenilinase reduces or inhibits the cleavage of the presenilin- luciferase fusion-protein and thus causes an increase of the full-length fusion-protein which binds to the coated microtiter plate via antigen-antibody interactions and causes a signal or a signal increase, respectively.

Material and methods

The DNA encoding the reporter luciferase is fused to the N-terminus of the DNA encoding the full-length PS in one of the following ways:

(i) direct fusion to the N-terminus of PS with or without nucleic acid deletions in the presenilin gene or (ii) fusion to the N-terminus via a spacer (6-150 base pairs) between the luciferase gene and the presenilin N-terminus PS with or without nucleic acid deletions in the presenilin gene or (iii) fusion into the N-terminus of presenilin, as shown in figure 15 (upper section) for the expressed protein, with or without nucleic acid deletions in the presenilin gene. Said fusion nucleic acid constructs are prepared according to standard procedures for cloning and sub-cloning (e.g. PCR, ligation, plasmid amplification in bacteria and purification) known in the art (see e.g. Sambrook et al., 1989, Molecular Cloning: A Laboratory Manual, 2^nd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York and Bertram, S. and Gassen, H.G. Gentechnische Methoden, G. Fischer Verlag, Stuttgart, New York, 1991). Cells of the cell line H4 (obtainable under ATCC HTB-148) are stably transfected with with one of said transcribed constructs with lipofectAMTNE (Life Technologies, Inc ) following the manufacturer's indications.

The resulting cell lines expressing the fusion-proteins consisting of presenilin 1 (PSl) or presenilin

2 (PS2), respectively, and of said reporter luciferase, are named H4/Luc-PS-l and H4/Luc-PS-2)

Said cell lines are plated onto 96-well microtiter plates

The stably transfected cells are grown to confluence and incubated with the test substances for 8-

16 hours The cells are washed with phosphate-buffered saline (PBS) and then protein extracts from the cells are made according to an optimized protocol provided by Boehringer Mannheim

GmbH, Mannheim, Germany We additionally added 1 % Triton to the extraction buffer, since it could be demonstrated that the stable complex formed between the two PS fragments is disrupted by the addition of this detergent (Capell et al., 1998) The non-disrupted PS-fragment complex would otherwise lead to the same signal as non-cleaved full-length PS.

Microtiter plates are pre-coated with the monoclonal antibodies specific for the C-terminus of

PSl (loop region) termed BI.3D7 or PS2 (loop region) termed BI.HF5C, respectively (Steiner et al, 1999; the following antibodies may be used instead: SI 82 (C-20), cat # scl244 and STM2 (C-

20), cat # scl456; both Santa Cruz Biotechnology, Inc , Santa Cruz, California/ USA).

Said protein extracts are added to the microtiter plates and incubated according to standard procedures (Immunochemistry 1, A Practical Approach, edited by A.P.Johnstone and M W

Turner, Practical Approach Series, IRL Press, at Oxford University Press, Oxford, New York,

Tokyo). The plates are washed intensively according to standard procedures, the wash is preserved procedures (Immunochemistry 1, A Practical Approach, edited by A.P.Johnstone and

M.W Turner, loc. cit.)

The bound material is tested for luciferase activity according to the protocol provided by

Boehringer Mannheim GmbH, Mannheim, Germany

Only if the test substance (i e a substance according to the present invention) is actually capable of reducing or inhibiting the activity of the presenilinase (see fig 15B + inhibitor, right side), the luciferase fused to the N-terminus is bound to the microtiter plates and gives rise to a signal due to the presence of full-length presenilin protein

If no substance capable of reducing or inhibiting the activity of the presenilinase is present (see fig 15B - inhibitor, left side), the presenilinase-luciferase fusion-protein is cleaved by the endogenous presenilinase, the C-terminus is bound to the microtiter plates via antibody-antigen interaction and the luciferase fused to the N-terminus is washed away during the washing steps. In this case, no signal is detected

Additionally, the wash (the unbound material) is tested for luciferase activity as a control according to the protocol provided by Boehringer Mannheim GmbH, Mannheim, Germany

Example 3 - Method for identifying substances capable of reducing or eliminating the formation of stable complexes between N-terminal and C-terminal PS-fragments

This example illustrates a method for identifying a substance or substances capable of reducing or eliminating the activity of the presenilinase wherein said substance is capable of reducing or eliminating the formation of stable complexes between N-terminal and C-terminal PS-fragments. The assay is essentially carried out as disclosed in example 2, however, the following modifications are made'

The stably transfected cells as disclosed in example 2 are grown to confluence. The cells are washed with phosphate-buffered saline (PBS) and then protein extracts are made from said cells Since the interactions between N-terminal and C-terminal fragments are SDS- and Triton-labile, CHAPS extracts are made according to a published protocol (Capell et al , 1998). This extraction method does not disturb complex generation

Microtiter plates are pre-coated with the monoclonal antibodies specific for the C-terminus of PSl (loop region) termed BI 3D7 or PS2 (loop region) termed BI.HF5C, respectively (Steiner et al., 1999; the following antibodies may be used instead S182 (C-20), cat # scl244 and STM2 (C- 20), cat # scl456, both Santa Cruz Biotechnology, Inc , Santa Cruz, California/ USA) Said protein extracts are added to the pre-coated microtiter plates and incubated according to standard procedures (Immunochemistry 1, A Practical Approach, A P Johnstone and M W Turner, loc. cit.) After incubation, the plates are washed following standard procedures (Immunochemistry 1, A Practical Approach, A P Johnstone and M W Turner, loc. cit.) Since no Triton or SDS is present, N- and C-terminus interact in a stable complex, which cannot be disturbed by washing and thus does not give rise to a luciferase signal because of the presence of luciferase fused to the N-terminus The luciferase activity is detected according to the protocol provided by Boehringer Mannheim GmbH, Mannheim. Germany

The plates are then incubated with test substances for 30 minutes to several hours After incubation the plates are washed intensively according to standard procedures (Immunochemistry 1, A Practical Approach, edited by A.P.Johnstone and M.W. Turner, Practical Approach Series,

IRL Press, at Oxford University Press, Oxford, New York, Tokyo).

In the case that a substance disrupts the PS-fragment interaction, the luciferase fused to the N- terminus is washed away leading to a reduction or inhibition of the luciferase activity and thus a smaller signal is measured.

The wash is preserved and also tested for luciferase activity as a control according to the protocol provided by Boehringer Mannheim GmbH, Mannheim, Germany.

Example 4 - Methods of reducing or eliminating the generation of presenilin fragments and thereby Aβ

This example exemplifies methods for the reduction or elimination of presenilinase activity at the DNA-level by mutagenizing the PS gene and thus modifying the substrate of the presenilinase. Thus, the mutated PS gene leads to the reduction of PS fragments and also Aβ which are linked to the pathology of neurodegenerative diseases, preferably AD.

Furthermore, a method for identifying a substance capable of reducing or eliminating the presenilinase activity is disclosed. Said method is similar to the method disclosed in example 2. Alzheimer's disease (AD) is characterized by the accumulation of senile plaques composed of Amyloid β-peptide (Aβ) (Price and Sisodia, 1998). The Presenilin-1 (PSl) protein plays a central role in Aβ generation, since autosomal dominant mutations increase the production of the pathologically relevant 42 amino acid Aβ42 (Price and Sisodia, 1998) and a deletion of the PS l gene in mice reduces physiological Aβ generation (De Strooper, B. et al., 1998). To investigate the amyloidgenic function of the homologous PS2 we created a dominant negative mutation by mutagenizing an evolutionary conserved Asp residue within the putative trans-membrane domain (TM) 7. The mutant PS2 is not endoproteolytically processed and inhibits accumulation of endogenous presenilin fragments. Cells expressing this dominant negative mutation exhibit significant deficits in processing of βAPP. Strikingly, Aβ secretion was severely diminished, therefore demonstrating that PS2 promotes the physiological generation of Aβ. Human PSl is critically required for Aβ generation, since neurons lacking the PSl gene secrete significantly less Aβ (De Strooper, B. et al, 1998). Moreover, neurons lacking PSl accumulate C-terminal fragments of βAPP, which are the immediate precursors for amyloidogenesis (Haass and Selkoe, 1993). It is therefore believed that PSl affects γ-secretase function either directly or by influencing cellular transport of βAPP and/or the proteases involved in βAPP processing (De Strooper, B. et al, 1998). Although Alzheimer's associated mutations in the PSl homologous PS2 gene increase the pathological production of Aβ42 (Price and Sisodia, 1998), the role of PS2 in physiological Aβ generation is still unknown. A deletion of the PS2 gene can not be used to investigate the normal function of PS2 in amyloidogenesis, since endogenous PSl expression will abolish effects on βAPP processing. Indeed, a gene knock out of PS2 causes no dramatic phenotype (Boeve et al, 1998), whereas a knock out of PSl leads to embryonic lethality (Shen et al, 1997; Wong et al, 1997). In order to investigate a functional role of PS2 in Aβ generation, we now generated a dominant negative mutation, which not only inhibits the function of overexpressed PS2, but also blocks accumulation of endogenous presenilins. It was demonstrated recently that the same Asp residue is important for the amyloidogenic function of human PSl (Wolfe et al, 1999). The critical role of such Asp residues appears to be conserved during evolution, since the corresponding Asp residue is required for the amyloidogenic function of zebrafish PSl (Leimer, U., Lun, K., Romig, H., Walter, J., Grunberg, J., Brand, M. and Haass, C. Expression, endoproteolytic processing, and amyloidogenic function of zebrafish (Danio rerio) presenilin- 1, submitted.). The corresponding Asp residue is also conserved in the human PS2 gene.

We therefore mutagenized Asp366 to Ala and stably transfected the cDNA construct in human kidney 293 (HEK293) cells overexpressing Swedish mutant β-Amyloid precursor protein (βAPP), a cell line, which is frequently used to analyze the influence of presenilins on Aβ generation (Steiner et al, 1999). Upon metabolic labeling with ³⁵-S-methionine, expression of the mutant holoprotein was observed in individual cell lines (Fig. 13a). To identify the proteolytically generated C-terminal fragment (CTF) of PS2 (Price and Sisodia, 1998), cell lysates from untransfected HEK293 cells or from a cell line stably overexpressing PS2 Asp366 Ala (clone 1 1) were immunoprecipitated with an antibody to the large cytoplasmic loop of PS2. PS2 CTFs were detected by immunoblotting using a PS2 specific monoclonal antibody. This revealed robust levels of the CTF in cells expressing endogenous PS2. However, in the cell line overexpressing the PS2 Asp366Ala cDNA endoproteolysis was completely abolished (Fig. 13b). Similar data were obtained in independent cell lines overexpressing PS2 Asp366Ala (data not shown). The lack of PS2 CTFs also demonstrates that endogenous PS2 is completely displaced by the overexpressed mutant variant. Displacement of endogenous PS fragments upon ectopic expression is consistent with previous findings demonstrating coordinate PS expression (Thinakaran et al, 1997). We next investigated if overexpressing of PS2 Asp366Ala also inhibits the accumulation of PSl CTFs. Lysates of cells expressing endogenous presenilins or cell lines overexpressing the PS2 Asp366Ala mutation were immunoprecipitated with an antibody to the cytoplasmic loop of PSl and PSl CTFs were detected by immunoblotting using a monoclonal PSl specific antibody. This revealed that overexpressing almost completely inhibited the formation of endogenous PSl fragments (Fig. 13c).

Therefore the mutation inhibits endoproteolytic processing of PS2, and simultaneously reduces accumulation of both, endogenous PSl and PS2 fragments.

We now analyzed the effect of the mutant PS2 on βAPP metabolism. Cell lines expressing wt PS2 or the corresponding PS2 mutation were metabolically labeled with ³⁵S-methionine and Aβ was immunoprecipitated. In independent cell lines expressing PS2 Asp366Ala a significantly diminished Aβ generation was observed (Fig. 14a). The decrease of Aβ generation in cell lines expressing mutant PS2 was accompanied by a dramatic increase of proteolytic fragments of βAPP, which correspond to the β-secretase and α-secretase cleaved intermediates (Haass and Selkoe, 1993; Fig. 14b).

These data demonstrate that the dominant negative PS2 mutation severely affects βAPP processing. Similar to a gene deletion ofPSl, the dominant negative mutation of PS2 significantly reduces Aβ generation and causes the accumulation of C-terminal proteolytic derivatives of βAPP therefore suggesting that not only PSl but also PS2 is involved in physiological Aβ generation. The critical Asp residue might be required for the structural integrity of PS2. Mutant PS2 could interfere with the transport of βAPP and/or its processing enzymes to the cellular compartments where proteolysis occurs. This is consistent with recent findings demonstrating that a PSl knock out affects transport of selected membrane bound proteins including βAPP (Naruse et al. 1998) and simultaneously inhibits Aβ generation (De Strooper, B. et al, 1998; Naruse et al. 1998). An alternative explanation of these results would be that presenilins are directly required to stimulate γ-secretase activity or exhibit such an activity by themselves (Wolfe et al, 1999).

Methods

Cell culture and cell lines. HEK293 cells stably expressing PS2 Asp366Ala were generated by transfection of the previously described cell line expressing the Swedish mutation (Steiner et al, 1999). Mutagenesis. The cDNA encoding PS2 Asp366Ala was constructed according to a previously described protocol (Steiner et al, 1999). The mutant cDNA was cloned into the pcDNA3.1/Zeio(-) expression vector (Invitrogen) and sequenced to verify successful mutagenesis.

Antibodies. The polyclonal and monoclonal antibodies against amino acids 263-407 of PSl (3027, BI.3D7) and amino acids 297-356 of PS2 (3711, BI.HF5C) were described previously (Steiner et al, 1999). Antibody 3926 to synthetic Aβ (Leimer et al, submitted), or antibody 6E10 (product 300-10, Senentek Pic.) specific for amino acids 1-16 of Aβ, or antibodies specific for Aβ40 or Aβ42 (catalogue nos. 44-348 and 44-344, respecively, QCB, Quality Control Biochemicals, Inc., Hopkinton, USa) and antibody 5818 to the cytoplasmic domain of human βAPP (Leimer et al, submitted) or antibodies specific for the C-terminus of Aβ (SAD 3138, Labgen) were used. Analysis of PS and βAPP metabolites. After starvation for lh in methionine- and serumfree MEM HEK293 cells were metabolically labeled with 700 μCi ³⁵ S-methionine (Promix, Amersham) in methionine- and serum free MEM for 2h. Cell extracts were prepared and subjected to immunoprecipitation of PS as described (Steiner et al, 1999). For the analysis of Aβ in conditioned media, cells were metabolically labeled with 450 μCi ³⁵S-methionine (Promix, Amersham) for 2h, and chased for 2h in medium containing excess amounts of unlabeled methionine. Conditioned media were immunoprecipitated with antibody 3926 and separated on 10-20 % Tris-Tricine gels (Novex). To analyze βAPP-CTFs, cell lysates were separated on 10-20 % Tris-Tricine gels (Novex) and analyzed by immunoblotting using antibody 5818 (Leimer et al, submitted).

Combined immunoprecipitation/westernblotting of PS. Cell extracts were prepared and subjected to immunoprecipitation using the polyclonal antibodies 3027 (to PSl) or 3711 (to PS2) (Steiner et al, 1999). Immunoprecipitated PS proteins were identified by immunoblotting using the monoclonal antibodies BI.3D7 and BI.HF5C (Steiner et al, 1999). Bound antibodies were detected by enhanced chemiluminescence (ECL, Amersham). References

Alnemri, ES, Livingston, DJ, Nicholson, DW, Salvesen, G, Thornberry, NA, Wong, WW, and Yuan, JY (1996). Human ICE/CED-3 protease nomenclature. Cell 87, 171.

Alton NK, Vapnek D (1979). Nucleotide sequence analysis of the chloramphenicol resistance transposon Tn9. Nature 282, 864-869.

Beaton et al. (1991), in: Eckstein, F. (ed.) Oligonucleotides and analogues - A practical approach - Oxford, JRL Press, 109- 135.

Berger J, Hauber J, Hauber R, Geiger R, Cullen BR (1988). Secreted placental alkaline phosphatase: a powerful new quantitative indicator of gene expression in eukaryotic cells. Gene 66, 1-10.

Been MD (1994). Cis- and trans-acting ribozymes from a human pathogen, hepatitis delta virus. Trends Biochem Sci 19, 251-256.

Boeve, C, Craessaerts K, Vanderstichele H, Serneels L, Checler F, von Figura K, Van Leuven F, Annaert W, Saftig P, De Strooper B (1998). Processing of APP in presenilin 1 and presenilin 2 deficient neurons. Neurobiol Aging 19, 283.

Boado RJ, Tsukamoto H, Pardridge WM (1998). Drug delivery of antisense molecules to the brain for treatment of Alzheimer's disease and cerebral AIDS. J Pharm Sci 87: .1308-1315.

Borchelt DR, Thinakaran G, Eckman CB, Lee MK, Davenport F, Ratovitsky T, Prada CM, Kim G, Seekins S, Yager D, Slunt HH, Wang R, Seeger M, Levey Al, Gandy SE, Copeland NG, Jenkins NA, Price DL, Younkin SG (1996). Familial Alzheimer's disease-linked presenilin 1 variants elevate ABetal-42/1-40 ratio in vitro and in vivo. Neuron 17, 1005-1013.

Borchelt DR, Ratovitski T, Van Lare J, Lee MK, Gonzales V, Jenkins NA, Copeland NG, Price DL, Sisodia SS (1997). Accelerated amyloid deposition in the brains of transgenic mice coexpressing mutant presenilin 1 and amyloid precursor proteins. Neuron 19, 939-945. Capell A, Grunberg J, Pesold B, Diehlmann A, Citron M, Nixon R, Beyreuther K, Selkoe DJ, Haass C (1998). The proteolytic fragments of the Alzheimer's disease-associated presenilin- 1 form heterodimers and occur as a 100-150-kDa molecular mass complex. J Biol Chem 273, 3205-3211

Chalfie M, Tu Y, Euskirchen G, Ward WW, Prasher DC (1994). Green fluorescent protein as a marker for gene expression. Science 263, 802-805.

Chartier-Harlin, MC, Crawford, F, Houlden, H, Warren, A, Hughes, D, Fidani, L, Goate, A, Rossor, M, Roques, P, Hardy, J, and Mullan, M (1991). Early-onset Alzheimer's disease caused by mutations at codon 717 of the beta amyloid precursor protein gene. Nature 353, 844-846.

Chinnaiyan, AM, and Dixit, VM (1996). The cell-death machine. Curr Biol 6, 555-562.

Chu G and Berg P (1985). Rapid assay for detection of Escherichia coli xanthine-guanine phosphoribosyltransferase activity in transduced cells. Nucleic Acids Res 13, 2921-2930.

Cullen BR (1986). Trans-activation of human immunodeficiency virus occurs via a bimodal mechanism. Cell 46, 973-982.

Deng GM, Pike CJ, Cotman CW (1996). Alzheimer-associated presenilin-2 confers increased sensitivity to apoptosis in PC12 cells. FEBS Lett 397, 50-54.

De Strooper B, Saftig P, Craessaerts K, Vanderstichele H, Guhde G, Annaert W, Von Figura K, Van Leuven F (1998). Deficiency of presenilin-1 inhibits the normal cleavage of amyloid precursor protein. Nature 391, 387-390.

Duff K, Eckman C, Zehr C, Yu X, Prada CM, Perez-Tur J, Hutton M, Buee L, Harigaya Y, Yager D, Morgan D, Gordon MN, Holcomb L, Refolo L, Zenk B, Hardy J, Younkin S (1996). Increased amyloid-Beta42(43) in brains of mice expressing mutant presenilin 1. Nature 383, 710- 713.

Eckstein F (1985). Nucleoside phosphorothioates. Annu Rev Biochem 54, 3.67-402. Engebrecht J, Silverman M. (1984). Identification of genes and gene products necessary for bacterial bioluminescence. Proc Natl Acad Sci USA 81, 4154-4158.

Fraley R and Papahadjopoulos D (1981). New generation liposomes - The engineering of an efficient vehicle for intracellular delivery of nucleic acids. Trends Biochem Sci 6, 77-80.

Goate, A, Chartier-Harlin, MC, Mullan, M, Brown, J, Crawford, F, Fidani, L, Giuffra, L, Haynes, A, Irving, N, James, L, et al. (1991). Segregation of a missense mutation in the amyloid precursor protein gene with familial Alzheimer's disease. Nature 349, 704-706.

Gorman CM, Moffat LF, Howard BH (1982). Recombinant genomes which express chloramphenicol acetyltransferase in mammalian cells. Mol Cell Biol 2, 1044-1051.

Gossen M, and Bujard H (1992). Tight control of gene expression in mammalian cells by tetracycline-responsive promotors. Proc Natl Acad Sci USA 89, 5547-5551.

Haass C and Selkoe DJ (1993). Cellular processing of beta-amyloid precursor protein and the genesis of amyloid beta-peptide. Cell 75, 1039-1042.

Haass C. (1997). Presenilins: genes for life and death. Neuron, 18, 687-690.

Haass C, Grunberg J, Capell A, Wild-Bode C, Leimer U, Walter J, Yamazaki T, Ihara I, Zweckbrormer I, Jakubek C, Baumeister R (1998). Proteolytic processing of Alzheimer's disease associated proteins. J Neural Transm Suppl 53, 159-67.

Hampel A, Tritz R, Hicks M, Cruz P (1990). Ηaiφin' catalytic RNA model: evidence for helices and sequence requirement for substrate RNA. Nucleic Acids Res 18, 299-304.

Haseloff, J, and Gerlach, WL (1988). Simple RNA enzymes with new and highly specific endoribonuclease activities. Nature 334, 585-591.

Hendriks, L, van Duijn, CM, Cras, P, Cruts, M, Van Hul, W, van Harskamp, F, Warren, A, Mclnnis, MG, Antonarakis, SE, Martin, JJ, et al. (1992). Presenile dementia and cerebral haemorrhage linked to a mutation at codon 692 of the beta-amyloid precursor protein gene. Nature Genet 1, 218-221.

Henthorn P, Zervos P, Raducha M, Harris H, Kadesch T (1988). Expression of a human placental alkaline phosphatase gene in transfected cells: use as a reporter for studies of gene expression. Proc Natl Acad Sci USA 85, 6342-6346.

Janicki S, Monteiro MJ (1997). Increased apoptosis arising from increased expression of the Alzheimer's disease-associated presenilin-2 mutation (N141I). J Cell Biol 139, 485-495.

Jarrett, JT, Berger, ET, and Lansbury, PT (1993). The carboxy terminus of the beta amyloid protein is critical for the seeding of amyloid formation: implications for the pathogenesis of Alzheimer's disease. Biochemistry 32, 4693-4697.

Jarvis TC, Wincott FE, Alby LJ, McSwiggen JA, Beigelman L, Gustofson J, DiRenzo A, Levy K, Arthur M, Matulic-Adamic J, Kaφeisky A, Gonzalez C, Woolf TM, Usman N, and Stinchcomb DT (1996) Optimizing the cell efficacy of synthetic ribozymes. J Biol Chem 271, 29107-29112.

Joyce (1992). In: Murray J.A.H. (Ed.,) Antisense RNA and DNA, Wiley-Liss, New York 353- 372.

Kim TW, Pettingell WH, Jung YK, Kovacs DM, Tanzi RE (1997). Alternative cleavage of Alzheimer-associated presenilins during apoptosis by a caspase-3 family protease. Science 277, 373-376.

Kim TW, Pettingell WH, Hallmark OG, Moir RD, Wasco W, and Tanzi RE (1997). Endoproteolytic cleavage and proteasomal degradation of presenilin 2 in transfected cells. J Biol Chem 272, 11006-11010.

Koizumi M, Hayase Y, Iwai S, Kamiya H, Inoue H, Ohtsuka E (1989). Design of RNA enzymes distinguishing a single base mutation in RNA. Nucleic Acids Res 17, 7059-7071. Koizumi M, and Ohtsuka E (1992). In: Murray J.A.H. (Ed.), Antisense RNA and DNA, Wiley- Liss, New York, 373-381.

Levy-Lahad E, Wasco W, Poorkaj P, Romano DM, Oshima J, Pettingell WH, Yu CE, Jondro PD, Schmidt SD, Wang K, Crowley AC, Fu Y, Guenette SY, Galas D, Nemens E, Wijsman EM, Bird TD, Schellenberg GD, Tanzi RE (1995). Candidate gene for the chromosome 1 familial Alzheimer's disease locus. Science 269, 973-977.

Lisziewicz J, Sun D, Smythe J, Lusso P, Lori F, Louie A, Markham P, Rossi J, Reitz M, Gallo RC (1993). Inhibition of human immunodeficiency virus type 1 replication by regulated expression of a polymeric Tat activation response RNA decoy as a strategy for gene therapy in AIDS. Proc Natl Acad Sci USA 90, 8000-8004.

Liu Z, and Carmichael GG (1994). Nuclear antisense RNA. Mol Biotechnol 2, 107-118.

Loetscher H, Deuschle U, Brockhaus M, Reinhardt D, Nelboeck P, Mous J, Gruenberg J, Haass C, Jacobsen H (1997). Presenilins are processed by caspase-type proteases. J Biol Chem 272, 20655-20659.

Luban J, Goff SP (1995). The yeast two-hybrid system for studying protein-protein interactions. Curr Opin Biotechnol 6, 59-64.

Mann, DMA, Iwatsubo, T, Cairns, NJ, Lantos, PL, Nochlin, D, Sumi, SM, Bird, TD, Poorkay, P, Hardy, J, Hutton, M, Prihar, G, Crook, R, Rossor, MN, and Haltia, M (1996). Amyloid beta protein (Abeta) deposition in chromosome 14-linked Alzheimer's disease: predominance of Abeta42(43). Ann Neurol 40, 149-156.

Mannino RJ, and Gould-Fogerite S (1988). Liposome mediated gene transfer. BioTechniques 6, 682-690.

Marcus-Sekura, CJ. (1988). Techniques for using antisense oligodeoxyribonucleotides to study gene expression. Anal. Biochem. 172, 289-295. Martin, SJ, and Green, DR (1995). Protease activation during apoptosis: death by a thousand cuts. Cell 82, 349-352.

Miller et al. (1991), in: Eckstein, F. (ed.) Oligonucleotides and analogues - A practical approach - Oxford, JRL Press, 137-154.

Morgan RA, and Anderson WF (1993). Human gene therapy. Annu Rev Biochem 62, 191-217.

Mullan, M, Crawford, F, Axelman, K, Houlden, H, Lilius, L, Winblad, B, and Lannfelt, L (1992). A pathogenic mutation for probable Alzheimer's disease in the APP gene at the N-terminus of beta-amyloid. Nature Genet 1, 345-347.

Murrell, J, Farlow, M, Ghetti, B, and Benson, MD (1991). A mutation in the amyloid precursor protein associated with the hereditary Alzheimer disease with a guanine to thymine missence change at position 1924 of the APP gene. Science 254, 97-99.

Naruse S, Thinakaran G, Luo JJ, Kusiak JW, Tomita T, Iwatsubo T, Qian X, Ginty DD, Price DL, Borchelt DR, Wong PC, Sisodia SS (1998). Effects of PSl deficiency on membrane protein trafficking in neurons. Neuron 21, 1213-1221.

Okochi M, Ishii K, Usami M, Sahara N, Kametani F, Tanaka K, Fraser PE, Ikeda M, Saunders AM, Hendriks L, Shoji SI, Nee LE, Martin JJ, Van Broeckhoven C, St George-Hyslop PH, Roses AD, Mori H (1997). Proteolytic processing of presenilin- 1 (PS-1) is not associated with Alzheimer's disease with or without PS-1 mutations. FEBS Lett 418, 162-6.

Paolella G, Sproat BS, Lamond Al (1992). Nuclease resistant ribozymes with high catalytic activity. EMBO J 11, 1913-1919

Pieken WA, Olsen DB, Benseler F, Aurup H, Eckstein F (1991). Kinetic characterization of ribonuclease-resistant 2'-modified hammerhead ribozymes. Science 253, 314-317.

Podlisny, MB, Citron, M, Amarante, P, Sherrington, R, Xia, W, Zhang, J, Diehl, T, Levesque, G, Fraser, P, Haass, C, Koo, EH, Seubert, P, St. George-Hyslop, P, Teplow, DB, and Selkoe, DJ (1997). Presenilin proteins undergo heterogenous endoproteolysis between Thr291 and Ala299 and occur as stable N- and C-terminal fragments in normal and Alzheimer brain tissue. Neurobiol Dis 3, 325-337.

Price DL, Sisodia SS (1998). Mutant genes in familial Alzheimer's disease and transgenic models. Annu Rev Neurosci 21, 479-505.

Rogaev El, Sherrington R, Rogaeva EA, Levesque G, Ikeda M, Liang Y, Chi H, Lin C, Holman K, Tsuda T, Mar L, Sorbi S, Nacmias B, Piacentini S, Amaducci L, Chumakov I, Cohen D, Lannfelt L, Fraser PE, Rommens JM, St George-Hyslop PH (1995). Familial Alzheimer's disease in kindreds with missense mutations in a gene on chromosome 1 related to the Alzheimer's disease type 3 gene. Nature 376, 775-778.

Rossi JJ (1993). Introductory remarks on the general application of antisense RNAs and ribozymes. Methods 5, 1-5.

Ruffner, DE, Stormo, GD, and Uhlenbeck, OC (1990). Sequence requirements of the hammerhead RNA self-cleavage reaction. Biochemistry 29, 10695-10702.

Sambrook et al. (1989). Molecular Cloning: A Laboratory Manual, 2^nd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.

Scheuner D, Eckman C, Jensen M, Song X, Citron M, Suzuki N, Bird TD, Hardy J, Hutton M, Kukull W, Larson E, Levy-Lahad E, Viitanen M, Peskind E, Poorkaj P, Schellenberg G, Tanzi R, Wasco W, Lannfelt L, Selkoe D, Younkin S (1996). Secreted amyloid Beta-protein similar to that in the senile plaques of Alzheimer's disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer's disease. Nature Med 2, 864-870.

Schumperli D, Howard BH, Rosenberg M (1982). Efficient expression of Escherichia coli galactokinase gene in mammalian cells. Proc Natl Acad Sci USA 79, 257-261.

Searle PF, Stuart GW, Palmiter RD (1985). Building a metal-responsive promoter with synthetic regulatory elements. Mol Cell Biol 5, 1480-1489. Shen J, Bronson RT, Chen DF, Xia W, Selkoe DJ, Tonegawa S (1997). Skeletal and CNS defects in Presenilin- 1 -deficient mice. Cell 89, 629-639.

Sherrington R, Rogaev El, Liang Y, Rogaeva EA, Levesque G, Ikeda M, Chi H, Lin C, Li G, Holman K, Tsuda T, Mar L, Foncin J, Bruni AC, Montesi MP, Sorbi S, Rainero I, Pinessi L, Nee L, Chumakov I, Pollen D, Brookes A, Sanseau P, St George-Hyslop PH (1995). Cloning of a gene bearing missense mutations in early-onset familial Alzheimer's disease. Nature 375, 754-760.

Shirotani K, Takahashi K, Tabira T (1999). Effects of presenilin N-terminal fragments on production of amyloid beta peptide and accumulation of endogenous presenilins. Neurosci Lett 262, 37-40.

Shirotani K, Takahashi K, Ozawa K, Kunishita T, Tabira T (1997). Determination of a cleavage site of presenilin 2 protein in stably transfected SH-SY5Y human neuroblastoma cell lines. Biochem Biophys Res Commun 240, 728-731.

Steiner H, Capell A, Pesold B, Citron M, Kloetzel PM, Selkoe DJ, Romig H, Mendla K, Haass C (1998). Expression of Alzheimer's disease-associated presenilin- 1 is controlled by proteolytic degradation and complex formation. J Biol Chem 273, 32322-32331.

Steiner H, Romig H, Grim MG, Philipp U, Pesold B, Citron M, Baumeister R, Haass C (1999). The biological and pathological function of the presenilin- 1 Deltaexon 9 mutation is independent of its defect to undergo proteolytic processing. J Biol Chem 274, 7615-7618.

Symons RH (1992). Small catalytic RNAs. Annu Rev Biochem 61, 641-671.

Thinakaran, G, Borchelt, DR, Lee, MK, Slunt, HH, Spitzer, L, Kim, G, Ratovitzky, T, Davenport, F, Nordstedt, C, Seeger, M, Hardy, J, Levey, Al, Gandy, SE, Jenkins, NA, Copeland, NG, Price, DL, and Sisodia, SS (1996). Endoproteolysis of presenilin 1 and accumulation of processed derivates in vivo. Neuron 17, 181-190. Thinakaran G, Harris CL, Ratovitski T, Davenport F, Slunt HH, Price DL, Borchelt DR, Sisodia SS (1997). Evidence that levels of presenilins (PSl and PS2) are coordinately regulated by competition for limiting cellular factors. J Biol Chem 272, 28415-28422.

Thomson JB, Tuschl T, Eckstein F (1993). Activity of hammerhead ribozymes containing non- 5 nucleotidic linkers. Nucleic Acids Res 21, 5600-5603.

Toh Y, Yamamoto M, Endo H, Misumi Y, Ikehara Y (1989). Isolation and characterization of a rat liver alkaline phosphatase gene. A single gene with two promoters. Eur J Biochem 182, 231- 237.

10

Tsang SY, Nakanishi M, Peterlin BM (1988). B-cell-specific and interferon-gamma-inducible regulation of the HLA-DR alpha gene. Proc Natl Acad Sci USA 85, 8598-8602.

Vito P, Wolozin B, Ganjei JK, Iwasaki K, Lacana E, and D'Adamio L (1996). Requirement of the i5 familial Alzheimer's disease gene PS2 for apoptosis. J Biol Chem 271, 31025-31028.

Weintraub HM (1990). Antisense RNA and DNA. Scientific American, 262, 34-40.

de Wet JR, Wood KV, DeLuca M, Helinski DR, Subramani S (1987). Firefly luciferase gene: 20 structure and expression in mammalian cells. Mol Cell Biol 7, 725-737.

Wisniewski T, Dowjat WK, Permanne B, Palha J, Kumar A, Gallo G, Frangione B (1997). Presenilin-1 is associated with Alzheimer's disease amyloid. Am J Pathol 151, 601-610.

Wolfe MS, Xia W, Ostaszewski BL, Diehl TS, Kimberly WT, Selkoe DJ (1999). Two transmembrane aspartates in presenilin-1 required for presenilin endoproteolysis and gamma- secretase activity. Nature 398, 513-517.

Wolozin B, Iwasaki K, Vito P, Ganjei JK, Lacana E, Sunderland T, Zhao B, Kusiak JW, Wasco o W, D'Adamio L (1996). Participation of presenilin 2 in apoptosis: enhanced basal activity conferred by an Alzheimer mutation. Science 274, 1710-1713. Wong PC, Zheng H, Chen H, Becher MW, Sirinathsinghji DJ, Trumbauer ME, Chen HY, Price DL, Van der Ploeg LH, Sisodia SS (1997). Presenilin 1 is required for Notch 1 and Dill expression in the paraxial mesoderm. Nature 387, 288-292.

5 Wu HN, Lee JY, Huang HW, Huang YS, Hsueh TG (1993). Mutagenesis analysis of a hepatitis delta virus genomic ribozyme. Nucleic Acids Res 21, 4193-4199.

Yamada O, Yu M, Yee JK, Kraus G, Looney D, Wong-Staal F (1994). Intracellular immunization of human T cells with a haiφin ribozyme against human immunodeficiency virus type 1. Gene ιo Ther 1, 38-45.

Yu M, Ojwang J, Yamada O, Hampel A, Rapapport J, Looney D, Wong-Staal F (1993). A haiφin ribozyme inhibits expression of diverse strains of human immunodeficiency virus type 1 Proc Natl Acad Sci USA 90, 6340-6344. [published erratum appears in Proc Natl Acad Sci USA ,5 (1993) 90, 8303],

Yuan Y, Altman S (1994). Selection of guide sequences that direct efficient cleavage of mRNA by human ribonuclease P. Science 263, 1269-1273.

2o Zuker M. (1989). Computer prediction of RNA structure. Methods Enzymol 180, 262-288.

Claims

claims

1. Method for identifying a substance capable of reducing or eliminating the activity of the presenilinase comprising: a) cultivating a cell or a cell line to express said presenilinase activity and a fusion-protein, said fusion-protein comprising the full-length presenilin 1 or presenilin 2 and a reporter b) incubating said cell or cell line with a test substance c) measuring the quantity of the full-length fusion-protein employing the reporter, and d) comparing the quantity of full-length fusion-protein obtained in c) to the quantity of full- length fusion-protein measured for a control

2. Method according to claim 1, wherein the presenilinase is specific for presenilin 1.

3. Method according to claim 1, wherein the presenilinase is specific for presenilin 2.

4. Method according to any one of claims 1 to 3, wherein the substance capable of reducing or eliminating the activity of the presenilinase reduces or eliminates the autoproteolytical cleavage of the presenilin.

5. Method according to any one of claims 1 to 4, wherein the substance capable of reducing or eliminating the activity of the presenilinase reduces or eliminates the activity of ╬│-secretase.

6. Method according to any one of claims 1 to 5, wherein the substance prevents cleavage of presenilin 1 into fragments of regulated endoproteolytic cleavage wherein the N-terminal fragment (NTF) is approximately 21-28 kDa in size and the C-terminal fragment (CTF) is approximately 16-24 kDa in size.

7. Method according to any one of claims 1 to 5, wherein the substance prevents cleavage of presenilin 2 into fragments of regulated endoproteolytic cleavage wherein the N-terminal fragment (NTF) is approximately 28-30 kDa (NTF) kDa in size and the C-terminal fragment (CTF) is approximately 20-25 kDa in size.

8. Method according to any one of claims 1 to 7, wherein the reporter is fused to the N-terminus of the presenilin.

9. Method according to any one of claims 1 to 8, wherein measuring the quantity of full-length fusion-protein employing the reporter comprises: i) immobilizing antibodies specific for the C-terminal portion of presenilin 1 or presenilin

2 to a solid surface ii) extracting the protein from said cell or cell line after cultivation iii) incubating said protein extracts with said antibodies, and iv) measuring the amount of full-length fusion-protein bound by said antibodies by detection of the reporter.

10. Method according to any one of claims 1 to 9, wherein any unbound fusion-protein fragments are measured by detection of the reporter after incubating said protein extract. s

11. Method according to any one of claims 1 to 10, wherein the cell or cell line is expressing said fusion-protein at such a level that little or no full-length fusion-protein is detected when no substance capable of reducing or eliminating the activity of the presenilinase is present.

12. Method according to any one of claims 1 to 11, wherein the reporter is luciferase.

13. Method according to any one of the claims 1 to 12, wherein the method is a high throughput o screening assay (HTS).

14. Substance capable of reducing or eliminating the activity of the presenilinase, identifiable with a method according to any one of the claims 1 to 13.

15. Substance according to claim 14, wherein the substance is selected from the group consisting of: s a) a substance capable of reducing or eliminating the enzymatic activity of the presenilinase or b) a substance capable of reducing or eliminating the expression of the presenilinase at the translational or transcriptional level or c) a substance capable of reducing or eliminating the expression of the substrate of the 0 presenilinase or d) a substance capable of reducing or eliminating the formation of complexes between presenilin fragments.

16. Substance according to any one of claims 14 or 15, wherein the substance is capable of reducing the amount of presenilin fragments and preventing neuronal cell death and/or capable 5 of reducing the deposition of A╬▓ and/or reducing the formation of amyloid plaques.

17. Substance according to any one of claims 14 to 16. wherein the presenilinase is specific for presenilin 1.

18. Substance according to any one of claims 14 to 16, wherein the presenilinase is specific for presenilin 2. o

19. Substance according to any one of claims 14 to 18, wherein the substance reduces or eliminates the autoproteolytical cleavage of the presenilin.

20. Substance according to any one of claims 14 to 19, wherein the substance reduces or eliminates the activity of ╬│-secretase.

21. Substance according to any one of claims 14 to 20, wherein the substance prevents cleavage of presenilin 1 into fragments of regulated endoproteolytic cleavage wherein the N-terminal fragment (NTF) is approximately 21-28 kDa in size and the C-terminal fragment (CTF) is approximately 16-24 kDa in size.

22. Substance according to any one of claims 14 to 20, wherein the substance prevents cleavage of presenilin 2 into fragments of regulated endoproteolytic wherein the N-terminal fragment (NTF) is approximately 28-30 kDa (NTF) kDa in size and the C-terminal fragment (CTF) is approximately 20-25 kDa in size.

23. Substance according to any one of claims 14 to 22, wherein the substance is an antisense- oligonucleotide or a ribozyme.

24. Pharmaceutical composition comprising a substance according to any one of claims 14 to 23 and a pharmaceutically acceptable carrier therefor.

25. Use of a substance according to any one of claims 14 to 23 in the manufacture of a medicament for the treatment of neurodegenerative diseases.

26. Use of a substance according to any one of claims 14 to 23 in the manufacture of a medicament for the treatment of Alzheimer's disease.

27. Use of a substance according to any one of claims 14 to 23 in the manufacture of a medicament for the treatment of familiar Alzheimer's disease.