WO2006007864A1 - Treating neurodegenerative conditions - Google Patents

Abstract

The present invention relates to the use of compounds capable of inhibiting protein aggregate formation and capable of depolymerising protein aggregates for the preparation of a pharmaceutical composition for treating neurodegenerative conditions such as Alzheimer disease.

Description

Treating Neurodegenerative Conditions

The present invention relates to the use of compounds capable of inhibiting protein aggregate formation and capable of depo- lymerising protein aggregates for the preparation of a pharma¬ ceutical composition for treating neurodegenerative conditions such as Alzheimer disease.

Alzheimer's disease (AD) is the most common cause of dementia in the middle-aged and the elderly and is responsible for about 50% of all cases of senile dementia in North America and Western Europe (Iqbal, K. and Grundke-Iqbal, I. 1997) . In Alzheimer's disease two main proteins or fragments thereof form abnormal polymers (review, Selkoe 2003) . Proteins precipitated in amy¬ loid plaques between cells largely consist of polymerised AB- peptide. The microtubule-associated protein tau occurs inside the cells and produces neurofibrillary tangles ((Lee et al. , 2001), Buee et al . 2002) . It is believed that these insoluble aggregates or their oligomeric precursors are responsible for the neuronal degeneration that leads to the cognitive impair¬ ment typical for the disease. The distribution of the neurofi¬ brillary changes has been used for the staging of Alzheimer' s disease (Braak and Braak, 1991) which is part of the guide¬ lines for post mortem diagnosis (Ball and Murdoch, 1997) . The Braak staging is based on the appearance of tau in an aggre¬ gated state which in addition is chemically modified in sev¬ eral ways (phosphorylation, truncation, glycation; Johnson and Bailey, 2002) . Whether these modifications are the cause, con¬ sequence, or merely byproducts of neuronal degeneration is still a matter of debate. For example, different kinases and pathways of phosphorylation have been suggested to be respon¬ sible for early stages of degeneration in neurons (Brion et al., 2001; Liu et al. , 2002; Maccioni et al. , 2001; Zhang and Johnson, 2000) , but in vitro the phosphorylation of tau does not appear to promote aggregation (Eidenmuller et al. , 2001; Schneider et al. , 1999) . Examples from other protein aggrega¬ tion diseases suggest that an increase in concentration drives the protein into aggregation which in turn causes the toxic effects (Bonifacio et al. , 1996; Goldberg and Lansbury, 2000; Rochet and Lansbury, 2000; Shtilerman et al. , 2002) . Con¬ versely, measures that reduce the concentration of oligomers and aggregates alleviate the diseases (Beirao et al. , 1999; Lambert et al. , 2001; Sanchez et al. , 2003; Schenk et al. , 1999) .

There are different views on whether cytotoxicity due to Aβ ag¬ gregation is transmitted from outside the cell or acts during the maturation of pre-fibrillary oligomers within the cells (for review see Glabe, 2001) . Since the discovery of inherited tau pathologies (FTDP-17) they were studied in animal and cel¬ lular models (for review see Hutton et al. , 2001)) . Consider¬ able progress has been made in creating tau pathology in transgenic mice (Duff et al. , 2000; Gotz, 2001; Higuchi et al., 2002a; Higuchi et al. , 2002b; Hutton, 2001; Lewis et al. , 2000) or other organisms (Hall et al. , 2000; Jackson et al. , 2002; Kraemer et al. , 2003; Wittmann et al. , 2001), but these do not yet reflect the full spectrum of the human pathology, and it is not clear what role tau protein and its aggregation plays in cytotoxicity. Much of the evidence for cytotoxicity of intracellular aggre¬ gates comes from other neurodegenerative diseases like Parkin¬ son's and Huntington' s disease (for reviews see Goedert et al, 1998; Voiles and Lansbury, 2003) . In Parkinson's disease the cytotoxicity of α-synuclein has recently been traced back to pre-fibrillary oligomers that bind to membranes (Caughey and Lansbury, 2003) . In Huntington's disease it is reported that aggregated protein can be found in the nucleus (Bates, 2003) , possibly affecting gene transcription, and in a mouse model it was shown that disaggregation of polymers leads to a prolonged life-time (Sanchez et al. , 2003) .

In the case of tau there is the paradoxon that the protein is intrinsically highly soluble, yet it can aggregate into in¬ soluble polymers. The soluble form of tau is characterised as a natively unfolded protein with mostly random coil conforma¬ tion, as judged by CD or FTIR-spectroscopy, small angle X-ray scattering, gel filtration and limited proteolysis (Schweers et al., 1994; Friedhoff et al. , 1998; von Bergen et al. , 2000) . However, the tau sequence contains certain motifs that may undergo a conformational change towards β-sheet structure. This can drive the protein into filaments that are indistin¬ guishable from those of Alzheimer's brain. Since the intracel¬ lular aggregation of tau in AD correlates with the clinical progression of the disease it seemed likely that inhibition or even reversal of the tau aggregation would protect or rescue the affected neurons

Substances that inhibit tau formation were consequently identi¬ fied in the art. U.S. patent No. 6,479,528 for example discloses that certain inhibitors of fatty acid oxidation also inhibit tau filament formation. Further, WO 03/007933 discloses naphtoqui- none-type compounds and their use to modulate the aggregation of proteins associated with neurodegenerative disease. Wischik et al. (1996) describes inhibition of tau aggregation by phenothi- azines.

Surprisingly, the compounds of the present invention were found to be more effectiv or efficient, respectively, in relation to drugs known in the art.

Due to the importance of neurodegenerative conditions today it there is a considerable need for new pharmaceutical composi¬ tions for the treatment of neurodegenerative conditions. The present invention specifically addresses this problem.

The above problem is solved by the use of compounds capable of inhibiting protein aggregate formation and capable of depoly- merising protein aggregates for the preparation of a pharma¬ ceutical composition for treating a neurodegenerative condi¬ tion.

Using a novel screening assay the inventors surprisingly iden¬ tified a group of specific compounds that are capable of in¬ hibiting protein aggregate formation and capable of depolym- erising pre-formed protein aggregates.

The compounds of the present invention are piperazines having a molecular weight of about 423.56 to about 509.65.

Compounds of the present invention are capable of inhibiting protein aggregate formation. The feature "capable of inhibit¬ ing protein aggregate formation" as used in the present appli¬ cation refers to the inherent activity of a compound to de¬ crease protein aggregate formation in vitro in comparison to a control reaction in the absence of the compounds . The in vitro test is preferably a thioflavine S fluorescence assay, such as the assay illustrated in Example 2 of the present application. A compound is capable of inhibiting protein aggregate forma- tion in that assay if a preferably more than >30%, preferably more than >40%, preferably more than >50%, preferably more than >60%, preferably more than >70%, more preferred >80% and most preferred >90% decrease of the signal is obtained.

Depolymerising pre-formed protein aggregates is a further im¬ portant aspect of the medical use of the compounds of the pre¬ sent invention. The feature "capable of depolymerising protein aggregates" as used in the present application refers to the inherent activity of a compound to depolymerise protein aggre¬ gates in an in vitro assay, such as in the thioflavine S fluo¬ rescence assay as illustrated in Example 3. A compound is ca¬ pable of depolymerising protein aggregates if in comparison to a control reaction in the absence of respective compounds preferably more than >30%, preferably more than >40%, prefera¬ bly more than >50%, preferably more than >60%, more preferred >70% and most preferred >80% decrease of the signal is ob¬ tained.

In a preferred embodiment of the present invention the com¬ pounds are used to inhibit protein aggregates that comprise paired helical filaments (PHFs) consisting of tau protein. The tau protein belongs to a class of microtubule-associated pro¬ teins (MAPs) expressed in mammalian brain that regulate the extensive dynamics and rearrangement of the microtubule net¬ works in the cells . The abnormal aggregation of tau in the form of PHFs is one of the hallmarks of Alzheimer's disease. Aggregation occurs in the cytoplasm and will therefore be toxic for neurons .

According to a further preferred embodiment the compounds are used to inhibit protein aggregates comprising Aβ protein, prion protein, α-synuclein, serum amyloid, transthyretin, hunting- tin, insulin or antibody light chain. According to an especially preferred aspect the present inven¬ tion is directed to the medical use of a compound having the following general formula:

wherein Rl and R2 is selected from H and

and R3 is selected from H, OCH₃ and F.

R4 is selected from H and CH₃, or R3 and R4 are connected to form a condensed pyrrole ring.

R5, if any, is selected from H and OCH₃.

Rβ may be H and R7 may be H, or Rβ and R7 may be connected to form a condensed phenyl ring.

R8 is selected from CH₂CH₂OH, CH₂Ph and C(O)OCH₂CH₃, and X¹, X' ', X"¹, and X' ' ' ' are selected from N and C.

The compound preferably has one of the following formulas:

In an alternative embodiment of the invention the medical use comprises the use of a compound having the following general formula :

wherein R9 is selected from

and RlO is selected from H and NO2.

RIl is selected from an N-morpholino group, iV-pyrrolidino group and OCH₃.

In a preferred embodiment the compound may have one of the formulas :

According to a further aspect compound may be selected from the group of compounds listed in Table 1.

In a preferred embodiment of the invention the compound is se¬ lected from Table 3.

The present invention is based on a method of screening for compounds that are capable of inhibiting PHF formation and ca¬ pable of depolymerising PHFs. Briefly such a method can be de¬ scribed as follows .

First a random library is screened for identifying compounds that are capable of inhibiting protein aggregate formation and capable of depolymerising protein aggregates. Any assay suit¬ able for assessing the capability of inhibiting protein aggre- gate formation or the capability of depolymerising pre-formed protein aggregates can be used.

The compounds that are identified as compounds capable of in^¬ hibiting protein aggregate formation and capable of depolym¬ erising protein aggregates, are then used for carrying out an in silico search to identify potential further compounds. In a second in vitro screen these potential new candidates are tested for their capability to inhibit protein aggregate for¬ mation and/or depolymerise protein aggregates .

For example, the described method may comprise the following steps :

In a first screen initially a thioflavine S assay (Example 2) is used to screen a random library for identifying compounds which are capable of inhibiting the PHF formation. The assay is based on the fluorescence of thioflavine S that is in¬ creased by binding to PHFs .

The compounds identified are then tested using a thioflavine S assay for their ability to depolymerise pre-formed PHFs (Exam¬ ple 3) in a second step.

Additional assays can be used for testing the ability to in¬ hibit PHF formation or the ability to depolymerise PHFs . Such assays comprise the tryptophan fluorescence assay (Li et al. (2002)) . This assay is independent of exogenous dyes. It re¬ lies on the change of the emission maximum of a tryptophan in¬ troduced instead of tyrosine 310 whose emission maximum is sensitive to the burial in a more hydrophobic surrounding upon PHF formation (see Example 6) .

Further assays suitable for the present invention are electron microscopy, filter assay or a pelleting assay. Electron mi- croscopy has been used previously to analyse PHF formation (Wille et al. 1992;; Schweers et al, 1995; Friedhoff et al, 1998a; Friedhoff et al . , 1998b) A filter assay has first been used for the analysis of huntingtin aggregates (Heiser et al. , 2000) , recently an application for tau-aggregates has been reported (Dou et al, 2003) .

In the next step compounds that inhibit PHF formation and de- polymerise PHFs are used to define patterns for an in silico homology search of a virtual library of chemical structures . For obtaining reasonable results from an in silico search, compounds which should build the basis of the search have to be selected carefully and several parameters have to be de^¬ fined.

The compounds were selected with regard to common three dimen¬ sional properties (lipophilie, shape and HH-binding ability) and chemical stability. Compounds with a molecular weight higher than 500 were excluded as well as structures with highly polar and reactive groups, for example SH-groups, hal- ide and azo-structures . The number of freely rotateable bonds was minimizez.

In a preferred embodiment the compounds are selected with re^¬ gard to common three-dimensional structure (e.g., shape and binding activity) and chemical stability. Parameters such as size, number of freely rotatable bonds, and inclusion or ex¬ clusion of specific groups, such as highly polar or reactive groups, should be defined. In a preferred embodiment of the invention, compounds with a molecular weight higher than 500 are excluded as well as structures with highly polar and reac¬ tive groups - for example SH-groups, halide and azo struc¬ tures, and the number of freely rotatable bonds is minimised. The compounds, identified by the in silico search, are subse^¬ quently tested in vitro for their ability to inhibit PHF for^¬ mation and depolymerise PHFs with the above methods .

As shown in Example 13, using this strategy leads to a sub^¬ stantial increase of the fraction of compounds that are capa¬ ble of depolymerising protein aggregates (Fig.13) .

The present invention is further directed to the preparation of pharmaceutical compositions for the treatment of neurode^¬ generative conditions .

In a preferred embodiment the neurodegenerative condition is Alzheimer's disease. Alzheimer's disease is characterised by two characteristic types of protein deposits, the first type consists of amyloid precursor protein (APP) and the second type of neurofibrillary tangles of paired helical filaments (PHFs) . The compounds and pharmaceutical compositions of the present invention are particularly suitable for the treatment of Alzheimer's disease.

In a further preferred embodiment the present invention is di¬ rected to the use of a compound of formula LSA (above) for the preparation of a pharmaceutical composition for treating Alz¬ heimer's disease.

In an alternative embodiment the invention is directed to the use of a compound of formula LSB (above) for the preparation of a pharmaceutical composition for treating Alzheimer's dis¬ ease.

In yet another embodiment the invention is directed to the use of a compound selected from the group of compounds shown in Table 1 for the preparation of a pharmaceutical composition for treating Alzheimer's disease. The invention further contemplates the medical use of the com^¬ pounds for treating other neurodegenerative conditions, such as those selected from the group of tauopathies consisting of CBD (Cortical Basal Disease) , PSP (Progressive Supra Nuclear Palsy) , Parkinsonism, FTDP-17 (Fronto-Temporal Dementia with Parkinsonism linked to chromosome 17) , Familiar British Demen¬ tia, Prion Disease (Creutzfeld Jakob Disease) and Pick's Dis^¬ ease.

The term "taupathies" as used herein refers to pathologies characterised by aggregated tau into paired helical filaments leading to neurodegeneration.

According to the present invention the pharmaceutical composi¬ tion may be administered orally or parenterally.

In a further aspect the invention is directed to the use of the compounds for the preparation of a pharmaceutical composi¬ tion that is administered as part of a sustained release for¬ mulation resulting in slow release of the compound following administration. Such formulations are well known in the art and may generally be prepared using well known technology, for example, by implantation at the desired target site, e.g. in the brain (Sheleg et al . , 2002) .

The pharmaceutical compositions of the invention may comprise additional compounds such as a pharmaceutically acceptable carrier, diluents, stabilising agents, solubilisers, preserv¬ ing agents, emulsifying agents and the like.

The following Examples illustrate the inhibition of protein aggregate formation and the depolymerisation of pre-fornαed protein aggregates . The invention also comprises a Tau transgenic mouse line with a genetic switch that can be operated at will and that permits the control of the Tau gene activity quantitatively and re- versibly in a temporal, spatial, and tissue-specific manner. The transgenic mice allows the expression of human tau iso- forms (or mutants thereof) or its domains in the central nerv¬ ous system (CNS) of mice to determine the effects of tau over- expression. Examples are the effects on the intracellular transport of vesicles and cell organelles in neurons, on the binding of tau to microtubules, and on the aggregation of tau into Alzheimer paired helical filaments (PHFs) .

Conditional expression of genes in eukaryotic cell systems and mice can be achieved by the tet-regulated system (Furth et al., 1994) . The regulation is done through the tetracycline- regulated transactivator (tTA) (Gossen et al. , 1995) . Fig. 14 (adapted from Gossen et al. , 1995) illustrates the mechanism of action of the Tc-controlled transactivator by the tetracy- clin derivative doxycyclin (Dox) .

The rtTA system is a variant of the tTA system. It is identi¬ cal with the exception of 4 amino acid exchanges in the tetR moiety. These changes convey a reverse phenotype to the re¬ pressor (rtetR) . The resulting rtTA requires doxycyclin for binding to tetO and thus for transcription activation. (Gossen et al . , 1995) . Tissue specificity of these systems is achieved by placing the tTA or rtTA gene under the control of a tissue specific promoter (P_sp) , for example the CaMKIIα-promotor for expression in the CNS.

The invention also comprises inducible cell lines for studying the aggregation of Tau protein that is characteristic of Alz¬ heimer's disease and related tauopathies . This allows one to study the toxicity of Tau in cells either in the soluble or aggregated state, the dissolution of Tau aggregates after switching off the Tau gene expression, and the efficiency of small molecule aggregation inhibitors identified by an in vi¬ tro screen.

EXAMPLES

Chemicals and proteins used:

Heparin (average MW of 6000) , poly-glutamate (average MW of 600 or 1000), thioflavine S was obtained from Sigma. Full- length tau isoforms htau23, htau24 and constructs of the re¬ peat domain of tau (Fig. 1) were expressed in E. coli and pu¬ rified by making use of the heat stability and FPLC Mono S (Pharmacia) chromatography as described. The purity of the proteins was analysed by SDS-PAGE, protein concentrations were determined by the Bradford assay. Emodin, Daunorubicin and Adriamycin were obtained from Merck (Germany) . PHF016 was ob¬ tained from ChemBridge (USA) and PHF005 was obtained from In- terchim (France) . All experiments presented here were carried out with freshly dissolved compounds.

EXAMPLE 1 PHF formation in vitro

Assembly of synthetic PHFs from tau protein (K19, 10 μM) was performed at 37 ⁰C in the presence of polyanions (heparin; 5 μM) in 50 mM NH₄Ac, pH 6.8. Assembly was followed either quali¬ tatively by electron microscopy or quantitatively by fluores¬ cence assay using thioflavine S. PHF-formation from tau iso¬ forms htau23 and htau24 was carried out in PBS-buffer pH 7.4, 50 μM protein, and 12.5 μM heparin. The samples were incubated at 50 °C for 10 days. In the case of hTau24 and K18, DTT was added at a final concentration of 1 mM each day in order to avoid intra-molecular disulfide crosslinking (Barghorn et al. , 2000) . EXAMPLE 2

Screening of compounds capable of inhibiting PHF formation with the thioflavine S assay

PHF formation was monitored by a thioflavine S fluorescence assay (Friedhoff et al. , 1998a) adapted to a 384 well format. 60 μM of each substance was tested for its inhibitory effect on PHF formation. Using an automated pipetting system (Cybi- WeIl, CyBio, Jena, Germany) 50 mM NH₄Ac, 10 μM protein (K19) , 60 μM compound and 5 μM heparin were mixed in 50 μl volume in a 384 well plate (black microtiter 384 plate round well, Ther- moLabsystems, Dreiich, Germany) and incubated overnight at 37 °C. As a control the protein was replaced with H₂O to measure the possible fluorescence of the compounds. As a second con¬ trol the reaction mixture without compound was treated in the same way.

After incubation with the compounds thioflavine S was added to the buffer to a final concentration of 20 μM and the signal was measured at excitation at 440 nm and emission at 521 nm in a spectrofluorimeter (Ascent; Labsystems, Frankfurt) .

Hits were defined by a >90% decrease of the signal in compari¬ son to the (second) control reaction without compound.

EXAMPLE 3

Screening of compounds capable of depolymerising PHFs with the thioflavine S assay

Depolymerisation of PHFs was monitored by the thioflavine S fluorescence. 60 μM of each compound was tested for its abil¬ ity to depolymerise pre-formed PHFs. Using an automated pipet¬ ting system (Cybi-Well, CyBio, Jena, Germany) 50 mM NH₄Ac, 10 μM PHF (K19) , 60 μM compound and 5 μM heparin were mixed in 50 μl volume in a 384 well plate (black microtiter 384 plate round well, ThermoLabsystems, Dreiich, Germany) and incubated overnight at 37 ⁰C. As a control the reaction mixture without compound was treated in the same way.

After incubation thioflavine S was added to the buffer to a final concentration of 20 μM and the signal was measured at excitation at 440 nm and emission at 521 nm in a spectro- fluorimeter (Ascent; Labsystems, Frankfurt) .

Hits were defined by a >80% decrease of the signal in compari¬ son to the control reaction without compound.

EXAMPLE 4

Inhibition of PHF formation using various concentrations of compounds and various constructs

This Example describes the ability of the five compounds Adriamycin, Daunorubicin, Emodin, PHF005 and PHF016 (Fig. IA- E) to inhibit PHF formation. Additionally to the construct K19 (Fig. II) the four repeat construct Kl8 (Fig. IH) and the re¬ lated full length isoforms htau23 (three repeat, no inserts, Fig. IG) and htau24 (four repeats, no inserts, Fig. IF) were also used.

Using fixed protein concentrations of Kl9 the compounds were tested in a concentration range from 0.01 nM to 100 μM (Fig. 2A) and IC₅O values are determined (Table 2) . Inhibitory ef¬ fects begin to appear at concentrations around 0.1 μM (ratio of protein to compound = 100) and reach nearly complete inhi¬ bition at 100 μM compound concentration (ratio pro¬ tein/compound = 0.1) . Overall, the curves of Fig. 2A decay fairly steeply over a compound concentration range of 2-3 or¬ ders of magnitude. The values of half-maximal inhibition (IC₅₀) range from 1.0-17.6 μM, which means that all compounds inter¬ fere with PHF aggregation of Kl9 already at substoichiometric concentrations.

The four repeat construct K18 was tested under the same condi¬ tions (Fig. 2B) . The compounds exhibit IC₅₀ concentrations be¬ tween 0.1 and 0.6 μM, except for PHF005 whose IC₅₀ is 2.7 μM. However, the decay of the curves of K18 is more gradual than those of K19, extending over 3-4 orders of magnitude of com¬ pound concentration (compare Fig. 2A) .

The study was then extended to the natural three and four re¬ peat isoforms htau23 and htau24 (Fig. IG,F) . PHF formation of these proteins was assayed in the presence of 0.1, 1, 10 and 60 μM compound (Fig. 2C,D) . For htau23 a clear dose dependent inhibition was observed (Fig. 2C) . The compounds can be subdi¬ vided into two groups . The more effective compounds are adria- mycin, daunorubicin and emodin which are capable to inhibit PHF formation about 50% at 0. lμM and -90 % at 60 μM. Compounds PHF016 and PHF005 are less inhibitory, they show only a slight effect at low concentration and a moderate one (~50%) at 60μM. In the case of htau24 (4 repeats) the compounds show generally a lower efficiency of inhibition than for htau23 (Fig. 2D) , but the internal ranking stays the same. The more active com¬ pounds emodin, daunorubicin and adriamycin reach inhibition levels of 70-90% at 60μM concentration. PHF016 and PHF005 show clear differences in their capacity to influence PHF forma¬ tion; only a small effect is seen with 4-repeat htau24, com¬ pared to htau23.

All the polymerisation reactions described so far used heparin as a cofactor for inducing PHF assembly because otherwise the process would be impracticably slow (Goedert et al. , 1996; Perez et al., 1996) . In order to rule out a potential influ¬ ence of heparin on the efficiency of the compounds we used the 4-repeat construct K18/ΔK280 which carries one of the muta-. tions observed in frontotemporal dementia (van Swieten et al. , 1999) and is capable of polymerising into PHFs without a poly- anionic cofactor (von Bergen 2001) (Fig. 2E) . The resulting IC50 values for the inhibition of filament formation from K18/ΔK280 are significantly higher than for Klδwt. The most effective ones are emodin, adriamycin and PHFOl6 which range from 1.3 to 3.9.μM. Daunorubicin which was very active in the case of K19 exhibits an IC50 of 48.μM and PHF005 which was the least efficient inhibitor of K19 and K18 filament formation fails nearly completely. The differences in inhibition effects for K18 (with heparin) and K18/ΔK280 (without heparin) could be caused either by a difference in conformation and/or pro¬ tein-protein interactions, or perhaps by an interaction be¬ tween the compound and the cofactor heparin.

EXZVMPLE 5

Depolymerisation of PHFs using various concentrations of com¬ pounds and various constructs

The ThS assay was used to analyse the ability of the five com¬ pounds (see Example 4) to depolymerise pre-formed PHFs made from the repeat domain constructs Kl9 and Kl8 as well as from isoforms htau23 and htau24, containing 3 or 4 repeats, respec¬ tively.

The depolymerisation of K19 filaments (Fig. 4A) follows a similar concentration dependence as the inhibition experiment, with consistently similar or slightly higher DC50 values than the corresponding IC50 concentrations (Table 2) . The ratios of IC50/DC50 range from 0.2-1.2. By contrast, K18 filaments ap¬ pear to be much more stable and therefore depolymerise less readily, resulting in higher DC50 values between ~6.5 and 43 μM (Fig. 4B) . Here, too, the concentration dependence for K19 is steeper than for Kl8 (compare Fig. 4A, B) , similar to that of assembly inhibition (Fig. 2A, B) . Thus the relationship be¬ tween assembly inhibition and disassembly promotion (IC50 vs. DC50) is less apparent for K18 than for K19, suggesting that the second repeat R2, present only in K18, confers higher sta¬ bility to the polymer.

All compounds are also able to dissolve PHFs made of K18/ΔK280 without heparin (Fig. 4E) in a dose dependent manner, but ex¬ hibiting higher DC50 values than PHFs made from K18. Simi¬ larly, the compounds show a lower activity in depolymerising PHFs made from K18/ΔK280 (Fig. 4E), compared to inhibition of polymerisation, consistent with the experiments on K19 and K18. Emodin, daunorubicin and adriamycin show DC50 values be¬ tween 2.2 and 22.0 μM, whereas the DC50 values of PHF016 and PHF005 are not accurately detectable due the low efficiency of depolymerisation under these conditions. The higher DC50 val¬ ues for K18/ΔK280 point to the higher stability of PHFs formed by this mutant.

PHFs assembled from the full length three repeat isoform htau23 are also sensitive for disaggregation (Fig. 4C) .The DC₅₀ values range from 7.0 to 60μM. All values are higher than the IC50 values, but the internal ranking of the compound stays the same, Emodin, daunorubicin and adriamycin (DC₅₀ range 7.0- 13.2μM) have a much stronger effect than PHFOl6 and PHF005. (DC₅₀ >60μM) . This is consistent with the similar rank¬ ing of compounds in the assembly inhibition assay of full- length tau isoforms (Fig. 2C, D) .

By contrast, even the most potent compounds in depolymerising htau23 filaments (emodin, daunorubicin and adramycin, Fig. 4C) are only weak PHF breakers for htau24 filaments (Fig. 4D) . All compounds exhibit a comparable low efficiency, the best values are achieved for PHF016 and PHF005 with DC50 values of 39.2 and 10.8μM respectively. At the lowest concentration (O.lμM) none of the compounds are able to decrease the level of ThS fluorescence significantly, whereas at the highest concentra¬ tion (60μM) the ThS fluorescence is decreased to a range of 10-55%. PHFOlβ and PHF005 are more active in depolymerising htau23 than htau24 filaments. This difference can be explained both by an increased stability of four repeat isoforms and by a isoform specific mode of action of the compounds.

Table 2: IC₅₀/DC₅₀ values of inhibition of PHF aggregation/ de- polymerisation of PHFs from tau and tau constructs

EXAMPLE 6 Tryptophan fluorescence spectroscopy

In order to exclude a possible distortion of the data by the dye the results of the ThS assays can be confirmed by a tryp¬ tophan fluorescence assay (Li et al. , 2002) . It allows the de¬ tection of the molecular environment of a tryptophan intro¬ duced instead of tyrosine 310 whose emission maximum is sensi¬ tive to the burial in a more hydrophobic surrounding upon PHF formation. Therefore the mutants K19/Y310W (Fig. II) and K18/Y310W (Fig. IH) that contain a single tryptophan within the core of the PHF structure were created. In the soluble protein the emission maximum lies at ~354 nm, whereas it shifts to 340 nm upon PHF formation (Fig. 3A, compare first and second entry) . The emission peak can be shifted back by incubation at high concentrations of GuHCl which is due to the disassembly of the PHFs (Fig. 3A, fourth entry) .

The fluorescence experiments were performed on a Spex Fluoro- max spectrophotometer (Polytec, Waldbronn, Germany) , using 3 mm x 3 mm micro cuvettes from Hellma (Muhlheim, Germany) with 20 μl sample volumes. A tryptophan emission spectrum scans from 300 to 450 nm at fixed excitation wavelength of 290 nm. The slit widths were 5 nm, the integration time was 0.25 sec¬ ond, and the photomultiplier voltage was 950 V. For fluores¬ cence inhibition assay, 60 μM compounds were incubated with K19/Y310W construct (10 μM) or K18/ΔK280/Y310W and heparin (2.5 μM) in PBS, pH 7.4 three days at 37 ⁰C.

In the Trp fluorescence assay the inhibition of PHF assembly becomes apparent if the emission maximum of Trp310 remains higher than that of the control without any compound, because Trp310 remains in a more solvent-accessible hydrophilic envi¬ ronment. The three repeat tau construct Kl9 (at lOμM) is pre¬ vented from polymerisation by about 90 % by the presence of all compounds at a concentration of 60μM (Fig. 3A, note that entries 5-9 retain their values around 354 nm, similar to the control #1) . By contrast the four repeat tau construct K18/Y310W is inhibited to this high extent only by PHF005 (Fig. 3B, entry #9) . Emodin, daunorubicin and adriamycin can prevent PHF formation to about 70% at 60 μM (Fig. 3B, entries #5, 6, 7), whereas PHF016 achieves only 25% inhibition (#8) . The trend becomes even more pronounced in the case of K18/ΔK280, where all compounds show a lower activity. The in- ternal ranking stays roughly the same as with K18; PHF005 (#9) is the best, PHFOlβ (#8) the worst inhibitor. Emodin, daunoru- bicin and adriamycin (#5, 6, 7) showed a level of ~30-50% in¬ hibition. The apparent degrees of inhibition differ somewhat between the ThS fluorescence and the intrinsic Trp fluores¬ cence assays, but this may be due to the different origins of the signal. In the ThS assay the dye has to bind to the fila¬ ments, which requires a minimal length of the fibres. The tryptophan fluorescence assay depends on the local surrounding of the residue and is therefore less dependent on the length of the filaments .

For the fluorescence depolymerisation assay, 60 μM inhibitor compound were added to pre-formed PHFs (10 μM) and incubated overnight at 37 ⁰C. PHFs were formed by incubation of tau con¬ struct K19/Y310W (50 μM) or K18/ΔK280/Y310W with 12.5 μM hepa¬ rin in volume of 100 μl at 37 ⁰C in PBS, pH 7.4. Incubation time was three days . The formation of aggregates was observed as a shift of the emission maximum from ~354 nm to ~340 nm.

Judging by the tryptophan assay the compounds were able to dissolve K19 filaments (Fig. 5A) with the exception of dauno- rubicin (Fig. 5A, entry #6) . All other compounds yielded emis¬ sion maxima of the protein after treatment around 350-353nm, close to the value of soluble tau, indicating a depolymerisa¬ tion efficiency of about 80-90%. In the case of K18 filaments (Fig. 5B) all compounds show a significantly lower efficiency of depolymerisation, only PHF005 is a strong inhibitor in these conditions (80%) , whereas emodin, adriamycin and PHFOlβ exhibit not more than 10% efficiency.

This ranking is consistent with the assembly inhibition assay (Fig. 3B) . In the case of K18/ΔK280 (Fig. 5C) the efficiency of disassembly is further reduced, but the ranking remains comparable to that of K18 (compare Fig. 5B) , as well as to the assembly inhibition assay (Fig. 3C) . In these cases, PHF005 remains the most potent agent for depolymerising PHFs (entry #9) .

The striking differences to the results obtained by thiofla- vine S assay can be explained by the different approaches of the assays. It is not known under what conditions ThS binds to PHFs, or how long the filaments have to be to become detect¬ able. In the case of the tryptophan assay the local environ¬ ment of every tryptophan is measured. It is therefore possible that in the ThS assay the long filaments are overrepresented, or that the tryptophan assay discriminates not between soluble and aggregated forms of tau, but only between more or less hy¬ drophobic environments.

EXAMPLE 7 Filter trapping assay

The effect of compounds on the depolymerisation of PHFs was analysed using a filter trapping assay. This assay monitors aggregated tau which is trapped on a membrane filter, whereas the soluble protein is washed through. Therefore the technique preferably detects larger filaments, similar to the ThS assay.

Aggregates of tau were trapped by filtration through a PVDF- membrane (pore diameter 0.45μm, Schleicher and Schuell, Dϋren, Germany) adapted to 9β-well dot blot apparatus. The PVDF- membrane was wetted with methanol and rinsed with PBS-buffer before incorporated into the dot blot apparatus . The samples were pipetted into lOOμl of PBS and filtered. The membrane was washed three times with PBS before taken out of the apparatus and blocked with 5% milk powder in PBS for 30 minutes in a ro¬ tational shaker at room temperature. The polyclonal antibody K9JA was used as primary antibody and incubated at a dilution of 1:20.000 at room temperature for one hour. A secondary anti-rabbit antibody conjugated with horseradish peroxidase (Dako, Hamburg, Germany) was diluted 1:2000 and incubated for 30 minutes at 37⁰C. After three times washing with TBS-Tween the signal was detected using the ECL system (Amersham Pharma¬ cia) and pictures were taken with the digital gel documenta¬ tion system Fuji film BAS3000 (Raytest, Straubenhardt, Ger¬ many) . Quantification of the signals was performed with the AIDA-software package (Raytest, Straubenhardt, Germany) .

Representative results are shown for htau23 (Fig. 5D) . The compounds shows similar depolymerising activities as with the ThS assay; emodin was most effective with a DC₅₀ of ~ 0.5 μM.

EXAMPLE 8 Depolymerisation of PHFs at prolonged incubation times

Depolymerisation data were typically obtained after 12 hours of incubation, but one is also interested in the effects of longer incubation times and lower compound concentrations which yielded only small effects after 12 hours. Fig. 7A and 7B show the time course of depolymerisation of Kl9 PHFs in the presence of 0.5 μM adriamycin or PHF005 during 28 days. Nearly no effects are seen after 12 hours, consistent with the other experiments (Fig. 3A) but interestingly the depolymerisation still continues and results in a final depolymerisation of -20-30% after 28 days. This result suggests that even low con¬ centrations of inhibitors can be used for depolymerisation us¬ ing prolonged incubation times .

EXAMPLE 8 Electron microscopy Protein solutions diluted to 0.1-10 μM were placed on 600-mesh carbon-coated copper grids for 1 min and negatively stained with 2 % uranyl acetate for 45 sec. The specimen was examined in a Philips CM12 electron microscope at 100 kV (Eindhoven, Netherlands) .

Fig. 6 shows the electron micrographs of hTau23-PHFs and hTau24-PHFs treated with different compounds for overnight.

EXAMPLE 10 Aggregation of Aβ fibres

Besides the activity of the compounds towards tau fibres, their specificity is an important issue, i.e. the ability to discriminate between different types of aggregates. Therefore, an analysis of the influence of the compounds (60 μM) on amy¬ loid fibrils made from the Aβl-40 peptide, both in terms of in¬ hibition of de novo filament formation and depolymerisation was performed (Fig. 8A-B) . These fibres are also abundant in Alzheimer brain and contain a core of cross-β-structure, but are located outside the cells, in contrast to the intracellu¬ lar PHFs.

Commercial human Aβl-40 was obtained from Calbiochem (Schwal- bach, Germany) and stored at -20 ⁰C. The Aβ peptide was rou¬ tinely dissolved in 100% DMSO to obtain a 2 mM stock solution, which was subsequently stored frozen at -20 ⁰C. 5 μl from the 2 mM Aβ stock solution was added to 90 μl of 25 mM phosphate buffer containing 120 mM NaCl, and 0,02% sodium azide, final pH 7,4 and 5 μl of 100% DMSO so that the final DMSO concentra¬ tion was 10% v/v, and the protein concentration was adjusted to 100 μM. Incubations were at room temperature. In order to accelerate aggregation tubes were put on a lab shaker and agi- tated at moderate speed. For analysis 5 μl of this solution were added to 45 μl 10 mM phosphate buffer containing 6 μM thioflavine T, pH 6,0, after 30 minutes incubation at room temperature the fluorescence was measured at 504 nm emission by an excitation of 409 nm. To correlate fibril morphology with the fluorescence signal, aliquots of the Aβl-40 solutions were simultaneously prepared for electron microscopy. The in¬ hibition of fibril formation and disassembly of pre-formed Aβ- fibrils were carried out in triplicates with 60 μM compound and 10 μM protein.

Most of the compounds show an inhibition of Aβ filament forma¬ tion of about 90% (Fig. 8A) and a depolymerisation activity of about 85%, except PHF005 which reached ~50% inhibition in the assembly and disassembly assay. Thus PHF005 appears to inter¬ fere more specifically with filaments made from tau, whereas the other compounds are promiscuous in terms of inhibiting β- sheet structures from different sources .

EXAMPLE 11

Light scattering for analysis of the influence of tau on mi¬ crotubule assembly

The repeat domain of tau is not only important for PHF aggre^¬ gation but also for the physiological function of microtubule binding. Microtubule polymerisation assays were performed in the absence and presence of compounds (Fig. 9) .

The ability of tau to promote microtubule assembly was moni¬ tored by light scattering at 350 nm in a Tecan spectrophotome¬ ter model Safire (Tecan, Crailsheim, Germany) . Tau protein (10 μM) was mixed with tubulin dimer (30 μM) and GTP (1 mM) at 4⁰C in polymerisation buffer (100 mM Na-PIPES pH 6.9, 1 mM EGTA, 1 mM MgSO₄, 1 mM DTT) with a final volume of 40 μl. Htau40 and inhibitor compounds (60 μM) were added last. After rapid mix¬ ing, the samples were pipetted into a Greiner transparent flat bottom 384 well plate (4 mm path length) , which was prewarmed at 37⁰C. The reaction was started by incubating the cooled components of the reaction at 37⁰C. The assembly of tubulin into microtubules was monitored over time by a change in tur¬ bidity. Three parameters were extracted from curves. The maxi¬ mum turbidity at steady state, the rate of assembly, and the lag time between the temperature jump and the start of the turbidity rise.

Tubulin (at 30 μM) without tau serves as a negative control which is unable to self-assemble into microtubules because it is below the critical concentration. However, in the presence of tau (10 μM) tubulin polymerises within 4 min. In the pres¬ ence of compounds (60 μM) the rate and extent of polymerisa¬ tion are not significantly affected, except for daunoribucin. The same is true for Congo Red, an Aβ fibre inhibitor (Podlisny et al . , 1998), used as a further control. These data suggest that the tested compounds influence specifically the patho¬ logical aggregation of tau protein, but not its interaction with microtubules .

EXAMPLE 12 Assays of tau aggregation in cells

A crucial test for the application of inhibitors is their ef¬ fect in cell models of tauopathy. A neuroblastoma (N2a) cell line which allows inducible expression of the tau construct K18ΔK280 under the control of the tet-on transactivator was generated. This construct was chosen because it contains the FTDP17 mutation ΔK280 in the 4-repeat domain K18 which pro- motes the formation of β structure and therefore aggregates readily, even in the absence of polyanionic inducers (von Ber¬ gen et al., 2001; {Barghorn, 2002 #2261}) .

The tau construct K18/ΔK280 was expressed in the mouse neuro¬ blastoma cell line N2a in an inducible manner under the con¬ trol of the reverse tetracycline-controlled transactivator (rtTA) as described elsewhere (Gossen & Bujard, 2002; Biernat et al., 2004) . The inducible N2a/K18ΔK280 cells were cultured in MEM medium supplemented with 10% fetal calf serum, 2 rtiM glutamine and 0.1% nonessential amino acids. The expression of K18/ΔK280 was induced by addition 1 μg doxycyclin per 1 ml me¬ dium. The effect of aggregation inhibition was observed by adding the inhibitor emodin (15 μM) . After 3-7 days the cells were harvested and tested for tau aggregation, thioflavin S fluorescence, and viability.

The levels and solubility of the K18/ΔK280 tau protein were determined by the method of Greenberg and Davies (1990) which makes use of the insolubility of protein aggregates after treatment with sarkosyl. The supernatant and sarcosyl- insoluble pellets were analysed by Western blotting with the pan-tau antibody K9JA and analysed by densitometry. Aggrega¬ tion of tau in cells was tested by the fluorescence of thio- flavine S. ThS signals were scored in three independent fields containing 40 cells each.

Fig. 1OA shows SDS blots of the cell extract after 7 days. The pellet of the untreated control (- emodin) shows the typical "smear" at higher molecular weight which is characteristic of aggregation in Alzheimer's disease as well (Fig. 1OA, lane 2) . However, emodin strongly suppresses the aggregates, leaving tau mostly in the soluble state (Fig. 1OA, lane 4) . Quantifi¬ cation of the sarkosyl-insoluble fraction shows a 5-fold re- auction by emodin, from 14% of the total cellular tau down to 3% (Fig. 10B) . Similar results were obtained by staining the cells with ThS (to show aggregated material) and with an anti¬ body against total tau (to show the level of tau expression) (Fig. 11) . The level of tau expression was comparable without or with emodin (compare Fig. 11 left, top and bottom) . How¬ ever, whereas the ThS signal is strong in the tau expressing cells, it becomes very weak in the presence of emodin, consis¬ tent with the absence of aggregates (Fig. 11 middle, top and bottom) . There are fewer ThS responsive cells, and fluores¬ cence intensity is much lower as well. The merged images il¬ lustrate that a large fraction of cells contain visible aggre¬ gates (green-yellow in superposition) , whereas the ThS signal is hardly visible in the emodin-treated cells (Fig. 11, right) . The quantification of the images is shown in Fig. 1OC.

EXAMPLE 13

Identification of compounds capable of inhibiting PHF forma¬ tion and capable of depolymerising pre-formed PHFs

In a first screen 200.000 compounds were analysed for their influence on PHF aggregation from the tau construct Kl9 using the thioflavine S assay as described in Example 2. 1266 com¬ pounds - corresponding to 0.6% of the library- were able to inhibit PHF aggregation to an extent higher than 90%. Out of these 1266 compounds, 77 were also able to disassemble PHFs with an efficiency of more than 80% (measured as described in Example 3), corresponding to 0.04% of the total library.

In the next step the 77 best compounds in the experimental PHF depolymerisation assay were used for an in silico search for potential PHF inhibitors. For this in silico search several criteria were set. The compounds were selected with regard to common three-dimensional properties (shape and binding abil- ity) and chemical stability. Compounds with a molecular weight higher than 500 were excluded as well as structures with highly polar and reactive groups - for example SH-groups, hal- ide and azo-structures . The number of freely rotateable bonds was minimised. A search of three million chemical structures yielded 300 compounds, of which 241 were further tested.

The compounds were obtained from different companies, tested for solubility in 100% DMSO and in aqueous buffers and ana¬ lysed with respect to absorbance and fluorescence. The fluo¬ rescence of 66 substances interfered with the ThS assay. Therefore a second screen with 175 compounds was performed by testing their capability to inhibit PHF assembly and for PHF disassembly by the ThS assay.

Figure 12A shows that the percentage of inhibitory compounds was similar in the first and in the second Thioflavine S screen, whereas the fraction of depolymerising substances was increased >40 fold in the second screen (Fig. 12B) . These two observations can be explained by the fact that the in silico search was performed on the basis of the substances that are capable of inhibiting assembly as well as inducing disassem¬ bly. The results are confirmed by the analysis of the distri¬ bution of efficiencies of inhibition and depolymerisation (Fig. 13A, B) . The results show that the efficiency of inhibi¬ tion is not altered by the selection of the compounds for the second screen but the average efficiency of depolymerisation is increased.

EXAMPLE 14 Generation of Tau transgenic mice

Doubly transgenic mice for the conditional expression of transgenic Tau constructs in the CNS are created by crossing the tTA transgenic mice (where the expression of tTA transac- tivator is driven by the CAMKII-α promoter, termed CamKIIα-tTA mice) and transgenic mice carrying the tau transgene (termed Tau-BiTetO mice) .

Generation of transgenic Tau-BiTetO mice:

For the generation of this type of transgenic mice it is nec¬ essary to construct plasmids carrying the bidirectional tetO responsive promoter followed by both a tau isoform (or mutant) in one direction and luciferase reporter sequences in the other (Baron at al. , 1995) . the pBI-5 plasmid-derivatives car¬ rying tau isoforms or mutants were constructed by inserting the tau cDNA sequence containing the CIaI site at 5' and Sail site at 3' terminus in the appropriate restriction sites available in the multiple cloning site of the pBI-5 vector. The pBI-5 plasmid (Fig. 15) was originally constructed in H.Bujard's laboratory (Baron et al. , 1995), but is now avail¬ able from Clontech under the name pBI-L. The bidirectional Tet vectors are used to simultaneously express two genes under the control of a single TRE (tetracycline-responsive element) con¬ sisting of seven direct repeats of a 42-bp sequence containing the tetO (tetracycline operator) followed downstream and up¬ stream by the minimal CMV promoter (PminCMv) • pBI-L can be used to indirectly monitor the expression of tau protein by following the activity of the reporter gene lucif¬ erase expressed at the same time downstream of TRE.

The sequences encoding the Tau isoforms or mutants htau40/ΔK280, htau40/ΔK280/2P, K18/ΔK280 and K18/ΔK280/2P were amplified by PCR from E.coli expresssion vectors pNG-2, (pNG- 2/ htau40/ΔK280, pNG-2/ htau40/ /2P, pNG-2/ K18/ΔK280, and pNG-2/ K18/ΔK280/2P ) and supplied with CIaI and Sail re¬ striction sites at the N- and C-terminus, respectively. ΔK280 means a deletion of amino acid lysine 280 in the tau protein sequence, with corresponding nucleotides 838-840 deleted from the Tau gene sequence. This Tau mutation was detected in a Dutch family afflicted with frontotemporal dementia, (FTDP-17, Rizzu et al. , 1999) . As shown previously (Barghorn et al. , 2000) , this mutant possesses a particularly high tendency to aggregate into PHFs . The abbreviation /2P stands for two iso- leucine to proline mutations at positions 277 and 308 of the Tau protein sequence (I277P, I308P) . These mutations inhibit the aggregation of Tau to PHFs because the prolines act as beta-sheet breakers in critical regions of the Tau molecule. The Tau construct Clal-Sall restriction fragments were intro¬ duced into CIaI and Sail digested pBI-L vector .

Before microinjection, the 1384 nucleotide long E.coli frag¬ ments of the pBI-5 vectors were removed by digestion with Xmnl and Drdl restriction enzymes and separated on agarose gels. The linearized plasmid fragments carrying the Tau genes were microinjected into single cell embryos.

The second tTA transgene mice line (CamKIIα-tTA mice) is al¬ ready available in the Lab. of Prof. H. Bujard. The tTA transgene is under the control of the calcium/calmodulin ki¬ nase Ilα (CAMKIIa) promoter (Mayford et al. , 1996) . This tTA line allows the restricted, conditional high expression of tTA transactivator in the CNS, particularly in the hippocampus and the cortex.

Generation of doubly transgenic progeny:

The Tau-BiTetO mice were crossed with CamKIIα-tTA mice to re¬ sult in doubly transgenic progeny constitutively expressing both transgenes , tau construct of interest and transactiva¬ tor tTA. This expression can be regulated by the presence of doxycycline, which turns off the tau gene transcription. EXAMPLE 15 Analysis of transgenic mice

Biochemical analysis :

The inducible transgenic mice KT1/K2.1 expressing a mutant htau40/ΔK280 protein exhibits neurofibrillary tangle pathology in the cortex and in the hippocampus. Fig.17 illustrates the biochemical analysis of neurofibrillary pathology and sarco- syl-insoluble tau in the cortex. Transgenic sarcosyl insoluble tau protein begins to accumulate in the cortex after 4 months of expression and its amount increases continuously till 8 months of age (Fig. 16 b) .

Histochemical analysis of brain sections:

The neurofibrillary pathology in the hippocampus of the in¬ ducible transgenic mice KT1/K2.1 is illustrated with immuno- histochemistry images following staining with conformational specific antibody MCl and Alzheimer specific phospho-KXGS-tau antibody 12E8, (Fig. 17) .

Conformational- and phospho-specific tau antibodies revealed an age - related progression between 5 to 8 month of trans¬ genic tau protein expression. Non of these antibodies bind to normal mice tau in control hippocampal sections (Fig. 17a) .

EXAMPLE 16

Generation of inducible mouse neuroblastoma (N2a) cell lines expressing Tau constructs

As a basis for a cell model the 4-repeat construct K18 con^¬ taining the FTDP-17 mutation ΔK280 was choosen because this has a high tendency for aggregation. Previous studies have shown that in vitro this construct K18/ΔK280 can assemble into PHFs even without the facilitation by polyanions (Barghorn et al., 2000) . As a control a variant K18/ΔK280/2P containing the two point mutations I277P and I308P was choosen because these mutations interrupt beta structure and therefore prevent the aggregation of tau. N2a cell lines expressing the tau con¬ structs K18/ΔK280 and K18/DelK280/2P were generated using the Tet-On expression system (Urlinger et al. , 2000) where protein synthesis is switched on by the addition of doxycyclin to the culture medium.

In the cell culture study, the aggregation of Tau was measured in the form of the aberrant, sarcosyl insoluble tau species which is pelletable after sarcosyl extraction (Greenberg & Da- vies, 1990) and can be analyzed by quantitative Western blot analysis. We have found a pronounced aggregation of K18ΔK280 protein which can be seen by comparing supernatants and pel¬ lets after sarcosyl extraction (Fig. 18) . The sarcosyl insolu¬ ble high-molecular-weight aggregates run as an immunoreactive smear in SDS gels. (Fig. 18, lane 3) .

EXAMPLE 17 Staining of Tau aggregates in cells by Thioflavin-S

To confirm by an independent method whether the inducible ex¬ pression of the Tau construct K18/ΔK280 in N2a cells induces aberrant aggregates indirect immunofluorescence experiments were carried out. Cells were stained with the fluorescent dye thioflavine-S (ThS) , followed by staining with the polyclonal antibody K9JA that recognizes all tau isoforms independently of phosphorylation. Thioflavin-S is known as a marker of in¬ soluble protein aggregates containing β-pleated sheets ("amy¬ loids") . In control cells without induction of K18/ΔK280 pro¬ tein, ThS-positive cells (unspecific binding ) were rare (-2%, Fig. 19) . After induction of K18/ΔK280 for 3 days ThS- positive aggregates of the Tau construct were formed in 28% of the cells.

EXAMPLE 18

Application of inducible "tau" cell line for testing of Tau aggregation inhibitors

The inducible N2a cell line expressing the Tau construct K18/ΔK280 can be used for testing the inhibition of tau aggre¬ gation by low molecular weight compounds. This is illustrated in Fig. 20 for the example of emodin. In the control case K18/ΔK280 was induced in N2a cells with doxycyclin, in the test case the induction was performed in the presence of 15 μM emodin. The analysis was done by two methods:

(a) Sarcosyl extraction of cells and analysis of soluble and aggregated Tau by quantitative Western blot analysis (densi¬ tometry) : Fig. 20a (lane 2) shows an example of the formation of sarcosyl insoluble high-molecular-weight aggregates of K18/ΔK280 in N2a cells not treated with emodin. They run as an immunoreactive "smear" in the SDS gel. The densitometric analysis of supernatant/pellet fractions demonstrates that 14% of the expressed K18/ΔK280 protein was found in the sarcosyl insoluble pellet (Fig. 20b) . By contrast, the super¬ natant/pellet analysis of cells treated with 15 μM emodin

(Fig. 20a, lanes 3, 4) shows that the immunoreactive smear of the pellet fraction in the SDS gel has disappeared, and sig¬ nificantly less material (3%) was found in the pellet fraction

(Fig. 20b) .

(b) Indirect immunofluorescence using ThS staining: ThS stain¬ ing of N2a cells transfected with K18/ΔK280 reveals the in¬ hibitory influence of emodin on the formation of aberrant tau aggregates . Two parallel cell cultures were incubated, one with 1 μg/ml doxycyclin (to induce the expression of the pro¬ tein) , another with 1 μg/ml doxycyclin and 15 μM emodin for 3 days. The quantitative analysis of N2a cells after induction of K18/ΔK280 for 3 days and staining with ThS revealed aggre¬ gates containing tau in 28 % of the cells (Fig. 20c) . By con¬ trast, treatment with doxycyclin and emodin resulted in only 15% cells with ThS signal (Fig. 20c) . This results indicate the inhibitory effect of emodin on tau aggregation in cell culture. An immunofluorescence image of double staining with Thioflavin-S and the tau antibody K9JA in Tet-On inducible N2a/K18/ΔK280 cells is shown in Fig.21.

EXAMPLE 19

Selection of N2a, Tet-On, G418-resistant cell line:

N2a cells were cotransfected with both the pUHD172-l (encoding the rtTA , origin: H.Bujard Lab.) and pEU-1 (encoding G418 re¬ sistance, a derivative of pRc/CMV, Invitrogen) Plasmid DNA (20 : 1; lμg/well of 6-well plates) using the DOTAP transfection reagent (Roche) . The cells were cultured in Eagle's Minimum Essential Medium (MEM) supplemented with 10% defined fetal bovine serum and subjected to G418 (600 μg/ml) and selection. The cells were fed with fresh media every 4 days for 3-4 weeks when single colonies appeared. Clones were tested for the induction level by transient transfection of pUHG 16-3 plasmid and induction of β-galactosidase was measured. The pBI-5 plasmid was also transiently transfected into these cells and the luciferase assay showed 23Ox induction.

Generation of inducible Tet-On, N2a/ Kl8/DelK280 cell line: The K18/ΔK280 DNA fragment was inserted into the bidirectional vector pBI-5 (pBI-5 is an unpublished derivative of pBI-2, Baron et al . , 1995) . The pBI-5/K18/ΔK280 plasmid with pX343 (a plasmid encoding the hygromycin resistance) were used for the cotransfection procedure of N2a/Tet-0n , G418-resistant cells with the aid of DOTAP (20 : 1; lμg/well of 6-well plates) . The cells were seeded at 4xlO⁵ cells per well. On the following day cells were transferred to 100-mm dishes and selected with 100 μg/ml of hygromycin and 600 μg/ml of G418. Clonal cell line were screened for inducible Kl8/DelK280 expression by measur¬ ing of luciferase activity with the luciferase assay and im¬ munofluorescence for tau protein with the Tau antibody K9JA.

Induction of K18/DelK280 expression in Tet-On N2a cells:

The inducible N2a/K18DelK280 cells were cultured in MEM medium supplemented with 10% fetal calf serum, 2 mM glutamine and 0.1% nonessential amino acids. The expression of K18/ΔK280 was induced by addition 1 μg doxycyclin per 1 ml medium. The in¬ duction was continued over 7 days and the medium was changed 3 times, always complemented with doxycyclin or with doxycyclin plus emodin.

Isolation of soluble and insoluble fractions of K18/ΔK280 protein from TetOn inducible N2a/K18/ΔK280 :

Tau aggregation assays:

For tau solubility assays the cells were collected by pellet¬ ing during centrifugation at 1000xg for 5 minutes. The levels and solubility of K18/ΔK280 tau protein were determined fol¬ lowing Greenberg and Davies (1990) . The cells were homogenized with Heidolph homogenizer DIAX900 in 10 vol (w/v) of buffer consisting of 10 mM Tris-HCl (pH 7.4), 0.8 M NaCl , 1 mM EGTA, and 10% sucrose. The homogenate was spun for 20 min at 20000xg, and the supernatant was retained. The pellet was re- homogenized in 5 vol of homogenization buffer and re- centrifuged. Both supernatants were combined, brought to 1% N- laurylsarcosinate (w/v) and incubated for 1 hr at room tem¬ perature while shaking and centrifuged at 100 OOOxg for 1 hr. The sarcosyl-insoluble pellets were resuspended in 50 mM Tris- HCl (pH 7.4), 0.5 ml per g of starting material. The super¬ natant and sarcosyl - insoluble pellet samples were analyzed by Western bloting. The amount of material loaded for super¬ natant and sarcosyl insoluble pellet represented 0.75% and 15% of total material present in the supernatant and pellet re¬ spectively (the ratio of supernatant and sarcosyl-insoluble pellet was always 1:20) . For quantification of the Tau level in each fraction, the Western blots were probed with antibody K9JA and analyzed by densitometry.

Quantitation of cells with induced aberrant K18/ΔK280 Tau ag¬ gregation using ThS staining:

Tet-On inducible N2a/K18/Δ280 cells were treated with 1 ug/ml doxycyclin for 3 days . After that the cover slips were fixed with 4% paraformaldehyde in PBS and incubated with the 0.01% ThS. Thereafter cells were washed three times in ethanol (70%) . In the next step the samples were blocked with 5% BSA and and treated with 0,1% Triton X-100. Finally the cells were incubated with rabbit polyclonal Tau antibody K9JA and secon¬ dary anti-rabbit antibody labeled with Cy5. Cells containing distinct ThS signals indicating the presence of insoluble ag¬ gregated material with β-pleated sheets were scored in three independent fields containing 40 cells each.

References

Ball, M.J., and Murdoch, G.H. (1997) . Neuropathological crite¬ ria for the diagnosis of Alzheimer's disease: are we really ready yet? Neurobiol Aging 18, S3-12. Barghorn, S., Zheng-Fischhofer, Q., Ackmann, M., Biernat, J., von Bergen, M., Mandelkow, E. M., and Mandelkow, E. (2000) . Structure, microtubule interactions, and paired helical fila^¬ ment aggregation by tau mutants of frontotemporal dementias . Biochemistry 39, 11714-11721.

Bates, G. (2003) . Huntingtin aggregation and toxicity in Huntington's disease. Lancet 361, 1642-1644.

Beirao, I., Lobato, L., Riedstra, S., Costa, P. M., Pimentel, S., and Guimaraes, S. (1999) . Ineffectiveness of dialysis in transthyretin (TTR) clearance in familial amyloid polyneuropa¬ thy type I, in spite of lower stability of the TTR Met30 vari¬ ant. Clin Nephrol 51, 45-49.

Bonifacio, M. J., Sakaki, Y., and Saraiva, M. J. (1996) . 'In vitro' amyloid fibril formation from transthyretin: the influ¬ ence of ions and the amyloidogenicity of TTR variants. Biochim Biophys Acta 1316, 35-42.

Braak, H., and Braak, E. (1991) . Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol (Berl) 82, 239- 259.

Brion, J. P., Anderton, B. H., Authelet, M., Dayanandan, R., Leroy, K., Lovestone, S., Octave, J. N., Pradier, L., Touchet, N., and Tremp, G. (2001) . Neurofibrillary tangles and tau phosphorylation. Biochem Soc Symp, 81-88.

Caughey, B., and Lansbury, P. T., Jr. (2003) . Protofibrils, Pores, Fibrils, and Neurodegeneration: Separating the Respon¬ sible Protein Aggregates from the Innocent Bystanders. Annu Rev Neurosci. Dou, F., Netzer, W. J., Tanemura, K., Li, F., Hartl, F. U., Takashima, A., Gouras, G. K., Greengard, P., and Xu, H. (2003) . Chaperones increase association of tau protein with microtubules. Proc Natl Acad Sci U S A 100, 721-726.

Duff, K., Knight, H., Refolo, L. M., Sanders, S., Yu, X., Pic- ciano, M., Malester, B., Hutton, M., Adamson, J., Goedert, M., et al. (2000) . Characterization of pathology in transgenic mice over-expressing human genomic and cDNA tau transgenes . Neurobiol Dis 7, 87-98.

Eidenmuller, J., Fath, T., Maas, T., Pool, M., Sontag, E., and Brandt, R. (2001) . Phosphorylation-mimicking glutamate clus¬ ters in the proline-rich region are sufficient to simulate the functional deficiencies of hyperphosphorylated tau protein. Biochem J 357, 759-767.

Friedhoff, P., Schneider, A., Mandelkow, E. M., and Mandelkow, E. (1998a) . Rapid assembly of Alzheimer-like paired helical filaments from microtubule-associated protein tau monitored by fluorescence in solution. Biochemistry 37, 10223-10230.

Friedhoff, P., von Bergen, M., Mandelkow, E. M., Davies, P., and Mandelkow, E. (1998b) . A nucleated assembly mechanism of Alzheimer paired helical filaments. Proc Natl Acad Sci U S A 95, 15712-15717.

Glabe, C. (2001) . Intracellular mechanisms of amyloid accumu¬ lation and pathogenesis in Alzheimer's disease. J MoI Neurosci 17, 137-145.

Goedert, M., Jakes, R., Spillantini, M. G., Hasegawa, M., Smith, M. J., and Crowther, R. A. (1996) . Assembly of microtu¬ bule-associated protein tau into Alzheimer-like filaments in¬ duced by sulphated glycosaminoglycans . Nature 383, 550-553. Goedert, M., Spillantini, M. G., and Davies, S. W. (1998) . Filamentous nerve cell inclusions in neurodegenerative dis¬ eases. Curr Opin Neurobiol 8, 619-632.

Goldberg, M. S., and Lansbury, P. T., Jr. (2000) . Is there a cause-and-effeet relationship between alpha-synuclein fibril- lization and Parkinson's disease? Nat Cell Biol 2, E115-119.

Gotz, J. (2001) . Tau and transgenic animal models. Brain Res Brain Res Rev 35, 266-286.

Hall, G. F., Chu, B., Lee, G., and Yao, J. (2000) . Human tau filaments induce microtubule and synapse loss in an in vivo model of neurofibrillary degenerative disease. J Cell Sci 113 ( Pt 8) , 1373-1387.

Heiser, V., Scherzinger, E., Boeddrich, A., Nordhoff, E., Lurz, R., Schugardt, N., Lehrach, H., and Wanker, E. E. (2000) . Inhibition of huntingtin fibrillogenesis by specific antibodies and small molecules: implications for Huntington' s disease therapy. Proc Natl Acad Sci U S A 97, 6739-6744.

Higuchi, M., Ishihara, T., Zhang, B., Hong, M., Andreadis, A., Trojanowski, J., and Lee, V. M. (2002a) . Transgenic mouse model of tauopathies with glial pathology and nervous system degeneration. Neuron 35, 433-446.

Higuchi, M., Lee, V. M., and Trojanowski, J. Q. (2002b) . Tau and axonopathy in neurodegenerative disorders . Neuromolecular Med 2, 131-150.

Hutton, M. (2001) . Missense and splice site mutations in tau associated with FTDP-17: multiple pathogenic mechanisms. Neu¬ rology 56, S21-25. Hutton, M., Lewis, J., Dickson, D., Yen, S. H., and McGowan, E. (2001) . Analysis of tauopathies with transgenic mice. Trends MoI Med 7, 467-470.

Iqbal, K. & Grundke-Iqubal, I. (1997) . Elevated levels of τ and ubiquitin in brain and cerebrospinal fluid in Alzheimer' s dis¬ ease. Int. Psychogeriatr. 9 (Suppl. 1), 289-296

Jackson, G. R., Wiedau-Pazos, M., Sang, T. K., Wagle, N., Brown, C. A., Massachi, S., and Geschwind, D. H. (2002) . Human wild-type tau interacts with wingless pathway components and produces neurofibrillary pathology in Drosophila. Neuron 34, 509-519.

Johnson, G. V., and Bailey, C. D. (2002) . Tau, where are we now? J Alzheimers Dis 4, 375-398.

Kraemer, B. C, Zhang, B., Leverenz, J. B., Thomas, J. H., Trojanowski, J. Q., and Schellenberg, G. D. (2003) . Neurode- generation and defective neurotransmission in a Caenorhabditis elegans model of tauopathy. Proc Natl Acad Sci U S A 200, 9980-9985.

Lambert, M. P., Viola, K. L., Chromy, B. A., Chang, L., Morgan, T. E., Yu, J., Venton, D. L., Krafft, G. A., Finch, C. E., and Klein, W. L. (2001) . Vaccination with soluble Abeta oligomers generates toxicity-neutralizing antibodies. J Neuro- chem 79, 595-605.

Lee, V. M., Goedert, M., and Trojanowski, J. Q. (2001) . Neuro¬ degenerative tauopathies. Annu Rev Neurosci 24, 1121-1159.

Lewis, J., McGowan, E., Rockwood, J., Melrose, H., Nacharaju, P., Van Slegtenhorst, M., Gwinn-Hardy, K., Paul Murphy, M., Baker, M., Yu, X., et al. (2000) . Neurofibrillary tangles, amyotrophy and progressive motor disturbance in mice express¬ ing mutant (P301L) tau protein. Nat Genet 25, 402-405.

Li, L., von Bergen, M., Mandelkow, E. M., and Mandelkow, E. (2002) . Structure, stability, and aggregation of paired heli¬ cal filaments from tau protein and FTDP-17 mutants probed by tryptophan scanning mutagenesis. J Biol Chem 277, 41390-41400.

Liu, F., Iqbal, K., Grundke-Iqbal, I., and Gong, C. X. (2002) . Involvement of aberrant glycosylation in phosphorylation of tau by cdk5 and GSK-3beta. FEBS Lett 530, 209-214.

Maccioni, R. B., Otth, C, Concha, II, and Munoz, J. P. (2001) . The protein kinase Cdk5. Structural aspects, roles in neurogenesis and involvement in Alzheimer's pathology. Eur J Biochem 268, 1518-1527.

Perez, M., Valpuesta, J. M., Medina, M., Montejo de Garcini, E., and Avila, J. (1996) . Polymerization of tau into filaments in the presence of heparin: the minimal sequence required for tau-tau interaction. J Neurochem 67, 1183-1190.

Rochet, J. C, and Lansbury, P. T., Jr. (2000) . Amyloid fi- brillogenesis : themes and variations. Curr Opin Struct Biol 10, 60-68.

Sanchez, I., Mahlke, C, and Yuan, J. (2003) . Pivotal role of oligomerization in expanded polyglutamine neurodegenerative disorders. Nature 421, 373-379.

Schenk, D., Barbour, R., Dunn, W., Gordon, G., Grajeda, H., Guido, T., Hu, K., Huang, J., Johnson-Wood, K., Khan, K., et al. (1999) . Immunization with amyloid-beta attenuates AIz- heimer-disease-like pathology in the PDAPP mouse. Nature 400, 173-177.

Schneider, A., Biernat, J., von Bergen, M., Mandelkow, E., and Mandelkow, E. M. (1999) . Phosphorylation that detaches tau protein from microtubules (Ser262, Ser214) also protects it against aggregation into Alzheimer paired helical filaments. Biochemistry 38, 3549-3558.

Selkoe, D. J. (1989) . The deposition of amyloid proteins in the aging mammalian brain: implications for Alzheimer's dis¬ ease. Ann Med 21, 73-76.

Sheleg, S.V., Korotkevich, E.A., Zhavrid, E.A., Muravskaya, G.V., Smeyanovich, A.F., Shanko, Y.G., Yurkshtovich, T.L., Bychkovsky, B. and Belyaev, S.A. (2002) . Local chemotherapy with cisplatin-depot for glioblastoma multiforme. J. Neuroon- col. 60(I)₁ 53-59

Shtilerman, M. D., Ding, T. T., and Lansbury, P. T., Jr. (2002) . Molecular crowding accelerates fibrillization of al- pha-synuclein: could an increase in the cytoplasmic protein concentration induce Parkinson's disease? Biochemistry 41, 3855-3860.

Schweers, 0., Mandelkow, E. M., Biernat, J., and Mandelkow, E. (1995) . Oxidation of cysteine-322 in the repeat domain of mi- crotubule-associated protein tau controls the in vitro assem¬ bly of paired helical filaments . Proc Natl Acad Sci U S A 92, 8463-8467.

van Swieten, J. C, Stevens, M., Rosso, S. M., Rizzu, P., Joosse, M., de Koning, I., Kamphorst, W., Ravid, R., Spil- lantini, M. G., Niermeijer, and Heutink, P. (1999) . Phenotypic variation in hereditary frontotemporal dementia with tau muta¬ tions. Ann Neurol 46, 617-626.

Voiles, M. J., and Lansbury, P. T., Jr. (2003) . Zeroing in on the pathogenic form of alpha-synuclein and its mechanism of neurotoxicity in Parkinson's disease. Biochemistry 42, 7871- 7878.

von Bergen, M., Barghorn, S., Li, L., Marx, A., Biernat, J., Mandelkow, E. M., and Mandelkow, E. (2001) . Mutations of tau protein in frontotemporal dementia promote aggregation of paired helical filaments by enhancing local beta-structure. J Biol Chem 276, 48165-48174.

Wille, H., Drewes, G., Biernat, J., Mandelkow, E. M., and Man¬ delkow, E. (1992) . Alzheimer-like paired helical filaments and antiparallel dimers formed from microtubule-associated protein tau in vitro. J Cell Biol 118, 573-584.

Wischik, C. M., Edwards, P. C, Lai, R. Y., Roth, M., and Har¬ rington, C. R. (1996) . Selective inhibition of Alzheimer dis¬ ease-like tau aggregation by phenothiazines . Proc Natl Acad Sci U S A 93, 11213-11218.

Wittmann, C. W., Wszolek, M. F., Shulman, J. M., Salvaterra, P. M., Lewis, J., Hutton, M., and Feany, M. B. (2001) . Tauopa- thy in Drosophila: neurodegeneration without neurofibrillary tangles. Science 293, 711-714.

Zhang, J., and Johnson, G. V. (2000) . Tau protein is hyper- phosphorylated in a site-specific manner in apoptotic neuronal PC12 cells. J Neurochem 75, 2346-2357. Baron U, Freundlieb S, Gossen M, Bujard H (1995) . Co- regulation of two gene activities by tetracycline via a bidi¬ rectional promoter. Nucleic Acids Res. (17) : 3605-3606.

Furth, P.A., St. Onge, L., Boger, H., Gruss, P., Gossen, M., Kistner, A., Bujard, H., Hennighausen, L. (1994) Temporal control of gene expression in transgenic mice by a tetracy- cline-responsive promoter. Proc Natl Acad Sci U S A, 91:9302- 9306.

Gossen M, Freundlieb S, Bender G, Muller G, Hillen W, Bujard H. (1995) . Transcriptional activation by tetracyclines in mam¬ malian cells. Science 268:1766-1769.

Greenberg SG, Davies P (1990) . A preparation of Alzheimer paired helical filaments that displays distinct tau proteins by polyacrylamide gel electrophoresis . Proc Natl Acad Sci U S A. 87, 5827-5831.

Mayford, M., Bach, M.E., Huang, Y.Y., Wang, L., Hawkins, R.D., Kandel, E.R. (1996) . Control of memory formation through regu¬ lated expression of a CaMKII transgene. Science, 274:1678- 1683.

Rizzu P, Van Swieten JC, Joosse M, Hasegawa M, Stevens M, Tib- ben A, Niermeijer MF, Hillebrand M, Ravid R, Oostra BA, Goedert M, van Duijn CM, Heutink P. (1999)

High prevalence of mutations in the microtubule-associated protein tau in a population study of frontotemporal dementia in the Netherlands . Am J Hum Genet. 64, 14-21.

Urlinger S, Baron U, Thellmann M, Hasan MT, Bujard H, Hillen W (2000) . Exploring the sequence space for tetracycline- dependent transcriptional activators: novel mutations yield expanded range and sensitivity. Proc Natl Acad Sci U S A. 97:7963-7968.

Brief Description of the Figures

Fig. 1: Structure of inhibitor compounds, tau isoforms and constructs .

A-E, Inhibitor compounds:

(A) Emodin (1, 3, 8-Trihydroxy-6-methyl-anthraquinone) ;

(B) PHF016 (1,2, 5, 8-Tetrahydroxy-anthraquinone) ;

(C) PHF005 (1-Phenyl-l- (2, 3, 4-trihydroxy-phenyl) -meth- anone) ;

(D) Daunorubicin (8-Acetyl-lO- (4-amino-5-hydroxy-β- methyl-tetrahydro-pyran-2-yloxy) -6, 8, 11-trihydroxy-l- methoxy-7, 8, 9,lO-tetrahydro-naphthacene-5,12-dione) ;

(E) Adriamycin (10- (4-Amino-5-hydroxy-β-methyl-tetra- hydro-pyran-2-yloxy) -6, 8, ll-trihydroxy-8- (2-hydroxy- ethanoyl) -l-methoxy-7, 8,9, 10-tetrahydro-naphthacene- 5, 12-dione;

(F-I) Tau isoforms and constructs:

(F) htau24, a four repeat isoform of tau lacking the two N-terminal inserts (numbering of the amino acids according to the longest isoform htau40) ;

(G) htau23, the fetal three repeat isoform lacking the two N-terminal repeats and the second repeat (exon 10);

(H) construct Kl8 comprising the four repeats in the microtubule binding domain;

(I) construct K19 containing three repeats. In H and I the hexapeptide motifs PHF6 (third repeat) and PHFβ* (second repeat) that promote the formation of β- structure are highlighted. The position of the point mutation Y310W in the third repeat is indicated. Fig. 2: Inhibition of PHF formation monitored by ThS- fluorescence.

(A) extent of aggregation of tau construct Kl9 (lOμM) plotted vs. inhibitor concentration (range IfM- 60μM) . The extent of aggregation was measured by the thioflavin S fluorescence assay and the degree of in¬ hibition was plotted as percentage of control. All measurements were performed in triplicates. Adriamycin

(open circles) , daunorubicin (filled squares) , emodin (open triangles), PHF016 (filled diamonds) and PHF005 (open diamonds) exhibit only small differences over the concentration range from 10 pM to O.lmM. The sym¬ bols are used consistently in Figures 2A-E. The fits were calculated as four parameter logistic curves, the IC₅₀ values are summarised in Table 2. Half-maximal inhibition occurs in the range of 1-7 μM.

(B) inhibition of aggregation of construct K18

(C) isoform htau23

(D) isoform htau24,

(E) construct K18/ΔK280.

Fig. 3: Inhibition of PHF aggregation monitored by tryptophan fluorescence assay.

(A-C) fluorescence emission maximum of the single tryp¬ tophan W310 inserted by site-directed mutagenesis into tau constructs K19 (Fig. 3A), K18 (Fig. 3B) and K18/ΔK280 (Fig. 3C) . Fully solvent-accessible Trp has an emission maximum at ~355 nm, a blue-shift to lower wavelengths is an indicator of PHF aggregation. Solu¬ ble tau constructs (10 μM) and tau or PHFs exposed to denaturing conditions (4M GuHCl) show the maximum of fully exposed Trp, aggregated PHFs show a maximum of 341 nm (typical of Trp buried in the interior) , and tau aggregated in the presence of inhibitors (60 μM) show intermediate values, depending on the degree of inhibition. Note that by this assay, all compounds are efficient inhibitors for the aggregation of the 3- repeat construct K19, but the 4-repeat construct K18 and its mutant K18/ΔK280 mutant are much less respon¬ sive to the inhibitors.

Fig. 4: Disassembly of pre-formed PHFs induced by inhibitor compounds and monitored by ThS fluorescence.

Tau constructs and isoforms K19, K18, hTau23, hTau24 (lOμM) were first aggregated into PHFs for 48 hours in the presence of 2.5μM heparin (except K18/ΔK280) and the polymers separated from the soluble tau by cen- trifugation of Ih at 100,00Og, redissolved and then exposed to the inhibitors overnight at 37⁰C at the in¬ dicated concentrations (range 0.001-200 μM) . The com¬ pounds are capable of disassembling PHFs with varying efficiencies (see Table 2) .

(A) construct K19, (B) K18, (C) isoform htau23, (D) isoform htau23, (E) K18/ΔK280 (no heparin) . All meas¬ urements were performed in triplicates . The symbols represent adriamycin (open circles) , daunorubicin (filled squares) , emodin (open triangles) , PHFOlβ (filled diamonds) and PHFO05 (open diamonds) .

Fig. 5: Disassembly of preformed PHFs measured by tryptophan fluorescence shift assay and filter assay. Experiments were performed with tau constructs con¬ taining the Y310W mutation as in Fig. 3.

(A) K19, (B) K18, (C) K18/ΔK280 (assembled without heparin) . Note that PHF aggregation is largely re¬ versible for Kl9 (except for daunorubicin) , but only partially for K18 and K18/ΔK280.

(D) Depolymerisation of PHFs from htau23 measured by filter assay. The bars show the fraction of polymer¬ ised material trapped on the PVDF membrane. Black bar = control, untreated PHFs. The groups of bars show disassembly by emodin, daunorubicin, adriamycin, PHFOlβ, PHF005 as a function of compound concentra¬ tion.

Fig. 6: Electron microscopy of inhibited and disassembled

PHFs.

Fig. 7: Time course of PHF disassembly at low inhibitor con¬ centrations .

PHFs were formed as above (see Fig. 4; 10 uM construct K19, 2.5μM heparin, overnight) and then exposed to 0.5μM adriamycin or PHF005. Note that in spite of the low inhibitor concentrations there is a gradual de¬ crease of PHFs . Untreated controls were measured in parallel and subtracted as background.

Fig. 8: Effect of PHF inhibitors on Aβ fibre aggregation and disassembly. Aβ peptide 1-40 (lOμM) was incubated with moderate shaking overnight at room temperature and incubated with various compounds (60μM) overnight.

(A) inhibition of fibre aggregation is most efficient in the case of emodin, daunorubicin, and PHF0016.

(B) disassembly of pre-formed fibrils.

Fig. 9: Effect of compounds on microtubule binding

30μM tubulin dimer was incubated in a microtiter plate at 37⁰C in the absence and presence of htau40 (10 μM) and 60μM compound. Absorbance was taken at 350nm and plotted versus time. The symbols refer to adriamycin (open circles) , daunorubicin (filled squares) , emodin (open triangles) , PHFOl6 (filled diamonds) and PHF005 (open diamonds. All curves (except tubulin only) show microtubule assembly within a few minutes.

Fig.10: Effect of the aggregation inhibitor emodin on tau ag¬ gregation in cells .

(A) Western blotting of fractionated lysates from in¬ ducible N2a cells expressing tau (K18/ΔK280) after sarkosyl extraction. Sarcosyl insoluble K18/ΔK280 tau was detected in these cells after 7 days of induction. The sarcosyl-soluble (S) and insoluble pellet frac¬ tions (P) were separated by high speed centrifugation. The pellets obtained from cells incubated without (-) and with 15 μM emodin (+) were resuspended in Tris- EDTA buffer in a volume equivalent to 5% of the ex¬ tracts . Note that the amount of material loaded for supernatant and pellet represents 1% and 20% of the total extracted material, respectively. (B) Histogram of sarcosyl insoluble tau (K18/ΔK280) from cells grown without emodin or with 15 μM emodin (see Fig.1OA, lanes 2, 4) .

(C) Histogram of number of N2a cells expressing K18/ΔK280 (after induction with doxycyclin) with dis¬ tinct thioflavine S signal in cell cultures induced without emodin (+Dox) or with 15 μM Emodin (+ Dox, + Emo) . Note that emodin inhibits the aggregation about 2-fold as measured by ThS.

Fig. 11: Tau expression and aggregation in N2a cells.

N2a cells were induced to express K18/ΔK280 and fixed after 3 days. They were sequentially double stained with Thioflavin-S (green) and the pan-tau antibody K9JA (red) .

Top row, without emodin, bottom row, with 15 μM emo¬ din. Left, immunofluorescence with tau antibody, mid¬ dle, ThS staining, right, merge. Note the reduced Th-S staining of cells in the presence of 15 μM emodin (middle, top and bottom) .

Fig. 12: Fractions of inhibiting and depolymerising compounds in the first and second screen.

(A) Fractions of compounds which exhibited an inhibi¬ tory effect >90% at 60 μM concentration.

(B) Fractions of depolymerising compounds with an ac¬ tivity >80%.

Fig. 13: Histograms of the activity of compounds in terms of inhibition and reversal of PHF formation (A) The distribution of compounds in percent is plot¬ ted against their efficiency to inhibit PHF assembly at a concentration of 60 μM. For both the first screen

(200.000 compounds, blue bars) and the second screen (175 compounds, red bars) a peak at 10-20% efficiency appears, i.e. a large number of compounds has a mild effect, but only few reach an efficiency close to 100%.

(B) Distribution of compounds plotted against their efficiency of depolymerising pre-formed PHFs. Note the difference between the first screen (blue bars) and the second screen (red bars) . The compounds from the first screen show a peak at 30-40% efficiency, whereas the compounds of the second screen exhibit a maximum at 60-70%, indicating that the average efficiency has been improved.

Fig. 14: tTA and rtTA tetracycline gene regulation system

tTA is a fusion protein composed of the repressor (tetR) of the TnIO Tc-resistance operon of Escherichia coli and a C-terminal portion of protein 16 of herpes simplex virus that functions as strong transcription activator. tTA binds in the absence of doxycyclin (but not in its presence) to an array of seven cognate ope¬ rator sequences (tetO) and activates transcription from a minimal human cytomegalovirus (hCMV) promoter, which itself is inactive.

Fig. 15: pBI-5 plasmid map

The pBI-5 plasmid was originally constructed in H.Bujard's laboratory (Baron et al. , 1995), but is now available from Clontech under the name pBI-L. The bi- directional Tet vectors are used to simultaneously ex¬ press two genes under the control of a single TRE (te- tracycline-responsive element) consisting of seven di¬ rect repeats of a 42-bp sequence containing the tetO (tetracycline operator) followed downstream and upstream by the minimal CMV promoter (PmincMv) ■ pBI-L can be used to indirectly monitor the expression of tau protein by following the activity of the reporter gene luciferase expressed at the same time downstream of TRE.

Fig. 16: Analysis of neurofibrillary pathology and sarcosyl-in- soluble tau in the cortex of the inducible transgenic mice KT1/K2.1

(A) The phosphorylation independent tau-antibody K9JA shows the expression of htau40/ΔK280 in the brains of transgenic mice after induction between 4 and 8 months .

(B) The phosphorylation independent tau-antibody K9JA shows the transgenic sarcosyl insoluble htau40/ΔK280 protein. Aggregation of the protein begins in cortex after 4 months of induction.

Fig. 17: Histochemical analysis of brain sections

Low magnification views of the hippocampus showing:

(A) a control mouse,

(B) a transgenic mouse expressing human tau40/ΔK280 in pyramidal neurons which are immunostained by the anti¬ body MCl which recognizes an Alzheimer like conforma¬ tion of tau

(C) human tau40/ΔK280 immunopositive pyramidal neurons following staining with phospho-tau antibody 12E8, which detects phosphorylated tau protein at the KXGS motifs in the repeats (Ser 262 and Ser 356) .

Fig. 18: Aggregation of K18/ΔK280 protein in N2a cells after 5 days of induction of K18/ΔK280 by doxycycline

Blots comparing supernatants (lanes 1, 2) and pellets (lane 3, 4) after sarcosyl extraction of tau. The ex¬ pression of K18/ΔK280 leads to the formation of sarco¬ syl insoluble high-molecular-weight aggregates which run as an immunoreactive smear in SDS gels (lane 3) . By contrast, only a small amount of the double proline mutant of K18/ΔK280/2P was found in the sarcosyl inso¬ luble pellet (lane 4) .

Fig. 19: Thioflavin-S positive N2a cells without and after in¬ duction of K18/ΔK280 with doxycylin

In control cells without induction of K18/ΔK280 prote¬ in, cells positive for Thioflavin-S (unspecific bin¬ ding) are rare (-2%) . After induction of K18/ΔK280 for 3 days ThS positive aggregates are formed in 28% of the cells.

Fig. 20: Analysis of Tau aggregation

(A) Western blotting of fractionated lysates obtained from inducible N2a/K18/ΔK280 cells after sarcosyl ex¬ traction. Sarcosyl-insoluble Tau was detected after 7 days of induction. The sarcosyl - soluble (S) and - insoluble pellet (P) fractions were separated by cen- trifugation at high speed. The pellets obtained from cells incubated without (-) and in the presence of 15 μM Emodin (+) were resuspended in TE buffer at a volu¬ me equivalent to 5% of the extracts . Note that the amount of material loaded for supernatant and pellet represented 1% and 20% of the total material extrac¬ ted, respectively.

(B) Histogram of the sarcosyl insoluble K18/ΔK280 pro¬ tein fraction obtained from cells grown without emodin (compare Fig. 2OA, lane 2) and in the presence of 15 μM emodin (compare Fig. 2OA, lane 4) .

(C) Histogram of the number of inducible N2a/K18/ΔK280 cells with distinct thioflavine S signal in cell cul¬ tures induced in the absence of emodin (+ Dox) and in¬ duced in the presence of 15 μM emodin (+ Dox, + Emo) .

Fig. 21: Immunofluorescence imaging of Tau aggregates in cells

Double staining with Thioflavin-S and Tau antibody K9JA in Tet-On inducible N2a/Kl8/ΔK280 cells. The cells were fixed 3 days post induction and sequential¬ ly double stained with Thioflavin-S (green) and tau antibody K9JA. The staining ThS intensities of cells induced with doxycyclin in the presence of 15 μM emo¬ din are distinctly lower than in cells induced without emodin (compare the quantitative analysis in Fig. 20C) .

Table 1

Nr. 45

Nr. 75 Nr. 76

Nr. 77

GruppeO_Mol071044_00

GruppeO_Mol071044_02

GruppeO_Mol071044_023

GruppeO_Mol071044_031

GruppeO_Mol071044_022

GruppeO_Mol071044_011

GruppeO_Mol071044_035

GruppeO_Mol071044

GruppeO_Mol071044

Gruppe4_Mo!081483_001

Gruppe4_Mol08148

Gruppe7_Mol148265_018

Gruppe3_Mol126819_016

Gruppe3_Mol126819_013

Gruppe3_Mol126819_004

Gruppe3_Mol126819_014

Gruppe3_Mol126819_001

Gruppe3_Mol126819_015

Gruppe7_Mol148265_018

Gruppe7_Mol148265_011

Gruppe7_Mol148265_0

Gruppe7_Mol148265_057

Compound B4C3

Claims

Claims

1. Use of a compound capable of inhibiting protein aggregate formation and capable of depolymerising protein aggre¬ gates for the preparation of a pharmaceutical composition for treating a neurodegenerative condition.

2. Use according to claim 1 wherein the compound has the general formula LSA

wherein

Rl and R2 are selected from H and

R3 is selected from H, OCH₃, and F;

R4 is selected from H and CH₃, or R3 and R4 are connected to form a condensed pyrrole ring;

R5, if any, is selected from H and OCH₃,. R6 may be H and

R7 may be H, or R6 and R7 may be connected to form a con¬ densed phenyl ring; R8 is selected from CH₂CH₂OH, CH₂Ph and C(O)OCH₂CH₃, and;

X¹ , X", X¹ ' ¹, and X" ^{• •} are selected from N and C.

3. Use according to claim 2 wherein the compound with the general formula LSA is selected from

4. Use according to claim 1 wherein the compound has the formula LSB

wherein R9 is selected from

RlO is selected from H and NO₂, and R11 is selected from an iV-morpholino group, N-pyrrolidino group and OCH₃.

5. Use according to claim 4 wherein the compound with the general formula LSB is selected from

6. Use according to claim 1 wherein the compound is selected from

7. Use according to any one of claims 1-6 wherein the pro¬ tein aggregate comprises PHFs consisting of tau protein.

8. Use according to any one of claims 1-6 wherein the protein aggregate comprises Aβ protein, prion protein, α- synuclein.

9. Use according to any one of claim 1-8, wherein the neuro¬ degenerative condition is Alzheimer disease.

10. Use according to any one of claims 1-8, wherein the neu¬ rodegenerative condition is selected from the group of Tauopathies consisting of CBD (Cortical Basal Disease) , PSP (Progressive Supra Nuclear Palsy) , Parkinsonism, FTDP- 17 (Fronto-Temporal Dementia with Parkinsonism linked to chromosome 17) , Familiar British Dementia, Prion Disease (Creutzfeld Jakob Disease) and Pick's Disease.

11. Use according to any one of claims 1-10, wherein the pharmaceutical composition is administered orally or par- enterally.

12. Use according to any one of claims 1-11, wherein the pharmaceutical composition is administered as part of a sustained release formulation or administered by depot implantation.

13. N2a cell line expressing a mutant of tau in a inducible fashion that polymerizes in the cell into aggregates, that can be visualized by thioflavine S.

14. N2a cell line according to claim 13, wherein the the tet- on system is used for regulation of expression.

15. N2a cell line according to claims 13 and 14, wherein the mutants of tau are constructs comprising the four microtu¬ bule binding repeats of tau and bear the deletion at K280 in one case and in the other the the isoleucines 277 and 308 are additionally mutated into prolines (I277P and I308P) .

16. N2a cell line according to claim 15, wherein the mutant of tau is the mutant of K18 bearing the deletion at K280 and the isoleucines 277 and 308 are additionally mutated into prolines (I277P and I308P) which can be used to analyze the neurotoxicity of tau.

17. Use of the N2a cell line according to claims 13 to 16 for testing conditions that are designed to attenuate or to inhibit the aggregation process.

18. Use according to claim 17, wherein the aggregation is the aggregation of the mutant of tau.

19. Use of claim 17 , wherein the conditions are a compound or a protein or an antibody or molecules of other classes, such as fatty acids, nucleotides, ribonucleic acids.

20. Use of the N2a cell line acccording to claims 13 to 19 for identifying agents suitable for treating Alzheimers dis^¬ ease and other neurodegenerative diseases such as tauopa- thies (parkinsonism, fronto temporal dementias, picks dis^¬ ease, ccrticobasal degeneration, prion disease)

21: Use according to claim 20, wherein the agent is are a com¬ pound or a protein or an antibody or molecules of other classes, such as fatty acids, nucleotides, ribonucleic ac¬ ids .

22. Transgenic non-human animal which expresses mutants of tau that polymerize in neurons into aggregates, that can be visualized by thioflavine S.

23. Transgenic non-human animal according to claim 22, wherein the animal is a transgenic mouse.

24. Animal according to claims 22 and 23, wherein the expres¬ sion of mutants of tau is inducible.

25. Animal according to claims 22 to 24, wherein the tet-off system is used for regulationof expression.

26. Animal according to claims 22 to 25, wherein the mutants of tau are

K18dK280, Kl8dK280 I277P I308P, htau40dK280, or htau40dK280 I277P I308P.

27. Use of the animal according to claims 22 to 26 for ana¬ lyzing the neurotoxicity of tau indepently from aggrega¬ tion.

28. Use of the animal according to claims 22 to 26 for testing conditions that are designed to attenuate or to inhibit the aggregation process within neurones.

29. Use according to claim 28, wherein the conditions are a compound or a protein or an antibody or molecules of other classes, such as fatty acids, nucleotides, ribonucleic ac¬ ids.

30. Use of the animal acccording to claims 22 to 29 for iden¬ tifying agents suitable for treating Alzheimers disease and other neurodegenerative diseases such as tauopathies

(parkinsonism, fronto temporal dementias, picks disease, ccrticobasal degeneration, prion disease) .

31. Use of the animal according to claims 22 to 30, for ob¬ taining primary hippocampal cultures for performing the uses analagously to claims 17 to 21.