N-Phenylamine Derivatives as Aggregation Inhibitors in Cell Models of

Current Alzheimer Research, 2007, 4, 397-402

397

N-Phenylamine Derivatives as Aggregation Inhibitors in Cell Models of Tauopathy M. Pickhardt, J. Biernat, I. Khlistunova, Y.-P. Wang, Z. Gazova, E.-M. Mandelkow and E. Mandelkow* Max-Planck-Unit for Structural Molecular Biology, c/o DESY, Notkestrasse 85, 22607 Hamburg, Germany Abstract: Cell models of tauopathy were generated in order to study mechanisms of neurodegeneration involving abnormal changes of tau. They are based on neuroblastoma cell lines (N2a) that inducibly express different forms of the repeat domain of tau (tauRD), e.g. the 4-repeat domain of tau with the wild-type sequence, the repeat domain with the
K280 mutation ("pro-aggregation mutant"), or the repeat domain with
K280 and two proline point mutations ("anti-aggregation mutant"). The data indicate that the aggregation of tauRD is toxic, and that aggregation and toxicity can be prevented by low molecular weight compounds, notably compounds based on the N-phenylamine core. Thus the cell models are suitable for developing aggregation inhibitor drugs.

Keywords: Alzheimer's disease, paired helical filaments, tau, drug screening. INTRODUCTION Several brain diseases are accompanied by the aggregation of the microtubule-associated protein tau ("tauopathies"), including Alzheimer's disease (AD) and frontotemporal dementias with parkinsonism (FTDP-17) [1, 2]. The generation of models for tau pathology has been difficult since the overproduction of tau by itself does not lead to neurofibrillary aggregation but to axonal transport defects that show up, for example, as motor neuron disease in mice if tau is expressed in the wrong cell types [3]. For generating neurofibrillary pathology it has been necessary to enhance the toxicity of tau by FTDP17 mutations or by combining tau mutations with enhanced A load [4-10]. But there is a debate on whether tau aggregation is toxic, and whether removal of aggregates is beneficial [11, 12]. We developed cell lines, based on neuroblastoma N2a cells that allow the inducible expression of the repeat domain of tau [13, 14]. Three variants are expressed, the wildtype sequence, a "pro-aggregation mutant" (
K280) which has a high tendency of aggregation, and an "anti-aggregation mutant" containing additional proline residues [15]. The cells illustrate that the aggregation of tauRD is toxic, and that removal of aggregates is beneficial. Fragmentation by a thrombin-like protease is a prelude to aggregation, whereas phosphorylation in the repeat domain bears no obvious relationship to aggregation. Inhibitor compounds reduce aggregation and toxicity, and therefore the cell model can be used in the search for drugs for tau pathology in AD. RESULTS Development of Lead Structures for Inhibition of Tau Aggregates Starting with a library of 200.000 compounds we carried out a primary screen for compounds capable to inhibit tau *Address correspondence to this author at the Max-Planck-Unit for Structural Molecular Biology, c/o DESY, Notkestrasse 85, 22607 Hamburg, Germany; Tel: +49-40-89982810 Fax: +49-40-89716822; E-mail: [email protected] 1567-2050/07 $50.00+.00

aggregation and to induce disassembly of tau aggregates (Fig. 1). The assay used in the first screen was based on the dye thioflavine S whose fluorescence is increased by binding to PHFs [16]. In the primary screen we first tested a concentration of 60
M of each substance with regard to its inhibitory effect on PHF formation. Using an automated pipetting system, 50 mM NH4Ac, 10
M protein (tau construct K19), 60
M compound and 5
M heparin were mixed in 50
l volume in the wells of a 384 well plate and incubated overnight at 37 °C. Thioflavine-S was then added to the reaction mixture to a final concentration of 20
M and the signal was measured fluorimetrically, with excitation wavelength of 440 nm and emission at 521 nm. Among the substances tested, 1266 (equivalent to 0.6% and belonging to several chemical families and a series of singletons) showed aggregation inhibition at an efficiency of 90% or higher. Of these, 77 compounds (= 0.04% of the library) were able to induce PHF disassembly with 80% efficiency. To verify the data we tested these compounds with several additional assays (Fig. 1). The screens and assays were initiated using the threerepeat tau construct K19 and then extended to the four-repeat constructs K18 and K18K280 as well as to the tau-isoforms htau23 (three repeats, no inserts) and htau24 (four repeats, no inserts) (Fig. 2). In order to rule out a potential influence of heparin on the efficiency of the compounds we also used the 4-repeat construct K18
K280 which carries one of the mutations observed in frontotemporal dementia and which is capable of aggregating into PHFs without a polyanionic cofactor. The efficiency of the compounds in vitro and their possible cytotoxicity in cell models were improved by chemical modification and retesting (Fig. 1). In this iterative process we are able to optimize the structure-activity-relationship as well as the cell viability of the compounds. As representative chemical groups we choose here compounds which belong to the chemical family of the N-phenylamines and anthraquinones (for examples see Fig. 3). They share a tricyclic aro©2007 Bentham Science Publishers Ltd.

398 Current Alzheimer Research, 2007, Vol. 4, No. 4

Pickhardt et al.

matic ring system with some modifications in the side groups.

Fig. (3). Chemical groups and lead structures: The naming core structure of N-phenylamines and anthraquinones are shaded. Two examples of modifications for each group are shown.

Fig. (1). Screening procedure: A compound library was screened for inhibition of PHF formation (first assay). The most active compounds belong to several chemical groups, including Nphenylamines, anthraquinones and a series of singletons. They were analyzed by various secondary assays, including assays for their toxicity and activity in cells.

The results on the anthraquinone and N-phenylamine series, their structures, IC50 and DC50 values were described in recent publications [18, 13]. The results were confirmed by electron microscopy. Fig. 4 illustrates examples of PHFs – made from the construct K19 – in the process of disassembly after overnight incubation with increasing inhibitor concentrations of B4D5. Filaments are seen in various stages of shortening or breaking reveal a concentration dependent degree of dissolution of the PHF structure.

Fig. (4). Depolymerisation of tau filaments can be monitored by electron microscopy: Electron micrographs of K19-PHFs treated with different compound concentrations of B4D5 for 12 hours, showing the breakdown of PHFs into smaller fragments.

Fig. (2). Tau-isoforms and constructs used in the assay: htau24, a four repeat isoform of tau lacking the two N-terminal inserts (numbering of the amino acids according to the longest isoform htau40, 441 residues); htau23, the fetal three repeat isoform lacking the two N-terminal repeats and the second repeat (exon 10); construct K18 comprising the four repeats in the microtubule binding domain; construct K19 containing three repeats. The hexapeptide motifs VQIINK (second repeat) and VQIVYK (third repeat) that promote the formation of -structure are highlighted. The deletion mutation
K280 ("pro-aggregation mutant") strongly accelerates PHF aggregation in vitro and in cells, the proline mutations I277P and I308P ("anti-aggregation mutant") strongly inhibit aggregation [17, 15].

INDUCIBLE CELL MODEL FOR TAU-INDUCED PATHOLOGY One of the important issues in the field is the generation of cell and animal models that reproduce the key features of tau pathology and allow one to test strategies for preventing it. The problem of tau aggregation has been particularly intriguing because tau is a highly soluble protein, it does not aggregate readily in physiological conditions, and therefore most attempts to achieve robust aggregation of wildtype tau in cells or animals have not been successful. This problem can be circumvented by expressing variants of tau with a higher amyloidogenic potential. Such cell models are of special value for the development of inhibitory compounds be-

N-Phenylamines as Tau Aggregation Inhibitors

cause they represent an intermediate level of complexity, closer to reality than the in vitro assays but still amenable to reasonably rapid testing. The cell model which we have developed is based on a cell line (mouse neuroblastoma N2a cell) in which tau can be inducibly expressed by the addition of doxycyclin (tet-ON system) for studying the aggregation of tau protein that is characteristic of Alzheimer disease (Fig. 5). The cell model allows one to study the toxicity of tau to cells in the soluble or aggregated state, the appearance of tau aggregates after switching on the tau gene expression, the dissolution of tau aggregates after switching the gene expression off again, and the efficiency of small molecules (identified in an in vitro screen) to prevent the formation of tau aggregates and dissolve them again [13]. The cells express the K280 mutant of the 4-repeat tau construct K18 because this has a high tendency for aggregation and was found to form PHFs in vitro even in the absence of polyanionic inducing cofactors [19]. To confirm whether the inducible expression of the K18K280 tau construct in N2a cells induces aberrant tau aggregates one can apply indirect immunofluorescence in combination with thioflavine-S staining (Fig. 5). This dye is known as a marker for amyloidlike aggregates with -pleated sheets. After induction of K18K280 for 5 days, thioflavin-S positive aggregates containing tau can be observed in about 14 % of the cells.

Fig. (5). Thioflavin S staining colocalizes with tau signal in inducible N2a cells: Expression of tau K18K280 tau in N2a cells induces aberrant tau aggregates as seen by indirect immunofluorescence. Cells were stained with thioflavine-S followed by staining with a polyclonal phosphorylation-independent pan- tau antibody (K9JA). Thioflavin-S is a marker for insoluble protein aggregates containing -sheet structure.

For an aggregation based cell model of tau pathology it is crucial to check whether the filaments generated in the cells resemble those in Alzheimer's disease. After induction of thioflavine S positive tau aggregates the cells are extracted with sarcosyl. The sarcosyl insoluble pellet is further purified by iodixanol density gradient centrifugation. This medium allows higher densities than sucrose and has been shown to be inert towards proteins [20]. We tested PHFs from recombinant tau protein and confirmed that filaments can be easily isolated (Fig. 6). Most of the protein is concentrated in fraction 4 corresponding to 40% of iodixanol. This fraction was used for electron microscopy. It contains fibrils which are labeled by immuno-gold with tau antibody K9JA

Current Alzheimer Research, 2007, Vol. 4, No. 4

399

conjugated to 8 nm gold particles (Fig. 6, lower panel), confirming their identity as bona fide PHF structures.

Fig. (6). Separation of sarcosyl insoluble Tau filaments by iodixanol gradient centrifugation: Left: The sarcosyl-insoluble PHFs were found in fraction 4 (40% iodixanol). Right: Electron micrographs of PHFs of fraction 4 immuno-gold labeled with an antibody against tau. C = control of recombinant K18
K280, M = protein marker. Bar = 100 nm.

Before testing the anti-aggregation effect of the Nphenylamine derivatives on tau expressing cells the toxicity of the compounds on control N2a cells was assayed. Neurotoxicity was determined by an LDH assay kit which reports on the leakiness of membranes in degenerating cells (Roche, Mannheim, Germany). After incubation of the cells for 24 h with 15
M compound, the degree of cell lysis can be determined by LDH release. Only compounds which show a low level of intrinsic cytotoxicity were considered to be suitable for determining the toxicity caused by the aggregation of tau. Examples are shown in Fig. 7 where we compare the toxicity of the anthraquinone compounds with those of the Nphenylamine series which show a much lower level of toxicity. For quantification the values are plotted as percentage of the control of 100 % cell lysis by Triton X-100. The value of the untreated DMSO-control was defined as background (0% cytotoxicity). The inducible N2a cell lines were based on the tau construct K18 which contains the 4-repeat domain of tau (Fig. 2). This domain forms the core of Alzheimer PHFs and can be polymerized into PHFs in vitro [21]. The mutation K18K280 has an enhanced tendency to form -structure and aggregation [19]. The mutation represents one of the tau mutations of FTDP-17 [22, 23]. In addition we generated a double proline mutant (PP – mutant) with prolines in each of the two hexapeptide motifs (I277P, I308P) that nucleate structure. Since prolines disrupt -strands they inhibit aggregation [17]. For establishing the Tet-On inducible N2a cells we generated the host N2a clone with a stably integrated reversed

400 Current Alzheimer Research, 2007, Vol. 4, No. 4

tetracycline transactivator rtTA-S2 under the control of the CMV promoter [24]. Inducible cell lines were then generated by stable transfection of vectors encoding the tauRD constructs. Doxycyclin was chosen at 1
g/ml for inducing tauRD expression. The yield after 2 days was ~1.2-1.8
g from 1x106 cells.

Pickhardt et al.

all cases the aggregates disappeared again, showing that the aggregation is reversible. Again the aggregation is most pronounced with the
K280 mutant [13]. We conclude that the aggregation of tauRD can be reversed by lowering the concentration, indicating that the protein subunits are exchangeable between the aggregated and soluble state. This is in contrast to the seemingly irreversible aggregation of tau in Alzheimer neurofibrillary tangles, which may be the result of secondary modifications. INHIBITION OF TAU AGGREGATION INDUCIBLE N2A CELLS BY N-PHENYLAMINES

Fig. (7). Determination of toxicity of compounds on N2a cells by the LDH assay: To analyze the cytotoxicity of the compounds an LDH-assay was performed. Note that the N-phenylamines are less cytotoxic than the anthraquinones.

For focussing on the aggregation process we searched for constructs which bind weakly to microtubules (to avoid transport defects) yet are able to aggregate; both conditions are met by the repeat construct K18. We also wanted to observe aggregation separately from the expression of tau, which can be achieved by staining with thioflavin S (ThS), biochemical analysis, and electron microscopy (Fig. 4) [18]. Among the tauRD constructs tested, only the
K280 mutant induced a strong reaction with ThS, consistent with its strong tendency for spontaneous aggregation in vitro. Aggregation was absent from cells in the case of the anti-aggregation PPmutant. In addition, filaments resembling those of AD PHFs were present in the cells (Fig. 6). The aggregates were also demonstrated by sarkosyl extraction [25]. During the first 10 days of K18wt expression the protein remains mostly soluble and intact. The insoluble fraction is minor up to day 11, when the sample reveals both lower cleavage products and higher aggregates. The phosphorylation of tau in the repeats (at KXGS motifs) was probed with antibody 12E8 (not shown). For all tauRD variants the phosphorylation remained mainly in the supernatant. This suggests that phosphorylation at KXGS motifs does not enhance the tendency to aggregate [26]. Fragmentation was also less apparent in the phosphorylated protein. It is an open question whether the toxicity of tau is due to tau aggregates. We therefore expressed different tau constructs for varying periods and observed cell degeneration by the LDH assay. The expression of soluble tauRD has no noticeable effect on the viability of the cells (data not shown). In the case of the pro-aggregation
K280 mutant the toxicity is ~2-fold higher than in the control. We conclude that the aggregation of tauRD causes toxicity. Next we asked whether tau aggregates can be removed again. The expression of tauRD was first induced for 5 days by doxycyclin until aggregates were clearly present, then doxycyclin was removed. In

IN

After having established a neuronal cell model which mimicks tau aggregation and toxicity we used the cell model for testing the most promising compounds. A variety of assay conditions for tau aggregation inhibitors were tested; the data shown here (Fig. 8A and B) were obtained by incubating the cells with 10
M compound, switching on the tau expression by doxycyclin, and scoring for ThS fluorescence after 5 days. To assess the depolymerization of aggregates, the expression of tau was induced for 5 days, then compounds were added, and the level of ThS fluorescence was measured after 2 more days.

Fig. (8AB). N-Phenylamines inhibit and disassemble Tau aggregates in inducible N2a cells. (A) The treatment of the tau expressing cells with the N-phenylamines leads to a decrease of tau aggregates as shown in the ThS staining but not to a reduction of the amount of total tau. (B) Quantification of the drug effects: The efficiency is around 60 to 70% for inhibition and around 40 to 45% for disassembly of tau aggregates in the cells.

Fig. 9 illustrates the general experience that in our cell model the level of toxicity is correlated with aggregation; conversely, when the degree of aggregation is lowered by drugs, the toxicity is lowered as well. We note here that the question of aggregation vs. toxicity is a prominent issue in the scientific debate, not only for the case of Alzheimer's disease but for other protein aggregation diseases as well. There appears to be no universal answer to this issue, however, at least in our cell model there is a clear relationship between the two parameters. DISCUSSION The aggregation and phosphorylation of tau are hallmarks of Alzheimer's disease and related tauopathies [1, 2]. We developed several cell models, based on the wellcharacterized neuronal cell line N2a, where the expression of tauRD can be switched on and off. The cell models were made in three variants, expressing the repeat domain of tau either in the wildtype sequence, or with pro-aggregation or

N-Phenylamines as Tau Aggregation Inhibitors

anti-aggregation mutations. The aggregation can be measured by ThS fluorescence, the aggregates can be visualized by electron microscopy, and incipient aggregates can be revealed by a sarkosyl-insoluble pellet and its accompanying smear at intermediate molecular weights which is characteristic for AD tau. These tools allowed us to test whether the aggregation of tauRD is toxic to cells, whether aggregation and toxicity can be reversed, and whether the aggregation can be inhibited by drugs.

Current Alzheimer Research, 2007, Vol. 4, No. 4

401

cell models are suitable for developing small-molecule inhibitors of tau aggregation and toxicity. One particular aim of this study was to test the ability of the N-phenylamine class of compounds (Fig. 3) to inhibit or to reverse the aggregation of tau in a cellular environment. These compounds had been identified in an in vitro screen for tau aggregation inhibitors. For example, compounds B4A1 and B4D5 have shown IC50 and DC50 values in the low
M range in vitro [13]. The results showed that these compounds are also able to inhibit the aggregation of tau within the cell models. In this regard, the N-phenylamine compounds are similar to the anthraquinone class of compounds [18]. However, the major advantage of the Nphenylamines is their lower toxicity to cells (about 37% lower than the anthraquinones). A caveat in these comparisons is that, in contrast to the in vitro assays, it is not known what fraction of the compounds enter into the cells, i.e. the effective cellular concentration remains to be determined. Given that tau is a cytosolic protein, the permeability of the neuronal plasma membrane will be a major determinant for the applicability of compounds as therapeutic agents in AD, in addition to the blood-brain barrier. This issue is currently under investigation. ACKNOWLEDGEMENTS

Fig. (9). Rescue from toxicity of aggregating K18
K280 in the inducible N2a cells in the presence of inhibitors: Bars illustrate LDH release after 2 days of expression of K18
K280 induced with doxycyclin in the absence or presence of small molecule inhibitors B1C11 and B4D3. The inhibitors reduce LDH release nearly to control levels.

Our results suggest that toxicity is related to aggregation, not merely to the expression of tauRD. Notably, the toxicity is already high before abundant aggregation. It is therefore possible that toxicity is caused by species smaller than PHFs, for example oligomers, or even by tau fragments that nucleate the aggregation [14]. This is reminiscent of the aggregation of A or other peptides where oligomers, rather than polymers, are considered the main toxic species [27-29]. Since small aggregates of tauRD can be toxic it would be important to find methods to counteract aggregation. The cell's own clearing capacity can be tested by silencing the expression of tauRD, and indeed soluble tauRD disappears again, and aggregated tauRD disappears as well. Thus the cell possesses mechanisms to break down the aggregates. Some proposed catabolic pathways include the ubiquitin-proteasome system [30, 31], calpain [32, 33] or autophagy [34]. As shown elsewhere [35], aggregates of tauRD in vitro are remarkably labile and can therefore be disintegrated into subunits which could be digested by various cellular proteases. This explains why toxic aggregates can be removed by the cell by lowering the production of tauRD. The consequence is that it should be possible to enhance the cell's self-cleaning capabilities by inhibitory drugs. Such compounds can be identified by in vitro screening [36, 18, 37, 38], and some of them have been shown to suppress aggregates in cells (Fig. 8,9). Importantly these cells also recovered from the toxicity of tau. Thus, the

We thank Dagmar Drexler and Olga Petrova for excellent technical assistance. We are indebted to H. Bujard and K. Schönig (Univ. Heidelberg) for reagents of the Tet-On system. This work was supported by grants from the Deutsche Forschungsgemeinschaft. ABBREVIATIONS AD

= Alzheimer's disease

FTDP-17

= Frontotemporal dementia with parkinsonism linked to chromosome 17

LDH assay = Lactate dehydrogenase assay for cell toxicity MT

= Microtubule

PHF

= Paired helical filament

TauRD

= Tau repeat domain

REFERENCES [1] [2]

[3]

[4]

[5] [6]

Lee VM, Goedert M, Trojanowski JQ. Neurodegenerative tauopathies. Annu Rev Neurosci 24:1121-1159 (2001). Mandelkow E, von Bergen M, Biernat J, Mandelkow E-M. Structural Principles of Tau and Alzheimer Paired Helical Filaments. Brain Pathol 17:83-90 (2007). Terwel D, Lasrado R, Snauwaert J, Vandeweert E, Van Haesendonck C, Borghgraef P, et al. Changed conformation of mutant Tau-P301L underlies the moribund tauopathy, absent in progressive, nonlethal axonopathy of Tau-4R/2N transgenic mice. J Biol Chem 280:3963-3973 (2005). Lewis J, Dickson DW, Lin WL, Chisholm L, Corral A, Jones G, et al. Enhanced neurofibrillary degeneration in transgenic mice expressing mutant tau and APP. Science 293:1487-1491 (2001). Gotz J, Chen F, van Dorpe J, Nitsch RM. Formation of neurofibrillary tangles in P301l tau transgenic mice induced by Abeta 42 fibrils. Science 293:1491-1495 (2001). Oddo S, Caccamo A, Shepherd JD, Murphy MP, Golde TE, Kayed R, et al. Triple-transgenic model of Alzheimer's disease with plaques and tangles: intracellular Abeta and synaptic dysfunction. Neuron 39:409-421 (2003).

402 Current Alzheimer Research, 2007, Vol. 4, No. 4 [7]

[8] [9]

[10] [11] [12] [13]

[14]

[15]

[16]

[17]

[18]

[19]

[20] [21]

[22]

Vogelsberg-Ragaglia V, Bruce J, Richter-Landsberg C, Zhang B, Hong M, Trojanowski JQ, et al. Distinct FTDP-17 missense mutations in tau produce tau aggregates and other pathological phenotypes in transfected CHO cells. Mol Biol Cell 11:4093-4104 (2000). DeTure M, Ko LW, Easson C, Yen SH. Tau assembly in inducible transfectants expressing wild-type or FTDP-17 tau. Am J Pathol 161:1711-1722 (2002). Ferrari A, Hoerndli F, Baechi T, Nitsch RM, Gotz J. beta-Amyloid induces paired helical filament-like tau filaments in tissue culture. J Biol Chem 278:40162-40168 (2003). Santacruz K, Lewis J, Spires T, Paulson J, Kotilinek L, Ingelsson M, et al. Tau suppression in a neurodegenerative mouse model improves memory function. Science 309:476-481 (2005). LaFerla FM, Oddo S. Alzheimer's disease: Abeta, tau and synaptic dysfunction. Trends Mol Med 11:170-176 (2005). Ashe KH. Mechanisms of memory loss in Abeta and tau mouse models. Biochem Soc Trans 33:591-594 (2005). Khlistunova I, Biernat J, Wang Y, Pickhardt M, von Bergen M, Gazova Z, et al. Inducible expression of Tau repeat domain in cell models of tauopathy: aggregation is toxic to cells but can be reversed by inhibitor drugs. J Biol Chem 281:1205-1214 (2006). Wang YP, Biernat J, Pickhardt M, Mandelkow E, Mandelkow EM. Stepwise proteolysis liberates tau fragments that nucleate the Alzheimer-like aggregation of full-length tau in a neuronal cell model. Proc Natl Acad Sci USA 104: 10252-10257 (2007). von Bergen M, Barghorn S, Li L, Marx A, Biernat J, Mandelkow EM, et al. Mutations of tau protein in frontotemporal dementia promote aggregation of paired helical filaments by enhancing local beta-structure. J Biol Chem 276:48165-48174 (2001). Friedhoff P, Schneider A, Mandelkow EM, Mandelkow E. Rapid assembly of Alzheimer-like paired helical filaments from mictotubule-associated protein tau monitored by fluorescence in solution. Biochemistry 37: 10223-10230 (1998). von Bergen M, Friedhoff P, Biernat J, Heberle J, Mandelkow EM, Mandelkow E. Assembly of tau protein into Alzheimer paired helical filaments depends on a local sequence motif ((306)VQI VYK(311)) forming beta structure Proc Natl Acad Sci USA 97:5129-5133 (2000). Pickhardt M, Gazova Z, von Bergen M, Khlistunova I, Wang Y, Hascher A, et al. Anthraquinones inhibit tau aggregation and dissolve Alzheimer's paired helical filaments in vitro and in cells. J Biol Chem 280:3628-3635 (2005). Barghorn S, Zheng-Fischhofer Q, Ackmann M, Biernat J, von Bergen M, Mandelkow EM, et al. Structure, microtubule interactions, and paired helical filament aggregation by tau mutants of frontotemporal dementias. Biochemistry 39: 11714-11721 (2000). Graham JM. OptiPrep density gradient solutions for macromolecules and macromolecular complexes. ScientificWorldJournal 2: 1547-1550 (2002). Wille H, Drewes G, Biernat J, Mandelkow EM, Mandelkow E. Alzheimer-like paired helical filaments and antiparallel dimers formed from microtubule-associated protein tau in vitro. J Cell Biol 118:573-584 (1992). Rosso SM, van Swieten JC. New developments in frontotemporal dementia and parkinsonism linked to chromosome 17. Curr Opin Neurol 5:423-428 (2002).

Received: April 30, 2006

Accepted: October 27, 2006

Pickhardt et al. [23] [24]

[25]

[26]

[27]

[28]

[29]

[30] [31] [32]

[33]

[34]

[35]

[36] [37]

[38]

D'Souza I, Schellenberg GD. Regulation of tau isoform expression and dementia. Biochim Biophys Acta 1739:104-115 (2005). Urlinger S, Baron U, Thellmann M, Hasan MT, Bujard H, Hillen W. Exploring the sequence space for tetracycline-dependent transcriptional activators: novel mutations yield expanded range and sensitivity. Proc Natl Acad Sci USA 97:7963-7968 (2000). Greenberg SG, Davies P. Preparation of Alzheimer paired helical filaments that displays distinct tau proteins by polyacrylamide gel electrophoresis. Proc Natl Acad Sci USA 87:5827-5831 (1990). Schneider A, Biernat J, von Bergen M, Mandelkow E, Mandelkow EM. Phosphorylation that detaches tau protein from microtubules (Ser262, Ser214) also protects it against aggregation into Alzheimer paired helical filaments. Biochemistry 38:3549-3558 (1999). Lambert MP, Barlow AK, Chromy BA, Edwards C, Freed R, Liosatos M, et al. Diffusible, nonfibrillar ligands derived from Abeta142 are potent central nervous system neurotoxins. Proc Natl Acad Sci USA 95:6448-6453 (1998). Walsh DM, Klyubin I, Fadeeva JV, Cullen WK, Anwyl R, Wolfe MS, et al. Naturally secreted oligomers of amyloid beta protein potently inhibit hippocampal long-term potentiation in vivo. Nature 416:535-539 (2002). Bucciantini M, Giannoni E, Chiti F, Baroni F, Formigli L, Zurdo J, et al. Inherent toxicity of aggregates implies a common mechanism for protein misfolding diseases. Nature 416:507-511 (2002). David DC, Layfield R, Serpell L, Narain Y, Goedert M, Spillantini MG. Proteasomal degradation of tau protein. J Neurochem 83:176185 (2002). Kosik KS, Shimura H. Phosphorylated tau and the neurodegenerative foldopathies. Biochim Biophys Acta 1739:298-310 (2005). Delobel P, Leroy O, Hamdane M, Sambo A, Delacourte A, Buee L. Proteasome inhibition and Tau proteolysis: an unexpected regulation. FEBS Lett 579:1-5 (2005). Brown MR, Bondada V, Keller JN, Thorpe J, Geddes JW. Proteasome or calpain inhibition does not alter cellular tau levels in neuroblastoma cells or primary neurons. J Alzheimers Dis 7:15-24 (2005). Nixon RA, Wegiel J, Kumar A, Yu WH, Peterhoff C, Cataldo A, et al. Extensive involvement of autophagy in Alzheimer disease: an immuno-electron microscopy study. Neuropathol Exp Neurol 64:113-122 (2005). Li L, von Bergen M, Mandelkow EM, Mandelkow E. Structure, stability, and aggregation of paired helical filaments from tau protein and FTDP-17 mutants probed by tryptophan scanning mutagenesis. J Biol Chem 277:41390-41400 (2002). Chirita C, Necula M, Kuret J. Ligand-dependent inhibition and reversal of tau filament formation. Biochemistry 43:2879-2887 (2004). Taniguchi S, Suzuki N, Masuda M, Hisanaga S, Iwatsubo T, Goedert M, et al. Inhibition of heparin-induced tau filament formation by phenothiazines, polyphenols, and porphyrins. J Biol Chem 280:7614-7623 (2005). Pickhardt M, Larbig G, Khlistunova I, Coksezen A, Meyer B, Mandelkow EM, et al. Phenylthiazolyl-hydrazide and its derivatives as inhibitors of tau aggregation in vitro and in cells. Biochemistry, in press (2007).