Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M507753200 on October 24, 2005

J. Biol. Chem., Vol. 281, Issue 2, 1205-1214, January 13, 2006
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
281/2/1205    most recent
M507753200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Khlistunova, I.
Right arrow Articles by Mandelkow, E.-M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Khlistunova, I.
Right arrow Articles by Mandelkow, E.-M.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Inducible Expression of Tau Repeat Domain in Cell Models of Tauopathy

AGGREGATION IS TOXIC TO CELLS BUT CAN BE REVERSED BY INHIBITOR DRUGS*

Inna Khlistunova1, Jacek Biernat1, Yipeng Wang, Marcus Pickhardt, Martin von Bergen, Zuzana Gazova, Eckhard Mandelkow, and Eva-Maria Mandelkow2

From the Max-Planck-Unit for Structural Molecular Biology, DESY, Notkestrasse 85, 22607 Hamburg, Germany

Received for publication, July 18, 2005 , and in revised form, September 22, 2005.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
We generated several cell models of tauopathy in order to study the mechanisms of neurodegeneration in diseases involving abnormal changes of tau protein. N2a neuroblastoma cell lines were created that inducibly express different variants of the repeat domain of tau (tauRD) when exposed to doxycycline (Tet-On system). The following three constructs were chosen: (i) the repeat domain of tau that coincides with the core of Alzheimer paired helical filaments; (ii) the repeat domain with the deletion mutation {Delta}K280 known from frontotemporal dementia and highly prone to spontaneous aggregation; and (iii) the repeat domain with {Delta}K280 and two proline point mutations that inhibit aggregation. The comparison of wild-type, pro-aggregation, and anti-aggregation mutants shows the following. (a) Aggregation of tauRD is toxic to cells. (b) The degree of aggregation and toxicity depends on the propensity for beta-structure. (c) Soluble mutants of tauRD that cannot aggregate are not toxic. (d) Aggregation is preceded by fragmentation. (e) Fragmentation of tauRD in cells is initially due to a thrombin-like protease activity. (f) Phosphorylation of tauRD (at KXGS motifs) precedes aggregation but is not correlated with the degree of aggregation. (g) Aggregates of tauRD disappear when the expression is silenced, showing that aggregation is reversible. (h) Aggregation can be prevented by drugs and even pre-formed aggregates can be dissolved again by drugs. Thus, the cell models open up new insights into the relationship between the structure, expression, phosphorylation, aggregation, and toxicity of tauRD that can be used to test current hypotheses on tauopathy and to develop drugs that prevent the aggregation and degeneration of cells.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
A number of human diseases are characterized by an age-related increase of abnormal protein deposits. This includes a group of diseases, termed "tauopathies," where the microtubule-associated protein tau aggregates in a highly phosphorylated form. The best known examples are Alzheimer disease (AD),3 where tau deposits in the brain occur together with aggregates of the Abeta peptide, and FTDP-17, where mutant forms of tau accumulate abnormally (1, 2). These changes lead to the loss of synapses, dying out of cell processes, and cell death. Hence, there is a strong interest in studying the pathways of abnormal aggregation and the development of cell and animal models. In the case of the amyloid pathology in AD, such models have been generated based on the overproduction of the Abeta peptide, and these models have led to drugs and treatments capable of reducing the plaque load (3). Generating reliable models for tau pathology has been more difficult. Overproduction of tau as such 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 (4, 5). To generate neurofibrillary pathology, it has been necessary to enhance the toxicity of tau by mutations known from familial FTDP-17 or by combining tau mutations with enhanced Abeta load (6-12). These models allowed insights into the early steps of tangle formation, such as the relationship between tau conformation, phosphorylation, and aggregation. However, it is still ambiguous whether tau aggregation is toxic, what the mechanisms of toxicity are, and whether removal of aggregates is beneficial. Furthermore, efficient screening for tau aggregation inhibitors is not possible with mice or neuronal cell cultures derived from them. These questions can be addressed by inducible cell models where the tau expression can be switched on and off at defined time points. The creation of such cell models allows one to analyze the consequences of tau expression and aggregation and to search for the causes of toxicity and for suitable drugs.

Here we describe N2a cell lines that allow the inducible expression of the repeat domain of tau, based on the Tet-On system (13). Three variants of tauRD are expressed, the wild-type sequence, a pro-aggregation mutant ({Delta}K280), and an anti-aggregation mutant (containing additional proline residues). The cells show robust aggregation of tau, including Alzheimer-like paired helical filaments. 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. Exposing the cells to inhibitor compounds reduces aggregation and toxicity, and even cells with aggregates are rescued again. Thus, the cell model will be useful in the search for drugs for AD and other tauopathies.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Selection of an N2a, Tet-On, G418-resistant Cell Line—The N2a cells used here were characterized previously (14). The cells were cotransfected both with pUHD172-1 plasmid (encoding the rtTA, obtained from H. Bujard, Heidelberg, Germany) and pEU-1 plasmid (encoding G418 resistance, a derivative of pRc/CMV; Invitrogen) (ratio 20:1; 1 µg/well of 6-well plates) using the N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium salts (DOTAP) transfection reagent (Roche Applied Science). The cells were cultured in Eagle's minimum essential medium 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 the pUHG 16-3 plasmid, and the induction of beta-galactosidase was measured. The pBI-5 plasmid was also transiently transfected into these cells, and the luciferase assay showed 230-fold induction.

Generation of Inducible Tet-On Cell Lines—The DNA fragments encoding the appropriate constructs of the tau repeat domain (K18WT, K18{Delta}K280, and K18{Delta}K280/PP) were inserted into the bidirectional vector pBI-5 between ClaI and SalI restriction sites (pBI-5 is an unpublished derivative of pBI-2) (15). The pBI-5/K18-derived plasmids with pX343 (plasmid encoding the hygromycin resistance) were used for cotransfection of N2a/Tet-On, G418-resistant cells with the aid of N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium salts (DOTAP) (20:1 plasmid ratio; 1 µg/well of 6-well plates). The cells were seeded at 4 x 105 cells per well. The following day cells were transferred to 100 mm dishes and selected with 100 µg/ml hygromycin and 600 µg/ml G418. Clonal cell lines were screened for the inducible expression of K18 derivatives by measuring luciferase activity with the luciferase assay and immunofluorescence for tauRD protein with the polyclonal pan-tau antibody K9JA.

Induction of K18 Derivatives in Tet-On N2a Cells—The inducible N2a/K18 cells were cultured in Eagle's minimum essential medium supplemented with 10% fetal calf serum, 2 mM glutamine, and 0.1% nonessential amino acids. The expression of K18 constructs was induced by adding 1 µg of doxycycline per 1 ml of medium. The induction was continued for 7-11 days, and the medium was changed three times (always complemented with doxycycline). For solubility assays, the cells were collected by pelleting during centrifugation at 1000 x g for 5 min. The levels and solubility of K18{Delta}K280 were determined following Ref. 16. The cells were homogenized (DIAX900, Heidolph, Schwabach, Germany) in 10 volumes (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 20,000 x g, and the supernatant was retained. The pellet was re-homogenized in 5 volumes of homogenization buffer and re-centrifuged. Both supernatants were combined, brought to 1% N-laurylsarcosinate (w/v), and incubated for 1 h at room temperature while shaken, followed by centrifugation at 100,000 x g for 1 h. The Sarkosyl-insoluble pellets were resuspended in 50 mM Tris-HCl (pH 7.4), 0.5 ml/1 g of starting material. The supernatant and Sarkosyl-insoluble pellet samples were analyzed by Western blotting. The amount of material loaded for supernatant and Sarkosyl-insoluble pellet represented about 0.5 and 15% of the total material present in the supernatant and pellet, respectively (the ratio between supernatant and Sarkosyl-insoluble pellet was always 1:30). For quantification of tau levels in each fraction, the Western blots were probed with pan-tau antibody K9JA and analyzed by densitometry (LAS 3000 and AIDA software, Raytest, Straubenhardt, Germany).

Quantitation of Aggregation of tauRD in Cells by Thioflavine S Staining—Tet-On inducible and undifferentiated N2a cells were treated with 1 µg/ml doxycycline for 5, 7, or 9 days in a culture dish. Then the cells were trypsinized and transferred to coverslips and incubated over-night. The cells on the coverslips were fixed with 4% paraformaldehyde in PBS for 15 min, then permeabilized with 80% MeOH for 6 min at -20 °C, incubated with 0.1% thioflavine S for 5 min, and washed three times in ethanol (50%). The samples were incubated with antibody K9JA in 5% goat serum (PBS). The secondary anti-rabbit antibody labeled with Cy5 was also diluted with 5% goat serum in PBS and incubated for 45 min. The cells were washed twice with PBS, once with water, and mounted. Cells containing distinct ThS signals indicating the presence of insoluble aggregated material with beta-pleated sheets were scored in many independent fields containing a total of 500 cells.

Density Gradient Centrifugation—A discontinuous density gradient of the nonionic medium iodixanol was created by layering different concentrations of iodixanol in a centrifugation tube (from bottom to top: 850 µl of 50, 40, and 30%; 2.2 ml of 20%; 850 µl of 10%; and 300 µl of 5%). The samples were applied in a volume of 50 µl, and centrifugation was performed in a Sorvall TV865 rotor (Kendro Laboratory Products, Langenselbold, Germany) in an ultracentrifuge (Beckman Coulter Optima 80K) for 3 h at 350,000 x g at 4 °C. Samples were released from the bottom of the tube by punctating the top and the bottom with a syringe, and droplets were collected. The numbering of fractions ranged from high to low density.

Electron Microscopy—The protein solutions were placed on 600-mesh carbon-coated copper grids for 5 min. The grids were washed twice with PBS/albumin/buffer (1%, pH 7.4). For immunogold labeling, the grids were incubated on a drop of antibody/gold solution (prediluted 1:50 in PBS) for 1 h in a chamber with a water-saturated atmosphere. After washing two times with PBS and distilled water, the grids were negatively stained with 2% uranyl acetate for 1 min. The specimens were examined in a Philips CM12 electron microscope at 100 kV (Eindhoven, Netherlands). For colloidal gold-antibody complexes, the colloidal sol was prepared by the reduction of tetrachloroauric acid by sodium citrate (17). To form the gold-protein complex, 2.5 ml of gold sol and 500 µl of protein solution were mixed and stirred for 15 min. Bovine serum albumin was added to final concentration of 0.1% and stirred for an additional 10 min. The solution was centrifuged at 32,000 rpm for 15 min at 4 °C, and the soft pellet was removed carefully.

Isolation of Peptides and Mass Spectrometry—For mass spectrometry peptides were eluted from the polyacrylamide gels after Coomassie staining. The bands of interest were excised, and then the peptides were eluted (18). Briefly, the gel pieces were cut into smaller pieces and totally destained by incubation in destaining solution (50 mM NH4HCO3, 50% acetonitrile) at room temperature under vigorous shaking. To replace the wash solution, the gel pieces were centrifuged in a benchtop centrifuge at 16,100 x g for 5 min, and the supernatant was removed carefully by using gel loader tips. The gel pieces were then dehydrated by incubation in 100 µl of 100% acetonitrile for 15 min. After discarding the supernatant, the elution was started by adding at least 40 µl of extraction buffer (formic acid/acetonitrile/isopropyl alcohol/H2O in a mixture of 50:25:15:10). Passive elution was carried out for 3 h at room temperature with vigorous shaking in a horizontal shaker (Eppendorf Thermomixer, Eppendorf, Hamburg, Germany). The supernatant was removed and pooled with the solution from a second round of elution carried out for 30 min. The gel pieces were dehydrated with 100% acetonitrile, and the supernatant was added to the pool of the elutions. The elution pool was dried, and the pellet was dissolved in 4 µl of a freshly prepared saturated matrix solution (sinapinic acid in 50% acetonitrile, 0.5% trifluoroacetic acid) for mass spectroscopy. This was performed with a surface-enhanced laser desorption ionization mass spectrometry instrument (PBSI, Ciphergen Biosystems, Fremont, CA). The samples were applied to chips covered with a hydrophobic surface (H4 chips, Ciphergen Biosystems) without any further pretreatment of the target and dried at 37 °C. The measurement was performed in the positive ion mode with various settings of laser power and sensitivity. A mixture of bovine insulin (Sigma), bovine insulin beta-chain (Sigma), ubiquitin (Sigma), ACTH (18-39) (Sigma), and tau constructs K19 and K18 were used for calibration of the instrument.

N-terminal Sequencing—Peptides were blotted on polyvinylidene difluoride membranes by semidry blotting and briefly stained with Coomassie. After destaining with intensive destainer, the bands were cut out. The membranes were completely destained with 0.1% triethylamine in methanol for about 1-3 min. Afterward, the membrane pieces were washed twice with 100% methanol for 1 min by vortexing. The membrane pieces were then dried and used for N-terminal sequencing with a protein sequencer (4cl, Applied Biosystems, Foster City, CA).


Figure 1
View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 1.
Tau constructs used for the generation of cell models. The top bar shows hTau40, the longest isoform of the human central nervous system (441 residues), containing 4 repeats of ~31 residues each, which are involved in microtubule binding and PHF aggregation. The tauRD constructs are based on the 4-repeat domain (construct K18, (M)Gln244-Glu372) It contains two hexapeptide motifs that promote PHF aggregation by formation of beta-structure (275VQIINK280 in R2 and 306VQIVYK311 in R3). The mutant K18{Delta}K280 lacks K280 (={Delta}K mutant) and therefore aggregates more readily. The double proline mutant K18{Delta}K280-PP contains I277P and I308P replacements in the hexapeptide motifs (= PP mutant) that inhibit aggregation because they disrupt a potential beta-strand.

 
Neurotoxicity Assay—Neurotoxicity was assessed using an LDH assay kit (Roche Applied Science) according to the manufacturer's specifications. The activity of LDH was measured spectrophotometrically at 492 nm. Cell death was calculated as percent of LDH released into medium, compared with total LDH obtained after total cell lysis. Toxicity values were determined in 1-day intervals. After N days of doxycycline-induced protein expression, the medium with 10% serum was exchanged for medium with 1% serum (after washing with PBS), and after 1 additional day the medium was collected for LDH determination.

Aggregation and Inhibition Assay in Cells—The N2a/K18{Delta}K280 cells were grown in Nunc flasks in Eagle's minimum essential medium supplemented with G418 (300 µg/ml) and hygromycin (100 µg/ml). The protein expression in the control sample was induced by addition of 1 µg/ml doxycycline (final concentration), and cells were incubated for 5 days. In the inhibition assay, the aggregation inhibitor compound was added together with doxycycline at a final concentration of 10 µM. After 5 days of protein expression, the cells were transferred to glass coverslips coated with polylysine, fixed with 3.7% paraformaldehyde in PBS, and briefly permeabilized with 80% MeOH. Next the cells were incubated with 0.01% thioflavine S, followed by incubation with rabbit antibody K9JA and secondary anti-rabbit antibody labeled with Cy5. For assaying the dissolution of preformed tau aggregates, the inducible N2a cells were incubated with 1 µg/ml doxycycline for 5 days. Then the medium was exchanged for new medium containing 1 µg/ml doxycycline and 10 µM of the inhibitor compound, and the incubation was continued for 2 more days. Transfer of cells onto coverslips and staining with ThS and tau antibody was performed as above. Finally, the cells showing ThS staining were scored in independent fields containing at least 500 cells.

Identification of Proteases and Inhibitors—The protein fragments generated in N2a cells were determined by sequencing and mass spectrometry. For identification of the protease, N2a cells were incubated with different protease inhibitors as follows: caspase inhibitor Z-VAD-FMK, 40 µM (Promega, Mannheim, Germany); proteasome inhibitor MG132, 1 µM (Calbiochem); calpain inhibitor ALLN, 10 µM (Calbiochem); and thrombin inhibitor PPACK (Biomol, Hamburg, Germany). In particular, for testing thrombin inhibition in cells the expression of K18{Delta}K280 was induced by doxycycline in the presence of 10-100 µM PPACK, followed by analysis of fragments. For thrombin cleavage in vitro, K18{Delta}K280 (1 mg/ml) was incubated with purified human {alpha}-thrombin (Hematologic Technologies, Essex Junction, VT) at final concentrations of 1-10 units/µg tau. The reaction mixtures were incubated at 37 °C for 3 h. The proteolytic reaction was stopped by Laemmli sample buffer and boiling for 5 min. Samples were blotted and analyzed by N-terminal sequencing and matrix-assisted laser desorption ionization time-of-flight.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Constructs and Neuronal Cell Lines for Studying Expression and Aggregation of TauRD—Tau derivatives used for establishing inducible N2a cell lines were based on the 4-repeat tau construct K18, which comprises the domain that is essential for microtubule binding and for PHF aggregation (Fig. 1). The repeat domain forms the core of Alzheimer PHFs and can be assembled into bona fide PHFs in vitro (19, 20). For studying the effect of aggregation, the mutation K18{Delta}K280 was particularly suitable because in vitro studies had shown a pronounced increase in the tendency to form beta-structure and aggregation, even without polyanionic cofactors (21). This mutation is one of the tau mutations in a case of FTDP-17 (22, 23). As a control we generated the PP mutant with prolines inserted into each of the two hexapeptide motifs (I277P and I308P) that normally nucleate beta-structure during aggregation. The prolines disrupt beta-strands and thus inhibit aggregation (24, 25).

To establish Tet-On inducible N2a cells, we first generated the host N2a clone with a stably integrated reversed tetracycline transactivator rtTA-S2 under the control of the cytomegalovirus promoter (26). Inducible cell lines were then generated by stable transfection of pBI-5 bifunctional vectors encoding the tauRD constructs and a reporter luciferase gene controlled by the Pbi-1 promoter. As a result, the following four cell lines were established: N2a/Tet-On/pBI-5 (mock), N2a/Tet-On/K18WT, N2a/Tet-On/K18{Delta}K280, and N2a/Tet-On/K18{Delta}K280/2P.

The cells were selected for clones with the lowest background of luciferase activity and tauRD expression. A robust induction was obtained at doxycycline concentrations above ~0.5 µg/ml. We therefore chose 1 µg/ml as a standard concentration for inducing tauRD expression. After 2 days of expression, the yield of 1 x 106 cells was comparable for the different constructs (~1.6 µg for K18WT, 1.8 µg for K18{Delta}K280, and 1.2 µg for K18{Delta}K280-PP).

Formation of PHF-like Aggregates in N2a Cells Expressing TauRD—One of the key questions of the experiments was the aggregation of tauRD and its effects on the cells. In principle, one could envisage various ways of how tauRD might affect cell behavior, e.g. overstabilization of microtubules and suppression of dynamics (27), inhibition of microtubule-based transport (28, 29), or aggregation, as suggested by the neurofibrillary pathology of Alzheimer disease and other tauopathies. In order to focus on the aggregation process, we required constructs that bind only weakly to microtubules (in order to avoid effects due to transport inhibition), yet are able to aggregate into PHFs; both conditions are met by the repeat domain. Furthermore, we needed cells in which aggregates could be formed reliably, and the tauRD expression could be induced and silenced at variable time points. Next, it was necessary to observe aggregation independently of the expression of soluble protein. As shown in Figs. 2, 3, 4, this can be achieved by fluorescence microscopy, biochemical analysis, and electron microscopy. The left panels of Fig. 2 demonstrate that each of the tau constructs expresses well in the cells after induction by doxycycline, as judged by immunofluorescence with a pan-tau antibody (K9JA). The same cells were probed by ThS, a dye that characteristically fluoresces when amyloid structures with a high content of cross-beta-structures are formed. ThS fluorescence is a faithful marker of the aggregation of tau and its derivatives in vitro (30) and can be used to screen for inhibitors of aggregation (31). Of the three tauRD constructs tested, only the {Delta}K280 mutant induced a strong reaction with ThS (~9% of the cells after 9 days of expression; see Fig. 2, b and d). This is consistent with the observation that this mutant of tauRD has a strong tendency for spontaneous aggregation in vitro, in contrast to other variants of tau that are normally highly soluble and need polyanions like heparin or RNA for aggregation. The wild-type tauRD construct K18 displayed a much weaker aggregation (~2% of cells after 9 days; Fig. 2, a and d), whereas aggregation was nearly absent from the inhibitory PP mutant (0.3% of cells; Fig. 2, c and d). We concluded that ThS fluorescence reflects tau aggregation not only in vitro but also in cells. Filaments resembling those of AD PHFs are present in the cells (Fig. 3) and can be demonstrated by negative stain electron microscopy of cell extracts. The identity of PHF-like filaments was shown directly after enrichment by iodixanol density gradient centrifugation and subsequent immunogold labeling with tau antibody (Fig. 3c).


Figure 2
View larger version (27K):
[in this window]
[in a new window]
  		FIGURE 2.
Expression and aggregation of tauRD in inducible N2a cells. Left, tau expression monitored by immunolabeling with antibody K9JA and rhodamine secondary antibody. Right, aggregates monitored by staining with thioflavine S. N2a cells expressing K18 tau constructs as follows: a, K18 WT; b, K18{Delta}K280; and c, K18{Delta}K280-PP. Note that appreciable aggregation occurs only in b. d, histogram showing the fraction of N2a cells with tauRD aggregates. Note that the {Delta}K mutant shows strong aggregation (black bars), whereas the {Delta}K-PP mutant is almost negative even after 9 days (white bar).

 
On the biochemical level, the presence of aggregates can be demonstrated by Sarkosyl extraction of the cells and analysis of soluble and insoluble components (16). The resulting proteins were separated by SDS-gel electrophoresis and blotted with antibodies against total tau or phosphorylated tau (Fig. 4). N2a cells contain very little endogenous tau (~0.1 ng/106 cells), which therefore does not interfere with the analysis. Moreover, the tauRD constructs have a lower Mr than full-length tau and are thus clearly identifiable. The polyclonal antibody K9JA was raised against the repeats and C-terminal domain and recognizes mainly the repeat domain of tau. The monoclonal antibody 12E8 was raised against tau phosphorylated at the KXGS motifs in the repeats and recognizes in particular Ser(P)262 in R1 (and to a lesser extent Ser(P)356 in R4) (32). Because these are the major phospho-sites in the repeats and because other domains of tau are absent, the analysis of K18 tau constructs does not suffer from complications due to the multiple Ser-Pro or Thr-Pro motifs of full-length tau that can be phosphorylated by proline-directed kinases. Thus, Fig. 4a illustrates that during the first 10 days of K18WT expression, the protein remains mostly in the soluble fraction and remains intact. The insoluble fraction is minor before day 11, at which time the sample reveals both lower molecular weight cleavage products and higher aggregates (Fig. 4a, lane 16).


Figure 3
View larger version (80K):
[in this window]
[in a new window]
  		FIGURE 3.
Electron microscopy of tau aggregates from Alzheimer brain and cell models. a, negative stain EM of PHFs purified from Alzheimer brain tissue by immunochromatography (gift of P. Davies), shown as a reference. b, pelleted Sarkosyl extracts of inducible N2a/Tet-On cells expressing K18{Delta}K280. c, Sarkosyl extracts enriched in PHFs using an iodixanol gradient. Fibers were immunogold-labeled with tau antibody K9JA conjugated to 8-nm gold particles.

 
A more detailed picture of the aggregation process is obtained from the pro-aggregation mutant K18{Delta}K280 (Fig. 4b). From day 4 onward, the expression of this construct leads to a noticeable high molecular weight smear in the Sarkosyl-insoluble pellet, reminiscent of Alzheimer aggregates (16, 33). Significantly, there are also pronounced lower molecular weight breakdown products, mainly in two bands at 9.9 and 7.3 kDa (fragments F2 and F3, see below). They are clearly smaller than the parent protein (F0 = K18, molecular mass = 12.4 kDa) and fragment F1 (~11.3 kDa) in soluble tauRD fractions (e.g. Fig. 4, a-c, lanes 3). The fragmentation becomes visible already at day 2, well before the appearance of aggregates. We concluded that aggregation is strongly correlated with cleavage and that tau fragments rapidly accumulate in the pellet. The quantitation of supernatants and pellets (Fig. 4d) confirms the gradual increase of aggregates, with ~13% of the {Delta}K280 mutant appearing in the Sarkosyl-insoluble fraction by day 7. In contrast to the {Delta}K mutant, the PP mutant shows almost no tendency to aggregate (Fig. 4c). This is consistent with the insertion of beta-breaking prolines into the hexapeptide motifs in R2 and R3, which therefore cannot support beta-strand interactions. The PP mutant also shows almost no fragments in the pellet, supporting the view that fragmentation is related to aggregation.

The phosphorylation of the KXGS sites in the repeats was probed with antibody 12E8 (Fig. 4, e-h). The phosphorylation of these motifs in neuronal tau is important for regulating neurite outgrowth (14), and they are the major sites in the constructs used here (because the phosphorylation sites in adjacent domains are absent). For all tauRD variants the phosphorylation remains concentrated in the supernatant rather than the pellet (Fig. 4, e-g). This argues that phosphorylation at KXGS motifs does not enhance the tendency to aggregate, consistent with earlier observations (34). Fragmentation is also less apparent in the phosphorylated protein. Fragment F1 is only found in the soluble fraction and is phosphorylated. F2 is mainly found in the insoluble fraction (pellet) and is also phosphorylated (at a reduced level), whereas F3 in the insoluble fraction is not phosphorylated.


Figure 4
View larger version (58K):
[in this window]
[in a new window]
  		FIGURE 4.
Kinetics of tauRD expression, aggregation, and phosphorylation in cell models. Lanes labeled p denote Sarkosyl-insoluble tauRD species that are pelletable after Sarkosyl extraction and measured by blot analysis and densitometry. s indicates the corresponding soluble proteins. a-c, blot analysis with antibody (Ab) K9JA (independent of phosphorylation). a, expression of K18; b, K18{Delta}K280 mutant; c, K18{Delta}K280-PP mutant. Note that in a a fraction of the K18 protein gradually becomes insoluble (p lanes) but retains its normal apparent molecular weight until day 11 when a higher molecular weight smear appears. c, there is very little aggregation of the PP mutant. By contrast, the {Delta}K mutant (b) shows pronounced aggregation in the pellet, combined with increasing proteolysis (lower bands in the p lanes) and a high molecular weight smear of aggregates. d, densitometric quantitation of protein aggregation (% Sarkosyl-insoluble tau). e-g, blots with antibody 12E8 (recognizes phosphorylated Ser(P)262/Ser(P)356). Note that phosphorylation at the 12E8 site occurs primarily on tauRD with normal molecular weight in the soluble fraction, indicating that the fragments and higher molecular weight smear are mostly unphosphorylated and that phosphorylation of intact protein occurs before it aggregates. h, densitometric quantitation of phosphorylation at the 12E8 epitope (ratio of 12E8/K9JA antibody signals). Note that the degree of phosphorylation increases sharply at day 2 (preceding aggregation) and then decreases again.

 


Figure 5
View larger version (44K):
[in this window]
[in a new window]
  		FIGURE 5.
Identification of tau fragments and cleavage by thrombin. Cell extracts were separated by SDS-PAGE and blotted with antibody K9JA. a, tau cleavage in N2a cells. Expression of the K18{Delta}K280 mutant was induced with doxycycline for 5 days, and soluble and Sarkosyl-insoluble tau fractions were separated by centrifugation. The supernatant (s, lane 2) shows intact tau K18{Delta}K and fragment F1. The pellet (p, lane 3) shows insoluble tauRD and fragments F2 and F3. Lanes 1 and 4 show recombinant K18{Delta}K for comparison. Lanes 5-12 show parallel experiments with N2a cells in the presence of different concentrations of thrombin inhibitor PPACK (10-100 µM), which suppresses the cleavage of tauRD and the accumulation of Sarkosyl-insoluble material. b, quantitation of the fraction of Sarkosyl-insoluble K18{Delta}K280 from inducible N2a cells treated with different amounts of PPACK. The fraction of accumulated Sarkosyl-insoluble material decreased ~50% at a PPACK concentration of 100 µM. c, cleavage of tauRD in vitro.73 µM K18{Delta}K samples were incubated with increasing thrombin concentrations for 3 h, which generated the major fragment F1 (lanes 3-7). Fragmentation was inhibited with the thrombin inhibitor PPACK (1 mM) (lane 8). For comparison, lane 2 shows the fragments generated in cells (as in a). d, quantitative densitometric analysis of in vitro thrombin cleavage shown in c.

 
The Fragmentation of TauRD Precedes Aggregation—The results given so far suggest that fragmentation of tauRD, rather than phosphorylation, is important for aggregation in the inducible cell models. To verify this, we analyzed the major protein bands by N-terminal sequencing and mass spectrometry (Fig. 5). As expected, the top band (F0 = 12.4 kDa) corresponds to the intact K18 construct (M)Gln244-Glu372 (130 amino acid residues; see Table 1). Fragment F1 from soluble K18 runs at ~11.3 kDa; its observed mass is 12,087 kDa. N-terminal sequencing reveals a peptide starting at Ser258. The peptide Ser258-Glu372 (115 residues) would have a calculated molecular mass of 12,034 kDa, in good agreement with the observed mass. Thus, F1 represents the major part of K18, lacking the N-terminal 15 residues. The cleavage occurs between Lys257 and Ser258. Most surprisingly, the major components of fragments F2 and F3 also start at Ser258, indicating that this Lys-Ser motif represents a major target for the responsible protease. The C-terminal ends of the peptides can be deduced from the masses (Table 1). F2 (9.9 kDa) represents the peptides Ser258-Val363 (106 residues) and Ser258-Pro364 (107 residues); F3 (Mr 7.3 kDa) is Ser258-Lys353 (96 residues) and Ser258-Ile360 (103 residues).


View this table:
[in this window]
[in a new window]
  		TABLE 1
Characterization of tau fragments

The fragments were identified by N-terminal sequencing and matrix-assisted laser desorption ionization time-of-flight after elution from the SDS gels. F0 = K18 is the parent protein; fragment F1 is the major fragment of soluble tau in cells, and it is identical to the major fragment generated by thrombin cleavage in vitro. F2 and F3 occur mainly in the insoluble fraction of tau fragments in cells. The fragments start at the same N-terminal cleavage site (Ser258, thrombin) but differ in length. Mr = apparent relative molecular weight in the SDS gel; Mobs = molecular mass determined by mass spectrometry; Mcalc = mass calculated for the best fitting peptide.

 
Which protease(s) are responsible for the fragments? The site Lys257/Ser258 has been reported as a major cleavage site by thrombin (35), and it is consistent with the specificity of thrombin; indeed the treatment of cells with the thrombin inhibitor PPACK inhibited the fragmentation of tauRD (Fig. 5, a and b, and Table 1). Inhibition of several other proteases was tested, e.g. caspase inhibitor Z-VAD-FMK, proteasome inhibitor MG132, or calpain inhibitor ALLN, but there was no effect on the fragments generated (data not shown). By contrast, the C-terminal ends of the fragments did not conform to unique protease specificities (except possibly the Lys353/Ile354 site, another potential thrombin target). This could be because of the activity of carboxypeptidases or autolytic processes, such as nonenzymatic cleavage behind Asn (33) which would generate heterogeneous C-terminal ends (Table 1).

To check the suspected role of thrombin, we performed digestion experiments with K18{Delta}K280 in vitro (Fig. 5, c and d). Again, the major N-terminal cleavage site was at Lys257/Ser258, which generated a single fragment (F1) extending to the end of the protein (Ser258-Glu372); a minor cleavage site was Lys254/Asn255, generating the slightly longer component Asn255-Glu372. Other fragments were negligible in this experiment, confirming that the repeat domain contains only one dominant thrombin site. By implication, the C-terminal digestion that occurs in cells originates from other proteases.

Toxicity and Aggregation—In AD brains, the aggregation of tau correlates well with the degeneration of neurons (36). However, it is not clear whether the toxicity is because of tau aggregates as such or to some other related events. We approached this issue by inducing the expression of different tau constructs in N2a cells for different times and observing the extent of cell death by the LDH release assay, which reflects the leakiness of membranes in degenerating cells. The results (Fig. 6) show that the expression of soluble tauRD has no noticeable effect on the viability of the cells, i.e. K18wt and the anti-aggregation PP mutant have the same level of LDH release (~10%) as the control cells even after 4 days. At this point these two proteins show only a negligible extent of aggregation (see Fig. 4). We concluded that soluble tauRD is not toxic to N2a cells, as seen by the LDH assay during the first few days. However, the situation is different for the pro-aggregation {Delta}K280 mutant. In this case the toxicity is about 2-fold higher than in the control (Fig. 6, 3rd bar). Most interestingly, the toxicity is elevated from day 1 onward, earlier than the appearance of aggregation. We concluded that aggregating species of tauRD cause toxicity, even at a stage where the aggregates are not yet detected by ThS fluorescence in cells (e.g. pre-aggregated oligomers).

The Aggregation of Tau Is Reversible—An important question in the field is whether tau aggregates, once formed, can be removed again. This question can be addressed by using the inducible N2a cell lines. The expression of tauRD was first induced for 5 days by the addition of doxycycline until aggregates were clearly present in the Sarkosyl-insoluble pellet, and then doxycycline was removed, and the cells were assayed for aggregation after 2-6 days in terms of ThS fluorescence and biochemical analysis (Fig. 7). In all cases the aggregates disappeared again, showing that, in principle, the aggregation is reversible, at least during the early stages. As in Fig. 4, the degree of aggregation is most pronounced with the {Delta}K mutant (~10-15% of expressed protein; Fig. 4d). It displays both the high molecular weight smear and the lower molecular weight triplet, consisting of the original protein (F0 = K18{Delta}K280) and the two fragments characteristic of the aggregated pellet (F2 and F3) (Fig. 7b). When the expression of tauRD was silenced by removal of doxycycline, all these bands as well as the ThS fluorescence in cells and the PHFs in the Sarkosyl pellet disappeared. This takes place independently of the proteasome inhibitor MG132, arguing that the proteasome is not responsible for tau degradation in these conditions. It is remarkable that the rate of disappearance is similar for all components, with a half-time of ~3 days. Similar results are observed for K18wt and the PP mutant (Fig. 7, a and c).


Figure 6
View larger version (15K):
[in this window]
[in a new window]
  		FIGURE 6.
Cell toxicity of tauRD expression and aggregation. Tau construct K18 or mutants were induced for 1-4 days; LDH release was measured as an indicator of cell death, calculated as percent of total LDH (media + lysate). Bar 1, LDH release of N2a mock cells without doxycycline-induced tau expression (control). Bar 2, N2a cells expressing K18; bar 3, {Delta}K mutant; bar 4, PP mutant. Note that the expression of the {Delta}K mutant strongly increases toxicity, whereas wild-type K18 and the PP mutant remain around control values. K18{Delta}K280 is toxic even during the first 2 days of expression when ThS-positive aggregates are not yet visible (see Fig. 4), suggesting that soluble and/or oligomeric forms are also toxic.

 
In conclusion, the aggregation of tauRD can clearly be reversed simply 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 long term secondary modifications and cross-linking. The similar rate of disappearance of all aggregated components would be explained if they all were part of a common supramolecular structure whose disassembly would make the subunits vulnerable to degradation.


Figure 7
View larger version (33K):
[in this window]
[in a new window]
  		FIGURE 7.
Reversibility of aggregation of tauRD. Expression of K18 or mutants was first induced for 5 days by doxycycline, then switched off by removal of doxycycline, and assayed for expression and aggregation. Left, blots of supernatants (s) and Sarkosyl-insoluble pellets (p); right, quantitation of supernatants and pellets by densitometry (fractions of tau retained after withdrawal of doxycycline, black and gray bars, respectively). a, K18; b, K18{Delta}K280 mutant; c, K18{Delta}K280/PP mutant. Note that in each case the amount of soluble and insoluble material, formed during the 5 days of tau expression, decreases again after silencing the tau gene.

 
TauRD Aggregation in Cells Can Be Inhibited by Small Molecule Inhibitors—The fact that tau aggregation can be inhibited and reversed suggests that there is initially a labile equilibrium between soluble and aggregated forms of tau. One would therefore expect that it should be possible to reverse or suppress the aggregation in cells by compounds that interfere with the self-assembly of tau. We had previously screened a compound library for compounds that inhibit the aggregation of tauRD in vitro or are even capable of disassembling preformed aggregates (31). Selected compounds were tested on the inducible N2a cell lines (Fig. 8). They belonged to the class of N-phenylamines and showed IC50 and DC50 values in the micromolar range in vitro (Fig. 8, d and e). TauRD (K18{Delta}K280) was first induced for 5 days, resulting in a robust expression of the protein and formation of aggregates (as judged by ThS staining; Fig. 8, a, top row and b, black bars). In parallel experiments, the induction of tauRD was done in the presence of inhibitor compounds. In this case the aggregation was strongly reduced (~30-40%), even though the level of protein expression remained the same as in the controls (Fig. 8, a, 2nd and 3rd row, and b, light gray bars). Finally, in order to test whether aggregates in cells can be disrupted again, the expression of tauRD was first induced for 5 days, and then compounds were added for 2 more days. At that point, the level of aggregation was reduced to ~50% of the controls (Fig. 8b, gray bars). Most interestingly, this procedure also reduced the toxicity of tauRD nearly to control levels (Fig. 8c). These experiments showed that aggregates of tauRD can be prevented by inhibitor compounds, that even pre-formed PHFs can be dissolved again in the N2a cell model, and that the removal of the aggregates is beneficial for the cells.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The abnormal aggregation and hyperphosphorylation of tau are hall-marks of Alzheimer disease and other tauopathies (2, 37). It is often assumed that phosphorylation of tau promotes aggregation; this would be consistent with the reaction of phospho-specific antibodies in AD brain tissue but difficult to reconcile with the observation that phosphorylation of tau in the repeat domain tends to inhibit aggregation (34). To clarify these issues, we developed a new cell model for the neurofibrillary pathology in AD which combines the following features. 1) The system is based on a neuronal cell line (N2a), which is well characterized, can be differentiated, and contains only low amounts of endogenous tau that does not interfere with exogenous tauRD. 2) The expression of tauRD can be switched on and off in order to observe the appearance of aggregates. 3) The cell model was designed in three variants, expressing the repeat domain of tau either in the wild-type sequence or with a pro-aggregation or anti-aggregation mutations. 4) The aggregation of tauRD can be monitored in cells biochemically or by electron microscopy.


Figure 8
View larger version (30K):
[in this window]
[in a new window]
  		FIGURE 8.
Inhibition of aggregation and toxicity of tauRD by aggregation inhibitors. a, N2a cells were induced to express K18{Delta}K280 for 5 days. Left column, staining for tau (antibody K9JA-TRITC); center, staining for aggregates (thioflavine S); right, merged images. Note that the {Delta}K280 mutant aggregates readily and therefore generates numerous ThS-positive cells (top row, center). The middle and bottom rows show experiments in the presence of aggregation inhibitor compounds B4D3 and B1C11, which strongly reduce the number of ThS-positive cells. b, quantification of effects of aggregation inhibitor compounds, normalized to control without inhibitor (100%). Black bars, ThS fluorescence of control cells with aggregated K18{Delta}K280 after 5 days; gray bars, in the presence of inhibitor compounds during tauRD induction the aggregation is reduced to ~30%; dark gray bars, tauRD induction for 5 days and then incubation of cells with compounds for 2 more days, which causes the depolymerization of aggregates down to ~50%. c, rescue from toxicity of aggregating K18{Delta}K280 in the inducible N2a cells measured by LDH release. Bars illustrate LDH release after 2 days of expression of K18{Delta}K280 induced with doxycycline in the absence or presence of small molecule inhibitors B1C11 and B4D3. The inhibitors reduce LDH release nearly to control levels. d, structure of inhibitor compounds. e, dose-response curves measured in vitro (for methods see Pickhardt et al. (31)). Left, inhibition of aggregation (IC50 values around 10 µM); right, dissolution of preformed aggregates (DC50 values ~5-10 µM).

 
With these tools we addressed several issues as follows. 1) Is the aggregation of tauRD toxic to cells? 2) Can the aggregation of tauRD and its toxicity be reversed? 3) What is the relationship between phosphorylation of tauRD and its aggregation? 4) Can the aggregation be prevented or reversed by drugs?

Indicators for Pathological Aggregation—One difficulty in the field has been to find reliable indicators of abnormal aggregation in tau-expressing cells. Even strong tau overexpression generally does not cause PHFs but rearrangements of the cytoskeleton (38). This problem can be overcome in part by FTDP-17 mutations of tau (6-10, 12, 39) by adding the cytotoxic Abeta peptide (11) or by expressing truncated variants of tau. Here we used the 4-repeat domain K18 with the {Delta}K280 mutation because it aggregates avidly without requiring polyanionic inducers (21, 25). The aggregation can be measured by three markers.

  1. ThS fluorescence of cells reports the presence of cross-beta-structure. By contrast, untransfected cells show no ThS signal, and the strength of the ThS signal correlates with the propensity of the protein to form beta-structure (Fig. 2).
  2. PHFs can be visualized by electron microscopy. They tend to be masked by other cell components but are enriched by an iodixanol gradient and then identified by immunogold labeling with a tau antibody (Fig. 3).
  3. The presence of PHFs in the Sarkosyl-insoluble pellet is accompanied by a smear at intermediate molecular weights, which is also characteristic for AD tau (16) and can be explained by nonenzymatic fragmentation and cross-linking (33).

Aggregation and Phosphorylation—In AD, tau hyperphosphorylation occurs together with aggregation. Therefore, we asked how these two parameters are related in the cell models. The tight correlation between the beta-structure propensity of tauRD and the aggregation in cells argues that PHF formation is indeed based on an intrinsic property of the protein. In particular it involves the same hexapeptide interaction motifs in repeats R2 and R3 whose beta-propensity can be either enhanced (as in the {Delta}K280 mutant) or diminished (as in the PP mutant).

Next we asked whether aggregation is linked to phosphorylation. Multiple site phosphorylation is a common feature of tauopathies, but there is a debate on whether it is a cause of aggregation. In our case we found no obvious relationship (Fig. 4). Some degree of phosphorylation at the KXGS motifs was observed on all variants of tauRD, and it was present at early time points long before aggregation and was concentrated in the soluble rather than the insoluble fraction. We conclude that phosphorylation of the repeat domain of tau does occur in the cell, but it does not predict aggregation. One should note here that many of the phosphorylation sites of AD tau, notably the SP or TP motifs, lie outside the repeat domain and do not show up here. The main phosphorylation sites in tauRD are the KXGS motifs (phosphorylated by the kinase MARK and detected by the antibody 12E8), which promote the dissociation from microtubules (29, 40). Generally, it is reasonable to assume that breaking tau-microtubule bonds is a prerequisite for tau-tau aggregation, especially because both interactions are based on similar interaction motifs (41). However, it appears that phosphorylation in the repeat domain cannot be considered as a precursor of PHF-tau as such, in agreement with our earlier observations that phosphorylation of tauRD at the KXGS motifs inhibits aggregation (34).

Tau Fragmentation Versus Aggregation—An unexpected finding was the involvement of specific tau fragments and proteases in the aggregation process. There are three main fragments as follows: F1 remains mostly in solution, and F2 and F3 become enriched in the Sarkosyl-insoluble pellet (Figs. 4 and 5). These fragments derive mainly from the same N-terminal cleavage site, Lys257-Ser258, but have different C-terminal ends (between Lys353 and Val364; Table 1). The N-terminal site is suggestive of a thrombin-like activity, in agreement with Ref. 35. In vitro digestion with thrombin confirmed the major cleavage at the Lys257-Ser258 bond, which can be inhibited by the thrombin inhibitor PPACK but not by other protease inhibitors (proteasome inhibitor MG132, calpain inhibitor ALLN, and caspase inhibitor Z-VAD-FMK). The C-terminal cleavage site is less well defined, probably due to other proteases because fragments F2 and F3 are not generated by thrombin in vitro (Fig. 5). This suggests the following scenario. Soluble tauRD can be cleaved by thrombin within the first repeat, generating fragment F1 from Ser258 to the end of the repeat domain that remains largely soluble. Later fragment F1 becomes digested from its C-terminal end, generating F2 and F3 that aggregate readily. The high molecular weight smear appears subsequently, presumably due to covalent modification and cross-linking of tau in the aggregates (33).

Why should fragments F2 and F3 be more prone to aggregation? As mentioned before, the repeat domain aggregates better than full-length tau because it contains the hexapeptide motifs responsible for aggregation (24). As shown recently, fragments F2 and F3 are enriched in motifs with a propensity for beta-structure (41), which would explain why they aggregate and presumably also nucleate the aggregation of uncleaved tauRD.

Proteolytic processing plays a prominent role in Alzheimer disease. The Abeta peptides are generated through inappropriate cleavage by beta- and {gamma}-secretases (1). Cleavage of tau occurs early in the disease process, e.g. after Glu391 (by an unknown protease (42), or after Asp421 by caspase 3 (43, 44), and the removal of the C-terminal tail is thought to expose the repeat domain and facilitate its aggregation. Cleavage of tau by calpain has been observed as well, following the exposure of neurons to Abeta and generating a toxic fragment (Glu45-Arg230), which might interfere with the physiological function of tau (45). The unusual feature of our present data is the activity of a protease that is normally considered to be extracellular. Nevertheless, the specificity and inhibition of the protease identify it as a thrombin-like enzyme, in agreement with Ref. 35.

Aggregation and Toxicity—One key question was whether the aggregation of tauRD is toxic to the cell or whether the degeneration is caused by other events. The results shown here argue that toxicity is directly related to aggregation and not merely to the expression of tauRD. Thus, the pro-aggregation {Delta}K280 mutant showed more than twice the LDH release than the control or the anti-aggregation mutant (PP). We note that the toxicity is already high at an early time point before abundant aggregation. It is therefore possible that toxicity is caused by species smaller than full-blown PHFs, such as oligomers of tauRD. This is reminiscent of the aggregation of Abeta or other peptides where oligomers, rather than polymers, are considered the main toxic species (46-48).

Reversibility of Aggregation and Anti-aggregation Drugs—If aggregates of tauRD are toxic, it would be important to find methods to counteract aggregation. In the first instance, one might ask whether the cell has a clearing mechanism for aggregated tau. This can be tested by silencing the expression of tauRD (Fig. 7), and indeed soluble tauRD disappears over a period of ~4 days. More importantly, aggregated tauRD also disappears in cells as seen by the ThS staining, and the same applies to all components of the Sarkosyl pellet. The cell must therefore possess mechanisms to break down the aggregates. Several catabolic pathways have been proposed for PHFs in Alzheimer disease, including the ubiquitin-proteasome system (49, 50) and degradation by calpain (51, 52) or macroautophagy (53). In our cell model, the disappearance of aggregates after switching off the expression of tauRD was not altered in the presence of proteasome inhibitors (MG132), implying that other catabolic mechanisms were active. As shown elsewhere (54), aggregates of tauRD formed in vitro are remarkably labile (in contrast to aged Alzheimer tangles) and can therefore be disintegrated into subunits, which could be digested by various cellular proteases. In summary, the encouraging lesson is that toxic aggregates can be removed by the cell once the production of tauRD is lowered.

The practical consequence is that it should be possible to enhance the self-cleaning capabilities of the cell by drugs that inhibit aggregation. Such compounds can be identified by in vitro screening (31, 55, 56). The examples of Fig. 8 show that the approach is successful in suppressing aggregates in cells. In this case, the expression of tauRD continued without interruption, but the aggregates were reduced by ~60% within 2 days. More significantly, the LDH assay showed the cells actually recovered from the toxic insult (Fig. 8c). Thus, the cell model is suitable for testing the effectiveness of inhibitor drugs of tau aggregation.

Tau and Intracellular Transport—So far we have discussed the aggregation of the repeat domain of tau in cells without considering other potential toxic roles of full-length tau. However, tau regulates microtubules for their roles in neurite outgrowth and axonal transport. In this context, tau plays an ambiguous role, as shown previously (28, 29). Tau binds to the surface of microtubules, but in doing so it competes for the binding of motor proteins (especially kinesin) and therefore tends to inhibit their motion. The cell tunes these opposing requirements by the phosphorylation of tau (57). An imbalance in this system, which increases tau on the microtubule surface (e.g. overproduction or underphosphorylation), can disrupt axonal transport so that synapses and neurites become starved and decay. This toxicity is an early phenomenon, prior to aggregation. The effect could contribute to the loss of synapses and the dying back of cell processes observed in AD. It is presently not clear which of the toxic effects of tau, traffic disruption or aggregation, plays the dominant role, but we note that certain aspects of tau expression in transgenic mice are well explained by the traffic jam hypothesis, for example the motor neuron axonopathies when tau is expressed in the spinal cord (4). In our cell model, the problem of traffic inhibition is circumvented because the repeat domain of tau and its mutants bind weakly to microtubules and instead aggregate more avidly (21). This allows us to analyze the effects of aggregation separately from traffic inhibition. In the future it will be important to sort out the inter-play between the different modes of tau toxicity in neurons.


   FOOTNOTES
 
* This work was supported by grants from the Deutsche Forschungsgemeinschaft, Max-Planck-Gesellschaft, and Institute for the Study of Aging. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

1 Both authors contributed equally to this work. Back

2 To whom correspondence should be addressed. E-mail: mandelkow{at}mpasmb.desy.de.

3 The abbreviations used are: AD, Alzheimer disease; FTDP-17, frontotemporal dementia with parkinsonism linked to chromosome 17; LDH, lactate dehydrogenase; PHF, paired helical filament; TauRD, tau repeat domain; Z, benzyloxycarbonyl; FMK, fluoromethyl ketone; PBS, phosphate-buffered saline; wt, wild type; ThS, thioflavine S; PPACK, D-phenylalanyl-L-prolyl-L-arginine chloromethyl ketone. Back


   ACKNOWLEDGMENTS
 
We thank Dagmar Drexler and Olga Petrova for excellent technical assistance. We are grateful to U. Aebi and B. Maco (Biocenter Basel) for help with immunogold labeling.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Selkoe, D. J. (2004) Nat. Cell Biol. 11, 1054-1061
  2. Lee, V. M., Goedert, M., and Trojanowski, J. Q. (2001) Annu. Rev. Neurosci. 24, 1121-1159[CrossRef][Medline] [Order article via Infotrieve]
  3. Schenk, D., Hagen, M., and Seubert, P. (2004) Curr. Opin. Immunol. 16, 599-606[CrossRef][Medline] [Order article via Infotrieve]
  4. Terwel, D., Lasrado, R., Snauwaert, J., Vandeweert, E., Van Haesendonck, C., Borghgraef, P., and Van Leuven, F. (2005) J. Biol. Chem. 280, 3963-3973[Abstract/Free Full Text]
  5. Zhang, B., Maiti, A., Shively, S., Lakhani, F., McDonald-Jones, G., Bruce, J., Lee, E. B., Xie, S. X., Joyce, S., Li, C., Toleikis, P. M., Lee, V. M., and Trojanowski, J. (2005) Proc. Natl. Acad. Sci. U. S. A. 102, 227-231[Abstract/Free Full Text]
  6. Lewis, J., Dickson, D. W., Lin, W. L., Chisholm, L., Corral, A., Jones, G., Yen, S. H., Sahara, N., Skipper, L., Yager, D., Eckman, C., Hardy, J., Hutton, M., and McGowan, E. (2001) Science 293, 1487-1491[Abstract/Free Full Text]
  7. Gotz, J., Chen, F., van Dorpe, J., and Nitsch, R. M. (2001) Science 293, 1491-1495[Abstract/Free Full Text]
  8. Oddo, S., Caccamo, A., Shepherd, J. D., Murphy, M. P., Golde, T. E., Kayed, R., Metherate, R., Mattson, M. P., Akbari, Y., and LaFerla, F. M. (2003) Neuron 39, 409-421[CrossRef][Medline] [Order article via Infotrieve]
  9. Vogelsberg-Ragaglia, V., Bruce, J., Richter-Landsberg, C., Zhang, B., Hong, M., Trojanowski, J. Q., and Lee, V. M. (2000) Mol. Biol. Cell 11, 4093-4104[Abstract/Free Full Text]
  10. DeTure, M., Ko, L. W., Easson, C., and Yen, S. H. (2002) Am. J. Pathol. 161, 1711-1722[Abstract/Free Full Text]
  11. Ferrari, A., Hoerndli, F., Baechi, T., Nitsch, R. M., and Gotz, J. (2003) J. Biol. Chem. 278, 40162-40168[Abstract/Free Full Text]
  12. Santacruz, K., Lewis, J., Spires, T., Paulson, J., Kotilinek, L., Ingelsson, M., Guimaraes, A., DeTure, M., Ramsden, M., McGowan, E., Forster, C., Yue, M., Orne, J., Janus, C., Mariash, A., Kuskowski, M., Hyman, B., Hutton, M., and Ashe, K. H. (2005) Science 309, 476-481[Abstract/Free Full Text]
  13. Gossen, M., and Bujard, H. (2002) Annu. Rev. Genet. 36, 153-173[CrossRef][Medline] [Order article via Infotrieve]
  14. Biernat, J., Wu, Y. Z., Timm, T., Zheng-Fischhofer, Q., Mandelkow, E., Meijer, L., and Mandelkow, E. M. (2002) Mol. Biol. Cell 13, 4013-4028[Abstract/Free Full Text]
  15. Baron, U., Freundlieb, S., Gossen, M., and Bujard, H. (1995) Nucleic Acids Res. 17, 3605-3606
  16. Greenberg, S. G., and Davies, P. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 5827-5831[Abstract/Free Full Text]
  17. Slot, J. W., and Geuze, H. J. (1985) Eur. J. Cell Biol. 38, 87-93[Medline] [Order article via Infotrieve]
  18. Feick, R. G., and Shiozawa, J. A. (1990) Anal. Biochem. 187, 205-211[CrossRef][Medline] [Order article via Infotrieve]
  19. Wischik, C. M., Novak, M., Edwards, P. C., Klug, A., Tichelaar, W., and Crowther, R. A. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 4884-4888[Abstract/Free Full Text]
  20. Wille, H., Drewes, G., Biernat, J., Mandelkow, E. M., and Mandelkow, E. (1992) J. Cell Biol. 118, 573-584[Abstract/Free Full Text]
  21. Barghorn, S., Zheng-Fischhofer, Q., Ackmann, M., Biernat, J., von Bergen, M., Mandelkow, E. M., and Mandelkow, E. (2000) Biochemistry 39, 11714-11721[CrossRef][Medline] [Order article via Infotrieve]
  22. Rosso, S. M., and van Swieten, J. C. (2002) Curr. Opin. Neurol. 5, 423-428
  23. D'Souza, I., and Schellenberg, G. D. (2005) Biochim. Biophys. Acta 1739, 104-115[Medline] [Order article via Infotrieve]
  24. von Bergen, M., Friedhoff, P., Biernat, J., Heberle, J., Mandelkow, E. M., and Mandelkow, E. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 5129-5133[Abstract/Free Full Text]
  25. von Bergen, M., Barghorn, S., Li, L., Marx, A., Biernat, J., Mandelkow, E. M., and Mandelkow, E. (2001) J. Biol. Chem. 276, 48165-48174[Abstract/Free Full Text]
  26. Urlinger, S., Baron, U., Thellmann, M., Hasan, M. T., Bujard, H., and Hillen, W. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 7963-7968[Abstract/Free Full Text]
  27. Bunker, J. M., Wilson, L., Jordan, M. A., and Feinstein, S. C. (2004) Mol. Biol. Cell 15, 2720-2728[Abstract/Free Full Text]
  28. Stamer, K., Vogel, R., Thies, E., Mandelkow, E., and Mandelkow, E. M. (2002) J. Cell Biol. 156, 1051-1063[Abstract/Free Full Text]
  29. Mandelkow, E. M., Thies, E., Trinczek, B., Biernat, J., and Mandelkow, E. (2004) J. Cell Biol. 167, 99-110[Abstract/Free Full Text]
  30. Friedhoff, P., von Bergen, M., Mandelkow, E. M., Davies, P., and Mandelkow, E. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 15712-15717[Abstract/Free Full Text]
  31. Pickhardt, M., Gazova, Z., von Bergen, M., Khlistunova, I., Wang, Y., Hascher, A., Mandelkow, E. M., Biernat, J., and Mandelkow, E. (2005) J. Biol. Chem. 280, 3628-3635[Abstract/Free Full Text]
  32. Seubert, P., Mawal-Dewan, M., Barbour, R., Jakes, R., Goedert, M., Johnson, G. V., Litersky, J. M., Schenk, D., Lieberburg, I., Trojanowski, J. Q., and Lee, V. M. Y. (1995) J. Biol. Chem. 270, 18917-18922[Abstract/Free Full Text]
  33. Watanabe, A., Hong, W. K., Dohmae, N., Takio, K., Morishima-Kawashima, M., and Ihara, Y. (2004) J. Neurochem. 90, 1302-1311[CrossRef][Medline] [Order article via Infotrieve]
  34. Schneider, A., Biernat, J., von Bergen, M., Mandelkow, E., and Mandelkow, E. M. (1999) Biochemistry 38, 3549-3558[CrossRef][Medline] [Order article via Infotrieve]
  35. Arai, T., Guo, J. P., and McGeer, P. L. (2005) J. Biol. Chem. 280, 5145-5153[Abstract/Free Full Text]
  36. Braak, H., and Braak, E. (1991) Acta Neuropathol. 82, 239-259[CrossRef][Medline] [Order article via Infotrieve]
  37. Binder, L. I., Guillozet-Bongaarts, A. L., Garcia-Sierra, F., and Berry, R. W. (2005) Biochim. Biophys. Acta 1739, 216-223[Medline] [Order article via Infotrieve]
  38. Barlow, S., Gonzalez-Garay, M. L., West, R. R., Olmsted, J. B., and Cabral, F. (1994) J. Cell Biol. 126, 1017-1029[Abstract/Free Full Text]
  39. Perez, M., Hernandez, F., Gomez-Ramos, A., Smith, M., Perry, G., and Avila, J. (2002) Eur. J. Biochem. 269, 1484-1489[Medline] [Order article via Infotrieve]
  40. Drewes, G., Ebneth, A., Preuss, U., Mandelkow, E. M., and Mandelkow, E. (1997) Cell 89, 297-308[CrossRef][Medline] [Order article via Infotrieve]
  41. Mukrasch, M. D., Biernat, J., von Bergen, M., Griesinger, C., Mandelkow, E., and Zweckstetter, M. (2005) J. Biol. Chem. 280, 24978-24986[Abstract/Free Full Text]
  42. Novak, M., Kabat, J., and Wischik, C. M. (1993) EMBO J. 12, 365-370[Medline] [Order article via Infotrieve]
  43. Gamblin, T. C., Chen, F., Zambrano, A., Abraha, A., Lagalwar, S., Guillozet, A. L., Lu, M., Fu, Y., Garcia-Sierra, F., LaPointe, N., Miller, R., Berry, R. W., Binder, L. I., and Cryns, V. L. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 10032-10037[Abstract/Free Full Text]
  44. Rissman, R. A., Poon, W. W., Blurton-Jones, M., Oddo, S., Torp, R., Vitek, M. P., LaFerla, F. M., Rohn, T. T., and Cotman, C. W. (2004) J. Clin. Investig. 114, 121-130[CrossRef][Medline] [Order article via Infotrieve]
  45. Park, S. Y., and Ferreira, A. (2005) J. Neurosci. 25, 5365-5375[Abstract/Free Full Text]
  46. Lambert, M. P., Barlow, A. K., Chromy, B. A., Edwards, C., Freed, R., Liosatos, M., Morgan, T. E., Rozovsky, I., Trommer, B., Viola, K. L., Wals, P., Zhang, C., Finch, C. E., Krafft, G. A., and Klein, W. L. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 6448-6453[Abstract/Free Full Text]
  47. Walsh, D. M., Klyubin, I., Fadeeva, J. V., Cullen, W. K., Anwyl, R., Wolfe, M. S., Rowan, M. J., and Selkoe, D. J. (2002) Nature 416, 535-539[CrossRef][Medline] [Order article via Infotrieve]
  48. Bucciantini, M., Giannoni, E., Chiti, F., Baroni, F., Formigli, L., Zurdo, J., Taddei, N., Ramponi, G., Dobson, C. M., and Stefani, M. (2002) Nature 416, 507-511[CrossRef][Medline] [Order article via Infotrieve]
  49. David, D. C., Layfield, R., Serpell, L., Narain, Y., Goedert, M., and Spillantini, M. G. (2002) J. Neurochem. 83, 176-185[CrossRef][Medline] [Order article via Infotrieve]
  50. Kosik, K. S., and Shimura, H. (2005) Biochim. Biophys. Acta 1739, 298-310[Medline] [Order article via Infotrieve]
  51. Delobel, P., Leroy, O., Hamdane, M., Sambo, A., Delacourte, A., and Buee, L. (2005) FEBS Lett. 579, 1-5[CrossRef][Medline] [Order article via Infotrieve]
  52. Brown, M. R., Bondada, V., Keller, J. N., Thorpe, J., and Geddes, J. W. (2005) J. Alzheimer's Dis. 7, 15-24[Medline] [Order article via Infotrieve]
  53. Nixon, R. A., Wegiel, J., Kumar, A., Yu, W. H., Peterhoff, C., Cataldo, A., and Cuervo, A. M. (2005) Neuropathol. Exp. Neurol. 64, 113-122[Medline] [Order article via Infotrieve]
  54. Li, L., von Bergen, M., Mandelkow, E. M., and Mandelkow, E. (2002) J. Biol. Chem. 277, 41390-41400[Abstract/Free Full Text]
  55. Chirita, C., Necula, M., and Kuret, J. (2004) Biochemistry 43, 2879-2887[CrossRef][Medline] [Order article via Infotrieve]
  56. Taniguchi, S., Suzuki, N., Masuda, M., Hisanaga, S., Iwatsubo, T., Goedert, M., and Hasegawa, M. (2005) J. Biol. Chem. 280, 7614-7623[Abstract/Free Full Text]
  57. Johnson, G. V., and Stoothoff, W. H. (2005) Biochim. Biophys. Acta 1739, 280-297[Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
A. C. LeBoeuf, S. F. Levy, M. Gaylord, A. Bhattacharya, A. K. Singh, M. A. Jordan, L. Wilson, and S. C. Feinstein
FTDP-17 Mutations in Tau Alter the Regulation of Microtubule Dynamics: AN "ALTERNATIVE CORE" MODEL FOR NORMAL AND PATHOLOGICAL TAU ACTION
J. Biol. Chem., December 26, 2008; 283(52): 36406 - 36415.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurosci.Home page
T. L. Spires-Jones, A. de Calignon, T. Matsui, C. Zehr, R. Pitstick, H.-Y. Wu, J. D. Osetek, P. B. Jones, B. J. Bacskai, M. B. Feany, et al.
In Vivo Imaging Reveals Dissociation between Caspase Activation and Acute Neuronal Death in Tangle-Bearing Neurons
J. Neurosci., January 23, 2008; 28(4): 862 - 867.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurosci.Home page
M.-M. Mocanu, A. Nissen, K. Eckermann, I. Khlistunova, J. Biernat, D. Drexler, O. Petrova, K. Schonig, H. Bujard, E. Mandelkow, et al.
The Potential for -Structure in the Repeat Domain of Tau Protein Determines Aggregation, Synaptic Decay, Neuronal Loss, and Coassembly with Endogenous Tau in Inducible Mouse Models of Tauopathy
J. Neurosci., January 16, 2008; 28(3): 737 - 748.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
K. Eckermann, M.-M. Mocanu, I. Khlistunova, J. Biernat, A. Nissen, A. Hofmann, K. Schonig, H. Bujard, A. Haemisch, E. Mandelkow, et al.
The beta-Propensity of Tau Determines Aggregation and Synaptic Loss in Inducible Mouse Models of Tauopathy
J. Biol. Chem., October 26, 2007; 282(43): 31755 - 31765.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
Y. P. Wang, J. Biernat, M. Pickhardt, E. Mandelkow, and E.-M. Mandelkow
Stepwise proteolysis liberates tau fragments that nucleate the Alzheimer-like aggregation of full-length tau in a neuronal cell model
PNAS, June 12, 2007; 104(24): 10252 - 10257.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
B. Bandyopadhyay, G. Li, H. Yin, and J. Kuret
Tau Aggregation and Toxicity in a Cell Culture Model of Tauopathy
J. Biol. Chem., June 1, 2007; 282(22): 16454 - 16464.
[Abstract] [Full Text] [PDF]

Home page
 ScienceHome page
E. D. Roberson, K. Scearce-Levie, J. J. Palop, F. Yan, I. H. Cheng, T. Wu, H. Gerstein, G.-Q. Yu, and L. Mucke
Reducing Endogenous Tau Ameliorates Amyloid {beta}-Induced Deficits in an Alzheimer's Disease Mouse Model
Science, May 4, 2007; 316(5825): 750 - 754.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurosci.Home page
K. N. Green, H. Martinez-Coria, H. Khashwji, E. B. Hall, K. A. Yurko-Mauro, L. Ellis, and F. M. LaFerla
Dietary Docosahexaenoic Acid and Docosapentaenoic Acid Ameliorate Amyloid-{beta} and Tau Pathology via a Mechanism Involving Presenilin 1 Levels
J. Neurosci., April 18, 2007; 27(16): 4385 - 4395.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
A. I. Iliev, S. Ganesan, G. Bunt, and F. S. Wouters
Removal of Pattern-breaking Sequences in Microtubule Binding Repeats Produces Instantaneous Tau Aggregation and Toxicity
J. Biol. Chem., December 1, 2006; 281(48): 37195 - 37204.
[Abstract] [Full Text] [PDF]

Home page
 JCBHome page
M. E. King, H.-M. Kan, P. W. Baas, A. Erisir, C. G. Glabe, and G. S. Bloom
Tau-dependent microtubule disassembly initiated by prefibrillar {beta}-amyloid
J. Cell Biol., November 20, 2006; 175(4): 541 - 546.
[Abstract] [Full Text] [PDF]

Home page
 ScienceHome page
E. D. Roberson and L. Mucke
100 Years and Counting: Prospects for Defeating Alzheimer's Disease
Science, November 3, 2006; 314(5800): 781 - 784.
[Abstract] [Full Text] [PDF]

Home page
 BrainHome page
H.-W. Klafki, M. Staufenbiel, J. Kornhuber, and J. Wiltfang
Therapeutic approaches to Alzheimer's disease
Brain, November 1, 2006; 129(11): 2840 - 2855.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
S.-i. Yamada, J.-i. Niwa, S. Ishigaki, M. Takahashi, T. Ito, J. Sone, M. Doyu, and G. Sobue
Archaeal Proteasomes Effectively Degrade Aggregation-prone Proteins and Reduce Cellular Toxicities in Mammalian Cells
J. Biol. Chem., August 18, 2006; 281(33): 23842 - 23851.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
281/2/1205    most recent
M507753200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Khlistunova, I.
Right arrow Articles by Mandelkow, E.-M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Khlistunova, I.
Right arrow Articles by Mandelkow, E.-M.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2006 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement