To study the direct effect of 3-APS on tau protein without the interference of other neuronal proteins, non neuronal cells, HEK 293, lacking tau were stably transfected with human tau cDNA. The tau-expressing cells were identified by immunofluorescence using an antibody raised against tau. Figure 1 indicates that upon tau expression the transfected cells show some microtubules bundles (see arrows and inset in Figure 1A), and that the actin-stress fibers dissappear. Moreover, some aggregates, where tau and actin colocalize, were found at the cell membrane (see below). These aggregates could be similar to those recently reported in Drosophila cells 24. In addition, the presence and quantitation of tau mRNA by RT-PCR, (Figure 1B) was determined, and the expression of tau protein was also analyzed by Western blot (Figure 1C).

Quantitation of tau protein expressed in these cells, indicated that tau was about 0.1 % of the total soluble cell protein. Since the content of tubulin in these cells was about 2 %, the ratio tau/tubulin was in line to that described for neuronal cells 25.

Increasing amounts of 3-APS were added to untransfected or transfected HEK 293 cells. Figure 2 indicates that upon addition of increasing amounts of 3-APS, a slight increase in tau staining was observed in tau transfected cells (see Figures 2A, 2C). This was not due to an increase in the amount of tau, as determined by Western blot (not shown), but probably, to an increase in tau aggregation in treated cells. Since tau aggregates are usually stained with Thioflavin S (Th-S) 26, Th-S staining was performed on the transfected cells after adding increasing amounts of 3-APS. Figure 2B also shows that there was a clear increase in Th-S staining dependent upon addition of increasing amounts of 3-APS (see also Figure 2D).

To confirm the increase in tau aggregates in 3-APS-treated cells, detergent insoluble aggregates were isolated and the presence of tau protein in those aggregates was determined. Western blot analysis revealed the presence of tau in the detergent insoluble fraction, and also showed that the amount of tau in the aggregates increased with the concentration of added 3-APS (Figure 3). When the same experiments, shown in Figures 2 and 3, were done in the presence of a sGAG, like heparin, instead of 3-APS, no differences were found with respect to controls (absence of heparin or 3-APS), when tau aggregation was tested. It can be explained by the fact that 3-APS, but not heparin, can enter into the cell.

To test if 3-APS directly facilitated the assembly of tau protein, purified recombinant tau protein was mixed with 3-APS and, after incubation of the mixture (in the conditions described in Materials and Methods), the aggregates were visualized under electron microscopy. Figure 4A shows the presence of short fibrillar polymers, upon addition of 3-APS; together with fewer but longer and larger fibrillar polymers (inset of the figure). No polymers were found in the absence of added tau. Some of the longer and larger 3-APS induced polymers were similar to those assembled in the presence of heparin (Figure 4B). When tau protein was mixed with both 3-APS and heparin both types of polymers were observed (Figure 4C). In addition, we studied the formation of these aggregates by immunofluorescence analysis 27, and we found that they could be identified by Th-S staining (Figure 4D). The result shown in Figure 4D suggests that the increased Th-S staining found upon 3-APS addition could be due to the presence of tau aggregates.

The previous observation suggests that 3-APS binds to tau. This binding could affect the interaction of tau with microtubules and could result in a disorganization of microtubule network. To test this possibility, immunofluorescence analyses to visualize tubulin and tau were carried out. No main differences were found upon addition of 3-APS in microtubule network. Moreover, the in vitro binding of tau to microtubules was studied in the absence or presence of 3-APS. Our results indicated that in the presence of 0.5 mM 3-APS there is a slight decrease of in vitro tau binding to microtubules (Additional File 1).

Although the presence of 3-APS does not affect to the organization of microtubule cytoskeleton, it seems to alter some aspects related to tau-actin aggregates. In Figure 1, it was shown that tau expression in HEK 293 cells resulted, in keeping with previously reported data 28, in the loss of actin stress fibers and in the presence of some aggregates in lamellipodia-like structures that are stained with phalloidin (aggregated actin) and tau antibodies (recently, in PC12 cells, tau has been localized to lamellipodia-like structures, where it associates with actin 29). These aggregates are shown in Figure 5A. Upon addition of 3-APS these aggregates disappeared (Figure 5B) suggesting that addition of 3-APS could result in the disasembly of tau-actin aggregates.

Since 3-APS is not affecting to the microtubule network and to the binding of tau to microtubules, it is possible that it does not bind directly to the microtubule binding sites, found in tau molecule that are needed for tau-tau interaction 4. To look for the binding site of 3-APS that results in tau aggregation we used different tau fragments (see Figure 6A). Figure 6B shows that tau aggregates were mainly found when tau 3RC peptide was mixed with 3-APS. Very few polymers were found when tau 3R peptide was tested and no polymers were observed for tau peptide containing the amino-terminal half region of tau protein. As can be seen also in the figure, larger polymers were found for the tau variant in which the N-terminal region was missing (tau 3RC) compared to the polymers found when the whole tau was assembled. These data suggest that 3-APS could bind to the C-terminal half region of tau protein, explaining the lack of interference of 3-APS addition with tau binding to microtubules, since tau may remain bound to microtubules and, also may bind to 3-APS through to a more distal C-terminal region. It should be indicated that tau binding to microtubules is mainly through the first and second tubulin binding repeats present in tau molecule 30.

It has been proposed that amyloid beta peptide fragment Aβ_1–28 binds to 3-APS 19. The sequence of this peptide being DAEFRHDSGYEVHHQKLVFFAEDVGSN K. The bold motif has been suggested to be the binding site for 3-APS 19. Thus, we have looked for a similar motif in tau molecule and it has been found a related motif NIHHK in tau 3RC peptide. Thus, we have analyzed the polymerization of a tau peptide containing the residues 317–335, being NIHHK the residues 327 to 331 present in that peptide. Figure 6B (inset) shows that the indicated tau peptide (317–335) is able to assemble in the presence of 3-APS. This experiment suggests a posible direct interaction between tau peptide (residues 317–335) and 3-APS.

To HEK 293 cells stably transfected with tau, increasing amounts of 3-APS were added, and the cell viability was measured. The data from Figure 7 show that, after four days of 3-APS addition, the number of surviving cells was similar at the different 3-APS concentrations tested in the cell cultures and in controls without 3-APS. In every cell culture, the presence of cells containing tau aggregates was calculated based on Th-S staining, as previously shown in Figure 2D for HEK 293 cells expressing tau protein.

After looking at the effect of 3-APS in tau-expressing non neuronal cells, we have tested the effect of 3-APS in human SH-SY5Y neuroblastoma cells. These neuroblastoma cells express a very little amount of tau protein in undifferentiated cells state. However, upon differentiation (see methods), a high increase of tau protein was found in those cells (Figure 8A). When undifferentiated neuroblastoma cells were treated with 3-APS, no differences in phalloidin staining (actin cytoskeleton) were found (Figure 8B). In fact, stress fibers were observed (probably due to the low expression of tau protein). In addition, a very weak Th-S staining was observed in the absence or presence of 3-APS. Upon differentiation of SH-SY5Y cells with dibutyril cyclic AMP, stress fibers disappear (Figure 8C) and very few actin-tau membrane complexes, compared to those found in HEK 293 cells stably transfected with tau, were observed. When these differentiated neuroblastoma cells were treated with 3-APS none of those actin-tau complexes were found (Figure 8C). Moreover, in 3-APS treated cells an increase in Th-S staining was observed (Figure 8D). Also, an increase in detergent insoluble tau aggregates was found upon treatment of the differentiated SH-SY5Y cells with 3-APS (Figure 8E).

In addition, the effect of 3-APS on primary cultures of mouse hipoccampal neurons was analyzed. Figure 8F shows and increase in Th-S staining in 3-APS treated neurons. The effect of 3-APS addition in neurons viability was also studied, but not changes in cell viability was found in the absence or presence of 3-APS (Figure 8G). Finally, no differences in morphology of hipocacmpal neurons were observed, upon 3-APS addition.

3-Amino-1-propanesulfonic acid (3-APS) (Ref. A4147), Phenylmethylsulfonyl fluoride (PMSF), EDTA, EGTA, 2-Mercaptoethanol, MES, HEPES, Trizma Base, NaF, Na₃VO₄ and SDS were purchased from Sigma. Triton X-100, Tween-20, NaCl and MgCl₂ were obtained from Merck. Acrylamide/bisacrylamide solution and Bradford reagent was supplied by Bio-Rad. Protran Nitrocellulose Transfer Membrane was from PerkinElmer. The chemiluminiscent detection kit (Western Light) was purchased from Tropix. DAPI (Ref. 268298) was obtained from Calbiochem. Thioflavin S (Th-S) (Ref. T-1892) was purchased from Sigma.

For immunoblot analysis, we used anti-β-tubulin (1/2000; Sigma), anti-β-actin (1/2500; Sigma), and 7.51(anti-tau antibody; a gift from Dr. C. M. Wischik, UK) antibodies followed by relevant secondary antibodies (1/2000; DAKO).

For immunofluorescence, we used anti-β-tubulin (1/500; Sigma), Tau-5 (1/500; Chemicon) and T14(1/500; Zymed Laboratories). The secondary antibodies (Molecular Probes) were used at 1/1000.

Actin staining was done by using phalloidin TRITC (1/300; Sigma). 0,01% Thioflavin S in PBS 1× was used for staining of treated or untreated cell cultures.

Total RNA from HEK 293 and HEK 293 tau expressing cells were prepared using the reagent TRIzol (Invitrogen) and following the supplier's protocol. Reverse transcription was performed using the first cDNA synthesis kit (Roche Applied Science) on 5 μg of RNA with oligo(dT) primers. PCR was performed with the primers for tau R1 (5'-GGCGAATTCGGATCCATGCCAGACCTGAAGAATG-3') and R2 (5'-GGCCTGC AGTTACTCGCGGAAGGTCAGCTTGTGGG-3'). The amplifications were performed basically with the following protocol: 30 cycles of 94°C for 45s, 55°C for 1 min, and 72°C for 1 min. The PCR products were resolved on a 1% agarose gel and stained with ethidium bromide. As a loading control, RT-PCR for actin was performed for each sample. The used primers for murine actin were H1 (5'-GCATGGAGTCCTG TGGCATCCACG-3') and H2 (5'-GGGTGTAACGCAACTAAGTCATAG-3'). Differences among groups were analysed by Student's unpaired t-test to determine significant differences between means.

HEK 293 and HEK 293 tau expressing cells were washed once with phosphate-buffered saline (PBS), placed on ice, and then homogenized in a buffer containing: 20 mM HEPES, pH 7.4; 100 mM sodium chloride (NaCl); 100 mM sodium fluoride (NaF); 1% Triton X-100; 1 mM sodium orthovanadate (Na₃VO₄); 5 mM EDTA; and the Complete protease inhibitor cocktail (Roche Diagnostics, Barcelona, Spain). After determination of the protein content via Bradford assay, samples containing the same amount of protein were mixed with electrophoresis buffer containing sodium dodecyl sulfate (SDS), boiled for 5 min, and separated by gel electrophoresis in the presence of SDS on 10% acrylamide gels. The proteins were then transferred to nitrocellulose membranes by following standard procedures, and the membranes were blocked with 10% nonfat dried milk in PBS, 0.2% Tween-20 (PBST). The blocked membranes were incubated overnight with primary antibodies diluted in blocking solution at 4°C. The membranes were then rinsed three times in PBST and incubated with the corresponding peroxidase-conjugated secondary antibody for 1 hr at room temperature. The immunoreactive proteins were visualized by using an enhanced chemiluminescence detection system (Amersham), and subsequent densitometric analysis was performed with an imaging densitometer (GS-710 model; Bio-Rad, Hercules, CA). Western blot analysis were also done using an antibody raised aginst actin (Sigma) as loading control. Protein extracts prepared from differetiated SH-SY5Y cells were processed and analysed by Western Blot using tau antibody 7.51 (1/100) as previously indicated.

Treated HEK 293 tau expressing cells, or SH-SY5Y neuroblastoma cells, were homogenized in RIPA buffer (50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 5 mM EDTA, 1% Nonidet P-40, 0.25% sodium deoxycholate, 1% SDS, 5 mM 4-(2-aminoethyl) benzene-sulfonyl fluoride hydrochloride, 1 μg/ml proteases inhibitors (Complete protease inhibitor cocktail (Roche Diagnostics, Barcelona, Spain)), 1 mM NaF, 1 mM Na₃VO₄, and 1 mM β-glycerophosphate) without thawing by using a polytron homogenizer (Kinematica, Kriens, Switzerland) at its highest speed for 30 s. Before centrifugation to fractionate the detergent soluble from detergent insoluble material, an aliquot from each sample was taken to determine if each sample contains, or not, a similar amount of actin (loading control). To do that we have used an antibody raised against actin. Tau protein aggregates from homogenates were isolated by centrifugation (4°C) for 20 min at 20000–30000 g (table-top centrifuge at 15000 rpm). One fourth of the supernatant volume, after centrifugation, was taken to characterize the protein present in that fraction. The whole aggregated protein present in the insoluble fraction was then diluted in O+ buffer (62.5 mM Tris-HCl pH 7.0, 10% (w/v) glycerol, 5% (v/v) β-mercaptoethanol, 2.3 % (w/v) SDS, 1 mM EDTA, 1 mM EGTA, 1 mM NaF, 1 mM Na₃VO₄ 1 mM PMSF (phenylmethyl-sulfonyl fluoride), 1 μg/ml proteases inhibitors (Complete protease inhibitor cocktail (Roche Diagnostics, Barcelona, Spain)), boiled for 5 min, and separated by gel electrophoresis in the presence of SDS on 10% acrylamide gels. Then, the protein content was analized by western blot as previously described.

After treatments, HEK 293 and HEK 293 tau expressing cells were fixed with either cooled methanol (-20°C) (tubulin immunofluorescence) or 4% paraformaldehyde (actin and tau immunofluorescence) for 20 min at 4°C or 37°C, respectively, and then washed with buffer A (0.1 M MES; 2 mM EGTA; 0.5 mM MgCl₂) or with phosphate buffered saline (PBS), respectively. Fixed cells were incubated with 1 M glycine 30 min then permeabilized with 0.2% Triton X-100 in PBS or buffer A for 5 min at room temperature. The cover slips were blocked with 1% BSA/PBS or Buffer A for 1 h at room temperature and incubated in primary antibodies in 1% BSA, in PBS, or in buffer A, for 1 h at room temperature. After washing three times with PBS or with buffer A, the secondary antibodies were incubated for 1 h, at room temperature. DAPI (1 μg/ml) staining was performed 10 minutes before finishing secondary antibody incubation. Finally, the covers lips were washed three times with PBS or with buffer A and once with H₂O, and mounted with FluorSave Reagent (Calbiochem).

Fluorescence microscopy was used to measure fluorescence intensity. After staining with 0.01 % Th-S and T14 antibody cell were observed on a Zeiss Axiovert200 fluorescent microscope. T14 antibody and thioflavin-S fluorescence images were captured through a 100× objetive on a high resolution CCD camera (SPOT RT Slider, Diagnostic). The images were saved for later analysis and quantitation. Camera exposure and light settings were keot constant during each experiment. The fluorescence intensity measurements were carried out using the image analysis software Metamorph 6.1 r6 (Universal Imaging).

A similar protocol was followed for SH-SY5Y neuroblastoma cells and primary cultures of hippocampal neurons.

For in vitro assembly of tau protein with microtubules, purified tubulin was assembled in the presence of taxol 42 and mixed with tubulin, after incubation of the mixture, the polymerized protein was isolated by centrifugation in Airfuge (Beckman) at room temperature, for 15 min at 100000 g. The protein present in supernatant and pellet was analyzed by gel electrophoresis followed by western blot using anti-tau and anti-tubulin antibodies.

Cell viability was assessed by calcein-propidium iodide uptake 43. Calcein/acetoxymethyl ester is taken up and cleaved by esterases present in living cells, yielding yellowish-green fluorescence. In contrast, propidium iodide is taken up only by dead cells, which then exhibit orange-red fluorescence. Briefly, cells were incubated for 30 min with 8 μM propidium iodide (Sigma) and 1 μM calcein/acetoxymethyl ester (Molecular Probes). The cultures were then rinsed once with Hanks balanced salt solution containing 2 mM CaCl₂, and the cells were visualized by fluorescence microscopy with a Zeiss Axiovert 135 microscope. Three randomly selected fields were analyzed per well (100–200 cells/field) in at least three independent experiments. Cell viability was expressed as the percentage of calcein-positive cells with respect to the total number of cells.

Recombinant human tau (whole molecule), N-terminal tau (residues 1–251), peptide 3RC (containing three tubulin binding motifs and the carboxyl-terminal region), peptide 3R (containing three tubulin binding motifs), have been synthesized and purified as previously reported 4. Tau peptide containing residues 317–335 was obtained as previously indicated (meter referencia 33).

Filaments were grown by vapor diffusion in hanging drops in the standard way used for protein crystallization as previously indicated 44. In a typical experiment, 0.5–2 μg of peptide was resuspended in 10–15 μL of buffer A (0.1 M MES (pH 6.4), 0.5 mM MgCl₂, and 2 mM EGTA) containing 50 mM NaCl and 3-APS at concentrations ranging from 1 to 4 mM. In other assays, a similar amount of tau (0.5–2 μg) but in the presence of 0.5 mg/mL heparin was resuspended in buffer A 4. The reservoir in this case contained 0.2 M NaCl in buffer A. Filaments were obtained after incubation for 4 days at 4°C. The samples we revisualized by electron microscopy as described 4. Electron micrographs were obtained at a magnification of 80000 on Kodak SO-163 film. Micrographs were digitized using an Eikonix IEEE-488 camera with a pixel size equivalent to 7 Å in the specimen plane. Processing and measurements were performed using the Digital micrograph 2.1 software from Gatan. Several standards were used for the control of the measurements. Alternatively, tau polymers were stained with Th-S and visualized by immunofluorescence 27.