Construction of mutants
Constructs of tau mutants (tauN, tauPRD, tauMTBD, tauC, tauΔMTBD and tauΔPRD&MTBD) were prepared by PCR or megaprimer PCR amplification  using human tau23 clones as templates. Primers used are listed in Additional file 1. They were then subcloned into the prokaryotic expression vector pET-28a(+) (Novagen, Germany) as NcoI and XhoI fragments. These five mutants each contained a His-tag at the C-terminus. The clones were transformed into E. coli BL21 (DE3) cells after their nucleotide sequences had been confirmed by sequencing.
Expression and purification of tau and its mutants
The tau mutants with His-tags were purified with Ni-NTA resin columns (QIAGEN, Holland) according to the manufacturer's instructions except that cell lysates were boiled for 5 min and centrifuged before loading on the resin. Protein samples were concentrated and then purified further with HiTrap desalting columns (Amersham Pharmacia Biotech, Switzerland). Each purified mutant exhibited a single protein band on Tris-Tricine gels (Additional file 6-a, b). Low molecular weight protein markers (SIBAS, Shanghai, China), mixed with aprotinin (MW 6,500), were used as molecular markers.
The shortest isoform of human tau (tau23) was purified with Q-Sepharose and SP-Sepharose chromatography (Amersham Pharmacia Biotech, Switzerland) as described by Goedert and Jakes . Protein concentrations were measured with BCA protein assay kits (Pierce, USA).
Western blotting of tau or its mutants
Tau mutants were run on Tris-Tricine gels. The bands on the gels were verified by staining with monoclonal anti-His antibodies (Novagen, Germany) (Additional file 1-b). Tau was run on a 12% SDS-PAGE (Additional file 6-a) and verified by staining with tau-13 antibodies (Santa Cruz, USA). All proteins and mutant peptides employed in this work showed single bands on SDS-PAGE gels or Western blots (Additional file 1).
Rabbit skeletal muscle global alpha actin was purified as described by Spudich and Watt . To improve purity, actin was assembled in a buffer containing a low ATP concentration (0.2 mM) during purification. Human platelet G-actin (a mixture of 85% beta and 15% gamma isoforms) was obtained from Cytoskeleton Inc. and was resuspended in buffer A (2 mM Tris-HCl pH 8.0, 0.2 mM ATP, 0.5 mM beta-mercaptoethanol, 0.2 mM CaCl2, and 0.005% NaN3). G-actin proteins were stored at -70°C after freezing in small volumes in liquid nitrogen. These samples were thawed rapidly with gentle agitation under running water at room temperature . Both skeletal muscle and platelet G-actin showed single bands on SDS-PAGE gels (Figure 2 and Additional file 6). To obtain F-actin, G-actin was polymerized in a buffer containing 100 mM KCl, 2 mM MgCl2 and 0.2 mM ATP at 25°C for about 90 min followed by centrifugation (80,000 g, 4°C, and 3 h). The pellet was resuspended in buffer F (100 mM KCl, 1 mM MgCl2, 0.1 mM CaCl2, 0.2 mM ATP, 1 mM Tris-HCl, pH 8.0, 1 mM NaN3).
Solid phase assays
Experimental conditions were as described by Farias and co-workers  with some modifications. The protein (skeletal muscle G-actin, platelet G-actin, skeletal muscle F-actin or platelet F-actin, 5 μg/ml, 50 μl, in 20 mM Tris pH 8.0 buffer) was coated on 96-well microtiter plates (Costar, USA), and incubated at 37°C for 2 h to allow adhesion to the polystyrene surface. After washing three times with PBST (PBS containing 0.2% Tween 20), the sites were saturated by incubation with 200 μl of blocking agent (PBS containing 5% non-fat milk) at 37°C for 30 min. Wells were then washed with binding buffer (50 mM Hepes containing 0.5 mM EGTA and 0.5 mM MgCl2, pH 7.5) and 50 μl of different concentrations of tau or its mutants were added. After incubation at 37°C for 45 min, wells were washed with PBST. Primary antibodies (tau-13, Santa Cruz, USA) for tau, and an anti-His-Tag monoclonal antibody (Novagen) for tau mutants were added and incubated at 37°C for 40 min. Plates were washed with PBST before the addition of the second antibody (1:1,000 dilution, 100 μl) labelled with horseradish peroxidase (HRP), and incubated at 37°C for 30 min. After the wells were washed with PBST, binding of actin with tau was detected using 100 μl of TMB buffer (6 μg/ml TMB, 0.045% H2O2, 0.1 M PB, pH 6.0) for 10 min, and the reaction was terminated with 50 μl of 2 M H2SO4. The binding of tau with actin was recorded by measuring the net change in absorbance at 450 nm by using an automatic solid phase assay plate reader (Thermo, USA). Correas and co-workers did not add excess ATP to the reaction in their study of the interaction of G-actin with tau. Similarly, Roger and coworkers  did not use excess ATP either in studying the bundling of tau with F-actin. Furthermore, Sattilaro and colleagues  have reported that formation of MAP-2-actin bundles is inhibited by millimolar concentrations of ATP. Thus, in this work, excess ATP was not used in the reaction of tau with G-actin or F-actin.
Tau or its mutants was incubated with G-actin or F-actin (from skeletal muscle or platelet) in binding buffer in the presence or absence of NaCl (50 - 500 mM) at 37°C for 40 min, and then centrifuged at either 25,000 g for 30 min or 100,000 g for 1 h at 4°C. The pellet was resuspended and then boiled for 5 min, followed by electrophoresis on SDS-PAGE gels.
Atomic force microscopy (AFM) analysis
G-actin, tau and its mutants were incubated in binding buffer at 37°C for 40 min, and G-actin alone and tau alone were used as controls. Samples were diluted 2 - 20 times with 20 mM Tris-HCl (pH 8.0). A 10 μl drop of the protein sample was deposited on freshly cleaved mica, allowed to stand for 5 min in air, and then washed with three 200 μl aliquots of buffer solution before drying for 4 min in a stream of nitrogen. Tapping mode AFM was performed using a Nanoscope IIIa Multimode-AFM (Veeco Instruments, USA) under ambient conditions. Silicon tips (TESP, Switzerland) with a resonance frequency of about 250 kHz were used at a scan rate of 1-2 Hz. Once the tip was engaged, the set point value was adjusted to minimize the force exerted on the sample while maintaining the sharpness of the image.
Tau (or its mutants) and F-actin were incubated in binding buffer in the presence or absence of NaCl (50 - 500 mM) at 37°C for 40 min. To observe the interaction of F-actin with tau or its mutants, F-actin alone was used as a control under the same conditions. Samples were placed on 300-mesh carbon-coated copper grids for 1 min, washed with H2O and negatively stained with 1% uranyl acetate for 1 min. The specimens were examined with a Tecnai 20 electron microscope (Philips, Holland).