Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 5, 9964

Tau Co-Organizes Dynamic Microtubule and Actin Networks


Tau Co-Organizes Dynamic Microtubule and Actin Networks

Auréliane Elie et al. Sci Rep.


The crosstalk between microtubules and actin is essential for cellular functions. However, mechanisms underlying the microtubule-actin organization by cross-linkers remain largely unexplored. Here, we report that tau, a neuronal microtubule-associated protein, binds to microtubules and actin simultaneously, promoting in vitro co-organization and coupled growth of both networks. By developing an original assay to visualize concomitant microtubule and actin assembly, we show that tau can induce guided polymerization of actin filaments along microtubule tracks and growth of single microtubules along actin filament bundles. Importantly, tau mediates microtubule-actin co-alignment without changing polymer growth properties. Mutagenesis studies further reveal that at least two of the four tau repeated motifs, primarily identified as tubulin-binding sites, are required to connect microtubules and actin. Tau thus represents a molecular linker between microtubule and actin networks, enabling a coordination of the two cytoskeletons that might be essential in various neuronal contexts.


Figure 1
Figure 1
Tau interacts with microtubules and F-actin, and forms hybrid bundles of both polymers. (a) Scheme of tau isoform used in this study. Tau projection domain contains a proline-rich region (P1). The C-terminal domain consists of a proline-rich region (P2), 4 repeat motifs (R1-R4) and a C-terminal extension comprising a pseudo-repeat motif (R’). (b) Tau binding affinity for microtubules. Increasing concentrations of microtubules were incubated with 0.5 μM tau in BRB80-K buffer. Percentage of bound tau was plotted against microtubule concentration and fitted with a hyperbolic function. The Kdapp value was 280 ± 52 nM (mean ± SD, five independent experiments). (c) Tau binding affinity for F-actin. Increasing concentrations of F-actin were incubated with 0.5 μM tau in AP buffer. Percentage of bound tau was plotted against F-actin concentration and fitted with a hyperbolic function. The Kdapp value was 241 ± 43 nM (mean ± SD, three independent experiments). (d) Time course of actin bundling (4 μM) with increasing tau concentrations, monitored by light scattering at 400 nm. a.u., arbitrary units. (e) Quantification of actin bundles by low speed sedimentation after 1 hour incubation with various tau concentrations. Data represent the mean ± SD (three independent experiments). (f) TIRF microscopy of actin (0.4 μM) polymerized with or without 0.4 μM tau. Time 00:00 (min:sec) indicates start of acquisition. Arrowheads, F-actin bundles; scale bar, 10 μm. (g) Tau induces microtubules/F-actin complexes. Tau (4 μM) was mixed with either 2 μM taxol-stabilized microtubules, 2 μM phalloidin-stabilized F-actin or both polymers before low-speed centrifugation. F-actin sedimented only with microtubules and tau. P, pellet; SN, supernatant; MT, microtubule. (h, i) Visualization of tau-induced complexes between phalloidin-stabilized F-actin (red) and taxol-stabilized microtubules (green). (h) Microtubules (0.3 μM) and F-actin (0.2 μM) were mixed for 30 minutes with or without tau (0.5 μM) before fixation and observation. Scale bar, 10 μm. (i) Microtubules (2 μM) and F-actin (4 μM) were incubated with increasing concentrations of tau before centrifugation at 50,000xg to sediment single/bundled microtubules but not F-actin alone. Scale bar, 10 μm.
Figure 2
Figure 2
Tau coordinates dynamic microtubules and actin filaments. (a) TIRF microscopy of co-polymerizing microtubules and F-actin. In the control (upper panel), 20 μM tubulin dimers (green) were co-assembled with 0.4 μM actin monomers (red). Most actin filaments were floating out of the TIRF field due to a lower medium viscosity compared to conditions used for actin assembly (see Methods). Addition of 1 μM fascin (middle panel) induces the apparition of a network of F-actin bundles randomly distributed around microtubules. Tau at 0.7 μM (lower panel) co-organizes growing microtubules and F-actin. In this last condition, we decreased tubulin concentration to 5 μM to avoid extensive microtubule self-assembly induced by tau. White stars show zippering events between microtubules and actin filaments, arrows point actin filaments growing along microtubules. Time 00:00 (min:sec) indicates start of acquisition. Scale bar, 10 μm. (b-e) Detailed views of typical microtubule/actin coupling events observed in the presence of tau: guided growth of F-actin along the microtubule wall (b), straightening of curved actin filaments by a growing extremity of a microtubule (c) and single microtubules elongating along actin bundles (d, e). White arrows and arrowheads indicate extremities of growing F-actin and microtubules, respectively. Time 00:00 (min:sec) indicates start of event. Scale bar, 5 μm. (f) Curvature of actin filaments co-assembled with tubulin and either fascin or tau. Boxes and whiskers represent 25–75 and 5–95 percentiles, respectively and lines within boxes indicate median values (3.89 and 0.094 degree.μm−1 with fascin and tau, respectively). Data were compared using the Mann-Whitney test (n = 346 and 354 actin filaments for fascin and tau, respectively). Quantification was performed on fixed time-point images displaying similar microtubule densities. (g) Percentage of microtubule surface co-aligned with F-actin as a function of microtubule network surface in the presence of tau or fascin (three representative curves). a.u., arbitrary units. (h, i) Kymographs of co-aligned growing microtubules and F-actin in the presence of tau. Schemes of microtubule/actin co-organization are drawn on the right. Horizontal bars, 10 μm; vertical bars, 3 min.
Figure 3
Figure 3
Tau microtubule-binding sites are required for microtubule/actin crosslinking. (a) Schematic representation of the different tau constructs used in this study. (b and c) Microtubule/F-actin cross-linking activity of tau proteins. Microtubules and F-actin were co-assembled from tubulin dimers and actin monomers in the presence of the various tau proteins and analyzed by TIRF microscopy (see Methods for details). (b) Quantification of the percentage of microtubule surface co-localizing with F-actin from TIRF images at fixed time points exhibiting similar microtubule densities (see examples in (c)). Fascin that does not bind to microtubules serves as a negative control. For statistical analysis, we used a non-parametric Kruskal-Wallis ANOVA test followed by post-hoc Dunn comparisons. P values calculated in comparison to the 4R-tau conditions are 0.0626 (1R-tau), 0.0009 (0R-tau) and 0.0286 (Fascin). Error bars represent standard deviations (three to six independent experiments). (c) Examples of fixed time point images used to quantify the percentage of microtubule surface co-aligned with F-actin in various conditions. Images were taken at similar microtubule densities. The masked images (lower panel) represent the microtubule area that co-localizes with actin. Scale bar, 10 μm.
Figure 4
Figure 4
Model for tau-mediated coordination of microtubule and actin cytoskeletons. Tau promotes the co-alignment and coupled growth of microtubules (green) and actin filaments (red). We suggest that one tau molecule would bind both actin and microtubules simultaneously and that the distribution of tubulin-binding repeated sites of tau would be required to establish the structural link between the two cytoskeletal polymers (inset). At least two repeats are necessary for the crosslinking activity of tau, with one repeat interacting with microtubules and the other one with actin filaments.

Similar articles

See all similar articles

Cited by 30 PubMed Central articles

See all "Cited by" articles


    1. Rodriguez O. C. et al. Conserved microtubule – actin interactions in cell movement and morphogenesis. Nat. Cell. Biol. 5, 599–609 (2003). - PubMed
    1. Hoogenraad C. C. & Bradke F. Control of neuronal polarity and plasticity–a renaissance for microtubules? Trends Cell Biol. 19, 669–76 (2009). - PubMed
    1. Geraldo S. & Gordon-Weeks P. R. Cytoskeletal dynamics in growth-cone steering. J. Cell Sci. 122, 3595–604 (2009). - PMC - PubMed
    1. Rochlin M. W., Dailey M. E. & Bridgman P. C. Polymerizing microtubules activate site-directed F-actin assembly in nerve growth cones. Mol. Biol. Cell. 10, 2309–2327 (1999). - PMC - PubMed
    1. Dent E. W. & Kalil K. Axon Branching Requires Interactions between Dynamic Microtubules and Actin Filaments. J. Neurosci. 21, 9757–9769 (2001). - PMC - PubMed

Publication types