Tau is an intrinsically disordered protein, which diffuses on microtubules1. In neurodegenerative diseases, collectively termed tauopathies, malfunction of tau and its detachment from axonal microtubules are correlated with axonal degeneration2. Tau can protect microtubules from microtubule-degrading enzymes such as katanin3. However, how tau carries out this regulatory function is still unclear. Here, using in vitro reconstitution, we show that tau molecules on microtubules cooperatively form cohesive islands that are kinetically distinct from tau molecules that individually diffuse on microtubules. Dependent on the tau concentration in solution, the islands reversibly grow or shrink by addition or release of tau molecules at their boundaries. Shielding microtubules from kinesin-1 motors and katanin, the islands exhibit regulatory qualities distinct from a comparably dense layer of diffusible tau. Superprocessive kinesin-8 motors penetrate the islands and cause their disassembly. Our results reveal a microtubule-dependent phase of tau that constitutes an adaptable protective layer on the microtubule surface. We anticipate that other intrinsically disordered axonal proteins display a similar cooperative behaviour and potentially compete with tau in regulating access to the microtubule surface.
Subscribe to Journal
Get full journal access for 1 year
only $18.75 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Source data for Figs. 1–5 and quantifications given in the main text (island density) have been provided as Supplementary Table 1. Example raw videos are available at BioStudies with accession number S-BSST266. All other data supporting the findings of this study are available from the corresponding authors on reasonable request.
Code used to determine single-molecule intensities (GNU General Public License v.3, for further information see repository) is available at https://doi.org/10.5281/zenodo.3270568. Code used to create kymographs (MIT license, for further information see repository) is available at https://doi.org/10.5281/zenodo.3270572. All other custom written code is available from the corresponding authors on request.
Hinrichs, M. H. et al. Tau protein diffuses along the microtubule lattice. J. Biol. Chem. 287, 38559–38568 (2012).
Kneynsberg, A., Combs, B., Christensen, K., Morfini, G. & Kanaan, N. M. Axonal degeneration in tauopathies: disease relevance and underlying mechanisms. Front. Neurosci. 11, 572 (2017).
Qiang, L. Tau protects microtubules in the axon from severing by katanin. J. Neurosci. 26, 3120–3129 (2006).
Morris, M., Maeda, S., Vossel, K. & Mucke, L. The many faces of tau. Neuron 70, 410–426 (2011).
Gao, Y.-L. et al. Tau in neurodegenerative disease. Ann. Transl. Med. 6, 175–175 (2018).
Iqbal, K., Liu, F. & Gong, C.-X. Tau and neurodegenerative disease: the story so far. Nat. Rev. Neurol. 12, 15–27 (2016).
Drechsel, D. N., Hyman, A. A., Cobb, M. H. & Kirschner, M. W. Modulation of the dynamic instability of tubulin assembly by the microtubule-associated protein tau. Mol. Biol. Cell 3, 1141–1154 (1992).
Chaudhary, A. R., Berger, F., Berger, C. L. & Hendricks, A. G. Tau directs intracellular trafficking by regulating the forces exerted by kinesin and dynein teams. Traffic 19, 111–121 (2018).
Dixit, R., Ross, J. L., Goldman, Y. E. & Holzbaur, E. L. F. Differential regulation of dynein and kinesin motor proteins by Tau. Science 319, 1086–1089 (2008).
Vershinin, M., Carter, B. C., Razafsky, D. S., King, S. J. & Gross, S. P. Multiple-motor based transport and its regulation by Tau. Proc. Natl Acad. Sci. USA 104, 87–92 (2007).
Seitz, A. et al. Single-molecule investigation of the interference between kinesin, tau and MAP2c. EMBO J. 21, 4896–4905 (2002).
Trinczek, B., Ebneth, A., Mandelkow, E. M. & Mandelkow, E. Tau regulates the attachment/detachment but not the speed of motors in microtubule-dependent transport of single vesicles and organelles. J. Cell Sci. 112, 2355–2367 (1999).
Ebneth, A. et al. Overexpression of tau protein inhibits kinesin-dependent trafficking of vesicles, mitochondria, and endoplasmic reticulum: implications for Alzheimer’s disease. J. Cell Biol. 143, 777–794 (1998).
Gamblin, T. C., Berry, R. W. & Binder, L. I. Tau polymerization: role of the amino terminus. Biochemistry 42, 2252–2257 (2003).
Tan, R. et al. Microtubules gate tau condensation to spatially regulate microtubule functions. Nat. Cell Biol. https://doi.org/10.1038/s41556-019-0375-5 (2019).
Makrides, V., Massie, M. R., Feinstein, S. C. & Lew, J. Evidence for two distinct binding sites for tau on microtubules. Proc. Natl Acad. Sci. USA 101, 6746–6751 (2004).
Wegmann, S. et al. Tau protein liquid–liquid phase separation can initiate tau aggregation. EMBO J. 37, e98049 (2018).
Schneider, R., Korten, T., Walter, W. J. & Diez, S. Kinesin-1 motors can circumvent permanent roadblocks by side-shifting to neighboring protofilaments. Biophys. J. 108, 2249–2257 (2015).
Telley, I. A., Bieling, P. & Surrey, T. Obstacles on the microtubule reduce the processivity of Kinesin-1 in a minimal in vitro system and in cell extract. Biophys. J. 96, 3341–3353 (2009).
Jiang, K. et al. Microtubule minus-end regulation at spindle poles by an ASPM–katanin complex. Nat. Cell Biol. 19, 480–492 (2017).
Varga, V., Leduc, C., Bormuth, V., Diez, S. & Howard, J. Kinesin-8 motors act cooperatively to mediate length-dependent microtubule depolymerization. Cell 138, 1174–1183 (2009).
Leduc, C. et al. Molecular crowding creates traffic jams of kinesin motors on microtubules. Proc. Natl Acad. Sci. USA 109, 6100–6105 (2012).
Kellogg, E. H. et al. Near-atomic model of microtubule–tau interactions. Science 360, 1242–1246 (2018).
McVicker, D. P., Hoeprich, G. J., Thompson, A. R. & Berger, C. L. Tau interconverts between diffusive and stable populations on the microtubule surface in an isoform and lattice specific manner. Cytoskeleton 71, 184–194 (2014).
Hernández-Vega, A. et al. Local nucleation of microtubule bundles through tubulin concentration into a condensed Tau phase. Cell Rep. 20, 2304–2312 (2017).
Bechstedt, S. & Brouhard, G. J. Doublecortin recognizes the 13-protofilament microtubule cooperatively and tracks microtubule ends. Dev. Cell 23, 181–192 (2012).
Sing, C. E., Olvera de la Cruz, M. & Marko, J. F. Multiple-binding-site mechanism explains concentration-dependent unbinding rates of DNA-binding proteins. Nucleic Acids Res. 42, 3783–3791 (2014).
Lansky, Z. et al. Diffusible crosslinkers generate directed forces in microtubule networks. Cell 160, 1159–1168 (2015).
Samsonov, A., Yu, J.-Z., Rasenick, M. & Popov, S. V. Tau interaction with microtubules in vivo. J. Cell Sci. 117, 6129–6141 (2004).
Bechstedt, S., Lu, K. & Brouhard, G. J. Doublecortin recognizes the longitudinal curvature of the microtubule end and lattice. Curr. Biol. 24, 2366–2375 (2014).
Monroy, B. Y. et al. Competition between microtubule-associated proteins directs motor transport. Nat. Commun. 9, 1714 (2018).
Mitra, A., Ruhnow, F., Girardo, S. & Diez, S. Directionally biased sidestepping of Kip3/kinesin-8 is regulated by ATP waiting time and motor-microtubule interaction strength. Proc. Natl Acad. Sci. USA 115, E7950–E7959 (2018).
Nitzsche, B. et al. Studying kinesin motors by optical 3D-nanometry in gliding motility assays. Methods Cell Biol. 95, 247–271 (2010).
Braun, M. et al. Adaptive braking by Ase1 prevents overlapping microtubules from sliding completely apart. Nat. Cell Biol. 13, 1259–1264 (2011).
Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676–682 (2012).
Ruhnow, F., Zwicker, D. & Diez, S. Tracking single particles and elongated filaments with nanometer precision. Biophys. J. 100, 2820–2828 (2011).
We thank A. Akhmanova and K. Jiang for the generous gift of the katanin plasmid, R. McKenney for feedback and sharing of data, V. Henrichs, I. Zhernov and L. Grycova for help with protein preparation, and Y. Bobrova, S. Dijkstra and C. Bräuer for technical support. We acknowledge the financial support from the Czech Science Foundation (grant no. 18-08304S to Z.L. and 17-12496Y to M.B.), the Introduction of New Research Methods to BIOCEV (CZ.1.05/2.1.00/19.0390) project from the ERDF, the institutional support from the CAS (RVO: 86652036) and the Imaging Methods Core Facility at BIOCEV, an institution supported by the Czech-BioImaging large RI projects (LM2015062 and CZ.02.1.01/0.0/0.0/16_013/0001775, funded by MEYS CR) for their support in obtaining imaging data presented in this paper.
The authors declare no competing interests.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Integrated supplementary information
(A) Distribution of time between the addition of tau-mEGFP and the formation of the islands (3 experiments per condition, n = 610 nucleation events). Bars show the median; error bars show the minimum and the maximum value. Same experiments as presented in (B) and (D). (B) Histograms of island growth velocities at different tau concentrations in solution (Methods, n = 2131 velocity traces). Same experiments and data representation as in (A). (C) Distribution of the lengths of islands 5 and 75 minutes after the addition of 40 nM tau-mEGFP, n = 91 microtubules in 3 experiments. Distribution of microtubule lengths is shown in the inset. Same experiments as presented in (E). (D) Fraction of microtubule length covered by tau islands at different concentrations of tau-mEGFP in solution. Solid lines and shaded areas (shown also in Fig. 1e) represent the median, minimum and maximum values calculated over the whole fields of view (n = 3 experiments). Boxplots represent the same statistics calculated over individual microtubules (n = 40 microtubules for 20 nM tau, n = 32 microtubules for 40 nM tau and n = 59 microtubules for 100 nM tau). The same data as presented in Fig. 1e, same experiments as presented in (A). (E) Island assembly does not cease even 75 minutes after the addition of 20 nM tau. The same data representation as in (D), same experiments as presented in (C). (F) Relative difference in tau density within islands i) just after the island nucleation upon the addition of tau and 30 seconds later and ii) just after the removal of tau and 30 seconds later (n = 91 microtubules in 16 experiments). Bleaching was not negligible in the case where tau was absent from solution – for a precise estimation of tau unbinding see (G). (G) Exemplary time trace of tau-mCherry density inside and tau-mEGFP density outside the islands after removal of tau (mCherry- or mEGFP-labelled, respectively) from solution, analogous to the results presented in Fig. 1h (n = 6 microtubules). Single exponential fits are indicated by solid lines. This experiment had been repeated 4 and 3 times for islands and surroundings, respectively, with similar results. Time constants derived from fits for all experiments are shown in Fig. 2b and Supplementary Fig. 2d. (H) Distribution of island disassembly velocities upon removal of tau from solution yielding the average velocity of 6.6 ± 5.2 nm/s (average ± SD, corresponding to 2.9 ± 2.3 tau molecules unbinding per second, Methods). Colors encode experiments (n = 335 velocity traces on 90 islands in 4 experiments). For a description of all box-plot elements see Methods. (I) Frequency of fissions occurring within islands upon removal of tau from solution yielding the average 0.20 ± 0.14 fissions per micron of island length per hour (average ± SD). Colors encode n = 4 experiments - same as in (H). (J) Schematics showing the tau constructs used in this study.
(A) Exemplary time trace of normalized tau-mCherry and tau-mEGFP density inside and outside the islands after exchange of 20 nM tau-mCherry for 20 nM tau-mEGFP (n = 5 microtubules). Photo-bleaching during this experiment was negligible (Methods). The dotted line indicates that the sample was out of focus at the time. (B) Exemplary time trace and fit of tau-mCherry density inside and outside the islands after exchange of 20 nM tau-mCherry for 20 nM tau-mEGFP (n = 5 respectively 7 microtubules). Photo-bleaching during these experiments was negligible (Methods). This experiment had been repeated 5 and 2 times for islands and surroundings, respectively, with similar results. Time constants derived from fits for all experiments are shown in Fig. 2b and Supplementary Fig. 2d. (C) Analogous estimation of tau residence times as in (B) for an exemplary experiment in which 20 nM tau-mCherry was exchanged for 100 nM tau-mEGFP (n = 5 microtubules). The photo-bleaching during this experiment was negligible (Methods). This experiment had been repeated 4 and 2 times for islands and surroundings, respectively, with similar results. Time constants derived from fits for all experiments are shown in Fig. 2b and Supplementary Fig. 2d. (D) Boxplots of dwell times of tau outside the islands at different tau concentrations in solution (at 0 nM tau, dwell time was 1.9 ± 0.3 s, average ± SD, n = 20 microtubules in 3 experiments, at 20 nM tau, the dwell time was 2.7 ± 1.3 s, average ± SD, n = 11 microtubules in 2 experiments, for 100 nM tau, the dwell time was 1.2 ± 0.7 s, average ± SD, n = 10 microtubules in 2 experiments; for dwell times within islands see Fig. 2b). Typical fits and time-traces underlying these data are shown in (B) and (C) and in Supplementary Fig. 1g. The data points are colour-coded by experiments. For a description of all box-plot elements see Methods. (E) Histogram of fluorescence intensities of single tau-mEGFP particles bound to microtubules in experiments as presented in Fig. 2c, showing that tau-mEGFP is associated with the microtubule in monomeric form (n = 1861 molecules inside the islands, n = 2210 molecules outside the islands, 1 experiment). (F) Mean square displacement over time of single tau-EGFP molecules inside and outside the islands yielding tau diffusion constants of 0.27 ± 0.15 µm2s-1 (linear fit coefficient ± 95% confidence bounds, n = 3660 molecules, 3 experiments) outside the islands and 0.027 ± 0.016 µm2s-1 (linear fit coefficient ± 95% confidence bounds, n = 2558 molecules, 3 experiments) inside the islands. For comparison, the analogously estimated diffusion constant of non-motile kinesin-1 molecules tightly bound to the microtubule in the presence of AMP-PNP (in the absence of ATP) was 0.014 ± 0.016 µm2s-1 (linear fit coefficient ± 95% confidence bounds, n = 126 molecules, 1 experiment).
(A) Exemplary time-traces of tau-mEGFP density outside and inside the islands during subsequent cycles of i) addition of increasing concentrations of tau-mEGFP followed by ii) removal of tau-mEGFP from solution. Experiment such as presented in Fig. 3c. This experiment had been repeated 11 times with similar results. (B) The density of tau-mEGFP on the microtubules saturates at high (µM) tau-mEGFP concentrations (total of 29 and 27 experiments for islands and surroundings, respectively). Horizontal lines indicate the three quartiles. (C) At tau-mEGFP concentrations in the range of 20 nM to approximately 1 µM, the tau-mEGFP density inside the islands scales linearly with the tau-mEGFP density outside the islands. Points indicate medians, error bars indicate the first and third quartiles (both axes). Data from the experiments presented in (B).
Velocities (A) and dwell times (B) of kinesin-1 molecules moving processively outside the islands formed at various tau-mCherry concentrations in solution (3 experiments at 0 nM, 6 experiments at 20 nM, 5 experiments at 40 nM, 3 experiments at 100 nM; numbers of molecules shown in box; no processive movement was observed inside the islands). (C) Kinesin-1 landing rates inside and outside the islands formed at various tau-mCherry concentrations in solution (7 experiments at 0 nM, 10 experiments at 20 nM, 7 experiments at 40 nM, 6 experiments at 100 nM). (D) Exemplary time-traces of normalized tubulin signal of the regions surrounding the islands (blue line) decaying during its katanin-mediated disassembly. Single exponential fits to the data (black lines) yield a tubulin residence time of 17 ± 6.4 s (average ± S.D., n = 17 microtubules in 2 experiments) in presence of 20 nM tau-mCherry and 34 ± 23 s (average ± S.D., n = 18 microtubules in 2 experiments) in presence of 800 nM tau-mCherry. Photo-bleaching during this experiment was negligible (Methods). (E) Box plot of katanin-mediated disassembly velocities of stretches of microtubules covered by tau islands (evaluated per island boundary) and the rate of katanin-generated cuts occurring within them. Experiments had been conducted at two different concentrations of tau in solution (3 experiments per condition, n = 56 microtubules at 20 nM tau and 33 at 800 nM tau). (F) Multichannel fluorescence kymograph showing the motion of Kip3–GFP (green) inside and outside the tau islands (red). Red arrow indicates the accumulation of Kip3–GFP in front of the island. White arrows indicate Kip3–GFP molecules speeding up as they leave the island. Scale bar 2 µm. This experiment had been repeated 7 times with similar results. (G) Quantification of velocities of single Kip3–GFP motors moving inside and outside the islands at 10 nM tau-mCherry in solution (n = 136 Kip3–GFP molecules inside islands, n = 184 outside, 3 experiments). (H) Quantification of Kip3–GFP density inside and outside the islands 100 s after the addition of Kip3–GFP (n = 98 microtubules in 6 experiments). The presented Kip3–GFP densities are normalized to the median of the density in the surroundings (within the same experiment). (I) Velocities of Kip3–GFP-driven disassembly of tau islands established at two different tau-mCherry concentrations (3 experiments per condition, n = 46 islands at 20 nM, n = 52 islands at 100 nM). For a description of all box-plot elements see Methods.
Supplementary Fig. 5. (A) Tau-mCherry islands on microtubules after the removal of 800 nM tau-mCherry from solution (upper panel); the same microtubule in presence of re-introduced 800 nM tau-mCherry in solution (middle panel); the same microtubule in presence of 200 nM katanin-GFP and 800 nM tau-mCherry. This experiment had been repeated 4 times with similar results. (B) Densities of tau-mCherry outside the islands, inside the islands and in the regions of high curvature (after the removal of 0.8 µM tau from solution). Points are colour-coded by experiments (n = 15 high-curvature regions in 4 experiments), horizontal lines indicate the three quartiles, weighted such that each experiment has equal weight. (C) Tau density profile along the microtubule after removing 800 nM tau from solution. X-axis is centred on the point of highest curvature. Data are colour-coded by microtubule and the density is normalized to the 90th percentile of the density-values of the respective microtubule. The red line represents the median of n = 30 microtubules (7 experiments), the first and third quartile are indicated by the shaded area. At 800 nM tau, there always were islands adjacent to microtubule bends, which is why the tau density to the left and right of the curved microtubule region is high even though tau has been removed from solution. A clear decrease in the tau density is apparent at the point of highest curvature. (D) Probability of severing of highly curved microtubule regions and straight island-covered microtubule regions, within 150 s after the addition of katanin. The bars represent the probability averaged over n = 4 experiments (29 bends and 82 straight microtubules), error bars represent the S.D. Curved microtubule regions were always cut. (E) Fluorescence micrograph showing the tau-mEGFP signal on microtubules at 20 nM tau-mEGFP in solution. Tau islands have higher tau-mEGFP densities than the tau-mEGFP regions localizing to microtubule regions with high curvature. See quantification and number of experiments in (H). (F) Kymograph showing the uniform increase of tau-mEGFP density along the whole region of the microtubule curve upon the addition of tau-mEGFP in solution. This is in strong contrast to islands localized on straight microtubules, which grew at their boundaries. See quantification and number of experiments in (G). (G) Exemplary time trace of tau density in the regions of high curvature after addition of 20 nM tau-mEGFP in solution (n = 5 microtubules, thick line and shaded area indicate the three quartiles). In contrast to the islands, in the high-curvature regions the tau density increases immediately after the addition of tau-mEGFP in solution (similar to the behaviour in the regions surrounding the islands). We observed similar time scales in a total of n = 20 high-curvature regions in 5 experiments (6 ± 2 s for high-curvature regions, 8 ± 6 s for surroundings, weighted average ± S.D., each experiment has equal weight). (H) Densities of tau-mEGFP outside the islands, inside the islands and in the regions of high curvature at 20 nM tau-mEGFP in solution (n = 11 high-curvature regions in 5 experiments). Points are colour-coded by experiments, horizontal lines indicate the three quartiles, weighted such that each experiment has equal weight. Scale bars, vertical 10 min, horizontal 2 µm.
Supplementary Figures 1–5, Supplementary Table title and legend and Supplementary Video titles and legends.
Experiment statistics source data.
Tau island formation.
Tau island disassembly.
Tau turnover inside and outside of islands.
Tau islands shield microtubules from severing enzymes.
Disassembly of islands by superprocessive kinesins.
Tau cohesion within islands, not the mere density of tau, shields microtubules from being severed.
Preferential binding of tau to curved microtubules is distinct from island formation.
About this article
Nature Cell Biology (2019)
Trends in Cell Biology (2019)