Tau hyperphosphorylation results in more diffuse microtubule interactions and reduces tau envelope formation in vitro.

A) Structure of 4R0N tau highlighting 14 disease-associated S/T residues mutated to E in pseudo-hyperphosphorylated (E14) tau or to A in phospho-resistant (AP) tau (Hoover et al., 2010). B) Schematic illustrating the main steps involved in preparing tau-GFP containing cell extracts and performing reconstitution assays with 500 nM tau (see Table 1). Linescans of tau intensity along a microtubule were analyzed using a Gaussian Mixture Model (GMM) to define intensity thresholds distinguishing envelopes from gaps. C) Representative images of WT, AP, and E14 tau-GFP on taxol-stabilized microtubules (WT: n=183, AP: n=170, E14: n=170). The indicated n values represent total number of samples over 3–4 replicates. Below the images, corresponding tau intensity plots along microtubules are shown, with horizontal lines indicating the GMM threshold (orange) and mean background fluorescence intensity (blue). Asterisks mark peaks identified as envelopes. Histograms show the distributions of tau fluorescence intensity fitted with a GMM. Insets show Bayesian Information Criterion (BIC) analyses used to determine whether a unimodal or multimodal distribution best describes the data. D and E) Plots quantifying the effects of hyperphosphorylation on D) tau envelope intensity and E) percentage of microtubule length covered by envelopes. Error bars indicate 95% CI. Statistical significance was assessed using Student’s t-test (*** p < 0.0001). Scale bars are 10 µm.

Tau hyperphosphorylation reduces the formation of envelopes in live neurons.

A) Representative images of axons with tau envelopes present in iPSC-derived MAPT-KO neurons expressing WT (n=39), AP (n=39), or E14 (n=28) tau-GFP. Each image is a max projection of 10 frames from live timelapse imaging. Images are all oriented so that the soma is towards the left and the distal axon is towards the right. Below, corresponding tau intensity plots from discrete locations (labeled 1–6) are shown. Horizontal lines indicating the GMM threshold (orange) and mean background fluorescence intensity (blue). Histograms of tau fluorescence intensity, fitted with a GMM to define the envelope threshold, are displayed for selected areas (labeled i–iii) marked by magenta ROIs. Insets show BIC analyses for multimodal distributions. B–C) Bar plots comparing B) tau envelope intensity and C) the mean envelope width of WT, AP, and E14 tau. D–E) Bar plots quantifying tau envelope frequency within D) discrete locations and E) the percentage of total length of axons imaged containing tau envelopes. Error bars indicate 95% CI. F) Timelapse images of fluorescence recovery after photobleaching assays of MAPT-KO neurons expressing WT(n=20), AP(n=21), or E14 (n=20) tau-GFP. Circles indicate the bleached regions. Note that tau envelopes are not clearly evident due to shorter exposure times needed for FRAP experiments and due to selection of higher tau-GFP expressing cells to ensure measurable recovery dynamics. G) Fluorescence recovery curves of WT (blue), AP (green), and E14 tau (red). Shaded regions indicate SD. The characteristic recovery (1) and mobile fraction (M) are indicated on the plot for each tau construct. H) Schematic illustrates how tau phosphorylation influences its dissociation from microtubules. Hyperphosphorylated E14 tau dissociates more readily than WT tau or the phospho-resistant AP tau, which remain more stably bound (Fig S2). I) Plots show how the ratio of tau signal in axons of neurons expressing WT (n= 37), AP (n=32), and E14 (n=36) tau over background intensity varies for each tau construct across the proximal, mid, and distal axonal regions. In the bottom plot, blue lines indicate a decrease in the signal from the proximal towards the distal axon, and red lines indicate an increase in tau signal towards the distal axon. The top plot shows the means and grey bars represent SD. Orange bars show means. Wilcoxon signed rank test was used to determine the pairwise comparison of tau intensity between each axonal region as shown and the p-values are indicated in the above inset. The scale bars are 10 µm (A) and 5 µm (F). (*p < 0.05, **p < 0.001, ***p < 0.0001).

Tau phosphorylation differentially regulates kinesin-1 and kinesin-3 motility.

A and C) Kymographs of constitutively active Janelia Fluor 554 (JFX554)-labelled A) KIF5C(1–560)-JFX554 and C) KIF1A(1–393)-JFX554 motors moving along taxol-stabilized microtubules incubated with mock cell lysate or lysates containing 500 nM WT, AP, or E14 tau-GFP (magenta). KIF5C (mock: n=1016, WT: n=792, AP: n= 1078, E14: n=1203) KIF1A (mock: n=864, WT: n=173, AP: n=218, E14: n=305) over 3–4 replicates. B and D) Bar graphs show the fraction of time that kinesin motors are paused or exhibit processive motility. Error bars indicate 95% CI. E–F) Bar graph shows the bootstrapped mean run length and mean velocity of E) KIF5C and F) KIF1A ± WT, AP, E14 tau-GFP. Error bars indicate SEM. G) Schematic illustrating the approach used to assess the impact of tau on kinesin-microtubule attachment and detachment. Motors were observed attaching to or detaching from microtubules either within tau envelopes or outside of them. H) Paired sample plots show the attachment and detachment frequencies of KIF5C (top) and KIF1A (bottom). Red lines indicate increased frequency inside envelopes compared to outside and blue lines indicate decreased frequencies inside of envelopes. Insets for KIF1A show a zoomed-in view of the attachment and detachment frequencies. Blue bars indicate 95% CI, orange lines mark mean values, and grey bars denote SD. I) Schematic illustrating the impact of tau on the detachment kinetics of kinesin motors. KIF5C is weakly inhibited by WT and phospho-resistive tau compared to hyperphosphorylated tau that has a similar dissociation rate compared to KIF5C in mock conditions. Conversely, KIF1A is more strongly inhibited by tau hyperphosphorylation and dissociates at a faster rate compared to WT and AP tau. (*p < 0.05, **p < 0.001, ***p < 0.0001). Horizontal scale bars are 5 µm, vertical scale bars are 5 sec.

Tau perturbations impact the trafficking of lysosomes in neurons.

A) Schematic of a neuron indicating the regions where tau’s effects on lysosome transport were assessed. Lysosome transport was recorded in proximal, mid, and distal axonal regions. Example images of WT tau-GFP expression are shown from each region. Proximal regions were defined as ∼50 µm from the soma, mid-axonal regions as ∼halfway along the axon, and distal regions as ∼50 µm from the axon terminal. Images are all oriented so that the soma is towards the left and the distal axon is towards the right. The number of cells analyzed under each condition are indicated in Table 2. B) Images show mid-axonal tau-GFP signal and max projections of lysosomes (lys) in control, MAPT-KO, and MAPT-KO neurons expressing WT, AP, and E14 tau-GFP. Below kymographs show anterograde and retrograde lysosome transport for each condition. C) Bar plot shows the fraction of anterograde and retrograde long-distance trajectories for control (CTL), MAPT-KO (KO), and MAPT-KO neurons expressing WT, AP, or E14 tau-GFP. Above the plot, asterisks represent statistical significance tested for anterograde (black) and retrograde (red) trajectories (* p < 0.05, ** p < 0.001, *** p < 0.0001). D and E) Plots show the D) mean absolute displacements and the E) frequency of lysosomes in the proximal, mid, and distal axon for each condition. Blue bars indicate 95% CI, orange lines mark mean values, and grey bars denote SD. Statistical significance is indicated by asterisks above the at the top of plots (*p < 0.05, **p < 0.001, ***p < 0.0001). Sample variance significance was determined using a two-sample F-test for equal variances († p < 0.05, ‡p < 0.001). Scale bars are 10 µm.

Tau hyperphosphorylation relieves inhibition of processive lysosome motility in neurons.

A) Shown are examples of lysosome trajectories segmented into stationary (red), diffusive (blue), and processive (green) motility based on run length. Stationary segments were defined as periods between two reversal events with a run length (RL) < 0.16 µm; diffusive segments as those with run lengths between 0.16 µm and 1.2 µm; and processive segments as those > 1.2 µm. B) Plots show the fraction of time spent in each transport mode for control (CTL), MAPT-KO (KO), and MAPT-KO neurons expressing WT, AP, and E14 tau-GFP in the proximal, mid, and distal axon. Error bars represent 95% CI. Statistical significance is indicated above the plots for comparison of the fraction of processive (proc; green) diffusive (diff; blue), and stationary (stat; red) time for each condition. C–D) Bar plots show C) the mean reversal frequency of lysosomes, D) and the fraction of time of anterograde or retrograde directed processive motility. E–F) Bar plots show E) the mean run lengths and F) mean velocities of processive runs of lysosomes for each condition in proximal, mid, and distal axonal regions. Error bars in C), E), and F) indicate SEM and in D) indicate 95% CI. The number of trajectories and cells analyzed under each condition are indicated in Table 2. G) Schematic summarizing the impact of tau hyperphosphorylation on bidirectional lysosome transport in neurons. Low levels of tau phosphorylation reduce processive anterograde transport throughout the axon but has a weaker impact on retrograde transport, whereas hyperphosphorylated tau enhances lysosome motility in both directions across the axon, similar to the effects observed in tau knockout conditions. Statistical significance is shown above each plot, where red asterisks indicate comparisons of retrograde transport and black asterisks indicate comparisons of anterograde transport. (* p < 0.05, ** p < 0.001, *** p < 0.0001).