Phosphorylation in the amino terminus of tau prevents inhibition of anterograde axonal transport

Abstract

Alzheimer's disease (AD) and other tauopathies are characterized by fibrillar inclusions composed of the microtubule-associated protein, tau. Recently, we demonstrated that the N-terminus of tau (amino acids [aa] 2-18) in filamentous aggregates or N-terminal tau isoforms activate a signaling cascade involving protein phosphatase 1 and glycogen synthase kinase 3 that results in inhibition of anterograde fast axonal transport (FAT). We have termed the functional motif comprised of aa 2-18 in tau the phosphatase-activating domain (PAD). Here, we show that phosphorylation of tau at tyrosine 18, which is a fyn phosphorylation site within PAD, prevents inhibition of anterograde FAT induced by both filamentous tau and 6D tau. Moreover, Fyn-mediated phosphorylation of tyrosine 18 is reduced in disease-associated forms of tau (e.g., tau filaments). A novel PAD-specific monoclonal antibody revealed that exposure of PAD in tau occurs before and more frequently than tyrosine 18 phosphorylation in the evolution of tangle formation in AD. These results indicate that N-terminal phosphorylation may constitute a regulatory mechanism that controls tau-mediated inhibition of anterograde FAT in AD.

Fig. 1

Schematic diagram of tau constructs used. (A) The longest human adult isoform of tau, ht40 (WT tau), contains exons 2 and 3 (E2 and E3) and four MTBRs. PAD (AEPRQEFEVMEDHAGTY) is responsible for inhibiting anterograde FAT (black box). Seven putative SH3 binding domains exist in tau between amino acids 176–236 (vertically hatched box) (Lee et al., 1998). (B) The Y→E/F tau mutants are full-length tau constructs with single pseudophosphorylation mutations (E – glutamic acid) at Y18 or Y29. A mutation control protein consisted of a Y→F (phenylalanine) mutation at residue 18. (C) The pY18 tau protein is full-length tau phosphorylated specifically at Y18 by fyn kinase. To ensure that only Y18 was phosphorylated by fyn, all of the other tyrosines in tau (Y29, Y197, Y310, and Y394) were mutated to F. (D) The AT8 tau protein was pseudophosphorylated at S199, S202 and T205 by mutagenesis of S→E and T→E to mimic the phosphoepitope recognized by the AT8 antibody (Biernat et al., 1992; Goedert et al., 1995) in hyperphosphorylated tau. (E) The Δ144–273 mutation contains a deletion of amino acids 144–273 (the proline-rich region and MTBR1), which was found in a patient with an autosomal dominant familial form of FTD (Rovelet-Lecrux et al., 2009). (F) 6D tau is a truncated, non-canonical N-terminal isoform resulting from alternative splicing in exon 6 and containing a unique 11 amino acid C-terminal sequence (in box). (G) Single pseudophosphorylation mutants of 6D tau were generated at Y18E, Y29E, and T17E. The Y18F 6D mutant protein was used as a mutation control.

Fig. 2

Pseudophosphorylation of Y18 prevents tau filaments from inhibiting anterograde FAT. (A) Representative electron micrographs illustrate that WT tau, Y18E tau, Y18F tau and Y29E tau produced comparable quantities of morphologically similar filaments (see Supplementary Fig. S1A). Scale bar = 250 nm. (B–F) Vesicle motility assays in isolated squid axoplasm. Individual velocity (µm/sec) measurements (arrowheads) are plotted as a function of time (minutes). (B) Perfusion of WT tau monomers did not affect anterograde (►) or retrograde () FAT rates. (C) In contrast, perfusion of WT tau filaments selectively inhibited anterograde, but not retrograde FAT. (D) Filaments composed of Y18E tau did not affect FAT. (E) Y18F tau filaments (a non-phosphomimicking control mutation) selectively inhibited anterograde FAT (see Supplementary Fig. S1B). (F) Y29E tau filaments specifically decreased anterograde FAT. (G) Quantification of anterograde and retrograde FAT rates in B–F indicates that WT, Y18F and Y29E filaments significantly reduce the rate of anterograde FAT relative to WT tau monomers in isolated axoplasms (one-way ANOVA; * p<0.01). Retrograde FAT was unaffected by all tau constructs tested.

Fig. 3

Fyn-mediated phosphorylation of Y18 prevents the inhibitory effect of tau filaments on anterograde FAT. (A) A tau mutant with a single Y at position 18 (all other tyrosines were mutated to F) was incubated with fyn kinase to phosphorylate Y18 (pY18 tau) prior to polymerization. An electron micrograph confirming that pY18 tau can polymerize into filaments. Scale bar = 500 nm. (B) Immunoblots using the phosphorylation-independent R1 tau antibody (total tau) and the 9G3 antibody, a phospho-Y18-specific antibody, confirmed the phosphorylation status of Y18 in the filaments used in motility assays (shown in panel C; see Supplementary Fig. 2). (C) Unlike WT tau filaments (Fig. 2C), pY18 tau filaments had no effect on FAT (► anterograde; retrograde). (D) Box plots summarizing the statistical comparisons of data shown in panel C (one-way ANOVA; * p<0.01 compared to WT tau monomer and pY18 tau filaments).

Fig. 4

Tau monomers are more readily phosphorylated than tau filaments at Y18. (A) Representative blots of in vitro kinase reactions (0–60 minute incubation) using fyn kinase with either WT tau monomer (M) or filaments (F). Antibodies used were R1 for total tau and 9G3 for pY18 tau. (B) Quantification of Y18 phosphorylation demonstrated an increased rate of phosphorylation in tau monomers, when compared to tau filaments (two-way repeated measure ANOVA; * p<0.05; error bars = +SEM). (C) Representative electron micrographs of tau filaments taken at 0 and 60 minutes after incubation with fyn kinase confirmed that tau filaments did not depolymerize upon phosphorylation at Y18. Scale bar = 500 nm. (D) Representative blots of a spin down assay performed after 10 minutes incubation with fyn kinase suggest that phosphates were added to tau proteins incorporated into the pre-formed tau filaments. (NonPh – prior to fyn incubation; Ph – post fyn incubation; S – soluble fraction; P – pellet fraction). (E) Quantification of pY18 reactivity demonstrates that pelleted tau filaments (PhFP) were phosphorylated at Y18, but at levels ~40% lower than soluble tau in the polymerized sample (PhFS). The monomeric tau sample (PhMS and PhMP) confirms the efficacy of separating soluble monomers from filamentous tau species since no monomeric tau was present in the pelleted fraction (two-way ANOVA; * p<0.05; error bars = +SEM).

Fig. 5

Disease-associated tau modifications impair phosphorylation at Y18. (A) WT tau was rapidly phosphorylated at Y18 by fyn kinase. In contrast, fyn-mediated phosphorylation of Y18 in AT8 tau and Δ144–273 tau was reduced. Antibodies used were Tau7 for total tau and 9G3 for pY18 tau. (B) Analysis of pY18 tau levels by ELISA demonstrates that while the maximal level of phosphorylation was similar, the rate of Y18 phosphorylation in AT8 tau was significantly reduced compared to WT tau (two-way repeated measure ANOVA; * p<0.05 compared to WT tau). Both the rate and maximal level of phosphorylation at Y18 in Δ144–273 tau was dramatically reduced compared to WT tau and AT8 tau (# p<0.05 compared to WT and AT8 tau).

Fig. 6

Y18 pseudophosphorylation prevents the inhibitory effect of 6D tau on anterograde FAT. (A) Perfusion of 6D tau inhibits anterograde FAT (►) without affecting retrograde FAT (). (B) Pseudophosphorylation at Y18 in 6D tau (Y18E) prevents this effect. (C) Perfusion of Y18F 6D tau inhibits anterograde FAT to an extent comparable to 6D tau. (D) Perfusion of Y29E 6D tau inhibits anterograde FAT to a lesser extent than 6D tau. (E) Box plots summarizing the data shown in panels A–D and the statistical comparisons (one-way ANOVA; * p<0.01 compared to WT tau, Y18E 6D, Y29E 6D and T17E 6D; # p<0.01 compared to Y18E 6D) (see also Supplementary Fig. S4).

Fig. 7

PAD exposure precedes and is more abundant than Y18 phosphorylation in human brains. (A) The specificity of TNT1 (red) for PAD was tested by Western blot using recombinant WT tau and Δ2–18 tau. Additionally, phosphorylation at Y18 does not interfere with TNT1 reactivity (red) as indicated by similar reactivity between Y4F and pY18 Y4F tau. (B–C) The specificity of TNT1 for tau proteins was confirmed using soluble (Sol) and insoluble (PHF) fractions isolated from the frontal cortex of control (n=3) and AD (n=3) brains. (B) Note that in denaturing conditions on Western blots TNT1 (red) reacts with all tau samples from controls and AD brains. (C) However, when samples were run in non-denaturing conditions on dot blots TNT1 (red) shows remarkable selectivity for soluble and insoluble tau from AD brains, not from controls. In A–C, R1 (green) was used to label total tau. (D) Immunohistochemistry for PAD exposed tau (TNT1 positive, brown) and phospho-Y18 tau (9G3 positive, blue) in controls (Braak stages I–II) reveals that Y18 phosphorylation occurs in more mature compact TNT1 positive tau inclusions (▸), rather than the early stage of tau accumulation indicated by diffuse cytoplasmic staining in “pre-tangle” neurons (→). Scale bars are 50µm. (E) Immunofluorescence confirmed the lack of phospho-Y18 tau (green) in the early pre-tangle neurons that are diffusely stained with TNT1 (red) in control brains (Braak stages I–II). Scale bars are 20µm. (F) Immunohistochemistry in hippocampal sections shows that TNT1 pathology (brown) is substantially increased in mild AD (Braak stages III–IV) and severe AD (Braak stages V–VI), compared to control cases (Braak stages I–II). The 9G3 positive pathology (blue) does increase with disease severity but not to the same magnitude as TNT1. Additionally, 9G3 shows nearly complete co-localization with TNT1 in all stages (TNT1+ neurons →; TNT1+9G3 neurons ▸). Scale bars are 50µm. (G–I) High magnification of the boxed images in (F) depict single (TNT1 only; brown) and double stained (TNT+ 9G3; brown and blue) neurons in control (G), mild AD (H) and severe AD (I) cases. Scale bars are 20µm. (J–M) Immunofluorescence confirmed the co-localization of TNT1 (red) and 9G3 (green) in hippocampal neurons containing mature compact tau inclusions from control (J) and severe AD (K) cases. Neuritic plaques consistently exhibit co-localization of TNT1 and 9G3 in mild (L) and severe AD (M), but TNT1 immunoreactivity is more extensive. Scale bars are 20µm.

Fig. 8

Schematic diagram of the proposed role for N-terminal phosphorylation of tau in cargo delivery and aberrant inhibition of anterograde FAT in disease. (A) Normally, tau localizes to microtubules where, in addition to stabilizing microtubules, it is capable of regulating local cargo delivery through PAD-mediated activation of the PP1-GSK3 cascade. Under normal conditions, cargoes are transported by conventional kinesin along microtubules when tau is phosphorylated at Y18 (by fyn or other tyrosine kinasaes) because PAD cannot activate the PP1-GSK3 cascade. In contrast, cargo is delivered at locations where intact unmodified PAD is exposed. (B) In disease, tau aggregation and other modifications expose PAD in tau no longer attached to microtubules, which initiates aberrant activation of the PP1-GSK3 cascade, inhibition of anterograde FAT and ultimately neuron dysfunction/degeneration. Neurons can mitigate the effects of abnormal PAD presentation in disease-modified tau species through Y18 phosphorylation (by fyn or other tyrosine kinases), which renders pathogenic tau non-toxic to FAT.