Journal: The EMBO Journal

Article Title: Clustering of Tau fibrils impairs the synaptic composition of α3‐Na + /K + ‐ ATP ase and AMPA receptors

doi: 10.15252/embj.201899871

Figure Lengend Snippet: A–C Time‐ and concentration‐dependent clustering of Fib‐Tau in primary neurons. Representative images are shown for certain conditions to illustrate Fib‐Tau clustering time dependence (A, top row) and concentration dependence (A, bottom row). Quantification of the number of Fib‐Tau clusters per μm 2 (B) or fluorescence intensity of clusters (C, indicating size, refer to ). At low concentrations (up to 0.72 nM), the density of clusters increased with time (between 10 and 60 min) but the increase in intensity was small. At high concentrations of Fib‐Tau (≥ 1.8 nM), both density and size increased with increasing time. Box‐plot represents median, interquartile range, and 10–90% distribution; one‐way ANOVA with Dunnett's post hoc test, number of images analyzed from three cultures (from left to right: 25, 25, 25, 25, 70, 45, 45, 45, 45, 45, 30, 40, 40, 40, 40, and 40 images). D–F Single‐particle tracking using quantum dots (SPT‐QD) of biotin‐tagged Fib‐Tau. Representative single molecule trajectories of Fib‐Tau following 10‐ or 60‐min exposure are shown (D). Note after 60‐min exposure (0.36 nM), single molecules are more confined suggesting they are trapped and clustered. Quantification of diffusion coefficient (E) and explored area (F, extracted from mean squared displacement (MSD), see ) shows that both these parameters decrease after 60‐min exposure to Fib‐Tau. Unpaired t ‐test, n is averaged value per cells imaged in three experiments (10 min: 22, 60 min: 23). C Neurons were exposed for 60 min to Fib‐Tau (0.36 nM) labeled with both biotin and ATTO‐488 (red). Cell surface‐exposed biotin was labeled using streptavidin‐550 (green) followed by live imaging. Note that most of the clusters of ATTO‐488 (red) are co‐labeled with streptavidin‐550 (green) indicating that the clusters are at the cell surface. H–J Clearance of Tau clusters from neurons. Neurons were exposed (0.36 nM) to ATTO‐550‐labeled Fib‐Tau for 10 min, and the unbound fibrils were washed. Cells were fixed immediately (10 min) or allowed to recover in culture medium for 60 min. Two representative images (H) and quantifications (I, J) show that following 60‐min recovery most of the Tau clusters disappear/dissociate as indicated by a decrease in their density. Box‐plot represents median, interquartile range, and 10–90% distribution; unpaired t ‐test, n is number of images analyzed from three cultures (49 images). Data information: * P < 0.05; ** P < 0.01; *** P < 0.001; ns = not significant. Scale bar, 5 μm in (G), 2 μm everywhere else.

Article Snippet: To pull down biotin‐labeled Fib‐Tau‐1N3R and 1N4R together with their partner proteins, 100 μl streptavidin magnetic beads (Pierce, Waltham, MA) were washed three times in binding buffer (50 mM Tris–HCl pH 7.5, 150 mM NaCl, 0.1% Triton X‐100, complete protease inhibitor cocktail).

Techniques: Concentration Assay, Fluorescence, Single-particle Tracking, Diffusion-based Assay, Labeling, Imaging

Journal: The EMBO Journal

Article Title: Clustering of Tau fibrils impairs the synaptic composition of α3‐Na + /K + ‐ ATP ase and AMPA receptors

doi: 10.15252/embj.201899871

Figure Lengend Snippet: Strategy used to purify and identify neuron intrinsic membrane proteins with extracellular domain that interact specifically with Fib‐Tau‐1N3R. Fib‐Tau was labeled 1 h with 10 molar equivalents of NHS‐S‐S‐Biotin. Mouse cortical neuron cultures were exposed for 10 min to biotinylated Fib‐Tau (14.4 nM). Fresh protein extracts from those neurons were incubated with streptavidin magnetic beads to pull down Fib‐Tau together with their specific protein partners. Unexposed neuron extracts were used as a control. Proteins bound to the streptavidin magnetic beads were eluted with Laemmli buffer and subjected to short migration on a SDS–PAGE gel. After Coomassie blue staining, proteins were subjected to in‐gel digestion using trypsin and subsequently identified by nanoLC‐MS/MS analysis, using a nanoLC‐TripleTOF mass spectrometer. Relative quantification between control and exposed neuron samples was performed using a label‐free MS‐based approach. Six independent replicates were analyzed. Venn diagram of 968 proteins identified in Fib‐Tau pull‐downs only (red), in control pull‐downs only (gray), or in both samples (overlap). Of the 92 proteins identified in both samples, 45 proteins were significantly enriched in Fib‐Tau pull‐downs ( t ‐test with P ‐values < 0.05, Benjamini–Hochberg, fold change > 2). Distribution of the 372 synaptic and membrane protein interactors of Fib‐Tau identified in the pull‐down experiments. Locations of proteins at the levels of subcellular structures were annotated using the Gene Ontology Cell Component annotation tool of AMIGO 2 ( http://amigo.geneontology.org/amigo/landing ). Distribution of Fib‐Tau interactors in the plasma membrane, pre‐synaptic membrane, post‐synaptic membrane, pre‐synapse, and post‐synapse is shown. List of synaptic and plasma membrane proteins with extracellular domains significantly enriched in pull‐downs from neurons exposed to Fib‐Tau. For each identified protein, the name of the protein, the gene name, the P ‐value ( t ‐test with Benjamini–Hochberg correction), and the fold change corresponding to the ratio of spectral counts between exposed neuron and control samples are given. In an independent analysis, after 10‐min exposure of neurons to biotinylated Fib‐Tau, a cross‐linking step was performed during 20 min using 1 mM of DTSSP added in the culture medium, in order to cross‐link the protein complexes formed at the cell surface using a membrane impermeable cross‐linker. After cross‐linking, proteins were analyzed and identified exactly as non‐cross‐linked samples. Proteins identified with at least two peptides are labeled “+”, and the other are labeled “−”. Co‐immunoprecipitation of exogenous biotin‐labeled Fib‐Tau with α3‐NKA, GluA2, and GluN1. α3‐NKA, GluA2, and GluN1 were immunoprecipitated using specific antibodies as described in the section. The presence of Fib‐Tau in the immunoprecipitate was assessed using a slot blot apparatus and nitrocellulose membranes probed with streptavidin‐HRP. A 2.4‐, 2.3‐, and 1.8‐fold enrichment in Tau band intensity is observed in α3‐NKA, GluA2, and GluN1 immunoprecipitates, respectively, compared to controls performed with pre‐immune goat or rabbit IgGs. Co‐immunoprecipitates of exogenous biotin‐labeled Fib‐Tau with anti‐α3‐NKA‐, GluA2‐, and GluN1‐specific antibodies were also subjected to SDS–PAGE and Western blot analysis. The presence of Fib‐Tau in the immunoprecipitates was assessed by probing the nitrocellulose membranes with streptavidin‐HRP. Fib‐Tau co‐immunoprecipitates with α3‐NKA and GluA2‐AMPA receptor but not with GluN1‐NMDA receptor. Network describing the interconnectivity of intrinsic membrane proteins extracellularly exposed (presented in panel D, labeled in yellow) and post‐synaptic proteins (proteins with a post‐synapse or a post‐synaptic membrane annotation, presented in panel C and labeled in blue) that interact with Fib‐Tau‐1N3R. This Fib‐Tau‐1N3R interactome was input in the String database (String v10, https://string-db.org/ ) and exported to Cytoscape (version 3.5.1 at http://www.cytoscape.org/ ) to visualize interactions between the identified proteins. A total of 121 proteins were evaluated. We set parameters to only detect interactions that were validated experimentally or described in databases. The thickness of the line corresponds to the confidence of interaction (thin lines, > 0.4; medium lines, > 0.7; thick lines, > 0.9).

Article Snippet: To pull down biotin‐labeled Fib‐Tau‐1N3R and 1N4R together with their partner proteins, 100 μl streptavidin magnetic beads (Pierce, Waltham, MA) were washed three times in binding buffer (50 mM Tris–HCl pH 7.5, 150 mM NaCl, 0.1% Triton X‐100, complete protease inhibitor cocktail).

Techniques: Membrane, Labeling, Incubation, Magnetic Beads, Control, Migration, SDS Page, Staining, Tandem Mass Spectroscopy, Mass Spectrometry, Quantitative Proteomics, Clinical Proteomics, Immunoprecipitation, Dot Blot, Western Blot

Journal: The EMBO Journal

Article Title: Clustering of Tau fibrils impairs the synaptic composition of α3‐Na + /K + ‐ ATP ase and AMPA receptors

doi: 10.15252/embj.201899871

Figure Lengend Snippet: Venn diagram of 1,065 proteins identified in Fib‐Tau‐1N4R pull‐downs only (red), in control pull‐downs only (gray), or in both samples (overlap). Of the 88 proteins identified in both samples, 63 proteins were significantly enriched in Fib‐Tau‐1N4R pull‐downs ( t ‐test with P ‐values < 0.05, Benjamini–Hochberg, fold change > 2). Distribution of the 379 synaptic and membrane protein interactors of Fib‐Tau‐1N4R identified in the pull‐down experiments. Locations of proteins at the levels of subcellular structures were annotated using the Gene Ontology Cell Component annotation tool of AMIGO 2 ( http://amigo.geneontology.org/amigo/landing ). Distribution of Fib‐Tau interactors in the plasma membrane, pre‐synaptic membrane, post‐synaptic membrane, pre‐synapse, and post‐synapse is shown. Comparison of synaptic and plasma membrane proteins with extracellular domains significantly enriched in pull‐downs from neurons exposed to Fib‐Tau 1N4R and 1N3R. For each identified 1N4R protein, the name of the protein, the gene name, the P ‐value ( t ‐test with Benjamini–Hochberg correction), and the fold change corresponding to the ratio of spectral counts between exposed neuron and control samples are given. Co‐immunoprecipitation of exogenous biotin‐labeled Fib‐Tau‐1N4R with α3‐NKA, GluA2, and GluN1. α3‐NKA, GluA2, and GluN1 were immunoprecipitated using specific antibodies as described in the section. An 1.8‐, 1.2‐, and 2.5‐fold enrichment in Tau band intensity is observed in α3‐NKA, GluA2, and GluN1 immunoprecipitates, respectively, compared to controls performed with pre‐immune goat or rabbit IgGs. Detection of co‐immunoprecipitation by SDS–PAGE and Western blotting. Co‐immunoprecipitation of exogenous biotin‐labeled Fib‐Tau‐1N4R with anti‐α3‐NKA‐, GluA2‐, and GluN1‐specific antibodies. The presence of Fib‐Tau in the immunoprecipitate was assessed by probing the nitrocellulose membranes with streptavidin‐HRP. Overall, the signal was low with high background. Fib‐Tau (**) co‐immunoprecipitates with α3‐NKA and GluN1‐NMDA receptor but not with GluA2‐AMPA receptor.

Article Snippet: To pull down biotin‐labeled Fib‐Tau‐1N3R and 1N4R together with their partner proteins, 100 μl streptavidin magnetic beads (Pierce, Waltham, MA) were washed three times in binding buffer (50 mM Tris–HCl pH 7.5, 150 mM NaCl, 0.1% Triton X‐100, complete protease inhibitor cocktail).

Techniques: Control, Membrane, Clinical Proteomics, Comparison, Immunoprecipitation, Labeling, SDS Page, Western Blot