Ectosomes: a new mechanism for non-exosomal secretion of tau protein

Abstract

Tau is a microtubule-associated protein that aggregates in neurodegenerative disorders known as tauopathies. Recently, studies have suggested that Tau may be secreted and play a role in neural network signalling. However, once deregulated, secreted Tau may also participate in the spreading of Tau pathology in hierarchical pathways of neurodegeneration. The mechanisms underlying neuron-to-neuron Tau transfer are still unknown; given the known role of extra-cellular vesicles in cell-to-cell communication, we wondered whether these vesicles could carry secreted Tau. We found, among vesicles, that Tau is predominately secreted in ectosomes, which are plasma membrane-originating vesicles, and when it accumulates, the exosomal pathway is activated.

Figure 1. Characterisation of ectomal and exosomal fractions from rat primary embryonic cortical cells.

Once purified from embryonic primary cultures, vesicles obtained from EcEF (a, left panels) or ExEF (a, right panels) were observed by EM. A scale bar is indicated on the figure. (b) A semi-quantitative analysis (n = 3 independent experiments with n > 200 vesicles per experiment) was performed to determine the purity of the EcEF and ExEF. The results are expressed as the mean of percentage +/^_ standard deviation and indicated at the top of the bar chart (c) The exosomes were purified from the ExEF using a continous sucrose gradient and Alix and Flotillin-1 used as specific markers.

Figure 2. Endogenous Tau is released from rat primary embryonic cortical cells in non-exosomal vesicles: the ectosomes.

EcEF (a, left panels) and ExEF (a, right panels) obtained from primary embryonic cultures were immunolabelled with N-Ter antibodies (a, upper panels) or C-Ter antibodies (a, lower panels), and the association of Tau with vesicles was observed by EM using an 18 nm gold colloidal goat anti-rabbit antibody. The scale bar is indicated on the figure. (b) A semi-quantitative analysis (n = 2 or 3 independent experiments with n > 200 vesicles per experiment) was performed to quantify the percentage of ectosomes and exosomes immunopositive for Tau. The results are expressed as the mean of percentage +/^_ standard deviation and indicated at the top of the bar chart. The Chi² test was used to compare the presence of Tau in ectosomes and exosomes for each individual experiment: N-Ter: Chi²-1 = 100, Chi²-2 = 34; C-Ter: Chi²-1 = 21, Chi²-2 = 20, Chi²-3 = 13). The statistical tests indicated p < 0.001 for each condition. (c) The presence of endogenous Tau in the cell lysate (CL), EcEF and ExEF obtained from primary embryonic cultures was analysed by western blotting using either C-Ter or N-Ter antibodies. (d) A quantitative analysis by ELISA (n = 2 independent experiments) was performed to quantify the ratio of vesicular versus non-vesicular Tau. (e) Primary embryonic cultures were plated and maintained in culture for 3, 5 or 10 days (3 DIV, 5 DIV or 10 DIV). Total Tau was analysed by western blotting using the C-Ter antibody in the cell lysate (CL) or after fractionation (EcEF and ExEF) of conditioned media; the conditioned media were also analysed for LDH release (f).

Figure 3. Characterisation of ectosomal and exosomal fractions from cell lines.

After purification from N1E-115 cells stably overexpressing h1N4R, vesicles obtained from EcEF (a, left panels) or ExEF (a, right panels) were observed by EM. The scale bar is indicated on the figure. (b) A semi-quantitative analysis (n = 3 independent experiments with n > 200 vesicles per experiment) was performed to determine the purity of EcEF and ExEF. The results are expressed as the mean of percentage +/^_ standard deviation and indicated at the top of the bar chart. S = size. (c) The exosomes were purified from the ExEF using a continous sucrose gradient and Alix and Flotillin-1 used as specific markers.

Figure 4. A small portion of Tau is shifted to the classical secretory pathway when Tau over-accumulates in cells.

EcEF (a, left panels) and ExEF (a, right panels) obtained from stable N1E-115 overexpressing h1N4R cells were immunolabelled with N-Ter (a, upper panels) or C-Ter (a, lower panels) antibodies. The scale bar is indicated on the figure. The association of Tau to vesicles was observed by EM using an 18 nm gold colloidal goat anti-rabbit antibody. (b) A semi-quantitative analysis (n = 2 or 3 independent experiments with n > 200 vesicles per experiment) was performed to determine the percentage of ectosomes and exosomes immunopositive for Tau. The Chi² test was used to compare the presence of Tau in ectosomes and exosomes for each individual experiment: N-Ter: Chi²-1 = 29 (***, p < 0.001), Chi²-2 = 8 (**, p < 0.01), C-Ter: Chi²-1 = 22 (***, p < 0.001), Chi²-2 = 21 (p < 0.001), Chi²-3 = 2.5 (NS)). The results are expressed as the mean +/^_ standard deviation and indicated at the top of the bar chart. (c) Total Tau was analysed by western blotting using HT7, N-Ter and C-Ter antibodies in the cell lysate (CL) or after fractionation (EcEF and ExEF) of conditioned media. (d) The conditioned media obtained from native or differentiated N1-E115 cells over-accumulating h1N4R were analysed for LDH release. (e) Rat embryonic primary neurons (10 DIV) were infected or not (∅) by LVs encoding either GFP or h1N4R. Total Tau was analysed by western blotting using a N-Ter antibody in the cell lysate (CL) or after fractionation (EcEF and ExEF) of conditioned media.

Figure 5. Tau is inside the vesicles.

EcEF (upper panel) and ExEF (lower panel) obtained from stable N1E-115 overexpressing h1N4R cells (left part of the immunoblots) or from naive N1E-115 (right part of the immunoblots) were incubated with growing concentrations of NaCl (0.01 to 0.5M) before western blotting analyses using a total Tau N-ter antibody and a flotillin-1 antibody. EcEF and ExEF obtained from naive N1E-115 were previously incubated with recombinant h1N4R Tau. Cell lysates (CL) from both cell lines and recombinant h1N4R Tau were used as controls.

Figure 6. Neurofibrillary degeneration related to WT Tau in the rat brain supports vesicular Tau secretion in ISF.

LVs encoding h1N4R were bilaterally injected into the CA1 layer of rat brains (n = 4). Five months later, ISF was recovered by the push-pull method. Naive control rats were also included in the assay (n = 2). EcEF and ExEF from LVs-injected rats (respectively a, left panels and a, right panels) or naive rats (respectively c, left panels and c, right panels) were fractionated and immunolabeled with N-Ter antibodies (LVs-injected rats: a, upper panels, naive rats: c, upper panels) or C-Ter antibodies (LVs-injected rats: a, lower panels, naive rats: c, lower panels). The scale bar is indicated on the figure. The association of Tau with vesicles was observed by EM using a 18 nm gold colloidal goat anti-rabbit antibody. A semi-quantitative analysis was performed to determine the percentage of ectosomes (size > 100 nm) and exosomes (30 nm < size <70 nm) immunopositive for Tau (LVs-injected rats: b, naive rats: d). For LVs-injected rats 4 independent experiments were performed; the total number of vesicles evaluated was 108, 110, 24 and 105 for C-Ter in ExEF, N-Ter in ExEF, C-Ter in EcEF and N-Ter in EcEF, respectively. For naive rats, 2 independent experiments were performed; the total number of vesicles evaluated was 254, 297, 35 and 27 for C-Ter in ExEF, N-Ter in ExEF, C-Ter in EcEF and N-Ter in EcEF, respectively. The results are expressed as the mean +/^_ standard deviation and indicated at the top of the bar. The Chi² test was used to compare the presence of Tau in ectosomes and exosomes: LVs-injected rats, N-Ter: Chi² = 6.3 (*, p < 0.05); Naive rats, Nter: Chi² = 10.9 (*, p < 0.05); Naive rats, Cter: Chi² = 5.5 (*, p < 0.05).