HSV-1 infection induces phosphorylated tau propagation among neurons via extracellular vesicles

Abstract

Extracellular vesicles (EV), key players in cell-to-cell communication, may contribute to disease propagation in several neurodegenerative diseases, including Alzheimer's disease (AD), by favoring the dissemination of neurotoxic proteins within the brain. Interestingly, growing evidence supports the role of herpes simplex virus type 1 (HSV-1) infection in the pathogenesis of AD. Here, we investigated whether HSV-1 infection could promote the spread of phosphorylated tau (ptau) among neurons via EV. We analyzed the ptau species that were secreted via EV following HSV-1 infection in neuroblastoma cells and primary neurons, focusing particularly on T205, T181, and T217, the phosphorylation sites mainly associated with AD. Moreover, by overexpressing human tau tagged with GFP (htau^GFP), we found that recipient tau knockout (KO) neurons uptook EV that are loaded with HSV-1-induced phtau^GFP. Finally, we exploited an in vivo model of acute infection and assessed that cerebral HSV-1 infection promotes the release of ptau via EV in the brain of infected mice. Overall, our data suggest that, following HSV-1 infection, EV play a role in tau spreading within the brain, thus contributing to neurodegeneration.IMPORTANCEHerpes simplex virus type 1 (HSV-1) infection that reaches the brain has been repeatedly linked with the appearance of the pathognomonic markers of Alzheimer's disease (AD), including accumulation of amyloid beta and hyperphosphorylated tau proteins, and cognitive deficits. AD is a multifactorial neurodegenerative disease representing the most common form of dementia in the elderly, and no cure is currently available, thus prompting additional investigation on potential risk factors and pathological mechanisms. Here, we demonstrate that the virus exploits the extracellular vesicles (EV) to disseminate phosphorylated tau (ptau) among brain cells. Importantly, we provide evidence that the HSV-1-induced EV-bearing ptau can be undertaken by recipient neurons, thus likely contributing to misfolding and aggregation of native tau, as reported for other AD models. Hence, our data highlight a novel mechanism exploited by HSV-1 to propagate tau-related damage in the brain.

Keywords: Alzheimer’s disease; extracellular vesicles; herpes simplex virus; tau.

Fig 1

Characterization of EV isolated from SH-SY5Y neuroblastoma cell supernatants upon HSV-1 infection. (A and B) Representative TEM micrographs showing the membranous nature of the MV (A) and exo (B) isolated from supernatants of mock-infected neuroblastoma cultures 36 h p.i. Scale bars are indicated in the pictures. (C and D) Size distribution of collected vesicles in MV (C) and exo (D) samples. Semi-quantitative data were obtained from TEM images measuring the major diameter of n = 50 EV. (E) Representative immunoblots showing the protein levels for EV markers alix, flotillin, and TSG101; intracellular marker calnexin; tau protein levels; tau phosphorylation sites in T181, T205, and T217 in lysates of cells and EV from mock- (ctr) and HSV-1-infected neuroblastoma cultures harvested 36 h p.i.; long arrows indicate 50–75-kDa tau; arrowheads indicate high-MW tau. (F) Densitometric analyses representing the normalized fold-changes of ptau and tau protein levels compared with ctr, performed with ImageLab software and normalized to alix expression in MV and exo. Data are expressed as mean ± standard error of the mean (SEM) of three independent experiments performed in separate days. Statistical significance was calculated using one-sample t-test. *P < 0.05 vs ctr.

Fig 2

Tau phosphorylation and expression in primary cortical neurons. (A–C) Confocal immunofluorescence analyses of primary cortical neurons that were mock- or HSV-1-infected with increasing moi of virus for 24 h. Images in A on the left show representative neurons that were immunostained for ptau T217 or ptau T205 (green) and ptau T181 (red). Magnified images of ptau T205 immunofluorescence are shown in B. Cell nuclei were stained with DAPI (blue). Bar graphs on the right in A show the mean fluorescence intensity of ptau immunoreactive signals. Bar graphs in C show the mean ptau 205 fluorescence intensity measured in the nuclei of all the cells present in the analyzed fields (n > 50 cells/field). The intensity was normalized on the total mean fluorescent intensity of ptau T205 measured in the same field. Data are expressed as mean ± SEM and were collected at least from three independent experiments using cultures prepared on separate days. *P < 0.05 and **P < 0.01 vs ctr. Statistical significance was calculated by one-way analysis of variance (ANOVA) followed by Bonferroni post hoc test. (D) Representative immunoblots showing tau phosphorylation and protein levels in lysates from mock- (ctr) and HSV-1-infected neurons at the indicated moi harvested 24 h p.i. Actin expression level was used as sample loading control. (E) Bar graphs show the densitometric analyses of immunoreactive signals expressed as normalized fold-change of protein compared with ctr. Data are expressed as mean ± SEM of three independent experiments performed in separate days. *P < 0.05 vs ctr. Statistical significance was calculated using one-sample t-test. (F) ELISA assay for tau was performed on supernatants (left graph) and neuronal lysates (right graphs) collected 24 h after mock and HSV-1 infection. Bar graphs show mean tau protein levels expressed as micrograms (μg) or picograms (pg) of tau protein normalized to milliliters (mL) of supernatant or milligrams (mg) of total protein. Data were collected at least from three independent experiments using cultures prepared on separate days. *P < 0.05 vs ctr. Statistical significance was calculated by one-way ANOVA followed by Bonferroni post hoc test.

Fig 3

EV purification from primary cortical neurons. (A) TEM micrographs showing the membranous nature of the vesicles isolated from supernatants of primary cultures of neurons harvested 36 h p.i. Scale bars are indicated in the pictures. (B) Size distribution of collected vesicles in MV (upper graph) and exo (lower graph) samples. Semi-quantitative data were obtained from TEM images measuring the major diameter of n = 50 EV. (C, D, and F) Representative immunoblots showing the protein levels for (C) EV markers alix, flotillin, and TSG101; intracellular marker calnexin; and viral glycoproteins gB and gC; (D) tau protein levels; (F) tau phosphorylation in the indicated sites, in cells and EV lysates from mock- (ctr) and HSV-1-infected primary cultures harvested 36 h p.i. Blots in D and F represent different exposures of the same membranes: panels on the left show cropped images of lanes loaded with cell lysates (low exposure); panels on the right show cropped images of lanes loaded with MV and exo (high exposure). (E) ELISA for tau were performed on neuronal lysates, MV, and exo collected 36 h after HSV-1 infection. Bar graphs show mean tau protein levels expressed as pg of tau protein normalized on mg of total proteins. Data were collected at least from three independent experiments using cultures prepared on separate days. Statistical analysis was carried out by one-way ANOVA followed by Bonferroni post hoc test.

Fig 4

EV purification from N2A-htau^GFP-transfected cells. Representative immunoblots (A), gel densitometry analysis (B), and fluorescent images (C) showing the overexpression of human tau tagged with GFP (htau^GFP) in N2A cells. Cells were transfected for 24 h prior to being mock- or HSV-1-infected for 36 h with 1 moi of virus. (D) Bar graphs showing HSV-1 titers, calculated by standard plaque assay and expressed as PFU/mL. (E and F) Representative immunoblots showing MV (E) and exo (F) isolated from mock- (ctr) and HSV-1-infected cells. (G and H) Bar graphs show the densitometric analyses of immunoreactive signals detected in MV (G) and exo (H) and expressed as normalized fold-change of protein compared with ctr. Alix expression level was used as sample loading control. Data were collected at least from three independent experiments using cultures prepared on separate days and expressed as mean ± SEM. Statistical significance was calculated using one-sample t-test. *P < 0.05 vs ctr.

Fig 5

Recipient assay performed with EV purified from htau^GFP-N2A onto tau KO primary neurons. (A) Schematic representation of the experimental procedures for the recipient assay. Mouse neuroblastoma cells (N2A) were transfected for 24 h with an expression vector for htau^GFP. Following 36 h of mock and HSV-1 infection, EV secreted by transfected cells were collected, treated with UV to inactivate the virus, and layered on tau KO neurons for 24 h. (B) Representative high exposure immunoblots showing htau^GFP levels and its phosphorylation in T205 in primary KO neuron lysates collected after 24 h of recipient assay (i.e., EV layering). Actin expression level was used as sample loading control. Bar graphs on the right side of the panel show GFP expression normalized to actin and tau phosphorylation at T205 normalized to GFP immunoreactive signal. Data are expressed as mean ± SEM and represented as fold-changes compared with ctr. (C) Confocal immunofluorescence images of primary KO neurons that were used as recipients for MV isolated from transfected and non-transfected N2A cells. Arrows indicate GFP⁺ signal. Insets show higher magnification of boxes outlined in each panel. Data were collected at least from three independent experiments using cultures prepared on separate days and were expressed as mean ± SEM. Statistical significance was calculated by one-sample t-test. *P < 0.05 vs ctr.

Fig 6

EV purified from in vivo mice model of HSV-1 infection. (A and B) Representative images of TEM (A) and ScEM (B) analyses showing the EV isolated from in vivo samples. Scale bars are indicated in the figure. (C) Size distribution of collected vesicle semi-quantitative data was obtained from TEM images measuring the major diameter of more than 150 EV. (D) Representative immunoblots showing tau phosphorylation and protein levels in lysates of different fractions of EV purification from mock- (ctr) and HSV-1-infected mice. Alix expression level was used as sample loading control. (E) Densitometric analyses of immunoreactive signals in Fraction 1 (F1) are shown in the graphs: values represent the normalized fold-changes in protein levels from HSV-infected mice with respect to ctr (n = 5). Data are expressed as mean ± SEM. Statistical significance was calculated by one-sample t-test. *P < 0.05 vs ctr.