pubmed.ncbi.nlm.nih.gov

Pathological Tau Strains from Human Brains Recapitulate the Diversity of Tauopathies in Nontransgenic Mouse Brain - PubMed

  • ️Sun Jan 01 2017

Pathological Tau Strains from Human Brains Recapitulate the Diversity of Tauopathies in Nontransgenic Mouse Brain

Sneha Narasimhan et al. J Neurosci. 2017.

Abstract

Pathological tau aggregates occur in Alzheimer's disease (AD) and other neurodegenerative tauopathies. It is not clearly understood why tauopathies vary greatly in the neuroanatomical and histopathological patterns of tau aggregation, which contribute to clinical heterogeneity in these disorders. Recent studies have shown that tau aggregates may form distinct structural conformations, known as tau strains. Here, we developed a novel model to test the hypothesis that cell-to-cell transmission of different tau strains occurs in nontransgenic (non-Tg) mice, and to investigate whether there are strain-specific differences in the pattern of tau transmission. By injecting pathological tau extracted from postmortem brains of AD (AD-tau), progressive supranuclear palsy (PSP-tau), and corticobasal degeneration (CBD-tau) patients into different brain regions of female non-Tg mice, we demonstrated the induction and propagation of endogenous mouse tau aggregates. Specifically, we identified differences in tau strain potency between AD-tau, CBD-tau, and PSP-tau in non-Tg mice. Moreover, differences in cell-type specificity of tau aggregate transmission were observed between tau strains such that only PSP-tau and CBD-tau strains induce astroglial and oligodendroglial tau inclusions, recapitulating the diversity of neuropathology in human tauopathies. Furthermore, we demonstrated that the neuronal connectome, but not the tau strain, determines which brain regions develop tau pathology. Finally, CBD-tau- and PSP-tau-injected mice showed spatiotemporal transmission of glial tau pathology, suggesting glial tau transmission contributes to the progression of tauopathies. Together, our data suggest that different tau strains determine seeding potency and cell-type specificity of tau aggregation that underlie the diversity of human tauopathies.SIGNIFICANCE STATEMENT Tauopathies show great clinical and neuropathological heterogeneity, despite the fact that tau aggregates in each disease. This heterogeneity could be due to tau aggregates forming distinct structural conformations, or strains. We now report the development of a sporadic tauopathy model to study human tau strains by intracerebrally injecting nontransgenic mice with pathological tau enriched from human tauopathy brains. We show human tau strains seed different types and cellular distributions of tau neuropathology in our model that recapitulate the heterogeneity seen in these human diseases.

Keywords: human tauopathies; tau mouse model; tau strains; tau transmission.

Copyright © 2017 the authors 0270-6474/17/3711406-18$15.00/0.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.

Biochemical characterization of tau strains from human tauopathy brains. A, IHC was conducted on sections of mid-frontal cortex with anti-tau MAb PHF-1 on all AD and CBD cases as well as PSP Case 1, and with MAb CP13 in PSP Case 2. Cases are identified by numbers corresponding to those in Table 1. Scale bar, 50 μm. B, Western blots were performed using anti-tau antibodies 17025 (red), PHF-1 (green), anti-4R tau from CosmoBio (red), and RD3 (green), plus silver staining, on final supernatants from sequential extraction of three AD, three CBD, and two PSP cases from A. Final supernatant blots are representative of multiple extractions from the same case (Table 2). C, IHC of the lentiform nucleus for PHF-1 from PSP cases. Scale bar, 50 μm. D, Western blots for 17025 (red) and PHF-1 (green) on final supernatants from sequential extraction of the lentiform nucleus of PSP cases from C. E, Immuno-EM for PHF-1 (1:300) on AD-tau fibrils (AD Case 3; 0.9 μg/μl), CBD-tau fibrils (CBD Case 2; 0.55 μg/μl), and PSP-tau fibrils (PSP Case 1; 0.625 μg/μl) before (−) and after (+) sonication. Scale bar, 100 nm. F, Representative Western blots for PHF-1 (green) on the GuHCl denaturation assay for one AD, one CBD, and two PSP cases. G, Whole-lane quantification of the PHF-1 signals for each concentration of GuHCl as a percentage of the initial sample (0

m

GuHCl) is plotted as percentage PK resistance for each case. Two to three replicates of n = 3 cases of AD, n = 3 cases of CBD, and PSP Case 1 were used for this quantification. Two-way ANOVA with Bonferroni post hoc test was performed comparing PSP Case 1 to all other groups (Mean ± SEM plotted; Tauopathy factor: F = 21.96, df = 3, p = 0.0001; GuHCl concentration factor: F = 98.71, df = 4, p = 0.0001; Interaction: F = 3.739, df = 12, p = 0.0002; residual df = 70). Statistical significance is indicated at each concentration of GuHCl (

m

) where each tauopathy is compared with PSP Case 1 (*p < 0.05, **p < 0.01, ***p < 0.0001). H, The two cases of PSP were considered separately as they exhibited different properties using this assay. Whole-lane quantification of the PHF-1 signals for each concentration of GuHCl as a percentage of the initial sample (0

m

GuHCl) is plotted as percentage PK resistance for two to three replicates of each PSP case. Statistics are the same as described in F.

Figure 2.
Figure 2.

Human tauopathy lysates seeding endogenous mouse tau in primary non-Tg neurons. A, Left, Top, Representative images from immunocytochemistry (ICC) studies for anti-mouse tau-specific MAb T49 (green) on primary non-Tg hippocampal neurons treated with two different amounts of pathological tau from AD, CBD, and PSP cases (note: the PSP case is Case 1 only). Left, Bottom, Images from ICC for T49 (green) and MAP2 (red) overlay from AD-tau, CBD-tau and PSP-tau-treated neurons. Scale bar, 100 μm. B, Quantification of percentage area occupied by T49 signal/DAPI signal from ICC of neurons treated with AD-tau from three cases of AD (2 replicates), CBD-tau from three cases of CBD (2 replicates), and PSP-tau from one case of PSP (Case 1; 2 replicates) at different tau concentrations. One-way ANOVA with Tukey post hoc test comparing all groups was performed (Mean ± SEM plotted; F = 86.49, df = 24; *p < 0.05, **p < 0.01, **p < 0.001). C, Western blot for T49 on Sarkosyl-soluble supernatants (sup) and Sarkosyl-insoluble pellets (pel) extracted from neurons treated with AD-tau, CBD-tau, or PSP-tau. Western blot for GAPDH as loading control. D, Representative images from ICC for T49 (green) on primary non-Tg hippocampal neurons treated with PSP-tau from five different PSP cases. PSP Cases 1, 3, 4, and 5 were used to treat neurons with 15 ng of Sarkosyl pellet from the lentiform nuclei, and PSP Case 2 was used to treat neurons with 150 ng of final supernatant from frontal cortex. Scale bar, 100 μm. E, Top, Images are shown here of ICC for T49 from primary non-Tg hippocampal neurons treated with 200 ng of PSP-tau, or Tau5-immunodepleted PSP-tau from Case 1, or mock IgG-treated PSP-tau from Case 1. Bottom, Representative images are shown here of ICC for T49 from primary non-Tg hippocampal neurons treated with 100 ng of CBD-tau from n = 3 cases, or 17025-immunodepleted CBD-tau, or mock IgG-treated CBD-tau. Scale bar, 100 μm.

Figure 3.
Figure 3.

Different seeding potencies and cell-type specificity of human tau strains in non-Tg mice. A, Top, Left, Schematic coronal and sagittal sections of the mouse brain showing injection sites in dorsal hippocampus and overlying cortex (bregma: −2.5 mm; lateral: +2 mm; depth: −2.4 mm and −1.4 mm from the skull). Right, Representative images from IHC for anti-tau MAb AT8 (pS202/T205) for AD-tau Case 1 (1 μg/site, n = 3 mice), CBD-tau Case 1 (1 μg/site, n = 4 mice), and PSP-tau Case 1 (0.7 μg/site, n = 4 mice) injected non-Tg mice 3 months p.i. (hippocampus/cortex). Black boxes with insets indicate neuronal tau pathology and red boxes with insets indicate glial tau pathology. Scale bars: panel, 100 μm; inset, 10 μm. B, Representative images of IHC for AT8 on two additional cases of AD-tau injections (each injection at 1 μg/site, n = 3 mice each), two additional cases of CBD-tau injections (Case 2: 0.43 μg/site, n = 3 mice; Case 3: 0.9 μg/site, n = 3 mice), and one additional case of PSP-tau injections (Case 2: 0.21 μg/site, n = 4 mice) into non-Tg mice at 3 months p.i. Black boxes with insets indicate neuronal tau pathology and red boxes with insets indicate glial tau pathology. Quantification performed on two sections for each region on three AD cases (n = 9 mice total; white bars), three CBD cases (n = 10 mice total; red bars), and two PSP cases (n = 8 mice total; blue bars) with the following bregma based on brain atlas coordinates: ventral hilar and DG neurons: −3.80 and −3.64 mm; ventral CA3: −3.40 and −3.28 mm; entorhinal cortex: −3.80 and −3.64 mm. One-way ANOVA with Tukey post hoc analysis was performed for each region (Mean ± SEM plotted; ventral hilus ipsi: F = 6.593, df = 25, ANOVA, p = 0.0055; ventral hilus contra: F = 6.474, df = 25, ANOVA, p = 0.0059; DG ipsi: F = 38.33, df = 25, ANOVA, p = 0.0001; Ventral CA3 ipsi: F = 5.682, df = 25, ANOVA, p = 0.0099; Ventral CA3 contra: F = 5.05, df = 25, ANOVA, p = 0.0152; Entorhinal Cortex ipsi: F = 11.12, df = 25, ANOVA, p = 0.0004; Tukey post hoc: *p < 0.05, **p < 0.01, ***p < 0.001). Scale bars: panel, 100 μm; inset, 10 μm. C, Left, Representative images of IHC for AT8 or double-labeling IF for AT8 (red) and anti-oligodendrocyte-specific marker Olig-2 (green) from CBD-tau (1 μg/site; n = 4 mice) and PSP-tau (0.7 μg/site; n = 4 mice)-injected mice 3 months p.i. Insets show formation of AT8-positive inclusions in oligodendrocytes of fimbria (white matter tract). Scale bars: panel, 50 μm; inset, 10 μm. Right, Representative images of IHC for AT8 or double-labeling IF for AT8 (green) and anti-astrocyte-specific marker GFAP (red) from CBD-tau- (1 μg/site; n = 4 mice) and PSP-tau-(0.7 μg/site; n = 4 mice) injected mice 3 months p.i. Images show formation of AT8-positive inclusions in astrocytes of the hippocampus. Scale bars: panel, 50 μm; inset, 20 μm. Hipp, Hippocampus; Ctx, cortex; Ipsi, ipsilateral, Contra, contralateral; DG, dentate gyrus.

Figure 4.
Figure 4.

Early recruitment and maturation of tau aggregates in non-Tg mice. A, Representative images from IHC for AT8 for AD-tau- (1 μg/site; n = 3 mice), CBD-tau- (0.4 μg/site; n = 2 mice), and PSP-tau- (0.7 μg/site; n = 2 mice) injected non-Tg at 1 month p.i. (hippocampus/cortex). Black boxes with insets indicate neuronal tau pathology and red boxes with insets indicate glial tau pathology. Scale bars: panel, 100 μm; inset, 10 μm. B, Representative IHC images for AT8 near injection site and R2295 (mouse tau-specific antibody) for AD-tau- (1 μg/site; n = 3 mice), CBD-tau- (1 μg/site; n = 4 mice), and PSP-tau- (0.7 μg/site; n = 4 mice) injected non-Tg mice 3 months p.i. Scale bar, 50 μm. C, R2295 also recognizes glial tau pathology in astrocytes and oligodendrocytes, indicating endogenous mouse tau is seeded in glial cells. Left, Astrocytic plaque-like pathology from CBD-tau-injected mice. Middle, Tufted astrocyte-like pathology from PSP-tau-injected mice. Right, Oligodendroglial coiled body-like pathology from PSP-tau-injected mice. Scale bar, 20 μm. D, Representative images of IHC for anti-tau MAb AT180, MC1, and TG3 or IF for AT8 and ThS of the ventral hilus for the three AD cases injected into non-Tg mice 3 months p.i. Scale bar, 50 μm. E, Representative images of IHC for AT180, MC1, and TG3 or for AT8/ThS of the ventral hilus for the three CBD cases injected into non-Tg mice 3 months p.i. Scale bar, 50 μm. F, Representative images of IHC for AT180, MC1, and TG3 or for AT8/ThS of the ventral hilus for the two PSP cases injected into non-Tg mice 3 months p.i. Scale bar, 50 μm. G, Representative images of IHC for AT180, MC1, and TG3 or for AT8/ThS of astrocytes (CBD-tau Case 1 injected non-Tg mice 3 months p.i.) or oligodendrocytes (PSP-tau Case 1 injected non-Tg 3 months p.i.). Scale bar, 20 μm. Hipp, Hippocampus; Ipsi, ipsilateral; Contra, contralateral.

Figure 5.
Figure 5.

PSP-tau from different brain regions have similar seeding potency in vivo. A, Representative images of IHC of AT8 on PSP-tau extracted from either the frontal cortex (0.7 μg/site; n = 4), thalamus (0.35 μg/site; n = 3), or lentiform nucleus (0.0175 μg/site; n = 3) of Case 1 injected into non-Tg mice hippocampus/cortex 3 months p.i. Red boxes with insets indicate glial tau pathology in either the fimbria (oligodendrocytes) or hippocampus (astrocytes). Scale bars: panel, 100 μm; inset, 10 μm. B, IHC using other epitope-specific anti-tau antibodies AT180, TG3, and MC1 on PSP-tau from the frontal cortex, thalamus, or lentiform nucleus of Case 1 injected into non-Tg mice HP-Ctx 3 months p.i. Double-labeling IF for AT8/ThS shows tau inclusions are ThS-positive when seeded by PSP-tau extracted from any of the three regions. Scale bars: panel,100 μm; inset, 10 μm. Hipp, Hippocampus; Ctx, cortex; Ipsi, ipsilateral; Contra, contralateral.

Figure 6.
Figure 6.

Spatiotemporal transmission patterns of seeded neuronal tau aggregates from three tau strains. A, Left, Representative images from IHC for AT8 for AD-tau- (1 μg/site; n = 4 mice), CBD-tau- (1 μg/site; n = 4 mice), and PSP-tau- (0.7 μg/site; n = 4 mice) injected non-Tg mice 6 months p.i. Right, Representative images from IHC for AT8 for AD-tau- (1 μg/site; n = 3 mice), CBD-tau- (1 μg/site; n = 4 mice), and PSP-tau- (0.7 μg/site; n = 3 mice) injected non-Tg mice 9 months p.i. Scale bar, 100 μm. B, Quantification of AT8+ neurons in ventral hilus region (ipsilateral and contralateral to site of injection) of AD (n = 3 mice for 6 and 9 months), CBD (n = 4 mice for 6 and 9 months p.i.), and PSP-tau (n = 4 mice for 6 months p.i. and n = 3 mice for 9 months p.i.). One-way ANOVA with Bonferroni post hoc test was performed to compare groups (Mean ± SEM plotted; ventral hilus ipsi: F = 7.077, df = 38, ANOVA, p = 0.0001; ventral hilus contra: F = 7.393, df = 38, ANOVA, p = 0.0001; Bonferroni post hoc: *p < 0.05, **p < 0.01, ***p < 0.001). C, Semiquantitative analyses performed on a scale of 0 (gray) to 3 (red) for neuronal tau pathology and color coded onto heat maps. Scores were averaged between mice for each group at 3 months p.i. (n = 3 for AD-tau, n = 4 for CBD-tau, n = 4 for PSP-tau), 6 months p.i. (n = 4 for AD-tau, n = 4 for CBD-tau, n = 4 for PSP-tau), and 9 months p.i. (n = 3 for AD-tau, n = 4 for CBD-tau, n = 3 for PSP-tau). Hipp, Hippocampus; Ctx, cortex; Ipsi, ipsilateral; Contra, contralateral.

Figure 7.
Figure 7.

Spatiotemporal transmission of seeded glial tau aggregates in CBD-tau and PSP-tau-injected mice. A, Left, Representative images of IHC for AT8 showing neuronal and astrocytic tau pathology in CBD-tau-injected non-Tg mice at 3, 6, and 9 months p.i. (n = 4 mice for each time point). Scale bar, 100 μm. Right, Semiquantitative heat maps of same CBD-tau-injected mice shown in IHC images on left. Neuronal tau pathology is color-coded per region (as in Fig. 4). Astrocytic tau pathology was counted for each coronal section and averaged across all mice in each group, then depicted schematically on the heat maps as red stars (1 star per 3 tau-positive astrocytes). Oligodendroglial tau pathology was counted similar to the astrocytic pathology, then represented similarly as above on heat maps as purple ovals (1 oval per 5 tau-positive oligodendrocytes). B, Quantification of AT8-positive neurons and astrocytes of CBD-tau-injected mice at 3, 6, and 9 months p.i. (n = 4 mice at each time point; hippocampus/cortex) based on stereological cell counts of two sections in each region with following bregma-based coordinates: ventral hilar neurons: −3.80 and −3.64 mm. Linear regression was performed correlating neuron tau pathology (y-axis) with astrocyte tau pathology (x-axis) in each region, with r2 and p values noted for each regression analysis in the graphs. C, Left, Representative images of IHC for AT8 showing oligodendroglial tau pathology in PSP-tau-injected mice at 3 months (n = 4 mice), 6 months (n = 4 mice), and 9 months (n = 3 mice) p.i. Scale bar, 100 μm. Right, Semiquantitative heat maps of same PSP-tau-injected mice shown in IHC images on left. Neuronal tau pathology is color-coded per region (as in Fig. 4), with astrocytic tau pathology in red stars and oligodendroglial tau pathology in purple circles as in A. D, Quantification of AT8-positive oligodendrocytes in PSP-tau-injected mice at 3 months (n = 4 mice), 6 months (n = 4 mice), and 9 months (n = 3 mice) p.i. based on stereological counts of three sections in each region at the following bregma: fimbria: −1.06, −0.94, and −0.82 mm; corpus callosum −1.06, −0.94, and −0.82 mm. One-way ANOVA with Tukey post hoc analysis was performed across the time points (Mean ± SEM plotted; Fimbria contra: F = 4.642, df = 10; ANOVA, p = 0.0459; corpus callosum contra: F = 11.40, df = 10; ANOVA, p = 0.0046; Tukey post hoc: *p < 0.05, **p < 0.01, ***p < 0.001). Hipp, Hippocampus; Ctx, cortex; Ipsi, ipsilateral; Contra, contralateral.

Figure 8.
Figure 8.

Spatiotemporal transmission of tau pathology from additional injection site in non-Tg mice. A, Left, Schematic coronal and sagittal sections of the mouse brain showing dorsal thalamus injection site (bregma based coordinates: −2.5 mm; lateral: +2 mm; depth: −3.4 mm from the skull). Right, Anatomical neuronal connectome showing anterograde and retrograde connections from the dorsal thalamus. Brain regions with seeded tau pathology are bolded. B, Representative images from IHC for AT8 for AD-tau- (6.4 μg/site; n = 3 mice), CBD-tau- (0.64 μg/site; n = 3 mice), and PSP-tau- (1.1 μg/site; n = 3 mice) injected non-Tg mice 6 months p.i. Scale bars: panel, 50 μm; inset, 10 μm. C, Semiquantitative heat maps were created as described in Figures 4 and 7 on AD-tau-, CBD-tau-, and PSP-tau-injected mice into the thalamus 6 months p.i.

Similar articles

Cited by

References

    1. Arriagada PV, Growdon JH, Hedley-Whyte ET, Hyman BT (1992) Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer's disease. Neurology 42:631–639. 10.1212/WNL.42.3.631 - DOI - PubMed
    1. Ballatore C, Lee VM, Trojanowski JQ (2007) Tau-mediated neurodegeneration in Alzheimer's disease and related disorders. Nat Rev Neurosci 8:663–672. 10.1038/nrn2194 - DOI - PubMed
    1. Boluda S, Iba M, Zhang B, Raible KM, Lee VM, Trojanowski JQ (2015) Differential induction and spread of tau pathology in young PS19 tau transgenic mice following intracerebral injections of pathological tau from Alzheimer's disease or corticobasal degeneration brains. Acta Neuropathol 129:221–237. 10.1007/s00401-014-1373-0 - DOI - PMC - PubMed
    1. Braak H, Braak E (1991) Neuropathological staging of Alzheimer-related changes. Acta Neuropathol 82:239–259. 10.1007/BF00308809 - DOI - PubMed
    1. Braak H, Del Tredici K (2012) Where, when, and in what form does sporadic Alzheimer's disease begin? Curr Opin Neurol 25:708–714. 10.1097/WCO.0b013e32835a3432 - DOI - PubMed

Publication types

MeSH terms

Substances