Table 1.
Demographics and descriptive parameters of AD cases.
Fig 1.
Progression rates of AD, conformational profiles of insoluble tau protein, and schematic representation of experimental design.
(A) Kaplan-Meier cumulative survival analysis and duration of disease in pathologically verified AD (n = 22). (B) Conformational diversity of brain-derived Sarkosyl-insoluble tau in the cortex of AD determined in the brain tissue homogenates (n = 22) with conformational stability assay (CSA) [3]. The line and shade represent mean ± S.E.M. (C) Schematic representation of photochemical hydroxylation method and the antibodies used for monitoring different domains. The tau protein domains are not to scale with microtubule-binding repeats exaggerated. Features within the tau proteins and antibody epitopes: (N) inserts of acidic N-terminal domain (yellow); (P) proline-rich domain (green); (R) microtubule binding domains (MTBDs, blue) with four imperfect repeat regions separated by flanking sequences are not to scale, with microtubule-binding repeats exaggerated; C-terminal tail (grey); Eu, europium label; K18-recombinant tau fragment; RT QuIC- Real-Time Quaking-Induced Conversion. (d) Primary mouse cortical neurons inoculated at 7 DIV for 14 days and RA-differentiated SH-SY5Y cells inoculated for 3 days with 6 different AD brain-derived tau were applied as cell-based systems to evaluate tau seeding and transmission rates by conformation-dependent immunoassay (CDI), western blot (WB), and confocal laser scanning microscopy (CLSM); PID – post-inoculation days.
Fig 2.
Half-lives of different MTBDs and C-tail epitopes in tau isolated from frontal cortex of AD cases with different disease duration.
(A,B) Each curve and shade are an average ± S.E.M. obtained from duplicate photochemical footprinting experiments for Sarkosyl-insoluble tau isolated from one AD case (n = 22). The domains are labeled as outlined in Fig 1 with the epitope amino acid sequence in the parenthesis.
Fig 3.
Variable seeding potency and propagation rates of different conformers of tau.
(A) K18 substrates seeded with AD (n = 20) brain samples and monitored in real time by thioflavin T fluorescence; the samples were diluted 104-fold, and the curves are average of Thioflavin T fluorescence intensity in four wells at a given time point and unseeded K18 constructs were applied as negative controls of spontaneous aggregation. (B) Seeding activity of brain-derived AD tau (n = 20) expressed as lag phase (hrs) per ng; the lag phase was determined in four independent RT-QuIC seeding experiments. Each box encloses 50% of the data with median value displayed as a line, whiskers mark the minimum and maximum, and individual point indicate an outlier value outside the UQ + 1.5*IQR interval, where UQ is upper quartal, and IQR is inter-quartal range. Statistical significance was determined with ANOVA. (C) Linear regression model of seeding activity dependency on R4 domain half-life in individual AD brain samples (n = 20). (D) Representative confocal microscopy images of accumulating aggregates of mouse tau in mature neurons exposed to a single dose of AD brain-derived human tau (45ng/well) fourteen days earlier. The neurons were stained with antibodies specific to mouse tau (red), somatodendritic marker MAP2 (cyan), and axonal neurofilaments (SM1312, purple). (E) Regression analysis of concentration of Sarkosyl-insoluble tau accumulating in neurons and R4 domain half-life of AD brain-derived tau (n = 6). The neuronal accumulation of Sarkosyl-insoluble tau was measured by CDI in three different experiments.
Table 2.
Regression analysis of the impact of MTBD half-life on lag phase of seeding in RT QuIC.
Fig 4.
Propagation of tau protein misfolding in RA-differentiated SH-SY5Y cells expressing mature human tau protein.
(A) The timeline of RA-induced differentiation of SH-SY5Y neuroblastoma cells and their inoculation with AD brain-derived tau. (B) LDH assay of RA-differentiated SH-SY5Y cells inoculated with AD brain-derived tau in three concentrations (5, 15, and 30 ng/ well in 96 well-plate) and with controls (other neurological disorders, OND) in corresponding volume as tau per well. The LDH levels in media were measured from at least five wells in two independent experiments. Each box encloses the data with mean value ± SEM and individual points indicate N-fold change to non-treated wells (medium only, Ctrl). Statistical significance was determined with one-way ANOVA. (C) Western blot with triton-soluble and insoluble fractions of cell lysates of differentiated SH-SY5Y cells inoculated with AD brain-derived tau or non-treated (medium, Ctrl) for 1 and 72 hours. (D) Representative confocal images of max intensity projections of 15 z-stacks (0.40 µm each) of RA-differentiated SH-SY5Y inoculated with AD brain-derived tau (30ng/well) for 1 hour and 3 days, OND (* corresponding volume added) for 3 days and cells in medium only. Neuronal marker Tuj1 (grey), human tau (clone RTM49, aa2-44, green) in 1% Triton-fixed cells, scale bar is 50 µm. (E) Integrated Density of tau fluorescence expressed as x-fold to medium levels were measured in wells inoculated with AD brain-derived tau for 1 h and 3 d in three concentrations (5, 15, and 30 ng) and with OND control added in corresponding volumes. Cultures inoculated with no inoculates (lipofectamine) were included as controls. Each box encloses the data with mean value ± SEM and individual points indicate N-fold change to non-treated wells (medium only, Ctrl). Statistical significance was determined with student t-test (P-values: ** P ≤ 0.01, **** P ≤ 0.0001). At least three areas from two wells in three independent experiments were captured. (F) Representative stimulated emission depletion (STED) microscopy images of max projections of 12 z-stacks (0.35 µm each) after deconvolution and brightness corrections show actin (red), PSD95 (magenta), and human tau (green) in non-treated cultures (medium, control) and inoculated with AD brain-derived tau. Scale bar is 20 µm, cropped images 10 µm.
Fig 5.
Transmission rate of human tau misfolding and calcium homeostasis in differentiated SH-SY5Y cells inoculated with AD brain-derived tau.
(A) Representative confocal images of max intensity projections of 15 z-stacks (0.40 µm each) of RA-differentiated SH-SY5Y inoculated with AD brain-derived tau (30ng/well) for 1 hour and 3 days, OND (* corresponding volume added) for 3 days and cells in medium only. Neuronal marker Tuj1 (grey), tau (clone 16040D, aa269-281, green) in 1% Triton-fixed cells, scale bar is 50 µm. (B) Integrated Density of tau fluorescence expressed as x-fold to medium levels. Each box encloses the data with mean value ± SEM and individual points indicate N-fold change to non-treated wells (medium only, Ctrl). Statistical significance was determined with student t-test (P-values: * P ≤ 0.05, ** P ≤ 0.01, *** P ≤ 0.001). At least three areas from two wells in three independent experiments were captured. (C) Linear regression analysis of levels of IntDensity fluorescence of tau in differentiated SH-SY5Y and concentration of Sarkosyl-insoluble tau accumulating in mouse primary neurons both inoculated with AD brain-derived tau (n = 6). (D) Non-linear regression analysis of levels of IntDensity fluorescence of tau in differentiated SH-SY5Y inoculated with AD brain-derived tau for 3 days and R4 domain half-life of AD brain-derived tau (n = 6). (E) Representative confocal images of max intensity projections of 15 z-stacks (0.40 µm each) of RA-differentiated SH-SY5Y inoculated with AD brain-derived tau (30ng/well) for 1 hour and 3 days, OND (* corresponding volume added) for 3 days and cells in medium only. Neuronal marker Tuj1 (grey), hyperphosphorylated tau (clone AT8, red) in 1% Triton-fixed cells, scale bar is 50 µm. (F) Integrated Density of AT8-positive tau fluorescence expressed as x-fold to medium levels. Each box encloses the data with mean value ± SEM and individual points indicate N-fold change to non-treated wells (medium only, Ctrl). Statistical significance was determined with student t-test (** P ≤ 0.001). At least three areas from two wells in three independent experiments were captured. (G) Representative confocal images of nuclei staining (Hoechst 33342, blue), baseline Fluo-8 immunofluorescence at 180 s time point and at the fluorescence peak at 210 s, 30 s after KCl-induced calcium influx. Scale bar is 50 µm. (H) Non-linear regression analysis of Fluo-8 fluorescence intensity in ratio to baseline levels at max peak (medium controls, t = 210 s) in differentiated SH-SY5Y inoculated with AD brain-derived tau for 3 days and R1 domain half-life of AD brain-derived tau (n = 6).
Fig 6.
The role of variable structural exposure of fourth repeat tau domain (R4, red) within the MTBDs in the propagation of tau conformers in neuronal cultures.