Fig 1.
Differentiation of hiPSC into cortical neurons and efficient transduction with AAV-TAU-P301L (A) Immunostaining for OCT4 and NANOG shows that iPSC0028 is pluripotent. Scale bar represents 50μm. (B-C) Immunostaining for Nestin and PAX6 revealing NPC stage at DIV25. Scale bar = 50μm for both. (D-F) Immunostaining on DIV70 visualizes the neuronal marker TUBB3 and cortical markers TBR1 and CTIP2 (D) as well as the dendritic marker MAP2 (E-F) together with either vGLUT2 (E) or vGAT (F). Scale bar = 25μm. (G-I) Representative traces of intrinsic neuronal properties of DIV70 neurons showing evoked responses in current clamp (G) as well as sodium and potassium currents (H) in voltage clamp (n = 13 cells). (I) Example of spontaneous EPSCs recorded at a holding of -65mV in the presence of 50μM PTX in voltage clamp mode. (J) Quantitative RTPCR data showing that transduced neurons express both 3R and 4R TAU mRNA, represented by an increased 4R/3R TAU ratio compared to non-transduced control cells (P = 0.04; n≥3 from different experiments). Values were normalized to PGK1 before analyses. * P<0,05 (K) Western Blot with a 4R TAU specific antibody depicts 2N4R TAU bands, only in transduced (2N4R-P301L) cells. TAU ladder and marker for band sizes are represented by (T) and (M) respectively. (L) Immunostaining with 4R TAU specific antibody confirms the presence of the 2N4R P301L TAU on the cellular level, only in transduced neurons, while total TAU (red) is present also in control neurons. Scale bar = 25 μm. DAPI stains the nuclei
Fig 2.
AlphaLISA optimizations on human brain extracts for total TAU, TAU aggregation and phosphorylation.
(A, B) AlphaLISA on 2 different AD brain extracts show high hTAU10 (A) and AT8 (B) TAU aggregation signals compared to control brain samples. (C, D) AT8/hTAU10 (C) AlphaLISA on these AD brain extracts reveals high levels of phosphorylated TAU compared to control brain samples while both AD and control brain extracts display high HT7/hTAU10 (D) levels. Decreasing signals with increasing dilutions suggest no hooking of the samples. Representative curve of 1 experiment with 2 technical replicates is shown as RFU (relative fluorescence units) ± SD. (E, F) Western Blot on soluble (S) and insoluble (IS) fractions of control and AD brain extracts after Sarkosyl extraction shows HT7-positive (E) and AT8-positive (F) bands only in the Sarkosyl insoluble pellets of both AD patients, confirming the presence of TAU aggregates. M represents Magic Marker (band sizes) and T represents TAU ladder with all 6 TAU isoforms. All experiments have been confirmed at least twice.
Fig 3.
Optimization of dynamic range of hTAU10 aggregation AlphaLISA.
(A) K18 sonication significantly improves seeding potency (P<0.001, n≥3 independent experiments). Seeds were added at week 1. In all further experiments, sonicated K18 is used. (B) K18 seeding does not induce aggregation in control (no virus) and WT virus (AAV WT TAU 2N4R) transduced neurons (P = NS; n≥3 independent experiments). (C) Weekly repeated K18 seeding of K18 significantly increases the dynamic range (P<0.001 for both 1xF versus 2xF (wk 1+2) and 1xF versus 3xF (wk1+wk2+wk3); n≥3 independent experiments). (D) Finally, also the timing of seeding has an effect on the aggregation potency. Addition of K18 at week 2 (wk2) significantly increases TAU aggregation compared to week 1 (P<0.001, n≥3 independent experiments) while addition of fibrils at week 3 shows significantly less aggregation (P<0.001; n≥3 independent experiments) probably due to the shorter (1 week) K18 incubation period before AlphaLISA. ***P<0,001; 2-way-ANOVA with Dunnett’s post hoc.
Fig 4.
K18 seeding induces TAU aggregation and hyperphosphorylation in human TAU-P301L neurons.
Optimal timelines are shown for AAV transduction, 96w final plating, K18 seeding and final assay. (A, B) AlphaLISA data show that K18 seeding induces an increase in both hTAU10 (P<0,001; A) and AT8 (P<0,001; B) TAU aggregation assays. (C, D) AlphaLISA results demonstrate around 2-fold increase in TAU phosphorylation (AT8/hTAU10; P<0,001; C) while total TAU levels remain unchanged (HT7/hTAU10; P = NS; D). (E) CellTiter-Glo® results showing that general cell health is unaffected after K18 addition (P = NS). For all assays in (A-E): *** P<0,001; 1-way ANOVA with Tukey’s post hoc; n≥3 independent experiments. (F) Representative blot of Native PAGE followed by Western blot showing two monomeric HT7-positive TAU bands (around 66kDa) in all conditions. Notably, non-migrated HT7-positive TAU proteins (>1236kDa) in K18-seeded samples suggest the presence of TAU aggregates. (G, H) Representative Western blots after Sarkosyl extraction showing soluble (S) and insoluble (IS) fractions after blotting with antibodies against total TAU (HT7; G) and hyperphosphorylated TAU (AT8; H). Aggregates are only present in the insoluble pellet after addition of 6nM or 50nM of K18 fibrils. Note the presence of monomeric 3R and 4R TAU protein in the soluble fraction. (I) Immunostaining for AT8 and neuronal HuC/D after 1%Triton/PFA fixation, to remove monomeric TAU, reveals AT8-positive neurons after K18 seeding. Scale bar = 25μm.
Fig 5.
Model validation: autophagy inducers reduce TAU hyperphosphorylation and aggregation in TAU-P301L neurons.
(A) CellTiter-Glo® data showing that rapamycin is not toxic (P = NS). (B, C) Rapamycin dose-dependently reduces hTAU10 (P<0,001; B) and AT8 aggregated TAU measured with AlphaLISA (P = 0,025 at 10 nM and P<0.001 at 1 μM; C) and compared to DMSO. (D, E) Also general TAU phosphorylation is reduced (AT8/hTAU10; P<0,001; D) to a similar extent as the reduction in total TAU (HT7/hTAU10; P<0,001; E). (F) CellTiter-Glo® results show that trehalose is highly toxic at 250 mM (P<0,001). (G, H) AlphaLISA results reveal that trehalose significantly reduces hTAU10 (P<0,001) and AT8 TAU aggregation levels versus control (P = 0,006 at 31,5 mM and P = 0,014 at 125 mM). (I, J) Only at 125 mM of trehalose, both phosphorylated TAU (P<0,001, I) and total TAU levels (P<0,001, J) are decreased. ***P<0,001; **P<0,01; *P<0,05; #P<0,001 due to toxicity; 1-way ANOVA with Dunnett’s post hoc; n≥3 independent experiments