Fig. 3: Tau strain was inefficient in cross-seeding non-corresponding isoform.

a Schematic showing the cross-seeding experiment paradigm whereby different tau strains extracted from human tauopathy brains were injected into the hippocampus and overlying cortex of either 4R tau-expressing (WT) or 3R tau-expressing (T44mTauKO) mice, respectively, to examine whether tau strains could cross-seed non-corresponding tau isoforms. The injection dose for strain_P0 were 1 μg per site. b representative IHC staining with MAb AT8 on brain sections from WT mice injected with control brain lysate at 9 m.p.i., and the brain sections from T44mTauKO mice injected with control brain extracts at 6 m.p.i. No significant tangle-like tau pathologies were observed in any of mice injected with the control lysates. c Representative IHC staining with AT8 on brain sections from WT mice injected with similar amounts of distinct human tau strains (P0) at 3 and 6 m.p.i. d Representative IHC staining with AT8 on brain sections from T44mTauKO mice injected with similar amounts of different human tau strains (P0) at 3 and 6 m.p.i. Note that AD-tau induced most abundant tau pathologies, including massive neuropil-like tau pathologies, which masked the NFTs in the hippocampal DG regions. Inserts in b–d are images with lower magnifications. Quantification of the AT8-positive cells in WT or T44mTauKO mice, respectively, injected with different human tau strains are shown in e for 3 m.p.i. and f for 6 m.p.i. time points. Both ipsilateral and contralateral sides were quantified together for each mouse. Data are presented as mean ± s.e.m. n = 3–7 in each group, and each dot represents a mouse. One-way ANOVA with Sidak’s multiple comparisons tests were performed. *p < 0.05; **p < 0.001; ***p < 0.001. Source data is available as a Source Data file.