Plasmid constructs of human tau. All types of human tau constructs are contained four microtubule-binding repeats but without exons 2 and 3. T4 is wild type of tau in pcDNA3.1(–). T4C3 is mimicking caspase-cleaved tau form which was deleted the last 20 amino acids at D421 in pcDNA3.1(+). Mutant constructs T4–2EC (S396E/S404E) is generated by mutating the serine 396 and 404 sites to glutamic acid to mimic phosphorylation.

Tau is degraded via the autophagic pathway. Representative immunoblot (A) and quantitative analysis (B) of intracellular tau levels. Tau constructs were transiently transfected into CHO cells. One day after transfection, 3-MA (5 mM) or trehalose (200 mM) was added to the medium immunobloted and incubated for 24 hrs. Lysates were immunobloted with tau (5A6), LC3, or β-actin antibodies. (B) The intensity of monomers and oligomers was quantitated by a densitometer and calculated as a fold to untreated control of each tau. Results are plotted as mean±SD from three independent experiments (n=3). 3-MA treatment caused the significant accumulation of oligomers in all tau species. Trehalose treatment exhibited decreased intracellular tau levels. Increased LC3-II levels indicate facilitated autophagic activity with trehalose treatment. Arrows indicate monomeric (∼50 kDa) and oligomeric (∼150 kDa) tau, respectively. ^∗p< 0.05, ^∗∗p<0.01 to each control.

GSK3 β-phosphorylated T4 is less efficient in trehalose-facilitated autophagic degradation. (A) Representative immunoblots of tau levels with enzyme-inactive GSK3β (kinase-dead (k/D), left) or constitutively active GSK3β (S9A, right), respectively. Tau and each GSK3β constructs were transiently co-transfected into CHO cells. One day after transfection, 3-MA (5 mM) or trehalose (200 mM) was added to the medium and incubated for 24 hrs. Lysates were immunobloted with tau (5A6), LC3, GSK3β, or β-actin antibodies. (B) Quantitative analysis of tau levels. The intensity of monomers and oligomers was quantitated and calculated as a fold to untreated control of each tau. Results are plotted as mean±SD from three independent experiments (n=3). In the presence of kinase-dead GSK3β, 3-MA treatment resulted in the shift of monomeric tau to oligomeric tau in all tau species and trehalose treatment showed significantly decreased intracellular tau level of all tau species. In the presence of active GSK3β, 3-MA treatment resulted in the increased accumulation of oligomeric tau in all tau species. Trehalose treatment resulted in facilitated clearance of T4C3 and T4–2EC. However, phosphorylated T4 by GSK3β was not efficiently degraded, which indicates phosphorylation hinders autophagic degradation. LC3-II levels indicate facilitated autophagic activity with trehalose treatment. Arrows indicate monomeric (∼50 kDa) and oligomeric (∼150 kDa) tau, respectively. ^∗p<0.05, ^∗∗p<0.01 to each control.

Activ me GSK3β and 3-MA treatment result in the formation of thioflavin-S-positive inclusions of tau in a co-operative manner. Each tau and active GSK3β or enzyme-inactive GSK3β constructs were transiently co-transfected into CHO cells on coverslips. One day after transfection, 3-MA (5 mM) or trehalose (200 mM) was added to the medium and incubated for 24 hrs. Cells fixed with 3% PFA were immunostained with an antibody that recognizes total tau (red) and counterstained with thioflavin-S (green). When tau and thioflavin-S staining co-localize, the merged image is yellow/orange. 3-MA treatment or active GSK3β exhibited thioflavin-S-positive tau inclusion in all tau species. Inhibition of autophagic activity and GSK3β-induced phosphorylation facilitated aggregate formation of tau in a co-operative manner. The scale bar shows 50 μm.

Co-operative facilitation of sarcosyl-insoluble aggregation with active GSK3β and 3-MA treatment. Cells were transiently transfected with each tau construct in the presence of constitutively active GSK3β or kinase-dead GSK3β. One day after transfection, the cells were treated with 3-MA for 24 hrs. Cell lysates were homogenized and then fractionated into RAB, RIRA, Sarkosyl-soluble, and Sarkosyl-insoluble fractions. After sarkosyl fractionation assay, each fraction was separated by SDS-PAGE and immunobloted with total tau antibody (5A6). Even in the absence of active GSK3β, partitioning into the sarkosyl-insoluble fraction was observed in T4C3, an aggregation-prone tau species, whereas not observed in T4 and T4–2EC. In the presence of active GSK3β, sarkosyl-insoluble aggregates were observed in all tau species and 3-MA treatment resulted in the increased amount of sarkosyl-insoluble aggregates.