Figure 1.
DYRK1A is required for efficient phosphorylation of pT231, pS396 and 12E8 tau.
H4 neuroglioma cells overexpressing four repeat tau (4R0N) were transfected with siRNA targeting the DYRK1A transcript. Silencing of DYRK1A was confirmed with anti-DYRK1A antibody (top panel). Percent control values represent the average of three independent siRNA transfections and westerns. NS refers to the non-silencing control. 12E8 refers to the dual phosphorylation epitope pS262/pS356.
Figure 2.
Harmine, a DYRK1A antagonist, inhibits tau phosphorylation.
Shown in (A) is the toxicity profile of harmine against the H4 neuroglioma cell line. The IC50 for viability was 12 µM. In (B), are results of a dose-response treatment of H4-tau cells with harmine at the indicated concentrations. Reductions to total tau and all three phosphorylated forms of tau tested were observed. % control values represent the amount of the respective tau forms present following treatment with 8 µM harmine, although 0.8 µM harmine also reduced phospho-tau levels significantly. Harmine treatment was performed as described in the Methods. For (C), results for moclobemide, an MAO-A selective antagonist, are shown. % control values represent the amount of each form of tau present following treatment with the highest 500 µM concentration.
Figure 3.
Multiple β-carboline derivatives affect levels of total tau and phosphorylated tau.
Each indicated compound was tested at the concentrations shown below each compound name. % control values represent the effect seen at the highest concentration tested for each compound. 9-ethylharmine clearly showed the most potent effect in this assay, significantly reducing total and phospho tau levels at a 1 µM concentration.
Figure 4.
Quantification of the inhibition of tau phosphorylation by multiple β-carbolines.
Quantification of tau phosphorylation data from the H4 cells is shown for each compound tested. The phospho-tau data have been normalized to account for any changes to total tau levels. Effects on total tau are indicated in each graph. Significance at p<0.05, as assessed by Student's T-test, is indicated by asterices above the error bars on the graphs. Error bars (standard deviation) from three independent replicates are shown.
Figure 5.
Harmine inhibits the DYRK1A catalyzed direct phosphorylation of tau protein on serine 396.
Shown in (A) are results of an in vitro phosphorylation assay utilizing recombinant DYRK1A and tau proteins. A doublet pS396 tau phosphorylation is observed only in the presence of tau, DYRK1A, and ATP. In (B), harmine potently inhibits the direct phosphorylation of tau protein by DYRK1A with an IC50 of 0.7 µM.
Figure 6.
Structurally distinct β-carboline derivatives inhibit DYRK1A-dependent pS396 tau phosphorylation with varying affinities.
Shown are in vitro phosphorylation results for all compounds in this study. The compounds tested are indicated above the respective western results for each compound. The concentrations (in µM) are indicated at the top of the first panel and are the same for each compound tested. The IC50 values calculated from these assays are indicated in the right column, next to the western results for each compound.
Figure 7.
Quantification of β-carboline affinities for inhibition of pS396 phosphorylation in vitro.
Quantification of the in vitro phosphorylation data at each drug concentration tested is shown. Error bars represent the standard deviation of three independent replicates. Significance at p<0.01, as assessed by Student's T-test, is indicated by a single asterix above the error bars on the graphs. Significance at p<0.001 is indicated by two asterices above the error bars.