a) GFP expression of HEK-293T cells transfected with pCMV-GIN-ZEO-shRNAmirSCR.GFP (SCRmiR). b) CDK5 immunofluorescence in HEK-293T cells transfected with SCRmiR. c) GFP expression of HEK-293T cells transfected pCMV-GIN-ZEO-shRNAmirCDK5.GFP (CDK5miR). d) CDK5 immunofluorescence in HEK-293T cells transfected with CDK5miR; arrows point to GFP+ cells in panel c. Decrease of the CDK5 immunoreactivity is observed in transfected cells. Green: GFP fluorescence, red: Alexa 594. 60×, scale bar: 20 μm. n=6. e) CDK5 Western blotting in HEK-293T transfected cells with SCRmiR and CDK5miR. A representative plot is shown. βIII-Tubulin was used as loading control. Data is presented as mean ± S.E.M. n=6, * = p<0,05. f) GFP expression of neuronal primary cultures transduced with LV- SCRmiR. g) Immunofluorescence of CDK5 in transduced neurons with LV-SCRmiR. h) GFP expression of neuronal primary cultures transduced with LV-CDK5miR. i) CDK5 immunofluorescence in neurons transduced with LV-CDK5miR. Arrows point to GFP+ transduced neurons seen in panel h. Decrease of CDK5 immunoreactivity in transduced neurons with LV-CDK5miR is observed. Green: GFP fluorescence, red: Alexa 594. 60×, scale bar: 20μm. n=6, ** = p<0,001. j) Quantification of the fluorescence intensity of CDK5 immunoreactivity of neurons transduced with LV-SCRCDK5miR and LV-SCRmiR, using the software image scope pro, Media cybernetics. RU= Relative Units, n=6, ** = p<0,001.

a) GFP expression of hippocampal neurons transfected with pCMV-GIN-ZEO.shRNAmirSCR.GFP (SCRmiR). b) Immunofluorescence of phosporylated tau with AT-8 antibody in neurons transfected with SCRmiR expressed in pCMV-GIN-ZEO-GFP plasmid. c) GFP expression of hippocampal neurons transfected with pCMV-GIN-ZEO.shRNAmirCDK5.GFP (CDK5miR). d) Immunofluorescence of phosporylated tau with AT-8 antibody in neurons transfected with CDK5miR; A decrease in AT-8 immunoreactivity was detected. Green: GFP fluorescence, red: Alexa 594. 60×, scale bar: 20 μm. n=4. e) Zoom inserts from b and d. Arrows indicate neuritic processes of transfected cells. The phospho-tau (AT-8) immunoreactivity decrease in neurites transfected with CDK5miR (GFP positive cells). f) Quantification of the fluorescence intensity of AT-8 immunoreactivity in neurons transfected with CDK5miR and SCRmiR, using the software image Scope-Pro (Media cybernetics). RU= Relative Units, n=4, * = p<0,05. g) CDK5, PHF-1, p35, and p25 were evaluated by Western Blotting in neuronal primary cultures transduced with AAV-CDK5miR and AAV-SCRmiR. β-actin was used as loading control. Representative blots are shown. h) Densitometric quantification from g, p25/p35 Ratio was calculated. Data are presented as mean ±S.E.M. p25 and p35 densitometry values were previously normalized to β-actin. n=6, * = p< 0,05. i) CDK5, PHF-1 tau, pSer199 tau and Tau-5 Western blotting from hippocampal lysates of C57BL6 mice injected with LV-CDK5miR and LV-SCRmiR. Representative blots are shown. j) Densitometry of Western blots in LV-CDK5miR and LV-SCRmiR treated mice was measured and normalized to actin. Data is plotted as the percentage of change between control (SCRmiR) and treated (CDK5miR) groups and presented as mean ±S.E.M. n=4, * = p< 0,05, **=p<0,001

3xTg-AD mice (18–23 months old) were sacrificed three weeks after injection with AAV. Silencing of CDK5 expression by CDK5miR in the Dentate gyrus, CA1, CA2 and CA3 regions of 3xTg-AD mice. GFP represent the transduced areas in the hippocampus. 15×, scale bar: 100 μm. CDK5 immunofluorescence 40×, scale bar 50 μm. n=6.

a) CDK5, PHF-1, p35, and p25 Western blotting from hippocampus of 3xTg-AD mice treated with SCRmiR and CDK5miR. β-actin as loading control. Representative blots are shown. Densytometric quantification was done, RU= Relative units. n=3, *= p<0.05. b) CDK5 Immunoprecipitation (IP), Wb: CDK5, p35 of neuronal primary cultures. A band corresponding to the IgG heavy chain was detected and used as loading control. Negative for IgG immunoprecipitation was used as an internal control. Representative blots are shown. Densytometric quantification of CDK5 and p35 proteins was done, RU= Relative units. n=3, *= p<0.05. c) CDK5 IP, Wb: CDK5 from hippocampus of 3xTg-AD mice treated with SCRmiR and CDK5miR. d) Temporal course for CDK5 kinase activity was detected by phosphorylated histone antibody. A band corresponding to the IgG heavy chain was detected. e) Product formation using CDK5 IP from hippocampi of transgenic and e) wild type mice was measured by densitometry and plotted against the reaction time. Linear regression was used and the fitting line was drawn (GraphPad Software). SCRmiR is shown as closed circles and a straight line, and CDK5miR is shown as open circles and a dotted line. Data is presented as mean ±S.E.M. n=3. *=p<0,05. g) Phospho-histone formation was calculated as indicated above in several assays where substrate concentration was varied and normalized to CDK5 densitometry in the immunoprecipitation. Reaction rates were measured as the slope of the curve and Lineweaver-burk plot is graphed. Each value is presented in the dot-plot and lineal regression was used to fit the model. CDK5miR is shown as closed circles and a straight line, and SCRmiR is shown as open circles and a dotted line.

a) Phospho-tau (PHF-1) immunofluorescence showing neurofibrillary tangles (NFTs) positive cells and b) phospho-tau (AT-8) immuoreactive cells present in the CA1 area of hippocampus of 23 month old 3xTg-AD mice three weeks post-injection with AAV-SCRmiR. c) PHF-1 immuofluorescence showing positive cells in CA1 in hippocampus of 18 month old 3xTg-AD mice, three weeks after hippocampal injection with AAV-SCRmR. d) PHF-1 immunofluorescence and e) AT-8 immunoreactive cells in the CA1 area of 23 months old and f) 18 months old 3xTg-AD mice three weeks after injection with CDK5miR. n=6, Red pseudo-color: nucleus staining with Hoechst, green pseudo-color: PHF-1 IF, Alexa 594. 60×, Bar: 50 μm; Inserts a, c, d: 100×, scale bar: 20 μm.

In this figure, left side images a), c) and e) correspond to AAV-SCRmiR and the right side b), d) and f) correspond to AAV-CDK5miR injection. Representative PHF-1 channels of hippocampal CA1 region of 23 months old 3xTg-AD mice, imaged three weeks after a) AAV-SCRmiR or b) AAV-CDK5miR injection. Inserted panels show magnified 3D visualizations of CA1 cell nuclei, generated using bioView3D software (Kvilekval et al., 2010), where, PHF-1 channel is shown in red and Hoechst is in blue. Arrows point towards cells automatically classified as high PHF-1. Plots c) and d) show average PHF-1 intensities within detected cells where point color represents the classification result, red indicates high PHF-1 and green indicates low PHF-1. The red line separates the classes. Plots e and f show the same data sorted according to the average intensity, where the plot portion colored in red is classified as high PHF-1. Note, that the CDK5miR image set was acquired at 1.5 times higher PMT voltage and offset than the SCRmiR. The bar graph g) shows a significant difference in the percentage of CA1 cells classified as high PHF-1 for both conditions (left). The density of the detected nuclei (right) shows a tendency to increase in old mice treated with CDK5miR; h) Intensity fluorescence quantification of PHF-1 also decreased after CDK5miR treatment. Data is presented as mean ±S.E.M. n=8 to controls and n=10 to treated. * p< 0,05, ** p< 0,001.