Panel A: Representative immunoblots from mock and p25 stably transfected SH-SY5Y cells after induction. The panel shows p25, cdk5, pTau(S235) (a cdk5 substrate), BACE1 (∼70kDa), full length APP (APP fl∼110 kDa) and APP-CTFs including α-CTF (C83), and β-CTF (C99). Brain lysate from a BACE1 knock-out mouse was run alongside to confirm signal specificity (not shown). Data are shown in triplicate. Panel B: FRET assay for BACE1 activity in lysates from mock (MK) and p25 stably transfected (p25) SH-SY5Y cells. Controls for assay validation included a BACE1 transfected HEK-293 cell line (B1). All samples were assayed with, or without EMD 565788, a specific BACE1 inhibitor (Inh). The signal from p25 cells was significantly higher than similarly treated mock cells at all time points after 60 minutes (p<0.01). Panel C: ELISA results for Aβ40 and 42 from 18h-conditioned medium from mock transfected and induced p25 cells. Panel D: The level of BACE1 mRNA from mock and p25 SH-SY5Y cells was quantified by qPCR using a 3′ probe. Data are shown following normalization to actin mRNA levels. Similar results were obtained using the 5′ probe (data not shown). All results were analyzed with Student's t-test, *p<0.05 ** p<0.01 *** p<0.001.

Panel A: Schematic of the 3.2kb BACE1 promoter/luciferase fusion clone, indicating the position of +1 transcription start site. PC12 cells, stably transfected with the BACE1-pGL4.14 construct, were transiently transfected with a p25-GFP expression construct, a cdk5 expression construct, or mock vector. P25-GFP over-expression increased activity of the BACE1 promoter as determined by firefly luciferase activity. All cells were co-transfected with pRL-SV40 renilla luciferase control vector. Cdk5 over-expression without activator did not significantly affect activity of the BACE1 promoter. Over-expression of p25-GFP and cdk5 did not result in significant toxicity as determined by LDH assay (data not shown). Panel B: Schematic of the BACE1P6 and BACE1P8 CAT fusion clones, indicating the position of +1 transcription start site. Normalized CAT levels in N2a cells transiently transfected with p25/GFP or mock vector, and co-transfected with promoter constructs P6/CAT or P8/CAT. Levels of CAT were normalized to co-transfected GFP levels, which were similar among all groups. Data show n=3 wells per cell group. Transfections and assays were repeated in triplicate with essentially similar results. Panel C: An illustration of the BACE1 promoter region and 5′–UTR from −1056 to +364. Positive and neutral non–induced functional elements are indicated. The positions of the BACE1P6 and BACE1P8 inserts are indicated, as are the locations of putative STAT1/3/6 and MEF2 transcription factor binding sites. Note the absence of all putative sites in the BACE1P8 sequence. ** p<0.01.

Panel A: Representative immunoblots of BACE1, STAT3, pSTAT3(Y705) and tubulin from mock and p25 stably transfected SH-SY5Ycells. Panel B: Immunoblots of mock SH-SY5Y cells following treatment with interferon-α (IFN-α) and/or a cell permeable STAT3 inhibitor (Inh). Panel C: Statistical analyses of the level of BACE1 and pSTAT3(Y705) are represented as bar graphs which show an increase in relevant protein levels following IFN-α treatment, and a decrease following exposure to the STAT3 inhibitor. Panel D: Images of agarose gels following ChIP using two different sets of primers (BACE1-U, BACE1-D). PCR was performed following immunoprecipitation with, or without (No Ab), the STAT3 antibody in the presence, or absence of IFN-α. Total sheared DNA from the same cells (Input) was used as a positive control. Panel E: ChIP samples were also examined by qPCR using the BACE1-U primers, with the relative expression levels graphically represented following normalization. Vehicle treated cells immunoprecipitated with STAT3 antibody expression was set to1. ** p<0.01 *** p<0.001.

Panel A: Representative immunoblot analysis of pTau(S235), BACE1, STAT3, pSTAT3(S727) and tubulin (loading control) in brain lysate from Ntg and p25 mice (n=10−11 for each group, two mice shown). BACE1 and pSTAT3/STAT3 levels were significantly higher in p25 mice compared with Ntg mice, as shown in the scatterplots. Panel B: Immunoblot analysis of cdk5 deficient mice compared with wild type mice. Reduced levels of BACE1 and pSTAT3(S727) were observed in cdk5 null mice (KO), compared with wild type control mice (WT); tubulin is shown as loading control. Several embryonic (day 16.5) brains were pooled to generate lysate. Panel C: Immunoblot analysis of BACE1 levels in mice with targeted ablation of the cdk5 responsive site at STAT3(727) (SA). STAT3 SA mice showed no detectable pSTAT3(S727). BACE1 was significantly reduced in homozygous SA mice at 4 days of age compared to strain and age matched controls (n=5 for SA mice, n=6 for WT, two mice for each shown). * p<0.05.

Panel A: Representative immunoblots of pTau(S235), BACE1, STAT3, pSTAT3(S727) and tubulin in brain lysate from vehicle (V) and CP-681301 (CP) treated Ntg mice. BACE1 mRNA and protein levels, together with pSTAT3(S727)/STAT3 and pTauS235 are shown as scatterplots. N=6 in each group. There was a significant difference (p<0.05) between vehicle and CP-681301 treated mice for the proteins of interest. Panel B: Three groups of young (5 day old) mice were analyzed for Aβ and BACE1 levels: Ntg mice treated with vehicle (Veh; n=7), p25 mice treated with vehicle (p25/veh; n=7) and p25 mice treated with CP-681301 (p25/CP; n=6). Aβ40 and 42 levels were elevated in p25 mice relative to Ntg mice. Administration of CP-681301 to p25 mice significantly reduced both peptides (p<0.01). BACE1 was elevated in p25 mice relative to Ntg mice but was significantly reduced by CP-681301. The relative activity of cdk5 was shown by phosphorylation of Tau at the S235 site. Total tau levels indicated that equivalent levels of protein were loaded. All results were analyzed with Student's t-test, * p<0.05 ** p<0.01.

Panel A: ELISA analysis of soluble murine Aβ40 and Aβ42 from Ntg and p25 mice at 18−22 months of age. Levels of both Aβ40 (p<0.01) and Aβ42 (p<0.05) were significantly increased in p25 mice compared to Ntg controls. P25 mice also had significantly elevated BACE1 mRNA and protein levels compared to Ntg. Representative immunoblots of p35/25, BACE1, APP-full length, APP α-CTF and β-actin (loading control) from Ntg and p25 mice are shown. Panel B: Effects of the cdk5 inhibitor CP-681301 on Aβ and BACE1 protein levels in adult Ntg mice. Levels of Aβ40 (p<0.01), Aβ42 (p<0.001) and BACE1 (P<0.05) were significantly reduced in CP-681301 treated (CP) mice, compared to vehicle treated (V) controls. A representative immunoblot of BACE1 and β-actin (loading control) is shown in the right hand panel. * P<0.05 ** p<0.01 *** p<0.001.

Panel A: APP processing was examined in double-transgenic p25/APP mice and single transgenic APP littermates. Levels of full length APP (APP fl) were not altered but there was an increase in the level of β-CTF (C99) relative to α-CTF (C83) in the p25/APP mice compared to controls (immunoblot shows bands recognized by antibody C1/6.1 from an intact membrane). Panel B: The level of α-secretase cleaved sAPPα fragments were unchanged in p25/APP mice compared to APP controls, but the fragments generated by BACE1 cleavage (sAPPβ) were significantly increased in p25/APP mice compared to controls. BACE1 was also increased in the double-transgenic mice. β-actin was used as a loading control (n=7 p25/APP, n=6 APP, 2 mice of each group shown). Panel C: Scatterplots of sAPPβ and BACE1 levels (values are normalized to APP group average = 1) were significantly higher (p<0.05) in the p25/APP mice than the single transgenic group. * p<0.05.