Fig 2: Oligomeric Aβ42 treatment induces GSK-3 activation and pathogenic phosphorylation of both APP and tau in primary rat cortical neurons. Primary rat cortical neurons were treated with 2.5 or 5 µm oligomeric Aβ42 for 24 h, and neuronal lysates were analyzed by Western blot using P-GSK3αβ (A) and total GSKαβ (B) antibodies. C, The bar graph represents the quantitative analysis of P-GSK3αβ levels normalized to total GSK3αβ expression. Membranes were then probed with P-Thr668 APP (D) and phospho-tau PHF-1 (F) antibodies. Finally, the membranes were reprobed with Actin antibody (H), which served as loading control. The bar graphs represent the quantitative analysis of P-Thr668 APP levels normalized to Actin (E) and PHF-1 levels normalized to Actin (G). Statistical analysis was performed using Student’s t test, and the data represent the mean ± SEM from three independent experiments (n = 3); *p < 0.05. I, Neurons were treated with indicated concentrations of Aβ and immunostained with neuron-specific MAP2 mouse monoclonal antibody and P-Thr668-APP rabbit polyclonal antibody. Immunostaining was visualized with Alexa Fluor 594 (red) and 488 (green) fluorophores, respectively. Hoechst staining was used to visualize nuclei (blue). The staining was analyzed with AxioVision Rel 4.8 software for Zeiss microscope. The corresponding bar graph (J) represents the qualitative analysis of the corrected total cell fluorescence for P-APP (n = 30 neurons/group). Statistical analysis was performed using ANOVA, and the data are representative of three independent experiments. Magnification 63×.

Fig 3: Aphidicolin inhibits Aβ-mediated Ras-MAPK activation, cyclin D1 expression, and APP phosphorylation. A, B103-695 cells were treated with or without 2.5 μm Aβ, in the presence or absence of 5 μg/ml aphidicolin or 20 μm roscovitine for 24 h. On Aβ treatment, as expected expression of Ras and Cyclin D1 as well as phosphorylation of ERK were increased. Aphidicolin treatment diminished both basal and Aβ-induced expression of Ras and Cyclin D1 as well as activation of ERK. Roscovitine did not exhibit any significant effect. Data shown are representative of two independent experiments. B–I, Primary rat neurons were treated with or without 2.5 μm Aβ in the presence or absence of 5 μg/ml aphidicolin or 20 μm roscovitine and analyzed by immunofluorescence using P-ERK and MAP2 (B), Ras and MAP2 (D), Cyclin D1 and MAP2 (F), or P-APP and MAP2 (H) antibodies. Hoechst was used for visualization of nuclei. The corresponding bar graphs represent the qualitative analysis of the corrected total cell fluorescence for P-ERK (C), Ras (E), cyclin D1 (G), and P-APP (I; n = 30 neurons/group). Statistical analysis was performed using ANOVA, and the data are representative of three independent experiments. Magnification 63×.

Fig 4: TrkA colocalizes with and binds to APP but not APPpT668. (A, B). Confocal microscopy images of untreated (CTR) and NGF‐treated (100 ng mL−1, 1 h) septal neurons are shown. Double immunofluorescence staining was performed with (A) anti‐APP C‐ter (Y188; red) and anti‐Trk (green) or with (B) specific anti‐APPpT668 (red) and anti‐Trk antibodies (green). Representative images are from 50 neurons per experimental group. Scale bar: 5 μm. (C, D). IP with a specific anti‐TrkA antibody was performed on extracts from control mouse hippocampal tissue (C) or untreated (CTR) and NGF‐treated (NGF; 100 ng mL−1, 1 h) SH slices from C57/Bl6 mice (D). Immunoprecipitates (IPs) were then probed using WB with antibodies against APP (22c11) or APPpT668. Triplicate data from hippocampal tissues and SH slices of at least 4–5 control C57/Bl6 mice per experimental group were used for the analysis. (E). IP with the anti‐APP C‐terminal antibody was conducted on hippocampal extracts from ShcC KO mice (KO; n = 4) and age‐matched control mice (WT; n = 4). IPs were probed with anti‐pTrkA and with anti‐APP (22c11) antibodies. (F). Confocal microscopy images of medial septum (MS) and hippocampal CA1 (CA1) in coronal brain sections of ShcC KO mice (n = 3 mice). Double immunofluorescence staining was performed with specific anti‐APPpT668 (red) and anti‐Trk antibodies (green). Scale bars: 25 μm. IPs, immunoprecipitate; Ft, flowthrough; Lys, lysate; WT, wild‐type; MS, medial septum; DB, diagonal band of Broca; HP, hippocampus; CC, Corpus Callosum; CA, Cornus Ammonis.

Fig 5: APPpT668 is elevated and conversely APP‐TrkA binding is reduced in the AD human hippocampus. (A). A representative WB with specific antibodies against APPpT668 and total APP on human hippocampal extracts from control (CTR), AD and other neurological disease (OND). (B, C) The graph shows the quantification of the APPpT668 level, reported as percentage of the level observed in the CTR (319.5 ± 46.1% of CTR, **P < 0.01; B) or OND group (202.4 ± 15.6% of OND, *P < 0.05; C). (D). IP with anti‐APP C‐terminal antibody was conducted on hippocampal extracts from CTR, OND and AD hippocampi. IPs were probed using WB with specific anti‐TrkA and anti‐APP N‐Ter (22c11) antibodies. (E). IP with anti‐APP C‐terminal antibody was conducted on hippocampal and cerebellar extracts from the AD2 subject. IPs were probed using WB with a specific antibody against TrkA and the anti‐APP (22c11) antibody. β‐actin was used as a loading control. CTR1‐3 = Subjects without neurological diseases. Mean age at death: 60 ± 7 years. OND1 = Huntington's Disease (HD); OND2 = spinocerebellar ataxia 17 (SCA17); OND3 = counter‐lateral hemisphere of unilateral intracerebral bleeding. AD1‐3 = familial AD (Met146Leu mutation, PS1 gene), Clinical Dementia Rating (CDR) stage 5, Braak stage VI. Mean age at death: 48 ± 8.2 years. IPs = immunoprecipitates; HP = hippocampus; Crb = cerebellum; CTR, age‐matched controls without neurological diseases; AD = Alzheimer's disease; OND = other neurological disease.