Fig. 3

Label-free quantitative proteomics reveals common and distinctive alterations in iPSC-based AD models and postmortem hippocampi from AD patients a and b Characterization of amyloid-β deposition and phosphorylation of tau protein in postmortem hippocampi of AD patients and non-demented controls. Immunostaining for H31L21 show presence of Aβ deposits in postmortem hippocampi of AD patients. H31L21 immuno-positive area was quantified relative to the total area. Immunostaining for AT8 show increased tau phosphorylation in postmortem hippocampi of AD patients. AT8 immuno-positive area was quantified relative to the total area. Results are presented as mean ± S.E.M. N = 3–4 animals. P value: **** p < 0.0001. Statistical analysis by t-test. CHPMBT: control human postmortem brain tissue; ADHPMBT: AD human postmortem brain tissue. c Dot blot immunoassay performed using OC antibody specific to amyloid fibrils showing their presence in postmortem hippocampi of AD patients and non-demented controls, but not in human grafts (b). Scale bars: 200 µm. d Western blotting analysis of phosphorylation of tau protein with actin blot included as a loading control. N = 3–4 human samples. e Comparison of the biological pathways dysregulated in AD iPSC-derived human spheroids (ADHS), AD iPSC-derived human grafts, and AD human postmortem brain tissues with respect to the control conditions (p-value < 0.01). p-values are denoted as the –log10 (p-value). The grey color symbolizes the absence of that specific biological pathway. f Principal component analysis of the normalized protein intensities that participate in key signaling pathways affected in AD. The proteomic profiles of each biological pathway are unique for each AD iPSC-derived model and PMBT, showing a clear separation between them. g Selected proteins significantly altered across the different AD models. Median fold-change (AD/Control) of normalized intensities; One-way ANOVA p-value < 0.05; Benjamini–Hochberg FDR < 0.05); Tukey HSD FDR < 0.05. h Quantification of dysregulated Rab proteins in each AD model. Bar graphs show the median of normalized protein intensities. Statistical significance was assessed by a two-tailed t-test; p-value < 0.05; Benjamini–Hochberg FDR < 0.05 and the log2 fold-change cutoffs of ± 0.5. i Quantification of dysregulated markers of senescence across the different AD models. One-way ANOVA p-value < 0.05; Benjamini–Hochberg FDR < 0.05; Tukey HSD FDR < 0.05