Fig 2: Stability of the binding of apoE to polystyrene as a function of pH, salt and detergent concentrations. Analysis of the stability of the binding of plasma apoE to polystyrene surfaces was challenged by varying pH (A), increasing salt concentrations (B) or detergent (0–0.5% polysorbate 20 or Triton X-100) (C) and detected by polyclonal pan-apoE antibody. Light and dark gray bars represent samples analyzed in wells blocked with a BSA-based or Superblock blocking solutions, respectively. Error bars represent the standard deviation of duplicated measures.

Fig 3: Analysis of the presence/absence of the apoE4 isoform in two cohorts of 230 (study cohort) and 50 (validation cohort) plasma samples by ELISA or by a turbidimetric assay based on LEIT. (A) Analysis of the presence of apoE4 in the study cohort of 230 plasma samples by ELISA setup stratified by APOE genotype. (B) Analysis of the presence of apoE4 in the validation cohort of 50 plasma samples by ELISA stratified by APOE genotype. (C) Analysis of the presence of apoE4 in the validation cohort of 50 samples by a turbidimetric assay stratified by APOE genotype. Error bars represent median with interquartile range.

Fig 4: Binding of apoE4 to previously “blocked” ELISA plates. Recombinant apoE4 (dark gray bars), plasma from an APOE ε3/ε4 donor (light grey bars) or PBS (open bars) was allowed to interact with ELISA plate wells previously treated with different suitable blocking solutions: PBS (PBS with no blocking solution), BSA (0.25% BSA solution in 15 mM borate buffer containing 100 mM NaCl, pH 8.5), Superblock (Superblock T20 (ThermoFisher Scientific)), Polysorbate 20 (1% polysorbate 20 in PBS), Triton X-100 (1% Triton X-100 in PBS), Skim milk (5% skim milk in PBS containing 0.1% polysorbate 20), ODGP (8 mM Octyl b-D-glucopyranoside in PBS. Error bars represent the standard deviation of duplicated measures performed in two independent experiments.

Fig 5: Organoids containing either APOE3/3 or APOE4/4 microglia exposed to rApoE4 secrete pro-inflammatory IL-8; however, only organoids containing APOE3/3 microglia exhibit an increase in NfL release, a clinical biomarker of AD. (A) Organoids with either APOE3/3 or APOE4/4 microglia exposed to rApoE4 protein show comparable levels of IL-8 secretion. However, comparison of organoids with either APOE3/3 or APOE4/4 microglia exposed to rApoE4 and Aβ demonstrates that (B) only APOE3/3 MG containing organoids release neuronal cell injury marker NfL into the supernatant, despite comparable levels of IL-8 secretion. GFAP release was also measured (C), though it did not reach the level of significance via ANOVA; direct t-tests between APOE3/3 MG + rApoE3 + Aβ vs APOE3/3 + rApoE4 + Aβ were statistically significant (p = 0.0248). (D) Characterization of the cell-specific contributions to the IL-8 pro-inflammatory response in the APOE3/3 organoids exposed to rApoE4 + Aβ shows that the disease induction model is primarily microglial-driven. Furthermore, (E) microglial lipid accumulation is also induced, as compared to induction with LPS + IFNγ. ELISA measurements includes biological replicates only (n = 3 organoids per condition) 24 h after model induction; IF volumetric quantification was performed using 10× confocal imaging using 2 organoids per condition 3 days after model induction. Statistical analysis was performed using one-way ANOVA and Tukey’s (A–C,E) or Dunnett’s post hoc test (D); p-value levels of significance are as follows: *, p ≤ 0.05; **, p ≤ 0.01; ****, p ≤ 0.0001.