Extended Data Fig. 1: NF-κB pathway-mediated B2M overproduction in activated microglia.
a, Quantification of Aβ42 levels in human cortex by ELISA (AD, n = 27; control, n = 8; two-sided unpaired Mann Whitney test). b, c, Immunoblot (b) and quantification of B2M, APP and β-actin (as loading control) in the hippocampus from 3- to 15-month-old 5×FAD mice and WT mice (n = 3 mice per group; B2M: two-sided two-way ANOVA followed by Bonferroni’s multiple comparisons test; comparing each month with the 3-month-old WT). d, Strategy of humanized B2M knock-in (B2MKI/KI) mouse generation. e, Quantification of body weights of 2-month-old WT and B2MKI/KI mice (WT, n = 7 mice; B2MKI/KI, n = 8 mice; two-sided unpaired t-test). f, Representative immunostaining for microglia (Iba1), astrocytes (GFAP), neurons (TUJ1) and B2M (HA) in hippocampal CA1 region of 2-month-old B2MKI/KI mice (expressing human B2M with a C-terminal HA tag). g, Quantification of the percentages of double-positive cells in each cell type (microglia, n = 458 cells; astrocyte, n = 550 cells; neuron, n = 370 cells). h, Experimental schematic of B2m mRNA expression analysis in primary microglial cultures. i, Relative B2m mRNA expression levels in primary microglia under differing treatment conditions are shown in (h) (Control and oAβ, n = 3 biologically independent samples; LPS, n = 4 biologically independent samples; two-sided one-way ANOVA followed by Tukey’s multiple comparisons test). j, Immunohistochemical analysis of B2M (anti-B2M antibody) and Aβ (thioflavin S) colocalization in 5×FAD mouse hippocampal sections. Similar results were observed in three independent experiments. The data represent means in (f) and mean ± s.e.m. in (a, c, g and i). The P values are indicated on the graphs. For detailed statistical information, see Source Data.
