FPS1, FPS2, and FPS3 derived from the primary screen. FPS-ZM1 derived from the secondary screen. Each structure shows the tertiary amide group (red) and its functional moieties, hydrophobic group (blue), electron-rich aromatic group (green), and electron-poor group (pink). For details see Results.

(A) Influx of circulating ¹²⁵I-Aβ40 (1.5 nM) across the BBB in 15- to 17-month-old APP^sw/0 mice with and without anti-RAGE antibody (40 μg/ml), NI IgG (40 μg/ml), FPS2 (200 nM), or FPS-ZM1 (200 nM) alone or after preincubation with sRAGE or sLRP. Values are mean ± SEM; n = 4–6 mice per group. (B and C) Detection of FPS-ZM1 in the brain of 15 month old APP^sw/0 mice 5 minutes after an I.V. administration. FPS-ZM1 was found in RAGE-IP brain fraction (B), but not in RAGE-depleted brain (C). Representative results from 6 experiments are shown. (D and E) Aβ40 (D) and Aβ42 (E) levels in the cortex and hippocampus. (F) Thioflavin S–positive amyloid deposits in the brain (left) and quantification of thioflavin S–positive area in the cortex and hippocampus (right). Scale bar: 350 μm. (G–J) CBF response to whisker stimulation (G), NOL (H), NOR (I), and active operant response (J) in 15-month-old APP^sw/0 mice treated with vehicle, FPS2, or FPS-ZM1 for 2 months. Nontransgenic littermate controls (gray columns) are shown in panels G–J. Values are mean ± SEM. n = 4–6 mice per group.

(A) Confocal microscopy analysis of Iba1-positive microglia (green) and methoxy X04–positive amyloid-β (blue) and merged images. Scale bar: 50 μm. (B) Microglia numbers in the cortex. (C–E) Relative mRNA expression levels of Tnfα, Il1β, Il6, and Ccl2 determined by qRT-PCR (C), TNF-α (D), and IL-1β (E) protein levels determined by ELISA in the cortex and hippocampus. (F) Relative mRNA expression levels of Tnfα, Il1β, Il6, and Ccl2 in laser-captured microglia and neurons determined by qRT-PCR. APP^sw/0 mice were treated with vehicle, FPS2, or FPS-ZM1 for 2 months starting at 15 months of age. Values are mean ± SEM. n = 5 mice per group.

(A) ¹²⁵I-Aβ40 (5 nM) binding to immobilized human sRAGE in the presence of FPS2 or FPS-ZM1 (10–500 nM). (B) ¹²⁵I-HMGB1 (5 nM) or ¹²⁵I-S100B (5 nM) binding to sRAGE in the presence of FPS-ZM1 (10–1,000 nM). In A and B, K_i represents inhibitory constant. (C) ¹²⁵I-Aβ40 (5 nM) binding to immobilized human recombinant RAGE V domain (Vd) or C1C2 domain (C1C2d) with and without FPS-ZM1 (200 nM) and to sRAGE with and without RAGE-specific C1 domain (anti-C1d), C2 domain (anti-C2d), or V domain (anti-Vd) antibodies (20 μg/ml), NI-IgG, and FPS-ZM1 (100 nM). (D) Aβ40 and Aβ42 binding to immobilized human sLRP with and without FPS-ZM1 (1 μM) or RAP (1 μM). (E) Aβ40-induced (1 μM) TBARS in RAGE-CHO cells in the presence of vehicle (closed triangle) or various concentrations of FPS2 (white circles) and FPS-ZM1 (black circles). (F) Aβ40-induced (1 μM) NF-κB activation in RAGE-CHO cells with and without FPS2 and FPS-ZM1. (G–I) BACE1 mRNA (G), BACE1 protein (H), and secreted sAPPβ (I) levels determined by qRT-PCR, immunoblotting, and ELISA, respectively, in SH-SY5Y cultures treated with vehicle or Aβ40 (1 μM) with or without FPS2 or FPS-ZM1 (50 nM) after transduction with Ad. GFP or a mutant Ad. IκB-α (S32, 36A), and/or transfection with scrambled siRNA or RAGE-siRNA. All values are means ± SEM. n = 3–5 independent experiments. β-actin was used as a loading control in H.

(A–D) Bace1 mRNA (A), BACE1 protein (B and C), and sAPPβ (D) levels determined in the cortex and hippocampus by qRT-PCR, immunoblotting, and ELISA, respectively. Quantitative densitometry was used to determine relative abundance of BACE1 using β-actin as a loading control. (E) Nuclear NF-κB p65 relative levels in the cortex and hippocampus. APP^sw/0 mice were treated with vehicle, FPS2, or FPS-ZM1 for 2 months starting at 15 months of age. Non-Tg, nontransgenic littermate controls. Values are mean ± SEM. n = 5–6 mice per group.