AZ-4217: a high potency BACE inhibitor displaying acute central efficacy in different in vivo models and reduced amyloid deposition in Tg2576 mice

Abstract

Aβ, the product of APP (amyloid precursor protein), has been implicated in the pathophysiology of Alzheimer's disease (AD). β-Site APP cleaving enzyme1 (BACE1) is the enzyme initiating the processing of the APP to Aβ peptides. Small molecule BACE1 inhibitors are expected to decrease Aβ-peptide generation and thereby reduce amyloid plaque formation in the brain, a neuropathological hallmark of AD. BACE1 inhibition thus addresses a key mechanism in AD and its potential as a therapeutic target is currently being addressed in clinical studies. Here, we report the discovery and the pharmacokinetic and pharmacodynamic properties of BACE1 inhibitor AZ-4217, a high potency compound (IC50 160 pM in human SH-SY5Y cells) with an excellent in vivo efficacy. Central efficacy of BACE1 inhibition was observed after a single dose in C57BL/6 mice, guinea pigs, and in an APP transgenic mouse model of cerebral amyloidosis (Tg2576). Furthermore, we demonstrate that in a 1 month treatment paradigm BACE1 inhibition of Aβ production does lower amyloid deposition in 12-month-old Tg2576 mice. These results strongly support BACE1 inhibition as concretely impacting amyloid deposition and therefore potentially an important approach for therapeutic intervention in AD.

Figure 1.

Structure of AZ-4217.

Figure 2.

Concentration-dependent reduction of Aβ40 generation in cortical primary neurons from guinea pig and C57BL/6 and Tg2576 mice. Concentration-dependent reduction of Aβ40 generation in cortical primary neurons from guinea pig (filled circles), C57BL/6 (squares), and Tg2576 (triangles) mice, after AZ-4217 treatment. AZ-4217 completely inhibited the Aβ40 production in the three cell systems, with the highest potency observed in the guinea pig cells. Data shown as mean ± SEM.

Figure 3.

Time- and dose/concentration-dependent inhibition of Aβ generation in plasma and brain of C57BL/6 mice. Time- and dose-dependent inhibition of Aβ generation in C57BL/6 mice after a 25 μmol/kg (diamonds), 50 μmol/kg (squares), 100 μmol/kg (triangles), or 200 μmol/kg (filled circles) oral dose. Mean ± SD observed and fitted time-plasma exposure profiles (A). Mean ± SD observed and fitted time-Aβ40 in brain profiles (B). Plasma exposure versus observed mean ± SD and fitted Aβ40 in brain (C). Mean ± SD observed and fitted time-Aβ42 in brain profiles (D). Unbound plasma versus free brain concentrations as measured in individual mice at the different doses and time points (E). Only values above LLOQ are shown, the mean levels (±SEM, n = 59) of Aβ40 and Aβ42 in the vehicle-treated groups were 4352 ± 136 and 1072 ± 31 pg/g tissue, respectively. The effect on sAPPβ was also studied and displayed maximum reduction at 6 h after a 100 μmol/kg dose (F). Data are presented as mean values ± SEM (**p < 0.01, compared with vehicle).

Figure 4.

Time- and dose-dependent inhibition of Aβ generation in guinea pig AZ-4217 was given to guinea pigs via oral gavage at doses 25 μmol/kg, 50 μmol/kg, or 100 μmol/kg and effects were observed after 3 h (A–E). Inhibition of plasma Aβ40 (A) and Aβ42 (B) seemed to reach a plateau, with Aβ42 reduced down to the LLOQ of the assay. Brain Aβ40 (C) and Aβ42 (D) were reduced dose dependently. CSF Aβ40 (E) also displayed a dose-dependent reduction. The time–response effects on brain, CSF, and plasma Aβ40 (Aβ42 not shown) of AZ-4217 were studied with 50 μmol/kg (F), as well as the free concentration of AZ-4217 in plasma and brain (G). Data are presented as mean values ± SEM (*p < 0.05; **p < 0.01, ***p < 0.001, compared with vehicle).

Figure 5.

Acute and long-term treatment of Tg2576 mice with AZ-4217. Female Tg2576 (12 months at termination) mice were treated with 200 μmol/kg AZ-4217 acutely or repeatedly once daily for 7 and 28 d, terminated 4.5 h after last dose. Significant effects on brain hAβ40 and hAβ42 were only seen after 28 d of treatment in both the soluble (A, B) and the insoluble (C, D) brain pools. Target engagement was seen already after acute treatment both on brain sAPPβ_SWE, reduced, and brain sAPPα, elevated (E). The levels of sAPPβ_SWE (triangles) and sAPPα (squares) in DEA brain homogenates were stable in female Tg2576 mice between 3 and 24 months of age (F). Data are presented as mean values ± SEM (*p < 0.05; **p < 0.01, ***p < 0.001, compared with vehicle). The mean levels (pg/mg tissue ± SEM, n = 72/77) in the vehicle-treated groups were 396 ± 26 (soluble hAβ40), 207 ± 9 (soluble hAβ42), 2399 ± 193 (insoluble hAβ40), and 2961 ± 148 (insoluble hAβ42).

Figure 6.

Effects on endogenous mAβ in 12-month-old Tg2576 mice treated with AZ-4217. In the same Tg2576 brain extracts, as hAβ were measured, we also evaluated the effects on endogenous mAβ. Significant reductions of soluble endogenous brain mAβ40 and mAβ42 were observed after 7 d of treatment (A, B). As with hAβ-insoluble endogenous brain mAβ40, hAβ42 only displayed significant effects after 28 d of treatment (C, D). Data are presented as mean values ± SEM (*p < 0.05; **p < 0.01, ***p < 0.001, compared with vehicle). The mean levels (pg/mg tissue ± SEM, n = 77) in the vehicle-treated groups were 8 ± 0.5 (soluble mAβ40), 8 ± 0.4 (soluble mAβ42), 53 ± 3 (insoluble mAβ40), and 88 ± 5 (insoluble mAβ42).