APP mouse models for Alzheimer's disease preclinical studies

Abstract

Animal models of human diseases that accurately recapitulate clinical pathology are indispensable for understanding molecular mechanisms and advancing preclinical studies. The Alzheimer's disease (AD) research community has historically used first-generation transgenic (Tg) mouse models that overexpress proteins linked to familial AD (FAD), mutant amyloid precursor protein (APP), or APP and presenilin (PS). These mice exhibit AD pathology, but the overexpression paradigm may cause additional phenotypes unrelated to AD Second-generation mouse models contain humanized sequences and clinical mutations in the endogenous mouse App gene. These mice show Aβ accumulation without phenotypes related to overexpression but are not yet a clinical recapitulation of human AD In this review, we evaluate different APP mouse models of AD, and review recent studies using the second-generation mice. We advise AD researchers to consider the comparative strengths and limitations of each model against the scientific and therapeutic goal of a prospective preclinical study.

Keywords: APP transgenic; Alzheimer's disease; App knock‐in; amyloid precursor protein; amyloid β peptide.

Figure 1. Cortical pathology, neurological symptoms, and APP mouse models of AD

There are three neurological phases leading to the onset of AD and associated cortical pathology. The first phase is preclinical AD, where Aβ accumulates in cortex without neurological symptoms. The second phase is mild cognitive impairment (MCI), where tauopathy and neurodegeneration proceed with predementia symptoms. The third phase is AD, where neurodegeneration eliminates neurons and neuronal circuits in an irreversible manner with progressively serious symptoms of dementia. As models of preclinical AD, APP‐overexpressing mice or App knock‐in mice exhibit extensive Aβ pathology without tauopathy and neurodegeneration, for which there is a preventive window of approximately two decades. Modified from Ihara and Arai (2007). The pathology shown is from the cortex of a 9‐month‐old App ^{NL‐G‐F/NL‐G‐F} mouse. Blue: Aβ; red: microglia (Iba‐1); green: astrocyte (GFAP) (Saito et al, 2014).

Figure 2. Aβ proteostasis determined by the balance of production and degradation

The balance of anabolism and catabolism determines the steady‐state quantity of a given protein in a biological system. In FAD, increased anabolism of pathogenic Aβ (Aβ₄₂ and Aβ₄₃) in cortex results in pathological deposition. In SAD, the causes of Aβ accumulation are not fully understood, but an aging‐associated decrease in catabolism is a candidate mechanism (Saido & Iwata, 2006; Hellström‐Lindahl et al, 2008).

Figure 3. Proteolytic processing of APP in wild‐type and Tg mice

(A) APP processing by β‐secretase, α‐secretase, and η‐secretase pathways, respectively. (B) Western blot analysis of APP and APP‐derived fragments in wild‐type (WT), App ^NL‐F , and APP23 mice indicated that only APP23 mice produced APP and non‐Aβ APP fragments in substantial abundance. App ^NL‐F mice overproduce CTFβ; however, App ^NL mice produce the same amount without Aβ deposition (Saito et al, 2014), therefore serving as relevant negative controls. (C) Proteins that are overproduced in APP‐overexpressing mice.

Figure 4. Analogy between cancer immunity and protective mechanism in brain

Analogy between cancer and AD. Because of cancer immunity mechanisms, it generally takes cancer stem cells more than a decade to develop pathological cancer. Human brains may possess similar protective mechanisms for AD, which can explain why it takes Aβ amyloidosis decades to induce neurodegeneration, whose identification may improve animal models and protective medications.