Distinguishing pathological neurodegeneration from normal biological aging is a fundamental challenge in preclinical neuroscience. Presented at AD/PD 2023 in Gothenburg, this poster — developed in collaboration with Sylics and VU University Amsterdam — establishes a novel method for discriminating neurodegenerative phenotypes from healthy aging using automated home-cage behavior analysis. 

Das spontane Verhalten wurde über einen Zeitraum von 2,5 Tagen mithilfe von 58 Parametern inPhenoTyper™-Heimkäfigen unter Verwendung der AHCODA™-Software systematisch erfasst. Eine Hauptkomponentenanalyse (PCA) über mehrere Mausmodelle für Neurodegeneration hinweg – darunter Modelle für die Alzheimer-Krankheit und Tauopathien – ergab modellspezifische Verhaltensmerkmale, die sich bereits vor dem Auftreten der klassischen pathophysiologischen Krankheitsmerkmale deutlich vom natürlichen Alterungsprozess unterscheiden ließen. Diese Plattform bietet ein sensibles, hypothesenunabhängiges Werkzeug zur Charakterisierung von Verhaltensmerkmalen der Neurodegeneration und zur Erprobung neuartiger therapeutischer Behandlungen an sich frei bewegenden Tieren.

Aging is the most common risk factor for Alzheimer’s disease and most other neurodegenerative disorders. Neurodegenerative diseases and natural healthy aging both show gradual deterioration in memory, cognition, and circadian rhythm. Studying features that separate neurodegenerative phenotypes from biological aging allows deeper understanding of disease pathophysiology and its potential treatments. Here, we establish a novel method to investigate complex patterns of spontaneous behavior to identify hallmarks of neurodegeneration, allowing its discrimination from natural healthy aging. 

We performed systematic analysis of spontaneous home-cage behavior over 2.5 days (3 dark phases, 2 light phases) in widely used mouse models for neurodegenerative diseases at different ages. Mice were tested for 58 behavioral parameters in automated PhenoTyper™ home-cages, further analyzed using AHCODA™ software. Mutant mice were compared to wildtype littermates; compound-treated mice were compared to vehicle controls. Principal component analysis (PCA) was performed using R (FactoMineR package; 58 parameters). 

PCA identified four behavioral clusters: activity during the light phase, activity during the dark phase, dark/light behavior change, and long sheltering/resting behavior. Different Aβ- and tau-related transgenic lines showed distinct phenotypes, with phenotypes generally becoming more pronounced at later timepoints. Natural aging mice showed a clear age-related progression clearly discriminable from neurodegenerative disease models. 

Home-cage behavioral parameters are highly sensitive markers of neurodegenerative disease, detectable before classical pathophysiology emerges, and clearly distinguishable from normal aging. This approach allows discrimination between disease-carrying mutants, compound-induced models, and naturally aging mice — providing a platform to verify treatment efficacy in freely moving animals.