I. Morganstern1, D. Hu1, R. Zenowich1, L. Thiede1, D. Havas1, A. Ambesi2, M. Bansal2, T. Hanania1
1Behavioral Pharmacology, PsychoGenics Inc., Paramus, NJ, USA;
2Informatics, PsychoGenics Inc., Paramus, NJ, USA;
Aging is the leading risk factor for the chronic diseases that account for the bulk of morbidity, mortality, and health costs. Although there’s been a tremendous progress in understanding the underlying molecular processes important in the aging cascade, much remains unknown specifically with regard to the relationship of normal aging to more advanced pathological states such as Alzheimer’s disease (AD). In order to better understand this critical relationship and provide more robust pre-clinical models for drug development, the goal of the current study was two-fold; 1) to identify the prominent behavioral features related to standard aging in C57BL/6 mice using both standard and computational analyses; 2) to examine the behavioral similarities and distinctions of normal aging from that of two mouse models of AD, namely rTg4510 (Tauopathy) and APP/PS1 (Amyloidosis).
The test battery consisted of commonly used metrics for motor function (Open Field, Rotarod), depression (Forced Swim, Tail Suspension), anxiety (Marble Burying) as well as cognition (Y-maze, Fear Conditioning and Odor Habituation). Additionally, more complex behavioral patterns profiled using proprietary platforms such as SmartCube®, combined with computational analyses were utilized in order to identify an age-related behavioral signature. We specifically tested three age-groups of wild type C57/BL6 mice, young (7-10 weeks), middle-age (16-20 weeks) and old age (39-43 weeks) and AD mutant lines, rTg4510 and APP/PS1, at similar ages. The behavioral findings demonstrate age and line-dependent similarities as well as distinctions from normal aging mice. While behaviors related to motor function and depression were distinct in both AD lines compared to aging mice; behaviors related to cognition and anxiety demonstrated some similarity with that of normal aging. Using SmartCube® technology we identified distinct behavioral features that are directly-related to aging and progressed over time. Comparison of these age-related features to that of the two AD lines (rTg4510 and APP/PS1) demonstrated a clear relationship between aging and AD as well as pointed to several unique features that are specific to AD.
In summary, we demonstrate here the common and distinct features of aging and AD phenotypes using both standard and algorithm-based behavioral tests. More importantly, we are able to dissociate aging-specific behavioral features from mutant lines of Tauopathy and Amyloidosis and identify, with more precision, AD-specific features using advanced and un-biased computer vision systems. Such an approach would be extremely valuable when assessing novel potential therapeutic approaches for AD in addition to aging.