Three Pillars of Automated Home-Cage Phenotyping of Mice: Novel Findings, Refinement, and Reproducibility Based on Literature and Experience

Abstract

Animal models of neurodegenerative and neuropsychiatric disorders require extensive behavioral phenotyping. Currently, this presents several caveats and the most important are: (i) rodents are nocturnal animals, but mostly tested during the light period; (ii) the conventional behavioral experiments take into consideration only a snapshot of a rich behavioral repertoire; and (iii) environmental factors, as well as experimenter influence, are often underestimated. Consequently, serious concerns have been expressed regarding the reproducibility of research findings on the one hand, and appropriate welfare of the animals (based on the principle of 3Rs-reduce, refine and replace) on the other hand. To address these problems and improve behavioral phenotyping in general, several solutions have been proposed and developed. Undisturbed, 24/7 home-cage monitoring (HCM) is gaining increased attention and popularity as demonstrating the potential to substitute or complement the conventional phenotyping methods by providing valuable data for identifying the behavioral patterns that may have been missed otherwise. In this review, we will briefly describe the different technologies used for HCM systems. Thereafter, based on our experience, we will focus on two systems, IntelliCage (NewBehavior AG and TSE-systems) and Digital Ventilated Cage (DVC^®, Tecniplast)-how they have been developed and applied during recent years. Additionally, we will touch upon the importance of the environmental/experimenter artifacts and propose alternative suggestions for performing phenotyping experiments based on the published evidence. We will discuss how the integration of telemetry systems for deriving certain physiological parameters can help to complement the description of the animal model to offer better translation to human studies. Ultimately, we will discuss how such HCM data can be statistically interpreted and analyzed.

Keywords: DVC; IntelliCage; behavior; mice; phenotyping; telemetry.

Figure 1

The figure depicts the data flow from racks and mice that via moving on the electrode grid on the DVC^® boards generate events that then are summarized and displayed through remote access in any browser of choice.

Figure 2

Temporal representation of a hypothetical experimental workflow in a core facility for testing animals with the DVC^® system and the IntelliCage. Several phases are incorporated to gather the most of information by combining two systems (bold black line showing the days from the arrival of animals, thinner pink arrow representing the different phases of monitoring—testing in the IntelliCage may contain different protocols for learning, impulsivity, taste preference, stress, et cetera as explained in the text).

Figure 3

Characteristics and setup of the IntelliCage. (A) General view of the IntelliCage and mice in the cage; (B) movable version of two cages + laptop on a trolley; (C) mice standing on the shelters and reaching the food; (D) mice entering the corners; and (E) inner view of the conditioning corner (note two holes for nose poke, closed on the left and open on the right side, with nipple visible there).