Between and within laboratory reliability of mouse behaviour recorded in home-cage and open-field

Background
Reproducibility of behavioural findings between laboratories is difficult due to behaviour being sensitive to environmental factors and interactions with genetics. The objective of this study was to investigate reproducibility of behavioural data between laboratories using the PhenoTyper home cage observation system and within laboratory reproducibility using different lighting regimes.

New Method
The ambulatory activity of C57BL/6 and DBA/2 mice was tested in PhenoTypers in two laboratories under near identical housing and testing conditions (Exp. 1). Additionally activity and anxiety were also assessed in the open-field test. Furthermore, testing in either a normal or inverted light/dark cycle was used to determine effects of lighting regime in a within-laboratory comparison in Aberdeen (Exp. 2).

Results
Using the PhenoTyper similar circadian rhythms were observed across laboratories. Higher levels of baseline and novelty-induced activity were evident in Aberdeen compared to Utrecht although strain differences were consistent between laboratories. Open field activity was also similar across laboratories whereas strain differences in anxiety were different. Within laboratory analysis of different lighting regimes revealed that behaviour of the mice was sensitive to changes in lighting.

Comparison with existing methods
Utilisation of a home cage observation system facilitates the reproducibility of activity but not anxiety-related behaviours across laboratories by eliminating environmental factors known to influence reproducibility in standard behavioural tests.

Conclusions
Standardisation of housing/test conditions resulted in reproducibility of home cage and open field activity but not anxiety-related phenotypes across laboratories with some behaviours more sensitive to environmental factors. Environmental factors include lighting and time of day.