Physiological and behavioural responses of laboratory rats housed at different tier levels and levels of visual contact with conspecifics and humans

摘要

Laboratory rats are typically housed in clear or opaque cages on multi-tiered racks. Clear-walled cages allow a view of the room and facilitate visual social contact with neighbouring rats but may induce anxiety due to lack of visual cover. We hypothesized that degree of visibility of humans and conspecifics affects anxiety in individually housed rats. We evaluated physiological and behavioural responses of adult male Sprague-Dawley rats (N = 54 pairs housed in adjacent cages with the same cover treatment) to three cover levels: (1) no visual cover (clear cage walls); (2) partial cover (opaque stripes covering 60% of the wall area); (3) solid cover (100% opaque walls), and three tier levels: (1) lower; (2) middle; (3) upper, located 61, 101, and 141 cm above the floor, respectively. Chromodacryorrhea, an indicator of acute stress, was recorded following weekly body weight measurement and cage cleaning. Faecal corticosteroid, an indicator of chronic stress, was measured in Weeks 1 and 7. Behaviour during Anticipatory Reaction to Handling (ARH), Elevated Plus Maze (EPM) and Cat Odor (CAT) tests was assessed during Weeks 4 and 8. Growth rate, chromodacryorrhea, and behaviour in ARH tests were not affected by treatments. Rats housed in cages with partial cover on upper tiers tended to have lower faecal corticosteroid levels than those in clear cages on lower tiers (P = 0.08). In the CAT test, rats housed on lower tiers spent more time rearing (P = 0.046), and tended to make more visits to a cloth impregnated with cat odour (P = 0.051), suggesting that they were less anxious in this context than those housed on higher tiers. Rats in cages positioned further from the main area of human activity within the housing room showed a tendency towards increased anxiety in the EPM. When individually housed for 24 h in three interconnected cages varying in cover, rats (N = 8) spent more time resting in the opaque, than clear or partially-covered, cage units during the photoperiod (P < 0.05), and were more likely to remain in the opaque than the clear unit when people were present (P < 0.05), but otherwise used the three cage units equally. We found no evidence that rats preferentially used the discontinuous cover as a 'Venetian blind' for monitoring events outside their cage. The results confirm the value of an opaque-walled hiding/sleeping area for rat welfare, and suggest that controlling for levels of human exposure within rodent rooms could improve the external validity of biomedical research.