Long‐lasting housing environment manipulation and acute loss of environmental enrichment impact BALB/c mice behaviour in multiple functional domains

Abstract Understanding environmental influences on individuals' behaviour is challenging. Here we have investigated the housing impact of 9 weeks of enriched environment (EE) and social isolation (SI) and the impact of abrupt deprivation of EE (enrichment removal: ER) on BALB/c mice. Compared with the widely used C57BL/6 strain in research, BALB/c synthesises serotonin less efficiently due to a genetic variation and thus may potentially represent human populations at higher risk of stress‐related disorders. We assessed the effects of EE and SI by conducting a behavioural test battery and the effects of acute ER by monitoring homecage activities and social behaviour. We found that EE and SI impact BALB/c's physiological states and behavioural performances from lower to higher cognitive processes: increased body weight, increased rectal temperature, altered performance in motor and sensory tasks, the activity level in a novel environment and altered performance in tests of anxiety‐like behaviour, stress‐coping strategies and learning and memory. Furthermore, acute ER triggered stress/frustration‐like behaviour in BALB/c, with increased aggression, increased social distancing and disrupted daily/nightly activities. Our results demonstrate that long‐lasting housing manipulation such as EE and SI, impact behaviour via multilayered processes over a wide range of functional domains, and unforeseen change to a negative environment, ER, is a major stressor that causes behavioural and psychological consequences through environment–gene interactions, a model of direct relevance to human health.

recorded (max 60 seconds). Grip strength was measured by a Newton scale with a small wire mesh (W 35 mm, D 43 mm, O'HARA, Tokyo, Japan). Animals gripped the mesh by their forelimbs, and they were gently pulled with their tails until they could not grip it. The best score from three trials was recorded. Epilepsy test was performed with the experimenter juggling a bunch of keys for 2 seconds above animals to test the possibility of seizures (acute convulsion, rigidity or faint) caused by loud sounds.

Crawley's Social Interaction Test
The walls of chambers were transparent and had holes (W 50 mm, D 30 mm) which allowed animals to move freely among the three chambers. Two small wire cages (the shape of bottom was 1/4 circle of 100 mm radius, H 100mm) for stranger male C57BL/6N mice (seven-BEHAVIORAL IMPACT OF ENVIRONMENTS ON BALB/C MICE 2 eight weeks of age) were placed in the inner left corner of the left chamber and in the inner corner of the right chamber, respectively. Chambers were illuminated at 6 lux. Subject animals were positioned in the center chamber when the experiments were started. Subject animals were allowed to move freely during the test. The total distance traveled and time spent in the proximity of cages (60 mm around cages) by the tested animal was recorded.

Prepulse-Inhibition Test
Animals were constrained in a transparent cylinder (25 mm in diameter) during the test.
The box was illuminated at 3000 lux with under 50 dB background white noise. First, animals were habituated to the test cylinder and box for 300 seconds. Next, acoustic startle responses were measured by acoustic stimuli in the startle stimulus trials. Each stimulus was presented four times (total 16 stimuli) and average startle responses were recorded. Intervals between each stimulus were randomly selected from 5, 10, 15, 20, 25 seconds to prevent animals from predicting the timing of stimuli. Subsequently, prepulse inhibition of acoustic startle responses was measured in the prepulse inhibition trial. Each pair was presented four times (total 16 pairs) and average startle responses were recorded. The prepulse stimuli sound lasted 100 msec before the startle stimulus. Intervals between each paired stimulus were randomly selected from 5, 10, 15, 20, 25 seconds. Amplitudes were sampled from 400 msec following startle stimuli and averages of amplitudes of same four stimuli were recorded. The percent of prepulse inhibition was calculated as follows; PPI (%) = (amplitude to 120 dB in the startle stimulus trial -amplitude in the prepulse inhibition trial) / amplitude to 120 dB in the startle stimulus trial × 100.

Barnes Maze
The arena was 760 mm above the floor and illuminated at 1000 lux. A black escape box (W 160 mm, D 120 mm, H 60 mm) filled with white paper bedding material was placed under one of the holes as the target hole. The position of the target hole was fixed during individual tests and positions of target holes were randomly chosen for different individuals. Four spatial cues (a big blue rectangular, yellow sphere, red quadrangular pyramid, and black coil) were hung from the ceiling in the four corners of the experimental room. The arena was rotated every day to avoid animals using olfactory or proximal cues in the arena. The training session consisted of 16 trials (1 trial/day, 5 minutes). Animals were placed in the center area inside of white opaque cylinder (110mm in diameter, H 168 mm), and when the experiments started, the cylinder was taken out. The starting point was randomized within the center area. Animals moved freely on the arena during the test. The total distance traveled and the latency to enter the targeted hole was recorded. When animals did not enter the target hole within 5 minutes, they were guided to the target hole and left there for 30 seconds. The latency was recorded as 300 seconds in this case. One day after the training session, the first probe test for 5 minutes was conducted in the absence of the target hole, to confirm that the animals were only guided by the distal spatial cues. After the first probe test, one training trial was conducted as a retraining. A second probe test was conducted one week after the first probe test with the same procedure. Time spent around holes was recorded. Missing values in data were complemented by means of the value of the previous trial and next trial. When missing values were in the data of trial 16, they were complemented by the values of trial 15. If missing values were in sequential trials, the data was excluded from analysis. Here data had some missing values due to administrative failures in the training session (14 missing points/ total 445 data collecting points), and processed as mentioned above.

Fear-Conditioning Test
On the first day, conditioning was conducted in a test chamber illuminated at 100 lux (W The chamber was illuminated at 10 lux. Animals could move freely in the chamber during the test for 6 minutes. First, animals were habituated to the chamber 180 seconds (pre-cue phase).
Second, 180 seconds of 55 dB white noise was presented (180-360 second) without foot shock (cue phase). The percent of immobile time (freezing) was recorded.  Semi-EE where EE animals were kept during the behavioral test battery.

Figure S2
The Number of Vertical Activities in the Open-Field Test. Error bars represent standard errors of the mean. Asterisks represent adjusted p < .05.

Figure S3
The Latency to Enter the light chamber in the Light-Dark Box Test. Error bars represent standard errors of the mean.

Figure S4
The Number of Entries into Arms in the Elevated-Plus Maze. Error bars represent standard errors of the mean. Asterisks represent adjusted p < .05.

Figure S5
The

Figure S8
Stress Coping Strategy of ER animals. Percent of immobile time in the tail-suspension test. Error bars represent standard errors of the mean. Asterisks represent adjusted p < .05.

Figure S9
Single-Nucleotide Polymorphism C1473G of Tph2 Gene of BALB/cCrSlc Mice and a C57BL/6 Mouse. The white arrowhead indicates Tph2 product band (523 bp) and black arrowhead indicates genotype-specific product band (307 bp). NC represents no DNA input.