Mouse behavioral tests

Elevated-plus maze test: The apparatus is at 50 cm height and consists of two open arms and two closed arms radiating out at 90-degree angles. The closed arms are protected by 30 cm high walls, whereas the open arms have 2.5 cm high walls. Mice were placed in the center of the apparatus and allowed to explore for 5 min. ANY-Maze video system measures the number of entries and the time spent in each arm. (18 WT (13 males and 5 females) and 18 Top3β-KO (12 males and 6 females) were examined.

Light-dark box test: Equipment and test procedures were adapted from a previous protocol⁵². The light-dark box has a light compartment (26 cm long × 22.5 cm wide × 23.5 cm tall) with transparent walls illuminated by overhead fluorescent lights, a dark chamber (13 cm long × 22.5 cm wide × 23.5 cm tall) with black walls, and a small (5.5 × 5.5 cm) opening that connects different compartments. Mice were placed in the light box and allowed to explore the environment for 10 min. An overhead camera connected to ANY-Maze software recorded the number of transitions between compartments, a measure that is sensitive to anxiolytic drugs^52,53. 11 WT (6 males and 5 females) and 11 Top3β-KO (5 males and 6 females) were examined.

Three-chamber sociability test: Equipment and procedures were based on a previous publication⁵⁴. The sociability apparatus (59 cm long × 29 cm wide × 22 cm tall) constructed of clear Plexiglas was divided into 3 equal-sized chambers with small doorways (5 cm wide × 8.5 cm tall) connecting each side chamber to the central chamber. Target mice were placed in custom-built cylinders (6.5 cm diameter × 12 cm tall) with walls of 3mm-wide stainless-steel rods spaced 7.5 mm apart to permit visual and olfactory interaction between test and target mice while preventing aggressive interactions. Each test session began with a 10 min habituation of test mouse into the sociability apparatus. Test mice were briefly removed from apparatus, and then cylinders were placed in both side chambers, with one empty and one containing a 4–5-week-old male C57Bl/6 J wild-type target mouse. The test mouse was then returned to the center chamber to commence a 5-min sociability test. An overhead camera was used with ANY-Maze software to score the length of time in chambers by the test mouse. 12 WT and 12 Top3β-KO mice of all males are used in this assay.

Reciprocal social interaction: Social interaction test was performed in a 25 × 25 × 20 cm opaque Plexiglas chamber under dim white lighting. After 30 min of habituation alone in the chamber, a 4–5 weeks old male juvenile mouse (C57Bl/6J albino, homozygous for Tyrc-2J gene; Jax stock 000058) was added in the chamber with the testing animal. The social interaction behavior was recorded for 10 min using a video camera. Frequency and duration of each social parameter detected in the test subject mouse were subsequently automatically analyzed using Noldus Ethovision XT (Noldus Information Technology, Leesburg, VA). Because mice were identified by the software by different colored fur, no paint or other markings were required. Parameters scored included following, nose-to-nose sniffing, nose-to-anogenital sniffing, sniffing of other body regions, and close following. 11 WT (6 males and 5 females) and 11 Top3β-KO (5 males and 6 females) mice were examined.

Fear conditioning test: Mice were tested for fear conditioned memory using a Video Fear Conditioning system (Med Associates). Conditioning chambers consist of plexiglass enclosure with stainless-steel grid rod flooring. In the conditioned training, mice were placed into the chamber for a total of 300 s. After 120 s of habituation, mice were presented with a 30 s tone that co-terminated with a 2 s, 0.5 mA electric foot shock. This was repeated for a total of three tone-shock pairings with an interval of 30 s between tone presentations. Twenty-four hours after conditioning, the mice were tested by a contextual experiment in the same chamber for 300 s without tone and electric foot shock stimuli. Three hours after the contextual test, cued testing was performed. Total time of the cued test was 600 s and the testing environment was altered by a smooth plastic floor covering the metal rods and a triangle shaped plastic insert above the inside of the chamber. After 300 s of resting time, animals were exposed to 3 tones of 30 s each with 30 s intervals. The Video Freeze software measured and recorded freezing behavior displayed by the mice during testing. 9 WT (4 males and 5 females) and 10 Top3β-KO (4 males and 6 females) mice were examined.

Context discrimination test: Mice were tested for ability to distinguish between a shock-paired environment (context A) and a neutral environment (context B). The specificity of freezing response to the correct environment is impaired by lesion of dorsal hippocampal tissue²⁵, and enhanced following increased hippocampal neurogenesis⁵⁵. Two distinct contexts were used (“A” and “B”) that differed in floor texture, wall shape, white vs infrared lighting, and odor. Housing rooms and transport method to and from the fear conditioning room were also different depending on which context was being used. Prior to training, mice were placed in context A for 10 min in the morning and context B in the afternoon for 10 min. 24 h later, training began and mice were put in context A for a total of 180 s. After 148 s of habituation (pre-shock period), mice were shocked by a 2 s, 0.5 mA electric foot shock. Context A chamber consists of straight 90-degree walls with stainless-steel grid rod flooring. White lighting in the box was on and 2 μl of mint extract (McCormick; Hunt Valley, MD) was put in the collection tray of the box. Four hours later, the mice were placed in context B for 180 s without shock. Context B consists of smooth plastic floor covering the metal rods and a triangle shaped plastic insert above the inside of the chamber. White lighting in the box was off and it was illuminated by near infrared lighting only. Additionally, 2 μl of orange extract (McCormick) was put on the ceiling of the chamber. Video Freeze software (MED-Associates; St Albans, Vermont) was used to measure and record freezing behavior displayed by the mice during testing. All freezing scoring was done using default settings of the software. The analysis was made on the pre-shock time between context A and Context B on training and test days. 11 WT (all males) and 12 Top3β-KO, mice of all males were examined.

Morris water maze test: This test followed a previous protocol⁵⁶. The platform was 15 × 15 cm square and the water temperature was 23 °C. We ran one trial for each mouse before moving to the next trial, resulting in inter-trial intervals of approximately 15 min. The tub was filled with water to 0.5–1.0 cm above platform. Non-toxic white paint was used to render the platform invisible to mice in the water. Each animal performed 4 trials/day, starting from each of the four compass directions. During each day, mice were tested following a pseudorandom order that was maintained for all mice on that day. ANY-Maze software recorded and measured the time required to each mouse to reach the platform. Mice that did not reach the platform within 60 s, were gently guided towards it and left there for additional 15 s. 4 h following the last training, mice were placed in water without platform for the first probe trial. ANY-Maze recorded the time in each quadrant for 60 s. 7 WT and 7 Top3β-KO mice of all males were examined.

Open field test: The test mouse was placed on the square box (43 × 43 cm, height 30 cm) for open field test. Traveling distance was automatically measured by three 16-bean IR arrays for 30 min for each mouse. Time in center, number of center entries, and center distance were tracked. Recording and analysis were performed by Activity Monitor Software (Version 4.0, Med Associates, St. Albans, VT). 13 WT and 12 Top3β-KO mice of all males were used in the test.

Rotarod test: Five-lane rotarod (Med Associates, St. Albans, VT) was used to define motor function in Top3β-KO mice. Individual mouse was tested for 5 min per day for 3 days. After the mouse was placed on the rod for 10 s, the rod speed was accelerated from 3 to 30 rpm. We recorded the latency to the first fall and the total number of falls. 7 WT and 7 Top3β-KO mice of all males were tested.

Forced swim test: A cylinder beaker (10 cm in diameter and 25 cm height) with water (24~25 °C temperature) was used for the test. Individual mouse was placed in the water for 6 min. Their behaviors were recorded by a camera. The data were analyzed for time of immobility and mobility in the last 4 min of testing. Genotype information was blinded during the analysis. 7 WT and 7 Top3β-KO mice of all males were tested.

Spontaneous alternation: Y maze (each arm: 7.62 cm × 38.1 cm, height 12.7 cm) with non-visible gray color was used for spontaneous alternation test. Each mouse was placed in the center area to start the test. Their behaviors were recorded by performers. The arms that are entered and explored by the test mouse consecutively for 10 min are determined. 7 WT and 7 Top3β-KO mice of all males were tested.

Prepulse inhibition: Mice were contained in plexiglas cylinders that allowed moderate movement, which were housed within sound-attenuated cubicles. After a 5 min habituation period mice were presented with 40 ms pulses of white noise at increasing amplitudes, with 20–30 s intervals between each presentation. Noises were presented in order from low to high amplitude, with the full sequence repeated a total of five times. Startle response was measured by accelerometer that measured maximal force applied (Vmax) in the 50 ms period following noise presentation (SR-Lab; San Diego Instruments). The next day mice were tested for prepulse inhibition. Following habituation, mice were presented with 105 dB, 40 ms startle noise presentations in the presence or absence of 20 ms “prepulse” noise presentations that occurred 30 or 70 ms prior to the startle presentation and were 70 or 80 dB in amplitude. Each of the five possible trial types was presented in randomized order within a block, for six blocks. Data was analyzed and graphed as percent inhibition of Vmax response during no-prepulse trials. 6 WT and 5 Top3β-KO mice of all males were tested.

Buried food test: Procedures were based on a published protocol (Yang & Crawley 2009; Curr Prot Neurosci). Mice were habituated for 3 days with exposure to food overnight (50 mg pieces of cheese) by introducing three pieces of cheese per mouse per cage. Mice were fasted for 16–20 h prior to testing. One hour before testing, mice were transferred to new cages containing 5 cm deep corncob bedding, where they were individually housed. Mice were temporarily removed from the cage and a single 50 mg piece of cheese was buried at the bottom of one end of the cage. The mouse was returned to the cage and the time for the mouse to find and retrieve the food was measured. 6 WT and 5 Top3β-KO (all male) mice were tested.