Predictive validity of a non-induced mouse model of compulsive-like behavior.

A key to advancing the understanding of obsessive-compulsive disorder (OCD)-like symptoms is the development of spontaneous animal models. Over 55 generations of bidirectional selection for nest-building behavior in house mice, Mus musculus, resulted in a 40-fold difference in the amount of cotton used for a nest in high (BIG) and low (SMALL) selected lines. The nesting behavior of BIG mice appears to be compulsive-like and has initial face validity as an animal model for OCD in humans. Compulsive-like digging behavior was assessed; BIG male mice buried about three times as many marbles as SMALL male mice, strengthening face validity. Using the open field and elevated plus maze, SMALL male mice showed higher levels of anxiety/fear-like behavior than BIG male mice, indicating that compulsive-like and not anxiety-like behavior was measured. To establish predictive validity, chronic (4 weeks) oral administration of fluoxetine (30, 50 and 100mg/kg/day) and clomipramine (80 mg/kg/day), both effective in treating OCD, significantly reduced compulsive-like nest-building behavior in BIG male mice. Compulsive-like digging behavior was also significantly reduced by chronic oral fluoxetine (30 and 80 mg/kg/day) treatment in BIG male mice. General locomotor activity was not affected by chronic oral fluoxetine (30 and 80 mg/kg/day) treatment; chronic oral treatment with desipramine (30 mg/kg/day), an antidepressant not effective in treating OCD, had no effect on nesting behavior of BIG male mice, strengthening predictive validity. Together, the results indicate that these mice have good face and predictive validity as a non-induced mouse model of compulsive-like behavior relevant to OCD.