An approach to screen susceptible rats and efficacy of an antidepressive treatment after chronic stress.

Chronic stress during life has been considered a risk factor for the development of psychiatric illness in humans. Chronic unpredictable stress battery (CUSB) is an animal model to study depression through stress exposition in rodents. CUSB induces depression and anxiety behavioral markers that could be modulated with fluoxetine an antidepressant treatment. However, it is well known that not all subjects develop depression-like behaviors or do not respond to antidepressant treatments. The way to screen these individual differences in susceptibility to stress or the efficacy of antidepressant therapy in depression models is not well studied. Here we show that saccharine consumption, immobility and time spent in the center of the area could be useful as behavioral markers for screening susceptibility to stress and depression, also we show that fluoxetine treatment had different efficacy depending on the age when the stress occurred. First, we found that after CUSB (during adolescence, adulthood, or both) rodents show depression and anxiety profiles. Second, we found that fluoxetine (10 mg/kg) could recover depression but no anxiety profile in all the groups exposed to stress. Finally, we use the machine learning clustering method (k-means), to classify the subjects in all groups based on the individual effects modulated by stress exposure and antidepressant treatment to find the ratio of stress effects and pharmacological efficacy. Using altered behaviors as classifiers, we found three possible clusters, (1) Without alterations, (2) Moderated anxiety and depression profile (3) Serious anxiety and depression profile, suggesting two groups of susceptible animals with different intensities of altered behavior. Also, fluoxetine could change the ratio of rats previously classified in groups 2 or 3, showing the beneficial effects of antidepressant treatment. Our results demonstrate that CUSB has different consequences even in subjects that experienced the same stress protocol. Also, fluoxetine has different efficacy in recovering behavior associated with the age of exposure to stress. Finally, we suggest k-means as an easy and useful method to apply in susceptible rodent studies of depression and to study the individual efficacy of antidepressant treatment.