Enduring deficits in brain reward function after chronic social defeat in rats: susceptibility, resilience, and antidepressant response

Abstract

Background: Anhedonia, or diminished interest or pleasure in rewarding activities, characterizes depression and reflects deficits in brain reward circuitries. Social stress induces anhedonia and increases risk of depression, although the effect of social stress on brain reward function is incompletely understood.

Methods: This study assessed the following: 1) brain reward function in rats (using the intracranial self-stimulation procedure) and protein levels of brain-derived neurotrophic factor and related signaling molecules in response to chronic social defeat, 2) brain reward function during social defeat and long-term treatment with the antidepressants fluoxetine (5 mg/kg/day) and desipramine (10 mg/kg/day), and 3) forced swim test behavior after social defeat and fluoxetine treatment.

Results: Social defeat profoundly and persistently decreased brain reward function, reflecting an enduring anhedonic response, in susceptible rats, whereas resilient rats showed no long-term brain reward deficits. In the ventral tegmental area, social defeat, regardless of susceptibility or resilience, decreased brain-derived neurotrophic factor and increased phosphorylated AKT, whereas only susceptibility was associated with increased phosphorylated mammalian target of rapamycin. Fluoxetine and desipramine reversed lower, but not higher, stress-induced brain reward deficits in susceptible rats. Fluoxetine decreased immobility in the forced swim test, as did social defeat.

Conclusions: These results suggest that the differential persistent anhedonic response to psychosocial stress may be mediated by ventral tegmental area signaling molecules independent of brain-derived neurotrophic factor and indicate that greater stress-induced anhedonia is associated with resistance to antidepressant treatment. Consideration of these behavioral and neurobiological factors associated with resistance to stress and antidepressant action may promote the discovery of novel targets to treat stress-related mood disorders.

Keywords: Anhedonia; BDNF; ICSS; depression; mTOR; stress.

Figure 1

(A) ICSS reward thresholds were significantly elevated in rats during exposure to chronic social defeat (n=13) compared to non-stressed controls (n=14). (B) Reward thresholds remained elevated in a subgroup of rats – termed susceptible (n=6) – after termination of the stressor, while threshold elevations were transient and indistinguishable from non-stressed controls in another subgroup of rats, termed resilient (n=7). Susceptibility and resilience were determined by average threshold elevations during days 17–21 of social defeat being greater than or within two standard deviations of average baseline threshold levels, respectively (Criterion 1) and by cluster analysis (Criterion 2) (i.e., both criteria characterized stress responsiveness identically for each rat). (C) Three weeks after termination of social defeat, BDNF protein levels were decreased and phosphorylation of AKT and mTOR, expressed as a ratio with total levels of these proteins, were increased in the VTA. Only the pmTOR increase was specific for susceptibility. Levels of total AKT and mTOR were unaffected (not shown). * Reward thresholds in social defeat (A)/ susceptible (B) groups were significantly different from those of controls; @ Reward thresholds in resilient rats were significantly different from those of controls; # Reward thresholds in susceptible rats were significantly different from those of resilient rats; p<0.05.

Figure 2

(A) ICSS reward thresholds were significantly elevated in rats during exposure to chronic social defeat (n=37) compared to non-stressed controls (n=10). (B) Reward thresholds remained elevated in susceptible rats (n=25) during stress exposure, while threshold elevations returned to baseline in resilient rats (n=12). Susceptibility and resilience were determined by average threshold elevations during days 10–14 of social defeat being greater than or within two standard deviations of average baseline threshold levels, respectively (Criterion 1) and by cluster analysis (Criterion 2) (i.e., both criteria characterized stress responsiveness identically for each rat). * Reward thresholds in social defeat (A)/ susceptible (B) groups were significantly different from those of controls; # Reward thresholds in susceptible rats were significantly different from those of resilient rats. (C) Stress-induced threshold elevations were significantly reversed in a subgroup of fluoxetine-treated susceptible rats (n=7) compared to vehicle-treated susceptible rats (n=9) and fluoxetine-treated susceptible rats that did not respond to fluoxetine (n=9). Ineffectiveness and effectiveness of fluoxetine treatment were determined by average threshold elevations during days 24–28 of testing (i.e., days 10–14 of fluoxetine treatment) being greater than or within two standard deviations of average baseline threshold levels, respectively (Criterion 1). * Reward thresholds in “Ineffective fluoxetine” rats were significantly different from those of “Effective fluoxetine” rats; @ Reward thresholds in “Ineffective fluoxetine” rats were significantly different from those of vehicle-treated rats; # Reward thresholds of “Effective fluoxetine” rats were significantly different from those of vehicle-treated rats. (D) Repeated fluoxetine treatment significantly decreased immobility in the forced swim test in non-stressed controls, but not in resilient or susceptible rats. Vehicle-treated susceptible rats were also significantly less immobile than vehicle-treated non-stressed controls (p<0.05).

Figure 3

(A) ICSS reward thresholds were significantly elevated in rats during exposure to chronic social defeat (n=38) compared to non-stressed controls (n=17). (B) Reward thresholds remained elevated in susceptible rats (n=22) during stress exposure, while threshold elevations returned to baseline in resilient rats (n=16). Susceptibility and resilience were determined by average threshold elevations during days 10–14 of social defeat being greater than or within two standard deviations of average baseline threshold levels, respectively (Criterion 1) and by cluster analysis (Criterion 2) (i.e., both criteria characterized stress responsiveness identically for each rat). * Reward thresholds in social defeat (A)/ susceptible (B) groups were significantly different from those of controls; # Reward thresholds in susceptible rats were significantly different from those of resilient rats. (C) Stress-induced threshold elevations were significantly reversed in a subgroup of desipramine-treated susceptible rats (n=7) compared to vehicle-treated susceptible rats (n=7) and desipramine-treated susceptible rats that did not respond to desipramine (n=8). Ineffectiveness and effectiveness of desipramine treatment were determined by average threshold elevations during days 24–28 of testing (i.e., days 10–14 of desipramine treatment) being greater than or within two standard deviations of average baseline threshold levels, respectively (Criterion 1). * Reward thresholds in “Ineffective desipramine” rats were significantly different from those of “Effective desipramine” rats; # Reward thresholds of “Effective desipramine” rats were significantly different from those of vehicle-treated rats (p<0.05).

Figure 4

The degree of stress-induced reward threshold elevations in susceptible rats predicted the subsequent response to antidepressant treatment, with greater stress induced-anhedonia associated with less likelihood of an antidepressant-like effect with fluoxetine or desipramine. In contrast, reward thresholds remained elevated in vehicle-treated susceptible rats throughout testing, regardless of the degree of stress-induced threshold elevations. Data include susceptible rats treated with fluoxetine (Experiment 2), desipramine (Experiment 3) and vehicle (pooled between Experiments 2 and 3). * p<0.05; ** p<0.01 (significant correlation).