These data demonstrate that exposure to restraint stress can impair spatial delayed alternation performance. These results extend previous work showing that an anxiogenic benzodiazepine inverse agonist, FG7142, can similarly impair performance of this task. FG7142 has no effect on a control task, spatial discrimination, with similar motor and motivational demands, but not requiring working memory or PFC cognitive function [12
]. Thus, stress-induced deficits in delayed alternation performance appear to reflect impairments in working memory operations dependent on the PFC. These findings complement the growing literature demonstrating that restraint stress can alter performance of behaviors dependent on amygdala and hippocampus [8
A more noteworthy finding of this study was the identification of differences in sensitivity to restraint stress across the estrus cycle, where females in proestrus, but not estrus, were impaired by 60 min restraint. These data support previous work from this lab showing that a benzodiazepine inverse agonist, FG7142, was more effective at eliciting PFC dysfunction during proestrus than during estrus [15
]. As proestrus is characterized by high levels of estrogen, this work provides further evidence that estrogen can act to promote sensitivity to the PFC-impairing effects of stress. As expected, all animals demonstrated significant impairment with more severe stress, 120 min of restraint. This again mimics results from previous work, which shows that larger doses of FG7142 can cause impairment regardless of sex or estrus phase [14
]. In contrast, Figueiredo et al [22
] reported that animals in proestrus show a muted acute stress response (as measured by c-Fos expression) compared to that of males or females in estrus or diestrus. However, this study saw its most robust c-Fos induction in the cingulate and motor regions of the frontal cortex, with virtually no changes in the pre- or infralimbic regions in any group. The delayed alternation task is mediated by these latter two regions [11
]; thus stress-induced PFC dysfunction is likely manifest through different mechanisms or pathways than those associated with stress-induced c-Fos expression.
The increased sensitivity to stress in proestrus does not appear to be an artifact of group differences in time to complete the task. Specifically, it might be argued that an animal that takes longer to make each choice might be at a disadvantage in remembering which arm it had previously chosen, and perform more poorly than quicker animals. Estrogen and stress have been shown to affect locomotor behavior as measured by the open field test [23
], which could potentially confound the results of our study. However, there were no significant differences in task completion times between groups whose performance differed with stress, nor were there baseline differences in task completion times between animals in estrus and proestrus, suggesting a dissociation between these factors. Thus, locomotor activity likely did not affect cognitive performance in this paradigm. These results were also not likely due to differences in spatial ability, as acute pharmacological stress has been shown to have no effect on a T-maze spatial discrimination task [14
]. Moreover, it has been demonstrated that female rats' spatial ability can in fact be enhanced by acute restraint stress [19
]; thus stress effects on spatial ability are likely not contributing to the pronounced cognitive deficits observed here.
One factor that warrants consideration is the potential effect of progesterone on the results obtained in this study. In addition to estrogen, progesterone levels fluctuate with the estrus cycle, with high levels during estrus, and rising levels during proestrus. However, ovariectomized animals with only estrogen replacement have shown a sensitivity to stress comparable to animals in proestrus as currently reported [15
], suggesting that the primary effect is due to changes in estrogen levels. That said, the potential role of progesterone in modulating stress effects should be the subject of future experiments.
The present results may also be due in part to estrogen-corticosterone interactions. Sex differences have been found with respect to basal levels as well as stress-induced release of corticosterone, with female rats releasing more corticosterone than males after 60 min of restraint, and having greater basal levels during proestrus than diestrus [24
]. Corticosterone is released into the PFC during stress, but its contribution to stress-induced PFC impairment has yet to be thoroughly described. Recent work suggests that corticosterone can, indeed, disrupt working memory [25
], but the mechanisms by which this occurs are not known. Future experiments will address this issue.
The present study provides further evidence that female rats in proestrus are more sensitive to the PFC-impairing effects of acute stress. These findings hold clinical relevance in that stress-related mental illnesses such as Major Depressive Disorder (MDD), often characterized by abnormalities in PFC morphology and function [26
], are more prevalent in women than in men [29
]. Moreover, this gender discrepancy appears at puberty, maintains through childbearing years, and then declines after menopause [30
], suggesting that circulating estrogen might make women more susceptible to stress-induced dysfunction. Much work is needed before the exact nature of estrogen's role in the stress response will be fully understood, much less yield clinical applications. That said, stress-related disorders continue to be a major public health concern for women, and basic research plays an important role in understanding the biological mechanisms underlying these disorders.