Elsevier

Behavioural Brain Research

Volume 154, Issue 2, 5 October 2004, Pages 399-408
Behavioural Brain Research

Research report
Long-term effects of a single exposure to stress in adult rats on behavior and hypothalamic–pituitary–adrenal responsiveness: comparison of two outbred rat strains

https://doi.org/10.1016/j.bbr.2004.03.011Get rights and content

Abstract

We have previously observed that a single exposure to immobilization (IMO), a severe stressor, caused long-term (days to weeks) desensitization of the response of the hypothalamic–pituitary–adrenal (HPA) axis to the homotypic stressor, with no changes in behavioral reactivity to novel environments. In contrast, other laboratories have reported that a single exposure to footshock induced a long-term sensitization of both HPA and behavioral responses to novel environments. To test whether these apparent discrepancies can be explained by the use of different stressors or different strains of rats, we studied in the present work the long-term effects of a single exposure to two different stressors (footshock or IMO) in two different strains of rats (Sprague–Dawley from Iffa-Credo and Wistar rats from Harlan). We found that both strains showed desensitization of the HPA response to the same (homotypic) stressor after a previous exposure to either shock or IMO. The long-term effects were higher after IMO than shock. No major changes in behavior in two novel environments (circular corridor, CC and elevated plus-maze, EPM) were observed after a single exposure to shock or IMO in neither strain, despite the fact that shocked rats showed a conditioned freezing response to the shock boxes. The present results demonstrate that long-term stress-induced desensitization of the HPA axis is a reliable phenomenon that can be observed with different stressors and strains. However, only behavioral changes related to shock-induced conditioned fear were found, which suggests that so far poorly characterized factors are determining the long-term behavioral consequences of a single exposure to stress.

Introduction

The neurochemical, physiological and behavioral consequences of exposure to stress have been extensively studied for decades. In most studies with adult rats, it was assumed that the effects of exposure to stress would promptly vanish, and therefore, such effects were assessed at a few hours or days after the initial exposure to the stressor. Nevertheless, there is a notable exception: the behavioral deficit caused by exposure to inescapable electric shock, the so named learned-helplessness (LH) phenomenon.

In the late 1960s, it was reported that exposure of dogs to a single session of uncontrollable electric shock resulted in a performance deficit in an active avoidance task [23], [30]. Apparently similar effects were observed in several species, including man, and the authors interpreted their findings in terms of the LH hypothesis which assumes that subjects having a previous experience with an uncontrollable situation (non-contingent relationship between their behavior and the possibility to avoid the situation) will manifest a generalized deficit in establishing further associations between their behavior and their outcome [19]. Although these deficits, which appear to involve cognitive, motivational and emotional changes [17], have been usually observed over the next 24–48 h, there are many examples of longer lasting changes [22], [31], [41].

Recently, there has been a renewed interest in long-term effects of a single exposure to stress since Tilders lab reported in 1992 that a single session of footshock was able to induce changes in behavior that lasted for at least 4 weeks. These changes were as follows [36], [37]: (i) enhanced one-way avoidance-escape behavior; (ii) reduced immobility in the forced swimming test; and (iii) reduced activity in a novel environment under background noise and even more after sudden noise off. Interestingly enough, some of the effects of shock, particularly immobility induced by noise off appears to enhance with time elapsed since shock [37], suggesting a maturation process. Changes in behavior were also accompanied by an increased adrenocorticotropin hormone (ACTH) (but not corticosterone) response to a novel environment [35], which may imply that a single exposure to shock had caused a sensitization of both behavioral and physiological response to superimposed moderate stressors.

After that, other laboratories have reported long-term behavioral effects of a single exposure to stress. For instance, long-lasting inhibitory effects of 1 or 5 sessions of shock on behavior of rats in novel environments, suggestive of increased anxiety, have been found [3], [32], [34]. Using another stressor, exposure to a cat, Adamec et al. [1], [2] reported that single exposure of rats to a cat resulted in lower ratio of activity (time or number of entries), in the open versus the closed arms of the elevated plus-maze (EPM) that was observed from day 1 to day 21 after cat exposure and was maintained over time. All these more recent data suggest that a single exposure to stress seems to induce a long-term sensitization of hypothalamus–pituitary–adrenal (HPA) and behavioral responses to superimposed stressful situations, a phenomenon which may be similar to that observed in post-traumatic stress disorder (PTSD) patients, characterized by increased anxiety and greater physiological activation in response to stressful situations [25].

However, long-term stress-induced sensitization is likely to be a complex phenomenon. Firstly, the consequences of a single session of shocks appear to be critically dependent on the characteristics of the shock protocol used. Thus, the intensity and duration of shock sessions used in the LH paradigm (i.e. [18]) are far higher than those used, for instance, by van Dijken et al. [36], [37] and this could explain some discrepant results: for instance, improvement [36] versus deficit [18] in active avoidance tasks using mild footshock and severe tailshock, respectively. Secondly, long-term sensitization of the HPA axis after a single exposure both emotional and to systemic stressors has been consistently observed when the animals were exposed to a novel (heterotypic) stressor [13], [28], [29], [35]. In contrast, there are contradictory results regarding the direction of the changes in the HPA axis when animals previously subjected to a single stress session are confronted with the same (homotypic) stressor. Whereas sensitization of the response to a homotypic stressor has been found after exposure to stressor-like agents such as cytokines [10], [11], [12], [27] and amphetamine [29], [38], a normal response to the second exposure to the same stressor was observed after restraint or social defeat [5], [16].

In addition, in a series of papers we found that a single exposure to IMO caused a desensitization of the HPA response to the same, but not to novel (forced swimming) stressors, and this desensitization had features [7], [8], [21], [33] very different from those predicted by habituation, the phenomenon accepted to underlie adaptation to repeated stress. These particular characteristics were as follows: (a) desensitization of the HPA response to the same stressors progressively develops over the course of 4 weeks and (b) desensitization of the HPA response was higher after IMO than after restraint in tubes, suggesting that the higher the intensity of the stressor the higher the long-term desensitization. Both the slow maturational process of desensitization and the direct relationship to the intensity of the stressor are incompatible with the laws of habituation. Instead, the progressive changes over time are reminiscent of stress or drug-induced sensitization, despite acting in an opposite direction.

Exposure to immobilization (IMO) in wooden boards is much more severe than restraint in tubes or shock as evaluated by the physiological response they elicit [4], [20], and therefore, we thought that IMO could induce long-term changes in the responsiveness to novel environments in the same direction as shock. However, we were unable to find any long-term effect on behavior of rats (unpublished data). There are several possibilities to explain these, somehow, unexpected results. One of them is related to the type of stressor, as most paper reporting long-lasting behavioral effects used shock. Although in physiological terms IMO is stronger than shock, the possibility that some qualitative aspects of shock (intermittency, association to particular novel environments) may be responsible for the different consequences should be considered. More precisely, It may be that those emotional stressors eliciting long-term behavioral changes may also elicit sensitization of the HPA response even to the homotypic stressor. Unfortunately, none of the stressors found to induce long-term behavioral changes, including shock, has been studied regarding long-term changes in the HPA response to the same stressor so that this possibility remains to be tested. A second possibility is related to the existence of individual differences in susceptibility to stress. Thus, whereas most authors reporting long-term stress-induced sensitization used Wistar from Harlan [32], [35], we currently use Sprague–Dawley (S–D) rats from Iffa-Credo. Therefore, we hypothesized that the long-term behavioral consequences of a single exposure to stress and, perhaps the direction of changes in the responsiveness of the HPA axis to the homotypic stressor, may be critically dependent on the strain of rats used or the particular characteristics of the stressor. Then, in the present work we have tested these hypotheses by studying the long-term effects of a single exposure to IMO or shock on behavioral reactivity to novel environments and HPA response to the homotypic stressor in both Wistar (Harlan) rats and our S–D stock.

Section snippets

Animals and general procedure

Thirty-four male S–D and 36 male Wistar rats approximately 60 days old at the beginning of the experiment, were used. S–D rats were obtained from the breeding center of the Universitat Autònoma de Barcelona using original stocks from Iffa-Credo (Lyon, France) and Wistar rats from Harlan Interfauna Iberica (Sant Feliu de Codines, Spain). The animals were housed two per cage, under standard conditions of light (photoperiod from 08:00 to 20:00 h) and temperature (22±1 °C), for 2 weeks before and

Comparison of ACTH and corticosterone levels between stressed and control rats

To statistically demonstrate the lower effects of blood sampling per se (controls) versus stress (shock or IMO) plus blood sampling on plasma levels of HPA hormones on day 1, ANOVAs were done. The ANOVAs revealed for both ACTH and corticosterone as well as for the two stressful situations, significant effects of stress, sampling time and the interaction stress by sampling time (P at least <0.002). The interaction reflected the fact that differences between stressed and merely sampled animals

Discussion

The present results confirm our previous reports, which demonstrated that a single exposure to IMO reduced the HPA response to the same stressor some days later, and extend the previous results in two ways: (a) a single session of shock also caused a reduction of HPA response to the homotypic stressor, although such reduction was smaller than that caused by IMO; (b) the long-term effects of either shock or IMO were of similar magnitude in Wistar and S–D rats; and (c) no effect of shock or IMO

Acknowledgements

This work was supported by grants from DGICYT (SAF2002-00623) and CUR (2001SGR-00203). We thank Dr. W.C. Engeland and Dr. G. Makara for their gift of the ACTH and the corticosterone antisera, respectively. Thanks are given to Elena Capillas for their help with English.

References (41)

Cited by (56)

  • Acute exposure of rats to a severe stressor alters the circadian pattern of corticosterone and sensitizes to a novel stressor: Relationship to pre-stress individual differences in resting corticosterone levels

    2020, Hormones and Behavior
    Citation Excerpt :

    Blood samples were taken by tail-nick as described previously (Belda et al., 2004). This procedure is extensively used in our lab and others because very low resting levels of hormones are obtained under appropriate conditions (Belda et al., 2004; Vahl et al., 2005). Cage-mates were sampled simultaneously (two experimenters were sampling at the same time and a third was gently holding the two rats).

  • Sex differences in the programming of stress resilience

    2020, Stress Resilience: Molecular and Behavioral Aspects
  • Sex differences in the programming of stress resilience

    2019, Stress Resilience: Molecular and Behavioral Aspects
View all citing articles on Scopus
View full text