Adulthood stress responses in rats are variably altered as a factor of adolescent stress exposure

Abstract

Stress exposure during development may influence adulthood stress response severity. The present study investigates persisting effects of two adolescent stressors upon adulthood response to predator exposure (PE). Rats were exposed to underwater trauma (UWT) or PE during adolescence, then to PE after reaching adulthood. Rats were then exposed to predator odor (PO) to test responses to predator cues alone. Behavioral and neuroendocrine assessments were conducted to determine acute effects of each stress experience. Adolescent stress altered behavioral response to adulthood PE. Acoustic startle response was blunted. Bidirectional changes in plus maze exploration were revealed as a factor of adolescent stress type. Neuroendocrine response magnitude did not predict severity of adolescent or adult stress response, suggesting that different adolescent stress events may differentially alter developmental outcomes regardless of acute behavioral or neuroendocrine response. We report that exposure to two different stressors in adolescence may differentially affect stress response outcomes in adulthood. Acute response to an adolescent stressor may not be consistent across all stressors or all dependent measures, and may not predict alterations in developmental outcomes pertaining to adulthood stress exposure. Further studies are needed to characterize factors underlying long-term effects of a developmental stressor.

Keywords:

• Adolescence
• adulthood
• behavior
• development
• modulation
• predator exposure
• stress
• underwater trauma

Introduction

Traumatic stress exposure during development may function as a developmental insult, potentially leading to lasting changes in neurophysiological signaling processes that mediate stress response across the lifespan. The behavioral shift toward greater peer association, risk taking and novelty seeking typically observed during adolescence may influence exposure to stress during this developmental phase (Buckley et al., 2014; Klein & Matos Auerbach, 2002; Merrick et al., 2004; Spear, 2000). A small number of studies have used rodent models to better characterize developmental influences on adulthood stress response, and results have showed primarily a bifurcating trend following developmental stress exposure: either a reduction (Bazak et al., 2009; Horovitz et al., 2012; Tsoory & Richter-Levin, 2006; Wright et al., 2013) or a potentiation (Buwalda et al., 2013; Chen et al., 2014; Post et al., 2014; Ricon et al., 2012) of later-life behavioral and physiological responses to a challenge stressor exposure. Sex differences in behavioral outcomes have also been reported, with females exhibiting behavioral deficits while males did not (Ariza Traslavina et al., 2014).

While data support the influence of developmental stress over long-term outcomes relevant to stress response, little information is available to characterize the nature of altered outcomes as a function of the type and severity of the developmental stressor. Published studies reflect a variety of environmental conditions, stressors and developmental stages, leaving gaps in knowledge of the lasting outcomes of early-life stress exposure. Understanding the alterations of developmental outcomes that influence stress response in fully developed organisms may assist in the long-term to identify hallmarks of response to and recovery from trauma, and potential relevant pathways in disorders associated with persistent effects of trauma exposure such as post-traumatic stress disorder (PTSD).

The psychological impacts of combined developmental and adulthood stress exposures may vary. Psychological diagnoses affect as many as 19.3% of combat experienced troops (Hoge et al., 2004), but it should be noted that these outcomes may deviate from those expected in subpopulations with early-life trauma exposure (Cabrera et al., 2007; Iversen et al., 2007; Stein et al., 2005). Among the civilian population, a prevalence rate of 7.8% has been reported (Kessler et al., 1995), with early-life trauma potentially interacting with adulthood stress exposure in the later development of PTSD as well (Koenen et al., 2007). In both human studies and preclinical nonhuman studies, not all developmental stressors are equal, and therefore may not produce one universal change in adulthood stress response.

We previously reported that adolescent underwater trauma (UWT) exposure lessens conditioned suppression and extinction of cued startle potentiation in adult male rats. Here, we present a continuation of our studies on adolescent stress exposure, characterizing the effects of varying developmental stressors upon adulthood stress response. In order to characterize the effects of differing developmental stressors upon later life stress response, we exposed adolescent rats to UWT or predator stress, and then to predator stress in adulthood. We report that behavioral and neuroendocrine effects of stress in adulthood are reduced after adolescent stress exposure, and this reduction depends on the specific stressor used in adolescence.

Methods

Subjects

A total of 44 male Sprague-Dawley rats (Charles River Laboratories, Wilmington, MA) arrived at postnatal day 23 and were housed on a 12:12 light:dark schedule (lights on at 0600). Rats were acclimated to the facility for 1 week prior to handling and handled once daily for 4 days prior to experimental manipulations. Adolescent rats were pair-housed and had ad libitum access to food and water. After reaching 250 g in weight (between 50–53 days), rats were singly housed with mild food restriction to effect a 10% delay in the ad libitum growth curve and allow a maximum weight of 350 g, to match previous studies in our laboratory. Research was conducted in an AAALACi accredited facility in compliance with the Animal Welfare Act and other federal statutes and regulations relating to animals and experiments involving animals and adheres to principles stated in the Guide for the Care and Use of Laboratory Animals, NRC Publication, 2011 edition.

Experimental design

The timeline of the study is shown in Figure 1. Adolescent rats were assigned to one of three groups: UWT, predator exposure (PE), or a procedural control group consisting of sham exposure matching PE conditions in a predator-free laboratory area. The procedural control group was used to provide a comparison lacking the experimental stressor UWT or PE. It should not be interpreted to indicate a condition free of stress altogether. The experimental stressors are employed to model a direct threat to life, as an integral component of a traumatic stress exposure in the definition of PTSD in humans. Prior to any experimental stress exposure, a baseline test battery consisting of a blood draw for corticosterone testing, elevated plus maze and acoustic startle response was conducted. The different stress conditions during adolescence had different time scales. The exposures were timed to end together, so that the post-stress testing timelines were conducted at the same time for all groups. The test battery was repeated at six follow-up timepoints (shown at arrows in Figure 1); 1 h post-adolescent exposure, then 7 days after adolescent exposure, before adult PE, 1 h after adult PE, before adult odor cue exposure, and 1 h after adult odor exposure. A total of seven tests were conducted. Adolescent rats were exposed to UWT (n = 15), PE (n = 15) or sham (n = 14, one rat in the final cohort died prior to beginning the experiment) and then allowed to reach adulthood without further manipulations, excepting the test batteries at 1 h post and 7 days post. Upon reaching adulthood, all rats were exposed to PE, with test batteries before and after the exposure. One week after adult PE, all rats were exposed to predator odor (PO), with test batteries before and after.

Figure 1. Experimental timeline. Adolescent rats received either PE consisting of snake (SN), ferret (FE) and cat (CA), or UWT, or sham procedural control exposures. After separation and food restriction implementation (diamond, P50–53), subjects were permitted to reach adulthood unperturbed under normal housing conditions. During adulthood, all rats underwent PE exposure. Behavioral and neuroendocrine testing took place at multiple points across the experimental timeline as shown (arrows).

Test battery

Serum corticosterone

Rats were briefly restrained in a Plexiglas tube, and 0.2 mL of blood was collected by tail vein nick. Sample collections were completed within 2–5 min. Samples were collected before behavioral testing began, to rule out the effect of behavioral testing upon circulating corticosterone levels. Blood was stored on ice for 30 min, and then centrifuged at 3000 × g for 30 min at 4 °C. Serum was removed and stored at −80 °C until assayed. Corticosterone (ng/mL) was quantified using radioimmunoassay from GE Healthcare (Piscataway, NJ).

Elevated plus maze

Exploratory behavior on selected zones of the maze was measured using a Kinder Scientific automated elevated plus maze (Poway, CA). The maze consisted of four tinted plastic arms in a “plus” configuration, elevated 80 cm above the floor. Each arm was 50 cm long and 10 cm wide. Closed arms had walls 40 cm in height, open arms had no walls. Testing was conducted over five-minute sessions in dim lighting (open arms 7 l×, closed arms 1–2 l×, intersection 2 l×). Each test began by placing the rat in the center of the maze. Photobeam tracking was used to quantify maze exploration. Basic movements in the maze were used as the dependent measure. We have previously used the change in basic movements as an outcome measure to describe the impact of stress on exploratory behavior in rats (Genovese et al., 2014; Moore et al., 2012). Procedural control and stress-exposed groups are run an equivalent number of times, and the analysis consists of a comparison of the change in exploratory behavior across the full maze (all zones) between exposure groups. Procedural control groups in this study and others in our laboratory show a minimal change in basic movements across several repeated sessions on the EPM.

Acoustic startle response

Startle amplitude was measured in Kinder Scientific chambers (Poway, CA). After a five-minute acclimatization period, 30 trials were presented, with a 115 dB white noise startle pulse over a 70 dB white noise background. Inter-trial interval was pseudorandomized, with a range of 10–30 s and mean of 20 s. Startle response was recorded for each trial in Newtons (N).

Underwater trauma

UWT was performed as previously published (Moore et al., 2012, 2014) with approval of the WRAIR/NMRC IACUC. Briefly, rats swam for 40 s in a transparent 12 L tank of normal saline at room temperature. They were then fully submerged for 20 s. Rats were removed from the tank, briefly dried, and returned to the home cage. The test battery was run 1 h after UWT.

Predator exposure

A repeated, multiple predator exposure (PE) was performed by exposing rats to one of three live predator species per day for three sequential days. All predator exposures took place in the same housing facility, which housed all predators. Predator species used (in order of exposure) were a snake (red-tailed boa constrictor or ball python), a trio of black-footed ferrets, and one of two pairs of domestic cats. Rats were placed into a custom designed and built protective enclosure to allow perception of visual, auditory and olfactory cues associated with the predator, but prevent gross physical contact with the rat. The enclosure differed for each predator species to account for different predator sizes. The enclosure was then placed into the housing area of the predator, or in the case of the snake, into the dedicated feeding cage of the predator. For the snake exposure, a piece of dried shed snake skin was included inside the rat enclosure along with the rat to maximize odor cue presence. For the cat exposure, the rat enclosure was placed on a bed of soiled cat litter to maximize odor cue, and cat treats were placed near the enclosure to increase cat proximity during exposure procedures (Genovese et al., 2014). Exposures lasted 10 min for each predator. The test battery was run for an hour after completion of the exposure.

Predator odor exposure

One week after predator exposure, rats were exposed only to Predator odor (PO) in a re-exposure experience. PO exposures took place in the predator housing facility, in a standard rodent housing cage containing standard bedding. A beaker full of predator associated materials (snake skin, paper towels with ferret odor, cat fur and cat litter) covered with an air-permeable cheesecloth was placed into the housing cage, along with the rat, for 5 min.

Statistical analyses

Data are presented as mean ± SEM. All analyses were conducted using mixed model ANOVA (Proc Mixed) in SAS software version 9.3 (SAS Institute, Cary, NC). For behavioral outcomes, one-way ANOVA was performed with contrasts and least squares means testing to detect significantly nonzero change from baseline. The factor of group was tested. For corticosterone, repeated measures ANOVA was performed with Tukey adjusted post-hoc tests. Factors of group and time were tested, with a repeated measure of time. Alpha value of 0.05 was considered significant in all tests.

Results

Adolescent stress

Adolescent rats were tested before and after experimental stress (PE, UWT) or sham procedural control exposure (Figure 1). Blood samples, followed by behavioral testing, started 1 h after the onset of stress exposure. Exploratory activity on the EPM was not different from pre- to post-test timepoints in the sham group (Figure 2(A)). Stress during adolescence did not significantly alter the expression of acoustic startle during the adolescent phase (Figure 2(B)). The UWT group was significantly decreased from the sham group (ANOVA contrast sham vs UWT, F[1,42] = 5.61, p = 0.0225), and a negative change from pre- to post-test took place in the PE group (significantly nonzero least squares means test, t[1,42] = −3.97, p = 0.0003). Corticosterone responses to stress, depicted in Figure 2(C), showed an acute increase in UWT with a decrease following at 7 days (ANOVA contrast UWT pre vs post, F[1,78] = 4.11, p = 0.0460, post vs 7d, F[1,78] = 9.46, p = 0.0029). PE showed a nonsignificant trend toward increased CORT, with a decrease indicating a return to baseline at 7 days post-stress (ANOVA contrast PE post vs 7d, F[1,78] = 3.13, p = 0.0063). No significant effects were found on open or closed arm entries (Figure 2(D)).

Figure 2. Varying the adolescent stressor (PE, UWT or sham procedural control) causes varied behavioral and neuroendocrine changes. Behavior was evaluated before and after the adolescent stressor (pre, post, 7 days) (A) There is a negative change in basic movements (post–pre) on the EPM after PE or UWT exposure. *Significantly different from sham, p < 0.03 # significantly nonzero change from pre to post-test p < 0.0005. (B) No behavioral changes were observed using the acoustic startle monitor. (C) Corticosterone was unchanged across timepoints in sham rats. PE produced a nonsignificant increase at the Post timepoint, which dropped to significantly lower than Post at the 7 day timepoint. UWT produced a significant increase at the Post timepoint followed by a significant decrease at 7 days. *Post significantly differs from pre, p < 0.05. #7 day significantly differs from post, p < 0.01. (D) No significant effects were found in open and closed arm entries.

Adulthood predator exposure

After adolescent stress, rats reached adulthood without further experimental stress exposures. All adult rats were exposed to predator, which created three exposure groups: Sham–PE (sham adolescent exposure, adulthood PE), UWT–PE (adolescent UWT, adulthood PE), and PE–PE (adolescent PE, adulthood PE). Baseline pre-PE data were collected, and post-PE blood draws and behavioral testing started 1 h after the onset of PE (Figure 1). After adulthood predator exposure, exploratory behavior on the EPM varied as a factor of stress exposure group over time (ANOVA group × time interaction, F[2,41] = 6.69, p = 0.0031). EPM exploration significantly decreased in the Sham-PE group following the adulthood predator exposure (least squares means test, t[1,41] = −3.00, p = 0.0046). In contrast, the PE–PE group showed a significant increase relative to control (ANOVA contrast, F[1,41] = 13.03, p = 0.0008). The adolescent exposure to UWT reduced the response to adulthood PE, as no significant change was found in EPM exploratory activity across pre–post testing was observed in the UWT-PE group (Figure 3(A)). No significant change in acoustic startle was observed in the Sham-PE group. Startle magnitude was, however, significantly decreased in the PE–PE group compared with controls (ANOVA contrast, sham–PE vs PE–PE, F[1,41] = 4.73, p = 0.0355), and this effect persisted at least three days after the adulthood predator exposure (F[1,41] = 4.41, p = 0.0418). The UWT–PE group had significantly decreased startle from pre-test to post-test after adulthood predator exposure (least squares means test, t[1,41] = 2.09. p = 0.0432), which had returned to baseline at a three-day post-test timepoint (Figure 3(B)). In neuroendocrine analyses, no significant changes were observed in CORT resulting from the adulthood PE, though a trend toward an increase was observed only in the PE–PE group (Figure 3(C)). No significant effects were found on open or closed arm entries (Figure 3(D)).

Figure 3. Differences in adolescent stress exposure caused different responses to the adulthood PE challenge. Behavior was evaluated before and after the adulthood stressor (pre, post). (A) The Sham–PE group showed a decrease in basic movements after PE. The PE–PE group increased movements, while the UWT–PE group did not significantly change across the two testing timepoints. *Significantly different from Sham–PE p < 0.001, #significantly nonzero change p < 0.005. (B) The PE–PE group showed a lasting decrease in acoustic startle reflex after adulthood PE. The UWT–PE group had an acute decrease in startle at the post timepoint, but had resolved by 3 days later (at the PO “pre” timepoint). *Significantly different from Sham-PE group, p < 0.05 # significantly nonzero change across timepoints, p < 0.05. (C) No neuroendocrine changes were observed across these timepoints. (D) No significant effects were found in open and closed arm entries.

Predator odor exposure

One week after adulthood predator exposure, all rats were briefly exposed to PO from all three predator species in an enclosed environment once daily over three days (Figure 1). Blood samples and behavioral testing took place before the first and 1 h after the last PO session. No significant effects on behavioral tests were found (data not shown). Neuroendocrine responses changed across time as a factor of adolescent exposure (ANOVA group × time interaction, F[2,22] = 6.06, p = 0.0080). CORT levels in Sham–PE and PE–PE groups were increased on the post-test relative to baseline (ANOVA contrasts F[1,22] = 5.67, p = 0.0263 and F[1,22] = 8.30, p = 0.0087, respectively). The UWT–PE group, which began the PO with a high baseline CORT value, showed a decreasing trend in circulating CORT from pre- to post-testing sessions (Figure 4).

Figure 4. Exposure to PO after the adulthood live predator exposure produced neuroendocrine changes, but no behavioral changes were observed between pre and post timepoints. Sham–PE and PE–PE groups significantly increased CORT, while UWT–PE showed a decreasing trend. *Post significantly differs from pre, p < 0.03 **p < 0.01.

Discussion

We have previously shown that an adolescent UWT exposure, a stressful procedure that does not inflict physiological damage (Moore et al., 2012), produced a decrement in adulthood behavioral response to stress in rats, decreasing behavioral response to cued fear conditioning and post-extinction cued startle potentiation (Moore et al., 2014). Separately, we have employed PE in adult rats to determine the extent of behavioral change after varying exposure levels (Genovese et al., 2014), finding that a repeated exposure to a single predator results in habituation and reduced behavioral impact. We now report that an adolescent stress exposure produces enduring effects upon behavioral and neuroendocrine responses to a novel multispecies PE in adulthood, and that varying the adolescent stressor alters adulthood predator impact.

The primary finding in this work is a lasting effect of adolescent stress exposure upon adulthood stress response outcomes. Following the adolescent stressor, we evaluated the adulthood response to a predator exposure. Responses to the adulthood stressor depended on stress exposure during adolescence, with PE during adolescence producing a robust decrease in behavioral impact after adulthood PE, and UWT producing an intermediate decrease. During adolescence, both stressors produced an acute decrement in EPM exploratory behavior, with UWT causing a greater decrease than PE, and no change in acoustic startle responding. Both stressors produced an acute CORT response (though significant only in the UWT group). For the adulthood PE challenge, both the adolescent PE and UWT pre-exposures produced a significant decrease in startle, though the effect persisted for longer in the PE–PE exposure. Differences were observed between the adolescent pre-exposures in lasting outcomes related to EPM exploratory behavior and PO neuroendocrine response. Adolescent PE produced an increase in EPM exploratory behavior following adulthood PE, as opposed to a decrease in sham–PE rats. The UWT–PE produced an intermediate outcome, with no change in exploratory behavior after the adulthood PE. Taken together, it seems that re-presentation of the same stressor stimulus in both the developmental phases tested may have produced an adaptive habituation, but using different stressors also produces some level of behavioral and neuroendocrine adaptation.

These results suggest there does not appear to be a single universal outcome of stress insult, at least if exposure takes place during the mid-adolescent developmental stage used here. Further, acute responses during adolescence do not, alone, predict lasting effects observed in adulthood. Thus, the type and severity of the stress exposure and the developmental status of the individual, as well as other contributing factors such as environment and mitigating factors should be considered when evaluating the potential for enduring effects on the organism.

Prior studies have provided evidence for both increased and decreased response to adulthood stressors after early-life stress exposure. We have previously shown that adolescent UWT blunts adulthood fear conditioning responses (Moore et al., 2014). Several studies have found a similar reduction in adulthood stress response severity (Buwalda et al., 2013; Chen et al., 2014; Perrot-Sinal et al., 1999; Post et al., 2014; Ricon et al., 2012). However, the literature also presents a number of studies showing increased adulthood responsivity to a stressor after adolescent stress exposure using a variety of stressors and adulthood challenges (Bazak et al., 2009; Horovitz et al., 2012; Tsoory & Richter-Levin, 2006; Wright et al., 2013). These differing results are not necessarily in conflict with each other, as the different stressors used across the cited works and different ages of stress insult may result in different developmental adaptations. Interpretations of longitudinal behavioral outcomes may also vary: a reduced response may be interpreted as a resilient outcome, but in some cases may represent a maladaptive failure to make a relevant defensive response. A single set of directly comparable longitudinal studies is not available at this time, and further studies are needed to better characterize the effects of early-life stress on adulthood outcomes.

Blunted corticosterone response as a consequence of stress exposure has been observed in our previous work and that of other groups (Bazak et al., 2009; Bourke & Neigh, 2011; Moore et al., 2012, 2014). A decreased cortisol baseline value has also been reported in the studies of human subjects with PTSD (reviewed in Raison & Miller, 2003 and in Yehuda, 2006). No acute increase in CORT was detected 1 h after the final adulthood predator exposure. Ongoing studies in the laboratory are addressing a more rigorous timecourse for blood sampling in neuroendocrine studies of PE response. The increase in EPM exploration after adulthood stress in the PE–PE group was an unexpected outcome. A possible explanation is that the habituation developed during the adolescent predator exposure. Habituation to behavioral responses following the brief, repeated exposure to cats has been reported in Genovese et al. (2014), but a chronic 60-minute daily exposure to a cat produced minimal behavioral habituation (Blanchard et al., 1998). This, however, does not explain the diminished effect of adulthood predator after adolescent exposure to UWT. Predator odor exposure has been more frequently employed in rodent studies, also with variable results (reviewed in Staples, 2010). To our knowledge, no studies have carried repeated predator exposure across the developmental phases. A compelling outcome was the negative change in startle amplitude following a stress exposure. Although increased startle amplitude in preclinical studies is cited as a point of face validity relevant to PTSD as analogous to hypervigilance, a depression of the startle reflex has also been observed in rats (Gonzales et al., 2008) and following more severe stress exposures in humans (D'Andrea et al., 2013; McTeague et al., 2010). The low frequency of this directional change in startle in the preclinical literature suggests that the present model may be a novel and useful tool for probing neurobiological phenomena underlying responses to a more severe stress event in rats, and testing of potential pharmacological interventions.

Another observation in the present study is a lack of predictive value of the acute response to lasting outcomes. There is also a lack of a clear correlation between neuroendocrine response and behavioral response to a stress exposure in both adolescent and adult developmental stages. Adolescent rats presented an acute corticosterone response to stress exposure, and a decrease in EPM exploration that varied as a factor of stressor used. No effect was found in the acoustic startle measure. Individual corticosterone response did not predict the severity of behavioral change in response to the stress. Neurophysiological and behavioral measures used here do not strongly correlate in adolescent or adult rats. The PO, in which only predator scent materials were used, triggered a corticosterone response (but not behavioral) after the stress exposure in adulthood. These differential results again represent a disparate presentation of responses between neuroendocrine and behavior. HPA response in the absence of a behavioral response to a predator odor has also been reported previously (Morrow et al., 2002; Perrot-Sinal et al., 1999). It is possible that the HPA axis maintains a species-specific sensitivity to olfactory stimuli associated with a threatening event.

The present study employed a mild food restriction regimen, commonly used with appetitively motivated tasks such as conditioned suppression, in order to match previous studies from our laboratory (e.g. Moore et al., 2014). Varied methods of chronic mild food restriction have a mildly anxiolytic effect on numerous behavioral dependent measures (Genn et al., 2003; Hendriksen et al., 2015; Heiderstadt et al., 2000; Kenny et al., 2014; Lutter et al., 2008a, 2008b; Willette et al., 2012), including increased EPM open arm exploration (Guccione et al., 2012), but others report minimal effects on the EPM and open field (Levay et al., 2007). This anxiolytic effect may diminish over time (Yamamoto et al., 2009). Neuroendocrine signaling effects have been reported. A forced loss of 20% body weight and food-restricted maintenance at 80% body weight led to an increased baseline serum CORT, which was no longer elevated 21 days later (Heiderstadt et al., 2000). In the present study, all subjects were included in the mild food restriction regimen to slow body weight gain beginning at 250 g, to include procedural controls. This experimental condition may have had some influence on the behavioral and neuroendocrine outcomes of the adulthood predator exposure, which began 21–24 days after initiation of the food restriction regimen.

Components of predator exposure are increasingly used as a stressor in rat models, including predator scented materials (reviewed in Apfelbach et al., 2005; Staples, 2010) and live predator species (Aguiar & Guimaraes, 2009, 2011; Beijamini & Guimaraes, 2006; Campos et al., 2013a; Campos et al., 2013b; Cezario et al., 2008; Comoli et al., 2003; Lisboa et al., 2014; Martinez et al., 2011; Moreira & Guimaraes, 2008; Pentkowski et al., 2009). We have previously used different exposure parameters using live cats (Genovese et al., 2014). Few studies have used live ferrets (Baisley et al., 2011; Bakshi et al., 2012; Nanda et al., 2008; Roseboom et al., 2007) or snakes (de Paula et al., 2005; Khonicheva et al., 2008) and, to our knowledge, no studies have employed a multispecies predator exposure. The novel multiple predator species exposure presented in the present study produces a robust stress profile in rats.

This study opens several questions for further investigation. While both adolescent and adult stress exposures were included, the study design was not constructed to directly compare adolescent and adult responses to a multi-predator exposure. Potential mitigating factors, such as social housing or environmental enrichment, were not tested. Other factors, such as stage of development at time of initial stress exposure, and effects of order of exposure to different stressors, were also beyond the scope of the present work. Sex differences in effects of stress exposure have only been minimally evaluated (see Shansky, 2015 for recent review), and should also be addressed in future studies.

Disclaimer

Material has been reviewed by the Walter Reed Army Institute of Research. There is no objection to its presentation and/or publication. The opinions or assertions contained herein are the private views of the author, and are not to be construed as official, or as reflecting true views of the Department of the Army or the Department of Defense. Research was conducted in an AAALACi accredited facility in compliance with the Animal Welfare Act and other federal statutes and regulations relating to animals and experiments involving animals and adheres to principles stated in the Guide for the Care and Use of Laboratory Animals, NRC Publication, 2011 edition.

Funding information

This study was supported by the Military Operational Medicine Research Program, US Army Medical Research and Materiel Command. The funding organization had no role in the writing of this article or decision to submit for publication.

Acknowledgements

The authors thank Christina Johnson, Stefania Dobre, and Larry Simmons for laboratory support, and Reynold Francis and Dr. Cynthia Kuhn for assistance with the corticosterone RIA.

Disclosure statement

The authors report no conflicts of interest.