Journal of Psychiatric Research 60 (2015) 29e39

Contents lists available at ScienceDirect

Journal of Psychiatric Research journal homepage: www.elsevier.com/locate/psychires

Corticosterone mitigates the stress response in an animal model of PTSD Min Jia, Stanley E. Smerin, Lei Zhang, Guoqiang Xing, Xiaoxia Li, David Benedek, Robert Ursano, He Li* Department of Psychiatry, Center for the Study of Traumatic Stress, Uniformed Service University of Health Sciences (USUHS), 4301 Jones Bridge Rd., Bethesda, MD 20814, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Received 18 July 2014 Received in revised form 16 September 2014 Accepted 18 September 2014

Activation of glucocorticoid receptor signaling in the stress response to traumatic events has been implicated in the pathogenesis of stress-associated psychiatric disorders such as post-traumatic stress disorder (PTSD). Elevated startle response and hyperarousal are hallmarks of PTSD, and are generally considered to evince fear (DSM V). To further examine the efficacy of corticosterone in treating hyperarousal and elevated fear, the present study utilized a learned helplessness stress model in which rats are restrained and subjected to tail shock for three days. These stressed rats develop a delayed long-lasting exaggeration of the acoustic startle response (ASR) and retarded body weight growth, similar to symptoms of PTSD patients (Myers et al., 2005; Speed et al., 1989). We demonstrate that both pre-stress and post-stress administration of corticosterone (3 mg/kg/day) mitigates a subsequent exaggeration of the ASR measured 14 days after cessation of the stress protocol. Furthermore, the mitigating efficacy of pre-stress administration of corticosterone (3 mg/kg/day for three days) appeared to last significantly longer, up to 21 days after the cessation of the stress protocol, in comparison to that of post-stress administration of corticosterone. However, pre-stress administration of corticosterone at 0.3 mg/kg/ day for three days did not mitigate stress-induced exaggeration of the ASR measured at both 14 and 21 days after the cessation of the stress protocol. In addition, pre-stress administration of corticosterone (3 mg/kg/day for three days) mitigates the retardation of body weight growth otherwise resulting from the stress protocol. Congruently, co-administration of the corticosterone antagonist RU486 (40 mg/kg/ day for three days) with corticosterone (3 mg/kg/day) prior to stress diminished the mitigating efficacy of the exogenous corticosterone on exaggerated ASR and stress-retarded body weight. The relative efficacy of pre versus post administration of corticosterone and high versus low dose of corticosterone on stressinduced exaggeration of innate fear response and stress-retarded body weight growth indicate that exogenous corticosterone administration within an appropriate time window and dosage are efficacious in diminishing traumatic stress induced pathophysiological processes. Clinical implications associated with the efficacy of prophylactic and therapeutic corticosterone therapy for mitigating symptoms of PTSD are discussed, particularly in relation to diminishing hyperarousal and exaggerated innate fear response. Published by Elsevier Ltd.

Keywords: Stress Traumatic stress response Corticosterone Glucocorticoids Amygdala Hypothalamus

1. Objectives Exposure to traumatic events alters the function of neuronal circuitry in the prefrontal cortex, amygdala, hippocampus, and, particularly, the hypothalamic-pituitary-adrenal axis (HPA) (Adamec et al., 2005; Belda et al., 2008; Osterlund and Spencer, 2011; Weiss, 2007; Yehuda, 1997). Enhanced plasma

* Corresponding author. Tel.: þ1 301 295 3295; fax: þ1 301 295 1536. E-mail addresses: [email protected], [email protected] (H. Li). http://dx.doi.org/10.1016/j.jpsychires.2014.09.020 0022-3956/Published by Elsevier Ltd.

glucocorticoid concentrations have been observed in human subjects exposed to traumatic events (Resnick et al., 1997; Yehuda, 2009). The extent and time course of plasma glucocorticoid elevation is dependent on the intensity and duration of the traumatic stressor (Servatius et al., 1995, 2001). The elevation of plasma glucocorticoid concentration may be salutary since lower baseline cortisol levels have been associated with a higher incidence of PTSD (Hauer et al., 2009). In fact, several studies of PTSD patients do suggest that exogenously administered glucocorticoids diminish fear memory retrieval and other traumatic stress associated behaviors and

30

M. Jia et al. / Journal of Psychiatric Research 60 (2015) 29e39

symptoms (Aerni et al., 2004; Schelling et al., 2001; Zohar et al., 2011; Miller et al., 2011) and in animal models of PTSD (Cohen et al., 2008; de Quervain et al., 1998). The benifical effects of glucocorticoids in reducing PTSD associated symptoms have been observed in patients who received high doses of hydrocortisone following treatment for septic shock and major surgery (Schelling et al., 2003, 2006). The efficacy of glucocorticoids in psychiatric conditions has been further observed in clinic trials (Aerni et al., 2004; Schelling et al., 2006; Weis et al., 2006). Further studies demonstrate that exogenous glucocorticoids can interfere with the retrieval of traumatic memories (de Quervain et al., 1998; de Quervain, 2008). In two double-blind, placebo-controlled studies, pre-administered glucocorticoids reduced phobic fear in subjects with social phobia and spider phobia (Soravia et al., 2006). Furthermore, administration of hydrocortisone has been reported in some studies to decrease re-experiencing and avoidance symptoms in patients with PTSD or impaired retrieval of declarative memory (Aerni et al., 2004; de Quervain et al., 2000; de Quervain et al., 2003). Likewise, memory retrieval is diminished in a watermaze spatial task in corticosterone treated rats (de Quervain et al., 1998). Patients with PTSD are a heterogeneous population with different levels of trauma. PTSD symptoms may develop during various time frames post exposure and patients may present for treatment in various time frames after symptoms develop. To determine the efficacy of glucocorticoid in PTSD requires a population that is similar in its time prior to or post exposure and in the same stage of disease development. Experimentally, such a population can be found in the restraint and tail shock animal stress model which demonstrates both an exaggerated fear response e hyperarousal e one of the most prominent symptoms of PTSD (Tomb, 1994) and HPA-axis dysfunction. Hyperarousal and HPA-axis dysfunction together comprise the closest model we have for simulating PTSD (Servatius et al., 1995). In this animal model of PTSD, the onset of enhanced ASR is not immediately observed but is delayed two weeks from the stressor, similar to the delay of onset of some symptoms of PTSD (Andrews et al., 2007; Jia et al., 2012; Solomon et al., 1989). In prior studies we addressed pharmacotherapy for PTSD in this stress model by examining the efficacy of a1-adrenoceptor and 5-HT2A receptor antagonists in mitigating exaggeration of the ASR (Jiang et al., 2009; Manion et al., 2007; Zhang et al., 2005). In the present study we continue along this line by evaluating the efficacy of corticosterone before and immediately after exposure to restraint and tail-shock. As before, exaggeration of the ASR is monitored. In comparison with non-stressed control or stress-alone subjects, current results from this study demonstrate a differential efficacy of corticosterone (3 mg/kg/day and 0.3 mg/kg/day) administration before versus immediately after three stress-exposures on the measurements of innate fear (acoustic startle) response and gain of body weight 14 and 21 days after the stress protocol. The neuronal mechanisms associated with the differential effects of corticosterone administration pre and post-stress on delayed, exaggerated fear response and body weight gain, as well as the potential clinical implications for diminishing symptoms of PTSD, are discussed. 2. Materials and methods 2.1. Experimental animals Male Sprague-Dawley rats initially weighing between 80 and 100 g (Taconic Farms, Derwood, MD, USA) were used. The animals were equally assigned to each group based on their body weight and baseline startle response. Animals were housed two per cage in

a climate controlled environment with free access to food and water, and were maintained on a 12 h reverse light/dark cycle (lights on 18:00) at 22  C. All experimental procedures were approved by the Institutional Animal Care and Use Committee of the Uniformed Services University of the Health Sciences and conducted in accordance with their Guidelines and Regulations. 2.2. Acclimation Animals were acclimated for three days to both the animal facility and to the acoustic startle chamber. Three consecutive days prior to the initial measurements animals were briefly handled in the acoustic startle chamber for 5 min each day to acclimate them. 2.3. A baseline measurement Body weight and acoustic startle response measurements were taken one day before stress and/or other procedures as baseline measurements. Baseline body weights were 138 ± 7.3 g on average. Daily food consumption was measured. Since body weights between control group and stress group were significantly different after stress, to exclude the effect of body mass on food consumption results were expressed as food consumption (in mg) per gram of body mass. 2.4. Acoustic startle measurement Acoustic startle response (ASR) measurement (Blaszczyk, 2003) was conducted with a Startle Response Acoustic Test System (Coulbourn Instruments, Columbus, Ohio, USA). This system consists of weight-sensitive platforms in a sound-attenuated chamber. The pressure against the platform due to the animal's movement in response to sound stimuli was measured as a voltage change by a strain gauge inside each platform and recorded as the maximum response occurring within 200 ms of the onset of the startleeliciting stimulus (Jiang et al., 2011a). There were six types of stimulus trials: 100 dB alone, 100 dB with pre-pulse, 110 dB alone, 110 dB with pre-pulse, pre-pulse alone and no stimulus control. Each trial type was presented eight times. Trial types were presented in random order to avoid order effects and habituation. Inter-trial intervals ranged randomly from 15 to 25 s. In the current study the responses to 100 dB sound stimuli are presented. Among the eight trials only the maximum values were collected in the results and finally adjusted with the animal body weight of the same day to avoid the force difference due to different animal body weights on the platform, and adjusted with baseline. Animals were tested one day before stress or corticosterone as baseline reading and 0, 7, 14 and 21 days following the final day of the consecutive 3 days of the stress or corticosterone. 2.5. Stress Stress exposure consisted of a 2-h per day session of immobilization and tail-shocks for three consecutive days. Stressing was done in the morning (between 0800 and 1200). Animals were restrained by being immobilized in a ventilated plexiglass tube. Forty electric shocks (2e3 mA, 3s duration; Animal Shocker, Coulbourn Instruments, USA) were delivered to their tails at semi-random intervals of 150e210 s (Graphic State Notation software, Habitest Universal Link, Coulbourn Instruments, USA) (Jiang et al., 2011a). 2.6. Chemicals Corticosterone (Sigma) dissolved in 10% ethanol (3 mg/kg/day or 0.3 mg/kg/day) 30 min before or after stress was injected intra-

M. Jia et al. / Journal of Psychiatric Research 60 (2015) 29e39

peritoneally for each of three consecutive days with stress (stress plus corticosterone groups) or without stress (corticosterone alone group). The doses we used (3 mg/kg/day or 0.3 mg/kg/day) were in the range of the plasma corticosterone concentrations produced by stress (Cohen et al., 2006; Kaouane et al., 2012; Resnick et al., 1997). Ethanol (10%) as vehicle was injected intra-peritoneally 30 min before stress for each of three days of stress (in the stress plus vehicle group) or injected intra-peritoneally alone without stress (in the vehicle alone group). The volume and the concentration of the vehicle were kept the same at different doses of corticosterone. The corticosterone receptor antagonist Mifepristone (RU486) (Cayman Chemical Co.) (40 mg/kg/day) was dissolved in 20% ethanol and injected intra-peritoneally with corticosterone (3 mg/ kg/day) for each day of three consecutive days 30 min prior to stress. 2.7. Data analysis Analysis of variance (ANOVA) for repeated measures was performed on net weight gain and ASR with the factors of days, stress status and drug dosage. The one-way ANOVA Dunnett test and student t test (indicated in results) were used to assess significant post-hoc differences in individual groups. Statistical analysis was performed using SPSS software (SPSS Inc., Chicago, IL, USA). The data were represented as mean ± S.E.M. 3. Results 3.1. Stress of restraint and inescapable tail-shock for 3 days induced a delayed and exaggerated acoustic startle response Consistent with previous reports, stress induced a delayed and exaggerated ASR to the 100 dB sound stimulus measured on day 14 and day 21 after cessation of the stress protocol. Compared with control, the maximum value of the ASR to 100 dB sound stimuli, adjusted for body weight and baseline value, is significantly enhanced in the stress group on day 14 (p < 0.01, t test) and day 21 (p < 0.01, t test) but not on Day 7. The ASR value in the stress group was significantly reduced (p < 0.01, t test) (Fig. 1) immediately after cessation of the stress protocol.

31

3.2. Corticosterone attenuated the stress induced exaggeration of the ASR to 100 dB sound stimuli Corticosterone (3 mg/kg/day), whether injected 30 min before stress or 30 min after stress, attenuated the exaggerated and delayed ASR on day 14 after stress (p < 0.05, Fig. 2a), indicating the prophylactic efficacy of corticosterone administration at higher concentration. On day 21, a significant reduction of the exaggeration of the ASR was still observed in the group of stress plus corticosterone (3 mg/kg/day) injected 30 min before stress (p < 0.01, compared with stress alone) but not when injected after stress, indicating that pre-stress treatment with 3 mg/kg corticosterone has a longer efficacy than post stress treatment. Thus, the time frame for pharmacologic intervention is a critical determinant for optimal efficacy. Corticosterone (0.3 mg/kg/day) injected 30 min before stress did not show the reduction in exaggeration of the ASR on day 21 after stress (p > 0.05) (Fig. 2b). These results indicate that both time window for the administration of corticosterone and dosage are critical determinants. Clinically, multiple applications of lower dosage have been demonstrated to reduce the symptoms of PTSD (Aerni et al., 2004). In an attempt to address whether the efficacy of corticosterone is mediated through activation of the glucocorticoid receptor, the glucocorticoid receptor antagonist RU486 (40 mg/kg/day) was administrated with 3 mg/kg/day corticosterone prior to the stress protocol. While the effect of corticosterone on the exaggerated ASR was attenuated in the presence of RU486 on day 14 and 21, the attenuation was not significant (p > 0.05, compare either with the stress alone group or with the stress group receiving 3 mg/kg/day corticosterone prior to stress). 3.3. Stress of restraint and inescapable tail-shock for 3 days resulted in a reduction of body weight Animals during the three days of restraint and inescapable tailshock ceased to gain body weight. Although post stress the weight gain of the stress group paralleled that of the control group and corresponded to the pre-stress rate, the three day stress-induced body weight retardation was never compensated, as measured at 21 days following cessation of the stress protocol (Fig. 3a). Student's t-test showed significant differences at each data point from day 0 to day 21 after stress protocol (p < 0.01). In the stress group during the days of stress, food consumption was significantly reduced (p < 0.01, t test) compared with that in the control group (Fig. 3b), and remained lower after stress. In order to exclude the impact of the difference of body mass in food consumption, we divided the food consumption by the individual animal's body weight. Fig. 3c shows that the calculated food consumption per unit body mass in the stress group was significantly lower than that in the control group during the 3 days of stress (n ¼ 8 in each group, p < 0.01, t-test). After stress no significant difference in food consumption per unit body mass was observed between the two groups (n ¼ 8 in each group, p > 0.05, ttest). 3.4. The effect of corticosterone on stress reduced gain in body weight

Fig. 1. Percentage of control (mean ± S.E.M.) of peak startle amplitude (100 dB, body weight adjusted, represented as % of the baseline ASR) for groups of stress and control measured immediately after stress, 7, 14 and 21 days following stress. Asterisks (**) indicate significant differences between groups p < 0.01 (t test). Day-3: n ¼ 32 in control group and n ¼ 39 in stress group. Day 0: n ¼ 24 in both control and stress groups. Day 7, 14 and 21: n ¼ 32 in control group and n ¼ 39 in stress group. Horizontal bar indicates the days of stress.

Measured at 21 days after the cessation of stress, the retardation of body weight by the stress protocol was mitigated in the group pre-treated with corticosterone at 3 mg/kg/day (Fig. 4d). Body weight of groups injected with corticosterone alone or vehicle was not significantly different from the naïve control group over the days coeval with the 21 days post stress of the stress group. (p > 0.05, Fig. 4a, b, c, d). Since body weight of groups treated with

32

M. Jia et al. / Journal of Psychiatric Research 60 (2015) 29e39

Fig. 2. Mean ± S.E.M. of peak startle amplitude (body weight adjusted, represented as % of the baseline ASR) for groups of control, stress, stress plus corticosterone (3 mg/kg, 30 min before stress), stress plus corticosterone (3 mg/kg, 30 min after stress), stress plus corticosterone (0.3 mg/kg, 30 min before stress), stress plus RU486 and corticosterone (3 mg/kg, before stress), corticosterone (3 mg/kg), vehicle alone and stress plus vehicle, measured 14 (a) and 21 (b) days following treatments. n ¼ 32 control, n ¼ 39 stress, n ¼ 24 stress þ corticosterone (3 mg/kg, before stress), n ¼ 16 stress þ corticosterone (3 mg/kg, after stress), n ¼ 8 stress þ corticosterone (0.3 mg/kg, before stress), n ¼ 8 stress þ RU486 þ corticosterone, n ¼ 24 corticosterone, n ¼ 20 vehicle alone, n ¼ 16 stress þ vehicle. Asterisks (*) indicate significant differences between groups (p < 0.05) and (**) indicate p < 0.01 (versus stress).

high dose corticosterone (3 mg/kg/day) 30 min either before or after stress, or low dose corticosterone (0.3 mg/kg/day) 30 min prior to stress was still retarded at day 1 and day 7 after treatment compared with naïve control (Fig. 4a and b, p < 0.01), the mitigating effect of corticosterone on the stress retarded body weight had not appeared at these times and doses. In addition, body weight gain was also not affected in the stress group pre-treated with RU486 (40 mg/kg/day) and corticosterone (3 mg/kg/day) on day 1 and day 7 after stress, indicating that pharmacological manipulation with either agonist or antagonist of corticosterone has no immediate effect on the body weight gain within the first 7 days after stress. However, 21 days after stress the body weight gain of the groups injected with corticosterone 3 mg/kg/day, 0.3 mg/kg/day, or 3 mg/ kg/day þ RU486 before stress was no longer differentiable from naïve controls. By day 14 after stress the body weight of the stressed animals receiving even the lower dose of corticosterone (0.3 mg/kg/day) prior to stress was no longer differentiable (p > 0.05) from that of naïve controls (Fig. 4c). Meanwhile, at twenty-one days after stress (Fig. 4d), a significant reduction in body weight compared to naïve controls was still seen in the stress alone group (p < 0.01) and in the stressed groups injected with corticosterone after stress. Furthermore, while the body weight gains between the stressed group and stressed group pre-treated with high dose corticosterone (3 mg/kg/day) are significantly different (p < 0.05) on day 21,

the group pretreated with the lower dose of corticosterone (0.3 mg/ kg/day) showed no significant difference compared to the stress group measured on day 14 and day 21 after the cessation of stress (p > 0.05). This lack of effect suggests that a low dose of corticosterone prior to stress treatment does not mitigate growth retardation after stress. Furthermore, in the stressed group pre-treated with RU486 and 3 mg/kg/day corticosterone no significant difference appeared in comparison with the control group, stress alone group, and stress treated with corticosterone group (p > 0.05), indicating that mechanisms other than glucocorticoid mediated mechanisms may be also involved in mitigating the retardation of body weight gain induced by the stress protocol. Food consumption per unit body weight showed a significant reduction during stress in the stress group and the stress plus corticosterone (3 mg/kg/day) administered 30 min prior to stress group compared with the control group (p < 0.01, ANOVA, Tukey), but did not show a significant difference either one week or two weeks after stress (Fig. 5). 4. Discussion The efficacy of administration of high concentration corticosterone on mitigating the enhanced fear response and reduced body weight could be attributed to the signaling activation of both mineralocorticoid (MR) and glucocorticoid receptors (GR), which

M. Jia et al. / Journal of Psychiatric Research 60 (2015) 29e39

33

which certainly results in corticosterone blood levels above 10 nM, has no significant long term impact on the exaggerated fear response measured by acoustic startle response and reduced body weight 14 days after stress. A blood level of CORT in even our low concentration corticosterone rats is certainly above 10 nM. Thus, activation of membrane and/or cytosolic MR alone is not sufficient to mitigate stress-induced enhanced fear and reduced body weight. In addition, in a current microarray study examining the gene expression profile in predator-scents-stress exposure, corticosterone (25 mg/kg) post-stress treatment prevented anxiety and hyperarousal measured by startle and elevated plus maze 7 day later in both sexes of mice, supporting the observation (shown in Figs. 2 and 4) in our rat model of PTSD that pre- and post-administration of corticosterone are able to mitigate stress-induced anxiety abnormality (Daskalakis et al., 2014). 4.1. Fear and arousal Most of the current literature supports a therapeutic efficacy of activation of the glucocorticoid receptor in the relief of fear response in PTSD patients (de Quervain and Margraf, 2008; de Quervain, 2008; Hauer et al., 2014). Administration of cortisol results in a significant reduction of PTSD symptoms in critically ill or injured patients. From a clinical and animal morphologic study Zohar et al. demonstrated that early application of high dose hydrocortisone significantly reduced the risk of the development of PTSD (Zohar et al., 2011). In agreement, when we injected corticosterone in rats before traumatic stress exposure the incidence of delayed and exaggerated fear response was reduced and there was even an additional beneficial effect after stress e normalization of weight gain. In apparent contradiction with the above beneficial effect of corticosterone are a number of studies that have shown an increased levels of hydrocortisone in plasma in humans with PTSD (Lemieux and Coe, 1995; Resnick et al., 1997) and in animal stress models (Richardson Morton et al., 1990; Servatius et al., 1995). That the effect of corticosterone on stress symptomatology seems positive under some conditions and negative in others may be at least partially explained by differences between studies in four dimensions e (1) intensity of stress, (2) time of evaluation, (3) memory versus arousal, and (4) other agents possibly co-released with corticosterone. We consider these points in order:

Fig. 3. (a) Effect of the stress protocol on net body weight gain. Means ± S.E.M of net weight gains of both control and stress groups of Sprague-Dawley rats in three days of restraint-tail shock (measured before stress, immediately following stress and 1, 7, 14 and 21 days following stress). Asterisk (**) indicates significant differences between groups (**p < 0.01), (day e 7 control group n ¼ 25, stress group n ¼ 24; day e 3 control n ¼ 61, stress n ¼ 60; day 0 control n ¼ 53, stress n ¼ 52 stress; day 1 control n ¼ 52, stress n ¼ 52, day 7, day 14 and day 21 control n ¼ 61, stress n ¼ 60). Arrows indicate the days of stress. (b) Average daily food consumption of control and stress animals before stress, during stress, one week after stress and two weeks after stress (n ¼ 8 in stress group, n ¼ 8 in control group, *p < 0.05 and **p < 0.01 compared to control group, t test). (c) Unit body weight food consumption of stressed animals relative to that in control animals (as 100%) (n ¼ 8 in stress group, n ¼ 8 in control group, **p < 0.01 compared to control group, t test).

mediate rapid non-genomic as well as slow genomic mechanisms. Neuronal tissue binding studies demonstrate that MR has a high affinity binding site for glucocorticoids ranging from 0.1 to 10 nM concentrations (Reul et al., 1990). As indicated in Figs. 2 and 4, corticosterone pre-stress administration at 0.3 mg/kg (lower dose),

4.1.1. Intensity of stress Animal experiments so far reported employed only foot shocks, in small number (de Quervain et al., 1998, for instance, used three foot shocks), or a relatively short period of restraint (Kaouane et al., 2012, for instance, used 15 min of restraint), with such stress increasing the level of corticosterone in the plasma in about 30 min (de Quervain et al., 1998). However, the stress protocol used in our current experiment consists of 40 shocks delivered to restrained rats for 120 min each day for three days. The stress protocol we employ is clearly more repetitive, prolonged and intense than that of either de Quervain's or Kaouane's group (de Quervain et al., 1998; Kaouane et al., 2012) and therefore expected to produce longer lasting neurobiologic alterations. Our findings to date (Braga et al., 2004; Jiang et al., 2011a, 2011b; Manion et al., 2007; Xing et al., 2011; Jiang et al., 2009) show that with this more intense stress paradigm the neuronal circuitry engaged in exaggerated fear is altered for a period of at least weeks. In the current study our stress protocol substantially altered both innate fear response and body weight growth. As shown in Fig. 1, the amplitude of the acoustic startle response is significantly reduced when measured immediately after the termination of the stress protocol.

34

M. Jia et al. / Journal of Psychiatric Research 60 (2015) 29e39

4.1.2. Time of evaluation PTSD animal model studies in the literature evaluated the effect of stress soon after the stress while the level of corticosterone remains high (de Quervain et al., 1998; Kaouane et al., 2012). Kaouane et al., especially, found that either foot shock stress, systemic injection of corticosterone, or infusion of corticosterone in the hippocampus can induce PTSD-like memory impairments associated with neuronal activation within the hippocampal-amygdala circuitry in a time frame of 24 h. Since the neurobiologic consequence of traumatic stress can be delayed by weeks or more, we evaluated ASR two and three weeks after stress, at which point the level of corticosterone returns to baseline (Servatius et al., 1995). As shown in our current study (Fig. 2), administration of corticosterone attenuated stress-induced startle exaggeration two weeks after corticosterone treatment and the beneficial effect remained evident up to three weeks, indicating a delayed and long lasting efficacy of such therapy. So it is not unreasonable to expect that our measurements of fear memory and innate fear response would be different from those evaluated immediately after stress. Furthermore, the exogenous pre-treatment and post-treatment with corticosterone within the appropriate time window associated with traumatic events can mitigate the development of pathophysiological processes underlying the symptoms of PTSD. On the other hand, activation of the glucocorticoid receptor might exacerbate the symptoms of already established PTSD under certain conditions. For example, the efficacy of bolus administration of hydrocortisone (4 mg/kg) on combat-related PTSD symptoms was examined using a traumatic memory reactivation therapeutic regime (Suris et al., 2010). No significant improvement was found in the ‘impact of event’ score report (IES-R) and symptom subgroup scores immediately after the therapy. In addition, the patients in the group treated with hydrocortisone had a significantly higher quick inventory depression score (QIDS) immediately following the therapy. But QIDS scores one week after the visit returned to the baseline level with a significantly lower IES-R avoidance/numbing scores. These results show that while exogenous glucocorticoid induced activation of the glucocorticoid receptor might transiently exacerbate certain symptom of already established PTSD in the presence of a higher concentration of cortisol (Suris et al., 2010), one week later a beneficial effect of hydrocortisone treatment can manifest. 4.1.3. Memory versus arousal It is the fear memory component rather than the arousal component of PTSD that has mainly been dealt with in the literature. For instance, rate of forgetting is generally a function of time since the stress event (Wixted, 2004). In PTSD patients, traumatic memories remain and are often recalled vividly even years after the traumatic incident indicating that there is a failure of forgetting in PTSD patients. Systemic administration of corticosterone, even before a traumatic event, could promote the forgetting of (de Quervain et al., 1998) or impair formation of (Kaouane et al., 2012) those traumatic memories underlying PTSD. Our test, ASR, is not so much a test of memory, but rather a test for hyperarousal

Fig. 4. The comparison of net body weight gains for groups of control, stress, stress plus corticosterone (3 mg/kg, 30 min before stress), stress plus corticosterone (3 mg/ kg, 30 min after stress), stress plus corticosterone (0.3 mg/kg, 30 min before stress) and stress plus RU486 and corticosterone (3 mg/kg, before stress), corticosterone (3 mg/ kg), vehicle alone, stress plus vehicle, 1(a), 7(b), 14(c) and 21(d) days following

treatments. (a) n ¼ 52 control group, n ¼ 52 stress group, n ¼ 22 stress þ corticosterone (3 mg/kg, before stress), n ¼ 16 stress þ corticosterone (3 mg/ kg, after stress), n ¼ 8 stress þ corticosterone (0.3 mg/kg, before stress), n ¼ 8 stress þ RU486 þ corticosterone (3 mg/kg, before stress), n ¼ 20 corticosterone (3 mg/ kg), n ¼ 12 vehicle, n ¼ 8 stress þ vehicle. (b, c, d) n ¼ 61 control, n ¼ 60 stress, n ¼ 30 stress þ corticosterone (3 mg/kg, before stress), n ¼ 24 stress þ corticosterone (3 mg/ kg, after stress), n ¼ 8 stress þ corticosterone (0.3 mg/kg, before stress), n ¼ 8 stress þ RU486 þ corticosterone (3 mg/kg, before stress), n ¼ 36 corticosterone, n ¼ 20 vehicle, n ¼ 16 stress þ vehicle. Asterisks (*) indicate significant differences between groups (p < 0.05) and (**) indicate p < 0.01 (versus control group).

M. Jia et al. / Journal of Psychiatric Research 60 (2015) 29e39

35

addition, Golier et al. (2012) demonstrated that administration of RU486 (600 mg/day) in treating combat related PTSD is associated with acute increases in cortisol and ACTH level. These results suggest that therapeutic administration of RU486 (600 mg/day) may also induce a secondary effect on the HPA axis to enhance glucocorticoid release in addition to its primary antagonistic action on the glucocorticoid receptor. In summary most of the current literature supports a therapeutic efficacy of activation of the glucocorticoid receptor in the relief of fear response in PTSD patients (Aerni et al., 2004; de Quervain and Margraf, 2008; de Quervain, 2008) and our results in animals concur. Corticosterone must be administered at the right time window. Fig. 5. Body weight adjusted daily food consumption (mg/gm body weight) in control, stress, stress plus corticosterone (3 mg/kg) administered 30 min prior to stress, corticosterone (3 mg/kg) groups during stress, one week after stress and two weeks after stress (n ¼ 8 in all 4 groups) (*p < 0.05, **p < 0.01).

(Koch, 1999; Tomb, 1994) e a test of state. That de Quervain et al. and Kaouane et al. found exogenous corticosterone given post stress to be detrimental, while we found it to be beneficial, could have resulted from our looking at state of arousal while they looked at fear memory (de Quervain et al., 1998; Kaouane et al., 2012).

4.1.4. Agents co-released with corticosterone That exogenous corticosterone mitigates the stress response even though endogenous corticosterone is elevated by stress directs us to the realization that agents other than corticosterone rise in response to stress. It may be that these co-released agents and not corticosterone elevate fear and arousal. One such co-released agent is norepinephrine (Quirarte et al., 1998; Tanaka et al., 1991). In a memory paradigm based animal model studies of stress, glucocorticoid effects on stress memory have been demonstrated to involve noradrenergic activity (de Quervain et al., 2007; de Quervain et al., 2009; McGaugh, 2013; Roozendaal, 2002). In agreement with our ASR results, several authors have found in their animal models that beta-receptor mediated facilitation of memory by norepinephrine is suppressed by glucocorticoids (Borrell et al., 1984; de Quervain et al., 1998; Roozendaal et al., 2006). Suppression of facilitation of memory requires that corticosterone rises before the enhancement of memory by norepinephrine (Joels et al., 2011; Schwabe et al., 2012). The importance of timing during the cooperation of these agents during stress is further discussed in Section 4.2. In attempting to address whether the efficacy of corticosterone in attenuating exaggerated fear is mediated through glucocorticoid receptor activation, the commonly used glucocorticoid receptor antagonist Mifepristone (RU486) (40 mg/kg/day for three days) was co-administrated with corticosterone (3 mg/kg/day for three days) in the stressed group. Although corticosterone significantly attenuated stress-induced exaggeration of ASR when measured 14 days and 21 days after the cessation of our stress protocol (Fig. 2a), coadministered RU486 diminished but did not significantly reverse this attenuation (p > 0.05, compared to either stress or stress plus corticosterone 3 mg/kg/day). It has been reported that RU486 can be a partial agonist rather than pure antagonist of glucocorticoid receptors in certain conditions, as it is when administered at a high dose (30 mg/kg/day) for a long period (14 days) in a genetically obese, diabetic rat (Havel et al., 1996). Although RU486 has no agonistic effect observed in healthy human tissues, RU486 has been shown to have a partial agonist activity in human cancer cell lines (Fryer et al., 2000) and in primary adrenal insufficiency subjects in vivo (Laue et al., 1988). In

4.2. The timing of cortisol administration during processing of stress Corticosterone/cortisol administration modulates memory retention and retrieval and traumatic memory in animal and in human studies (de Quervain et al., 2000; de Quervain et al., 1998). The impaired retention was evident in the group with pretreatment 30 min before retention testing or after footshock stress but no impaired performance was observed when footshock stress was given 2 min or 4 h before testing (de Quervain et al., 1998). These time-dependent effects of stress on impairing retention performance is associated with circulating corticosterone levels at the time of testing (de Quervain et al., 1998). In addition, repeated oral administration of cortisol (10 mg) 1 h before stressor exposure reduces spider-induced phobic fear response (Soravia et al., 2006). In agreement with our ASR results shown in Figs. 2 and 4, corticosterone administration 30 min before restraint and tail shock has long lasting efficacy in mitigating stress-induced exaggerated fear on day 14 and retarded body weight gain on day 21 after the termination of the stress protocol in comparison to the corticosterone treatment 30 min after the stress (Figs. 2 and 4). In a more recent study using the predator-scent-stress exposure model, 1 h post-stress corticosterone administration prevented anxiety and hyperarousal 7 days later, confirming that pharmacological intervention in a proper time window associated with a traumatic event is the key to therapeutic efficacy (Daskalakis et al., 2014). Using microdialysis, norepinephrine concentration was observed to increase up to three fold in the basolateral amygdala during footshock stress as well as during the restraint and tail shock stress, which is the same stress protocol as used in our current study (Quirarte et al., 1998; Tanaka et al., 1991). Recent studies also demonstrated that glucocorticoids can interact with beta noradrenergic mechanisms in memory retrieval in vivo (Roozendaal et al., 2004). The characteristics of timing related to the efficacy of the corticosterone mediated effect on glutamatergic transmission was further investigated at the cellular level in the in vitro brain slice preparation demonstrating that evoked AMPA receptor mediated EPSCs as well as miniature EPSCs were enhanced after corticosterone exposure approximately 3e4 h after treatment (Karst and Joels, 2005). In addition, that such corticosterone exposure induced enhancement of glutamatergic synaptic transmission reveals a gradual suppression of the amplitude of the EPSC after 4 h treatment (Joels et al., 2012b; Karst and Joels, 2005). Furthermore, both evoked NMDA and AMPA receptor-mediated synaptic transmission were enhanced in a period of 1e4 h after stress (Yuen et al., 2011; Joels et al., 2012a). Viewing stress induced elevation of ASR as a measure of delayed fear memory, we hypothesize that in our experiments corticosterone acted by diminishing the action of norepinephrine. One locus of action is evidenced to be the basolateral amygdala. Pu et al. (2009), recording excitatory postsynaptic field potentials, and Liebmann et al. (2009), recording amplitude of excitatory postsynaptic currents from basolateral amygdala

36

M. Jia et al. / Journal of Psychiatric Research 60 (2015) 29e39

neurons, found that the facilitation of glutamatergic synaptic transmission by the noradrenergic beta agonist isoproteranol was diminished by corticosterone, but only if the corticosterone was applied ahead (2e4 h) of the noradrenergic agonist (Pu et al., 2009). This 2e4 h time course corresponds to the half hour that we gave corticosterone prior to stress, plus the two hours of the stress period. A plausible cellular mechanism underlying the interaction of corticosterone with noradrenergic signaling recently was also demonstrated by the enhancement of AMPA receptor currents by norepinephrine via the cAMP signaling pathway in hippocampal neurons (Zhou et al., 2012). Corticosterone applied prior to the activation of noradrenergic enhancement of memory consolidation in the amygdala circuitry may thus interfere with the traumatic stress-induced fear response in the current model of PTSD by affecting noradrenergic mediation of glutamatergic transmission. In addition, Osterlund and Spencer (2011) found that three hours of corticosterone pre-treatment reduced restraint-stress induced c-fos gene expression in the anterior pituitary (Osterlund and Spencer, 2011). Their work also suggests that corticosterone pretreatment could modulate HPA axis stress reactivity via a protein synthesis-dependent mechanism (Osterlund and Spencer, 2011). Although the current stress paradigm is more severe than that described in Osterlund and Spencer's study, pre-activation of glucocorticoid receptors in the brain is considered to dampen a stress-induced hyperactivation of the HPA axis (Herman et al., 2005). In addition to the activation of glucocorticoid receptor signaling, administration of corticosterone at 15 mg/kg and traumatic stress can induce distinct changes in inflammatory and metabolic pathway related gene expression profiles, including the NFkb, Bcl2, and TNF signaling pathways, which may prevent anxiety-like behaviors through direct binding to glucocorticoid response elements (GREs) in the nucleus (Li et al., in press; Datson et al., 2011; Gray et al., 2013). Therefore, corticosterone mediated short term and long term effects on neuronal functions appear to be necessary for optimal neurobiological adaptation to stressful events. 4.3. Multiple mechanisms in body weight loss by stress In the current experiments, the growth of the rats exposed to the three-day restraint/tail shock stress protocol ceased during the three day stress period and their body weight remained at a lower level than that of unstressed controls for at least 21 days following stress. This is consistent with previous reports including ours (Braga et al., 2004; Manion et al., 2007; Jiang et al., 2009). Such decrement in growth has also been observed in animals experiencing milder stress protocols (Chotiwat et al., 2010; Harris et al., 2002a; Smagin et al., 1999). Decrease in gain of body weight is present in patients with depression, which is highly comorbid with PTSD and may share some neurobiological mechanisms (Donini et al., 2003; Eberly and Engdahl, 1991; Gazewood and Mehr, 1998; Myers et al., 2005; Speed et al., 1989). Since lower body weight persists for weeks after cessation of stress (Harris et al., 2002b; Jiang et al., 2009), reduction in the decrement in growth consequent to stress, as found in our animal model, could be used as a sensitive indicator for evaluating the therapeutic efficacy of pharmacological agents. In Fig. 3a, gain of body weight ceased during the stress period and remained low after the period of stress, without any trend towards restoration of normal body weight. The body weight of stressed animals remained low even up to day 21 (Fig. 3a). According to “set-point theory”, after stress, food consumption should rebound and body weight should catch up to normal (Harris, 1990; Harris et al., 2002a; Wilson and Osbourn, 1960). In our experiments, stress seems to lower the “set-point” after stress for a long period of time. This long term depression of

the set point suggests that neuronal mechanisms are involved in the long term depression of body weight by stress. These neuronal mechanisms are proposed to be related to increased energy expenditure and reduced food intake during stress (Harris et al., 2002a, 2002b). Food consumption was, in fact, significantly decreased in our experiments during the days of stress, both absolutely (Fig. 3b) and when normalized relative to body mass (Fig. 3c). Food consumption in the stressed group remained below controls after stress too (Fig. 3b). At that point the decreased eating may be due to the decreased energy required to maintain the lower body mass in the stressed group, since after stress no significant difference was observed between stress and control groups when food consumption was normalized to body mass (Fig. 3c). The stressed rats immediately resumed food and water intake after stress. Since food consumed per body mass was normal in the stress group post stress, metabolic rate may normalize after stress. However, no additional food was consumed to make up the weight loss in the stressed group after stress. Body weight loss during stress and the sustained lower body weight after stress may have different mechanisms. In mild stress such as restraint only, there is a small loss of body weight during restraint, but after restraint body weight soon rebounds (Harris et al., 1998, 2002a). But, as described above in more severe stress conditions, the body weight of stressed animals may never be restored to that of control animals. Therefore, in severe stress, a lowering of the set point for body mass may take place and endure. 4.4. Corticosterone effects on stress induced body weight loss Elevation of endogenous corticosterone during and after stress blocks the rectification of body weight. It is well established that a high level of plasma corticosterone in the periphery is catabolic, with an increase in glycogenolysis, conversion of protein and fat to glucose, and metabolism of the resulting free glucose. This catabolism both releases energy for coping with stress and likely underlies the halting in the growth of the stressed rats. The plasma level of corticosterone was greatly enhanced immediately after the three day stress protocol in the stress alone group (Jia et al., 2012), which indicates that the elevation of endogenous corticosterone during and after stress blocks the rectification of body weight that would otherwise occur after a period of hypophagia. Similarly, stressed animals that were administered corticosterone after each of the three periods of stress did not normalize their body weight, even 21 days after stress (Fig. 4). A different mechanism emerges when corticosterone is administered before stress. When corticosterone is administered 30 min before stress at 3 mg/kg/day, no significant difference was found in the body weight compared to control group when observed on day 21 after the termination of the three day stress. Thus while corticosterone injection prior to stress did not prevent cessation of body weight gain during stress, it did block the lowering of the homeostatic set point for body weight by stress, ultimately leading to the normalization of body weight. Both hormonal and neuronal mechanisms may be involved in the body weight loss during stress and the post-stress reinitiation of weight gain. The neuropeptide corticotrophin-releasing factor (CRF), secreted from the parvocellular neurons of the paraventricular nucleus (PVN) suppresses food intake and result in body weight loss (Smagin et al., 1999). In addition, third ventricle CRF receptor antagonist administration prevents stress-induced weight loss (Smagin et al., 1999). Thus, homeostasis in food intake and energy consumption is controlled at least in part by CRF mediated neuroendocrine systems. Recently, stress-induced cessation of body weight gain and retarded body weight growth after stress were found to relate with demethylation of the CRF gene in the

M. Jia et al. / Journal of Psychiatric Research 60 (2015) 29e39

hypothalamus thus, increase CRF production (Elliott et al., 2010). In particularly, chronic social stress induced long-term demethylation of CRF genomic transcriptome sites in a subset of defeated mice that displayed social avoidance (Elliott et al., 2010). Futhermore, site-specific knockdown of CRF gene products attenuated stressinduced social avoidance indicating that chronic hyperactivation of the CRF system linked to stress-related emotional disorders such as anxiety, anorexia nervosa and depression via epigenetic mediated limbic hyperthalamus adrenalin (LHPA) axis regulation. (Elliott et al., 2010). Systemic injection of corticosterone in animal models results in a rapid rise of corticosterone concentration in brain (Pariante et al., 2004), as well as in blood (Dallman and Yates, 1969; Yehuda et al., 2006) which may induce the central release of CRF. Enhanced central CRF release would reduce food intake, corresponding to our results shown in Fig. 3b and c, as well as enhance catabolism in the periphery. As shown in Fig. 3c, food intake resumed to the control level after the three day stress and the daily weight gain returned to control level, suggesting that central CRF level and basal metabolism recovered after the three day stress. However post-stress, as shown in Fig. 3a, the stressed animals showed no tendency to make up for the weight not gained during stress. One plausible mechanism explaining why the stressed animals did not make up weight is that the stress protocol altered the homeostatic modulation in the central circuitry (Falkenstein et al., 2000; Heim et al., 2000; Liberzon et al., 1997; Yehuda et al., 2006). In particular, disruption of the canonical CRF >ACTH > glucocorticoid j> CRF feedback loop may prevent, as shown in Fig. 3c, the hyperphagia required to normalize body weight (Heim et al., 2000; Smagin et al., 1999). Besides CRF, a player entirely outside of the glucocorticoid system, serotonin, may be involved in the set point for body weight. It has been demonstrated that the 5-HT2A receptor inhibitor MDL 11,939 (Jiang et al., 2011a), like corticosterone, has no acute effect on body weight during three day stress, but eventually normalizes body weight gain 21 days later. That the glucocorticoid antagonist RU486 diminished but did not significantly reverse the effect of corticosterone on stress induced exaggerated ASR and body weight loss, and that low dose of corticosterone did not significantly reverse stress induced exaggerated ASR, suggests that besides the glucocorticoid system, other mechanisms such as the previously reported 5-HT2A and CRF mediated systems may also be involved in stress induced body weight regulation and exaggerated ASR (Jiang et al., 2011a; Smagin et al., 1999; Jiang et al., 2009). The interactions among the different systems involved in stress induced pathological alterations require further investigation. In conclusion our experiments show that corticosterone administration (3 mg/kg/day) 30 min prior to or after stress attenuates the enhanced ASR induced by three days of restraint and inescapable tail-shock. Administration of corticosterone (3 mg/kg/ day) prior to stress also results in normalization of body weight measured 21 days after stress, although such a regime does not acutely affect the cessation of body weight gain during three days of stress. Neuronal mechanisms may play an important role in the reversal of both enhanced ASR and body weight gain reduction observed when corticosterone is administered immediately prior to or after stress. The possibility exists that corticosterone administration elevates brain as well as plasma concentrations of corticosterone, thereby alleviating the cascade triggered by stress that otherwise results in hyperarousal. Corticosterone administration may thereby impede the development of key symptoms of PTSD and should be further considered as a possible preventive as well as therapeutic agent for traumatic stress-associated psychiatric disorders such as PTSD.

37

Role of funding source This work was supported by Congressionally Directed Medical Research Programs (CDMRP), Award #W81XWH-08-02-006 to Dr. He Li and in part by the Center for the Study of Traumatic Stress (CSTS) at the Uniformed Services University of the Health Sciences (USUHS). Contributors HL was responsible for the design of the study, interpretation of the data, and editing the manuscript. MJ did the experiments, participated in the data analysis and interpretation, and the manuscript preparation. SS, RU and DB revised the manuscript critically and added important intellectual content. GX, LZ, XL made contributions to biochemical measurement of stress response and revised the manuscript critically for important intellectual content. Conflict of interest All authors report no biomedical financial interests or potential conflicts of interest. Acknowledgments The authors gratefully thank Dr. Cara Olsen, a professional biostatistician for the assistance on statistical analysis and Ms. Eleanore H. Gamble for technical help. The opinions or assertions contained herein are the private ones of the authors and are not to be construed as official or reflecting the views of the Department of Defense or the Uniformed Services University of The Health Sciences. References Adamec RE, Blundell J, Burton P. Neural circuit changes mediating lasting brain and behavioral response to predator stress. Neurosci Biobehav Rev 2005;29: 1225e41. Aerni A, Traber R, Hock C, Roozendaal B, Schelling G, Papassotiropoulos A, et al. Low-dose cortisol for symptoms of posttraumatic stress disorder. Am J Psychiatry 2004;161:1488e90. Andrews B, Brewin CR, Philpott R, Stewart L. Delayed-onset posttraumatic stress disorder: a systematic review of the evidence. Am J Psychiatry 2007;164: 1319e26. Belda X, Rotllant D, Fuentes S, Delgado R, Nadal R, Armario A. Exposure to severe stressors causes long-lasting dysregulation of resting and stress-induced activation of the hypothalamic-pituitary-adrenal axis. Ann N Y Acad Sci 2008;1148: 165e73. Blaszczyk JW. Startle response to short acoustic stimuli in rats. Acta Neurobiol Exp (Wars) 2003;63:25e30. Borrell J, De Kloet ER, Bohus B. Corticosterone decreases the efficacy of adrenaline to affect passive avoidance retention of adrenalectomized rats. Life Sci 1984;34: 99e104. Braga MF, Aroniadou-Anderjaska V, Manion ST, Hough CJ, Li H. Stress impairs alpha(1A) adrenoceptor-mediated noradrenergic facilitation of GABAergic transmission in the basolateral amygdala. Neuropsychopharmacology 2004;29: 45e58. Chotiwat C, Kelso EW, Harris RB. The effects of repeated restraint stress on energy balance and behavior of mice with selective deletion of CRF receptors. Stress 2010;13:203e13. Cohen H, Matar MA, Buskila D, Kaplan Z, Zohar J. Early post-stressor intervention with high-dose corticosterone attenuates posttraumatic stress response in an animal model of posttraumatic stress disorder. Biol Psychiatry 2008;64:708e17. Cohen H, Matar MA, Richter-Levin G, Zohar J. The contribution of an animal model toward uncovering biological risk factors for PTSD. Ann N Y Acad Sci 2006;1071: 335e50. Dallman MF, Yates FE. Dynamic asymmetries in the corticosteroid feedback path and distribution-metabolism-binding elements of the adrenocortical system. Ann N Y Acad Sci 1969;156:696e721. Daskalakis NP, Cohen H, Cai G, Buxbaum JD, Yehuda R. Expression profiling associates blood and brain glucocorticoid receptor signaling with trauma-related

38

M. Jia et al. / Journal of Psychiatric Research 60 (2015) 29e39

individual differences in both sexes. Proc Natl Acad Sci U S A 2014;111(37): 13529e34. Datson NA, Polman JA, de Jonge RT, van Boheemen PT, van Maanen EM, Welten J, et al. Specific regulatory motifs predict glucocorticoid responsiveness of hippocampal gene expression. Endocrinology 2011;152:3749e57. de Quervain DJ. Glucocorticoid-induced reduction of traumatic memories: implications for the treatment of PTSD. Prog Brain Res 2008;167:239e47. de Quervain DJ, Aerni A, Roozendaal B. Preventive effect of beta-adrenoceptor blockade on glucocorticoid-induced memory retrieval deficits. Am J Psychiatry 2007;164:967e9. de Quervain DJ, Aerni A, Schelling G, Roozendaal B. Glucocorticoids and the regulation of memory in health and disease. Front Neuroendocrinol 2009;30: 358e70. de Quervain DJ, Henke K, Aerni A, Treyer V, McGaugh JL, Berthold T, et al. Glucocorticoid-induced impairment of declarative memory retrieval is associated with reduced blood flow in the medial temporal lobe. Eur J Neurosci 2003;17: 1296e302. de Quervain DJ, Margraf J. Glucocorticoids for the treatment of post-traumatic stress disorder and phobias: a novel therapeutic approach. Eur J Pharmacol 2008;583: 365e71. de Quervain DJ, Roozendaal B, McGaugh JL. Stress and glucocorticoids impair retrieval of long-term spatial memory. Nature 1998;394:787e90. de Quervain DJ, Roozendaal B, Nitsch RM, McGaugh JL, Hock C. Acute cortisone administration impairs retrieval of long-term declarative memory in humans. Nat Neurosci 2000;3:313e4. Donini LM, Savina C, Cannella C. Eating habits and appetite control in the elderly: the anorexia of aging. Int Psychogeriatr 2003;15:73e87. Eberly RE, Engdahl BE. Prevalence of somatic and psychiatric disorders among former prisoners of war. Hosp Community Psychiatry 1991;42:807e13. Elliott E, Ezra-Nevo G, Regev L, Neufeld-Cohen A, Chen A. Resilience to social stress coincides with functional DNA methylation of the Crf gene in adult mice. Nat Neurosci 2010;13:1351e3. Falkenstein E, Tillmann HC, Christ M, Feuring M, Wehling M. Multiple actions of steroid hormones e a focus on rapid, nongenomic effects. Pharmacol Rev 2000;52:513e56. Fryer CJ, Kinyamu HK, Rogatsky I, Garabedian MJ, Archer TK. Selective activation of the glucocorticoid receptor by steroid antagonists in human breast cancer and osteosarcoma cells. J Biol Chem 2000;275:17771e7. Gazewood JD, Mehr DR. Diagnosis and management of weight loss in the elderly. J Fam Pract 1998;47:19e25. Golier JA, Caramanica K, Demaria R, Yehuda R. A pilot study of mifepristone in combat-related PTSD. Depress Res Treat 2012;2012:393251. Gray JD, Rubin TG, Hunter RG, McEwen BS. Hippocampal gene expression changes underlying stress sensitization and recovery. Mol Psychiatry 2013. Harris RB. Role of set-point theory in regulation of body weight. FASEB J 1990;4: 3310e8. Harris RB, Mitchell TD, Simpson J, Redmann Jr SM, Youngblood BD, Ryan DH. Weight loss in rats exposed to repeated acute restraint stress is independent of energy or leptin status. Am J Physiol Regul Integr Comp Physiol 2002a;282: R77e88. Harris RB, Zhou J, Mitchell T, Hebert S, Ryan DH. Rats fed only during the light period are resistant to stress-induced weight loss. Physiol Behav 2002b;76: 543e50. Harris RB, Zhou J, Youngblood BD, Rybkin II, Smagin GN, Ryan DH. Effect of repeated stress on body weight and body composition of rats fed low- and high-fat diets. Am J Physiol 1998;275:R1928e38. Hauer D, Kaufmann I, Strewe C, Briegel I, Campolongo P, Schelling G. The role of glucocorticoids, catecholamines and endocannabinoids in the development of traumatic memories and posttraumatic stress symptoms in survivors of critical illness. Neurobiol Learn Mem 2014;112C:68e74. Hauer D, Weis F, Krauseneck T, Vogeser M, Schelling G, Roozendaal B. Traumatic memories, post-traumatic stress disorder and serum cortisol levels in longterm survivors of the acute respiratory distress syndrome. Brain Res 2009;1293:114e20. Havel PJ, Busch BL, Curry DL, Johnson PR, Dallman MF, Stern JS. Predominately glucocorticoid agonist actions of RU-486 in young specific-pathogen-free Zucker rats. Am J Physiol 1996;271:R710e7. Heim C, Ehlert U, Hellhammer DH. The potential role of hypocortisolism in the pathophysiology of stress-related bodily disorders. Psychoneuroendocrinology 2000;25:1e35. Herman JP, Ostrander MM, Mueller NK, Figueiredo H. Limbic system mechanisms of stress regulation: hypothalamo-pituitary-adrenocortical axis. Prog Neuropsychopharmacol Biol Psychiatry 2005;29:1201e13. Jia M, Meng F, Smerin SE, Xing G, Zhang L, Su DM, et al. Biomarkers in an animal model for revealing neural, hematologic, and behavioral correlates of PTSD. J Vis Exp 2012;68(10). http://dx.doi.org/10.3791/3361. Jiang X, Xing G, Yang C, Verma A, Zhang L, Li H. Stress impairs 5-HT2A receptormediated serotonergic facilitation of GABA release in juvenile rat basolateral amygdala. Neuropsychopharmacology 2009;34:410e23. Jiang X, Zhang ZJ, Zhang S, Gamble EH, Jia M, Ursano RJ, et al. 5-HT2A receptor antagonism by MDL 11,939 during inescapable stress prevents subsequent exaggeration of acoustic startle response and reduced body weight in rats. J Psychopharmacol 2011a;25:289e97. Jiang XL, Chen A, Smerin S, Zhang L, Li H. Pharmacology of 5-HT2 modulation of amygdala & hypothalamus. In: Anxiety disorders; 2011. p. 39e50.

Joels M, Fernandez G, Roozendaal B. Stress and emotional memory: a matter of timing. Trends Cogn Sci 2011;15:280e8. Joels M, Sarabdjitsingh RA, Karst H. Unraveling the time domains of corticosteroid hormone influences on brain activity: rapid, slow, and chronic modes. Pharmacol Rev 2012;64:901e38. Kaouane N, Porte Y, Vallee M, Brayda-Bruno L, Mons N, Calandreau L, et al. Glucocorticoids can induce PTSD-like memory impairments in mice. Science 2012;335:1510e3. Karst H, Joels M. Corticosterone slowly enhances miniature excitatory postsynaptic current amplitude in mice CA1 hippocampal cells. J Neurophysiol 2005;94: 3479e86. Koch M. The neurobiology of startle. Prog Neurobiol 1999;59:107e28. Laue L, Chrousos GP, Loriaux DL, Barnes K, Munson P, Nieman L, et al. The antiglucocorticoid and antiprogestin steroid RU 486 suppresses the adrenocorticotropin response to ovine corticotropin releasing hormone in man. J Clin Endocrinol Metab 1988;66:290e3. Lemieux AM, Coe CL. Abuse-related posttraumatic stress disorder: evidence for chronic neuroendocrine activation in women. Psychosom Med 1995;57:105e15. Li H, Li X, Smerin S, Zhang L, Jia M, Xing Q, et al. Mitochondrial gene expression profiles and metabolic pathways in the amygdala associated with exaggerated fear in an animal model of PTSD. In: Frontiers in neurology, 5(164); 2014. http:// dx.doi.org/10.3389/fneu.2014.00164. Liberzon I, Krstov M, Young EA. Stress-restress: effects on ACTH and fast feedback. Psychoneuroendocrinology 1997;22:443e53. Liebmann L, Karst H, Joels M. Effects of corticosterone and the beta-agonist isoproterenol on glutamate receptor-mediated synaptic currents in the rat basolateral amygdala. Eur J Neurosci 2009;30:800e7. Manion ST, Gamble EH, Li H. Prazosin administered prior to inescapable stressor blocks subsequent exaggeration of acoustic startle response in rats. Pharmacol Biochem Behav 2007;86:559e65. McGaugh JL. Making lasting memories: remembering the significant. Proc Natl Acad Sci U S A 2013;110(Suppl. 2):10402e7. Miller MW, McKinney AE, Kanter FS, Korte KJ, Lovallo WR. Hydrocortisone suppression of the fear-potentiated startle response and posttraumatic stress disorder. Psychoneuroendocrinology 2011;36:970e80. Myers MW, Kimbrell TA, Booe LQ, Freeman TW. Weight loss and PTSD symptom severity in former POWs. J Nerv Ment Dis 2005;193:278e80. Osterlund C, Spencer RL. Corticosterone pretreatment suppresses stress-induced hypothalamic-pituitary-adrenal axis activity via multiple actions that vary with time, site of action, and de novo protein synthesis. J Endocrinol 2011;208: 311e22. Pariante CM, Thomas SA, Lovestone S, Makoff A, Kerwin RW. Do antidepressants regulate how cortisol affects the brain? Psychoneuroendocrinology 2004;29: 423e47. Pu Z, Krugers HJ, Joels M. Beta-adrenergic facilitation of synaptic plasticity in the rat basolateral amygdala in vitro is gradually reversed by corticosterone. Learn Mem 2009;16:155e60. Quirarte GL, Galvez R, Roozendaal B, McGaugh JL. Norepinephrine release in the amygdala in response to footshock and opioid peptidergic drugs. Brain Res 1998;808:134e40. Reul JM, De Kloet ER, van Sluijs FJ, Rijnberk A, Rothuizen J. Binding characteristics of mineralocorticoid and glucocorticoid receptors in dog brain and pituitary. Endocrinology 1990;127:907e15. Resnick HS, Yehuda R, Acierno R. Acute post-rape plasma cortisol, alcohol use, and PTSD symptom profile among recent rape victims. Ann N Y Acad Sci 1997;821:433e6. Richardson Morton KD, Van de Kar LD, Brownfield MS, Lorens SA, Napier TC, Urban JH. Stress-induced renin and corticosterone secretion is mediated by catecholaminergic nerve terminals in the hypothalamic paraventricular nucleus. Neuroendocrinology 1990;51:320e7. Roozendaal B. Stress and memory: opposing effects of glucocorticoids on memory consolidation and memory retrieval. Neurobiol Learn Mem 2002;78:578e95. Roozendaal B, de Quervain DJ, Schelling G, McGaugh JL. A systemically administered beta-adrenoceptor antagonist blocks corticosterone-induced impairment of contextual memory retrieval in rats. Neurobiol Learn Mem 2004;81:150e4. Roozendaal B, Okuda S, de Quervain DJ, McGaugh JL. Glucocorticoids interact with emotion-induced noradrenergic activation in influencing different memory functions. Neuroscience 2006;138:901e10. Schelling G, Briegel J, Roozendaal B, Stoll C, Rothenhausler HB, Kapfhammer HP. The effect of stress doses of hydrocortisone during septic shock on posttraumatic stress disorder in survivors. Biol Psychiatry 2001;50:978e85. Schelling G, Richter M, Roozendaal B, Rothenhausler HB, Krauseneck T, Stoll C, et al. Exposure to high stress in the intensive care unit may have negative effects on health-related quality-of-life outcomes after cardiac surgery. Crit Care Med 2003;31:1971e80. Schelling G, Roozendaal B, Krauseneck T, Schmoelz M, De Quervain D, Briegel J. Efficacy of hydrocortisone in preventing posttraumatic stress disorder following critical illness and major surgery. Ann N Y Acad Sci 2006;1071:46e53. Schwabe L, Joels M, Roozendaal B, Wolf OT, Oitzl MS. Stress effects on memory: an update and integration. Neurosci Biobehav Rev 2012;36:1740e9. Servatius RJ, Brennan FX, Moldow R, Pogach L, Natelson BH, Ottenweller JE. Persistent hormonal effects of stress are not due to reduced food intake or exposure to stressed rats. Endocrine 2001;14:181e7. Servatius RJ, Ottenweller JE, Natelson BH. Delayed startle sensitization distinguishes rats exposed to one or three stress sessions: further evidence toward an animal model of PTSD. Biol Psychiatry 1995;38:539e46.

M. Jia et al. / Journal of Psychiatric Research 60 (2015) 29e39 Smagin GN, Howell LA, Redmann Jr S, Ryan DH, Harris RB. Prevention of stressinduced weight loss by third ventricle CRF receptor antagonist. Am J Physiol 1999;276:R1461e8. Solomon Z, Kotler M, Shalev A, Lin R. Delayed onset PTSD among Israeli veterans of the 1982 Lebanon War. Psychiatry 1989;52:428e36. Soravia LM, Heinrichs M, Aerni A, Maroni C, Schelling G, Ehlert U, et al. Glucocorticoids reduce phobic fear in humans. Proc Natl Acad Sci U S A 2006;103: 5585e90. Speed N, Engdahl B, Schwartz J, Eberly R. Posttraumatic stress disorder as a consequence of the POW experience. J Nerv Ment Dis 1989;177:147e53. Suris A, North C, Adinoff B, Powell CM, Greene R. Effects of exogenous glucocorticoid on combat-related PTSD symptoms. Ann Clin Psychiatry 2010;22: 274e9. Tanaka T, Yokoo H, Mizoguchi K, Yoshida M, Tsuda A, Tanaka M. Noradrenaline release in the rat amygdala is increased by stress: studies with intracerebral microdialysis. Brain Res 1991;544:174e6. Tomb DA. The phenomenology of post-traumatic stress disorder. Psychiatr Clin North Am 1994;17:237e50. Weis F, Kilger E, Roozendaal B, de Quervain DJ, Lamm P, Schmidt M, et al. Stress doses of hydrocortisone reduce chronic stress symptoms and improve healthrelated quality of life in high-risk patients after cardiac surgery: a randomized study. J Thorac Cardiovasc Surg 2006;131:277e82. Weiss SJ. Neurobiological alterations associated with traumatic stress. Perspect Psychiatr Care 2007;43:114e22. Wilson PN, Osbourn DF. Compensatory growth after undernutrition in mammals and birds. Biol Rev Camb Philos Soc 1960;35:324e63.

39

Wixted JT. The psychology and neuroscience of forgetting. Annu Rev Psychol 2004;55:235e69. Xing G, Carlton J, Zhang L, Jiang X, Fullerton C, Li H, et al. Cannabinoid receptor expression and phosphorylation are differentially regulated between male and female cerebellum and brain stem after repeated stress: implication for PTSD and drug abuse. Neurosci Lett 2011;502:5e9. Yehuda R. Sensitization of the hypothalamic-pituitary-adrenal axis in posttraumatic stress disorder. Ann N Y Acad Sci 1997;821:57e75. Yehuda R. Status of glucocorticoid alterations in post-traumatic stress disorder. Ann N Y Acad Sci 2009;1179:56e69. Yehuda R, Yang RK, Buchsbaum MS, Golier JA. Alterations in cortisol negative feedback inhibition as examined using the ACTH response to cortisol administration in PTSD. Psychoneuroendocrinology 2006;31:447e51. Yuen EY, Liu W, Karatsoreos IN, Ren Y, Feng J, McEwen BS, et al. Mechanisms for acute stress-induced enhancement of glutamatergic transmission and working memory. Mol Psychiatry 2011;16:156e70. Zhang ZJ, Jiang XL, Zhang SE, Hough CJ, Li H, Chen JG, et al. The paradoxical effects of SKF83959, a novel dopamine D1-like receptor agonist, in the rat acoustic startle reflex paradigm. Neurosci Lett 2005;382:134e8. Zhou M, Hoogenraad CC, Joels M, Krugers HJ. Combined beta-adrenergic and corticosteroid receptor activation regulates AMPA receptor function in hippocampal neurons. J Psychopharmacol 2012;26:516e24. Zohar J, Yahalom H, Kozlovsky N, Cwikel-Hamzany S, Matar MA, Kaplan Z, et al. High dose hydrocortisone immediately after trauma may alter the trajectory of PTSD: interplay between clinical and animal studies. Eur Neuropsychopharmacol 2011;21:796e809.