Juvenile stress-induced alteration of maturation of the GABA_A receptor α subunit in the rat

Abstract

Profound evidence indicates that GABA_A receptors are important in the control of physiological response to stress and anxiety. The α subunit type composition contributes significantly to the functional characterization of the GABA_A receptors. The α₂, α₃, α₅ subunits are predominately expressed in the brain during embryonic and early postnatal periods of normal rats, whilst α₁ are most prominent during later developmental stages. In the present study, we examined the long-term effects of juvenile stress on GABA α subunit expression in adulthood in the amygdala and hippocampus. We applied the elevated platform stress paradigm at juvenility and used the open-field and startle response tests to assess anxiety level in adulthood. Juvenile stress effects without behavioural tests in adulthood were also examined since previous studies indicated that the mere exposure to these tests might be stressful for rats, enhancing the effects of the juvenile exposure to stress. In adulthood, we quantitatively determined the level of expression of α₁, α₂ and α₃ in the hippocampus and amygdala. Our results indicate that subjecting juvenile stressed rats to additional challenges in adulthood results in an immature-like expression profile of these subunits. To test for potential functional implications of these alterations we examined the effects of the anxiolytic (diazepam) and the sedative (brotizolam) benzodiazepines on juvenile stressed and control rats following additional challenges in adulthood. Juvenile stressed rats were more sensitive to diazepam and less sensitive to brotizolam, suggesting that the alterations in GABA α subunit expression in these animals have functional consequences.

Introduction

The functional anatomy of emotional behaviour involves the hippocampus- and amygdala-based neurocircuits (McEwen and Sapolsky, 1995; Ninan, 1999). The amygdala is implicated in emotional behaviour, and emotional related memories (Kluver and Bucy, 1939; LeDoux, 1995). The hippocampus is involved in the storage of explicit memory (Eichenbaum, 1999; Squire, 1992) and in the initiation of spatio-temporal and contextual representations (Eichenbaum et al., 1992; Squire, 1992). Abnormalities in these brain pathways may be the cause of alterations of cognitive and emotional abilities (Brown et al., 1999; Sheline et al., 1998).

The juvenile period in rats is suggested to be of relevance to human childhood (Spear, 2000). At this age the pups become more independent and spend more time in social interactions. Yet, they are still sexually immature (Martin and Berthoz, 2002). Acute juvenile stress was found to increase vulnerability to stressful events in adulthood (Avital and Richter-Levin, 2005; Tsoory and Richter-Levin, 2006).

Accumulating data suggests that γ-aminobutyric acid (GABA) functioning abnormality is involved in the pathophysiology of mood disorders and that GABA neuromodulation contributes to the mechanism of action of anxiolytics, mood stabilizers and antidepressants (Biggo et al., 1995; Petty, 1995).

GABA_A receptor (GABA_AR) is an ionotropic receptor, allowing chloride ions to cross the neuronal membrane. The mature GABA_AR is assumed to be a pentameric assembly of subunits (α₁–α₆, β₁–β₃, γ₁–γ₃, δ, ε, θ, π and ρ) (Parsian and Cloninger, 1997). The composition of the assembly produces different receptor subtypes (McKernan and Whiting, 1996).

The α subunit type composition contributes significantly to the functional characterization of the GABA_ARs (Mohler et al., 1995a,b). Specific α subunits are suggested to be preferentially associated with the induction of certain types of behaviour, including anxiety (Gulinello et al., 2001; Rudolph et al., 1999). α Subunit expression varies dramatically between embryonic and postnatal ontogeny (Laurie et al., 1992; Poulter et al., 1992). Until the early postnatal period of the rat predominately α₂, α₃, and α₅ subunits are expressed in the brain, whilst α₁ is the most prominent thereafter (Fritschy et al., 1994; Poulter et al., 1999).

In the present study we examined the effects of juvenile stress on adulthood expression of GABA_AR α subunits in the amygdala and hippocampus. As a juvenile stress we applied the elevated platform stress paradigm, which is similar to a protocol named small platform stress (Pokk and Vali, 2001; Pokk and Zharkovsky, 1998), and which comprises several factors of stress like novelty, restraint, uncontrollability, and unpredictability (Avital and Richter-Levin, 2005). To validate the impact into adulthood of the exposure to juvenile stress we have employed the open-field and startle response tests. However, previous studies indicate that the mere exposure to these tests might be stressful for rats (Day et al., 2005; Prut and Belzung, 2003; van den Buuse et al., 2002). Subjecting juvenile stressed animals to an additional challenge in adulthood resulted in heightened anxiety levels (Tsoory and Richter-Levin, 2006). It was thus possible that subjecting an animal to the open-field and startle response tests in adulthood may enhance the effect of the juvenile exposure to stress. To test for this possibility, we added an additional group that was exposed to the juvenile stress, but not to the behavioural assessments in adulthood. Following the behavioural assessment, or at a similar time-point, we quantitatively determined the levels of expression of α₁, α₂ and α₃ subunits in the amygdala and hippocampus.

Materials and methods

Animals

Forty-four male Sprague–Dawley rats, aged 22 d, weighing 35–49 g were supplied by Harlan Laboratories (Jerusalem, Israel). The rats were maintained for the entire duration of the experiment on a 12-h light–dark cycle (lights on 07:00 hours); room temperature 22±2°C, two rats per cage (26×42×18 cm) on sawdust bedding, and provided with water and solid food pellets (Teklad Global Diet 2018S, Harlan Teklad Ltd, WI, USA) ad libitum.

Study design

The effects of juvenile stress on the mature expression of GABA_AR α subunits

Rats were randomly assigned to one of four groups:

1. Juvenile stress group (J): exposed to the platform stress paradigm at juvenility (aged 27–29 d).

2. Juvenile(+) stress group (J+): exposed to the platform stress paradigm at juvenility (aged 27–29 d) and to the open-field and startle reflex response tests, in adulthood (aged 60 d).

3. Control(+) (C+): exposed only to the tests in adulthood (aged 60 d).

4. Control group (C): not exposed to stress.

The amygdala and the hippocampus were harvested 24 h after the adulthood challenges, lysed by a lysis buffer and prepared for analysis utilizing immunoblotting.

Behavioural procedures

Repeated elevated platform stress

Following a period of 4 d habituation to the housing conditions, individual rats (aged 27 d) were placed for 30 min on an elevated black platform (12.0×12.0 cm), positioned in the middle of a water tank (diameter: 1.7 m, 50 cm high rim), with the top surface elevated 3 cm above the water surface. Animals were subjected to the elevated platform paradigm three times (inter-trial interval 1 h) for three consecutive days. Following the completion of the elevated platform stress paradigm, the rats were returned to their home cage and were not handled until the age of 60 d except for weekly cage sawdust bedding maintenance.

Open-field test

The open-field test was carried out according to methods described previously (Avital et al., 2001; Carli et al., 1989; Lemoine et al., 1990). For 3 min the number of line crossings was manually recorded. All rats were tested between 11:00 and 17:00 hours by the same experimenter.

Acoustic startle reflex test

The acoustic startle reflex test is used extensively to index fear and anxiety in rodents (e.g. Faraday and Grunberg, 2000) and humans (e.g. Riba et al., 2001). The startle reflex response (Patrick, 1994) was measured using an automated JR Startle Box (Hamilton-Kinder, Poway, CA, USA) positioned in a dimly lit room. Following exposure to the open-field test, animals were habituated for 30 min to the startle test room before placing them in the chamber. The animals were presented with eight auditory stimuli (40.0 ms of 115 dB noise stimulus over a 70-dB background), spaced by a 1-min interval. When no tone was delivered, a white noise was applied. The maximal startle reflex response for each animal was calculated as the average of the responses to the eight auditory stimuli. All rats were tested between 11:00 and 17:00 hours by the same experimenter.

Comparison of the sensitivity of adult juvenile stressed and control rats to the effects of benzodiazepines (BZs)

Sensitivity to the anxiolytic effect of the BZ diazepam

Twenty-four hours following the additional challenges in adulthood, rats were injected with diazepam solution (1 mg/kg diazepam, in saline i.p.; Teva Pharmaceuticals, Petach Tikva, Israel) or with vehicle. Twenty-five minutes later the rats were habituated for 5 min to the test room before placing them into the startle reflex response chamber. Three parameters were measured:

Pre-pulse inhibition (PPI) (Braff et al., 1999; Swerdlow et al., 1992): Animals were presented with five auditory stimuli (40.0 ms of 115 dB noise stimulus) spaced by 1-min intervals, each preceded by a ‘weak’ prepulse (e.g. 5–15 dB over a 70-dB background). When no tone was delivered, a white noise was applied. The maximal response of each animal was calculated as the average of the responses to the five auditory stimuli.

Startle reflex response: The animals are presented with five auditory stimuli (40.0 ms of 115 dB noise stimulus), spaced by 1-min intervals. When no tone was delivered, a white noise was applied. The maximal response for each animal was calculated as the average of the responses to the five auditory stimuli.

A third measure was calculated: the delta between startle reflex alone and PPI.

All rats were tested between 11:00 and 17:00 hours by the same experimenter.

Comparison of the sensitivity of adult juvenile stressed and control rats to the anaesthetic effect of the BZ brotizolam

Twenty-four hours following the additional challenges in adulthood rats were injected with 0.75 mg/kg i.p. brotizolam (RAFA Laboratories, Jerusalem, Israel), in order to determine the effects of juvenile stress on sensitivity to its sedative effects.

Immediately after the injection rats were placed into an automated locomotion test chamber and locomotion was followed up for 72 min. Behaviour was automatically recorded and subsequently analysed at 1-s intervals. The percent time freezing was calculated at 0–2 min, 5–7 min, and for 2 min every 10-min interval thereafter.

Brain tissue harvesting

The following brain regions were collected as described previously (Tsoory et al., 2008): Basolateral amygdala, (left and right pooled together) and the whole hippocampus. Immediately following decapitation, the brain was placed on an ice-cooled glass dish. First, the hippocampus was dissected out. Then a thick (∼1 mm) coronal slice was cut ∼6 mm anterior of the interaural plane (indicated by the rostal end of the cerebellum), and the basolateral amygdala were cut out: (1) cutting the ventral part of the slice just below the base of the optic tract; (2) cutting orthogonally to the first cut and in parallel to the corpus callosum about 2 mm medial to the rhinal fissure; (3) cutting out a small isosceles (∼1.5 mm) right triangle.

Biochemical methods

Homogenization

Brains of the two groups of animals were removed 24 h subsequent to the anxiety tests. The hippocampus and amygdala were immediately homogenized in a glass Teflon homogenizer in 300 µl (amygdala)/600 µl (hippocampus) lysis buffer (10 mm Hepes, 2 mm EDTA, 2 mm EGTA, 0.5 mm DTT, 10 µl/ml leupeptin, 10 µl/ml aprotinin). After the protein levels in each sample were determined (Bradford method, Bio-Rad Protein Dye Reagent; Bio-Rad, Hercules, CA, USA), we diluted them in SDS sample buffer (10% glycerol, 5% β-mercaptoethanol and 2.3% SDS in 62.5 mm Tris–HCl, pH 6.8), and boiled (100°C) for 5 min. Samples were then stored at −80°C until further analysis.

Western blot analysis

Aliquots in SDS sample buffer were subjected to SDS–PAGE (10% polyacrylamide) and immunoblot analysis.

Following blotting to a nitrocellulose membrane, lysate homogeneity lanes were stained by Ponceau staining. Blots were blocked with 5% BSA+5% Blotto solution (Chemicon, Temecula, CA, USA) for 1 h at room temperature, followed by 1 h incubation with the primary antibody at room temperature. Following three short washes in TBST washing buffer (0.9% w/v NaCl, 0.05% v/v Tween-20 and 100 mm Tris–HCl, pH 7.6); the blots were subsequently incubated for 1 h with an HRP-linked secondary antibody at room temperature before being reacted with ECL+ substrate (Amersham, Piscataway, NJ, USA).

Reagents

Antibodies: GABA_AR α₁ subunit [cat. no. AB5946 (1:2000) rabbit polyclonal; Chemicon]. GABA_AR α₂ subunit (N-19) [cat no. AB5948 (1:2500) rabbit polyclonal; Chemicon]. GABA_AR α₃ subunit [cat no. AB5594 (1:200) rabbit polyclonal; Chemicon]. Actin (N-19) [cat. no. sc-1616 (1:1500) polyclonal goat antibody; Santa Cruz Biotechnology, Santa Cruz, CA, USA]. Rabbit anti-goat (IgG) horseradish peroxidase (HRP) conjugated (Santa Cruz Biotechnology). Goat anti-rabbit (IgG) horseradish peroxidase (HRP) conjugated and the enhanced chemiluminescense (ECL+) kit were obtained from Amersham. All other chemicals were of analytical grade or the highest grade available.

Quantification

Quantification was performed using a CCD camera (XRS; Bio-Rad). The expression level for each depicted protein was calculated as the ratio between the band intensity of the protein divided by that of the normalized β-actin, a cytoskeletal protein used as an internal control. No differences were detected in β-actin levels throughout between the different groups.

Statistics

Results are expressed as means ±s.e.m. Behavioural indices were assessed by t test when two groups were compared or by one-way ANOVA (post-hoc Scheffé test) when four groups were compared. GABA_AR α subunit expression were assessed by two-way ANOVA and further analysed by one-way ANOVA. All post-hoc comparisons were made by Scheffé test.

Ethical approval

The experiments were approved by the institutional Animal Care and Use Committee of the University of Haifa, and adequate measures were taken to minimize pain or discomfort, in accordance with the guidelines laid down by the NIH (Bethesda, MD, USA) regarding the care and use of animals for experimental procedures.

Results

The effects of juvenile stress on stress-related behaviours in adulthood

The open-field and startle reflex response tests were conducted 31 d after exposure to juvenile stress.

Exposure to juvenile stress significantly decreased the activity level in the open-field test, measured as the total number of line crossings [J+ compared to C+: t(27)=2.34, p<0.05] (Figure 1), and increased startle response [t(27)=3.01, p<0.005] (J+ compared to C+: Figure 1b).

The effects of juvenile stress on adulthood GABA_AR α subunit expression in the amygdala

GABA_AR α₁ subunit expression

Two-way ANOVA for GABA_AR α₁ subunit expression indicated a significant main effect for juvenile stress alone [F(1, 38)=17.47, p<0.0001] but not for adulthood emotional challenge [F(1, 38)=0.244, n.s]. The interaction between juvenile stress and adulthood emotional challenge was not significant [F(1, 38)=0.724, n.s].

One-way ANOVA indicated a significant difference between groups in GABA_AR α₁ subunit expression, in the amygdala [F(3, 40)=7.79, p<0.0001] (Figure 2). Post-hoc Scheffé test revealed a significant decrease in GABA_AR α₁ in the J+ group compared to the C (p<0.05) and C+ (p<0.001) groups.

GABA_AR α₂ subunit expression

Two-way ANOVA for GABA_AR α₂ subunit expression indicated a significant main effect for adulthood emotional challenge [F(1, 38)=34.45, p<0.0001] but not for juvenile stress alone [F(1, 38)=1.92, n.s]. The interaction between juvenile stress and adulthood emotional challenge was also significant [F(1, 38)=9.96, p<0.001].

One-way ANOVA showed a significant difference between groups in GABA_AR α₂ subunit expression, in the amygdala [F(3, 40)=7.89, p<0.0001] (Figure 2). Scheffé post-hoc test revealed a significant increase in the expression of GABA_AR α₂ subunit in the J+ group compared to the J (p<0.003), C+ (p<0.005) and C (p<0.05) groups.

GABA_AR α₃ subunit expression

Two-way ANOVA for GABA_AR α₃ subunit expression indicated a significant main effect for juvenile stress [F(1, 38)=4.88, p<0.05] but not for adulthood emotional challenge [F(1, 38)=0.469, n.s]. The interaction between juvenile stress and adulthood emotional challenge was not significant [F(1, 38)=0.643, n.s].

One-way ANOVA showed a trend for a difference between groups in GABA_AR α₃ subunit expression, in the amygdala [F(3, 40)=2.67, p=0.06] (Figure 2).

An additional measure was calculated: the ratio between the expression levels of GABA_AR α₁ to α₂ or α₃ subunits.

The relative expression of GABA_AR α₁:α₂

Two-way ANOVA for the ratio between α₁:α₂ subunit expression indicated a significant main effect for juvenile stress [F(1, 38)=4.88, p<0.001] but not for adulthood emotional challenge [F(1, 38)=0.165, n.s]. The interaction between juvenile stress and adulthood emotional challenge was significant [F(1, 38)=5.82, p<0.05].

One-way ANOVA showed a significant difference between groups in the relative expression of GABA_AR α₁:α₂ subunits, in the amygdala [F(3, 40)=9.87, p<0.0001] (Figure 3). Scheffé post-hoc test revealed a significant decrease in the relative expression of GABA_AR α₁:α₂ subunits in J+ stressed rats compared to the J (p<0.05), C+ (p<0.0001) and C (p<0.05) groups (Figure 3).

The relative expression of GABA_AR α₁:α₃

Two-way ANOVA for the ratio between α₁:α₃ subunit expression indicated a significant main effect for juvenile stress [F(1, 38)=15.26, p<0.0001] but not for adulthood emotional challenge [F(1, 38)=0.014, n.s]. The interaction between juvenile stress and adulthood emotional challenge was not significant [F(1, 38)=0.492, n.s].

One-way ANOVA showed a significant difference between groups in the relative expression of GABA_AR α₁:α₃ subunits, in the amygdala [F(3, 40)=6.56, p<0.005] (Figure 3). Scheffé post-hoc test revealed a significant decrease in the relative expression of GABA_AR α₁:α₃ subunits in juvenile stressed rats compared to the C+ group (p<0.05) (Figure 3). Additionally, there was a significant decrease in the relative expression of GABA_AR α₁:α₃ subunits in J+ stressed rats compared to the C+ (p<0.001) and C (p<0.05) groups (Figure 3).

The effects of juvenile stress on adulthood GABA_AR α subunit expression in the hippocampus

GABA_AR α₁ subunit expression

Two-way ANOVA for GABA_AR α₁ subunit expression indicated a significant main effect for adulthood emotional challenge [F(1, 38)=23.67, p<0.0001] but not for juvenile exposure to stress alone [F(1, 38)=0.051, n.s]. The interaction between juvenile stress and adulthood emotional challenge was significant [F(1, 38)=17.68, p<0.01].

One-way ANOVA showed a significant difference between groups in GABA_AR α₁ subunit expression, in the hippocampus [F(3, 40)=14.12, p<0.0001] (Figure 4). Post-hoc Scheffé test revealed a significant increase in GABA_AR α₁ in the J+ group compared to J (p<0.0001), C+ (p<0.005) and C (p<0.05) groups.

GABA_AR α₂ subunit expression

Two-way ANOVA for GABA_AR α₂ subunit expression indicated a significant main effect for exposure to adulthood emotional challenge [F(1, 38)=34.45, p<0.0001] and for juvenile exposure to stress [F(1, 38)=6.49, p<0.01]. The interaction between juvenile stress and adulthood emotional challenge was also significant [F(1, 38)=14.2, p<0.01].

One-way ANOVA showed a significant difference between groups in, GABA_AR α₂ subunit expression, in the hippocampus [F(3, 40)=24.09, p<0.0001] (Figure 4). Scheffé post-hoc test revealed a significant increase in the expression of GABA_AR α₂ subunit in the J+ group compared to the J (p<0.0001), C+ (p<0.0001) and the C (p<0.0001) groups.

GABA_AR α₃ subunit expression

Two-way ANOVA for GABA_AR α₃ subunit expression revealed no significant main effect for exposure to adulthood emotional challenge [F(1, 38)=0.003, n.s] and not for juvenile exposure to stress [F(1, 38)=0.236, n.s]. The interaction between juvenile stress and adulthood emotional challenge was also not significant [F(1, 38)=0.64, n.s].

Additionally, one-way ANOVA in the hippocampus, indicated no significant differences between the groups [F(3, 40)=n.s.] (Figure 4).

An additional measure was calculated: the ratio between the expression levels of GABA_AR α₁ to α₂ and α₃ subunits.

The relative expression of GABA_AR α₁:α₂

Two-way ANOVA for the ratio between α₁:α₂ subunit expression indicated a significant main effect for adulthood emotional challenge [F(1, 38)=4.41, p<0.05] but not for juvenile exposure to stress [F(1, 38)=2.01, n.s]. The interaction between juvenile stress and adulthood emotional challenge was also not significant [F(1, 38)=2.21, n.s].

One-way ANOVA show a clear trend toward a difference between groups in the relative expression of GABA_AR α₁:α₂ subunits, in the hippocampus [F(3, 40)=2.67, p58] (Figure 5). Scheffé post-hoc test revealed a significant decrease in the relative expression of GABA_AR α:α₂ subunits in J+ stressed rats compared to the J (p<0.05), C+ (p<0.05) and C (p<0.05) groups (Figure 5).

The relative expression of GABA_AR α₁:α₃

Two-way ANOVA for the ratio between α₁:α₂ subunit expression indicated no significant main effect for adulthood emotional challenge [F(1, 38)=3.83, n.s.] and not for juvenile exposure to stress [F(1, 38)=0.039, n.s]. The interaction between juvenile stress and adulthood emotional challenge was also not significant [F(1, 38)=2.63, n.s].

In the hippocampus, no significant differences were found between the groups [F(3, 40)=n.s.] (Figure 5).

Comparison of the effects of BZs on behavioural outcomes in controls and juvenile stressed animals

The anxiolytic and sedative effects of BZs on behavioural outcomes were measured 31 d after exposure to juvenile stress and 24 h following the additional emotional challenges.

Sensitivity to the anxiolytic BZ diazepam

Anxiety-like behaviours were decreased following treatment with the anxiolytic diazepam in juvenile stressed animals compared to behaviour under vehicle administration, across the three parameters we measured. One-way ANOVA showed a significant difference between groups in PPI [F(3, 23)=5.86, p<0.05] (Figure 6a), startle reflex response test [F(3, 23)=8.09, p<0.01] (Figure 6a) and the delta between the startle reflex response and the PPI [F(3, 23)=6.65, p<0.05] (Figure 6a). Scheffé post-hoc test revealed an increase in the behavioural outcome of J+/vehicle rats compared to the juvenile/diazepam (p<0.05), C+/vehicle (p<0.05) and C/diazepam groups (p<0.05) (Figure 6). Nevertheless this difference diminished following diazepam administration (J+/diazepam group).

To further elucidate the difference between the anxiolytic effect of diazepam on juvenile stressed rats compared to the lack of effect in control animals we compared the delta between the behavioural responses of untreated (i.e. vehicle) and treated (i.e. diazepam injected) animals in each group.

Exposure to juvenile stress and an additional emotional challenge in adulthood significantly increased the effect of the anxiolytic diazepam in the three parameters we measured: PPI [J+ compared to C+: t(10)=2.15, p<0.05] (Figure 6b), startle reflex response test [J+ compared to C+: t(10)=2.69, p<0.05] (Figure 6b) and the delta between the startle reflex response and the PPI [J+ compared to C+: t(10)=2.61, p<0.05] (Figure 6b).

Sensitivity to the sedative BZ brotizolam

Exposure to the combination of juvenile stress and an additional challenge in adulthood significantly decreased the sedative effect of the BZ brotizolam on the two parameters we have measured: induction of sleep (measured as percent immobility immediately following drug administration) [J+ compared to C+: t(10)=3.95, p<0.005] (Figure 7), and the time it took the rat to recover from the drug-induced sleep (measured as percent immobility in minutes 60–62 and 70–72 following drug administration) [J+ compared to C+: t(10)=3.39, p<0.01; t(10)=6.81, p<0.0001] (Figure 7).

Discussion

The exposure of juvenile rats to stress was previously found to result in impaired coping behaviour in adulthood (Avital and Richter-Levin, 2005). Here we examine whether the exposure to juvenile stress modifies the expression of GABA_AR subunits in adulthood. Our results show that indeed, juvenile stress results in modulation of GABA_AR subunit expression both in the hippocampus and the amygdala.

We first further validated Avital and Richter-Levin's (2005) findings regarding the long-term behavioural impact of exposure to juvenile stress on anxiety levels. However, previous studies indicate that the mere exposure to the open-field and startle response tests are considered stressful to rats (Day et al., 2005; Prut and Belzung, 2003; van den Buuse et al., 2002). The anxious behaviour in the open-field test is triggered by three different factors: individual testing, agoraphobia (as the arena is very large relative to the animal's breeding or natural environment) (Prut and Belzung, 2003), and novelty. Introducing a rat to these stressful components led to the activation of stress responses, including elevation in blood pressure, heart rate and body temperature of the rat (van den Buuse et al., 2002). The acoustic startle reflex test is used extensively to index fear and anxiety in rodents (e.g. Faraday and Grunberg, 2000) and humans (e.g. Riba et al., 2001). Yet this apparatus also serves to induce a loud noise-related anxiety (Day et al., 2005). An acoustic tone at an intensity of 85 dBA or above is considered stressful, as determined by a significant increase in plasma ACTH and corticosterone levels (Day et al., 2005). Therefore we considered these tests not only as indicators of anxiety but also as challenges. Indeed, our results demonstrate that subjecting adult juvenile stressed rats to additional challenges in adulthood resulted in modulation of the α₁, α₂, and α₃ subunits, mainly in the amygdala, producing an immature-like expression profile. Juvenile stress exposure alone was not sufficient to result in a similar effect. Exposure only to the behavioural challenges in adulthood did not affect the GABA_AR α subunits expression profile. In contrast, J+ rats exhibited decreased α₁:α₂ expression ratio compared to all the other groups in both the amygdala and hippocampus. Their α₁:α₃ expression ratio was lower than that of both C and C+ rats only in the amygdala.

Taken together, it appears that the observed alterations did not stem from the juvenile stress alone or from the challenges in adulthood, but from the interaction of the two.

Regulation of anxiety is associated with the function of the GABA_AR (Crestani et al., 1999; Sanders and Shekhar, 1995). In fact, unique GABA_AR isoforms in specific brain regions are thought to differentially modulate anxiety (Crestani et al., 1999; Menard and Treit, 1999). Modulation of α subunits in the amygdala following stress exposure is particularly important given that the amygdala is believed to serve as an interface between the environment and effector organs generating behavioural responses associated with anxiety (Da Cunha et al., 1992). Generally, a pentameric CNS GABA_AR consists of at least one α subunit and one β subunit, and one or more γ, δ, or ε subunits (Fritschy and Mohler, 1995; Sieghart, 1992). Although little is known about the specific properties of each subunit, functional studies demonstrate that the subunit composition of the receptor determines its electrophysiological and pharmacological properties (Barnard et al., 1998; Narahashi, 1999), thus allowing a variety of adaptive changes (Olsen et al., 1999). Studies in rodents have shown that developmental changes are more pronounced among the α subunits than among the other subunits. Specifically, α₂ or α₃ expression at earlier periods of life is predominantly switched to α₁ expression in adults (Fritschy et al., 1994; Hornung and Fritschy, 1996; Laurie et al., 1992). Our results demonstrate that exposing juvenile stressed rats to additional challenges in adulthood resulted in modulation of GABA_AR subunit expression both in the hippocampus and the amygdala. However, while stressed animals exhibited an immature expression profile of these subunits in the amygdala, the stress-induced alterations were less distinguished in the hippocampus, i.e. α₃ expression was not affected. Kang et al. (1991) showed an increase in α₁ mRNA in the hippocampus but not in the cortex following an electroconvulsive shock. Conversely, 1–3 d after exposure to social stress there was an increase in α₁ mRNAs in the cortex but not in the hippocampus. In another study, swim stress induced a reduction of α₁ mRNA in the hippocampus following 14 d but not 7 d of exposure to stress (Montpied et al., 1993). Collectively, these findings suggest that in the hippocampus stress-induced changes of the GABA_AR subunit profile are dependent upon several factors, such as the nature and duration of the stress (Drugan and Holmes., 1991). It is therefore reasonable to assume that the expression profile of α₁ we observed in the hippocampus is strongly dependent on the specific stress paradigm and the time-frame employed here. Another possible explanation for the lack of changes in α₃ in the hippocampus could be the very low level of expression of α₃ there (Low et al., 2000). Other techniques may be required in order to investigate α₃ in the hippocampus.

Exposure to juvenile stress followed by additional challenges in adulthood induced a reduction in the ratio between α₁:α₃ in the amygdala but not in the hippocampus while the reduction in the relative expression between α₁:α₂ was significantly decreased in both regions. Recent knockout studies in mice showed that deletion of the α₁ subunit leads to an increase in the expression of α₂ and/or α₃ subunits. These changes are thought to be compensatory and have considerable implications in the treatment of mood and anxiety disorders (Kralic et al., 2002; Sur et al., 2001). In particular, these changes are significant with regards to the anxiolytic action of BZs that are thought to target mainly α₂- and α₃-containing receptors (Low et al., 2000, McKernan et al., 2000, Möhler et al., 2002, Rudolph and Möhler, 2004).

Diversity in BZ pharmacology is generated by heterogeneity of the α subunit of the GABA_AR (Pritchett et al., 1989a). BZs bind with different affinity to various α subunits of GABA_AR. Type I bind primarily to the α₁ subunit (BZ receptor 1, ω₁), while Type II bind to α₂, α₃ and α₅ (BZ receptor 2, ω₂) (Pritchett et al., 1989b). Animal studies point to the specific contribution of individual receptor subtypes to the pharmacological spectrum of BZs. Specifically, the sedative and anterograde amnesic effects of BZs are attributed mainly to α₁-containing receptor subtypes, while the anxiolytic action of BZs is associated with the α₂- and α₃-containing receptors (Low et al., 2000, McKernan et al., 2000; Möhler et al., 2002). Mutation studies in mice showed that the anxiolytic effect of diazepam (a non-selective GABA_AR agonist) is lost in α₂ but not in α₃ mutated mice (Low et al., 2000). The current results thus raise the possibility of functional alterations of the sensitivity of these animals to the effects of BZs.

Indeed diazepam at the dose tested affected adult juvenile stressed animals, compared to adult juvenile stressed animals injected with vehicle, while having no effect on control animals. However, since previous studies and current results have shown that juvenile stressed animals have elevated anxiety levels compared to control animals (Avital and Richter-Levin, 2005; Tsoory and Richter-Levin, 2006), it was possible that the greater sensitivity to diazepam among juvenile stressed animals was because these animals were anxious and thus susceptible to the anxious effects of diazepam while in controls there was a floor effect due to initial low levels of anxiety.

We further tested for potential functional implications of the alterations in subunit expression by examining the sedative effects of the BZ brotizolam. Juvenile stressed rats tested following additional emotional challenges in adulthood were less sensitive to the sedative effects of brotizolam compared to controls.

Given the strong association between the α₂ subunit, anxiety and anxiety-related drugs, it is especially interesting that exposure to juvenile stress followed by challenges in adulthood resulted in elevated expression of the α₂ subunit in both the amygdala and hippocampus. In both hippocampus and amygdala, the elevation of α₂ subunit was, at least partially, due to the interaction between juvenile exposure to stress and adulthood emotional challenge. This suggests that the exposure to juvenile stress induces sensitivity to emotional challenges in adulthood, as previously suggested (Avital and Richter-Levin, 2005; Tsoory and Richter-Levin, 2006).

Collectively our results provide evidence for a region- and subunit-specific regulation of GABA_AR expression profile within the limbic system following exposure to juvenile stress and additional challenges in adulthood. The functional consequences might be a faulty functioning of the inhibitory GABAergic system that may lead to increased anxiety levels (Kralic et al., 2002).

Acknowledgements

This work was supported by a 2002 NARSAD Independent Investigator award to G.R-L., and by the EU's PROMEMORIA grant no. 512012 to G.R-L. We thank Dr Liza Barki-Harrington and Dr Michael Tsoory for valuable comments on this paper.

Statement of Interest

None.

References