Impact of Chronic Stress on the Spatial Learning and GR-PKAc-NF-κB Signaling in the Hippocampus and Cortex in Rats Following Cholinergic Depletion
- 176 Downloads
Studies have shown that the removal of the cholinergic innervation to the hippocampus induces dysfunction of the hypothalamic–pituitary–adrenocortical axis and decreases the number of glucocorticoid receptors (GRs). Subsequent studies have revealed that the loss of cholinergic input to the hippocampus reduces the expression of GRs and activates nuclear factor-kappa B (NF-κB) signaling through interactions with the cytoplasmic catalytic subunit of protein kinase A (PKAc). We examined the effects of chronic stress on cognitive status and GR-PKAc-NF-κB signaling in rats with a loss of cholinergic input to the hippocampus and cortex. Male Sprague-Dawley rats received 192 IgG-saporin injections to selectively eliminate cholinergic neurons in their basal forebrain. Two weeks later, rats were subjected to 1 h of restraint stress per day for 14 days. Rats subjected to both chronic stress and cholinergic depletion showed more severe memory impairments compared to those that received either treatment alone. The reduction in nuclear GR levels induced by cholinergic depletion was unaffected by chronic stress. The activation of NF-κB signaling in the hippocampus and the cerebral cortex induced by cholinergic depletion was augmented by chronic stress, resulting in the increased expression of pro-inflammatory markers, such as inducible nitric oxide synthase and cyclooxygenase-2. The activation of NF-κB induced by cholinergic depletion appears to be aggravated by chronic stress, and this might explain the increased susceptibility of patients with Alzheimer’s disease to stress since activation of NF-κB is associated with stress.
KeywordsCholinergic neuron Stress Glucocorticoid receptor Nuclear factor-kappa B Spatial memory
Dysregulation of the negative feedback to the hypothalamic–pituitary–adrenocortical (HPA) axis in response to an acute stressor is related to alterations in hippocampal glucocorticoid receptor (GR) signaling . Animals in which basal forebrain cholinergic neurons (BFCN) have been selectively eliminated show HPA axis dysregulation, and decreases in GR mRNA and protein levels in the hippocampus and prefrontal cortex [2, 3, 4]. A series of studies has shown that selective immunotoxic lesions of the BFCN using 192 IgG-saporin lead to impairments in hippocampal function, including attention and stress regulation, along with alterations in GR and nuclear factor-kappa B (NF-κB) signaling [2, 3, 5, 6].
Activated NF-κB, induced by inflammatory stimuli or stress, is translocated to the nucleus and induces the transcription of pro-inflammatory proteins, such as inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2) . In tandem, stress-induced glucocorticoid secretion increases inhibitory-kappa B (I-κB) expression levels, which prevents the translocation of NF-κB from the cytoplasm to the nucleus  and inhibits the DNA-binding activity of NF-κB . This inhibitory control system might be important for preventing the prolonged expression of pro-inflammatory cytokines in the brain after stressful events.
In addition, Doucas et al.  reported cytoplasmic cross-repression between NF-κB and GR. They showed that GR-mediated inhibition of NF-κB transactivation is dependent on the catalytic subunit of PKA (PKAc) [10, 11, 12]. Interestingly, an imbalance among GR, PKAc, and NF-κB was observed in the hippocampus after cholinergic depletion, including decreased nuclear GR protein levels, reduced PKAc activity, a reduction in the interaction between GR and PKAc, an increase in the interaction between NF-κB p65 and PKAc, and activation of NF-κB signaling [4, 13]. In addition, activation of NF-κB signaling is associated with the behavioral effects of acute stress and depression , and it is also a relevant transcription factor for learning and memory . NF-κB is activated in the brains of rats following acute immobilization stress  and in mononuclear cells of human subjects following psychological stress . Chronic stress also exacerbates the lipopolysaccharide-induced activation of NF-κB .
The loss of cholinergic neurons activates hippocampal NF-κB signaling, which influences memory formation and synaptic plasticity [13, 15]. However, previous studies have reported that selective lesions in the BFCN do not affect hippocampal-dependent learning and memory [19, 20]. Based on these reports, chronic stress is expected to aggravate the activation of NF-κB induced by cholinergic depletion, which may impair or induce a greater degree of impairment in learning and memory. Therefore, the present study examined the impact of chronic stress on the effects of cholinergic depletion on spatial learning and alterations in GR and NF-κB signaling. In addition, the activation of NF-κB signaling was substantiated by evaluating the protein expression of pro-inflammatory markers.
Materials and Methods
The selective lesioning of cholinergic neurons in the basal forebrain was performed surgically in 62 rats under isoflurane anesthesia. Thirty-eight rats were used for behavioral testing, and 24 rats were used for molecular signaling analyses. The lesions were induced by injections of 192 IgG-saporin, as described previously . The immunotoxin 192 IgG-saporin (Batch #ATS 64-124, 2.6 mg/ml; Advanced Targeting Systems, San Diego, CA) was dissolved in sterile 0.01 M phosphate buffered saline (Dulbecco’s PBS) at a concentration of 0.25 μg/μl. The general surgical procedure was identical for all groups (sham-control, sham-stress, lesion-control, and lesion-stress). Rats were anesthetized using isoflurane and placed in a stereotaxic frame (Kopf Instruments, Tujunga, CA) fitted with an isoflurane gas anesthesia system and an incisor bar set 3.3 mm below the ear bars. The scalp was then incised and retracted, and holes were drilled at appropriate locations in the skull using a dental drill. Injections of Dulbecco’s PBS (sham surgery, n = 31) or 192 IgG-saporin (lesion surgery, n = 31) were made using a 28-gauge needle attached by a length of plastic tubing to a 2-μl Hamilton microsyringe (Sigma-Aldrich, St. Louis, MO).
Medial Septum/Vertical Limb of the Diagonal Band (MS/VDB) Surgeries
Two holes were drilled in the skull at stereotaxic coordinates anteroposterior (AP) +0.5 mm and mediolateral (ML) ±0.6 mm to bregma according to the Paxinos and Watson (2004) atlas. Injections were performed at two depths for each site: dorsoventral (DV) −7.8 mm and −6.2 mm from the skull surface at bregma. Cannulas were left in place for 30 s before the infusion began. Saline or immunotoxin were slowly infused (0.05 μl/min) into the injection site to yield a total volume of 0.3 μl in the sites at DV −7.8 mm and a total volume of 0.2 μl in the sites at DV −6.2 mm. The needle was left in place for 5 min after each 0.3-μl injection and for 3 min after each 0.2-μl injection to limit the diffusion up the needle track.
Nucleus Basalis Magnocellularis/Substantia Innominate (NBM/SI) Surgeries
Four holes were drilled into the skull at stereotaxic coordinates AP −0.75 mm, ML ±2.3 mm (medial sites), or ML ±3.3 mm (lateral sites) from bregma. Injections were made at DV −7.8 mm (medial sites) or DV −8.1 (lateral sites) from the skull surface measured from bregma. Saline or immunotoxin were slowly infused (0.1 μl/min) into the injection site. A total volume of 0.2 μl was infused into each site. Cannulas were left in place for 3 min after each injection.
Chronic Stress Paradigm and Body Weight
The stress procedure began 14 days after the surgery. Half of the animals from the sham and lesion groups were restrained in a Plexiglas tube, and the others were left undisturbed in their home cages. The animals in the stress groups were chronically subjected to the stressor for 1 h once a day for 2 weeks. Therefore, there were four groups: sham-control (n = 15), sham-stress (n = 16), lesion-control (n = 15), and lesion-stress (n = 16). The body weights in the stress groups were measured on the first day and the last day of the restraint treatment. The body weight in the control groups were measured at the corresponding time points.
Trunk blood was collected during the decapitation procedure, conducted approximately 24 h after the last stress session. Blood serum was separated by centrifugation (5000 rpm, 20 min at 4 °C) and stored at −70 °C. Serum corticosterone concentrations were measured using a corticosterone ELISA kit (IBL, Minneapolis, MN).
Hidden Platform Water Maze Task
Spatial memory was measured using the Morris water maze task . One day after the final restraint stress, rats were subjected to the spatial version of the water maze task. In a standardized procedure, using distal cues in the maze environment, the rats were trained to find the position of a camouflaged escape platform. The water maze consisted of a large, circular tank (diameter 183 cm; wall height 58 cm) filled with water (27 °C) made opaque by the addition of powdered milk (0.9 kg). A retractable white escape platform (height, 34.5 cm) was placed 2 cm beneath the water surface near the center of one of the four quadrants in the maze. White curtains completely surrounded the maze, and large geometric designs were attached to the curtains to provide spatial cues. Data were analyzed using an HVS Image Analyzing VP-116 video tracking system and an IBM PC with software developed by HVS Imaging (Hampton, UK).
The place training protocol has been described previously in detail . Briefly, the rats were trained for two to three trials per day for eight consecutive days with a 60-s intertrial interval. In each training trial, the rat faced the inside wall of the tank at one of four pseudorandomly varied starting points. The location of the platform remained constant, and in each training trial, the rats swam for 90 s or until they found the platform. If the rats did not find the platform within 90 s, then the trainer guided them to the platform. At the end of each trial, the rats were forced to stay on the platform for 30 s. During each trial, the distance of the rats from the escape platform was sampled 10 times per second, and these values were averaged in 1-s bins. The cumulative search error was then calculated as the summed 1-s averages of the proximity measures, corrected for the particular start location in each trial.
A probe trial was conducted after the 10th and 20th training trials to assess the development of spatial bias in the maze. During these probe trials, the rats swam with the platform retracted to the bottom of the pool for 30 s. Afterwards, the platform was raised to its normal position and made available for escape. The probe tests were used to assess the search strategy the rats used to navigate the maze. The proximity measure was obtained by sampling the position of the animal in the maze (10 times per second) to recode its distance from the escape platform in 1-s averages.
Upon the completion of the spatial learning assessments, one session with four trials of cue training was performed. Rats were trained to escape to a visible black platform 2 cm above the surface of the water. The location of the platform was varied from trial to trial to assess the animals’ sensorimotor and motivational functioning, independent of their spatial learning abilities. Each rat was allowed 30 s to reach the platform and was allowed to remain there briefly before a 30-s intertrial interval.
One week after the water maze task, the rats used for immunohistochemistry were euthanized using an overdose of ketamine HCl (30 mg/mg) and xylazine (2.5 mg/kg) and then intracardially perfused with 4% paraformaldehyde (PFA) in 0.01 M PBS (pH 7.4). After perfusion, the brains were dissected out and post-fixed with PFA (48 h) and cryoprotected in PBS containing 30% sucrose until they sank in the solution. The fixed brain tissue was frozen in powdered dry ice and stored at −80 °C. Brain tissues were cut into coronal sections (coronal plane = 40 μm) using a microtome. All sections were preserved in cryoprotectant (30% ethylene glycol, 25% glycerol, 25 mM phosphate buffer) and stored at −20 °C.
To confirm the lesions, basal forebrain sections were processed with either a polyclonal antibody for choline acetyltransferase (ChAT, 1:100; Chemicon, Temecula, CA) or a monoclonal antibody for parvalbumin (Parv, 1:500; Sigma-Aldrich) to assess immunoreactivity. Tissue sections were washed three times with PBS (pH 7.4) containing 0.3% Triton X-100 (PBS-T). Endogenous peroxidases were blocked by incubating the sections for 30 min in 3% H2O2/10% MeOH in PBS. The sections were incubated for 1 h at room temperature in PBS-T containing 10% fetal horse serum (PBS-T-S, normal serum; Vector Laboratories, Burlingame, CA) and then incubated overnight at 4 °C with the appropriate primary antibody. After three washes, the sections were incubated for 1 h in the appropriate biotinylated secondary antibody (Vector Laboratories) in PBS-T at room temperature. The sections were rinsed three times and incubated for 1 h in ExtrAvidin® peroxidase conjugate (1:1000; Sigma-Aldrich). Finally, the sections were visualized using the Vector® SG substrate kit (Vector Laboratories). The sections were mounted onto resin-coated slides, dehydrated through ascending concentrations of ethanol, defatted in xylene, and cover-slipped with Permount™ (Fisher, Pittsburgh, PA).
Preparation of Cytosolic and Nuclear Extracts
Approximately 24 h after the last stress, rats were sacrificed by decapitation. The brains were dissected out, and a coronal cut 2 mm posterior to the optic chiasm was made to partition the anterior portion of the brain, containing the BFCN, from the posterior part, containing the hippocampus and cortex. The anterior portions of the brains were post-fixed in 4% PFA for 5–6 days, cryoprotected in PBS containing 30% sucrose (2–3 days), frozen on powdered dry ice, and sectioned (coronal plane = 40 μm) on a microtome. These sections were then processed for ChAT and Parv immunostaining to confirm the presence of a selective cholinergic lesion. The posterior portion of the brain was microdissected, and the hippocampus and cerebral cortex were frozen.
Brain tissue from individual animals was homogenized with a tissue grinder (Radnoti, Monrovia, CA). Frozen tissues were homogenized in 0.3 ml buffer/100 mg tissue. The homogenization buffer consisted of 20 mM Tris-HCl (pH 7.5) buffer containing 10% glycerol, 50 mM NaCl, 1 mM ethylenediaminetetraacetic acid (EDTA), 1 mM ethylene glycol tetraacetic acid (EGTA), 2 mM dithiothreitol, 1 mM phenylmethane sulfonyl fluoride (PMSF), 1 mM Na3VO4, 20 mM NaF, and protease inhibitors. The homogenate was centrifuged for 10 min at 2000×g at 4 °C, and the resulting supernatant was used as the cytosolic fraction. The cytosolic preparation was centrifuged for 1 h at 105,000×g, and the supernatant was the cytoplasmic fraction. The pellet was washed twice in 0.5 ml homogenization buffer followed by additional low-speed centrifugation (10 min, 2000×g). The washed pellet was resuspended (1:1: mass/volume) in homogenization buffer containing 0.5 M KCl. The tissue was then incubated for 1 h in an ice bath with frequent vortexing and centrifuged for 10 min at 8000×g at 4 °C. The final supernatant was used as the nuclear extract .
Western Blot Analysis
An aliquot of the cytosolic or nuclear fraction was used for the determination of total protein concentration using the Bradford method. The samples were denatured in 1 M Tris-HCl, 50% glycerol, 10% SDS, 500 mM 2-mercaptoethanol, and 0.1% bromophenol blue and boiled at 100 °C for 5 min. For Western blot analysis, the proteins were separated by electrophoresis on 8–10% SDS-polyacrylamide gels. Equal amounts of protein were loaded per well. The proteins were electroblotted onto polyvinylidene fluoride membranes. The membranes were blocked with 5% skimmed milk in Tris-buffered saline with Tween-20 (TBS-T) for 1 h at room temperature. Incubation with the primary antibody was performed overnight at 4 °C in 5% skimmed milk in TBS-T. The anti-GR (1:2000), anti-glutamate decarboxylase-67 (anti-GAD-67; 1:500), anti-NF-κB p65 (1:2000), anti-I-κB (1:2000), anti-iNOS (1:2000), and anti-COX-2 (1:1000) antibodies were purchased from Santa Cruz Biotechnology (Dallas, TX). Anti-phospho-NF-κB p65 (Ser276; 1:2000), polyclonal anti-PKAc (1:2000), and polyclonal anti-phospho-PKAc (Thr197; 1:1000) antibodies were from Cell Signaling (Danvers, MA), and the anti-ChAT antibody (1:1000) from Chemicon. After three washes with TBS-T, the membranes were incubated for 1 h in horseradish-peroxidase conjugated secondary antibody in 5% skimmed milk in TBS-T. The membrane was washed three times with TBS-T, and the peroxidase was visualized on ECL hyperfilm (GE Healthcare, Chicago, IL) using ECL Plus chemiluminescent reagents (GE Healthcare). The levels of each protein were normalized to β-actin (1:1000; Sigma-Aldrich). The immunoblotting results were adjusted to the total amount of β-actin and quantified using the Image Gauge program (Fujifilm, Japan).
Co-Immunoprecipitation (Co-IP) and IP-Western Blot Analysis
Hippocampal and cortical tissues were homogenized in extraction buffer (10 mM Tris-HCl, pH 7.5, 5 mM EDTA, 150 mM NaCl, 1 mM sodium orthovanadate, 10 mM NaF, 1% Triton X-100, 10% glycerol, 1 mM PMSF, 1 mM MgCl2) supplemented with protease inhibitors (Calbiochem, San Diego, CA). The extracts were then clarified by centrifugation for 10 min at 8000×g, and the protein concentrations of the supernatants were determined using a BCA assay (Sigma-Aldrich). Next, samples containing 1 mg of total protein were taken for subsequent immunoprecipitation with specific antibodies against the PKA catalytic domain overnight, followed by an additional 3-h incubation at 4 °C with protein A-Sepharose beads. The immunoprecipitates were extensively washed using the extraction buffer and then subjected to SDS-PAGE and Western blot analysis using the procedures and concentrations described above. Membranes were also incubated with the polyclonal anti-phospho-PKAc primary antibody (Cell Signaling).
Search errors during behavioral training were analyzed using a two-factor repeated measures (group × trial block) analysis of variance (ANOVA) to evaluate acquisition in the hidden platform version of the water maze task, and performance during the probe trials was analyzed using a one-factor ANOVA. A two-factor repeated measures (group × changes in body weight) ANOVA was used to evaluate changes in body weight during chronic stress. A one-factor ANOVA was used to analyze levels of corticosterone and proteins measured using Western blot and IP-Western blot analyses. Post hoc analyses (Fisher’s least significant difference) were then used to evaluate the differences between the groups if necessary. Differences with a p value less than 0.05 were considered significant. All data were expressed as mean ± standard error of the mean (SEM).
Changes in Body Weight and Corticosterone Levels
A two-factor repeated measures ANOVA revealed that the body weights of rats increased significantly (F(1, 34) = 60.18, p < 0.001), and the effects of group were significant (F(3, 34) = 4.08, p = 0.01). The interaction between group and changes in body weight was also significant (F(3, 34) = 12.18, p < 0.001). The post hoc analyses revealed that the body weights of animals in the lesion-stress group increased significantly less than that in the other groups during stress (Fig. 1b). There were no significant effects of group found in the results from the corticosterone assay (Fig. 1c).
Confirmation of Cholinergic Lesions
Impairment in Spatial Memory Was Observed in Rats Following Cholinergic Depletion and Subsequent Chronic Stress
The relevance of the selective removal of the BFCN to memory impairments is controversial. A number of studies have reported that cholinergic depletion in rats induces marginal impairments in spatial memory , whereas in other studies, there were no effects of cholinergic lesions on spatial learning [19, 25]. The present study was therefore conducted to determine whether impairments in spatial memory in rats after cholinergic depletion would be enhanced by chronic stress or if marginal impairments in spatial memory in rats after cholinergic depletion would be aggravated by chronic stress. There were four groups: sham-control (n = 9), sham-stress (n = 10), lesion-control (n = 9), and lesion-stress (n = 10).
The average distance traveled to reach the platform during the 30-s probe trial was analyzed as a measure of spatial memory retention (Fig. 3b). This parameter is the mean distance traveled by each rat to the platform during the probe trial. One-factor ANOVAs on data from the first and the second probe trials revealed significant group effects (first trial—F(3, 34) = 4.29, p < 0.01; second trial—F(3, 34) = 7.30, p < 0.001). According to the post hoc analyses, the sham-control and the sham-stress rats showed better performance in the first probe trial than the lesion-control and the lesion-stress rats. In the second probe trial, the lesion-stress rats showed the worst performance in terms of spatial memory retention. The four trials with the platform visible were conducted 2 days after the spatial training was terminated. All of the rats in the four groups found the visible platform quickly. A one-factor ANOVA revealed no group effect, indicating that visual function and motor function were intact in all four groups.
The Reduction in Hippocampal GR Levels Following Cholinergic Depletion Was Not Altered by Chronic Stress
GR signaling in the hippocampus and frontal cortex is associated with the regulation of the negative feedback on the HPA axis after an acute stressor [1, 26, 27]. It has been reported that selective lesioning of the BFCN results in HPA-axis dysregulation and decreases in GR mRNA and protein levels in the hippocampus [2, 3, 4]. GR levels were measured using Western blot analysis of the cytosol and the nuclear fractions to assess the effects of chronic stress on the reduction in hippocampal GR levels, and changes in cortical GR levels induced by cholinergic depletion (n = 6 per group).
Effects of Chronic Stress on the Alterations in Hippocampal NF-κB Signaling Induced by Cholinergic Depletion
Cholinergic depletion alters GR-PKA-NF-κB p65 signaling in the hippocampus, leading to NF-κB activation [4, 11, 13]. Moreover, NF-κB signaling is activated in the rat brain by exposure to acute immobilization stress . Therefore, the present study examined the effects of chronic stress on the alterations in hippocampal NF-κB signaling induced by cholinergic depletion (n = 6 per group).
NF-κB p65 and I-κB
NF-κB p65 Phosphorylation
NF-κB p65 is phosphorylated at Ser276 by PKA . Phosphorylation of NF-κB p65 is increased in the hippocampus following cholinergic depletion via a reduced interaction between GR and PKAc [4, 11], indicating activation of NF-κB signaling. Based on previous reports, the present experiment examined the effects of chronic stress on the increase in phosphorylated levels of cytosolic NF-κB p65 induced by cholinergic depletion in the hippocampus. A one-factor ANOVA revealed a significant effect of group on levels of phosphorylated cytosolic NF-κB p65 (F(3, 20) = 13.38, p < 0.001). According to the post hoc analyses, the levels of phosphorylated NF-κB p65 in the sham-stress rats and the lesion-control rats were significantly increased compared with those of the sham-control rats. The lesion-stress rats showed the highest levels of NF-κB p65 phosphorylation (Fig. 5e, f).
PKAc and Its Phosphorylation at Thr197
A series of studies previously showed that cholinergic depletion decreases the activity of PKAc and increases the interaction between PKAc and NF-κB p65 in the hippocampus . The present study measured the phosphorylation status of PKA. No group differences in PKAc levels were observed in the cytosolic or nuclear fractions (Fig. 5g–i). However, a one-factor ANOVA showed a significant group effect on PKAc phosphorylation in the hippocampus (F(3, 20) = 20.65, p < 0.001). Post hoc analyses revealed that the reduction in PKAc phosphorylation at Thr197 was sequentially greater from the sham-stress rats to the lesion-control rats to the lesion-stress rats compared to the sham-control rats (Fig. 5j, k).
COX-2 and iNOS
NF-κB p65 is one of the major transcription factors involved in the inflammatory response . Increased phosphorylation of NF-κB p65 should augment the transcriptional activity of NF-κB p65. Translocation of NF-κB into the nucleus following the degradation of I-κB induces the transcription of inflammatory enzymes, such as COX-2 and iNOS . Therefore, in the present experiment, pro-inflammatory markers in the hippocampus were measured using Western blot analyses. A one-factor ANOVA showed a significant effect of group on levels of COX-2 and iNOS (F(3, 20) > 14.58, p < 0.001). According to the post hoc analyses, when compared to the sham-control rats, the sham-stress rats showed significantly higher levels of COX-2. COX-2 levels were also increased in the lesion-control rats, but to a lesser extent than in the sham-stress rats. The lesion-stress rats showed the highest levels of COX-2 expression (Fig. 5l, n). The pattern of iNOS expression across the treatments was similar to that of COX-2 (Fig. 5l, m).
Effects of Chronic Stress on the Alterations in Cortical GR and NF-κB Signaling Induced by Cholinergic Depletion
In the present study, a coronal cut was made 2 mm posterior to the optic chiasm to partition the brain. The anterior portion containing the BFCN was used to examine the absence of cholinergic neurons, and the posterior portion was used for Western blot analysis (n = 6 per group). The cortex did not include the prefrontal cortex. The present study examined GR and NF-κB signaling in the cortex following cholinergic depletion. To our knowledge, this is the first time that this has been examined.
Cortical GR Levels
The statistical analyses showed that there were no differences in the GR levels in the cortical cytosolic and nuclear fractions between the groups (Fig. 4c, d).
NF-κB p65 and I-κB
NF-κB p65 Phosphorylation
A one-factor ANOVA revealed that there was significant effect of group on the cortical levels of cytosolic NF-κB p65 phosphorylation (F(3, 20) = 15.06, p < 0.001). According to the post hoc analyses, sham-stress and lesion-control rats showed significantly increased levels of NF-κB p65 phosphorylation relative to the levels seen in sham-control rats; however, there were no differences between the sham-stress and lesion-control groups. The lesion-stress rats exhibited higher levels of NF-κB p65 phosphorylation than the other groups (Fig. 6e, f).
PKAc and Its Phosphorylation at Thr197
A one-factor ANOVA revealed that there were no effects of group on cortical cytosolic and nuclear PKAc levels (Fig. 5g–i). However, a one-factor ANOVA showed a significant effect of group on the levels of PKAc phosphorylation in the cortex (F(3, 20) = 16.91, p < 0.001). The post hoc analyses revealed that the sham-stress rats and the lesion-control rats showed significantly lower levels of PKAc phosphorylation at Thr197 than the sham-control rats, and the lesion-stress rats showed the lowest degree of phosphorylation (Fig. 6j, k).
COX-2 and iNOS
Similar to the experiment in the hippocampus, we measured cortical COX-2 and iNOS levels to confirm the activation of NF-κB signaling. A one-factor ANOVA showed that there was a significant effect of group on the levels of COX-2 (F(3, 20) = 9.91, p < 0.001). According to the post hoc analyses, COX-2 expression in the sham-stress rats was not significantly increased compared to the levels seen in the sham-control rats; however, the levels in the lesion-control rats were significantly increased compared to the sham-control rats. Lesion-stress rats showed higher levels of COX-2 than the other groups (Fig. 6l, n). A one-factor ANOVA showed that there was a significant effect of group on iNOS levels (F(3, 20) = 3.90, p < 0.05). The post hoc analyses showed that iNOS levels in the lesion-stressed rats only were significantly higher than that in the sham-control rats (Fig. 6l, m).
The present study examined the effects of chronic stress on the alterations in spatial learning and its underlying molecular signaling induced by cholinergic depletion. Rats that experienced only cholinergic depletion were mildly impaired in acquiring a spatial water maze task, and in the early session probe trial, which is consistent with the results of a previous study . Rats exposed only to chronic stress for 2 weeks showed no differences in spatial learning compared to sham-control rats. These results are consistent with earlier reports that mild or moderate chronic stress prior to learning does not affect spatial memory, as assessed using the water maze task [29, 30]. Most importantly, rats that experienced both cholinergic depletion and chronic stress exhibited severe impairments in the acquisition and retention of spatial memory tasks. Craig et al.  also reported cognitive impairments in rats that experienced cholinergic lesions in the medial septum and chronic stress; however, there was a discrepancy between the procedures and results in the present study and theirs. The cholinergic lesion in the study by Craig et al.  was limited to cholinergic neurons of the medial septum, and a variable restraint stress procedure was used. In addition, they used a training protocol to measure spatial working memory. That is, rats first received a hidden platform position and were subsequently trained with a new hidden platform location during subsequent sessions. Unlike our results, rats with cholinergic lesions of the medial septum that were exposed to chronic stress were not impaired in the first spatial reference memory task. However, these rats were impaired in the second and third training tasks conducted after the platform position was changed, which suggests impairment in spatial working memory.
Impairments in hippocampal-dependent memory and alterations in GR signaling have been reported in animals that are exposed to relatively long or severe chronic stress, indicating a connection between the GR and cognitive function . As the GR is responsible for mediating the adaptive stress response , altered GR signaling is associated with post-stress glucocorticoid hypersecretion . Interestingly, animals with cholinergic depletion display a post-stress prolonged decline in glucocorticoids in response to a single stress exposure, and altered GR signaling in hippocampus [2, 4]. In addition, exposure to glucocorticoids exacerbates ethycholine aziridinium-induced cholinergic lesions in the hippocampus . These reports suggest a link between stress, GR signaling, and cholinergic neurons . It is therefore expected that, even if relatively short term, chronic stress would not cause cognitive impairments by itself; however, it may aggravate HPA-axis dysfunction in rats following cholinergic depletion and lead to further changes in the altered GR signaling induced by cholinergic depletion. However, in the present study and the study by Craig et al. , no differences in corticosterone levels were observed in rats that were exposed to chronic stress and those that experienced both cholinergic depletion and chronic stress. Above all, in the present experiment, a further reduction in hippocampal nuclear GR levels was not observed in rats following cholinergic depletion and chronic stress in comparison to the rats that only experienced cholinergic depletion. It is possible that GR-associated signaling, or other pathways (e.g., NF-κB signaling), may be involved in the effects of chronic stress on the cognitive deficits induced by cholinergic depletion.
Many studies have extensively investigated the effects of cholinergic depletion on hippocampal GR signaling. From the initial report that showed that GR mRNA was decreased in the hippocampus and frontal cortex following cholinergic depletion [2, 5], a subsequent study showed that nuclear levels of hippocampal GR protein were decreased following cholinergic depletion . Based on reports that PKAc mediates the cross-repression between NF-κB and the GR in the cytosol , and that PKA phosphorylates the GR and NF-κB , we examined the status of PKAc and NF-κB signaling in the hippocampus following cholinergic depletion. A reduction in PKA activity and PKAc phosphorylation at Thr197 was observed in the hippocampus following cholinergic depletion. In addition, while the interaction between PKAc and GR was decreased, the interaction between PKAc and NF-κB increased, indicating activation of NF-κB. Activation of NF-κB was supported by further evidence such as the degradation of I-κB and the phosphorylation of NF-κB p65 at Ser276 [4, 13].
NF-κB signaling is associated with memory formation and synaptic plasticity , and plays a role in the behavioral effects of stress and depression . Hence, we examined the NF-κB signaling underlying the effects of chronic stress on the alterations in signaling induced by cholinergic depletion, focusing on PKAc and NF-κB. It is well reported that PKAc and NF-κB form the NF-κB-I-κB-PKAc complex with I-κB, an inhibitor protein for NF-κB. NF-κB is activated by the degradation of I-κB , and PKAc subsequently phosphorylates NF-κB p65. The phosphorylation of NF-κB p65 by PKA induces the transcriptional activity of NF-κB , leading to the expression of NF-κB target genes . Hence, based on the above reports and our previous studies, we examined GR and NF-κB signaling in the hippocampus and cortex following cholinergic depletion and chronic stress.
At first, the reduction in hippocampal nuclear GR levels induced by cholinergic depletion was not affected by chronic stress. However, the activation of NF-κB signaling induced by cholinergic depletion was enhanced by chronic stress, which was also substantiated by reductions in the levels of PKA phosphorylation and increased levels of the pro-inflammatory markers, such as iNOS and COX-2. NF-κB p65 is a major transcription factor involved in the inflammatory response . In addition, as a further confirmatory evidence, we measured nuclear NF-κB p65 levels in the hippocampus. But there were no group differences; even the stress rats exhibited slightly increased levels in NF-κB p65 in the nuclear extract of hippocampus regardless of cholinergic lesion. Considering the complexity of NF-κB signaling in inflammation  and the animal study indicating an association of stress-related behavior with NF-κB subunit p50 , further study should be conducted to intensively examine the status of NF-κB in the nucleus by measuring several nuclear markers including NF-κB subunit p50.
However, no changes in cytosolic or nuclear GR levels were found in the cortex in response to cholinergic depletion or chronic stress, or a combination of the two. Unlike hippocampal NF-κB signaling, only nuclear levels of cortical NF-κB p65 were increased in rats that experienced cholinergic depletion and chronic stress compared with control rats or rats that experienced either cholinergic depletion or chronic stress alone. Interestingly, the levels of NF-κB p65 phosphorylation were higher in the rats that experienced either cholinergic depletion or chronic stress alone than the control rats. A further increase was observed in the rats that experience both cholinergic depletion and chronic stress together. Activation of NF-κB in the cortex was also confirmed by a reduction in PKA phosphorylation, and increases in iNOS and COX-2. Thus, notable differences were found between the hippocampus and cortex in terms of the alterations in GR and NF-κB signaling induced by cholinergic depletion and chronic stress.
The first difference between the cortex and hippocampus in response to cholinergic depletion is the lack of involvement of GR signaling in the cortex. This difference might be primarily due to the distribution of GRs in the brain. High levels of GRs are present throughout the hippocampus, whereas the cortex has low levels of GRs. The second difference is an involvement of a different intracellular NF-κB signaling pathway in the cortex even though activation of NF-κB occurs in both the hippocampus and cortex. That is, based on the disparate findings regarding NF-κB signaling in the cortex and hippocampus, a different intracellular NF-κB signaling pathway might be found in the cortex. Since most studies on such signaling pathways have focused on the hippocampus, further research on NF-κB signaling in the cortex might elucidate these differences.
One common signal changed in both cortex and hippocampus in response to cholinergic depletion and chronic stress, i.e., the phosphorylation of PKAc was reduced. We previously reported that cholinergic depletion reduced PKAc activity without changing total PKA levels . In the present study, PKAc levels were not altered in response to cholinergic depletion or chronic stress or both in combination. However, the level of PKA phosphorylation at Thr197 was reduced in response to either cholinergic lesioning or chronic stress alone. Interestingly, further reductions in PKA phosphorylation were observed in the hippocampus and cortex when animals experienced both cholinergic depletion and chronic stress in combination. Hence, considering that PKAc signaling is critical in the normal regulation of stress and stress-related psychiatric disorders , it would be worthwhile determining whether PKA phosphorylation is a key event in mediating the combined effects of cholinergic depletion and chronic stress.
Numerous studies have revealed that the BFCN plays a key role in the regulation of stress and cognition in aging and Alzheimer’s disease (AD) [1, 41]. BFCN projections to cortical sites and the limbic system, including the hippocampus, are necessary for stress regulation and cognition. In particular, the hippocampus is involved in termination of the HPA-axis signaling in response to stress. Altered GR signaling is related to anxiety, depressive behaviors, and impairments in learning and memory [42, 43]. In addition, cholinergic deficits are associated with some neuropsychiatric changes in AD . The selective removal of the BFCN produces impairments in the negative feedback regulation of the stress response and alterations in GR-PKAc-NF-κB signaling [2, 4]. The present study has reported that impairments in spatial memory evidently appeared after chronic stress and that the alterations in PKAc-NF-κB signaling induced by cholinergic depletion are aggravated by chronic stress, as evidenced by lower cytosolic NF-κ and I-κB in the hippocampus of rats with cholinergic lesion and chronic stress, and this might explain the susceptibility of patients with AD to neuropsychiatric disorders such as depression following stress exposure.
This study was funded by the National Research Foundation of Korea grants 2011-0015725 and 2015M3C7A1031395 to J.S.H.
S.Y.L. and J.S.H. designed the research; S.Y.L. and W.H.C. performed the research; S.Y.L. and J.S.H. analyzed the data; S.Y.L. and J.S.H. wrote the paper.
Compliance with Ethical Standards
Conflict of Interest
The authors declare that they have no competing interests.
- 4.Lim CS, Kim YJ, Hwang YK, Banuelos C, Bizon JL, Han JS (2012) Decreased interactions in protein kinase A-glucocorticoid receptor signaling in the hippocampus after selective removal of the basal forebrain cholinergic input. Hippocampus 22(3):455–465. doi: 10.1002/hipo.20912 CrossRefPubMedGoogle Scholar
- 18.Munhoz CD, Lepsch LB, Kawamoto EM, Malta MB, Lima Lde S, Avellar MC, Sapolsky RM, Scavone C (2006) Chronic unpredictable stress exacerbates lipopolysaccharide-induced activation of nuclear factor-kappaB in the frontal cortex and hippocampus via glucocorticoid secretion. J Neurosci 26(14):3813–3820. doi: 10.1523/JNEUROSCI.4398-05.2006 CrossRefPubMedGoogle Scholar
- 23.Adzic M, Djordjevic J, Djordjevic A, Niciforovic A, Demonacos C, Radojcic M, Krstic-Demonacos M (2009) Acute or chronic stress induce cell compartment-specific phosphorylation of glucocorticoid receptor and alter its transcriptional activity in Wistar rat brain. J Endocrinol 202(1):87–97. doi: 10.1677/JOE-08-0509 CrossRefPubMedPubMedCentralGoogle Scholar
- 30.McDonald RJ, Craig LA, Hong NS (2008) Enhanced cell death in hippocampus and emergence of cognitive impairments following a localized mini-stroke in hippocampus if preceded by a previous episode of acute stress. Eur J Neurosci 27(8):2197–2209. doi: 10.1111/j.1460-9568.2008.06151.x CrossRefPubMedGoogle Scholar
- 42.Ridder S, Chourbaji S, Hellweg R, Urani A, Zacher C, Schmid W, Zink M, Hortnagl H et al (2005) Mice with genetically altered glucocorticoid receptor expression show altered sensitivity for stress-induced depressive reactions. J Neurosci 25(26):6243–6250. doi: 10.1523/JNEUROSCI.0736-05.2005 CrossRefPubMedGoogle Scholar