INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

CORTICOTROPIN-RELEASING factor (CRF) is a 41-amino acid peptide recognized as a major regulator of pituitary ACTH. This neuropeptide is also known to coordinate a wide variety of physiological responses caused by sustained stresses (11, 20, 31). Indeed, intracerebroventricular injection of CRF induces physiological, behavioral, and autonomic changes similar to those observed in severely stressed animals (11, 32). CRF is one of the most widely distributed neuropeptides throughout the brain, but the neuroendocrine CRF secreted into the hypophysial portal system originates mainly from the parvocellular division of the paraventricular nucleus (PVN) of the hypothalamus (41).

To exert its action, CRF binds to high-affinity membrane receptors (9) cloned by different groups (6, 24, 35) and is referred to as CRF₁ and CRF₂ receptors. These receptors belong to the recently described family of gut-brain neuropeptide receptors that also include receptors for calcitonin, vasoactive intestinal peptide, parathyroid hormone, secretin, pituitary adenylate cyclase-activating peptide, and glucagon (42), which are potentially activated by CRF and sauvagine (6). All of these receptors possess seven putative transmembrane domains and are positively coupled to adenylate cyclase (38). The CRF₁ receptor is a 415-amino acid protein and shares ~30% identity with all other members of the neuropeptide receptor family. CRF₁ receptor mRNA is predominantly expressed in the olfactory bulb, cerebral cortex, red nucleus, pontine gray, cerebellum, and pituitary, whereas a low signal is generally detected in the endocrine hypothalamus of unchallenged rodents (39). Two forms of the CRF type 2 receptor have been characterized: CRF₂ and CRF₂. The CRF₂ receptor is a 411-amino acid protein that shares ~71% identity with the CRF₁ receptor (23). The CRF₂ receptor is a 431-amino acid protein that differs from CRF₂ in that the first 34 amino acids in the NH₂-terminal extracellular domain are replaced by 54 different amino acids (23). CRF₂ is the predominant neuronal CRF₂ receptor subtype within the brain, whereas the CRF₂ form is localized in nonneuronal elements, the choroid plexus and cerebral blood vessels (23). CRF₂ receptor mRNA is generally confined to limbic structures, particularly within the lateral septal nuclei, specific amygdaloid nuclei, and ventromedial hypothalamic nuclei (22, 24, 34). In the periphery, CRF₂ receptor mRNA is expressed in both cardiac and skeletal muscle with lower levels observed in both lung and intestine (23, 34). This heterogeneous distribution of CRF₁ and CRF₂ receptor mRNAs suggests distinctive functional roles for each receptor in CRF-related brain and systemic systems.

We and others have recently observed a robust transcriptional activation of the gene encoding the CRF₁ (but not the type 2) receptor in the rat endocrine hypothalamus after different types of neurogenic and systemic stressors. Indeed, an acute immobilization session caused a strong and transient expression of the mRNA encoding CRF₁ receptor in the parvocellular neurosecretory zone of the rat PVN (39). These data together with the fact that exogenous CRF can stimulate transcription of neuroendocrine CRF (26) strongly support the concept that endogenous CRF participates as positive autoregulatory feedback onto neuroendocrine CRF neurons. Whether such effect is maintained during repeated sessions of homotypic neurogenic challenges has yet to be investigated. The general aim of the present study was therefore to compare the influence of an acute versus a chronic immobilization stress on 1) the mRNA encoding the immediate early gene (IEG) c-fos used here as an index of the postsynaptic neuronal activation (29); 2) the activity of neuroendocrine CRF cells and those expressing its type 1 receptor; and 3) colonic transit using fecal pellet output as a simple marker of stress-induced colonic motility disturbances. The stimulation of colonic motility is a classical visceral response to stress in animals and humans (5, 45), and consistent experimental evidence indicates that CRF action in the PVN is part of the underlying mechanisms through which stress alters colonic motor function (5, 45, 46). Indeed, the stimulation of colonic transit and fecal pellet output induced by CRF injected into the lateral ventricle or PVN is well correlated with the increase in intraluminal pressure and phasic contractility or frequency of spike burst recorded in the cecum and proximal colon of conscious fasted or fed rats (28).

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Animals. Experiments were performed on 40 adult male Sprague-Dawley rats weighing 230-260 g. Animals were individually housed under conditions of controlled illumination (14:10-h light-dark cycle; lights on at 6 AM and off at 8 PM), humidity, and temperature (20-22°C) for at least 7 days before the experiments. Animals had free access to a standard rat diet and water until the stress session and all protocols were approved by Laval University's Animal Welfare Committee.

Immobilization stress. Immobilization stress sessions were performed as previously described (39). Rats were submitted to a 90-min immobilization stress either acutely (n = 20) or chronically during 11 consecutive days (n = 20). Because of the diurnal variations in corticosteroid secretion and c-fos expression in specific brain areas (19), all stress sessions started at the same time during the day (from 2:00 to 3:30 PM) in animals deprived of food and water for the duration of the challenge. Immobilization was performed using adjustable individual restraining cages (Centrap cages, Fisher Scientific 01-282-10). Cages were adjusted very tightly so that animals were unable to move once placed in the immobilization apparatus, thus explaining the use of the term immobilization instead of restrain stress. At the end of the stress session, rats were either killed or placed in their home cages until killed. Control animals were freely moving animals never exposed to the stress.

In situ hybridization histochemistry. Rats were killed before (controls; time 0: afternoon at 2:00 PM) or immediately, 1.5, 3, 6, or 12 h after the end of the acute stress session or after the last chronic stress session. Animals were deeply anesthetized with an intraperitoneal injection of a mixture of ketamine hydrochloride (91 mg/kg) and xylazine (9 mg/kg) and then rapidly perfused transcardially with saline followed by 4% paraformaldehyde in 0.1 M borax buffer (pH 9.5 at 4°C). Brains were rapidly removed from the skull, postfixed for 4-8 days in the same fixative at 4°C, and then subsequently cryoprotected overnight at 4°C in the same fixative containing 10% sucrose. Frozen brains were mounted on a microtome (Reichert-Jung, Cambridge Instruments, Deerfield, IL), and 30-µm coronal sections were cut from the olfactory bulb to the end of the medulla. Sections were collected in a cold cryoprotectant solution (0.05 M sodium phosphate buffer, 30% ethylene glycol, 20% glycerol) and then stored at 20°C for further processing. Hybridization histochemical localization of CRF₁ receptor and c-fos transcripts was carried out in one in six series (every sixth section) of brain sections with the use of ³⁵S-labeled cRNA probes. Protocols for riboprobe synthesis, hybridization, and autoradiographic localization of the mRNA signals were adapted from those of Simmons et al. (43). All solutions were treated with diethylpyrocarbonate (DEPC) and sterilized to prevent RNA degradation. Tissue sections were mounted onto gelatin- and poly-L-lysine-coated slides, vacuum dried, fixed in 4% paraformaldehyde for 20 min, and digested by proteinase K [10 µg/ml in 100 mM Tris · HCl (pH 8.0) and 50 mM EDTA (pH 8.0) at 37°C for 25 min]. Brain sections were then rinsed in sterile DEPC-treated water followed by a solution of 0.1 M triethanolamine (TEA; pH 8.0), acetylated in 0.25% acetic anhydride in 0.1 M TEA, and dehydrated through graded concentrations of alcohol (50, 70, 95, and 100%). After being vacuum dried for a minimum of 2 h, 90 µl hybridization mixture (10⁷ cpm/ml) was spotted on each slide, sealed under a coverslip, and incubated at 60°C for 15-20 h in a slide warmer. Coverslips were then removed, and the slides were rinsed four times in 4× saline-sodium citrate (SSC; 150 mM NaCl and 15 mM Tris-NaCl citrate buffer, pH 7.0) at room temperature. Sections were digested by RNase A [20 µg/ml in a solution of 500 mM NaCl, 10 mM Tris · HCl (pH 8.0), and 1 mM EDTA (pH 8.0)] at 37°C for 30 min, rinsed in descending concentrations of SSC (2×, 1×, 0.5×), washed in 0.1× SSC for 30 min at 60°C and dehydrated through graded concentrations of alcohol. After a 2-h dried period under vacuum, sections were exposed at 4°C to X-ray films (Kodak) for 17 (c-fos mRNA) or 24 h (CRF₁ receptor mRNA). After development of the X-ray film, sections were defatted in xylene and dipped into NTB2 nuclear emulsion (Kodak; diluted 1:1 with distilled water). Slides were exposed for 7 (c-fos mRNA) or 14 days (CRF₁ receptor mRNA), developed in D19 developer (Kodak) for 3.5 min at 14-16°C, washed for 15 s in water, and fixed in rapid fixer (Kodak) for 5 min. Thereafter, sections were rinsed in running distilled water for 1 h, counterstained with thionin (0.25%), dehydrated through graded concentrations of alcohol, cleared in xylene, and placed under a coverslip with DPX mountant.

cRNA probe synthesis and preparation. The rat c-fos probe (2.0 kb) was generated from the EcoR 1 fragment of rat c-fos cDNA (Dr I. Verma, The Salk Institute, La Jolla, CA), subcloned into pBluescript SK-1 (Stratagene, La Jolla, CA), and linearized with Sma I and Hind III (Pharmacia) for antisense and sense probes, respectively (40). The rat CRF₁ receptor probe (1.3 kb) was generated from the Pst I-Pst I fragment of the rat prCRF PP1.3-BS cDNA [Dr. W. Vale, Peptide Biology Laboratory, The Salk Institute (35)] subcloned into pBluescript II SK (Stratagene) and linearized with BamH I and Hind III (Pharmacia) for antisense and sense probes, respectively (35). Radioactive cRNA copies were synthesized by incubating 250 ng of linearized plasmid in 6 mM MgCl₂, 40 mM Tris (pH 7.9), 2 mM spermidine, 10 mM NaCl, 10 mM dithiothreitol, 0.2 mM ATP/GTP/CTP, 200 µCi of [-³⁵S]UTP (DuPont-NEN, NEG039H), 40 U RNAsin (Promega, Madison, WI), and 20 U of T7 or T3 RNA polymerase for c-fos and CRF₁ receptor antisense and sense probes, respectively, for 60 min at 37°C. Unincorporated nucleotides were removed using the ammonium-acetate method; 100 µl of DNAse solution (1 µl DNAse, 5 µl of 5 mg/ml tRNA, 94 µl of 10 mM Tris-10 mM MgCl₂) was added, and 10 min later an extraction was accomplished using a phenol-chloroform solution. The probes were precipitated with 80 µl of 5 M ammonium acetate and 500 µl of 100% ethanol for 20 min on dry ice. After centrifugation (14,000 rpm) for 15 min, the supernatant was removed and the pellet was dried and then resuspended in 100 µl of 10 mM Tris-1 mM EDTA. A concentration of 10⁷ cpm probe was mixed into 1 ml of hybridization solution [in µl: 500 formamide, 60 5 M NaCl, 10 1 M Tris (pH 8.0), 2 0.5 M EDTA (pH 8.0), 50 20× Denhardt's solution, 200 50% dextran sulfate, 50 10 mg/ml tRNA, 10 1 M dithiothreitol (118 µl DEPC water  treated water per volume of probe used)]. This solution was mixed and heated for 5 min at 65°C before being spotted on slides.

Combination of immunocytochemistry with in situ hybridization. Protocol was performed as previously described (39). Immunocytochemistry (CRF-immunoreactive neurons) was combined to the in situ hybridization histochemistry protocol (c-fos and CRF₁ receptor mRNAs) to determine whether CRF neurons located mainly in the hypothalamic PVN express the gene encoding c-fos or CRF₁ receptor after a single or repeated immobilization challenge. Every sixth tissue slice was processed by means of the avidin-biotin peroxidase method. Briefly, slices were washed in sterile DEPC-treated 0.05 M potassium phosphate-buffer saline (KPBS) and incubated at 4°C with CRF antibody mixed in sterile KPBS, 0.4% Triton X-100, 0.25% heparine sodium salt USP (ICN Biomedicals, Aurora, OH), and 1% bovine serum albumin (fraction V, Sigma, St. Louis, MO). Sections were then incubated overnight with a rabbit anti-human/rat CRF serum (code PBL rc 70, 8/9/83 bleed), a generous gift of Dr. Wylie Vale (Peptide Biology Laboratory, The Salk Institute), used at a concentration of 1:10,000. Brain slices were rinsed in sterile KPBS and incubated for 60 min with a secondary biotinylated goat anti-rabbit IgG (1:1,500 dilution in KPBS containing heparine; Vector Laboratories). Sections were then rinsed with sterile KPBS and incubated with an avidin-biotin complex (Vectastain ABC elite kit, Vector Laboratories) for 60 min at room temperature. Thereafter, tissues were rinsed in sterile KPBS and reacted in a mixture containing sterile KPBS, the chromogen 3,3'-diaminobenzidine tetrahydrochloride (0.04%), and 0.003% hydrogen peroxide (H₂O₂). Sections were then rinsed in sterile KPBS, mounted on poly-L-lysine-coated slides, desiccated under vacuum overnight, and then processed for in situ hybridization as described above with the difference of dehydration (alcohol 50, 70, 95, 100%), which was shortened to avoid decoloration of CRF cells (brown staining), a 24- to 48-h exposure to X-ray film, a l0- to 15-day exposure to NTB2 nuclear emulsion, and the absence of counterstaining with thionin. The presence of c-fos and CRF₁ receptor transcripts was evidenced as agglomeration of silver grains in perikarya while CRF immunoreactivity within the cell cytoplasm was stained in brown.

Qualitative analysis. Anatomic identification of brain structures was essentially based on the atlas of Paxinos and Watson (33). The relative intensity of c-fos and CRF₁ mRNA signals throughout the brain of each animal was assessed on X-ray film images and graded according to the scale of undetectable (0), low (+), moderate (++), strong (+++), or very strong (++++) signal.

Quantitative analysis. Semiquantitative analysis of hybridization signals for CRF₁ receptor and c-fos mRNAs in the PVN was carried out in nuclear emulsion-dipped slides over the confines of cells of the hypothalamic PVN using a Zeiss Optical System (Axioscop) coupled to a multimedia personal computer (ASC computer) and Image software (Alcatel TITN Answare). The ratio luminosity of the signal to surface of the PVN (expressed in pixels) was analyzed on matched sections for all animals and expressed in arbitrary units. Because of the lack of basal expression of c-fos and CRF₁ receptor mRNAs in the PVN, each unilateral medial nucleus was digitized under bright field illumination and then subjected to densitometric analysis under dark field at a magnification of 10×, yielding measurements of mean refraction density in arbitrary units (RDAU). The RDAU was corrected for the average background in subtracting the signal measured immediately outside the PVN (27).

Fecal pellet output. Fecal pellet output was measured as previously described (5). The number of pellets emitted by the animals during the 90-min immobilization stress session was monitored at the end of the 90-min immobilization stress session (acute stress) or at each of the 11 stress sessions (chronic stress).

Statistical analysis. Data from Figs. 2 and 4 were analyzed by a two-way ANOVA, followed by a Tukey test procedure as post hoc comparisons (Statview 4.01). Factors were identified as follows: stress, which was composed of two levels (acute and chronic stress), and time during and post session, which was divided into six levels: before (control) and immediately, 1.5, 3, 6, and 12 h after the 90-min stress session. On the other hand, data from Fig. 6 were analyzed by a one-way ANOVA for each dependent variable.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Effect of immobilization stress on the distribution of c-fos mRNA in the rat brain. As previously described (7), basal levels of c-fos mRNA were observed in numerous brain regions of control animals (Table 1), predominantly in the dorsal endopiriform nucleus, cerebral cortex, hippocampus, thalamus (anterodorsal nucleus), suprachiasmatic nucleus, pontine gray, spinal trigeminal nucleus, and cerebellum. However, the acute stress session caused a robust expression of the IEG in numerous regions of the brain (Table 1), such as the cerebral cortex, lateral septum, parastriatal nucleus, septohypothalamic nucleus, amygdala (medial, anterior cortical, and basomedial nuclei), thalamus (paraventricular and central medial nuclei), hypothalamus (paraventricular, supraoptic, dorsomedial, retrochiasmatic, arcuate, and lateral nuclei), periaqueductal gray, laterodorsal tegmental nucleus, locus ceruleus (LC), lateral parabrachial nucleus, dorsal raphe nucleus, medial raphe nucleus, A5, medial vestibular nucleus, dorsal cochlear nucleus, lateral reticular nucleus, nucleus of the solitary tract, raphe pallidus, cuneate nucleus, and the ventrolateral medulla. In most of these regions, with some exceptions, the signal intensity peaked at the end of the session and greatly diminished or vanished at 90 min and 3 h postchallenge. In the PVN, very high c-fos mRNA levels were observed in the dorsomedial subdivision at the end of the stress session, whereas the magnocellular section exhibited barely detectable signals (Figs. 1 and 2). Whereas the signal regularly decreased in the parvocellular neurosecretory division of the PVN, c-fos expression was clearly evident in the magnocellular part 6 h after the end of the stress session and then vanished at 12 h (Figs. 1 and 2). In the supraoptic nucleus (SON), c-fos transcript was highly detectable at the end of the stress period, essentially in the dorsal part, and the signal was still positive 12 h after the end of the stress period in both the dorsal and ventral parts (data not shown). In the LC, the mRNA encoding the IEG was highly expressed at the end of the stress session and also, at a lesser extent, 90 min postchallenge but decreased at 3 h and totally vanished 6 h after the end of the single immobilization session (data not shown).

View larger version (71K): [in this window] [in a new window]   Fig. 1.   Expression of the gene encoding c-fos in the paraventricular nucleus (PVN) of the hypothalamus in acutely (left) vs. chronically (right) stressed animals. These photos depict dark-field photomicrographs of dipped autoradiographs of 30-µm coronal sections hybridized with c-fos cRNA probe throughout identical areas of the right PVN. Animals were killed before (A and G), immediately (B and H), 1.5 (C and I), 3 (D and J), 6 (E and K), and 12 h (F and L) after the end of the 90-min acute stress session or after the last session of the repeated exposures to immobilization. No c-fos transcript was observed before the stress session (A and G). In acutely stressed rats, dorsomedial parvocellular PVN exhibited a dense and selective c-fos signal at the end of the stress session (B). Signal regularly decreased in the parvocellular division, whereas it shifted to the magnocellular part 6 h after the end of the stress (E). In animals submitted to repeated challenges, immediate early gene (IEG) was not expressed in the PVN before the last stress session, whereas an attenuated signal was observed over the dorsomedial parvocellular neurons at the end of the last 90-min stress session (H). The transcript was also observed in the magnocellular PVN at an earlier time (90 min after the end of the stress session) in chronically than in acutely stressed animals (I). 3V, third ventricle; p, parvocellular part of the PVN; m, magnocellular part of the PVN. Magnification ×62.5.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Effect of an acute vs. a chronic 90-min immobilization stress session on the refraction density in arbitrary units (RDAU) for c-fos mRNA signal in the PVN of the rat hypothalamus. Results represent means ± SE. *** P < 0.001, * P < 0.05 vs. acutely stressed animals.

Stress-induced c-fos mRNA signal was greatly attenuated throughout the brain of animals submitted to repeated sessions of immobilization (Table 1), including within selective subdivisions of the hypothalamic PVN (Figs. 1 and 2). Indeed, the medial parvocellular division (neuroendocrine) of the hypothalamic PVN exhibited a low hybridization signal at the end of the last challenge. At that time, a clear and convincing signal was only detected in the ventromedial parvocellular PVN, whereas the magnocellular PVN displayed a low positive c-fos mRNA signal (Fig. 1). However, the gene encoding Fos was clearly expressed in the magnocellular subdivision of the PVN 90 min after the end of the last stress session, a signal that largely vanished 3, 6, and 12 h thereafter (Fig. 1). The RDAU data depicted by Fig. 2 did not necessarily reflect these subtle changes, because the semiquantitative image analysis was not performed for each specific subdivision of the nucleus but on the whole PVN. In the SON of chronically stressed animals, a robust expression was noted 90 min after the last challenge, still detectable 90 min after, and then vanished at time 6 h (data not shown). In numerous other regions of the brain that were clearly positive for the mRNA encoding the IEG after the single stress exposure, a lower signal was still detected at the end of the 11th session (Table 1) but was essentially gone thereafter (data not shown).

Expression of the gene encoding the CRF₁ receptor in the rat brain. As recently reported in control rats (38, 39), basal levels of CRF₁ receptor mRNA were observed in numerous regions of the brain (Table 2), such as the frontal cortex, cingulate cortex and piriform cortex, the medial and basolateral nuclei of the amygdala, the caudal division of the zona incerta, the red nucleus, the laterodorsal tegmental nucleus, the pontine gray, the cerebellum, the nucleus incertus, the spinal nucleus of the trigeminal nerve, the principal sensory nucleus of the trigeminal nerve, the suprageniculate nucleus and the external cuneate nucleus. A low to moderate signal was also observed in the medial septal nucleus, the nucleus of the diagonal band, the bed nucleus of the stria terminalis, the hippocampus, the SON, the dorsomedial nucleus of the hypothalamus, the arcuate nucleus of the hypothalamus, the parafascicular nucleus, the interpeduncular nucleus, the nucleus prepositus, the medial vestibular nucleus, and the spinal nucleus of the trigeminal nerve. Hybridized tissues with sense probe did not exhibit detectable signal in any of the regions that showed positive signal with antisense probe (data not shown).

In control rats, undetectable signal of CRF₁ receptor mRNA was observed in the PVN (Table 2; Figs. 3 and 4). In agreement with recently published data (39), animals exposed to an acute 90-min immobilization stress selectively stimulated the CRF₁ receptor gene transcript in the dorsomedial parvocellular PVN (Fig. 3). Indeed, the CRF₁ receptor mRNA signal was positive in this structure at the end of the 90-min stress session and highly expressed 90 min and 3 h thereafter, whereas the signal decreased at 6 h and almost totally vanished 12 h after the end of the stress session (Figs. 3 and 4). Several positive neurons were observed in the parvocellular PVN of acutely immobilized rats, whereas no or very few silver grains forming positive cells were detected in the magnocellular part of the PVN (Fig. 3). Moreover, no significant increase in the intensity of CRF₁ receptor signal was observed in other regions and nuclei of the brain, including the SON (Table 2).

View larger version (76K): [in this window] [in a new window]   Fig. 3.   Expression of the gene encoding the corticotrophin-releasing factor type 1 (CRF1) receptor in the PVN of the hypothalamus in acutely (left) and chronically (right) stressed animals. These photos depict dark-field photomicrographs of dipped autoradiographs of 30-µm coronal sections hybridized with CRF1 receptor cRNA probe throughout identical areas of the right PVN. Animals were killed before (A and G), immediately (B and H), 1.5 (C and I), 3 (D and J), 6 (E and K), and 12 h (F and L) after the end of the 90-min acute stress session or after the last session of the repeated exposures to immobilization. No CRF1 receptor transcript was observed before the stress session (A and G). In acutely stressed rats, the parvocellular subdivision of the PVN exhibited a dense and selective signal 1.5 (C), 3 (D) and 6 h (E) after the end of the acute stress session, whereas this signal was absent in chronically stressed animals at the same periods (I, J, K). Magnification ×62.5.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Differential effects of a single or repeated sessions of immobilization stress on the expression of the CRF1 receptor in the PVN of the rat hypothalamus. Results represent means ± SE of the RDAU. *** P < 0.001 vs. acutely stressed animals.

In chronically stressed animals, the CRF₁ receptor mRNA signal was also detected in the dorsomedial parvocellular PVN; however, the signal was very low compared with acutely stressed animals (Figs. 3 and 4). Indeed, the semiquantitative analysis showed that the CRF receptor signal was in the same order of magnitude than in acutely stressed animals at the end of the stress, but no significant increase of the refraction density was observed 90 min and 3 h after the end of the stress session (Fig. 4). Interestingly, very few positive neurons were detected in the parvocellular neurosecretory division of the hypothalamic PVN in chronically challenged rats (Fig. 3).

Induction of c-fos and CRF₁ receptor transcripts in CRF-ir neurons. In control rats, none of the CRF-immunoreactive (ir) neurons exhibited positive signals for Fos transcript in the parvocellular neurosecretory zone of the PVN. In acutely stressed rats, a large concentration of CRF-ir neurons were positive for the IEG, whereas very few CRF-ir neurons expressed the transcript in the endocrine hypothalamus of chronically stressed animals after the last session of immobilization (Fig. 5). In fact, 20.2% of the double-labeled neurons were found in the PVN of acutely stressed animals at the end of the stress, whereas only 5.4% of CRF-ir neurons expressed the transcript at the same postchallenge time in animals submitted to repeated sessions (Fig. 6).

View larger version (93K): [in this window] [in a new window]   Fig. 5.   Expression of c-fos (A and B) and CRF1 receptor (C and D) mRNAs in CRF-immunoreactive (ir) perikarya located in the hypothalamic PVN of acutely (left) and chronically (right) stressed animals. Animals were fixed with 4% paraformaldehyde-borax at the end of the stress session (CRF-ir/c-fos mRNA) or 90 min thereafter (CRF-ir/CRF1 mRNA). Immunocytochemistry (CRF protein, brown neurons) was performed on the same brain sections (30 µm) before in situ hybridization histochemistry (CRF1 receptor and c-fos mRNAs, silver grains). Solid arrowheads, CRF neurons expressing either the IEG c-fos or CRF1 receptor transcript; open arrowheads, CRF-ir perikarya alone. Note the number of silver grains delineating several CRF-containing neurons that express the gene encoding c-fos (A) or CRF1 receptor (C) mRNAs in acutely stressed animals, whereas very few CRF-containing neurons were positive for either transcripts in the PVN of chronically stressed animals (B and D). Magnification ×625.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Number of CRF-ir cells expressing either c-fos or CRF1 receptor mRNAs in the hypothalamic PVN immediately after a single stress exposure (acute stress) or 90 min after the end of the 11th session (chronic stress) of immobilization, respectively. *** P < 0.001, ** P < 0.01 vs. acutely stressed animals.

In control rats, none of the CRF-ir neurons exhibited a positive signal for CRF₁ receptor transcript in the parvocellular PVN. Numerous, but not all, CRF-ir neurons were positive for the CRF₁ receptor transcript in acutely stressed rats, a phenomenon that essentially vanished in the PVN of chronically stressed animals (Fig. 5); 35.8 and 37.8% of the CRF-ir neurons were positive for the mRNA encoding the CRF₁ receptor in acutely stressed animals 90 min and 3 h after the end of the stress session, respectively. At the same times, only 6.3 and 1.8% of CRF-ir neurons expressed the CRF₁ receptor transcript in chronically stressed animals (Fig. 6).

Effect of an acute and chronic immobilization stress on fecal pellet output. In acutely stressed animals, fecal pellet output measured at the end of the 90-min immobilization stress session was 12.5 ± 0.7. Fecal pellet output of animals exposed to repeated sessions of immobilization was in the same order of magnitude (12.1 ± 0.8) during the first challenge but significantly decreased (7.5 ± 0.9; P < 0.01) at the end of the 11th stress session. In fact, fecal pellet output decreased from the first to the 11th stress session in chronically stressed animals.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The present study provides the evidence that chronic immobilization stress caused an attenuation in the expression of the gene encoding the CRF₁ receptor within neuroendocrine CRF cells of the PVN, an event associated to a decrease in stress-induced postsynaptic activation of these neurons. Numerous studies have examined the regulation of CRF; however, little is known about the mechanisms controlling the expression of CRF receptors in the brain. Some studies have failed to show alterations in levels of CRF receptor ligand binding in the brain by chronic stress or corticosterone treatment (15, 16, 47), whereas others have demonstrated that levels of CRF receptors in the brain are decreased by repeated stress (2) or intracisternal injection of CRF (17). As previously reported (39), a selective and transient increase of CRF₁ mRNA expression was found in the parvocellular PVN of acutely challenged rats, whereas little notable changes were detected in other regions of the brain. These data are quite interesting because they suggest that sophisticated mechanisms take place in a site-dependent manner to regulate the CRF₁ receptor in stressed animals.

The neuropeptide CRF itself is believed to play an important role in triggering the expression of CRF receptor gene in the endocrine hypothalamus of the stressed rat brain. We have recently reported that the gene encoding the CRF₁ receptor was modulated in a positive manner by exogenous CRF selectively within the PVN. Indeed, central administration (intracerebroventricular) of human/rat CRF induced transcription of the CRF₁ receptor within the PVN, whereas rats treated with vehicle and/or D-Phe¹², a CRF antagonist, displayed low to undetectable levels of CRF₁ receptor mRNA within this hypothalamic structure (26). Using intronic probe technology, we have also shown that exogenous CRF activates transcription of CRF selectively within the PVN; the hybridization signal for the CRF heteronuclear RNA was observed exclusively in the PVN at 15 min post-CRF injection. This set of data therefore supports the hypothesis of an ultrashort positive feedback loop within the PVN through which CRF may modulate its own biosynthesis as well as that of its type 1 receptor. This may ultimately play an important role in the animal's ability to restore homeostasis during stressful situations. In the present study, rats submitted during 11 days to the same immobilization stress exhibited a decrease in CRF₁ mRNA only in the PVN compared with acutely stressed rats, whereas no modification was observed in other regions of the brain. These data argued for a selective adaptive mechanism modulating quite exclusively the CRF₁ receptor in the PVN, an event that may involve an attenuated CRF drive within the endocrine hypothalamus. The physiological relevance of this phenomenon has yet to be clarified, but it is possible that the CRF induced its type 1 receptor within the parvocellular cells is a mechanism involved to restore and prepare neuroendocrine CRF cells for the subsequent challenges. The role of the type 1 receptor in regulating PVN CRF neurons may be gradually reduced when the system is well adapted, which suggests that the CRF/CRF₁ receptors may have a potent influence for adjusting this essential neuroendocrine response to a novel, but not controlled and repeated, situation. It is therefore possible that the interaction between the ligand and its receptor modulate the transduction signals necessary to allow the biosynthesis of the peptide in the early stages of the acute stress conditions, but not during the chronic and adapted responses where the neuroendocrine CRF itself may have a more limited influence on the corticotroph axis.

In parallel to these neuroanatomical changes, we observed functional modifications as revealed by a decrease of fecal pellet output in chronically stressed animals. Numerous data argue for a role of CRF in stress-induced gastrointestinal motility disturbances (5, 45, 46). Microinjection of CRF into the central nervous system, either in the cerebrospinal fluid, the PVN, or the LC increases colonic motility through vagal pathways (46). In addition, we have previously shown that the effects of stress on the colonic transit were significantly prevented by pretreatment with central injection of CRF antagonist (5). In the present study, a decrease in functional modification of colonic transit was observed in chronically challenged rats, a phenomenon that paralleled the downregulation of CRF₁ receptor expression within CRF-containing cells of the neurosecretory PVN. The possibility therefore remains that the CRF₁-expressing neurons participate in the functional adaptation of stress-related responses after repeated homotypic stressful events. It is of interest to note that c-fos mRNA signal was still detected after the 11th stress session in the ventromedial parvocellular subdivision of the PVN, which is an autonomic-related group of cells likely to play an important role in the gastrointestinal response to stress. Although significantly attenuated, this process was in fact maintained in chronically challenged animals and may be CRF₁-receptor independent.

A widespread and robust induction of c-fos mRNA was observed in the brains of acutely stressed animals with a peak expression at the end of the stress session, a message that systematically decreased thereafter. The IEG was expressed in several regions known to be regulating the PVN and hypothalamic-pituitary-adrenal (HPA) axis, such as the lateral septal nucleus, medial preoptic area, bed nucleus of the stria terminalis, medial and central amygdaloid nuclei, cingulate cortex, and in the brain stem catecholaminergic cell groups (A1-2/C1-2). In the PVN of acutely challenged animals, c-fos distribution was essentially localized to the parvocellular division known to contain neuroendocrine CRF perikarya. Combination of in situ hybridization with immunohistochemistry showed that numerous, but not all, CRF perikarya expressed the IEG. C-fos mRNA generally peaked at the end of the 90-min stress session and then regularly decreased to return to basal levels in most brains regions, with some exceptions, which may have important functional implications. For example, very few neurons of the magnocellular PVN expressed c-fos mRNA in the early times, although 6 h after the end of the stress session the IEG had vanished in the parvocellular section but increased in the magnocellular PVN. In addition, peak expression of c-fos mRNA was observed within the LC at the end of the stress session and the signal remained high up to 3 h after the stress session and totally vanished 6 and 12 h thereafter. This may imply additional secondary activation related to attention and arousal, functions that have been proposed for this cell group (4, 12).

In chronically stressed animals, we observed a desensitization of the c-fos response across the brain. Attenuation in the signal intensity was observed in the parvocellular division of the PVN, whereas the signal was more rapidly induced in the magnocellular PVN of chronically challenged rats than after the acute immobilization stress. Although we did not characterize the neurochemical nature of these neurons, vasopressinergic neurons are likely candidates. There is compelling evidence for an increased role of vasopressin during repeated stress conditions as a major neurosecretagogue to trigger the HPA axis (1, 8). This concept is reinforced by the fact that there is a transcript-specific alteration in gene regulation within specific subdivisions of the hypothalamic PVN, in which CRF gene transcription is desensitized while AVP primary transcript is maintained after a repeated restraint (25).

Compared with the acutely stressed rats, a significant decrease of c-fos expression was also detected in catecholaminergic brain stem cell groups, such as the ventrolateral medulla A1/C1, the nucleus of the solitary tract A2/C2, and the locus ceruleus A6, after the 11th session of neurogenic challenge. Physiological investigations have demonstrated that the integrity of these cell groups that innervate the PVN (particularly the A1-2/C1-2) is necessary for the activation of HPA axis by different stressors, including ether (44) and immobilization (13, 14) stress. There is also considerable evidence that the HPA response is progressively reduced after repeated exposure to the same stressor. This phenomenon, called habituation, was shown to occur after exposure to various chronic intermittent stressors, such as handling (10), novelty (36), noise (3), and restraint (18, 30, 37). Habituation of brain stem catecholaminergic groups of cells to daily restraint stress was also observed; habituation was first observed in the A1/C1 group on day 5 and was more pronounced on day 10, whereas in the A2/C2 and the LC, habituation was delayed and was less pronounced than for A1/C1 at any time (21). The parallel time course of habituation for the A1/C1 cell groups and HPA axis suggests that the progressive lowering HPA axis response may result, at least in part, in a decreased A1/C1 input (the most important origin of the PVN catecholaminergic innervation). For ether and restraint stress, the integrity of the catecholaminergic innervation originating from the brain stem has been shown to be necessary for the stimulation of the HPA axis. Because habituation only developed after exposure to the same stressor, it is likely to result from familiarization to this particular stressor. Habituation of catecholaminergic-containing neurons of the brain stem, which is reflected by a decrease in c-fos expression in the ventrolateral medulla and the nucleus of the solitary tract, could participate in the attenuated activity of neuroendocrine CRF neurons and those expressing its type 1 receptor in the rat PVN after repeated exposures to the same neurogenic stressor.

Perspectives