Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The neuropeptide, corticotropin-releasing factor (CRF) coordinates the endocrine response to stress through its action as the major physiological regulator of the hypothalamic-pituitary-adrenal (HPA) axis [see Owens and Nemeroff (1991) for review]. In addition to the hypothalamic paraventricular nucleus (PVN), CRF-containing neurons are also localized to extrahypothalamic limbic structures and brainstem nuclei that subserve behavioral and autonomic regulatory functions. There is overwhelming evidence that suggests that CRF functions as a neurotransmitter in these regions. In this manner, it is hypothesized that CRF neuronal systems are strategically located to integrate not only the endocrine, but also the behavioral, immune, and autonomic responses to stress (Owens and Nemeroff, 1991).

Recently, the existence of two distinct CRF receptor subtypes with contrasting neuroanatomical distributions has been demonstrated: CRF₁ and CRF₂ [see Chalmers et al. (1996) for review]. In rats, CRF₁ is the predominant receptor within the pituitary, cerebellum, and neocortex. The CRF₂ receptor consists of at least two major splice variants, CRF_2A and CRF_2B. The former is more prevalent in subcortical regions, particularly the lateral septum (LS), ventromedial hypothalamus (VMH), and dorsal raphe nucleus (DRN), whereas the latter is more abundant in the periphery.

Another relatively recent and important discovery is that of urocortin, a second endogenous mammalian ligand for the CRF receptors (Vaughan et al., 1995; Donaldson et al., 1996a,b). Urocortin binds with equal affinity to both CRF receptor subtypes, but demonstrates approximately 10-fold higher affinity for CRF₂ receptors than does CRF itself [see Chalmers et al. (1996) for review]. In the rat, urocortin-containing perikarya and urocortin mRNA expression are most prominent in the Edinger-Westphal nucleus and the lateral superior olive (Vaughan et al., 1995), regions that do not contain CRF mRNA. The highest density of urocortin innervation is observed in the LS and DRN (Vaughan et al., 1995; Wong et al., 1996). It is of interest to note that these two regions nearly exclusively express CRF_2A mRNA. This overlapping distribution, along with the higher affinity of urocortin for the CRF₂ receptor as compared with CRF, provide evidence that urocortin may be the endogenous CRF₂ ligand.

There now exists a large body of evidence supporting the hypothesis that CRF is hypersecreted from hypothalamic as well as extrahypothalamic neurons in some patients with affective disorders [for reviews, see Arborelius et al. (1999) and Owens et al. (1999)]. After antidepressant treatment, measures of hyperactivity HPA axis and CRF function appear to normalize suggesting that hyperactivity of CRF neurons is a state marker for depression (Plotsky et al., 1998; Holsboer, 1999).

The hypothesis that CRF may also play a role in the pathophysiology of anxiety disorders derives mainly from preclinical findings [for review, see Arborelius et al. (1999)]. It is well known that central administration of CRF increases anxiety-like behaviors in rodents and transgenic mice that overexpress CRF exhibit anxiogenic behavior. Conversely, CRF receptor antagonists or CRF antisense oligonucleotides produce anxiolytic-like effects in the rat. Studies using CRF₁ receptor knockout mice and CRF₁ and CRF₂ receptor antisense oligonucleotides have revealed that the anxiogenic effect of CRF appears to be mediated by the CRF₁ receptor subtype (Heinrichs et al., 1997; Smith et al., 1998; Timpl et al., 1998). In humans, elevated concentrations of cerebrospinal fluid CRF have been reported in patients with some anxiety disorders, including post-traumatic stress disorder, obsessive-compulsive disorder, and Tourette's syndrome (see Plotsky et al., 1998; Holsboer, 1999).

Based on findings such as those described above, it has been hypothesized that CRF₁ receptor antagonists may represent a novel class of drugs for treatment of depression and/or anxiety disorders (Owens and Nemeroff, 1999). Recently, a selective, nonpeptide CRF₁ receptor antagonist, CP-154,526, was synthesized and characterized (Chen et al., 1997). CP-154,526 displays high affinity to CRF₁ receptors (K_i < 10 nM) and blocks CRF-stimulated adenylate cyclase activity in rodent pituitary and cortical membranes (Lundkvist et al., 1996; Schulz et al., 1996). As expected, acute administration of CP-154,526 or its analog antalarmin blocks CRF- and stress-induced elevations in plasma adrenocorticotropin (ACTH) and shows anxiolytic activity (Lundkvist et al., 1996; Schulz et al., 1996; Webster et al., 1996; Chen et al., 1997; Deak et al., 1999). Moreover, CP-154,526 exhibits antidepressant-like activity in the learned helplessness paradigm (Mansbach et al., 1997). Interestingly, but in contrast to standard antidepressant drugs that require repeated administration before clinical effects are observed, CP-154,526 showed antidepressant-like activity after a single dose. However, in the treatment of human depression and anxiety disorders drugs are usually administered during a prolonged period of time. Thus, the purpose of the present study was to investigate whether the anxiolytic effect and the blockade of the endocrine stress response observed after acute administration of CP-154,526 persists during chronic treatment. Moreover, does chronic blockade of CRF₁ receptors lead to adaptive alterations in central CRFergic function as assessed by measures of CRF and urocortin mRNA expression and CRF receptor mRNA expression and receptor binding?

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Male Sprague-Dawley rats (200-250 g body weight at the beginning of the study; Charles River laboratories, Raleigh, NC) were housed two to three per cage under controlled laboratory conditions (12:12 h light:dark cycle, lights on at 7:00 AM) with water and food available ad libitum. The behavioral experiments and the stress tests were performed between 8:00 AM and noon.

Drug Treatment. CP-154,526 (butyl-[2,5-dimethyl-7-(2,4,6-trimethylphenyl)-7H-pyrrolo[2,3-d]pyrimidin-4-yl]-ethylamine) was provided by Pfizer Inc. (Groton, CT).

Defensive Withdrawal Paradigm. The apparatus consisted of an open field (75 × 75 cm) with 50-cm high side walls made of white plastic. Each rat was placed in a black polyvinyl chloride tube (10 cm in diameter x 21 cm) that was closed at one end and placed at a distance of 20 cm from a corner of the open field with the open end of the tube facing the corner. The behavior of the rat was recorded by videotape for 10 min. The following parameters were monitored: 1) latency until the animal exited into the open field, 2) total time the animal stayed in the tube, and 3) number of entries into the open field. Entries were defined as all four paws in the open field (Takahashi et al., 1989). The light intensity in the open field was approximately 600 lux during the experiment. Because of the great interindividual variability in this model, we used a 95% confidence interval.

Airpuff Startle. Rats were implanted with chronic jugular vein cannulas 2 days before the stressor under aseptic conditions as previously described (Thrivikraman and Plotsky, 1993). Briefly, animals were anesthetized with an s.c. injection of a mixture of acepromazine (1.5 mg/kg; Tech America, Fermenta Animal Health Co., KS City, MO), ketamine (37 mg/kg; Vetalar, Aveco Co. Inc., Fort Dodge, IA), and xylazine (7.4 mg/kg; Rompun, Miles Laboratory Inc., Shawnee, KA). The jugular vein was exposed by blunt dissection, and a small incision was made using iridectomy scissors. The cannula, consisting of a piece of PE50 tubing (Clay Adams, Sparks, MD) with a tip of silicone tubing (T5715-3; Baxter, McGaw Park, IL), was inserted into the vein approximately 3 cm in the caudal direction, ligated to the vessel, and tunneled s.c. to emerge from the neck of the animal. The wounds were closed with metal clips, and the cannula was filled with sterile saline containing gentamicin (2.5 µg/100 g of body weight; Schein Pharmacy, Port Washington, NY). Thereafter, animals were individually housed in polyethylene buckets (28 cm in diameter and 37 cm high) containing regular wood chip bedding, under controlled laboratory conditions (12:12 h light:dark cycle, lights on at 7:00 AM) with free access to food and tap water.

CRF RIA. Regional brain samples for CRF RIA were stored at 80°C until assay. The CRF concentrations were determined by RIA as previously described in detail (Ladd et al., 1996). Briefly, duplicate aliquots from each sample were extracted in 1 M HCl, lyophilized, and then reconstituted in 200 µl of SPEAB buffer (100 mM NaCl, 50 mM Na₂HPO₄, 25 mM EDTA, 0.1% sodium azide, 0.1% BSA, and 0.1% Triton X-100, pH 7.3) and incubated at 4°C for 18 h with 100 µl of rabbit anti-CRF (Peninsula, Belmont, CA) at a final dilution of 1:70,000 in SPEAB buffer containing 1.0% normal rabbit serum. Next, 50 µl (17,000-20,000 cpm) of radiolabeled ¹²⁵I-Tyr^o-rat/human CRF (New England Nuclear, Boston, MA) were added to each tube. After incubation with radiolabeled CRF for 24 h at 4°C, 10 µl of goat anti-rabbit serum (Arnel Products, New York, NY) diluted 1:1 in SPEAB with 1% normal rabbit serum was added to precipitate bound CRF. A standard curve was prepared with rat/human CRF (Bachem, Torrance, CA) from 0.625 to 5120 pg/tube. The sensitivity of the assay was 1.25 pg/tube. CRF concentrations are expressed as picograms of CRF per milligram of protein, except for the median eminence (ME), which is expressed as picograms of CRF per ME.

CRF Receptor Binding. Single point CRF receptor binding assays were performed as previously described (Ladd et al., 1996) on individual tissue samples, dissected as described for the CRF RIA (vide supra) at a near-saturating concentration of ¹²⁵I-labeled ovine CRF (1 nM final concentration; 0.1 nM ¹²⁵I-labeled ovine CRF plus 0.9 nM ovine CRF). Specific CRF receptor binding was calculated by subtracting the mean counts per minute in triplicate pellets incubated with ¹²⁵I-labeled ovine CRF in the presence of 1 µM unlabeled rat CRF. The K_d of ¹²⁵I-labeled ovine CRF binding in our laboratory ranges from 0.25 to 0.6 nM. Tissue samples were homogenized in 4 ml of buffer (50 mM Tris-HCl, 10 mM MgCl₂, and 2 mM EDTA, pH 7.2, at 22°C) containing the peptidase inhibitors aprotinin (0.1%) and bacitracin (0.1 mM) and 0.1% BSA using a Polytron homogenizer (model 3100; Brinkmann Instruments, Westbury, NY) at 20,000 rpm for 10 s, followed by centrifugation at 32,000g for 10 min at 4°C. This procedure was repeated either two or four times, and the sample pellets were washed, capped, and frozen at 70°C. On the day of the assay, samples were resuspended and homogenized in buffer to yield a final concentration of approximately 150 µg of protein/100 µl. Aliquots (100 µl) of membrane homogenate were incubated for either 90 min or 3 h with 100 µl of ¹²⁵I-labeled ovine CRF (0.1 nM final concentration; New England Nuclear), 50 µl ovine CRF (0.9 nM final concentration), and 50 µl of either rat CRF (1 µM final concentration) to define nonspecific binding or incubation buffer (total binding). Specific binding as a percentage of total binding varies slightly among brain regions and represents 60% to 70% of total binding. Aliquots of the tissue homogenate were used to determine total protein content using BSA as the standard. After incubation, samples were microcentrifuged for 3 min at 12,000g, aspirated, then washed in ice-cold PBS (pH 7.4) containing 0.01% Triton X-100. The supernatant was removed, and the pellet was counted in an gamma counter (LKB, Rockville, MD) at 86% efficiency.

CRF Receptor Autoradiography. In vitro CRF receptor autoradiography as previously described (Skelton et al., 2000) was performed on 15-µm rat brain sections mounted on SuperFrost Plus slides (Fisher Scientific). Brain sections were fixed for 2 min in 0.1% paraformaldehyde followed by a 15-min incubation in assay buffer (50 mM Tris, 10 mM MgCl₂, 2 mM EGTA, 0.1% BSA, 0.1 mM bacitracin, and 0.1% aprotinin; pH 7.5) to remove endogenous CRF. Next, triplicate slides containing adjacent brain sections were incubated for 2 h at room temperature in one of three conditions: 1) 0.1 nM radiolabeled ¹²⁵I-sauvagine (DuPont-NEN) to determine total binding at both the CRF₁ and CRF₂ receptor subtypes, 2) 0.1 nM radiolabeled ¹²⁵I-sauvagine + 1 µM CP-154,526 to determine CRF₂ receptor-specific binding, or 3) 0.1 nM radiolabeled ¹²⁵I-sauvagine + 1 µM unlabeled sauvagine (American Peptide Company Inc., Sunnyvale, CA) to determine nonspecific binding. After the incubation, unbound radioligand was removed by two 5-min rinses in ice-cold (4°C) PBS + 1% BSA on a rotating platform at 60 rpm, followed by two brief dips in ice-cold ddH₂O. Slides were then rapidly dried with a blow dryer on the coldest setting and apposed to Kodak (Rochester, NY) Biomax MR film with ¹²⁵I microscale standards (Amersham Pharmacia Biotech, Piscataway, NJ) for 90 to 160 h.

In Situ Hybridization. Serial coronal brain sections (15 µm) were prepared on a cryostat at 18°C, thaw-mounted onto SuperFrost Plus slides under RNase-free conditions, and stored with Humi-Cap desiccant capsules (Life Technologies, Grand Island, NY) at 80°C until the assay. In situ hybridization was performed according to the procedures described by Simmons et al. (1989) with minor modifications. Briefly, slides were warmed in a step-wise manner to room temperature, postfixed in 4% paraformaldehyde (pH 7.4) for 20 min, and rinsed twice in 10 mM PBS (pH 7.4) for 2 min. Next, the slides were treated for 15 min with proteinase K (Promega Life Science, Madison, WI; 10 µg/ml in 0.1 M Tris with 50 mM EDTA) at room temperature, followed by a quick rinse in deionized water, 2.5 min in 0.1 M triethanolamine (pH 8.0), 10 min of acetylation (0.5% acetic anhydride in 0.1 M triethanolamine, pH 8.0), two rinses in 2× SSC, and dehydration through a graded ethanol series. The sections were then air-dried for at least 1 h before hybridization.

Image Analysis. Images from the in situ hybridization and receptor autoradiography films were digitized with a Dage-MTI CCD-72 (Michigan City, IN) image analysis system equipped with a Nikon camera. Semiquantitative analysis was performed using Scion Image (version 3.0b; Frederick, MD) or Analytical Imaging Station software (Imaging Research, Inc., St. Catherines, Ontario, Canada). Optical densities were calibrated against ¹⁴C standards (in situ hybridization films) or ¹²⁵I microscale standards (receptor autoradiography films) and expressed in terms of nanocuries per gram of tissue equivalent. For the purpose of quantifying mRNA levels, specific signal density was determined relative to neutral background density present in the same brain section. For the purpose of quantifying CRF receptor levels: CRF₁ receptor-specific binding was calculated by subtracting CRF₂ receptor binding from total binding, and CRF₂ receptor specific binding was calculated by subtracting nonspecific binding from CRF₂ receptor binding. In all cases, two to six sections per region were matched for rostrocaudal level according to the atlas of Paxinos and Watson (1986) and used to produce a single value for each animal.

Statistical Analysis. Adrenal weight, basal levels of ACTH and corticosterone, and the neurochemical data were analyzed by ANOVA. The effects on ACTH and corticosterone levels before and after airpuff startle were analyzed by two-way repeated measures ANOVA followed by Tukey's post hoc test. The different parameters of the defensive withdrawal test were analyzed using Kruskal-Wallis one-way ANOVA on ranks or the Mann-Whitney rank sum test. A P value of <.05 was considered significant. All values are presented as mean ± S.E.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of CP-154,526 on Defensive Withdrawal. Acute treatment with CP-154,526 did not affect the latency to leave the tube, total time spent in tube, or number of entries into open field in the defensive withdrawal test (Table 1). Animals treated chronically with the low dose of CP-154,526 (3.2 mg/kg/day) showed a decreased latency to initially leave the tube compared with vehicle-treated rats, although this effect did not reach statistical significance (Fig. 1A). However, the total time spent withdrawn in the tube was significantly decreased in CP-154,526-treated animals compared with vehicle-treated rats (Fig. 1A). There was no difference in the number of entries into the open field between animals treated with CP-154,526 and vehicle (Fig. 1B), suggesting that locomotor activity was not affected by the drug treatment. The animals receiving chronic treatment with the high dose of CP-154,526 (32 mg/kg/day) revealed signs of sickness behavior and stayed withdrawn in the tube most of the time. These rats were not included in the analyses.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Defensive withdrawal behavior after chronic treatment with 3.2 mg/kg/day CP-154,526 (CP 3.2; n = 8) compared with vehicle (n = 6). A, the latency of the animals to initially leave the tube and the total time the animals stayed in the tube (duration). B, the number of entries into the open field indicates the locomotor activity of the animals. Data are shown as means ± S.E. **P < .01, Mann-Whitney rank sum test.

Effects of CP-154,526 on Endocrine Responses to Airpuff Startle. As previously shown (Engelmann et al., 1996), airpuff startle induced a time-dependent increase in both plasma ACTH and corticosterone concentrations in vehicle-treated animals (Fig. 2, A and B; Fig. 3, A and B). An acute dose of 32, but not 3.2 mg/kg of CP154,526, completely blocked airpuff-induced increases of plasma ACTH and corticosterone levels (Fig. 2, A and B).

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Endocrine responses to airpuff startle after pretreatment with an acute dose of 3.2 mg/kg CP-154,526 (CP 3.2; n = 5), 32 mg/kg CP-154,526 (CP 32; n = 5), or vehicle (n = 4). Mean ± S.E. of plasma (A) ACTH and (B) corticosterone concentrations in blood samples taken immediately before and at different time points after the airpuff startle. A two-way ANOVA (repeated measures) revealed a significant treatment (P < .05) and time (P = .001) effect, but no significant interaction between treatment and time for ACTH, and a significant treatment (P < .01) and time (P < .001) effect as well as a significant interaction between treatment and time (P < .001) for corticosterone. **P < .01, ***P < .001 between CP-154,526 (32 mg/kg)- and vehicle-treated animals, Tukey's post hoc test.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Neuroendocrine responses to airpuff startle in rats treated chronically with CP-154,526 (3.2 mg/kg/day; n = 5) compared with vehicle-treated animals (n = 5). Mean ± S.E. of plasma (A) ACTH and (B) corticosterone concentrations in blood samples taken immediately before and at different time point after the airpuff startle. A two-way ANOVA (repeated measures) revealed a significant time effect for both ACTH (P < .001) and corticosterone (P < .001) but no significant treatment or interaction between treatment and time effect for either ACTH or corticosterone between the two treatment groups. *P < .05, **P < .01, ***P < .001 versus time point 0. +P < .05 between CP-154,526- and vehicle-treated animals, Tukey's test.

Effects of CP-154,526 on Adrenal Weight. There was no significant difference in adrenal weight between animals treated chronically with vehicle (41.6 ± 1.2 mg, n = 8), 3.2 mg/kg/day CP-154,526 (43.3 ± 1.8 mg, n = 9), or 32 mg/kg/day CP-154,526 (46.4 ± 2.0 mg, n = 10).

Effects of CP-154,526 on CRF₁ Receptor Binding and mRNA Expression. Chronic administration of CP-154,526 produced a significant and dose-dependent decrease in CRF₁ receptor binding in the parietal cortex as demonstrated by ex vivo autoradiography (PC; Fig. 4, A and B). A significant decrease in CRF₁ receptor binding was also observed in cerebellum homogenates; however, this effect was abolished after more washes and a longer incubation time (Fig. 5). These changes almost certainly represent residual drug bound to the receptors.

View larger version (117K): [in this window] [in a new window]   Fig. 4.   A, representative autoradiographs of coronal sections of rat brain showing a dose-dependent decrease in CRF1 receptor binding as revealed by 125I-sauvagine binding in the parietal cortex (PC) in rats treated chronically with CP-154,526; vehicle (upper panel), 3.2 mg/kg/day CP-154,526 (middle panel), and 32 mg/kg/day CP-154,526 (lower panel). Note that CRF receptor binding in the LS, which mainly express CRF2A receptors, and punctate binding associated with the cerebral vasculature is not affected by chronic CP-154,526 treatment. B, effects of chronic administration of 3.2 mg/kg/day CP-154,526 (CP 3.2; n = 9), 32 mg/kg/day CP-154,526 (CP 32; n = 7), or vehicle (n = 5) on CRF1 receptor binding (mean ± S.E.) in the parietal cortex. **P < .01, ***P < .001 versus vehicle, one-way ANOVA followed by Tukey's test.

View larger version (34K): [in this window] [in a new window]   Fig. 5.   Effects of chronic administration of 3.2 mg/kg/day CP-154,526 (n = 10), 32 mg/kg/day CP-154,526 (n = 11), or vehicle (n = 10) on CRF1 receptor binding (mean ± S.E.) in cerebellum homogenates after two washes and 90-min incubation time (Cer-1) and four washes and 3-h incubation time (Cer-2). *P < .05, ***P < .001 versus vehicle, one-way ANOVA followed by Tukey's test.

View larger version (199K): [in this window] [in a new window]   Fig. 6.   Representative autoradiographs showing CRF1 mRNA expression in coronal sections of rat brain in the parietal cortex (PC) and the basolateral amygdala (BLA; upper panel), and in the cerebellum (Cer; lower panel) observed with in situ hybridization.

Effects of CP-154,526 on CRF Content and mRNA Expression. Acute administration of CP-154,526 (3.2 or 32 mg/kg) did not affect CRF peptide content in the prefrontal cortex, frontal/parietal cortex, hippocampus, amygdala, ME, DRN, locus ceruleus (LC), and parabrachial nucleus. In addition, CRF gene expression in the PVN and the central nucleus of amygdala (CeA) did not change after a single dose of the drug (data not shown).

View larger version (57K): [in this window] [in a new window]   Fig. 7.   A, representative autoradiographs showing CRF mRNA expression in coronal rat brain sections in the parietal cortex (PC) and the bed nucleus of stria terminalis (BNST; upper panel), in the central nucleus of amygdala (CeA; middle panel), and in Barrington's nucleus (Bar; lower panel). B, representative autoradiographs showing CRF mRNA expression in the PVN after chronic treatment with vehicle (upper panel) and 32 mg/kg/day CP-154,526 (CP 32; lower panel). C, CRF mRNA expression (mean ± S.E.) in the PC, CeA, BNST, PVN, and Bar during chronic treatment with 3.2 mg/kg/day CP-154,526 (CP 3.2: n = 7, PC; n = 7, CeA; n = 7, BNST; n = 6, PVN; n = 6, Bar), 32 mg/kg/day CP-154,526 (CP 32: n = 10, PC; n = 8, CeA; n = 8, BNST; n = 6, PVN; n = 7, Bar), or vehicle (n = 6, PC; n = 9, CeA; n = 6, BNST; n = 7, PVN; n = 6, Bar). *P < .05, **P < .01, one-way ANOVA followed by Tukey's test.

Effects of CP-154,526 on CRF₂ Receptor Binding and mRNA Expression and Urocortin mRNA Expression. CRF₂ receptor binding in LS, VMH, and DRN was not altered by chronic administration of CP-154,526 (Fig. 8; Table 3). In addition, neither CRF_2A receptor mRNA expression in the LS and DRN nor urocortin mRNA expression in the Edinger-Westphal nucleus was altered by CP-154,526 administration (Figs. 9 and 10; Tables 4 and 5).

View larger version (123K): [in this window] [in a new window]   Fig. 8.   Representative autoradiographs of coronal rat brain sections showing CRF2 receptor binding as revealed by 125I-sauvagine binding in the presence of unlabeled CP-154,526 in the lateral septum (LS; upper autoradiograph), ventromedial hypothalamus (VMH; middle autoradiograph), and the dorsal raphe nucleus (DRN; lower autoradiograph).

View larger version (204K): [in this window] [in a new window]   Fig. 9.   Representative autoradiographs of coronal rat brain sections showing CRF2 receptor mRNA expression in the lateral septum (LS; upper panel) and the dorsal raphe nucleus (DRN; lower panel).

View larger version (110K): [in this window] [in a new window]   Fig. 10.   Representative autoradiograph of a coronal rat brain section showing urocortin mRNA expression in the Edinger-Westphal nucleus (EWN).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Previous studies investigating possible anxiolytic activity of a single dose of CP-154,526 or antalarmin have sometimes provided inconsistent results. For instance, CP-154,526 attenuates fear-potentiated startle (Schulz et al., 1996; Chen et al., 1997) and shows anxiolytic-like activity in the elevated plus-maze (Lundkvist et al., 1996), however, in one study CP-154,526 was without effect in this model (Griebel et al., 1998). Moreover, in the light/dark test and the mouse defensive test battery, CP-154,526 has been shown to have anxiolytic-like activity but not in conflict tests, i.e., punished lever pressing and punished drinking tests (Griebel et al., 1998). In the present study we used the defensive withdrawal test, because central CRFergic systems appear to be involved in the mediation of the behaviors in this model (Takahashi et al. 1989; Smagin et al. 1996). We observed that acute administration of either dose of CP-154,526 did not significantly affect defensive withdrawal behaviors. The lack of anxiolytic-like behaviors of CP-154,526 is in contrast to previous findings where central administration of the CRF receptor antagonist -helical CRF_9-41 produced clear anxiolytic-like behaviors in this model (Takahashi et al., 1989). This may be related to the different route of administration of the CRF receptor antagonists used in the two studies, i.e., s.c. versus i.c.v. However, in the present study we observed a tendency for a decrease in latency to initially leave the tube after acute treatment with both doses of CP-154,526 (Table 1). The lack of statistical significance of this effect may be related to the low number of animals in the control group. In contrast, rats treated chronically with the low dose of CP-154,526 spent significantly less time withdrawn in the tube than did control animals, suggesting anxiolytic activity of the compound during chronic treatment (Fig. 1A). As assessed using ex vivo autoradiography, chronic treatment with the low dose of CP-154,526 produced about a 25% blockade of CRF₁ receptors in the parietal cortex (see Fig. 4B). Thus, it is reasonable to believe that approximately the same degree of CRF₁ receptor blockade occurs in other brain regions, i.e., the amygdala which appears to mediate the anxiolytic effects of CRF₁ receptor antagonists (Swiergiel et al., 1993; Liebsch et al., 1995, 1999). Acute administration of CP-154,526, at least after the high dose, would also produce a substantial blockade of central CRF₁ receptors for a period of time; however, no effect in the defensive withdrawal model was observed. This suggests that other mechanisms in addition to CRF₁ receptor blockade may be involved in the anxiolytic-like effect observed during chronic treatment with the low dose of CP-154,526. One possible mechanism could be through a decrease in CRF synthesis (mRNA expression) as observed in Barrington's nucleus and the PVN of animals treated chronically with CP-154,526 (Fig. 7C). Neurons in Barrington's nucleus have been demonstrated to send projections to the LC, the origin of the main noradrenergic projections to the forebrain, some of which contain CRF (Valentino et al., 1996). Because local injection of CRF into the LC increases defensive withdrawal behavior (Butler et al., 1990) and a peptide antagonist attenuates it (Smagin et al. 1996), a decrease in CRFergic input to the LC from the Barrington nucleus may contribute to the anxiolytic-like effect observed in rats treated chronically with CP-154,526.

As previously reported, airpuff startle induced a robust and time-dependent increase in plasma ACTH and corticosterone (Engelmann et al., 1996). This effect was markedly attenuated by acute administration of the high, but not the low dose of CP-154,526. Previous studies have shown that a single dose of CP-154,526 or antalarmin attenuates CRF-induced elevations of plasma ACTH concentrations (Schulz et al., 1996; Webster et al., 1996) and foot shock-induced increase in plasma ACTH but not that of corticosterone (Deak et al., 1999; Owens and Nemeroff, 1999). In contrast to a previous study (Bornstein et al., 1998), no change in basal concentrations of ACTH or corticosterone was observed after chronic administration of CP-154,526 in the present study. The main difference between the two studies is the different route of administration used, i.e., two daily s.c. injections in the study by Bornstein et al. (1998) and osmotic minipumps in the present study.

During chronic treatment with the low dose of CP-154,526, application of an airpuff startle induced a similar time-dependent rise in plasma ACTH as seen in vehicle-treated animals. However, the corticosterone response returned to baseline levels faster than in control animals. Because this effect was not observed after a single dose of CP-154,526, it is probably not primarily mediated through CRF₁ receptor blockade but through some other mechanism developed during the chronic treatment. For example, corticosterone synthesis and release is subjected to a negative feedback at several different levels.

CRF₁ receptor binding densities in parietal cortex and cerebellum were dose dependently decreased by chronic CP-154,526 treatment. This effect could be washed off from cerebellar homogenates, suggesting that this effect is almost certainly due to residual drug bound to the receptors. In the present study, whether the decrease in CRF₁ receptor binding observed in the parietal cortex could be washed off was not studied. However, it is most likely that this effect in the parietal cortex is also a result of residual drug bound to CRF₁ receptors. Moreover, no change in gene expression of the CRF₁ receptor in the parietal cortex was found during the chronic CP-154,526 treatment, further supporting this notion. Nevertheless, these results suggest that, in the present chronic treatment regimen, CP-154,526 does penetrate into the brain.

Although chronic administration of CP-154,526 produced a substantial blockade of central CRF₁ receptors, no change in CRF₁ receptor mRNA expression was observed in the parietal cortex, basolateral amygdala, or the cerebellum. This is in contrast to the general pharmacological dogma that chronic blockade of a neurotransmitter receptor may produce a compensatory up-regulation in the synthesis of the receptor. A prolonged increase in CRF concentrations produced by immobilization stress has previously been shown to produce a down-regulation of pituitary CRF₁ receptors (Hauger et al., 1988). Interestingly, no change in CRF₁ receptor binding sites in several brain regions were observed after the same chronic stressor, despite the findings that chronic stress increases CRF content, and presumably release, in extrahypothalamic regions (Chappell et al., 1986). This suggests that CRF₁ receptors in the anterior pituitary are subjected to a negative feedback mechanism by CRF. In contrast, our finding that chronic treatment with a CRF₁ receptor antagonist does not change the synthesis of CRF₁ receptors in the parietal cortex, basolateral amygdala, or the cerebellum suggest that the synthesis of extrahypothalamic CRF₁ receptors are not subjected to a negative feedback mechanism by CRF. This lack of a compensatory up-regulation of CRF receptors after chronic blockade may contribute to the lack of tolerance to the anxiolytic effects demonstrated by CP-154,526. However, it should be noted that down-regulation of cell surface receptors by drug administration can occur in the absence of changes in mRNA expression (e.g., receptor internalization).

CRF mRNA expression decreased in the PVN and Barrington's nucleus after chronic administration of CP-154,526. CRF synthesis in the PVN appears to be subjected to a positive feedback mechanism through CRF₁ receptors (Imaki et al., 1996). Thus, it is possible that the observed decrease in CRF mRNA expression in the PVN in animals treated chronically with CP-154,526 may be due to a chronic blockade of this feedback mechanism. Whether a similar positive feedback mechanism through CRF₁ receptors also exists for CRF synthesis in the Barrington's nucleus is not known. Post-mortem studies have revealed increased CRF peptide and mRNA levels in the PVN in depressed patients (Raadsheer et al., 1994, 1995) and have been postulated to play a role in the hyperactivity of the HPA axis observed in many depressed patients. In view of the possible use of CRF₁ receptor antagonists in the treatment of depression, the present finding that a 14-day administration with CP-154,526 decreases CRF mRNA expression in the PVN is of considerable interest.

CRF₂ receptor binding or mRNA expression was not changed by chronic administration of CP-154,526 suggesting that chronic treatment with a selective CRF₁ receptor antagonist does not modify CRF₂ receptor functioning (Owens and Nemeroff, 1999; Smart et al., 1999). Similarly, urocortin mRNA expression in the Edinger-Westphal nucleus was not altered by CP-154,526 administration. Although CP-154,526 lacks appreciable affinity for the CRF₂ receptor, until recently it has been suggested that the urocortin-CRF₂ receptor system might represent a parallel stress-regulating system similar to the CRF-CRF₁ receptor system and that there might be considerable cross talk between them. The lack of chronic CP-154,526 administration to alter this system suggests that the above hypothesis may not be true. Indeed, we have recently observed evidence that these two systems might represent "antiparallel" systems (Skelton et al., 2000).

Although CP-154,526 is a potent CRF₁ antagonist, it is of interest to note that, in this study and others, plasma ACTH and corticosterone concentrations can still increase in response to stress and that basal concentrations are not greatly affected. Therefore, any concerns regarding potential serious adverse effects such as adrenal insufficiency, which might hinder clinical development, appear to be minimal at present.