The Newly Developed CRF1-Receptor Antagonists, NGD 98-2 and NGD 9002, Suppress Acute Stress-Induced Stimulation of Colonic Motor Function and Visceral Hypersensitivity in Rats

Corticotropin releasing factor receptor 1 (CRF1) is the key receptor that mediates stress-related body responses. However to date there are no CRF1 antagonists that have shown clinical efficacy in stress-related diseases. We investigated the inhibitory effects of a new generation, topology 2 selective CRF1 antagonists, NGD 98-2 and NGD 9002 on exogenous and endogenous CRF-induced stimulation of colonic function and visceral hypersensitivity to colorectal distension (CRD) in conscious rats. CRF1 antagonists or vehicle were administered orogastrically (og) or subcutaneously (sc) before either intracerebroventricular (icv) or intraperitoneal (ip) injection of CRF (10 µg/kg), exposure to water avoidance stress (WAS, 60 min) or repeated CRD (60 mmHg twice, 10 min on/off at a 30 min interval). Fecal pellet output (FPO), diarrhea and visceromotor responses were monitored. In vehicle (og)-pretreated rats, icv CRF stimulated FPO and induced diarrhea in >50% of rats. NGD 98-2 or NGD 9002 (3, 10 and 30 mg/kg, og) reduced the CRF-induced FPO response with an inhibitory IC50 of 15.7 and 4.3 mg/kg respectively. At the highest dose, og NGD 98-2 or NGD 9002 blocked icv CRF-induced FPO by 67–87% and decreased WAS-induced-FPO by 23–53%. When administered sc, NGD 98-2 or NGD 9002 (30 mg/kg) inhibited icv and ip CRF-induced-FPO. The antagonists also prevented the development of nociceptive hyper-responsivity to repeated CRD. These data demonstrate that topology 2 CRF1 antagonists, NGD 98-2 and NGD 9002, administered orally, prevented icv CRF-induced colonic secretomotor stimulation, reduced acute WAS-induced defecation and blocked the induction of visceral sensitization to repeated CRD.


Introduction
Corticotropin releasing factor (CRF), a 41-amino acid peptide originally isolated from ovine brain extract, is the principal mediator of the hypothalamic-pituitary-adrenal (HPA) stressresponse [1,2] CRF exerts its biological functions by activating two classes of B subfamily G-protein coupled receptors, CRF 1 and CRF 2 receptors [3]. Activation of brain CRF 1 signaling by CRF peptides plays a pivotal role in the behavioral, endocrine, immune, autonomic, and visceral responses to stress [2,[4][5][6]. One of the bodily systems susceptible to stress and stress-related peptides is the gastrointestinal tract [7]. Specifically, acute stressors and CRF injected into the brain or the periphery induces a rapid onset stimulation of colonic motor function in rodents, a response that is largely mediated by activating CRF 1 receptors in both the brain and the colon and reproducing symptoms of irritable bowel syndrome (IBS) with diarrhea (IBS-D) [8,9].
Preclinical and early clinical studies support the possibilities that pharmacological interventions targeting CRF 1 signaling may have potential therapeutic benefits in alleviating stress sensitive disorders [10,11]. For instance, the peptide CRF receptor antagonist, a-CRF 9-41 , injected into the circulation alleviates symptoms in a subclass of IBS patients [12]. As peptide compounds are less desirable in drug development, non-peptide small molecule CRF receptor antagonists are being developed to treat anxiety, depression, alcoholism, drug relapse and stress-related gastrointestinal diseases [10,[13][14][15]. Progress in the therapeutic use of nonpeptide CRF 1 antagonists, however, has been slow and largely disappointing due in part to the lack of consistency in their efficacy. For instance, chronic administration of a selective CRF 1 antagonist, R121919/NBI 30775, showed anxiolytic and antidepressant effects in the first open-label clinical study in patients with major depressive episodes [16]. NBI-34041 showed efficacy against the Trier social stress-induced endocrine response in placebo-controlled phase I and II clinical trials performed in healthy subjects [11]. There is also preliminary evidence that R317573 exerts anxiolytic effects in healthy subjects subjected to 7.5% carbon monoxide inhalation, an experimental model of anxiety [17]. Similarly, in a recent randomized, double-blind, placebo-controlled study, the selective CRF 1 antagonist GSK-GW876008 decreased brain regional activity associated with the emotional-arousal network during expectation of abdominal pain in IBS patients [14]. On the other hand, the CRF 1 antagonists, CP-316,311, showed no effect against depression in a 6-week randomized, placebo-controlled trial [18] and pexacerfont did not demonstrate efficacy compared to placebo for the treatment of generalized anxiety disorders in a multi-center clinical trial [19]. With regard to IBS, a double blind placebo-controlled clinical report showed the lack of effect of the CRF 1 selective antagonist BMS-562086 in ameliorating gastrointestinal symptoms in IBS-D patients [15].
Enthusiasm for the first generation of selective non-peptide CRF 1 antagonists, including CP-154,526 [20] and SSR125543A [21] was dampened by their pharmacokinetic properties. Overall the CRF 1 antagonists, with demonstrated high selectivity and potency in in vitro biological tests and preclinical assays, were highly lipophilic and hence less attractive for therapeutic use due to the potential risk of elevated tissue accumulation and prolonged half life [22,23]. Furthermore, the improvements in decreasing lipophilicity are not necessarily translated to higher oral bioavailability. Thus, to date there are very few CRF 1 receptor antagonists with high oral bioavailability and desirable pharmacokinetic profile.
In this study, we examined the antagonist action of NGD 98-2 and NGD 9002 on CRF-induced IBS-D-like symptoms, namely altered colonic motor function and visceral nociceptive hyperresponsiveness to colorectal distention (CRD) in conscious rats. We first delineated the doses at which orogastric (og) and subcutaneous (sc) administration of these compounds will antagonize CRF injected icv or intraperitoneally (ip)-induced stimulation of colonic propulsive motor function and diarrhea [26,27]. We then used the maximal effective oral dose of NGD 98-2 and NGD 9002 to assess whether this will counteract defecation induced by water avoidance stress (WAS) and the development of visceral hypersensitivity induced by repeated tonic CRD in rats known to involve activation of CRF 1 signaling [27][28][29].

Animals
Adult male Sprague-Dawley rats (Harlan, San Diego, California, USA) weighing 280-320 g were housed in group cages with free access to Purina rat chow and tap water. Animals were quarantined under controlled conditions of illumination (12 h light/dark cycle; lights on 06:00 h), temperature, and humidity for at least one week. Experiments started between 9 am and 10 am in non-fasted rats unless otherwise stated. Experimental protocols were approved by the Animal Care Committee of the Veteran Affairs Greater Los Angeles Healthcare System (#06-069-02) and the UCLA Animal Research Committee UCLA (ARC #2002-042).  [24]. For oral preparation, a day before the experiment, compounds were sonicated and suspended in 0.5% methylcellulose in distilled water with 0.1% triacetin (Sigma-Aldrich Co., St. Louis, MO) and placed on a magnetic stir plate overnight. For sc injection, NGD 98-2 and NGD 9002 were dissolved in dimethyl sulfonic acid (DMSO, Sigma-Aldrich):Tween-80:saline (1:1:8 ratio). Rat/human CRF and astressin (Clayton Foundation Laboratories for Peptide Biology, Salk Institute, La Jolla, CA, USA) were kept at -80 o C in powder form and diluted in saline just before administration. The selective non-peptide CRF 1 antagonist, CP-154,526 (Pfizer, Groton, CT, USA) [30] was diluted in a DMSO:Tween-80-:saline (1:1:8 ratio) solution as described before [31]. The pH of compound solutions at different concentrations was measured and respective vehicles were adjusted to be at the same pH.

Substance Administration
The volume of administration varied with the route of delivery: og, 5 or 10 ml/kg; sc, 1.5 ml/kg; ip, 1.0 ml/kg; icv, 10 ml/rat. The og gavage was performed using a stainless steel tubing (Cadence, Inc. Staunton, VA) in lightly hand-restrained rats and the sc injection was made into the loose skin of the back over the shoulders.
The icv injections were performed as in our previous studies [32]. Conscious lightly restrained rats with chronic icv cannula were injected through a 28 ga cannula (Plastics One Inc., Roanoke, VA, USA), 1 mm longer than the guide cannula. The injection cannula was connected to a 50 ml Hamilton syringe by a PE-50 tubing (Intramedic Polyethylene Tubing, Clay Adams, Sparks, MD, USA) filled with distilled water. A small air bubble (1 ml) was drawn at the distal end of the PE-50 tubing to separate the injected solution from the water and for visual monitoring of the icv injection which was performed slowly over a 60-sec period. At the end of experiments, animals were euthanized with sodium pentobarbital overdose followed with bilateral thoracotomy. In icv cannulated rats, the correct location of the cannula into the lateral ventricles was assessed by injecting icv 0.1% toluidine blue (10 ml) and the visualization of dye on the walls of lateral ventricles.
The regimens of compound administration were as follow: the sc injection of NGD 98-2 was performed 60 min before icv or ip CRF ( Fig. 2A and B) and og administration, 180 min before icv CRF (Fig. 3) or WAS. Oro-gastric or sc NGD 9002 was given 60 min before icv (Fig. 2 A) or ip CRF (Fig. 4) or WAS (Fig. 5). For CRD-induced visceral nociceptive responsivity, both NGD 98-2 and NGD 9002 were given og 40 min prior to the 1st CRD (Fig 6  and 7). These dosing regimens were based on our previous report showing that NGD 98-2 given orally 180 min before icv CRF prevents the CRF-induced increased locomotor activity or restraint stress-induced elevation of plasma ACTH levels [24] as well as pilot studies to assess optimal inhibitory effect on icv CRFinduced defecation.

Surgeries
4.1. Intracerebroventricular cannulation. ICV cannulation was performed as previously reported [32]. Rats were anesthetized with an ip injection of a mixture of ketamine (75 mg/kg; Fort Dodge Laboratories, Fort Dodge, IA, USA) and xylazine (5 mg/kg; Mobay Corporation, Shawnee, KS, USA). A chronic guide cannula (22 ga, Plastic One Products) was implanted into the right lateral brain ventricle according to coordinates from Paxinos and Watson [33] (mm from bregma: antero-posterior, 20.8; lateral, 21.5; dorsoventral, 23.5). The guide cannula was maintained in place by dental cement anchored by four stainless steel jewelry screws fixed to the skull. The cannula was capped with a dummy cannula for protection. Following icv cannulation, rats were housed singly and allowed to recover for at least 7-10 days during which they were trained to the experimental conditions of icv injection by handling them for 5 min per day for at least 3 consecutive days.
4.2. Abdominal muscle electrodes implantation. The implantation of electrodes was performed in rats anesthetized with an ip injection of a mixture of ketamine (75 mg/kg) and xylazine (5 mg/kg) as detailed in previous studies [34]. Under aseptic conditions, a 2-3 cm incision was made near the lower abdomen along the mid-line. The abdominal muscle layers and the peritoneum were opened and a group of 3 teflon coated silver electrodes were fixed in triangular pattern (5 mm apart) on the left side of the abdominal muscle 1-2 cm lateral to the mid line. The other end of the electrodes was fitted in a small plastic casing, which served as a jack to connect the electrodes to the recording device. The connecting side of the jack was then exteriorized on the right side of the flank (2-3 cm lateral to the mid-line) through a small (3 mm diameter) opening across the abdominal wall and the skin. The base of the jack was secured in place by suturing it onto the peritoneal side of the abdominal muscle. Rats were allowed to recover from surgery for 10-15 days.

Procedures
5.1. Water avoidance stress. The WAS was performed as described before [35] by placing the rat on a small cubic pedestal (8 cm height, 6 cm wide) positioned in the center of a plastic cage filled with room temperature water up to 7 cm height of the pedestal. To avoid contact with the water, the rat remains on the pedestal platform for the experimental period.

Measurements of abdominal contractions to
colorectal distention. Rats chronically fitted with electrodes on abdominal muscles were trained to the experimental conditions by placing them in Bollman cages for 2-3 h/day for at least 3 consecutive days before the study. On the day of the experiment, rats were briefly anaesthetized with isoflurane (3% in O 2 ), and a 6 cm long plastic balloon tied around an Intramedic PE-100 tubing (Becton Dickinson, Franklin Lakes, NJ, USA) was inserted intra-anally with the distal end positioned 1 cm proximal to the anus. The CRD in awake rats results in contractions of the abdominal and hind limb musculature and this visceromotor response (VMR) is validated as a quantitative measure of visceral hypersensitivity [36]. The protocol of CRD was similar to our previous studies showing the induction of visceral hypersensitivity [29,37]. CRD entailed inflating the CRD balloon with a barostat (Distender II, J&J Inc., Toronto, ON, Canada) at 60 mmHg twice for 10 min with a 30 min rest interval. The VMR to CRD was recorded as electromyography (EMG) signals and acquired using a Micro1401 A/D interface (Cambridge Electronic Design, Ltd, Cambridge, UK) connected to a Pentium IV class computer running Spike 2 data acquisition software. EMG signals were amplified, filtered (x10000, 300-5000 Hz), digitized, and rectified as detailed previously [38]. The basal area under the curve (AUC) of abdominal contractions recorded from EMG was calculated as the area under the rectified EMG signal trace for the 10 min period immediately preceding the first 10 min CRD. The AUC values of the EMG during the first and second distensions were computed and basal AUC subtracted to obtain the net AUC in response to CRD as described [37]. AUC of contractions in response to each 10 min distention was compared to each other. From these values, the percent change in AUC [DAUC (%)] was calculated by taking the difference between the 1 st and 2 nd distention AUC responses and dividing by the 1 st AUC and multiplying by 100.
5.3. Colonic motor function measurements. Defecation was monitored as described previously [31] by counting the number of fecal pellets excreted every 15 min for up to 2 h. The incidence of diarrhea was assessed for the 2-h period post CRF injection as percent of rats that developed one loose-watery stool or more from the total number of treated rats.

Experimental Protocols
All the experiments started between 9-10 am and were conducted in non-fasted conscious rats trained to the experimental conditions. Unless otherwise stated, antagonists were administered following a regimen of 60 min pretreatment period. In a separate experiment, rats (n = 8/group) were injected sc with either vehicle, NGD 98-2 (3, 10 or 30 mg/kg) or CP 154,526 (20 mg/kg) followed with ip saline or CRF (10 mg/kg) and fecal pellet and diarrhea responses were monitored for 60 min post ip injection. We previously reported the efficacy of CP 154,526 (20 mg/kg) against ip CRF (10 mg/kg) -induced stimulation of defecation in rats [39] and this CRF 1 antagonist at such a dose was used as a positive control.
For intraperitoneal CRF, in a separate set of experiments, NGD 9002 (3, 10 or 30 mg/kg) or its vehicle was administered orogastrically before ip injection of CRF (10 mg/kg) or saline (n = 8/group). The defecation response was monitored for 60-120 min. The peptide CRF antagonist astressin injected ip (30 mg/kg) immediately before ip CRF (10 mg/kg) was used as positive control. Astressin under these conditions is known to block ip CRF-mediated colonic response [31].
6.3. Effect of orally-administered NGD 98-2 or NGD 9002 on repeated colorectal distention-induced visceral nociceptive sensitization. After rectocolonic positioning of the balloon and recovery from the short anesthesia, rats were administered og with NGD 9002 (10 or 30 mg/kg, n = 8 or 17/ group) or NGD 98-2 (30 mg/kg, n = 10) or their vehicle (0.5% methylcellulose/0.1% triacetin, n = 22) and placed in Bollman cages. After a 30-min stabilization period and 10-min baseline recording, all groups were submitted to isobaric CRD (60 mm Hg for 10 min twice, with a 30 min rest interval) using a barostat. The abdominal contraction responses to the 1 st and 2 nd distention were compared within and between groups.

Statistical Analysis
Values are expressed as mean and standard error of the mean or mean % difference. The FPO response to different treatments or doses was compared using a one-way analysis of variance (ANOVA). Comparison between groups on diarrhea incidence (%) was made using Fisher's exact test. Two way ANOVA was used, to compare the two factor effects of NGD 9002 and NGD 98-2 (0, 3, 10, 30 mg/kg) on CRF or WAS-induced FPO responses. Similarly, where appropriate, IC 50 of NGD compounds was calculated using Graphpad Prism Software (GraphPad Software, Inc. La Jolla, CA). The AUC of abdominal contraction response to the 1 st vs 2 nd distension within a group were compared using before and after paired t-test or one-way repeated measures ANOVA and percent differences using t-test or ANOVA. For pair wise multiple comparisons, Student-Newman-Keuls Method or Fisher LSD test was used. P,0.05 was considered as a significant difference.

Orogastric Administration of NGD 98-2 or NGD 9002 Decreases Repeated Tonic Colorectal Distention-induced Visceral Nociceptive Hyper-responsivity
Two tonic colorectal distensions (60 mmHg for 10 min with a 30 min interval) increased the VMR monitored by increased EMG activity (Fig. 6A-B). A representative trace of the abdominal contraction response to tonic distensions in vehicle or NGD 98-2 (30 mg/kg, og) pretreated rats (-40 min before 1 st CRD) is shown in Fig. 6A and 6B respectively. The mean percent difference between the 1 st and 2 nd responses in the vehicle-pretreated rats was significantly higher than the mean percent difference in NGD 98-2-pretreated rats (60.4633.4% vs 238.2619.2%, p,0.05, n = 8-10/group, Fig. 6C). In addition, while 5 out of 8 vehiclepretreated rats (62.5%) had at least a 10% higher response to the 2 nd distention when compared to the 1 st , only 1 out of 10 rats (10%) pretreated with NGD 98-2 (30 mg/kg) had a 10% or higher response to the 2 nd CRD when compared to the 1 st .
In a separate study where the effect of NGD 9002 was evaluated in a larger number of rats (n = 8-22/group), repeated CRD resulted in a significantly increased 2 nd CRD response (p,0.05, Fig. 7A & 7C). As in the NGD 98-2 experiment, the mean percent difference between the 1 st and 2 nd responses in the vehiclepretreated rats (n = 22) was significantly higher than the mean percent difference of NGD 9002 at 30 mg/kg group (n = 17) (Fig. 7C). In addition, while 9 out of 22 vehicle pre-treated rats (41%) had at least a 10% higher response to the 2 nd CRD, when compared to the 1 st , none of the 8 rats pretreated with NGD 9002 at 10 mg/kg and only 2 out of 17 of the NGD 9002 at 30 mg/kg had a 10% or higher response to the 2 nd CRD compared with the 1 st (Fig. 7A-C).

Discussion
NGD 98-2 and NGD 9002 were identified as pre-clinical development candidates from drug discovery efforts spanning several series of topology 2 CRF 1 antagonists. These compounds display high affinity (less than 10 nM) to both human and rat CRF 1 receptors (Ki = 1.0 and 9.8 nM for NGD 98-2 and Ki = 2.3 and 4.3 nM for NGD 9002 respectively) and demonstrate oral efficacy [24,25]. The present study shows that NGD 98-2 and NGD 9002, administered orogastrically, dampen centrally or peripherally injected CRF-or acute WAS-induced colonic motor stimulation (defecation) and repeated tonic CRD-induced visceral sensitization in rats. These data indicate the effectiveness of orally administered NGD 98-2 and NGD 9002 against exogenous CRF and endogenous CRF released by acute WAS [27,35]-induced IBS-D-like symptoms in rats.
CRF injected icv at a dose of 10 mg/kg produced a significant and reproducible increase in FPO and incidence of diarrhea in 50% of rats. This is consistent with previous reports showing that icv CRF-induced enhanced colonic secretory and motor function [27,28,40]. Orogastric administration of the selective CRF 1 antagonists, NGD 98-2 [24] or NGD 9002 [25] dose-dependently reduced the icv CRF-induced FPO. NGD 98-2 also dosedependently prevented the incidence of diarrhea induced by icv CRF while NGD 9002 showed a non-significant trend. Previous studies in rats showed that icv injection of the CRF 1 antagonist, NBI-27914 blocked icv CRF-induced defecation [28,41]. The present data provide the first evidence that an orally administered CRF 1 antagonist abolished icv CRF induced stimulation of both colonic propulsive motor function and diarrhea. This supports the efficacy of orally administered new generation CRF 1 antagonists against centrally mediated CRF stimulatory effects on colonic secretory-motor function. We have recently reported that NGD 98-2 crosses the blood brain barrier upon orogastric administration to block icv CRF mediated increased locomotor activity in rats [24]. The present data extend these findings to the gastrointestinal tract. In addition, the fact that og administration of NGD 9002 has an in vivo IC 50 value of 4.3 mg/kg compared to 15.7 mg/kg for NGD 98-2 in blocking icv CRF-induced defecation, while both NGDs have similar Ki to rat CRF 1 [24,25] suggests that NGD 9002 and NGD 98-2 may have ability to cross the blood brain barrier to block exogenous CRF action in the brain. This is of significance because most of the selective CRF 1 antagonists used in prior studies to address the role of CRF 1 in central CRF-induced colonic stimulation had to be administered through systemic or central injections due to their poor oral bioavailability [28,31,41]. The few CRF 1 antagonists that show oral bioavailability were tested with either peripheral injection of CRF [34] or mainly in the context of icv CRF-induced behavioral outcomes such as increased locomotion, addiction, depression and anxiety [10,22,24]. The single double-blind placebo controlled clinical trial that evaluated the effect of an orally-administered CRF 1 antagonist, BMS-562086 on gastrointestinal outcomes showed no improvement in colonic transit in IBS-D predominant patients [15]. The current study is the first to demonstrate the efficacy of orally administered selective non-peptide CRF 1 antagonists to alleviate central CRF-induced colonic stimulation in rats. Of note, most of the CRF 1 antagonists tested in prior studies had limitations for clinical use in part due to their high lipophilicity [22,42].
Next we tested whether systemic or orogastric administration of NGD 9002 and NGD 98-2 would influence the colonic response to peripherally injected CRF. Intraperitoneal injection of CRF is well-established to act through distinct mechanisms than those initiated in the brain by central injection of CRF [43]. CRF injected ip or locally induces a direct activation of colonic cholinergic myenteric neurons, stimulates colonic 5-HT release and activates mast cells [44,45]. NGD 98-2 injected sc at 30 mg/ kg unlike lower doses (10 or 3 mg/kg) reduced defecation induced by ip CRF. Additionally NGD 9002 given orally at 30 mg/kg reduced icv CRF-induced defecation and showed a non-significant trend to reduce the incidence of diarrhea induced by ip CRF. Under the same conditions, the peptide CRF 1 /CRF 2 antagonist astressin [46] injected ip at 30 mg/kg completely prevented both the defecation and diarrhea as previously reported [31]. Consistent with these findings, the CRF 1 antagonist NBI 35965 injected sc or og (10-20 mg/kg) reduced ip or intravenous CRF induced stimulation of distal colonic transit in rats [34]. The present data are also in line with reported preventive action of other CRF 1 antagonists such as CP-154,526 given sc against ip CRF or the selective CRF 1 agonist, cortagine,-induced stimulation of colonic motor function [31,47].
The effects of central and peripheral CRF on the colon have long been suggested to mimic those produced by acute stressors [48]. In particular, acute exposure to WAS produced a reliable and significant increase in FPO which we demonstrated previously to involve activation of CRF 1 receptors located in both the brain and the colon [6]. In the present study, orogastric administration of NGD 9002 and NGD 98-2, while having no effect on defecation in a non-stressed rat, effectively reduced the enhanced colonic motor response caused by WAS. To date, with the exception of one study that tested the efficacy of an orogastrically-administered CRF 1 antagonist, NBI 35965 on defecation induced by WAS [34], the overwhelming majority of studies on acute stress-induced colonic motor stimulation have used sc injections of the nonpeptide selective CRF 1 antagonist CP-154,524 [31,49]. The present study adds, to the orally-active selective CRF 1 antagonists, two novel compounds for use in preclinical and potential clinical studies on gut motor response to stress.
Lastly, while non-selective and selective CRF 1 antagonists have been well documented to prevent stress-or icv or ip CRF-induced visceral sensitization in rodents [34,47,[50][51][52][53], the effects of orallyactive CRF 1 antagonists against nociceptive hyper-responsivity to tonic noxious colonic stimulation have not been studied. Repeated tonic CRD-induced visceral hypersensitivity has been established in rats [29,37,54], mice [55], and humans [56,57]. In rats, we previously reported that such a hypersensitization is not associated with colonic lesion and is prevented by the ip injection of a selective CRF 2 receptor agonist, urocortin-2 [37], a sc injection of a selective CRF 1 antagonist antalarmin [29] or by oral administration of pregabalin, a ligand to the a 2 d subunit of the voltagegated calcium channel [58]. In the present study, repeated tonic CRD resulted in an enhanced VMR to the 2 nd distention compared to the 1 st , indicating the development of acute hypersensitivity to a noxious visceral stimuli. This response is prevented by oral-administration of NGD 98-2 and NGD 9002, indicative of a role of CRF 1 in the response. The demonstration of the involvement of CRF 1 in repeated mechanical CRD-induced nociceptive hyper-responsivity is in line with previous studies using sc antalarmin [29].
The exact site(s) and mechanism(s) of action of the orally-active NGD compounds cannot be determined from the present study. Tonic CRD is shown to activate ERK phosphorylation in the lumbosacral spinal segments in rats and the response is blunted by CRF 2 activation [37]. Spinal ERK phosphorylation modulates neuronal excitability and plays an important role in hyperalgesia after noxious somatic stimuli and inflammation [59] suggesting a possible sensitization occurring during repeated CRD as well. On the other hand, activation of peripheral CRF 1 receptors by the CRF 1 selective peptide agonist, cortagine, causes visceral hyperalgesia in rats [47] suggesting a possible peripheral action of orallyadministered NGD compounds to prevent repeated tonic CRDinduced sensitization. Taken together, given that both central and peripheral sensitizations are reported to occur during visceral hyperalgesia [60] and the fact that the orally-administered NGD 98-2 crosses the blood brain barrier with high brain exposure leading to CRF 1 receptor occupancy assessed by autoradiography [24], it is likely that the compounds block CRF 1 -mediated central and peripheral visceral hyper responsiveness.  In summary, significant progress has been made in the design and development of non-peptide CRF 1 receptor antagonists [22,61]. However, high lipophilic characteristics and potential toxicity have hampered their translational applications [22,42,61]. Thus, of the numerous non-peptide small molecule CRF 1 antagonists developed, to date very few have moved forward to clinical trials [13,61]. NGD 98-2 and NGD 9002 are topology 2 CRF 1 antagonists with high affinity to CRF 1 , high oral bioavailablity and low lypophilicity. Orally-administered NGD 98-2 and NGD 9002 effectively prevented icv CRF-induced activation of defecation in rats. The present data show also that the selective CRF 1 antagonists blunt WAS-induced colonic motor activation and prevented repeated CRD-induced visceral hypersensitivity. The data strengthen the concept that activation of CRF 1 signaling pathway plays a role in mediating acute stressrelated stimulation of colonic motor function as well as visceral sensitization to CRD in healthy rats. Although, several preclinical studies have shown, early on, the role of CRF 1 receptors in stressrelated colonic stimulation and suggested a therapeutic potential of CRF 1 antagonists against stress-related diarrhea and visceral hypersensitivity [7,40,50], the report that a CRF 1 antagonist lacks efficacy to influence intestinal transit and diarrhea in IBS patients [15], raises concern on the potential use of CRF 1 antagonists to alleviate symptoms in IBS patients. Several reasons including the testing of CRF 1 antagonists in preclinical studies mainly in an acute stress-context, while in clinical studies the state of acute or chronic stress of patients is not well determined; the presence of over 11 splice variants of CRF 1 receptors in humans [62]; the existence of differences in binding kinetics, association and dissociation rate and efficacy of CRF 1 antagonists used [61,63] may account for the discrepancy. There is also a lack of information, in previous clinical studies, on whether the antagonist regimens used is optimum to block both central and peripheral Rats were chronically implanted with abdominal electrodes and ,10 days later were pretreated og with vehicle or NGD 98-2 (30 mg/kg). After 30 min of habituation and 10 min basal recording, rats were submitted to the first CRD (10 minutes at 60 mm Hg) followed by a 30 min rest and a 2 nd 10 min distention at 60 mmHg. Values are mean and SEM of percent differences between the first and second responses to CRDs of 9-10 rats/group. *p,0.05 versus vehicle, t-test or ANOVA, Student-Newman-Keuls. doi:10.1371/journal.pone.0073749.g006 Figure 7. Oro-gastric (og) pretreatment with NGD 9002 blunts repeated tonic colorectal distention (CRD)-induced visceral sensitization response in conscious rats. Rats were chronically implanted with abdominal electrodes and ,10 days later were pretreated og with vehicle or NGD 9002 (0, 10 or 30 mg/kg). After 30 min of habituation and 10 min basal recording, rats were submitted to the 1 st CRD (10 minutes at 60 mm Hg) followed by a 30 min rest and a 2 nd 10 min distention at 60 mmHg. Representative trace of abdominal muscle electromyogram (EMG) of rats pretreated with vehicle (A) or NGD 9002 (B). Percent difference in the area under the curve of contraction (AUC) between the 1 st and the 2 nd dissentions in saline and NGD 9002 treated rats is shown in C. Values are mean and SEM of percent differences between the first and second responses to CRDs of 8-22 rats/group. *p,0.05 versus vehicle, t-test or ANOVA, Student-Newman-Keuls. doi:10.1371/journal.pone.0073749.g007 CRF 1 receptors. In this context, the development of new CRF 1 antagonists such as NGD 98-2 and NGD 9002 that have better oral bioavailability opens new venues to understand the potential role and mechanisms of CRF 1 receptors in stress-sensitive functional bowel disorders such as IBS.