Molecular Psychiatry (2002) 7, 474–483  2002 Nature Publishing Group All rights reserved 1359-4184/02 $25.00 www.nature.com/mp ORIGINAL RESEARCH ARTICLE Marked suppression of gastric ulcerogenesis and intestinal responses to stress by a novel class of drugs KE Gabry1, GP Chrousos2, KC Rice3, RM Mostafa4, E Sternberg1, AB Negrao1, EL Webster1, SM McCann5 and PW Gold1

1Clinical Neuroendocrinology Branch, National Institute of Mental Health, Intramural Research Program, National Institutes of Health, Bethesda, MD, USA; 2Pediatric and Reproductive Endocrinology Branch, National Institute of Child Health and Human Development, Intramural Research Program, National Institutes of Health, Bethesda, MD, USA; 3Laboratory of Medicinal Chemistry, National Institute of , Digestive and Kidney Diseases, Intramural Research Program, National Institutes of Health, Bethesda, MD, USA; 4Department of Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA, USA; 5Pennington Biomedical Research Laboratories, Louisiana State University, Baton Rouge, LA, USA

When exposed to prolonged stress, rats develop gastric ulceration, enhanced colon motility with depletion of its mucin content and signs of physiological and behavioral arousal. In this model, we tested whether (fluoxetine and ), ( and ) or the novel nonpeptide corticotropin-releasing hormone (CRH) type-1 recep- tor (CRH-R1) antagonist, antalarmin, modify these responses. , bupropion, diazepam and antalarmin all suppressed stress-induced gastric ulceration in male Sprague– Dawley rats exposed to four hours of plain immobilization. Antalarmin produced the most pronounced anti-ulcer effect and additionally suppressed the stress-induced colonic hyper- motility, mucin depletion, autonomic hyperarousal and struggling behavior. Intraperitoneal CRH administration reproduced the intestinal but not the gastric responses to stress while vagotomy antagonized the stress-induced gastric ulceration but not the intestinal responses. We conclude that brain CRH-R1 and vagal pathways are essential for gastric ulceration to occur in response to stress and that peripheral CRH-R1 mediates colonic hypermotility and mucin depletion in this model. Nonpeptide CRH-R1 antagonists may therefore be prophylactic against stress ulcer in the critically ill and therapeutic for other pathogenetically related gas- trointestinal disorders such as and . Molecular Psychiatry (2002) 7, 474–483. doi:10.1038/sj.mp.4001031 Keywords: functional gastrointestinal disorders; corticotropin-releasing; type 1 receptor; peptic ulcer; irritable bowel; nonpeptide; antagonist

Introduction appear to be interrelated and to be associated with affective and disorders.5,6 Moreover, early life In addition to predisposing to mental illness and alter- adversity is associated with more prevalence of mood ing the endocrine milieu of the organism, stress fre- and anxiety disorders7,8 as well as predisposition to quently precipitates marked disturbance of visceral gastric ulceration9 and IBS10 later in adulthood. In spite functions. The gastrointestinal tract is one of the most of the fact that these stress-related gastrointestinal (GI) vulnerable viscera to aversive psychological stimuli.1 disorders constitute the most common presentation in Individual variability in responding to stress ranges gastroenterology and primary care clinics,11 our ability from nausea and vomiting to abdominal pain and to treat them is still limited as their underlying patho- diarrhea. Severe stress of life-threatening conditions, physiologic mechanisms remain to be deconvoluted.12 eg shock, head trauma, myocardial infarction, may be Since its identification as a classic component of the associated with stress ulcer (SU) development; a poten- stress response by Selye (1936)1 and establishing it as tially fatal condition whose pathophysiology is still a model of somatic consequences of stress by Brodie,13 obscure.2 Less severe, prolonged stress has been impli- stress-induced gastric ulceration has been widely used cated in the pathogenesis of peptic ulcer disease3 and in studying visceral responses to stress. This includes irritable bowel syndrome (IBS).4 These disorders exploring protective mechanisms against stress by measuring their potential for preventing stress ulcer- ation. The same model is also useful as a model of non- Correspondence: KE Habib, MD, PhD, Clinical Neuroendocrinol- ulcer dyspepsia and peptic ulcer, particularly in the ogy branch, IRP, NIMH, NIH Bldg 10, Rm 2D-46, Bethesda, MD 3,14 20892-11284, USA. E-mail: habibȰcodon.nih.gov context of their biopsychosocial nature, in man. Received 22 May 2001; revised 24 September 2001; accepted 13 Many researchers in the area of gastrointestinal drugs October 2001 screen their new compounds in the stress-induced gas- Marked suppression of gastric ulcerogenesis KE Gabry et al 475 tric ulcer model.15 Even though for more than a decade in an animal model that simultaneously reproduces the focus of the etiology of peptic ulcer shifted to hel- features of stress ulcer13 and IBS.4 Antalarmin, N-butyl- icobacter infection, many clinicians now believe that N-ethyl-[2,5,6-trimethyl-7-(2,4,6 trimethylphenyl)-7H- the role of Helicobacter pylori in the causation of pep- pyrrolo[2,3-d]pyrimidin-4-yl] amine, is a selective tic ulcer is no more than a partial one.3,16 Other equally CRH-R1 antagonist that penetrates the blood–brain bar- important factors include genetic makeup and vulner- rier and suppresses neuroendocrine, autonomic and ability to stress.17–19 Moreover, a recent study has behavioral responses to stress in primates.36 Other pyr- negated the involvement of H. pylori in stress ulcer rolopyrimidine compounds also exert bleeding in critically ill patients.2 effects42 and suppress colon hypermotility43 in rodents. The pivotal role of the brain in the pathogenesis of The major advantage of nonpeptide antagonists is their stress-induced gastric mucosal damage has been well potential clinical utility in humans.44 As opposed to established. Increased vulnerability to depression20 antagonists that require direct injections into and anxiety21 in experimental animals is paralleled the brain to access central CRH receptors, antalarmin with ulcer development. The same holds true for can neutralize CRH-R1 whether in the brain or in the humans.5,22 Moreover, classic antidepressants23,24 and periphery after oral administration.36 anxiolytics25,26 significantly reduce stress ulcer forma- Here we first conducted a dose-response study to tion, perhaps to a greater extent than that seen with examine the effects of antalarmin on stress-induced traditional therapies such as cimetidine and antacids.27 gastric ulceration in adult male Sprague–Dawley rats. Indeed, induction of gastric lesions in rodents by socio- To determine whether antalarmin’s effects are due to psychological stress is useful as an experimental model moderating the activity of the stress neurocircuitry, we to assess the anxiolytic activity of novel drugs and the tested it against other known antidepressants anxiolytic potency of a drug may be predicted by its (bupropion and fluoxetine) and anxiolytics (diazepam anti-ulcer effect.28 and buspirone). We chose the latter drugs for their non- While the role of corticotropin-releasing hormone classic effects and lesser anticholinergic (CRH) in mediating the stress-induced alteration of gas- activity.45 In addition to scoring gastric ulceration, we trointestinal secretory and motor responses is fairly evaluated colon specimens for mucin depletion and established,29–31 its role in the pathogenesis of stress counted the stool pellets as an index of enhanced colon ulcer has been debatable.32,33 CRH is a 41-amino acid motility in response to stress.46 Stress-induced peptide that was first isolated as the principal hypo- enhancement of intestinal transit in the rat has been thalamic hormone that mobilizes the pituitary-adrenal proposed as an animal model for irritable bowel syn- responses to stress34 and subsequently identified as a drome.4 Moreover, defecation scores have been long key player in the behavioral and autonomic responses known as valid representatives of fearfulness and anxi- to stress in rodents35 and nonhuman primates.36 Sev- ety in rodents.47 Plasma catecholamines and struggling eral studies have shown that both stress and central behavior were included as indices of autonomic and CRH administration decrease gastric acid secretion, behavioral arousal in this model. gastric emptying, and small bowel transit while mark- edly enhancing large bowel transit.29,37 Moreover, Experimental procedures immobilization stress of rats causes increased colonic Drugs mucin and prostaglandin E2 secretion, increased Antalarmin, N-butyl-N-ethyl-[2,5,6-trimethyl-7-(2,4,6 tri- colonic mucosal levels of cyclooxygenase-2 mRNA, methylphenyl)-7H-pyrrolo[2,3-d]pyrimidin-4-yl] amine, and degranulation of colonic mast cells, which are was synthesized in our laboratory as previously antagonized by alpha-helical CRH 9–41, a non specific described.48 For intraperitoneal (IP) injections, antalar- peptide CRH antagonist.31 Central administration of min was dissolved in cremophor/ ethanol/ saline, peptide CRH receptor antagonists prevents the gastro- 5:5:90 by volume, and injected (0.5 ml per animal) intestinal secretory and motor responses elicited by immediately after preparation. Diazepam (Roche Lab, either stress or exogenous CRH in a dose-dependent Nutley, NJ, USA) was suspended in 0.5% carboxyme- fashion.30,38 thylcellulose and injected intramuscularly.49 Buspi- The clinical development of nonpeptide antagonists rone (Sigma, St Louis, MO, USA), bupropion (Sigma) which cross the blood–brain barrier to reset brain CRH- and fluoxetine (Tocris Cookson, Ballwin, MO, USA) R1 for the treatment of anxiety40 and melancholic were dissolved in saline, and all were injected as 0.5 depression41 is currently the focus of extensive ml per animal IP. For the dose response experiment, research. We have recently reported that oral adminis- four animal groups were included into which antalar- tration of antalarmin, a member of this novel family of min was injected at doses of 0, 15, 30 and 45 mg kg−1. drugs, attenuates neuroendocrine and behavioral For the comparative study, six treatment groups were responses to stress in primates.36 Due to the significant included: saline (0.5 ml rat−1), antalarmin (20 mg kg−1), overlap in the neural circuitry that mediates mental diazepam (5 mg kg−1),50 buspirone (4 mg kg−1),51 bupro- and gastrointestinal responses to stress,39 we hypothes- pion (30 mg kg),52 and fluoxetine (10 mg kg−1).53 CRH ized that CRH-R1 plays a role in the pathogenesis of (human/rat CRF; Peninsula Lab, San Carlos, CA, USA) gastrointestinal injury and/or dysfunction during was dissolved in saline and injected IP (0.5 ml rat−1)at stress. The recent availability of non-peptide CRH-R1 a single dose of 5 ␮gkg−1 in freely moving animals 4 h antagonists enabled us to examine the role of CRH-R1 prior to their killing.

Molecular Psychiatry Marked suppression of gastric ulcerogenesis KE Gabry et al 476 Animal experiments mucosa in an area containing 10 parallel aligned crypts Adult male (200–250 g) Sprague–Dawley rats (Taconic were examined for the percentage of mucin-positive Farms, Germantown, NY, USA) were housed under goblet cells to the total number of colonocytes lining controlled conditions on a 12-h light/12-h dark cycle the superficial half of the crypts epithelium as for an average of 1 week after arrival to the animal described previously.46 Stomachs were washed in ice facility. Rats were housed 3 to a cage and provided cold saline and opened along the greater curvature to with food and water ad libitum except for the last 24 h wash the interior with phosphate buffered saline for preceding the experiment, when they were individu- scoring gastric ulcers as the sum of their cumulative ally housed in mesh bottom cages and deprived of length in mm per stomach with the aid of a magnifying food, but not water. In our experience, fasted rats lens and a micrometer.56 All samples were coded and develop stress ulcers more readily and consistently blindly scored. Because of the protective effect of food than non-fasted subjects, particularly if stress is against SU, stomachs containing food remnants were applied in the afternoon. All experiments were started excluded regardless of their ulcer scores. between the hours of 12:00 and 13:00 to minimize the influence of circadian rhythm on the parameters under Statistical analysis investigation. Animal procedures were approved by ANOVA followed by post-hoc (Dunnett’s Multiple the National Institute of Child Health and Human Comparison) test was used to study the effects of differ- Development Animal Care and Use Committee and ent treatments on plasma catecholamines, stress ulcer, conducted with strict adherence to the National Insti- intestinal mucin, struggling and defecation scores. tutes of Health Guidelines of Animal Use in Intra- ANOVA post-test for linear trend was used to express mural Research. the results of the dose-response study. On all Animals groups received half the dose of their drug occasions, a P level was considered statistically sig- intraperitoneally (IP) twice at −2 h and +2 h of starting nificant at two-tailed values of 0.05 or less. Statistical immobilization in the dose response experiment. A analysis and graphic presentations were aided by  pilot experiment showed better anti-ulcer responses if Prism version 3.0 for Macintosh (GraphPad , Ithaca, the drugs were administered in full doses twice at −24 NY, USA). and −2 h prior to immobilization (data not shown), and therefore this latter regimen was used for the compara- Results tive study. Subdiaphragmatic or sham vagotomy were Figures 1a and 1b show the dose response effects of performed 4 days prior to the experiment as pre- antalarmin on stress-induced gastric ulceration (SU). viously described.54,55 Antalarmin reduced SU scores in a dose-dependent Immobilization stress lasted for 4 h and was perfor- fashion, namely from 12.63 ± 1.711 at 0 mg kg−1 to med essentially as described by Brodie13 with slight 2.000 ± 1.323 at 45 mg kg−1. ANOVA post-test for linear modification. Briefly, awake rats were placed in the trend showed a slope of −1.788 (R2 = 0.539, P Ͻ0.001). prone position on specifically designed stainless steel In freely moving animals, none of the drugs signifi- plates equipped with semi-circular rings that prevent cantly caused any significant change in gastric ulcer- the animal from turning around to access its taped ation, colonic mucin or defecation scores as compared lower trunk and feet. The animal’s struggling behavior to saline treatment (data not shown), and therefore the was blindly scored for 5 min at the summits of immobi- effects of different drugs on the parameters under lization hours 1, 2 and 3, according to a previously investigation during stress are only presented. validated scale that scores one point for every act of In saline-treated animals, exposure to immobiliz- jaw movement, body rolling, tail flipping or vocaliz- ation stress precipitated gastric ulceration (from 0.125 ation, with simultaneous acts counted as the sum of ± 0.125 mm to 15.88 ± 2.463 mm, t = 5.41, P Ͻ 0.001), their individual components.49 The three time point colonic mucin depletion (from 13.4 ± 2.73 to 4.625 ± results were averaged and expressed as the struggling 0.944, t = 3.03, P Ͻ 0.01), and elevation of plasma epi- score per hour. Colonic motility was estimated by nephrine (from 8.40 ± 2.22 to 21.11 ± 2.55 ng ml−1, t = counting the number of fecal pellets expelled during 3.76, P Ͻ 0.01 ) and norepinephrine (from 5.45 ± 1.278 the immobilization period.29,46 to 10.23 ± 0.849 ng ml−1, t = 3.243, P Ͻ 0.01) as com- After the elapse of 4 h, rats were released from their pared to freely moving animals. restrainers and immediately guillotined. Trunk blood There were no significant differences between saline was collected in ice-chilled tubes and spun in a cooling treatment and any of the vehicles used for drug dissol- centrifuge at 1000 × g to obtain plasma for the assay of ution during stress on stress ulceration, defecation epinephrine and norepinephrine by high performance scores, struggling, colonic mucin, plasma epinephrine liquid chromatography coupled with electrochemical or plasma norepinephrine (Table 1), and therefore we detection as previously described.36 A sagittal lapar- used saline treatment as the control group. otomy was performed to collect stomachs and colonic The ability of antalarmin and other psychotropic segments 10 cm proximal to the anal orifice. The latter drugs to reduce stress ulcer are shown in Figures 2a specimens were fixed in formalin, embedded in paraf- and 2b. Significant reductions of stress ulceration from fin and stained with hematoxylin, eosin and alcian 15.88 ± 2.463 mm were achieved by antalarmin (to blue for histological estimation of mucin content. 1.392 ± 0.664 mm, P Ͻ 0.01) followed by diazepam (to Briefly, 7-␮m sections perpendicular through the 1.833 ± 0.726 mm, P Ͻ 0.01), bupropion (to 4.429 ±

Molecular Psychiatry Marked suppression of gastric ulcerogenesis KE Gabry et al 477 Table 1 Effects of different vehicles used for drug dissolution on the stress-induced gastric ulceration, colon mucin, stool pellet output, struggling behavior and plasma epinephrine and norepinephrine

Saline Vehicle A Vehicle B ANOVA comparison

Gastric ulceration 15.88 ± 2.463 12.5 ± 3.212 11.17 ± 3.24 F(2,20) = 0.77, P = 0.476 Colon mucin 4.625 ± 0.944 4.4 ± 1.077 4.333 ± 1.526 F(2,18) = 0.022, P = 0.978 Pellet output 12.13 ± 1.663 6.857 ± 1.471 8.833 ± 2.4 F(2,18) = 2.292, P = 0.13 Struggling 25 ± 3.525 24 ± 3.386 24.2 ± 4.164 F(2,18) = 0.02, P = 0.978 Plasma epinephrine 21.11 ± 2.55 20 ± 5.33 20.2 ± 4.091 F(2,15) = 0.022, P = 0.98 Plasma norepinephrine 10.23 ± 0.849 15.2 ± 5.44 19.4 ± 6.071 F(2,17) = 1.794, P = 0.196

Vehicle A used to dissolve antalarmin consisted of cremophor/ethanol/saline, 5:5:90 by volume, and injected (0.5 ml per animal) intraperitoneally. Vehicle B used for diazepam consisted of 0.5% carboxymethylcellulose and injected intramuscularly (0.5 ml per animal). The rest of the drugs were dissolved in saline and injected intraperitoneally (0.5 ml per animal). As apparent in the ANOVA comparison column, there were no significant differences among the different vehicles on any of the parameters under investigation.

1.706 mm, P Ͻ 0.01) and fluoxetine (to 7.438 ± 1.668 0.05) (Figure 5a) and mucin depletion (from 7.89 ± 1.07 mm, P Ͻ 0.01). Buspirone did not significantly alter to 3.98 ± 1.06 mucin-positive colonocyte per 100 SU scores (to 11.36 ± 2.497 mm, P Ͼ 0.05). colonocytes, t = 2.596, P Ͻ 0.05) (Figure 5b). The effects Both diazepam and antalarmin significantly of intraperitoneal CRH on stool pellet output were also inhibited bowel hypermotility, as indicated by the abolished by antalarmin pretreatment (compared to the stool pellet output (from 12.13 ± 1.663 to 5.875 ± 1.072 vehicle group, t = 0.068, P Ͼ 0.05), but were marginally SPO per 3 h, P Ͻ 0.05 by diazepam, and to 5.938 ± affected by vagotomy or sham operation (compared to 0.972, P Ͻ 0.05 by antalarmin) (Figure 3a), but only the vehicle group, t = 2.146 and 2.26, P = 0.07 and P = antalarmin (Figure 3b) significantly increased colonic 0.06, respectively) (Figure 5a). The effects of intraperi- mucin during exposure to stress (from 4.625 ± 0.944 to toneal CRH on mucin content were abolished by antal- 9.5 ± 1.195 mucin-positive colonocyte per 100 armin pretreatment (compared to the vehicle group, t colonocytes, P Ͻ 0.01). = 0.546, P Ͼ 0.05) but were not significantly affected Figures 4a, 4b and 4c show the effects of antalarmin by vagotomy or sham operation (compared to the and other psychotropic medications on struggling vehicle group, t = 4.181 and 3.06, P Ͻ 0.01 and P Ͻ scores, plasma epinephrine and norepinephrine, 0.02, respectively) (Figure 5b). respectively. As demonstrated in Figure 4a, antalarmin Figure 6 shows that compared to sham-operated rats, (20 mg kg−1) was as effective as diazepam (5 mg kg−1) vagotomized animals expressed significantly lower SU in decreasing the score of struggling; from 25.0 ± 3.525 scores in response to the same stressor (from 17.65 ± to 9.0 ± 2.591 (P Ͻ 0.001) and to 11.5 ± 3.065, P Ͻ 0.05), 3.85 to 3.86 ± 2.5, t = 3.048, P Ͻ 0.02), but their scores respectively. Buspirone (15.5 ± 3.157), bupropion (17.5 of stool pellet output and colonocyte mucin content ± 3.541) and fluoxetine (30.0 ± 3.117) did not signifi- were not statistically different (data not shown). CRH cantly reduce struggling behavior in this model (P Ͼ administration had no effect on SU in either sham- 0.05). Reductions of plasma epinephrine from 21.11 ± operated (t = 1.459, P Ͼ 0.05) or vagotomized subjects 2.55 ng ml−1 were achieved by buspirone (10.79 ± 0.651 (t = 0.619, P Ͼ 0.05). On the other hand, antalarmin ng ml−1, P Ͻ 0.001) followed by diazepam (12.38 ± pretreatment further attenuated SU in the sham-operated 1.143 ng ml−1, P Ͻ 0.001), then antalarmin (13.45 ± animals (t = 3.473 and 1.17, P Ͻ0.01 and Ͼ 0.05, for the 1.178 ng ml−1, P Ͻ 0.05). Least changes were obtained sham-operated and vagotomized rats respectively). with the antidepressants bupropion (14.35 ± 2.18 ng ml−1, P Ͼ 0.05) and fluoxetine (17.20 ± 2.31 ng ml−1, P − Discussion Ͼ 0.05). Plasma norepinephrine (10.23 ± 0.849 ng ml 1) significantly decreased by diazepam (4.168 ± 1.052 ng Here we show that antalarmin not only reduced gastric ml−1, P Ͻ 0.01) followed by buspirone (6.29 ± 0.544 ng ulcerogenesis in responses to stress in a dose-depen- ml−1, P Ͻ 0.05), then antalarmin (6.916 ± 0.463 ng ml−1, dent fashion, but also attenuated autonomic and P Ͻ 0.05). Non-significant changes were obtained with behavioral indices of arousal. The relatively new anti- the antidepressants bupropion (9.532 ± 2.56 ng ml−1, P depressants, bupropion and fluoxetine, as well as Ͼ 0.05) and fluoxetine (13.22 ± 1.277 ng ml−1, P Ͼ diazepam also exerted significant effects in attenuating 0.05). SU, supporting the notion that stress ulceration is a Compared to the saline-treated animals, intraperi- centrally mediated process57 and that it is their psycho- toneal CRH did not significantly change the animals’ tropic properties that contribute to their ulcerolytic stress ulcer scores in non-stressed subjects (from 0.167 effects. We used the relatively new antidepressants ± 0.167 to 0.167 ± 0.167, t = 0.0, P = 0.50). However, it bupropion and fluoxetine that are virtually devoid of significantly resulted in increased colon motility (from antihistaminic or anticholinergic actions to avoid the 4.3 ± 0.925 to 9.4 ± 1.75 SPO per 3 h, t = 2.577, P Ͻ argument that antidepressants attenuated stress ulcer-

Molecular Psychiatry Marked suppression of gastric ulcerogenesis KE Gabry et al 478 ogenesis via blocking cholinergic or histaminergic The notion that stress ulceration is a centrally receptors.45 It is unlikely that antalarmin produced its mediated phenomenon is further supported by the fail- protective effects by merely attenuating ure of peripherally administered CRH to produce SU- responses to stress because we have previously found like lesions in non-stressed subjects. On the contrary, that metyrapone does not modify the severity of stress IP CRH reproduced colonic responses to stress; an ulceration, at doses (40 mg kg−1 IM) that effectively effect that was abolished by antalarmin pretreatment, interfere with glucocorticoid synthesis in the same which is in line with previous studies.29,30,59,60 More- model.58 over, we have observed that the effects of IP CRH were not significantly different between vagotomized and sham-vagotomized animals, which further supports the notion that the intestinal effects of IP CRH were mediated by peripheral mechanisms. In spite of its known implication in virtually the entire cascade of the stress response36 and the repro- ducibility of stress ulceration as a central component of the animals’ responses to stress,1 the role of CRH in the pathophysiology of stress ulcer has been contro- versial,3,61 with some studies reporting an ulcerogenic effect of centrally administered CRH33 and others reporting attenuation of cold restraint-induced gastric mucosal lesions by the same treatment.32,62 The dis- agreement of our findings with the latter reports may be due to the different techniques utilized and their direct injections into the brain of excessive quantities of CRH (in the microgram range) considering its nat- urally occurring concentrations (being in the picogram range).36 At such concentrations, CRH acts locally as a proinflammatory agent and to increase the permeability of physiological barriers; both factors can significantly confound the roles attributed to the endogenous peptide.63 Because our focus is on stress ulceration, whose pathophysiology cannot be simply explained by the associated secretory and motor patterns,55,58 we chose not to collect data on gastric secretion and motility. Thus we avoided the trauma of laparotomy usually required to collect gastric secretion or contractility rec- ords. Moreover, the ability of antalarmin to cross the blood–brain barrier allowed us to avoid inflicting the trauma of intracerebroventricular injections to the ani- mals. The efficacy of antalarmin to antagonize stress ulceration under these circumstances indicates that endogenous CRH, via its type-1 receptor, mediates gas- tric ulcerogenesis caused mostly by plain immobiliz- ation, rather than other confounding physical stressors. Moreover, such findings indicate the potential clinical utility of nonpeptide CRH antagonists for the treatment of stress ulcer in the critically ill and perhaps peptic ulcer disease and non-ulcer dyspepsia. Vagotomy produced antalarmin-like ulcerolytic Figure 1 (a) Dose response (0, 15, 30, 45 mg kg−1 IP) effects effects, which agrees with previous reports.64 The of antalarmin on immobilization stress-induced gatric ulcer- vagus has been long implicated in mediating gastric ation (SU) and (b) a photograph showing the interior of the damage in response to stress.65 By virtue of its exten- rats’ stomachs after the same treatment regimen. Fasted adult sive projections to brain nuclei that participate in the = male Sprague–Dawley rats (n 8/group) received half the neurocircuitry of stress, the vagal efferents are well dose of their drug intraperitoneally twice at −2 h and +2hof positioned to mediate the gastric responses to stress.66 exposure to immobilization stress for 4 h, followed by immediate termination and collection of trunk blood and sto- Indeed, the stomach receives by far more vagal inner- machs. The latter were opened along the greater curvature vation (both afferent and efferent) than any other 67 and the sum of each stomach ulcer length (cumulative ulcer viscus. Efferent vagal fibers contain a variety of length) was measured with the aid of a magnifying lens and neurotransmitters that could be ulcerogenic, eg TRH, a micrometer. serotonin and opiate .65,68 Moreover, the vagus

Molecular Psychiatry Marked suppression of gastric ulcerogenesis KE Gabry et al 479 is essential for the maintenance of mast cells and enter- that have utilized different paradigms43,46 Moreover, ochromaffin-like cells in the stomach.54 The latter cells exogenous parenteral CRH reproduced the intestinal contain highly ulcerogenic substances as well. CRH is responses to stress, in agreement with previous stud- extensively distributed in the hypothalamus and lim- ies.30 This is of particular interest in the light of Fuku- bic system nuclei that project to the dorsal motor do’s finding of induction of colon motility in both IBS nucleus of the vagus (DMV) and CRH containing nerve patients and control subjects by CRH administered fibers are found in the DMV.66,67 intravenously, with significantly greater motility The effects of antalarmin in attenuating colonic indexes in IBS patients than in controls. Abdominal responses to stress were also marked. It significantly symptoms evoked by CRH in IBS patients lasted sig- suppressed both colonic hypermotility and mucin nificantly longer than those in the controls in the same depletion. This agrees with recently published studies study.69 In the light of Williams’ hypothesis of the use- fulness of enhanced bowel motility in response to restraint stress to model an important aspect of IBS, our findings here might suggest the potential clinical utility of CRH-R1 antagonists to combat IBS.4 Vagotomy—on the other hand—did not alter the intestinal responses to stress in this paradigm, again suggesting that the site of antalarmin action in modul- ating colonic responses to stress is primarily periph- eral. In the gut, antalarmin could have directly exerted its effects via a multiplicity of mechanisms. CRH-con- taining cells have been identified in the human, mon- key, cat, and rat stomach and small intestine.70,71 Local production of CRH has been detected in normal human colon enterochromaffin cells.72 In patients with ulcer- ative colitis, considerably enhanced expression of immunoreactive CRH in mucosal inflammatory cells has been reported. Intense staining with anti-CRH anti- body was also shown in mucosal macrophages and CRH mRNA was found to be expressed in mucosal epithelial cells.73 Binding and pharmacological studies have documented the presence of CRH binding sites in the circular smooth muscles of the guinea pig cecum.74 CRH binding sites have also been detected in the myenteric neurons from the guinea-pig ileum,74,75 where it evokes prolonged depolarization in the majority of these neurons in association with aug- mented excitability.75 In isolated distal rat colon prep- arations, CRH provokes a concentration-dependent increase of both mechanical and electrical peristaltic activity.76 belongs to the CRH family of pep- tides and CRH-R1 is primarily involved in mediating

Figure 2 (a) Effects of normal saline (NT; 0.5 ml), antalarmin (ANT; 20 mg kg−1 IP), diazepam (DZP; 5 mg kg−1 IM), buspi- rone (BSP; 4 mg kg−1), fluoxetine (FLX; 10 mg kg−1) and bupro- pion (BPR; 30 mg kg−1) injected 24 and 2 h before the experi- ment on immobilization stress-induced gastric ulceration. As shown, stress ulceration was reduced to Ͻ 9% by antalarmin, 11.5% by diazepam, 28% by bupropion and 47% by fluoxet- ine. (b) The interior of the rats’ stomachs after the same treat- ments. Animals groups (n = 6–12) received the same dose of their drug intraperitoneally twice at −24 h and −2h of exposure to immobilization stress for 4 h followed by immediate termination and collection of trunk blood and sto- machs. The latter were opened along the greater curvature and the sum of each stomach ulcer length (cumulative ulcer length) was measured with the aid of a magnifying lens and a micrometer. *P Ͻ 0.05, **P Ͻ 0.01. There was essentially no ulceration in freely moving animals whether or not they received any medication.

Molecular Psychiatry Marked suppression of gastric ulcerogenesis KE Gabry et al 480 both CRH and urocortin-induced colonic motor response.43 The expression of the urocortin gene in the gastrointestinal tract has also been investigated using reverse transcription polymerase chain reaction (RT- PCR) and in situ hybridization histochemistry. PCR has demonstrated the presence of urocortin mRNA in the rat brain, duodenum, small intestine, and colon. By in situ hybridization, urocortin-containing cells are exclusively localized to the submucosal plexus and myenteric plexus in the duodenum, small intestine and

Figure 3 Effects of normal saline (NT; 0.5 ml), antalarmin (ANT; 20 mg kg−1 IP), diazepam (DZP; 5 mg kg−1 IM), buspi- rone (BSP; 4 mg kg−1), fluoxetine (FLX; 10 mg kg−1) and bupro- pion (BPR; 30 mg kg−1) on immobilization stress-induced colonic motility as indicated by the stool pellet output (panel a) and colonic mucosal mucin content (panel b) in adult male Sprague–Dawley rats. Colon motility was decreased to 48% and 49% by diazepam and antalarmin, respectively. Colon mucin content was increased more than two-fold by antalar- min treatment. *P Ͻ 0.05. In freely moving animals, none of the drugs significantly caused any significant change in gas- tric ulceration, colonic mucin or defecation scores as com- pared to saline treatment.

Figure 4 Effects of normal saline (NT; 0.5 ml), antalarmin (ANT; 20 mg kg−1 IP), diazepam (DZP; 5 mg kg−1 IM), buspi- rone (BSP; 4 mg kg−1), fluoxetine (FLX; 10 mg kg−1) and bupro- pion (BPR; 30 mg kg−1) on struggling scores (panel a), plasma epinephrine (panel b) and norepinephrine (panel c) after exposure of adult male Sprague–Dawley rats to 4 h of immob- ilization stress. All drugs were injected in two full doses at −24 and −2 h of exposure to immobilization stress in the per- itoneal cavity except for diazepam that was administered intramuscularly. Assays utilized trunk blood samples col- lected immediately after releasing the animals from the restrainers. *P Ͻ 0.05, **P Ͻ 0.01. Exposure to this stressor elevated plasma epinephrine (from 8.40 ± 2.22 to 21.11 ± 2.55 ng ml−1, t = 3.76, P Ͻ 0.01 ) and increased plasma norepi- nephrine (from 5.45 ± 1.278 to 10.23 ± 0.849 ng ml−1, t = 3.243, P Ͻ 0.01) as compared to freely moving animals. There was naturally no struggling behavior in freely moving subjects.

Molecular Psychiatry Marked suppression of gastric ulcerogenesis KE Gabry et al 481 colon.77 More recently, urocortin and CRH-R1 mRNA have suggested to serve as a cognitive limb of the peripheral been detected in the human colonic mucosal as well.78 parasympathetic response and to play a role in dis- Several lines of evidence have emerged to indicate orders characterized by comorbidity of intestinal and bidirectional communication between brain and psychiatric symptoms, such as irritable bowel syn- enteric CRH systems. CRH pathways have been traced drome.82 between the rectum and the brainstem autonomic On the efferent side, CRH acts in the nuclei.79 A major symptom in IBS is the exaggerated to induce a long lasting stimulation of colonic transit pain sensitivity to colon distension, which is corre- and bowel discharge.83 This mechanism of action may lated with psychological stress.80 CRH released in the explain the known clinical efficacy of antidepressants gastrointestinal mucosa from immune cells or enter- in IBS. In this study only diazepam was effective in ochromaffin cells may play a role in the modulation of suppressing colonic responses to stress, perhaps via a rectal afferent function as it decreases the thresholds suppressive effect on CRH secretion. By virtue of their and increases the intensity for the sensation of dis- actions on the GABA//chloride iono- comfort in response to distension. These findings are phore complex, are potent inhibitors consistent with a dual effect of CRH on afferent path- of CRH secretion.84 Our findings here, however, should ways mediating perception of aversive rectal sen- not be taken as indices of the efficacy of the drugs sations in humans.81 CRH also mediates the increase in tested, because of their different doses and pharmaco- locus coeruleus spontaneous discharge during colonic kinetic profiles. We have not tested our drugs after mul- distension in rats.82 Activation of the locus coeruleus- tiple administrations to ensure their reaching of the noradrenergic system during colon distension has been steady state, which is beyond the scope of this study. Thus failure of a drug to attenuate a given stress- induced endpoint does not exclude its potential effi- cacy if administered differently. Even though antalarmin significantly attenuated adreno-medullary responses to stress as indicated by the reductions of plasma epinephrine and norepi-

Figure 6 Comparison of the effect of vagotomy vs sham vag- Figure 5 Effect of intraperitoneal CRH (5 ␮gkg−1) on adult otomy (sham.) on stress-induced gastric mucosal lesions in male Sprague–Dawley rats’ colon motility (as indicated by the fasted adult male Sprague–Dawley rats. Vagotomized animals stool pellet output: SPO) and mucin (indicated by the number expressed significantly lower SU scores in response to the of mucin-positive colonocytes %) in freely moving animals same stressor. CRH administration had no significant effect 4 h after the injections. *P Ͻ 0.05. The effects of intraperi- on SU in either sham-operated or vagotomized subjects. On toneal CRH were abolished by antalarmin pretreatment, but the other hand, antalarmin (Ant) pretreatment further attenu- were not affected by vagotomy or sham-vagotomy. *P Ͻ 0.05 ated SU in the sham-operated animals. Means ± SE, *P Ͻ 0.05 when compared to the CRH group. compared to the sham-operated group.

Molecular Psychiatry Marked suppression of gastric ulcerogenesis KE Gabry et al 482 nephrine, no alteration of catecholamine secretion was 12 Drossman DA et al. US householder survey of functional gastro- found in response to antidepressants that also attenu- intestinal disorders. Prevalence, sociodemography, and health impact. Dig Dis Sci 1993; 38: 1569–1580. ated stress ulceration. This supports our previous 13 Brodie DA. Experimental peptic ulcer. Gastroenterology 1968; 55: notion that suppression of sympathoadrenal responses 125–134. to stress may not account for the anti-ulcer effects of 14 Cho CH, Koo MW, Garg GP, Ogle CW. Stress-induced gastric ulcer- drugs, because sympathectomy actually enhances the ation: its aetiology and clinical implications. Scand J Gastroenterol 55 1992; 27: 257–262. ulcerogenic processes in response to stress. Other 15 Pare WP, Glavin GB. Restraint stress in biomedical research: a studies have shown that neither adrenalectomy nor review. Neurosci Biobehav Rev 1986; 10:339–370. hypophysectomy prevent the response of the intestine 16 Levenstein S. Stress and peptic ulcer: life beyond Helicobacter. BMJ to stress, suggesting that neither adrenal nor pituitary- 1998; 316:538–541. derived factors are responsible for mediating the effects 17 Habib KE, Gold PW, Chrousos GP. Neuroendorinology of stress, 4 Endocrinol Metab Clin North Am 2001; 30: 695–728. of stress on the gut. Therefore, the weight of evidence 18 Peters MN, Richardson CT. Stressful life events, acid hypersecre- supports the implication of primarily central mech- tion, and ulcer disease. Gastroenterology 1983; 84: 114–119. anisms in the pathogenesis of stress ulceration, and lar- 19 Bresnick WH, Rask-Madsen C, Hogan DL, Koss MA, Isenberg JI. gely peripheral mechanisms, presumably utilizing The effect of acute emotional stress on gastric acid secretion in normal subjects and duodenal ulcer patients. J Clin Gastroenterol CRH-R1 in the gut, in the colonic responses to stress. 1993; 17: 117–122. Taken together, we conclude that brain CRH-R1 20 Pare WP. Stress ulcer susceptibility and in Wistar Kyoto exerts an important role in transforming psychological (WKY) rats. Physiol Behav 1989; 46:993–998. adversity into visceral damage at least in the stomach. 21 Glavin GB. Vulnerability to stress ulcerogenesis in rats differing in Our data also indicate that gut CRH-R1 mediates anxiety: a dopaminergic correlate. J Physiol Paris 1993; 87:239–243. 22 Feldman M, Walker P, Green JL, Weingarden K. Life events stress colonic responses to stress. Novel nonpeptide CRH-R1 and psychosocial factors in men with peptic ulcer disease. A multi- antagonists may therefore provide an optimal treat- dimensional case-controlled study. Gastroenterology 1986; 91: ment for such gastrointestinal disorders associated 1370–1379. with stress as stress ulcer and possibly other pathogen- 23 Ries RK, Gilbert DA, Katon W. antidepressant therapy for etically related conditions such as peptic ulcer and irri- peptic ulcer disease. Arch Intern Med 1984; 144:566–569. 24 Mangla JC, Pereira M. Tricyclic antidepressants in the treatment of table bowel syndrome in humans. peptic ulcer disease. Arch Intern Med 1982; 142:273–275. 25 Shrivastava RK, Siegel H. The role of and benzodiazepine compounds in the treatment of irritable gut syndrome and peptic Acknowledgements ulcer disease. Psychopharmacol Bull 1984; 20:616–621. We thank Dr Rita J Valentino for her thoughtful dis- 26 Haggerty JJ Jr, Drossman DA. Use of psychotropic drugs in patients with peptic ulcer. Psychosomatics 1985; 26:277–280, 283–284. cussion and Mr Keith Zachman for his help in con- 27 Shrivastava RK, Shah BK, Siegal H. and cimetidine in the ducting the animal experiments. treatment of duodenal ulcer: a double-blind comparative study. Clin Ther 1985; 7:181–189. 28 Ogawa N, Hara C, Takaki S. Anxiolytic activity of SC-48274 com- References pared with those of buspirone and diazepam in experimental anxi- 1 Selye H. A syndrome produced by diverse nocuous agents. Nature ety models. Jpn J Pharmacol 1993; 61:115–121. 1936; 138:32–36. 29 Williams CL, Peterson JM, Villar RG, Burks TF. Corticotropin- 2 Schilling D et al. Low seroprevalence of Helicobacter pylori infec- releasing factor directly mediates colonic responses to stress. Am tion in patients with stress ulcer bleeding—a prospective evalu- J Physiol 1987; 253: G582–G586. ation of patients on a cardiosurgical intensive care unit. Intensive 30 Lenz HJ, Raedler A, Greten H, Vale WW, Rivier JE. Stress-induced Care Med 2000; 26:1832–1836. gastrointestinal secretory and motor responses in rats are mediated 3 Weiner H. Use of animal models in peptic ulcer disease. Psycho- by endogenous corticotropin-releasing factor. Gastroenterology som Med 1996; 58: 524–545. 1988; 95: 1510–1517. 4 Williams CL, Villar RG, Peterson JM, Burks TF. Stress-induced 31 Pothoulakis C, Castagliuolo I, Leeman SE. Neuroimmune mech- changes in intestinal transit in the rat: a model for irritable bowel anisms of intestinal responses to stress. Role of corticotropin- syndrome. Gastroenterology 1988; 94: 611–621. releasing factor and . Ann N Y Acad Sci 1998; 840: 5 Sjodin I, Svedlund J, Dotevall G, Gillberg R. Symptom profiles in 635–648. chronic peptic ulcer disease. A detailed study of abdominal and 32 Ray A, Henke PG, Gulati K, Sen P. The amygdaloid complex, cortic- mental symptoms. Scand J Gastroenterol 1985; 20:419–427. otropin releasing factor and stress-induced gastric ulcerogenesis in 6 Lydiard RB, Falsetti SA. Experience with anxiety and depression rats. Brain Res 1993; 624:286–290. treatment studies: implications for designing irritable bowel syn- 33 Nakane T, Kanie N, Audhya T, Hollander CS. The effects of cen- drome clinical trials. Am J Med 1999; 107: 65S–73S. trally administered on the development of gastric 7 De Bellis MD et al. Hypothalamic-pituitary-adrenal axis dysregul- lesions in the rat. Life Sci 1985; 36: 1197–1203. ation in sexually abused girls. J Clin Endocrinol Metab 1994; 78: 34 Vale W, Spiess J, Rivier C, Rivier J. Characterization of a 41-residue 249–255. ovine hypothalamic peptide that stimulates secretion of corticotro- 8 Heim C, Owens MJ, Plotsky PM, Nemeroff CB. Persistent changes pin and beta-endorphin. Science 1981; 213: 1394–1397. in corticotropin-releasing factor systems due to early life stress: 35 Sutton RE, Koob GF, Le Moal M, Rivier J, Vale W. Corticotropin relationship to the pathophysiology of major depression and post- releasing factor produces behavioural activation in rats. Nature traumatic stress disorder. Psychopharmacol Bull 1997; 33:185–192. 1982; 297:331–333. 9 Ackerman SH, Hofer MA, Weiner H. Early maternal separation 36 Habib KE et al. Oral administration of a corticotropin-releasing hor- increases gastric ulcer risk in rats by producing a latent thermoreg- mone significantly attenuates behavioral, neur- ulatory disturbance. Science 1978; 201:373–376. oendocrine, and autonomic responses to stress in primates. Proc 10 Almounajed G, Drossman DA. Newer aspects of the irritable bowel Natl Acad Sci U S A 2000; 97: 6079–6084. syndrome. Prim Care 1996; 23: 477–495. 37 Tache Y et al. Inhibition of gastric acid secretion in rats by intracer- 11 Russo MW, Gaynes BN, Drossman DA. A national survey of prac- ebral injection of corticotropin-releasing factor. Science 1983; 222: tice patterns of gastroenterologists with comparison to the past two 935–937. decades. J Clin Gastroenterol 1999; 29:339–343. 38 Martinez V, Rivier J, Wang L, Tache Y. Central injection of a new

Molecular Psychiatry Marked suppression of gastric ulcerogenesis KE Gabry et al 483 corticotropin-releasing factor (CRF) antagonist, astressin, blocks 61 Murison R, Bakke HK. The role of corticotropin-releasing factor in CRF- and stress-related alterations of gastric and colonic motor rat gastric ulcerogenesis. Ann N Y Acad Sci 1990; 597:71–80. function. J Pharmacol Exp Ther 1997; 280:754–760. 62 Tache Y. The peptidergic brain-gut axis: influence on gastric ulcer 39 Lydiard RB. Anxiety and the irritable bowel syndrome: psychiatric, formation. Chronobiol Int 1987; 4:11–17. medical, or both? J Clin Psychiatry 1997; 58:51–58; discussion 63 Chrousos GP. The hypothalamic-pituitary-adrenal axis and 59–61. immune-mediated inflammation. N Engl J Med 1995; 332:1351–1362. 40 Habib KE et al. CRH type-1 receptor antagonism attenuates 64 Cho CH, Ogle CW, Dai S. Acute gastric ulcer formation in response behavioral, endocrine and autonomic responses to stress in pri- to electrical vagal stimulation in rats. Eur J Pharmacol 1976; 35: mates. Neuropsychopharmacology 2000; 23(S2): S119. 215–219. 41 Gold PW, Goodwin FK, Chrousos GP. Clinical and biochemical 65 Hernandez DE. Neuroendocrine mechanisms of stress ulceration: manifestations of depression. Relation to the neurobiology of stress focus on thyrotropin-releasing hormone (TRH). Life Sci 1986; 39: (2). N Engl J Med 1988; 319:413–420. 279–296. 42 Chen YL et al. Synthesis and oral efficacy of a 4- 66 Weiner H. From simplicity to complexity (1950–1990): the case of (butylethylamino)pyrrolo[2,3-d]pyrimidine: a centrally active peptic ulceration–II. Animal studies. Psychosom Med 1991; 53: corticotropin-releasing factor1 receptor antagonist. J Med Chem 491–516. 1997; 40:1749–1754. 67 Grijalva CV, Novin D. The role of the hypothalamus and dorsal 43 Maillot C, Million M, Wei JY, Gauthier A, Tache Y. Peripheral vagal complex in gastrointestinal function and pathophysiology. corticotropin-releasing factor and stress-stimulated colonic motor Ann N Y Acad Sci 1990; 597:207–222. activity involve type 1 receptor in rats. Gastroenterology 2000; 119: 68 Habib KE, Negro PJ, Negro PP, Gold PW. Endogenous opioids con- tribute to gastric ulcerogenesis during immobilization stress in 1569–1579. male Sprague–Dawley rats. Recognition of ␤-endorphin immunore- 44 Gilligan PJ, Robertson DW, Zaczek R. Corticotropin releasing factor activity in the gastric mucosa. Gastroenterology 1999; 116: G0782. (CRF) receptor modulators: progress and opportunities for new 69 Fukudo S, Nomura T, Hongo M. Impact of corticotropin-releasing therapeutic agents. J Med Chem 2000; 43: 1641–1660. hormone on gastrointestinal motility and adrenocorticotropic hor- 45 Beaumont G. Adverse effects of antidepressants. Int Clin Psycho- mone in normal controls and patients with irritable bowel syn- pharmacol 1990; 5 Suppl 3: 61–66. drome. Gut 1998; 42:845–849. 46 Castagliuolo I et al. Acute stress causes mucin release from rat 70 Petrusz P et al. Corticotropin-releasing factor (CRF)-like immunore- colon: role of corticotropin releasing factor and mast cells. Am J activity in the gastro-entero-pancreatic endocrine system. Peptides Physiol 1996; 271: G884–G892. 1984; 5:71–78. 47 Hall CS. Emotional behavior in the rat. J Comp Psychol 1934; 18: 71 Wolter HJ. Corticotropin-releasing factor is contained within per- 385–403. ikarya and nerve fibres of rat duodenum. Biochem Biophys Res 48 Webster EL et al. In vivo and in vitro characterization of antalar- Commun 1984; 122:381–387. min, a nonpeptide corticotropin-releasing hormone (CRH) receptor 72 Kawahito Y et al. Local secretion of corticotropin-releasing hor- antagonist: suppression of pituitary ACTH release and peripheral mone by enterochromaffin cells in human colon. Gastroenterology inflammation. Endocrinology 1996; 137: 5747–5750. 1994; 106:859–865. 49 Ushijima I et al. The role of adenosinergic, GABAergic and benzo- 73 Kawahito Y et al. Corticotropin releasing hormone in colonic diazepine systems in hyperemotionality and ulcer formation in mucosa in patients with ulcerative colitis. Gut 1995; 37:544–551. stressed rats. Psychopharmacology 1986; 89: 472–476. 74 Iwakiri Y, Chijiiwa Y, Motomura Y, Osame M, Nawata H. Presence 50 Ida Y, Tanaka M, Tsuda A, Tsujimaru S, Nagasaki N. Attenuating of functional receptors for corticotropin releasing hormone in cae- effect of diazepam on stress-induced increases in noradrenaline cal circular smooth muscle cells of guinea pig. Life Sci 1997; 60: turnover in specific brain regions of rats: antagonism by Ro 15– 857–864. 1788. Life Sci 1985; 37:2491–2498. 75 Hanani M, Wood JD. Corticotropin-releasing hormone excites 51 de Boer SF, Slangen JL, van der Gugten J. Effects of chlordiazepox- myenteric neurons in the guinea-pig small intestine. Eur J Pharma- ide and buspirone on plasma catecholamine and corticosterone lev- col 1992; 211:23–27. els in rats under basal and stress conditions. Endocrinol Exp 1990; 76 Mancinelli R, Azzena GB, Diana M, Forgione A, Fratta W. In vitro 24:229–239. excitatory actions of corticotropin-releasing factor on rat colonic 52 Cooper BR, Hester TJ, Maxwell RA. Behavioral and biochemical motility. J Auton Pharmacol 1998; 18:319–324. effects of the antidepressant bupropion (Wellbutrin): evidence for 77 Harada S et al. Urocortin mRNA is expressed in the enteric nervous selective blockade of dopamine uptake in vivo. J Pharmacol Exp system of the rat. Neurosci Lett 1999; 267:125–128. Ther 1980; 215: 127–134. 78 Muramatsu Y et al. Urocortin and corticotropin-releasing factor 53 Caccia S, Fracasso C, Garattini S, Guiso G, Sarati S. Effects of short- receptor expression in the human colonic mucosa. Peptides 2000; and long-term administration of fluoxetine on the monoamine con- 21: 1799–1809. tent of rat brain. Neuropharmacology 1992; 31: 343–347. 79 Valentino RJ, Kosboth M, Colflesh M, Miselis RR. Transneuronal 54 Weichert RF, 3rd, Ellison JP, Woolverton WC, Drapanas T. Postvag- labeling from the rat distal colon: anatomic evidence for regulation of distal colon function by a pontine corticotropin-releasing factor otomy serotonin depletion in mucosal mast cells of the rat stomach. system. J Comp Neurol 2000; 417:399–414. Surg Forum 1970; 21:318–320. 80 Whitehead WE, Crowell MD, Davidoff AL, Palsson OS, Schuster 55 Habib KE, Negro PJ, Sciullo D, Dib A, Gold PW. Protective role of MM. Pain from rectal distension in women with irritable bowel the sympathoadrenal system against stress-induced gastric mucosal syndrome: relationship to sexual abuse. Dig Dis Sci 1997; 42: injury in the rat. Gastroenterology 1999; 116: G0783. 796–804. 56 Bakke HK, Murison R. Plasma corticosterone and restraint induced 81 Lembo T et al. Effects of the corticotropin-releasing factor (CRF) on gastric pathology: age-related differences after administration of rectal afferent nerves in humans. Neurogastroenterol Motil 1996; 8: corticotropin releasing factor. Life Sci 1989; 45:907–916. 9–18. 57 Glavin GB et al. The neurobiology of stress ulcers. Brain Res Brain 82 Lechner SM, Curtis AL, Brons R, Valentino RJ. Locus coeruleus Res Rev 1991; 16: 301–343. activation by colon distention: role of corticotropin-releasing factor 58 Habib KE. Role of brain-adrenal-gastric axis in modulating gastric and excitatory amino acids. Brain Res 1997; 756:114–124. responses to stress in albino rats. MSc, Minia University, 1994. 83 Monnikes H, Schmidt BG, Tebbe J, Bauer C, Tache Y. Microin- 59 Lenz HJ, Burlage M, Raedler A, Greten H. Central nervous system fusion of corticotropin releasing factor into the locus effects of corticotropin-releasing factor on gastrointestinal transit coeruleus/subcoeruleus nuclei stimulates colonic motor function in the rat. Gastroenterology 1988; 94: 598–602. in rats. Brain Res 1994; 644: 101–108. 60 Bueno L et al. Effects of corticotropin-releasing factor on plasma 84 Calogero AE. Neurotransmitter regulation of the hypothalamic and somatostatin levels and gastrointestinal motility in corticotropin-releasing hormone neuron. Ann N Y Acad Sci 1995; dogs. Gastroenterology 1986; 91:884–889. 771:31–40.

Molecular Psychiatry