- Email: [email protected]
Comparative antisecretory effect of loperamide somatostatin and the sigma ligand igmesine on IL-1β-induced colonic hypersecretion in rats
Intestinal Disorders A423
April 1998
• G1723 CRITICAL ROLE OF cAMP RESPONSE ELEMENT (CRE) IN HYPOXIA-INDUCED SECRETION OF TUMOR NECROSIS FACTOR ALPHA FROM INTESTINAL EPITHELIAL CELLS. C.T. Tavlor& S.P. Colgan. Center for Experimental Therapeutics and Repeffusion Injury, Dept. of Anesthesia, Brigham and Women's Hospital and Harvard Medical School, Boston MA 02115, USA. In a number of intestinal diseases, tissue hypoxia occurs in conjunction with ongoing inflammation. In parallel, intestinal inflammation is associated with increased levels of tumor necrosis factor-a (TNFct). Since the epithelium represents an active component of the intestinal immune response, we hypothesized that epithelia produce TNF~ under the influence of cellular hypoxia. Confluent monolayers of T84 colonic epithelial cells were exposed to hypoxia (pOz of 20 torr) for up to 48 hours (conditions we have previously demonstrated to be non-toxic). Polarized TNFa secretion and intracellular cAMP levels were determined by ELISA. Activation of the transcription factor NF~:B was examined by nuclear translocation of NFKB p65 subunit via western blotting of nuclear lysates. Phosphorylation of the cAMP response element binding protein (CREB) was determined by gel supershift assay. T84 cells did not secrete detectable amounts TNFct under normoxic conditions. Exposure to hypoxia (2 - 48 hours) induced a time-dependent release of TNFct (10.1_+1.9 pg/monolayer by 48h; n=5; p<0.05), similar to our positive control (100ng/ml PMA, 7.2 -+ 1.3 pg/monolayer). Cell surface TNFa was also elevated by 48 hours of hypoxia (OD405=0.21 ±0.01) when compared with control (OD405=0.10 ± 0.01). TNFet secretion was matched by a time-dependent increase in NFKB translocation (maximal by 24h). Interestingly, other epithelial NFKB-dependent endpoints (ICAM- 1 expression) proved negative in response to hypoxia, suggesting additional component(s) for induction of epithelial TNF-ct by hypoxia. Unlike 1CAM-l, the TNFa promoter bears a DNA binding motif for both NFKB and CREB. Thus, we investigated the influence of hypoxia on cAMP. Hypoxia decreased cytosolic cAMP levels in a time-dependent manner (35.0 _+26.1 fmol / monolayer vs 211.6 _+72.7 fmol / monolayer for controls; 48 hours hypoxia; p < 0.01; ANOVA). Pharmacologic elevation of intracellular cAMP (using 8-bomocAMP, 0 - 3mM) reversed hypoxia-elicited TNFct secretion in a concentration dependent manner (86.2% reversal with 3mM 8-bromo-cAMP) without influencing other NFKB-dependent gene products lacking CRE (ICAM-1). By gel supershift assay, hypoxia diminished TNFct-specific CREB phosphorylation. In summary, hypoxia elicits TNFct secretion from intestinal epithelial cells under positive and negative regulatory control by NFKB and CREB phosphorylation, respectively. We conclude that in conditions of inflammation, tissue hypoxia may play an important role in propagation of the inflammatory response. • G1724 COMPARATIVE ANTISECRETORY EFFECT OF LOPERAMIDE SOMATOSTATIN AND THE SIGMA LIGAND IGMESINE ON IL-lJ3INDUCED COLONIC HYPERSECRETION IN RATS. V. Theodorou+, M. Chovet*, J. Fioramonti, L. Bueno. Dept of Pharmacology, INRA, +ESAP, Toulouse; *Jouveinal Parke-Davis, Fresnes, France. IL-113 a cytokine released by macrophages and monocytes, is involved in diarrheal states associated with infectious or inflammatory bowel diseases and food allergy reactions. IL-113 alters ion transport in the ileum and triggers net water secretion in the proximal colon. Loperamide, somatostatin and igmesine, a sigma ligand, are antisecretory agents acting through different mechanisms. The aim of this study was to compare the antisecretory effect of loperamide somatostatin and igmesine on IL-l[3-induced colonic hypersecretion in anesthetized rats. Ten groups of six male Wistar rats were used. Under urethane anesthesia a proximal colonic segment (5 cm) was isolated and infused with a Ringer solution containing 5 laCi/l of [14C] PEG. After a 2 h equilibration period, net water flux in the isolated loop, was calculated from 14C concentration, measured in the effluent collected at 15 min intervals over 3 h. IL-113 (5 pg/kg) or vehicle (NaCI 0.9%) was administered intraperitoneally (i.p) 3 h after the start of infusion. IL-113 or saline administration was preceded (20 min) by intraperitoneal administration of either loperamide (1 mg/kg), somatostatin (1 ~g/kg), igmesine (1 mg/kg) or their vehicle. In another group, IL-113 administration was preceded by both the somatostatin receptor antagonist, anti-SRIF (10 ~tg/kg i.p) and igmesine (1 mg/kg i.p) administration. In a last group the effect p e r se of anti-SRIF (10 IJg/kg i.p) was determined. Net water absorption was +61.0 + 22.1 ~l/cm/h in controls. IL-113reversed the net water absorption into a net secretion during 30 min (-75.2 -+46.0 IJl/cm/h). Loperamide failed to modify the IL-113 secretory effect (-80.5 -+42.5 ~l/cm/h). In contrast somatostatin and igmesine significantly (p < 0.05) reduced the IL-1 [3-induced hypersecretion (+37.0 -+43.5, +38.3 ± 41.8 pl/cm/h respectively). Moreover, anti-SRIF reversed the antisecretory effect of igmesine on IL-l[3-induced colonic hypersecretion (-51.0-+39.2 IJl/cm/h). Loperamide, somatostatin, igmesine or anti-SRIF, had no effect p e r se on colonic net water flux (+54.3-+ 21.2, 49.3 ± 16.4, +51.3 ± 18.8, 58.2 ± 31.5 ~l.cm/h respectively). Conclusion: Somatostatin and igmesine, but not loperamide, block the IL-I[L induced hypersecretion, suggesting a role of somatostatin and sigma, but not
opiate receptors in this effect. The igmesine antisecretory effect involves a release of somatostatin. This research was funded in part by Jouveinal-Parke-Davis, France. • G1725 EFFECT OF LOCAL ADMINISTRATION OF A NON-PEPTIDE NIO RECEPTOR ANTAGONIST, SR 142801, ON C L O S T R I D I U M D I F F I C I L E - I N D U C E D DIARRHEA IN GUINEA PIGS. V. Theodorou +, P. Marche, G. Corthier*, X. Edmonds-Alt =, J. Fioramonti, L. Bueno. Dept of Pharmacology, INRA,; +ESAP, Toulouse,; *Lab. of Digestive Ecology, Jouy,; =SANOFI Research, Montpellier, France. Clostridium Difficile (CD) toxin A induces a fluid secretion in the ileum and an increase in intestinal permeability, involving sensory neurons and activation of tachykinin NK1 receptors. Diarrhea of CD toxins in mice has been found reduced by NK1 and NK2 receptor antagonists, but the role of NK3 receptors, has not been investigated. Since NK3 receptors have been localized in guinea pig large intestine, the aim of this study was to determine in this species, the effect of SR 142801 a non-peptide selective NK3 receptor antagonist, on CD toxin-induced diarrhea. Methods: Ten groups of eight male Dunkin Hartley guinea pigs were used. Under ketamine anesthesia (120 mg/kg) animals were chronically fitted with a catheter inserted in the jejunum (-40 cm from the ileo-cecal junction) and exteri~brized on the back of the neck, for intraluminal administrations. Eight days post surgery, the animals were individually housed and fecal materials were collected at 30 min intervals over 300 min. Fecal water was measured by weighting the feces, before and after heating at 120 ° C for 24 hours and expressed in gram. A mixture of CD toxins A and B (10 ng) or vehicle (NaC1 0.9%) was administered intrajejunally 10 min, after SR 142801 (0.1, 0.3, 1 and 10 mg/kg) or saline given by the same route. Results: Fecal water excreted from 30 to 300 min in controls was 0.68 ± 0.39 g. CD toxins, significantly increased (p < 0.05) fecal water excretion during the same period (2.13 -+0.33 g). All doses of SR 142801 (0.1, 0.3, 1 and 10 mg/kg) used, significantly (p < 0.05) blocked the CD toxin-induced fecal water increase for the 30 to 300 min period (0.43_+0.16, 0.99_+0.27, 0.87 ± 0.26, 0.84 -+0.28 g respectively). SR 142801 administered through the jejunal catheter at the same doses, had no effect p e r se on fecal water excretion (0.50 + 0.25, 0.68 _+0.27, 0.62 _+0.30, 0.50 +- 0.25 g respectively). Conclusion: SR 142801 exerts a potent antidiarrheal effect on CD toxininduced diarrhea. These results suggest that diarrhea induced by CD toxins in guinea pigs, is mediated by NK3 receptors activation, resulting from tachykinin and particularly NKB release. This research was funded by institutional support of INRA and ESAP.
• G1726 DIFLOUROMETHYLORNITHINE INHIBITS CRYPT FISSION. J.S. Thomnson, S.K. Saxena, J.G. Sharp. Dept of Surgery and Cell Biology and Anatomy, Omaha, VAMC and University of Nebraska Medical Center, Omaha, NE Crypt fission is a physiologic mechanism of crypt reproduction. It increases in pathophysiologic situations where intestinal regeneration is required e.g. radiation injury. Polyamine metabolism is important in the regulation of intestinal growth and recovery from injury in response to a variety of stimuli. The aim of the present study was to determine whether inhibition of polyamine synthesis by DFMO influences crypt fission. 48 rabbits under patch enteroplasty in the terminal ileum. One group served as controls and the other took 2% DFMO orally. Animals (n-6) from each group were sacrificed at 7, 14, 21 and 28 days. Normal ileum adjacent to the enteroplasty was studied. Crypt dissection was performed 2 hours after vincristine I.P. to determine crypt cell production rate (CCPR), crypt depth(CD), and proportion of bifurcating crypts(fission). Control Fission 7days
CCPR
DFMO CD
Fission 4.0±2.4*
CCPR
CD
7.9±0.6
251±3
ll.0±2.2
7.5±0.9
244±9
14days 34.0±4.0
14.9±0.7
241±7# 13.0±1.2"
12.6±0.7 354±31"#
21days
10.0±0.9
9.4±0.6
206±7
10.4±0. A6
9.0±0.8
292±5*
28 days 16.0±6.6
9.7±1.3
254±36
10.5±2.5
8.6±0.6
268±13
* P < 0.05 vs control # P < 0.05 vs 21 days DFMO administration decreased crypt fission. There was a corresponding increase in crypt depth. CCPR was similar in both groups. Mucosal ornithine decarboxylase activity (379 -+ 122 vs 39 -+9 sp. act. at 21 days, P < 05) and polyamine content were significantly lower in the DFMO group at 14 and 21 days (7.4±1.4 vs 1.6-+7 and 8.7-+.9 vs 48-+1.0 pmol/mg, control vs DFMO, P < .05). Conclusions: (1) DFMO administration inhibits crypt fission in stimulated intestinal epithelium. (2) This effect correlates temporally with reduced polyamine production. (3) Reduced crypt fission is another potential mechanism of inhibition of intestinal growth by altered polyamine metabolism.
April 1998
• G1723 CRITICAL ROLE OF cAMP RESPONSE ELEMENT (CRE) IN HYPOXIA-INDUCED SECRETION OF TUMOR NECROSIS FACTOR ALPHA FROM INTESTINAL EPITHELIAL CELLS. C.T. Tavlor& S.P. Colgan. Center for Experimental Therapeutics and Repeffusion Injury, Dept. of Anesthesia, Brigham and Women's Hospital and Harvard Medical School, Boston MA 02115, USA. In a number of intestinal diseases, tissue hypoxia occurs in conjunction with ongoing inflammation. In parallel, intestinal inflammation is associated with increased levels of tumor necrosis factor-a (TNFct). Since the epithelium represents an active component of the intestinal immune response, we hypothesized that epithelia produce TNF~ under the influence of cellular hypoxia. Confluent monolayers of T84 colonic epithelial cells were exposed to hypoxia (pOz of 20 torr) for up to 48 hours (conditions we have previously demonstrated to be non-toxic). Polarized TNFa secretion and intracellular cAMP levels were determined by ELISA. Activation of the transcription factor NF~:B was examined by nuclear translocation of NFKB p65 subunit via western blotting of nuclear lysates. Phosphorylation of the cAMP response element binding protein (CREB) was determined by gel supershift assay. T84 cells did not secrete detectable amounts TNFct under normoxic conditions. Exposure to hypoxia (2 - 48 hours) induced a time-dependent release of TNFct (10.1_+1.9 pg/monolayer by 48h; n=5; p<0.05), similar to our positive control (100ng/ml PMA, 7.2 -+ 1.3 pg/monolayer). Cell surface TNFa was also elevated by 48 hours of hypoxia (OD405=0.21 ±0.01) when compared with control (OD405=0.10 ± 0.01). TNFet secretion was matched by a time-dependent increase in NFKB translocation (maximal by 24h). Interestingly, other epithelial NFKB-dependent endpoints (ICAM- 1 expression) proved negative in response to hypoxia, suggesting additional component(s) for induction of epithelial TNF-ct by hypoxia. Unlike 1CAM-l, the TNFa promoter bears a DNA binding motif for both NFKB and CREB. Thus, we investigated the influence of hypoxia on cAMP. Hypoxia decreased cytosolic cAMP levels in a time-dependent manner (35.0 _+26.1 fmol / monolayer vs 211.6 _+72.7 fmol / monolayer for controls; 48 hours hypoxia; p < 0.01; ANOVA). Pharmacologic elevation of intracellular cAMP (using 8-bomocAMP, 0 - 3mM) reversed hypoxia-elicited TNFct secretion in a concentration dependent manner (86.2% reversal with 3mM 8-bromo-cAMP) without influencing other NFKB-dependent gene products lacking CRE (ICAM-1). By gel supershift assay, hypoxia diminished TNFct-specific CREB phosphorylation. In summary, hypoxia elicits TNFct secretion from intestinal epithelial cells under positive and negative regulatory control by NFKB and CREB phosphorylation, respectively. We conclude that in conditions of inflammation, tissue hypoxia may play an important role in propagation of the inflammatory response. • G1724 COMPARATIVE ANTISECRETORY EFFECT OF LOPERAMIDE SOMATOSTATIN AND THE SIGMA LIGAND IGMESINE ON IL-lJ3INDUCED COLONIC HYPERSECRETION IN RATS. V. Theodorou+, M. Chovet*, J. Fioramonti, L. Bueno. Dept of Pharmacology, INRA, +ESAP, Toulouse; *Jouveinal Parke-Davis, Fresnes, France. IL-113 a cytokine released by macrophages and monocytes, is involved in diarrheal states associated with infectious or inflammatory bowel diseases and food allergy reactions. IL-113 alters ion transport in the ileum and triggers net water secretion in the proximal colon. Loperamide, somatostatin and igmesine, a sigma ligand, are antisecretory agents acting through different mechanisms. The aim of this study was to compare the antisecretory effect of loperamide somatostatin and igmesine on IL-l[3-induced colonic hypersecretion in anesthetized rats. Ten groups of six male Wistar rats were used. Under urethane anesthesia a proximal colonic segment (5 cm) was isolated and infused with a Ringer solution containing 5 laCi/l of [14C] PEG. After a 2 h equilibration period, net water flux in the isolated loop, was calculated from 14C concentration, measured in the effluent collected at 15 min intervals over 3 h. IL-113 (5 pg/kg) or vehicle (NaCI 0.9%) was administered intraperitoneally (i.p) 3 h after the start of infusion. IL-113 or saline administration was preceded (20 min) by intraperitoneal administration of either loperamide (1 mg/kg), somatostatin (1 ~g/kg), igmesine (1 mg/kg) or their vehicle. In another group, IL-113 administration was preceded by both the somatostatin receptor antagonist, anti-SRIF (10 ~tg/kg i.p) and igmesine (1 mg/kg i.p) administration. In a last group the effect p e r se of anti-SRIF (10 IJg/kg i.p) was determined. Net water absorption was +61.0 + 22.1 ~l/cm/h in controls. IL-113reversed the net water absorption into a net secretion during 30 min (-75.2 -+46.0 IJl/cm/h). Loperamide failed to modify the IL-113 secretory effect (-80.5 -+42.5 ~l/cm/h). In contrast somatostatin and igmesine significantly (p < 0.05) reduced the IL-1 [3-induced hypersecretion (+37.0 -+43.5, +38.3 ± 41.8 pl/cm/h respectively). Moreover, anti-SRIF reversed the antisecretory effect of igmesine on IL-l[3-induced colonic hypersecretion (-51.0-+39.2 IJl/cm/h). Loperamide, somatostatin, igmesine or anti-SRIF, had no effect p e r se on colonic net water flux (+54.3-+ 21.2, 49.3 ± 16.4, +51.3 ± 18.8, 58.2 ± 31.5 ~l.cm/h respectively). Conclusion: Somatostatin and igmesine, but not loperamide, block the IL-I[L induced hypersecretion, suggesting a role of somatostatin and sigma, but not
opiate receptors in this effect. The igmesine antisecretory effect involves a release of somatostatin. This research was funded in part by Jouveinal-Parke-Davis, France. • G1725 EFFECT OF LOCAL ADMINISTRATION OF A NON-PEPTIDE NIO RECEPTOR ANTAGONIST, SR 142801, ON C L O S T R I D I U M D I F F I C I L E - I N D U C E D DIARRHEA IN GUINEA PIGS. V. Theodorou +, P. Marche, G. Corthier*, X. Edmonds-Alt =, J. Fioramonti, L. Bueno. Dept of Pharmacology, INRA,; +ESAP, Toulouse,; *Lab. of Digestive Ecology, Jouy,; =SANOFI Research, Montpellier, France. Clostridium Difficile (CD) toxin A induces a fluid secretion in the ileum and an increase in intestinal permeability, involving sensory neurons and activation of tachykinin NK1 receptors. Diarrhea of CD toxins in mice has been found reduced by NK1 and NK2 receptor antagonists, but the role of NK3 receptors, has not been investigated. Since NK3 receptors have been localized in guinea pig large intestine, the aim of this study was to determine in this species, the effect of SR 142801 a non-peptide selective NK3 receptor antagonist, on CD toxin-induced diarrhea. Methods: Ten groups of eight male Dunkin Hartley guinea pigs were used. Under ketamine anesthesia (120 mg/kg) animals were chronically fitted with a catheter inserted in the jejunum (-40 cm from the ileo-cecal junction) and exteri~brized on the back of the neck, for intraluminal administrations. Eight days post surgery, the animals were individually housed and fecal materials were collected at 30 min intervals over 300 min. Fecal water was measured by weighting the feces, before and after heating at 120 ° C for 24 hours and expressed in gram. A mixture of CD toxins A and B (10 ng) or vehicle (NaC1 0.9%) was administered intrajejunally 10 min, after SR 142801 (0.1, 0.3, 1 and 10 mg/kg) or saline given by the same route. Results: Fecal water excreted from 30 to 300 min in controls was 0.68 ± 0.39 g. CD toxins, significantly increased (p < 0.05) fecal water excretion during the same period (2.13 -+0.33 g). All doses of SR 142801 (0.1, 0.3, 1 and 10 mg/kg) used, significantly (p < 0.05) blocked the CD toxin-induced fecal water increase for the 30 to 300 min period (0.43_+0.16, 0.99_+0.27, 0.87 ± 0.26, 0.84 -+0.28 g respectively). SR 142801 administered through the jejunal catheter at the same doses, had no effect p e r se on fecal water excretion (0.50 + 0.25, 0.68 _+0.27, 0.62 _+0.30, 0.50 +- 0.25 g respectively). Conclusion: SR 142801 exerts a potent antidiarrheal effect on CD toxininduced diarrhea. These results suggest that diarrhea induced by CD toxins in guinea pigs, is mediated by NK3 receptors activation, resulting from tachykinin and particularly NKB release. This research was funded by institutional support of INRA and ESAP.
• G1726 DIFLOUROMETHYLORNITHINE INHIBITS CRYPT FISSION. J.S. Thomnson, S.K. Saxena, J.G. Sharp. Dept of Surgery and Cell Biology and Anatomy, Omaha, VAMC and University of Nebraska Medical Center, Omaha, NE Crypt fission is a physiologic mechanism of crypt reproduction. It increases in pathophysiologic situations where intestinal regeneration is required e.g. radiation injury. Polyamine metabolism is important in the regulation of intestinal growth and recovery from injury in response to a variety of stimuli. The aim of the present study was to determine whether inhibition of polyamine synthesis by DFMO influences crypt fission. 48 rabbits under patch enteroplasty in the terminal ileum. One group served as controls and the other took 2% DFMO orally. Animals (n-6) from each group were sacrificed at 7, 14, 21 and 28 days. Normal ileum adjacent to the enteroplasty was studied. Crypt dissection was performed 2 hours after vincristine I.P. to determine crypt cell production rate (CCPR), crypt depth(CD), and proportion of bifurcating crypts(fission). Control Fission 7days
CCPR
DFMO CD
Fission 4.0±2.4*
CCPR
CD
7.9±0.6
251±3
ll.0±2.2
7.5±0.9
244±9
14days 34.0±4.0
14.9±0.7
241±7# 13.0±1.2"
12.6±0.7 354±31"#
21days
10.0±0.9
9.4±0.6
206±7
10.4±0. A6
9.0±0.8
292±5*
28 days 16.0±6.6
9.7±1.3
254±36
10.5±2.5
8.6±0.6
268±13
* P < 0.05 vs control # P < 0.05 vs 21 days DFMO administration decreased crypt fission. There was a corresponding increase in crypt depth. CCPR was similar in both groups. Mucosal ornithine decarboxylase activity (379 -+ 122 vs 39 -+9 sp. act. at 21 days, P < 05) and polyamine content were significantly lower in the DFMO group at 14 and 21 days (7.4±1.4 vs 1.6-+7 and 8.7-+.9 vs 48-+1.0 pmol/mg, control vs DFMO, P < .05). Conclusions: (1) DFMO administration inhibits crypt fission in stimulated intestinal epithelium. (2) This effect correlates temporally with reduced polyamine production. (3) Reduced crypt fission is another potential mechanism of inhibition of intestinal growth by altered polyamine metabolism.