Role of cyclooxygenase enzymes in a murine model of experimental cholera

Abstract

Nonsteroidal anti-inflammatory drugs (e.g., indomethacin) inhibit and reduce the fluid secretion responses elicited by cholera toxin (CT), but it has not been conclusively determined which cyclooxygenase (COX) isoform is involved in CT's action. This study evaluated the role of the COX enzymes and their arachidonic acid metabolites in experimental cholera. Swiss-Webster mice were dosed with celecoxib and rofecoxib and challenged with CT in ligated small intestinal loops, and intestinal segments from mice deficient in COX-1 and COX-2 were challenged with CT. The effects of CT on fluid accumulation, prostaglandin E(2) production, mucosal tissue injury, and markers of oxidative stress were measured. Celecoxib and rofecoxib given at 160 micro g per mouse inhibited CT-induced fluid accumulation by 48% and 31%, respectively, but there was no significant difference among cox-1(-/-) and cox-2(-/-) mice in response to CT compared to wild-type controls. CT elevated tissue levels of oxidized glutathione and lipid peroxides and elicited small intestinal tissue injury in two of five cox-1(-/-) and four of five cox-2(-/-) mice. A role for COX-2 in CT's mechanism of action has previously been suggested by the effectiveness of COX-2 inhibitors in reducing CT-induced fluid secretion, but CT challenge of COX-1 and COX-2 knockout mice did not corroborate the pharmacological data. The results of this study show that CT induced oxidative stress in COX-deficient mice and suggest a tissue-protective role for arachidonic acid metabolites in the small intestine against oxidative stress.

FIG. 1.

Celecoxib and rofecoxib, selective inhibitors of COX-2, reduced fluid accumulation in CT-stimulated murine ligated intestinal loops. Groups (n = 6) of 6- to 8-week-old female Swiss-Webster mice underwent intestinal loop ligation. Loops were stimulated with CT (1 μg/loop). Celecoxib or rofecoxib was injected intraperitoneally at the time of challenge at the indicated doses. After a standard 6-h incubation period, mice were necropsied and fluid volume was determined. The vertical bars indicate one standard error above and below the arithmetic mean. The asterisk indicates a significant difference (P < 0.05) as determined by Dunnett's multiple-group comparison test.

FIG. 2.

CT challenge of cox-1 and cox-2 gene-deficient mice. Groups of five 6- to 8-week-old cox-1^−/−, cox-2^−/−, and wild-type 129Ola/C57BL6 female mice underwent intestinal loop ligation. Loops were stimulated with cholera toxin (1 μg/loop) and returned to the peritoneal cavity. Fluid volume (A) and luminal PGE₂ content (B) were determined upon necropsy after a 3- or 6-h incubation period. PGE₂ content was determined by ELISA on clarified fluids. The vertical bars indicate one standard error above and below the arithmetic mean.

FIG. 3.

Intestinal architecture after a 3-h (A) and 6-h (B) CT challenge of wild-type, cox-1^−/−, and cox-2^−/− mice. Groups (n = 5) of 6- to 8-week-old cox-1^−/− and cox-2^−/− mice and wild-type 129Ola/C57BL6 female mice underwent intestinal loop ligation. Loops were then injected with CT (1 μg/loop) or PBS and returned to the peritoneal cavity. After 3 h (A) or 6 h (B), the mice were sacrificed by cervical dislocation and intestinal loops were dissected and submitted for histochemistry (hematoxylin-eosin stains). (A) Wild-type mice after PBS (panel 1) and CT (panel 2) challenges. There is no alteration of the villous architecture or the intestinal wall. cox-1^−/− mice after PBS (panel 3) and CT (panel 4) challenges. After CT challenge, there is mild broadening of intestinal villi. Inflammatory cells are slightly increased in both sections. No evidence of necrosis or edema is present. The same changes are also present in cox-2^−/− mice after PBS (panel 5) and CT (panel 6) challenges. (B) Wild-type mice after PBS (panel 1) and CT (panel 2) challenges. The villi reveal mild to moderate edema and moderate inflammation of the lamina propria with villous broadening in animals challenged with PBS. These changes are most likely due to the ligation procedure. In animals challenged with CT, there is evidence of marked villous edema, inflammation, and broadening (arrowheads). No evidence of mucosal necrosis is present. In cox-1^−/− mice after PBS (panel 3) and CT (panel 4) challenges and cox-2^−/− mice after PBS (panel 5) and CT (panel 6) challenges, the intestinal mucosa revealed marked edema of villi with inflammation and focal villous necrosis (arrows) in animals challenged with CT. Animals challenged with PBS showed the same changes as described for wild-type animals challenged with PBS. Magnification, ×200.

FIG. 3.

Intestinal architecture after a 3-h (A) and 6-h (B) CT challenge of wild-type, cox-1^−/−, and cox-2^−/− mice. Groups (n = 5) of 6- to 8-week-old cox-1^−/− and cox-2^−/− mice and wild-type 129Ola/C57BL6 female mice underwent intestinal loop ligation. Loops were then injected with CT (1 μg/loop) or PBS and returned to the peritoneal cavity. After 3 h (A) or 6 h (B), the mice were sacrificed by cervical dislocation and intestinal loops were dissected and submitted for histochemistry (hematoxylin-eosin stains). (A) Wild-type mice after PBS (panel 1) and CT (panel 2) challenges. There is no alteration of the villous architecture or the intestinal wall. cox-1^−/− mice after PBS (panel 3) and CT (panel 4) challenges. After CT challenge, there is mild broadening of intestinal villi. Inflammatory cells are slightly increased in both sections. No evidence of necrosis or edema is present. The same changes are also present in cox-2^−/− mice after PBS (panel 5) and CT (panel 6) challenges. (B) Wild-type mice after PBS (panel 1) and CT (panel 2) challenges. The villi reveal mild to moderate edema and moderate inflammation of the lamina propria with villous broadening in animals challenged with PBS. These changes are most likely due to the ligation procedure. In animals challenged with CT, there is evidence of marked villous edema, inflammation, and broadening (arrowheads). No evidence of mucosal necrosis is present. In cox-1^−/− mice after PBS (panel 3) and CT (panel 4) challenges and cox-2^−/− mice after PBS (panel 5) and CT (panel 6) challenges, the intestinal mucosa revealed marked edema of villi with inflammation and focal villous necrosis (arrows) in animals challenged with CT. Animals challenged with PBS showed the same changes as described for wild-type animals challenged with PBS. Magnification, ×200.

FIG. 4.

Comparison of CT, inactive CT, CTB, and lipopolysaccharide challenges of cox-2^−/− mice intestinal segments. Mice deficient in the cox-2 gene (n = 2) were challenged with 1 μg of heat-inactivated CT (panel 1), CT B subunit (panel 2), or lipopolysaccharide (panel 3). After a standard 6-h incubation, mice were sacrificed, and 1-cm small intestine sections were fixed in 10% buffered formalin. Intestinal sections were stained with hematoxylin and eosin and examined for tissue injury.

FIG. 5.

Levels of oxidized glutathione and lipid peroxide metabolites in CT-challenged intestinal segments from cox-1^−/− and cox-2^−/− mice. Groups (n = 5) of 6- to 8-week-old cox-1^−/−, cox-2^−/−, and wild-type 129Ola/C57BL6 female mice underwent intestinal loop ligation. Loops were stimulated with cholera toxin (1 μg/loop) and returned to the peritoneal cavity. After a 6-h incubation period, mice were sacrificed by cervical dislocation and 1-cm intestinal sections were removed from the peritoneal cavity and assayed for oxidized glutathione (A) and lipid peroxides (B). Error bars represent one standard error above the mean. Statistical significance was determined by the Tukey test.