Figure 1.
Mast cell-deficient Sash mice are highly susceptible to NSAID-induced peptic ulcers.
A. Mean weight data is shown for 13 Sash and 10 wild type (WT) mice exposed to 200 ppm piroxicam in their diet. SEMs ranged from 0.4–1.6% of pre-treatment weight; error bars were omitted for clarity. The rapid weight loss correlates with decreased intake of food and water. Euthanasia for humane reasons was typically required on days 4–6. * indicates p≤0.05 relative to WT mice. Piroxicam-exposed Sash mice demonstrated marked gastric distension by gas and fluid (B, bottom half of panel C) compared with Sash mice not exposed to piroxicam (labeled “control” in panel C). Histologic examination showed ulceration at the gastroduodenal junction (E) of Sash mice, with marked inflammatory cell infiltration and edema that led to gastric outlet obstruction. WT mice (D) do not typically develop ulcers or changes in stomach size with piroxicam exposure.
Figure 2.
Mast cell-deficient mice are more susceptible to peptic ulceration when exposed to piroxicam.
Significantly more Sash mice developed grossly evident gastric outlet obstruction defined as gastric enlargement >4× normal (A) and histologically detected peptic ulcers (B) when exposed to piroxicam, compared with WT or Tnf−/− mice. Both the increased sensitivity to peptic ulcers and gastric outlet obstruction appeared to be mast cell-dependent, since they reversed when Sash mice were reconstituted with mast cells (Sash+GFP-MC), but were not TNF-dependent since Tnf −/− mice had a similar incidence of gastric ulceration and gastric outlet obstruction as WT mice. * indicates p≤0.05 relative to WT.
Figure 3.
Mast cells in various stages of degranulation are present at the site of peptic ulceration.
Hematoxylin and eosin- (panel A) and toluidine blue-stained sections (panels B–F) show a peptic ulcer located at the junction between the stomach (s) and duodenum (d). Arrowheads denote the extent of the ulcerated region. The intense purple staining adjacent to the arrowheads in panel B is derived from necrotic debris, inflammatory cells, and large numbers of metachromatic granules. Panel C shows the boxed region from panel B at higher magnification. Arrows point out mast cells in various stages of degranulation, as shown in panels D, E, F. Bar represents 200 µm in panels A and B and 100 µm in panel C.
Figure 4.
Effect of antihistamines on piroxicam-induced peptic ulceration.
A. Treatment with the H1 receptor antagonist diphenhydramine increased histologically detected peptic ulcers in WT mice exposed to piroxicam (black bars; n = 7–9 per group), to a level comparable in mast cell-deficient Sash mice treated with piroxicam alone (gray bar; n = 18). Diphenhydramine did not affect ulcer incidence in Sash mice exposed to piroxicam (nor shown; n = 6; p<0.50). B. Treatment with the H2 receptor antagonist ranitidine (R; dose = 1.8 mg/ml in drinking water) did not affect the incidence of piroxicam-induced peptic ulcer in either WT (black bars) or Sash mice (gray bars). The numbers of mice studied were: WT (n = 13); WT+R (n = 9); Sash (n = 18); Sash+R (n = 7). For both panels, * indicates p<0.05 compared to WT mice treated with piroxicam alone.
Figure 5.
Acid inhibition with the proton pump inhibitor omeprazole does not protect mast cell-deficient mice from development of NSAID-induced peptic ulceration or gastric outlet obstruction.
* indicates p<0.05 compared with WT mice treated with piroxicam alone. Gastric distension by large amounts of gas is readily apparent (arrowhead) in this gross photograph of an omeprazole-reated Sash mouse on day 5 of piroxicam treatment.
Figure 6.
NSAIDs differentially affect apoptosis and proliferation of gastric epithelial cells in vitro.
A. All three NSAIDs tested decreased proliferation of AGS cells compared with vehicle (DMSO) alone, as measured by changes in CFSE incorporation over 3 days of culture, with the largest decrease seen in piroxicam-treated cells (* indicates p<0.05 vs. vehicle; ** p≤0.04 vs. vehicle, ketorolac, or etodolac). B. Survival of AGS cells measured by a tetrazolium dye-based assay is significantly decreased by exposure to piroxicam, but not to keterolac or etodolac. Data shown (mean ± SD; n = 12) was obtained after 4 hrs of treatment. * indicates p<0.05 compared with DMSO control. C. Treatment with the NSAIDs ketorolac or etodolac, but not piroxicam, significantly increased apoptosis as indicated by increased caspase 3 and 7 activity, shown here at 24 hrs (mean ± SD; n = 3).