Liver response to indomethacin-induced intestinal injury.

The aim of our study was to evaluate the impact of impaired barrier function of the small intestine induced by indomethacin on biochemical markers of liver damage (serum levels of alanine aminotransferase, aspartate aminotransferase, bilirubin), and liver functional parameters (serum concentration of albumin, liver DNA synthesis). Indomethacin (Sigma) was administered in 2 injections in a dose of 7.5 mg/kg subcutaneously spaced 24 hours apart, rats were sacrificed 24, 48 or 72 hours after the second dose of indomethacin. Control rats received indomethacin vehicle (5% NaHCO3, pH 7.4, 1.0 ml/kg) in the same manner. Small intestine injury was approved by increased permeability (measured as a lactulose-mannitol index). Significant increase of small intestine DNA synthesis (estimated by incorporation of 3H thymidine) in indomethacin-treated rats 48 (p < 0.01) and 72 (p < 0.05) hours after the second dose of indomethacin documents induction of reparative process. All biochemical markers of liver injury were significantly decreased in indomethacin treated rats in all recorded intervals (p < 0.05). By contraries, serum concentration of albumin, which predicates about liver function, was in indomethacin-treated rats significantly decreased in all intervals (p < 0.01). To explain these contrarious results of indomethacin-induced impaired barrier function of the small intestine on the liver deserves further studies.

riods (6:00 a.m. to 6:00 p.m.). The animals had free access to standard laboratory rat chow (DOS 2B, Velaz, Prague, CR) and tap water. All experiments used in this study were approved by Animal Use and Care Committee of the Charles University, Faculty of Medicine in Hradec Králové. Indomethacin (Sigma) was administered in 2 injections in a dose of 7.5 mg/kg subcutaneously spaced 24 hours apart (1). Control rats received indomethacin vehicle (5 % NaHCO 3 , pH 7.4, 1.0 ml/kg) in the same manner. Control rats were sacrificed by exsanquination from abdominal aorta 24 hours after the second dose of vehicle, experimental animals divided into three groups were sacrificed 24, 48 and 72 hours after the second dose of indomethacin. Each group involves 10 animals.
For the estimation of DNA synthesis, the rats were labelled by an intravenous injection of 3 H thymidine (200 µCu/kg b.w. one hour before sampling). The radioactivity of samples was measured according to Short et al. (16) using a Delta 300 scintillation counter (Nuclear Chicago). The liver and intestine DNA concentrations were determined by the diphenylamine reaction (7) from which the specific activity of DNA was calculated. Activities of alanine aminotransferase (ALT), aspartate aminotransferase (AST) and bilirubin concentration were estimated on automatic analyser Modular Roche. Malondialdehyde (MDA) concentration was measured according to Hendrix and Assman (9) by thiobarbituric test. Another 24 rats divided into four groups (2 control and 2 experimental) were used for evaluation of intestine permeability. Small bowel per-meability was estimated using lactulose-mannitol test. The test solution (1 ml containing 100 mg of re-crystallised lactulose -VÚFB, Prague, Czech Republic, and 50 mg of D-mannitol -Merck) was administered by stomach tube to the rats after overnight fasting, than the animals were placed into metabolic cages, urine was collected for 6 hours and stabilised by thiomersal (Kulich, Hradec Králové, Czech Republic) (200mg/100 ml of urine). The concentrations of sugars were measured using capillary gas chromatography (Hewlett Packard 5890 II, equipped with flame ionisation detector and HP 3396A integrator) (6). The amounts of excreted sugars were expressed as the ratio of orally given substances.
The results are expressed as mean ± S.E.M. The statistical analyses were based on ANOVA and Student's t-test. Differences were considered as significant if the p value was less than 0,05.

Results
Indomethacin administration induced macroscopically detectable damage in the small intestine, focal ulceration along the mesenteric border vas evident. Small bowel permeability measured as lactulose-mannitol ratio was significantly increased (p< 0.05) 24 hours after second dose of indomethacin compared to control group (Fig. 1). Surprisingly, we found increased permeability for both sugars lactulose and mannitol, nevertheless the increase was higher in case of lactulose ( Fig. 1). Damage of small intestine    was followed by induction of reparative process, which is documented by increased DNA synthesis in the intestine of rats treated with indomethacin (Fig. 2). This increase was statistically significant in the interval 48 hours (p<0.01) and 72 hours (p<0.05) after indomethacin administration.
An unexpected finding concerns biochemical markers of liver damage. As shown in Fig. 3 serum activities of ALT and AST as well as serum concentration of bilirubin were significantly decreased (p< 0.05) in all recorded intervals in rats treated with indomethacin. By contraries serum concentration of albumin, which predicates about liver function was in indomethacin-treated animals significantly decreased in all intervals (p< 0.01) as depicts Fig. 4. Fig. 2 shows the changes in liver DNA synthesis. We found increased values of DNA synthesis after administration of indomethacin, 72 hours after the second dose of indomethacin DNA synthesis was doubled in comparison with control group, nevertheless due to high inter-individual differences this increase was not significant.
Triacylglycerols content in the liver as well as TAG serum concentrations were elevated in all groups treated with indomethacin (p< 0.05) (Fig. 5). Serum concentration of malondialdehyde was not influenced by indomethacin administration (Fig. 4).

Discussion
In the rat, indomethacin-induced enteropathy is characterised by distal jejunal and proximal ileal ulcerations, bowel wall thickening, adhesions, partial obstructions and inflammation (15). The precise mechanisms by which NSAIDs cause intestinal injury have not been fully elucidated. Full expression of NSAID-induced intestinal injury requires luminal bile, enteric bacteria and a high luminal concentration of NSAID.
In our experiments, indomethacin caused macroscopically detectable damage in the small intestine and increased intestinal permeability measured as lactulose-mannitol index. Indomethacin is a potent inhibitor of cyclooxygenase and the decreased production of endogenous prostaglandins could contribute to intestinal damage. The other candidates that can increase intestinal permeability are platelet activating factor (PAF), the most potent ulcerogenic factor, and leukotrienes. NSAIDs increase leukotrienes synthesis by diversion of the arachidonic acid cascade toward the lipo-oxygenase pathway (14). Nevertheless, Mion et al. (13) documented in their experiments in which they studied short term effects of indomethacin on intestinal permeability on a model of rat isolated vascularly perfused terminal ileum that short term effect of indomethacin on intestinal permeability is not a result of changed prostanoid or PAF metabolism. They suggest other mechanisms, which can be involved in pathogenesis of increased intestinal permeability. Namely, a) indomethacin, being a weak acid and a lipophylic molecule, can interfere directly with cell membranes; b) indomethacin has been shown to change synthe-sis of ATP through the anaerobic pathway (8), since epithelial tight junction permeability is energy dependent, lack of ATP can increase permeability; c) indomethacin can increase production of reactive oxygen species that induce lipid peroxidations. Results of our experiments did not confirm participation of increased lipid peroxidation rate. Serum concentrations of malondialdehyde did not change after indomethacin administration. Reactive oxygen species can play local role in the gut, but we did not measure concentration of lipid peroxidation markers in the small intestine.
We tested regenerative response of small intestine to indomethacin injury. DNA synthesis was measured in a 3cm long sections of intestine extending proximally from a point 20 cm proximal to the ileocecal junction. We found increased intestine DNA synthesis in rats treated with indomethacin, the values were statistically significant 48 and 72 hours after the second dose of indomethacin. Our results are in contrast with findings of Uribe et al. (19). They estimated gastrointestinal cell proliferation in sections of rats treated with low and high doses of parenteral indomethacin administration for 3 to 60 days and found proliferative response only in long-term, high-dose indomethacin treatment. Possible explanation of this difference could be higher dose of indomethacin which was used in our acute experiments and which probably caused intestinal injury severe enough to induce proliferative response.
The main aim of our study was to evaluate impact of indomethacin-induced intestinal injury on the liver. Unfortunately, there are not data in literature available concerning this matter. Firstly we focused on the biochemical markers of liver injury. To our surprise, serum activities of ALT and AST and serum concentration of bilirubin were significantly decreased in all rats treated with indomethacin. To explain this finding is difficult without further data. We speculate that this effect of indomethacin could be probably caused by inhibitory effect on cyclooxygenase activity and thus interference with the prostaglandins and cytokines network. On the other hand, the functional capacity of the liver was decreased as documented by significantly diminished serum concentration of albumin in indomethacin treated rats. This supports also finding of increased triacylclycerol content in the liver. The decreased functional capacity of the liver is accompanied by decreased ability to produce proteins required for lipoprotein synthesis. On the increased serum TAG concentration, which was found in all rats treated with indomethacin, could participate both increased intestinal permeability and also decreased ability of the liver to uptake lipids from the plasma. Liver damage is commonly followed by liver regeneration (18). DNA synthesis in the liver of indomethacin treated rats was increased, this increase was not statistically significant.
This is a pilot study concerning the response of the liver to indomethacin-induced intestinal injury. Since NSAIDs are the most widely prescribed of the antirheuma-tic drugs, attesting to their efficacy as analgesic, anti-inflammatory, and antipyretic agents we assume that relations between NSAID-induced intestinal injury and the liver deserves further studies.