Intestinal oxidative damage in inflammatory bowel disease: semi-quantification, localization, and association with mucosal antioxidants

Abstract

Intestinal inflammation is accompanied by excessive production of reactive oxygen and nitrogen metabolites. In order to counteract their harmful effects, the intestinal mucosa contains an extensive system of antioxidants. It has previously been shown that the levels of and the balance between the most important antioxidants are seriously impaired within the intestinal mucosa from inflammatory bowel disease (IBD) patients compared with normal mucosa. The present study investigated the consequences of this antioxidative imbalance by evaluating parameters of oxidative stress-related mucosal damage in the same tissue samples. The extent of apoptosis, peroxynitrite-mediated protein nitration (3-NT), and lipid peroxidation were assessed in relation to the expression of nitric oxide synthase (NOS) and the superoxide-producing enzyme xanthine oxidase (XO). In addition, bi- and multi-variate regression analyses were performed to associate these parameters with the levels of the antioxidants assessed previously. Apoptotic cell death was visualized by TUNEL staining in luminal epithelium of normal controls, and in IBD additionally in the inflammatory infiltrate and in deeper parts of the crypts, but its frequency was unrelated to the severity of inflammation. In Crohn's disease (CD), epithelial apoptosis levels were strongly associated with the expression of XO, implying a role for this enzyme in the regulation of epithelial cell homeostasis, although its levels were unaffected by intestinal inflammation and were comparable to those in normal control mucosa. 3-NT immunoreactivity was substantially increased in luminal crypt cells, neutrophils, and mononuclear cells in the inflamed mucosa of ulcerative colitis (UC) patients. The inflamed IBD luminal epithelium, but not the inflammatory cells, also contained increased amounts of NOS. The immunoreactivity of both 3-NT and NOS was significantly higher in UC than in CD. Unexpectedly, the increased 3-NT expression in UC was associated with neutrophilic myeloperoxidase and not with NOS, which suggests that 3-NT is formed in areas with a dense neutrophilic infiltrate via a peroxynitrite-independent oxidation pathway. Lipid peroxidation, as estimated by the malondialdehyde (MDA) concentration, was elevated in both the inflamed CD and the inflamed UC mucosa, and was identified in the luminal epithelium using a histochemical technique. In CD, lipid peroxidation was independently associated with the concentration of metallothionein and with Mn-superoxide dismutase activity, suggesting the involvement of hydroxyl radicals and superoxide anions. In UC, however, the amount of MDA was associated with epithelial catalase expression and neutrophilic myeloperoxidase activity, suggesting a hydrogen peroxide- and/or hypochlorous acid-mediated mechanism. The present study underlines the importance of oxidative stress in the pathogenesis of IBD and provides clues regarding the (anti)oxidants involved which indicate that this process evolves through diverging pathways in CD and UC.