Xanthine oxidoreductase exists in two functionally distinct forms, xanthine dehydrogenase (XDH, D-form) and xanthine oxidase (XOX, O-form). Under normal conditions, the larger part of the enzyme occurs in an NAD⁺-dependent dehydrogenase form that produces NADH and urate. The dehydrogenase can be transformed under various (patho)physiological conditions to an oxygen-dependent oxidase which produces reactive oxygen species (oxygen radicals and/or hydrogen peroxide) and urate.

The physiological role of the enzymes is still not completely understood. Some authors have ascribed a bactericidal function to xanthine oxidoreductase, whereas others have proposed an anti-oxidant function based on the production of urate. Moreover, the physiological role of the O-form of xanthine oxidoreductase (xanthine oxidase) has been discussed. Van den Munckhof described the possible function of this enzyme in the oesophagus and tongue, in the skin and ejková et al. (2001) in the cornea. These authors suggested that under normal conditions, xanthine oxidase is secreted by the epithelial cells and may play a role in anti-bacterial defence by generating superoxide anions and/or hydrogen peroxide.

Xanthine oxidoreductase and xanthine oxidase have been demonstrated in various tissues, including liver, skin, heart, kidney, and intestine. ejková and Lojda (1996) and ejková et al. (2002) demonstrated xanthine oxidase activity in rabbit and human cornea, and in human and bovine retina. ejková and Lojda (1996) and ejková et al. (2001) described that reactive oxygen species generated by xanthine oxidase may contribute to the damage of rabbit corneas irradiated with UV-B light. ejková et al. (1998) hypothesized that xanthine oxidase is present in the tear fluid and is involved in additional damage of superficial layers of corneal epithelium after prolonged wearing of contact lenses, long-lasting hypoxia, and rapid reoxygenation of the cornea after contact lens removal. discussed the involvement of xanthine oxidase in acute uveitis, whereas suggested that xanthine oxidase is involved in corneal damage after excimer laser therapy. Cekic et al (1999) hypothesized that xanthine oxidase has damaging effects on the lens in alloxan-induced diabetic rats and Kuriyama et al (2001) suggested that toxic oxygen species generated by xanthine oxidase are involved in retinal ischemia–reperfusion injury.

To examine the possible participation of xanthine oxidase-generated reactive oxygen species in oxidative eye damage, we decided to localize enzyme histochemically and immunohistochemically xanthine oxidoreductase and xanthine oxidase in the epithelium of bovine, pig, guinea-pig, and rat corneas, where these enzymes have not yet been demonstrated.

Bovine, pig, guinea-pig, and rat corneas of normal eyes were employed (6 corneas in every group). Immediately after the death of the animals (ages of the animals were as follows: cows, 12 years; pigs, six months; guinea-pigs, 2 months; rats, 3 months), the anterior eye segments were cut out, quenched in light petroleum and chilled with an acetone dry-ice mixture. Cryostat sections of 10 ?m thickness were cut on a cryostat (Leica CM 1900; Leica Instruments GmbH, Heidelberg, Germany) and transferred to glass slides.

Xanthine oxidase activity was detected by capturing hydrogen peroxide by cerium ions as described by. Shortly; sections were fixed in 2% glutaraldehyde at 4°C for 1 min, rinsed in distilled water and incubated at 37°C for 3 h. The incubation medium contained 100 mM Tris–maleate buffer (pH 8.0), 10 mM cerium chloride, 100 mM sodium azide, and 0.5 mM hypoxanthine. The visualization step was performed by incubating sections for 25 min at 37°C in 100 mM sodium acetate buffer (pH 5.3), containing 42 mM cobalt chloride, 100 mM sodium azide, 1.4 mM diaminobenzidine, and 0.6 mM hydrogen peroxide. After visualization, sections were mounted in glycerol jelly. Controls were performed by incubating in the medium without substrate or in the presence of substrate and 1 mM allopurinol, an inhibitor of xanthine oxidase.

Xanthine oxidoreductase activity was demonstrated by the tetrazolium salt method as modified by Frederiks and Bosch (1995). Shortly, unfixed sections were incubated at 37°C for 30 min in a medium consisting of 100 mM phosphate buffer (pH 8.0) containing 18% PVA, 0.5 mM hypoxanthine, 1 mM NAD, 5 mM tetranitro BT and 0.45 mM phenazine methosulphate (PMS). After incubation, sections were washed in 100 mM phosphate buffer (pH 5.3, 60°C), rinsed in distilled water and embedded in glycerol jelly. Hypoxanthine and/or NAD were omitted from the incubation medium in control incubations.

For the immunohistochemical localization of xanthine oxidoreductase and xanthine oxidase, the following primary antibodies were used: polyclonal rabbit anti-bovine xanthine oxidase antibody (Chemicon International, Temecula, CA, USA), polyclonal rabbit anti-human xanthine oxidase antibody (Biogenesis, Dorset, UK) and monoclonal mouse anti-human xanthine oxidase/xanthine dehydrogenase/aldehyde oxidase antibody (Lab Vision, Fremont, CA, USA). Unfixed cryostat sections were postfixed in acetone at 4°C for 4 min. Subsequently, anti-mouse/rabbit-HRP/DAB Ultravision Detection System (Lab Vision) was employed as recommended by the manufacturer: hydrogen peroxide block (12 min), ultra V block (5 min), primary antibody incubation (60 min), biotinylated goat anti-mouse/rabbit incubation (10 min), and streptavidin peroxidase incubation (12 min). The visualization was performed using freshly prepared DAB substrate/chromogen solution. The optimal dilution of the primary antibodies was 4 ?g/ml, and the incubation in a humidity chamber lasted for 60 min at room temp. After visualization, sections were mounted in Aquatex (Merck, Darmstadt, Germany). In controls, sections were incubated in the absence of primary antibody.

Xanthine oxidoreductase and xanthine oxidase (both the proteins and their enzymatic activities) were present in the corneal epithelium of bovine (1–4), pig (5–8), guinea-pig (9–12), and rat (13–16) eyes. As compared with the histochemical demonstration of xanthine oxidoreductase activity in bovine, pig, and rat corneal epithelium, the immunohistochemical demonstration of xanthine oxidoreductase in corneal epithelium of the same animals revealed more intense staining (compare 1 with 3, 5 with 7, and 13 with 15). Only in guinea-pig corneal epithelium are the enzyme histochemical and immunohistochemical staining patterns nearly identical (compare 9 with 11). It should be mentioned that the antibody used for the demonstration of xanthine oxidoreductase localized not only xanthine oxidoreductase, but also aldehyde oxidase in corneal epithelium. In immunohistochemical studies, polyclonal rabbit anti-human xanthine oxidase antibody and polyclonal rabbit anti-bovine xanthine oxidase antibody were used for the detection of xanthine oxidase in the corneal epithelium of all animals investigated. We detected xanthine oxidase protein in the epithelium of bovine, pig, guinea-pig, and rat corneas with both antibodies. We did not obtain significant differences in staining pattern using these two antibodies. The xanthine oxidase protein was demonstrated in 2, 6, 10 and 14 with the bovine xanthine oxidase antibody.