Changes in lysyl oxidase (LOX) distribution and its decreased activity in keratoconus corneas

Abstract

Inadequate cross-linking between collagen lamellae is a characteristic feature of keratoconus corneas. The formation of covalent bonds between collagen and elastin fibrils, which maintain the biomechanical properties of the cornea, is mediated by the cuproenzyme lysyl oxidase and four lysyl oxidase-like enzymes. The aim of this study was to determine the distribution of lysyl oxidase and the total lysyl oxidase activity (lysyl oxidase and the four lysyl oxidase-like enzymes) in control and keratoconic corneas. Seven control and eight keratoconic corneas were used for the imunohistochemical detection of lysyl oxidase in corneal cryosections using two different antibodies. The total lysyl oxidase activity in the culture medium of corneal fibroblasts from six explanted keratoconic and four control corneas was measured using a fluorometric assay in the presence and absence of the lysyl oxidase inhibitor beta-aminopropionitrile and determined as the production of H(2)O(2) in nM per μg of total protein. In the control tissue, the most intense signal for lysyl oxidase was present in the corneal epithelium, in which perinuclear dots brightly projecting from more or less homogenous cytoplasmic staining may represent the lysyl oxidase propeptide. Less intense staining was present in keratocytes, the extracellular matrix and in the corneal endothelium. The epithelium of the limbus and the perilimbal conjunctiva showed intense to very intense staining. The distribution of lysyl oxidase was clearly decreased in at least five of the eight keratoconic specimens. The most marked signal reduction was observed in the stromal matrix and in keratocytes. Moreover, the signal in pathological specimens revealed a more irregular pattern, including the presence of intra- and extracellular clumps in the epithelium. Interestingly, endothelial cells showed no or very weak staining in areas just beneath negative stromal tissue. The mean activity of total lysyl oxidase in the keratoconic samples (2.60 ± 2.23 nM H(2)O(2)/μg of total protein) was more than 2.5-fold lower than in control tissue (6.83 ± 2.53 nM H(2)O(2)/μg of total protein), and the decrease was statistically significant (p = 0.0178). The location of lysyl oxidase in the healthy cornea, limbus and perilimbal conjunctiva was described. We hypothesize that the restricted lysyl oxidase distribution in keratoconic corneas, and particularly the decrease of total lysyl oxidase activity in cultured keratoconic fibroblasts, is one potential reason for the inadequate collagen cross-linking that is a hallmark of this disease.