The transparent cornea is made up of an anterior stratified epithelium, a collagenous stromal matrix containing fibroblasts called keratocytes, Descemet's membrane and a single-layered endothelium. While the extracellular stroma has been studied intensively with respect to corneal transparency, relatively little attention has been given to the molecular basis for transparency of the corneal cells. Experiments have shown that corneal epithelial cells, like lens cells, contain unexpectedly high concentrations of a few metabolic enzymes. In the mouse, aldehyde dehydrogenase 3 (ALDH3) and transketolase (TKT) comprise 30-50% of the water-soluble protein of the corneal epithelium, an amount reminiscent of lens crystallins. In the last few years we have cloned the mouse ALDH3 cDNA and 5? flanking region of the gene, and initiated transgenic mouse experiments to identify the corneal-specific cis-control elements. This year we reported a 4 kbp sequence composed of 1 kbp of 5? flanking region, exon 1 and intron 1 of the ALDH3 gene will direct the chloramphenicol acetyltransferase (CAT) gene selectively to the corneal epithelium (with minor expression in the liver). This is the first corneal specific promoter fragment that can function with tissue specificity in transgenic mice and opens the door to molecular analyses of corneal development and disease. The TKT gene is also expressed at least 50 times more highly in the cornea than in other tissues, although it is essentially ubiquitously expressed throughout the body. In contrast to the ALDH3 promoter fragment, the TKT promoter was unable to direct corneal gene expression in transgenic mice. The TKT and ALDH3 genes have numerous inducible cis-control consensus sequences in their 5'flanking regions consistent with the possibility that induction by the environment (i.e. uv light) contributes significantly to their high expression in the cornea. This fits with our observation that ALDH3 and TKT expression is minimal in the mouse corneal epithelial cells until after eye opening 14 days after birth. Moreover, corneal TKT was induced several fold by exposing mice raised in the dark for 25 days after birth to light and in explanted eyes exposed to light derived from new- born mice. The use of inductive mechanisms involving the environment differs from the strictly developmental control processes used by the lens for the high, preferential expression of crystallin genes. Considerable time and effort has been spent this year to create transgenic mice that lack the ALDH3 or TKT gene. At present the constructs have been made and the gene is being recombined in embryonic stem cells with the mutant transgenes. The ultimate purpose of these knockout experiments is to demonstrate that enzymes may play strictly structural roles related to transparency in the cornea. Last year we showed that at least 50% of the water-soluble protein in the zebrafish cornea appeared to be gelsolin, a protein known to bind and cleave actin. This was exciting since it suggested that gelsolin is a new corneal ?crystallin? due to its abundance and since a gelsolin point mutation in humans has been associated with lattice type II corneal dystrophy. This year we have established that this protein is gelsolin by cDNA cloning and sequencing, that it is the intracellular form of the protein, and that it accumulates in the epithelial cells of the zebrafish cornea. Experiments have led to the development of a number of cell lines from mouse corneal epithelial cells. These contain a temperature sensitive simian virus 40 T-antigen that is driven by a promoter that is activated by gamma-interferon. At 33 degrees in the presence of gamma-interferon the cells proliferate, while at 37 degrees without gamma-interferon the cells reduce their growth rate. At present the cells are being tested for the induction of ALDH3 and TKT under reduced growth conditions. If these cells behave as differentiating corneal epithelial cells, they will be invaluable for studying the expression of the ALDH3, TKT and other corneal genes. Previous experiments have identified a 208 base pair promoter fragment that functions specifically in the lens of transgenic mice. This year we have shown that this lens-specific region also functions in the corneal epithelial cells of the transgenic mice, further linking the relationships of crystallins to the lens and cornea. Finally, in collaborative experiments with Dr. James Jester (University of Texas Southwestern Medical Center) we have shown that ALDH1 and TKT comprise over 30% of the water-soluble protein of the transparent rabbit corneal keratocytes. The keratocytes became reflective and specifically lost more than half of their ALDH1 and TKT after freeze-injury, a situation that persisted for several weeks and was not associated with dying cells. The abundance of ALDH1 and TKT in the keratocytes exceeds their expected role for catalysis per se and extends the concept of enzyme-crystallins to these fibrocytes.
Showing the most recent 10 out of 14 publications
