Michael V. Riley

Anion-sensitive ATPase in rabbit corneal endothelium and its relation to corneal hydration.

The activity of Mg2+-activated ATPase in the corneal endothelium of the rabbit is dependent on the anionic composition of the medium and is not inhibited by ouabain. Maximum anion-dependent Mg2+ ATPase activity was demonstrated in the presence of HCO3− ion, followed by acetate, Cl−, SO42−, SCN− and OCN−. The maximum rates of ATP hydrolyzed in the presence of Mg2+, Mg2++HCO3− and Mg2++Na++K+ were 5·61±0·7, 9·03±0·8 and 6·74±0·7 μmol hr−1 mg−1 protein, respectively. The major part of each of these ATPase activities was found in a crude mitochondrial fraction. Activity in the presence of HCO3− was inhibited by 65% when 50 mm-SCN− or 10 mm-OCN− was added to the medium. These concentrations of inhibitor, when added to a medium perfusing intact, isolated corneas, caused a deterioration of their ability to transport fluid out of the tissue. The results indicate that the anion-sensitive ATPase plays an important quantitative role in the energy metabolism of the endothelium and suggest that the postulated active bicarbonate transport across the cell layer may be influenced by this enzyme.

ATPases of ciliary epithelium: Cellular and subcellular distribution and probable role in secretion of aqueous humor

The distribution of ion-stimulated ATPases of the ciliary epithelium has been examined in tissues from bovine and rabbit eyes. In homogenates of tissues from both species, both Na,K- and anion-stimulated enzyme activities were found, but no K,H-stimulated activity was detected. The anion ATPase had a broad specificity for a number of anions, and was strongly inhibited by thiocyanate. Following separation of pigmented (outer) and non-pigmented (inner) layers of the bovine ciliary epithelium and isolation of the two cell types on density gradients, higher activities of both Na,K- and anion ATPases were found in the non-pigmented cells. Subcellular fractionation of a mixed population of cells showed that the anion ATPase was almost exclusively associated with a mitochondrial fraction, rather than with the plasma-membrane fraction containing the Na,K-ATPase. These results confirm histochemical studies of the distribution of Na,K-ATPase in the ciliary epithelium and support the concept that the inner, non-pigmented cell layer is chiefly responsible for the active transport of ions into the posterior chamber. It is concluded that this transepithelial transport can be driven only by the energy derived via the Na,K-ATPase, and that any subsequent anion or proton transport in the formation of aqueous humor is driven by the sodium gradient through exchange mechanisms.

Interactions of ascorbate and H2O2: implications for in vitro studies of lens and cornea

The interaction of ascorbate, hydrogen peroxide and oxygen has been examined in order to understand the equilibrium between these compounds that exists in the aqueous humor of the eye and their influence on function of the corneal endothelium. Ascorbate was found to promote corneal swelling when isolated corneas were perfused with a medium lacking glucose. This was found to be due to the rapid oxidation of ascorbate in the medium, yielding H2O2 which is toxic to the endothelial cells. In the absence of oxygen, or if EDTA was added to the medium, no H2O2 was produced from ascorbate, but ascorbate reacted with any pre-existing H2O2. Oxidation of ascorbate in the aqueous humor is limited by the presence of glutathione (and, possibly, other compounds) and no significant increase in H2O2 concentration occurs on standing in air. Nevertheless, the concentration of H2O2 in the aqueous is directly dependent on the concentration of ascorbate secreted in the aqueous humor. Therefore, there must be a dynamic equilibrium in this fluid between ascorbate, H2O2 and oxygen, and it may be modulated by glutathione. Each of these substances is important in redox reactions, including free-radical production or scavenging. Consequently, when studying the effects on corneal or lenticular function of other agents which cause or relieve oxidant stress, it is critical that the modifying effects of ascorbate and H2O2, as they occur in vivo, be considered. A perfusion system is described which permits an approximation in vitro of stable concentrations of ascorbate, H2O2, GSH and O2 similar to those found in the aqueous humor.

Relation of intracellular levels and redox state of glutathione to endothelial function in the rabbit cornea

Abstract The correlation between extracellular or intracellular glutathione and endothelial cell function was investigated in this study. The level of oxidized and reduced glutathione was measured in fresh or perfused corneal endothelium and Descemets membrane with the method of Tietze. From our present study, it is clearly shown that normal corneal hydration can be maintained over a wide range of intracellular glutathione concentrations. However, fluid transport in the cornea is greatly disturbed when intracellular glutathione levels drop to about one-third of the in vivo value. Moreover, when intracellular glutathione is either completely oxidized, with diamide, or completely reduced, with dithiothreitol, rapid swelling of the cornea is observed. This result suggests that there may exist an intracellular redox buffer system for the maintenance of normal hydration in the cornea.

Regional distribution of calcium in alloxan diabetic rabbit lens

A diabetic rabbit model was developed for investigation of cataractogenesis and other changes in the anterior segment. Rabbits were fasted, injected with 0.7 mg/kg alloxan, fed 1% glucose solution for 24 hrs and returned to a normal diet. Animals showing and maintaining blood glucose of greater than 300 mg% within two days were used in this study. Concomitant with increase in blood glucose was a rise in aqueous humor glucose and osmolality, together with a decrease in ascorbate concentration. Vacuoles and small discrete opacities developed, and in some cases, at longer time periods complete opacity of anterior or posterior aspects was found. Total calcium content of the whole lens increased up to 2-fold, especially after 60 days, and was correlated with a decrease in lens transmittance of a He/Ne laser beam and also with high osmolality of the aqueous humor. Free calcium was six-fold higher in opaque areas than clear areas, and was 100-fold higher in vacuoles. It is suggested that, in addition to the recognized role in sugar cataractogenesis of osmotic stress due to sorbitol accumulation in the lens, changes of intracellular calcium in localized areas of the lens and stresses imposed by changes in aqueous humor osmolality may also be important.

The role of the epithelium in control of corneal hydration.

Abstract A study has been made, in vitro, of the thickness of corneas whose epithelial surfaces were exposed to a variety of aqueous solutions. Corneas were isolated and mounted in such a way that damage to the endothelium by wrinkling or stretching was virtually eliminated. It was found, when freshly isolated corneas were perfused on the endothelial surface with a Ringer-type medium and the epithelium was covered with either the same mediu,, or with 0·15 m NaCl, that the normal thickness of the cornea could be maintained for periods of up to 6 hr. Under the same conditions, pre-swollen corneas (obtained by storing the enucleated eye at 5°C for 18 hr) deturgesced over a 3–4-hr period and maintained normal thickness for a further 3–4 hr. Identical results were obtained when the medium bathing the epithelial surface was substituted by solutions of sucrose, mannitol, urea, LiCl, MgSo 4 or choline chloride. The results confirm that the epithelium behaves as a semipermeable membrane, but provide no support for the postulate that the control of corneal thickness is dependent upon active transport of sodium ions by the epithelium from the tear film to the stroma. It is concluded that the primary role of the epithelium in regard to corneal hydration is as a barrier preventing access of the tear fluid to the stroma and that the site of active metabolic control of hydration is in the endothelium.

Pump and leak in regulation of fluid transport in rabbit cornea

Rabbit corneas were isolated, denuded of epithelium, and perfused on the anterior and posterior surfaces with Krebs Ringer-bicarbonate with additions of 50 microM H2O2, 125 microM BCNU, or 100 microM ouabain. The permeability of the corneal endothelium to labelled mannitol and inulin was determined by adding these compounds to the endothelial perfusate and measuring the rate of appearance of radioactivity in the anterior perfusate. Both H2O2 and BCNU increased the flux of mannitol and inulin across the endothelium in a time dependent manner, but ouabain had no effect. Additions of glucose with H2O2 or of GSH with BCNU prevented the observed changes in permeability. ATPase activities in the endothelia of intact, isolated corneas were also determined following incubation in the same media. The only observable effects of H2O2 and BCNU were slight reductions in the activity of Na+ + K+ ATPase. It is concluded that permeability changes, the leak, are more critical than active transport processes, the pump, in determining the rate and extent of swelling that results from exposure of the cornea to these agents.

Pyridine nucleotides of rabbit cornea with histotoxic anoxia: Chemical analysis, non-invasive fluorometry and physiological correlates

The pyridine nucleotides from both the epithelium and the endothelium of rabbit cornea were measured by the cycling assay. Sodium azide (10 mM) applied for 1 hr to induce histotoxic anoxia decreased the endothelial NAD+/NADH ratio from 4.62 to 1.49 and decreased the epithelial NAD+/NADH ratio from 2.56 to 1.08. The larger NAD+/NADH ratio for the endothelium as compared to the epithelium corresponds to a more oxidized state. The corresponding ratios for NADP+/NADPH were 1.2 for the endothelium and 0.70 for the epithelium. Sodium azide had no effect on the NADP+/NADPH ratio for the endothelium, but decreased the epithelial ratio to 0.62. Pyridine nucleotide fluorescence was measured with a difference corneal fluorometer on the perfused whole cornea preparation and the perfused everted corneal preparation. Sodium azide (10 mM) for 30 min resulted in a 19.4 +/- 0.7% increase in the pyridine nucleotide fluorescence from the whole corneal preparation and a 4.5 +/- 0.6% increase from the everted endothelial preparation. Corneal anoxia induced by stopping the perfusion on the endothelial side resulted in a 18.7 +/- 0.6% increase in pyridine nucleotide fluorescence for the whole corneal preparation. Sodium azide (10 mM) resulted in a 35% decrease in the transendothelial potential difference and a 76% decrease in the rate of transendothelial fluid transport. A comparison is made between invasive chemical analysis and real time, non-invasive fluorometry to measure histotoxic corneal anoxia.

Toxic effects of hydrogen peroxide on corneal endothelium

We have measured the effects of hydrogen peroxide on the ability of the isolated rabbit cornea to maintain normal hydration during perfusion, on the structure of the endothelial cells, and on the redox state of glutathione and the activity of the hexose monophosphate shunt in these cells. Swelling of isolated corneas was immediate and severe upon exposure to 50 microM H2O2 in the absence of glucose, or to 200 microM H2O2 in the presence of glucose. The presence of glucose with 50 microM H2O2 delayed the onset of swelling for two hours and markedly decreased its severity. The concentration of glutathione in the endothelium, and its redox state, were unaltered by 50 microM H2O2 in the presence of glucose, but in its absence glutathione loss was significant and the fraction in the oxidized state was greatly increased. H2O2 was removed from the medium by reaction with the cornea at a concentration-dependent rate. It was calculated that the reaction rate at 50 microM H2O2 was about twice that which could be accounted for by the increased yield of NADPH generated by the stimulation of the hexose monophosphate shunt. H2O2 may also react with endothelial cell membranes, which could account for the marked changes in cell structure seen in the scanning electron microscope.

Fluid and ion transport in corneal endothelium: insensitivity to modulators of Na(+)-K(+)-2Cl- cotransport.

The role of Na+-K+-2Cl-cotransport in ion and fluid transport of the corneal endothelium was examined by measuring changes in corneal hydration and uptake of86Rb by the endothelial cell layer. Isolated, intact rabbit corneas maintain normal hydration when they are superfused at the endothelial surface with bicarbonate ([Formula: see text])-Ringer solutions as a result of equilibrium between active ion and fluid transport out of the stromal tissue and leak of fluid into stromal tissue from the aqueous humor. Furosemide and bumetanide did not alter this equilibrium when they were added to the superfusion medium. Uptake of86Rb by the endothelium of the incubated cornea was increased 25% by bumetanide, but uptake in the presence of ouabain (70% less than that of controls) was not changed by bumetanide. In Na+-free medium, uptake of 86Rb was reduced by 58%, but it was unchanged in Cl--free medium. Calyculin A, a protein phosphatase inhibitor and activator of Na+-K+-Cl-cotransport, was without effect on86Rb uptake. Hypertonicity (345 mosmol/kg) increased uptake slightly, whereas hypotonicity (226 mosmol/kg) caused a 33% decrease. Neither of these changes was significantly different when bumetanide was present in the media. It is concluded that Na+-K+-2Cl-cotransporter activity is not exhibited by the in situ corneal endothelium and does not play a role in the ion and fluid transport of this cell layer. Its presence in cultured endothelial cells may reflect the reported importance of this protein in growth, proliferation, and differentiation.