Gustav Graff

Secretion of proinflammatory cytokines by human conjunctival epithelial cells.

The production of cytokines by human conjunctival epithelial cells following stimulation was investigated. Primary cultures of human conjunctival epithelial cells were characterized by morphology and keratin expression. Cultured epithelial cells were treated with varying concentrations of lipopolysaccharide, interleukin (IL)-1 beta, calcium ionophore A23187, or phorbol myristate acetate, and cytokine secretion was determined over specified intervals. Culture supernatants and cell lysates were analyzed by ELISA for IL-1 beta, IL-3, IL-4, IL-5, IL-6, IL-8, IL-11, IL-1 receptor antagonist (IL-1ra), tumor necrosis factor-alpha (TNF-alpha), and granulocyte-macrophage colony stimulating factor (GM-CSF). With the exception of IL-1ra, unstimulated conjunctival epithelial cells produced cytokines at relatively low or undetectable levels. IL-1ra was detected in both culture supernatants and cell lysates under basal conditions. In response to stimuli, conjunctival epithelial cells secreted the proinflammatory cytokines TNF-alpha, IL-6, IL-8, and GM-CSF in a dose- and time-dependent fashion. After stimulation, the intracellular levels of IL-1ra increased in these cells but the supernatant-associated levels remained unchanged. None of the other cytokines evaluated (IL-1 beta, IL-3, IL-4, IL-5, and IL-11) were detected in supernatants or lysates of resting or stimulated cells. These findings suggest that conjunctival epithelial cells may contribute to the pathogenesis of human ocular diseases by production of proinflammatory cytokines. Further evaluation of these cells as targets of therapy is warranted.

Improved myeloperoxidase assay for quantitation of neutrophil influx in a rat model of endotoxin-induced uveitis

Previously described models of endotoxin-induced uveitis quantify neutrophil influx into the eye using biochemical or direct cell count methods that result in an underestimation of ocular leukocyte accumulation following the inflammatory stimulus. We have optimized the rat model of endotoxin-induced uveitis by first overcoming interference in the biochemical assay of myeloperoxidase due to endogenous ocular reductants and cellular constituents containing free thiol functional groups. This was accomplished by simultaneously 1) extensively diluting soluble, interfering substances and 2) blocking tissue sulfhydril functional groups during tissue homogenization. Uveitis was induced in rats by subplantar injection of endotoxin. Twenty-four hours later, eyes were enucleated, homogenized, fractionated, and myeloperoxidase activity of neutrophils sedimenting with the membranous pellet was extracted. Previously published extraction procedures yielded only 40% of total assayable myeloperoxidase activity. Optimal recovery of myeloperoxidase activity (>twofold increase) was achieved only with two sequential extractions using 50 mM phosphate buffer (pH 7.4) containing 10 mM N-ethylmaleimide, and subsequent solubilization of myeloperoxidase activity by extraction with 0.5% hexadecyltrimethylammonium bromide in 50 mM phosphate buffer (pH 6.0). This modified extraction procedure and optimized myeloperoxidase assay conditions (300 microM hydrogen peroxide and 1.5 mM o-dianisidine) were then used to enhance the uveitis model. Maximum ocular neutrophil accumulation was observed at endotoxin doses of 100-200 microg. Total ocular neutrophil infiltrations ranged from 250,000 to 800,000 cells/globe. This leukocyte influx was inhibited dose-dependently by topical ocular administration of dexamethasone, with half-maximal inhibition observed at a concentration of 0.01%, w/v. Further validated by the correlation of biochemical results with histological evaluation, the refined methodology described in this report has application in assessing the ophthalmic therapeutic potential of antiinflammatory agents.

The Hydrolysis of the Prostaglandin Analog Prodrug Bimatoprost to 17-Phenyl-trinor PGF2α by Human and Rabbit Ocular Tissue

Bimatoprost (Lumigan), the ethyl amide derivative of the potent prostaglandin FP agonist 17-phenyl-trinor PGF(2alpha), has been reported to be a member of a pharmacologically unique class of ocular hypotensive agents. To confirm that bimatoprost, which is intrinsically active as an FP prostaglandin agonist, is also a prostaglandin analog prodrug, the hydrolysis of bimatoprost by ocular tissues was studied by incubating solutions containing bimatoprost with either human or rabbit ocular tissue. The ethyl amide group of bimatoprost was hydrolyzed by rabbit and human cornea, iris/ciliary body and Thasclera to produce the expected carboxylic acid product, 17-phenyl-trinor PGF(2alpha). The rate of hydrolysis by human and rabbit cornea and iris/ciliary body is similar, whereas the rate of hydrolysis by the sclera is slower in humans than in rabbits. These studies show that human and rabbit ocular tissue (cornea, iris/ciliary body and sclera) can convert bimatoprost to the potent prostaglandin FP agonist 17-phenyl-trinor PGF(2alpha). Separate in vitro studies clearly show that both bimatoprost and 17-phenyl-trinor PGF(2alpha) have affinity for and are agonists at the human FP receptor. Taken together, the data strongly suggests that the ocular hypotensive effect of bimatoprost can be attributed to its activity as a prostaglandin receptor agonist either directly or through its role as a prostaglandin agonist prodrug.

The effect of partial vitrectomy on blood-ocular barrier function in the rabbit

Purpose. To compare ocular vascular permeability in the rabbit after vitrectomy as assessed by contrast-enhanced magnetic imaging (CE-MRI) and measurements of aqueous and vitreous humor protein concentration. Methods. Partial vitrectomies were performed, irrigating with BSS ® or BSS PLUS ®. Post-operative vascular leakage was determined by CE-MRI following intravenous administration of gadolinium diethylenetriaminepentaacetic acid (Gd-DTPA). Aqueous and vitreous protein concentrations were quantified by standard biochemical assay. ERG evaluations were performed on postoperative days 1, 3, and 7. Results. Using BSS as irrigant, breakdown of the inner blood-retinal barrier (BRB) occurred in 4/7 eyes on post-operative day 1. The rate of Gd-DTPA leakage was significantly greater on postoperative day 1 than that in unoperated, control eyes, but declined ˜50% by day 3. At both time points, outer BRB breakdown was restricted to the sclerotomy wounds. No BRB leakage was detectable in control eyes. Blood-aqueous barrier (BAB) leakage was bilateral on day 1. Significantly greater Gd-DTPA leakage occurred in the operated eye than in the nonsurgical contralateral eye. On day 3, ˜40% bilateral reduction in leakage indicated resolution of BAB leakage. Notably, Gd-DTPA leakage of the BAB and BRB was significantly reduced in the BSS PLUS treated group. In contrast to MRI assessments, protein concentrations of the aqueous and vitreous in the surgical eye showed no detectable differences between BSS and BSS PLUS. Concurrent with the transient loss of ocular barrier function, ERG responses also declined. However, by day 7 greater than 90% recovery was noted in BSS PLUS treated animals but not in the BSS treatment group. Conclusions. CE-MRI is capable of detecting subtle changes in vascular permeability following ocular surgery. Advantages of using BSS PLUS compared to BSS as the irrigating solution can be detected using this technique. BSS PLUS’s protection of barrier function is consistent with a rapid recovery in retinal function not observed in BSS treated eyes.