Endothelium Of Cornea Research Articles

Viscoelastic protection from endothelial damage by air bubbles

Purpose: To determine whether viscoelastic materials effectively protect the corneal endothelium from air bubbles. Setting: Department of Ophthalmology, Yonsei University College of Medicine, Seoul, Korea. Methods: Human eye-bank and rabbit eyes had a standardized phacoemulsification procedure with or without viscoelastic material (Healon® [sodium hyaluronate 1.0%], Healon GV® [sodium hyaluronate 1.4%], or Viscoat® [chondroitin sulfate 4.0%−sodium hyaluronate 3.0%]). The integrity of the endothelium was examined after the procedure with F-actin staining and scanning electron microscopy. Rabbit eyes with and without viscoelastic material (Healon or Viscoat) had a standardized irrigation/aspiration (I/A) procedure. The mucinous layer of the endothelium was examined after the procedure with transmission electron microscopy. Results: In the phacoemulsification experiment without viscoelastic material, with Healon, and with Healon GV, the endothelium of human and rabbit corneas had many areas of cell loss in a pattern consistent with air-bubble damage. With Viscoat, endothelial cells remained intact. In the I/A experiment, the mucinous layer of Viscoat-exposed rabbit endothelium appeared thinner. In the same experiments without viscoelastic material or with Healon, the mucinous layer of the endothelium appeared normal. Conclusions: Viscoat effectively protected the endothelium from air-bubble damage. Viscoat appears to protect the endothelium by acting as a physical barrier. Its adherence is probably related to the way it interacts with the mucinous layer of the endothelium.

Distribution of cyclosporin A in the cornea after topical or oral administration.

The purpose of this study was to determine levels of cyclosporin A (CsA) in ocular tissues and fluids after oral or topical administration of CsA in humans requiring immunosuppression for penetrating keratoplasty and to evaluate the sensitivity of a high performance liquid chromatography coupled to electrospray mass spectrometry (LC-MS). High-risk penetrating keratoplasty was performed on 11 patients requiring 5 mg/kg of oral cyclosporin A and 9 patients with topical 0.5% (suspension, equivalent to 5 mg/ml)) preoperative immune suppression. Venous blood samples, as well as aqueous samples, were obtained for some patients, and the central 6.0-8.5 mm corneal buttons were removed and dissected into cornea epithelium, stroma and endothelium and analyzed for CsA levels by LC-MS. The mean values (concentration) of the specimens analyzed (obtained from 11 patients on oral CsA) showed 2.45 +/- 2.12 and 0.35 +/- 0.49 ng CsA/mg tissue in the epithelium and endothelium, respectively. A much smaller concentration, 0.13 +/- 0.14 ng CsA/mg tissue was observed in the stroma. There was a fair amount of CsA in the serum at the time of surgery but no detectable CsA in all aqueous humor samples analyzed. In cases of topical application, the mean concentration of CsA was similarly highest in the epithelium, 64.29 +/- 39.92 ng CsA/mg tissue, followed by the endothelium, 9.98 +/- 6.55 ng CsA/mg tissue, with the lowest levels found in the stroma. Topical application of CsA is able to achieve high levels of CsA in the corneal epithelium, higher than that achieved from oral administration of CsA.

Gap junctional communication in the human corneal endothelium and epithelium

Purpose. Gap junctional communication in the epithelium and endothelium of human corneas was studied. The influence of corneal storage on endothelial gap junctions was also examined. Methods. Donor human corneal cells were injected with carboxyfluorescein while surrounding cells were monitored for traces of fluorescence. Dye-spread coefficients were measured in corneal endothelial cells. Western blot and immunohistochemical analysis of the endothelium and epithelium was employed to determine if connexin 43 was present. Results. Dye coupling occurs in both the epithelium and endothelium of the human cornea. Epithelial dye coupling was extensive in the basal layers but less apparent in the superficial layers. Endothelial dye coupling was similar to that reported for rabbit corneas. Western blot and immunohistochemical analyses demonstrated the presence of connexin 43 in both cell types. Conclusions. Gap junctional communication occurs in the endothelium and epithelium of human corneas, and both cell types express connexin 43. These results are similar to previous rabbit studies, thereby strengthening the use of the rabbit cornea as a gap junction model.

The role of apoptosis in the pathogenesis of Fuchs endothelial dystrophy of the cornea.

To investigate the potential role of apoptosis in the pathogenesis of Fuchs endothelial dystrophy of the cornea. Twenty-one corneal buttons from patients with Fuchs dystrophy and 15 control corneas were studied. Apoptosis was assessed by the in situ end-labeling of double-stranded DNA breaks, and by immunohistochemical characterization of cellular markers associated with apoptosis (Fas, FasL, Bcl-2, and Bax). Expression of Bcl-2 and Bax mRNA in the corneal stroma and endothelium was separately analyzed by a semiquantitative reverse transcriptase polymerase chain reaction. Furthermore, cultivated keratocytes generated from diseased corneal buttons and donor rims were exposed to camptothecin, an apoptotic inducer, for 6 and 24 hours. They were then examined for protein and messenger RNA (mRNA) expression of apoptotic regulatory molecules. DNA fragmentation was seen in the epithelium, stroma, and endothelium in 6 of 7 corneas with Fuchs dystrophy. A statistically significant difference was identified in the expression of Bax and its mRNA in the stroma, but not in the endothelium of Fuchs dystrophy corneas. Following exposure to camptothecin, keratocytes from patients with Fuchs dystrophy responded with an increased level of Bax and a low level of Bcl-2. This trend was distinctively different from the response of normal keratocytes. The evidence in this study points to a disease-related disturbance in the regulation of apoptosis in Fuchs dystrophy. Our findings suggest that excessive apoptosis may be an important mechanism in the pathogenesis of Fuchs dystrophy.

Cornea-lens transdifferentiation in the anuran, Xenopus tropicalis.

Previously, the only anuran amphibian known to regenerate the lens of the eye was Xenopus laevis. This occurs during larval stages through transdifferentiation of the outer cornea epithelium under control of factors presumably secreted by the neural retina. This study demonstrates that a distantly related species, X. tropicalis, is also able to regenerate lenses through this process. A transgenic line of X. tropicalis was used to examine the process of cornea-lens transdifferentiation in which green fluorescent protein (GFP) is expressed in differentiated lens cells under the control of the Xenopus gamma1-crystallin promoter element. Unlike X. laevis, the process of cornea-lens transdifferentiation typically occurs at a very low frequency in X. tropicalis due to the rapid rate at which the inner cornea endothelium heals to recover the pupillary opening. The inner cornea endothelium serves as a key physical barrier that normally prevents retinal signals from reaching the outer cornea epithelium. If this barrier is circumvented by implanting outer cornea epithelium of transgenic tadpoles directly into the vitreous chamber of non-transgenic X. tropicalis larval eyes, a higher percentage of cases formed lenses expressing GFP. Lenses were also formed if these tissues were implanted into X. laevis larval eyes, suggesting the same or similar inducing factors are present in both species. When pericorneal ectoderm and posteriolateral flank ectoderm were implanted into the vitreous chamber, only in rare cases did pericorneal ectoderm form lens cells. Thus, unlike the case in X. laevis, competence to respond to the inducing factors is tightly restricted to the cornea epithelium in X. tropicalis. As controls, all these tissues were implanted into the space located between the inner and outer corneas. None of these implants, including outer cornea epithelium, exhibited GFP expression. Thus, the essential inductive factors are normally contained within the vitreous chamber. One explanation why this type of lens regeneration is not seen in some other anurans could be due to the rapid rate at which the inner cornea endothelium heals to recover the pupillary opening once the original lens is removed. These findings are discussed in terms of the evolution of this developmental process within the anurans.

Transferrin receptor-mediated gene transfer to the corneal endothelium.

The application of gene therapy to prevent allograft rejection requires the development of noninflammatory vectors. We have therefore investigated the use of a nonviral system, transferrin-mediated lipofection, to transfer genes into the cornea with the aim of preventing corneal graft rejection. Rabbit and human corneas were cultured ex vivo and transfected with either lipofection alone or in conjunction with transferrin. The efficiency of transfection, localization, and kinetics of marker gene expression were determined. Strategies to increase gene expression, using chloroquine and EDTA, were investigated. In addition to a marker gene, a gene construct encoding viral interleukin 10 (vIL-10) was transfected and its functional effects were examined in vitro. Transferrin, liposome, and DNA were demonstrated to interact with each other, forming a complex. This complex was found to deliver genes selectively to the endothelium of corneas resulting in gene expression. Treatment of corneas with chloroquine and EDTA increased the transfection efficiency eight-fold and threefold, respectively. We also demonstrated that constructs encoding vIL-10 could be delivered to the endothelium. Secreted vIL-10 was shown to be functionally active by inhibition of a mixed lymphocyte reaction. Our data indicate that transferrin-mediated lipofection is a comparatively efficient nonviral method for delivering genes to the corneal endothelium. Its potential for use in preventing graft rejection is shown by the ability of this system to induce vIL-10 expression at secreted levels high enough to be functional.

Effects of penetrating keratoplasty rejection on the endothelium of the donor cornea and the recipient peripheral cornea

PURPOSE: To determine the effects of penetrating keratoplasty rejection on the recipient endothelium. METHODS: Twenty transplanted corneas of 20 keratoconus patients were enrolled in this study. Ten of the corneas had undergone allograft rejection and the other 10 grafts had an uneventful postsurgical course. The endothelium of the donor cornea and the endothelium of the recipient peripheral cornea were evaluated by wide-field specular microscopy. RESULTS: The transplanted corneas that underwent allograft rejection showed a marked decrease in endothelial cell density not only in central donor cornea but also in the recipient peripheral cornea compared with that in the uneventful transplanted corneas. Additionally, the recipient endothelium had significant pleomorphism in the rejection group. There were no morphologic differences in the peripheral donor cornea between the control group and the rejection group. CONCLUSIONS: Our findings indicate that the recipient peripheral endothelium is also affected by allograft rejection, perhaps because of the contribution of recipient endothelium to the wound healing process after allograft rejection.

Kir2.1 Potassium channels and corneal epithelia

To determine the existence of inward rectifier (Kir2.1) sequences in mRNA from corneal epithelium and endothelium. cDNA library construction, cloning, PCR, patch clamp. RESULTS. Both the corneal epithelium and endothelium contain mRNA for Kir2.1 inwardly rectifying potassium channels. When the cDNA is transfected into Chinese hamster ovary cells, the channel currents match those expected from Kir2.1 inward rectifiers. The mRNA for Kir2.1 potassium channels exists in corneal epithelium and endothelium. Therefore, Kir2.1 inwardly rectifying potassium channels probably make some contributions to the resting voltages of cornea epithelium and endothelium. Previous data, however, suggest that they are probably not the dominant contributors in these preparations.

Clinical features and keratoplasty results in keratoconus complicated by acute hydrops

PURPOSE: Acute hydrops develops when Descemet's membrane and the endothelium of the ectatic cornea separates to allow aqueous humor to enter the stroma. The current study was done to determine if penetrating keratoplasty (PK) in keratoconus patients with resolved hydrops has a poor prognosis. METHODS: A retrospective chart review of 35 eyes of 35 patients with keratoconus who underwent PK after the resolution of acute hydrops (Group 1) and of 74 eyes of 69 patients who had PK without a previous history of acute hydrops (Group 2) was performed. The chi-square test was used for statistical analysis. RESULTS: The mean age in Group 1 was 21.1±11.0 years. The mean age in Group 2 was 23.2±14.4 years. Vernal keratoconjunctivitis was present in 11 patients in Group 1 (31%) and in 12 (17.4%) patients in Group 2. Loose sutures were present in two patients (6%) in Group 1; both resulted in loose suture-related corneal neovascularization. Loose sutures developed in five patients (7%) in Group 2. In Group 1, three patients (9%) had microbial keratitis. Microbial keratitis was seen in five patients (7%) in Group 2. Two patients (6%) in Group 1 developed late endothelial graft failure. In Group 2, one eye (1%) developed early and one eye (1%) late epithelial graft rejection; five eyes (7%) developed late endothelial graft rejection. There was no statistically significant difference between the two groups with respect to the incidence of graft rejection. At the last examination, 34 eyes in Group 1 had clear grafts; in the second group, graft clarity was achieved in 71 eyes. CONCLUSION: Penetrating keratoplasty in hydrops is successful in terms of graft clarity and visual outcome in patients with keratoconus after the resolution of hydrops. The patients should have preoperative and postoperative anti-allergic treatment and close follow-up for possible complications.

Kir2.1 Potassium channels and corneal epithelia

Purpose. To determine the existence of inward rectifier (Kir2.1) sequences in mRNA from corneal epithelium and endothelium. Methods. cDNA library construction, cloning, PCR, patch clamp. Results. Both the corneal epithelium and endothelium contain mRNA for Kir2.1 inwardly rectifying potassium channels. When the cDNA is transfected into Chinese hamster ovary cells, the channel currents match those expected from Kir2.1 inward rectifiers. Conclusions. The mRNA for Kir2.1 potassium channels exists in corneal epithelium and endothelium. Therefore, Kir2.1 inwardly rectifying potassium channels probably make some contributions to the resting voltages of cornea epithelium and endothelium. Previous data, however, suggest that they are probably not the dominant contributors in these preparations.

Functional human corneal equivalents constructed from cell lines.

Human corneal equivalents comprising the three main layers of the cornea (epithelium, stroma, and endothelium) were constructed. Each cellular layer was fabricated from immortalized human corneal cells that were screened for use on the basis of morphological, biochemical, and electrophysiological similarity to their natural counterparts. The resulting corneal equivalents mimicked human corneas in key physical and physiological functions, including morphology, biochemical marker expression, transparency, ion and fluid transport, and gene expression. Morphological and functional equivalents to human corneas that can be produced in vitro have immediate applications in toxicity and drug efficacy testing, and form the basis for future development of implantable tissues.

The spatial organization of apical junctional complex-associated proteins in feline and human corneal endothelium

Purpose. Previous studies suggest that proteins associated with the apical junctional complex (AJC) play essential roles in the development, maintenance and regulation of barrier function in transport epithelium and vascular endothelium. The goal of this study is to identify and determine the spatial organization of several major AJC-associated proteins in normal human and feline corneal endothelium. Methods. Fresh corneal tissue was obtained from 4 recipient buttons removed during penetrating keratoplasty (two from keratoconus patients, and two from patients with post-traumatic stromal scarring) as well as from 16 cat eyes. En bloc double- and triple-labeling of corneas was performed using phalloidin, and mouse, rat or rabbit antibodies to ZO-1, occludin, pan-cadherin, a-catenin, ß-catenin and plakoglobin (?-catenin). The 3-D localization of the proteins was then determined in situ using laser confocal microscopy. Results. Similar staining patterns were obtained for the corneal endothelium of normal cat corneas and fresh human buttons. Apically, f-actin was arranged into dense peripheral bands (DPB) in individual cells that were separated from those in adjacent cells. Diffuse phalloidin staining also extended from the DPB into the cytoplasm apically. Although weaker, phalloidin staining also appeared to be associated with the basolateral cell borders. The adherens junction protein, cadherin, formed a thin pericellular band at the apical cell junctions between the DPB. In addition, cadherin staining also appeared to extend along the basolateral cell borders in a convoluted pattern. Staining for a-catenin, ß-catenin and plakoglobin each showed a nearly identical organization as cadherin. ZO-1 formed a single apical band that was localized between the DPB; however, no basolateral ZO-1 staining was detected. Interestingly, the distribution of ZO-1 was discontinuous around the cell, with the largest gaps occurring at the Y-junctions between adjacent endothelial cells. Positive staining for occludin was not detected in either human or feline corneal endothelium. Conclusions. The composition and organization of the AJC of corneal endothelium appears to be different from that of classical transport epithelia; these findings may be related to functional differences between these two cell types.

The complications of contact lens wear.

Out of the approximately 30 million contact lens wearers in North America, about 6%--approximately 1.8 million individuals--develop a complication each year [Acta Ophthalmol 1984, 62:556-565]. These problems range from self-limiting to sight threatening, and require rapid diagnosis and treatment to prevent vision loss. Contact lens complications are as varied as they are common, involving the lids, conjunctiva, and all layers of the cornea (epithelium, stroma, and endothelium).

The effects of intraocular lidocaine on the corneal endothelium

The study aimed to evaluate the direct effect of intraocular lidocaine hydrochloride (HCl) 1% on corneal endothelial cell function, ultrastructure, and viability using an in vitro perfusion specular microscope system. Paired rabbit and human corneas were isolated and mounted in an in vitro specular microscope for endothelial perfusion evaluation. Corneas were perfused with a control solution (BSS Plus for humans, glutathione bicarbonate Ringer's [GBR] for rabbits) for a 1-hour stabilization period. After the stabilization period, one cornea of each matched pair was perfused with preservative-free lidocaine HCl 1% for 15 minutes followed by control solution for an additional 2 to 3 hours. The control cornea continued to receive either GBR or BSS Plus. Corneal thickness measurements were taken every 15 minutes throughout the perfusion period. Corneal swelling and deswelling rates were calculated by linear regression analysis. At the end of the experiment, corneas were fixed for scanning and transmission electron microscopy. In another group of corneas, the endothelial viability was assayed after direct perfusion with lidocaine HCl 1%. Lidocaine HCl 1% caused endothelial cell edema, which reversed on removal of lidocaine from perfusion media. Corneal swelling and deswelling rates did not differ significantly between the lidocaine and control groups. Electron microscopy showed the effects of transient endothelial cell edema with an otherwise normal mosaic pattern and ultrastructure for both treatment groups. Endothelial cell viability was maintained after the direct lidocaine exposure and a 2-hour washout. Lidocaine HCl 1% causes a transient endothelial cell edema to the in vitro perfused endothelium of human and rabbit corneas. Proper attention should be given to the type of lidocaine injected intraocularly (i.e., concentration, vehicle, preservatives, pH, osmolarity). Although lidocaine HCl 1% appears to be safe to both human and rabbit endothelium during short-term in vitro exposure, further in vivo and in vitro studies are needed to determine long-term effects of intraocular lidocaine on the corneal endothelium.

Ion transporters and receptors in cDNA libraries from lens and cornea epithelia

Purpose. To construct tissue-specific cDNA libraries containing copies of genes of ion transporting and receptor proteins in lens and cornea epithelia. To screen the libraries for the cDNA of these molecules and to deduce the protein sequence from the cDNA sequence.Methods. Lens and corneal cDNA libraries are not commercially available and are difficult to prepare from microdissected single animal eyes or cadaver human eyes due to the small amount of tissue. To overcome these problems, we refined an approach for preparing high efficiency, non-PCR amplified plasmid cDNA libraries, where only nanogram quantities of poly (A) RNA are required. Plasmids were electrotransformed into DH10B bacteria and routinely yielded >106 independent colonies with typically >90% recombinants. Randomly selected colonies were subjected to colony PCR analysis to determine the libraries average insert size (typically ≊1-kb). Primers to known genes were used in PCR amplification to check for representation of the genes in the cDNA libraries.Results. Using libraries from rabbit cornea epithelium and endothelium, cultured human lens epithelium, and αTN4 cells, we have found the libraries to contain the cDNA for 3 common housekeeping proteins expected to be at high copy numbers in all cells. In addition, we identified 22 rare proteins and for many we determined the complete coding regions. These include K+ channels, Cl− channels, Ca++ channels, Na+ channels, three exchangers, the Na+-HCO3− cotransporter, and three kinds of receptors.Conclusion. Cornea and lens libraries have been constructed from single cornea or lens preparations. The libraries often contain the cDNA sequences for ion transporting and receptor proteins which are expected to be relatively rare in these cells. Many of these cDNA's have now been sequenced and the encoded proteins identified. Curr. Eye Res. 17:708–719, 1998.

Ion transporters and receptors in cDNA libraries from lens and cornea epithelia

Purpose. To construct tissue-specific cDNA libraries containing copies of genes of ion transporting and receptor proteins in lens and cornea epithelia. To screen the libraries for the cDNA of these molecules and to deduce the protein sequence from the cDNA sequence.Methods. Lens and corneal cDNA libraries are not commercially available and are difficult to prepare from microdissected single animal eyes or cadaver human eyes due to the small amount of tissue. To overcome these problems, we refined an approach for preparing high efficiency, non-PCR amplified plasmid cDNA libraries, where only nanogram quantities of poly (A) RNA are required. Plasmids were electrotransformed into DH10B bacteria and routinely yielded >106 independent colonies with typically >90% recombinants. Randomly selected colonies were subjected to colony PCR analysis to determine the libraries average insert size (typically ≊1-kb). Primers to known genes were used in PCR amplification to check for representation of the genes in the cDNA libraries.Results. Using libraries from rabbit cornea epithelium and endothelium, cultured human lens epithelium, and αTN4 cells, we have found the libraries to contain the cDNA for 3 common housekeeping proteins expected to be at high copy numbers in all cells. In addition, we identified 22 rare proteins and for many we determined the complete coding regions. These include K+ channels, Cl− channels, Ca++ channels, Na+ channels, three exchangers, the Na+-HCO3− cotransporter, and three kinds of receptors.Conclusion. Cornea and lens libraries have been constructed from single cornea or lens preparations. The libraries often contain the cDNA sequences for ion transporting and receptor proteins which are expected to be relatively rare in these cells. Many of these cDNA's have now been sequenced and the encoded proteins identified. Curr. Eye Res. 17:708–719, 1998.

Heterogeneous expression of transketolase in ocular tissues

PURPOSE. Previous studies have shown that transketolase is preferentially expressed in the corneal epithelium and comprises up to 10% of the soluble protein of the mature mouse cornea. The aim of this study is to evaluate the expression and distribution of TKT in the different ocular tissues. METHODS. We have used in situ hybridization and immunohistochemistry to localize TKT mRNA and protein in the developing and adult mouse eye. RESULTS. TKT were found to be widely distributed throughout the adult mouse eye. Among the ocular tissues examined, the corneal epithelium exhibited the highest levels of TKT mRNA and protein. Within the epithelial layer, TKT mRNA and protein were differentially distributed with the highest expression occurring in basal cells and the lowest in apical cells, suggesting that TKT expression in the corneal epithelium may be differentiation-related. Enriched expression of TKT was also found in the cornea endothelium, lens epithelium, ciliary body, and iris. Low basal levels of expression were observed in the limbus and conjunctiva. In contrast to the adult eye, TKT expression in the one-day-old mouse eye was homogeneous at low, but detectable levels, suggesting that TKT expression is developmentally regulated in the cornea as well as in the other ocular tissues. In the healing corneal epithelium, TKT expression in the single cell layer of the leading edge was completely suppressed until the cells began to stratify, at which point TKT expression increased markedly. CONCLUSIONS. The results presented here suggest that TKT is differentially expressed and developmentally regulated in the various tissues that comprise the eye.

A histochemical study of the distribution of dextran 500 in human corneas during organ culture

PURPOSE. The aim of this histochemical study was to demonstrate the absorption of dextran 500 and its distribution in the cornea after storage under standard eye bank conditions. Furthermore, an attempt was made to distinguish between the soluble and insoluble parts of dextran 500 absorbed by the cornea, in order to see how much dextran remains in the cornea after transplantation. METHODS. Forty-nine fresh and 65 organ-cultured human corneas were investigated. The corneas were cultured for 28 days in a dextran-free medium, followed by a period of 1–14 days in a medium containing 5% dextran 500. Cryosections were stained by aqueous PAS and a modified alcoholic PAS to determine the amount of dextran. RESULTS. Dextran was not found in the epithelium, stroma or endothelium of fresh human corneas. By contrast, extra- and intracellular positive staining reactions were detected in corneas following storage in a medium containing dextran. Dextran 500 absorption was relatively diffuse in the epithelium after storage in a dextran medium. Initial accumulations were found in the stroma near Bowman's and Descemet's membranes and also in the central part of the cornea, as the period of culture in the medium containing dextran lengthened. We also observed interaction between the stroma and endothelium: decreasing amounts in the endothelium were followed by an increase of same in the stroma. Intracellular deposits of dextran were detected after only one day. A much greater part of the extracellular dextran than previously described was found to be insoluble. CONCLUSIONS. As the amount of dextran in the cornea increases over a longer storage time, we conclude that the period of storage in a medium containing dextran should be limited to four days. The fact that the cornea is saturated with dextran after seven days has been shown in further studies to interfere with mitochondrial function and may also cause severe post-operative swelling of the transplant, hence leading to a longer recovery period for the patient.

Ocular toxicity of intravitreous transforming growth factor-beta 1.

To evaluate the ocular toxicity of intravitreous transforming growth factor-beta 1 (TGF-beta(1)). Eyes of New Zealand white rabbits were injected intravitreally with TGF-beta(1) in doses of 20, 10, 5, 1 or 0.5 micrograms. Electroretinography and clinical examination were performed before and after the injection. At either 10 or 30 days after injection, the eyes were enucleated and examined histologically. The endothelium of fresh corneas, which had received 20 micrograms TGF-beta(1) or nothing, were checked 5 days after injection. Corneal opacity was caused by doses > or = 10 micrograms in 7 of 7 eyes; by doses of 5 micrograms in 3 of 4 eyes; and by doses of 1 microgram in 1 of 4 eyes. Doses > or = 5 micrograms also caused pannus formation in 15 of 16 eyes. Ten days after injection there was some decrease in B-wave amplitude on electroretinography. Thirty days after injection the electroretinographic responses had recovered, except in the eyes injected with 20 micrograms TGF-beta(1), which still showed a decrease in step 1. Histological examination of paraffin sections did not demonstrate significant change. Corneal thickness of the eyes receiving 20 micrograms TGF-beta(1) was twice that of the normal cornea and some corneal endothelial cells lost their hexagonal shape. TGF-beta(1) injected intravitreally at doses > or = 1 micrograms is toxic to the cornea. Retinal function is only slightly disturbed by 20 micrograms of TGF-beta(1).

Interferon gamma regulates platelet endothelial cell adhesion molecule 1 expression and neutrophil infiltration into herpes simplex virus-infected mouse corneas.

In a mouse model of herpes simplex virus (HSV) 1 corneal infection, tissue destruction results from a CD4+ T cell-mediated chronic inflammation, in which interleukin 2 and interferon (IFN) gamma are requisite inflammatory mediators and polymorphonuclear leukocytes (PMN) are the predominant infiltrating cells. In vivo neutralization of IFN-gamma relieved inflammation at least in part through a specific block of PMN extravasation into HSV-1-infected corneas. Intercellular adhesion molecule (ICAM) 1 and platelet endothelial cell adhesion molecule (PECAM) 1 were upregulated on the vascular endothelium of inflamed corneas. Reduced PMN extravasation in anti-IFN-gamma-treated mice was associated with a dramatic reduction of PECAM-1 but not ICAM-1 expression on vascular endothelium. PMN accumulated in the lumen of corneal vessels after in vivo IFN-gamma neutralization. PECAM-1 was readily detectable on PMN inside the vessels but was not detectable on PMN that extravasated into the infected cornea. Moreover, flow cytometric analysis revealed reduced PECAM-1 expression but elevated major histocompatibility complex class I expression on PMN that recently extravasated into the peritoneal cavity when compared with PMN in the peripheral blood. We conclude that IFN-gamma contributes to HSV-1-induced corneal inflammation by facilitating PMN infiltration; this appears to be accomplished through upregulation of PECAM-1 expression on the vascular endothelium; and PMN downregulate PECAM-1 expression during the process of extravasation.