A tissue culture technique for growing corneal epithelial, stromal, and endothelial tissues separately

In Vitro Expansion of Corneal Endothelial Cells for Clinical Application: Current Update

ABSTRACT

Endothelial dysfunction is one of the leading causes of corneal blindness and one of the common indications for keratoplasty. At present, the standard of treatment involves the replacement of the dysfunctional endothelium with healthy tissue taken from a donor. Because there is a paucity of healthy donor tissues, research on the corneal endothelium has focused primarily on expanding these cells in the laboratory for transplantation in an attempt to reduce the gap between the demand and supply of donor tissues for transplantation. To expand these cells, which are nonmitotic in vivo, various mitogens, substrates, culture systems, and alternate strategies have been tested with varying success. The biggest challenge has been the limited proliferative capacity of these cells compounded with endothelial to mesenchymal transition that alters the functioning of these cells and renders them unsuitable for human transplantation. This review aims to give a comprehensive overview of the most common and successful techniques used in the culture of the cells, the current available evidence in support of epithelial to mesenchymal transition (EMT), alternate sources for deriving the corneal endothelial cells, and advances made in transplantation of these cells.

The molecular basis of neurotrophic keratitis

Endogenous proliferation of corneal epithelial cells is regulated by a bidirectional control process characterized by an adrenergic, cAMP-dependent 'off', and a cholinergic, muscarinic cGMP-dependent 'on' response. The adrenergic receptor(s) are located in the plasma membrane (microsomal fraction), whereas the novel feature of the system is a cholinergic receptor specific for acetylcholine (ACH) located in the nuclear membrane. Exogenous substances which raise intracellular cAMP levels such as isoproterenol or PGE1, shut off epithelial mitosis: and, carbamylcholine or ACH raise intranuclear cGMP levels and increase mitosis by specific, regulatory stimulation of RNA-polymerase II activity. We believe that this regulatory system explains the transitory mitotic suppression induced by superficial corneal wounding (interruption of adrenergic fibres, chalone-effect); and the marked, permanent depression of epithelial mitosis associated with decreased intracellular ACH levels which are produced by total corneal denervation, and which results in neurotrophic keratitis.

Corneal epithelial adhesion strength to tethered-protein/peptide modified hydrogel surfaces

In this study, we investigated the suitability of microjet impingement for use on hydrogel materials to determine the cellular adhesion strength of corneal epithelial cells grown on novel hydrogels with extracellular matrix proteins (laminin and/or fibronectin) or a peptide sequence (fibronectin adhesion promoting peptide, FAP) tethered to their surface with poly(ethylene glycol) chains. The deformation of the hydrogel surface in response to the force of the microjet was analyzed both visually and mathematically. After the results of these experiments and calculations determined that no deformation occurred and that the pressure required for indentation (1.25 x 10(6) Pa) was three factors of 10 greater than the maximum pressure of the microjet, the relative mean adhesion strength of primary rabbit corneal epithelial cells grown on the novel poly(2-hydroxyethyl methacrylate-co-methacrylic acid) hydrogels was determined and compared with that of the same type of cells grown on control glass surfaces. Only confluent cell layers were tested. Cells grown on control glass surfaces adhered with a mean relative adhesion strength of 488 +/- 28 dynes/cm2. Under identical conditions, cells grown on laminin- and FAP-tethered hydrogel surfaces were unable to be removed, indicating an adhesion strength greater than 516 dynes/cm2. Cells grown on fibronectin- and fibronectin/laminin (1:1)-tethered surfaces showed significantly lower relative adhesion strengths (201 +/- 50 and 189 +/- 11 dynes/cm2, respectively), compared with laminin- and FAP-tethered surfaces (p = 0.001). Our results demonstrate that the microjet impingement method of cell adhesion analysis is applicable to hydrogel substrates. Additionally, analysis of our test surfaces indicates that fibronectin tethered to this hydrogel in the quantity and by the method used here does not induce stable ligand/receptor bonding to the epithelial cell membrane to the same degree as does laminin or FAP.

Corneal epithelial cell growth over tethered-protein/peptide surface-modified hydrogels

In this study, we investigated the corneal epithelial cell growth rate and adhesion to novel hydrogels with (1) extracellular matrix proteins [fibronectin, laminin, substance P, and insulin-like growth factor-1 (IGF-1)] and (2) peptide sequences [RGD and fibronectin adhesion-promoting peptide (FAP)] tethered to their surface on poly(ethylene glycol) (PEG) chains. The growth rate to confluence of primary rabbit cornea epithelial cells was compared for plain polymethacrylic acid-co-hydroxyethyl methacrylate (PHEMA/MAA) hydrogels, PHEMA/MAA hydrogels coated with extracellular matrix proteins or peptides, and PHEMA/MAA hydrogels with tethered extracellular matrix proteins or peptides on the surface. The development of focal adhesions by the epithelial cells grown on the surfaces was determined by F-actin staining. Little to no epithelial cell growth occurred on the plain hydrogel surfaces throughout the 15-day culture period. Of the coated hydrogels, only the fibronectin-coated surfaces showed a significant increase in cell growth compared to plain hydrogels (p < 0.009). However, even these surfaces reached a maximum of only 20% confluence. Laminin, fibronectin adhesion-promoting peptide (FAP), and fibronectin/laminin (1:1) tether-modified hydrogels all achieved 100% confluence by the end of the culture period, although the rates at which confluence was reached differed. F-actin staining showed that focal adhesions were formed for the laminin, FAP, and fibronectin/laminin tether-modified surfaces. The results support the hypothesis that tethering certain extracellular matrix proteins and/or peptides to the hydrogel surface enhances epithelial cell growth and adhesion, compared with that seen for protein-coated or plain hydrogel surfaces.

The use of a porcine organotypic cornea construct for permeation studies from formulations containing befunolol hydrochloride

The purpose of this study was to develop an organotypic cornea equivalent consisting of three different cell types (epithelial, stromal and endothelial cells) and to investigate its usefulness as in vitro model for permeation studies. The different cell types of a porcine cornea were selectively isolated and a multilayer tissue construct was created step-by-step in Transwell cell culture insert. Histology, basement membrane components (laminin, fibronectin) and surfaces of cornea construct were investigated to evaluate the degree of comparability to porcine cornea from slaughtered animals. The cornea construct exhibited similarities to the original cornea. Ocular permeation of befunolol hydrochloride from different formulations across the cornea construct was tested using modified Franz cells and compared with data obtained from excised cornea. The cornea construct showed a similar permeation behavior for befunolol hydrochloride from different formulations compared with excised porcine cornea. However, permeation coefficients K(p) obtained with the construct were about three to fourfold higher for aqueous formulations and same for the w/o-emulsion. The reconstructed cornea could be an alternative to excised animal tissue for drug permeation studies in vitro.

Murine keratocytes function as antigen-presenting cells

Keratocytes express MHC class I molecules constitutively, and keratocytes stimulated with IFN-gamma express MHC class II molecules. Unstimulated keratocytes constitutively express B7-1 and ICAM-1, as well as low levels of CD40 and 4-1BBL. These findings indicate that keratocytes may deliver both antigen-specific and costimulatory signals to CD4(+) and CD8(+) T cells. To demonstrate that keratocytes expressing B7-1 provide a costimulatory signal to T cells, CD4(+) or CD8(+) mouse T cells were incubated with anti-CD3 mAb and irradiated keratocytes. Enhanced proliferation of both CD4(+) and CD8(+) T cells occurred, and could be inhibited by anti-B7-1 mAb, indicating T cell costimulatory activity by B7-1 on the keratocytes. To demonstrate that keratocytes can deliver an antigen-specific signal, CD4(+) and CD8(+) T cells from herpes-infected mice were incubated with HSV-1-infected, irradiated keratocytes. The resulting T cell proliferation and production of Th1 cytokines (IL-2, IFN-gamma) indicated T cell activation by antigens presented by the infected keratocytes. These results show that keratocytes in the corneal stroma of the mouse can function as antigen-presenting cells and, thus, may play a role in immune-mediated stromal inflammation such as herpetic stromal keratitis.

Reconstruction of an in vitro cornea and its use for drug permeation studies from different formulations containing pilocarpine hydrochloride

The aim of the present contribution was to develop a functional three-dimensional tissue construct to study ocular permeation of pilocarpine hydrochloride from different formulations. The in vitro model was compared to excised bovine cornea. Modified Franz cells were used to study the transcorneal permeability. Analysis was performed by reversed-phase high-performance liquid chromatography. Comparisons of the permeation rates through excised bovine cornea and the in vitro model show the same rank order for the different formulations. The permeation coefficient, K(P), obtained with the cornea construct, is about 2-4-fold higher than that from excised bovine cornea. It is possible to reconstruct bovine cornea as an organotypic culture and also to use this construct as a substitute for excised bovine cornea in drug permeation studies in vitro.

Permeability of chemical delivery systems across rabbit corneal (SIRC) cell line and isolated corneas: a comparative study

The purpose of this study was to investigate the corneal permeability of phenylephrone chemical delivery systems (CDS) across isolated cornea and to evaluate the utility of the SIRC cell line (epithelial cells originating from rabbit cornea) as an in vitro model for predicting the ocular permeability. The effect of benzalkonium chloride (BAC) on the drug permeability through SIRC cell layers was also studied. The transport of phenylephrone CDS across the isolated cornea of the albino rabbit was measured at various pH values using a two-chamber glass diffusion cell, and the results were compared with the reported permeability values across SIRC cells of rabbit origin. Corneal membranes showed lower flux values for compounds, especially for hydrophilic compounds, than the SIRC cell line. A significant correlation was observed between the permeability coefficients through corneal membranes and SIRC cells. When the pH of the transport medium was increased, the permeability coefficients increased and lag times decreased in both in vitro models. Furthermore, both in vitro models showed significant correlation between permeability coefficients and lipophilicities of the drugs. The three esters, having higher lipophilic characteristics, showed higher permeability than phenylephrine HCl. The phenylacetyl ester of phenylephrone showed a three-fold increase in penetration across SIRC cell layers in the presence of 0.01% BAC. These results suggest that the use of SIRC cell layers can reasonably predict the permeability of ophthalmic drugs across corneal membranes.

Expression and activity of cell cycle-regulatory proteins in normal and transformed corneal endothelial cells

Corneal endothelial cells have a limited capacity for proliferation. Upon transformation with the SV40 large T antigen, however, these cells undergo division and grow rapidly. In order to gain insight into the control mechanisms that determine this proliferative switch, we investigated the expression level and activity of various known cell cycle-regulatory proteins in these cells. Primary human and rabbit corneal endothelial cells were transduced in vitro with a replication-defective adenovirus containing SV40 large T antigen, and subsequently the expression and activity of cell cycle-regulatory proteins was analyzed. Cells transduced with large T antigen exhibited strongly increased activity of cyclin-dependent kinases. This increase correlated with the elevated expression of various cyclin-dependent kinase subunits, such as cyclin A, and to a lesser extent, cyclin D, cdk2, and cdk4. Furthermore, the expression of two cyclin-dependent kinase inhibitors, p21(WAF1) and p27(KIP1), which was high in primary human cells (but not in primary rabbit cells), was strongly reduced in large T-antigen transduced cells. Thus, the remarkably low proliferative activity of normal human corneal endothelial cells appears to be regulated at two levels: the expression of certain cell cycle-regulatory proteins that are essential for cell cycle progression is extremely low (cyclin A) or somewhat low (cdk2 and cdk4); but the amount of p21 and p27, inhibitors of cell cycle progression, is very high. As a consequence, the enzymatic activity of cyclin-dependent kinase is below detectable levels. However, the growth-inhibitory status of these components is clearly reversible: upon transduction with large T antigen, the expression of cyclin A, cyclin D, cdk2, and cdk4 is induced, whereas the expression of p21 and p27 is inhibited, and the cells proliferate. Thus, our study provides insight into the molecular basis of the attenuated proliferation of corneal endothelial cells and suggests potential targets that could be manipulated for the purpose of therapeutic interventions aimed at renewed cell growth.

Different characteristics of endothelial cells from central and peripheral human cornea in primary culture and after subculture

Several methods for isolation and cultivation of human corneal endothelial cells have been described during the last few decades. In contrast to the situation in vivo, the cultured cells show mitogenic activity but often lose their typical morphological appearance. In this paper, we describe a technique to isolate and cultivate morphologically unchanged endothelium from the human cornea. This method revealed different characteristics of endothelial cells according to their position within the human cornea. Endothelial cells isolated from the central part have a morphology similar to that of cells in vivo (i.e., they are densely packed and show no mitogenic activity). In contrast, endothelial cells derived from the peripheral part of the cornea are characterized by mitogenic activity but their cell-to-cell attachment seems to be less tight than in vivo. The significance of these two different endothelial cell types for wound healing in the human cornea is discussed.

A method for separation and staining of flat mounts of human corneal endothelium

A technique is described in which sheets of corneal endothelium are removed from human donor corneo-scleral discs. Celloidin solution was applied to the endothelial surface, allowed to dry, peeled off with the attached endothelial cell layer and mounted on a glass slide. Following removal of the celloidin with acetone, this endothelial cell flat mount was then stained with H&E and monoclonal antibodies to cell adhesion molecules. A pilot study of endothelial cell adhesion molecule expression in flat mount preparations of 14 corneas showed constitutive neural cell adhesion molecule (NCAM) expression, but a lower degree of focal expression of intercellular adhesion molecule (ICAM)-1, vascular cell adhesion molecule (VACM)-1, P/E-selectin and HLA-DR.

Release of platelet activating factor (PAF) and eicosanoids in UVC-irradiated corneal stromal cells

Ultraviolet (UV) irradiation provokes acute inflammation of the eye, and can be used to model processes that occur in response to damage to the anterior segment. This study characterized ultraviolet-C (UVC, 254 nm) irradiation-induced PAF synthesis, and arachidonic acid (20:4) and eicosanoid release in rabbit corneal stromal cells maintained in vitro. PAF was measured by radioimmunoassay (RIA) after exposing cultured corneal stromal cells to UVC irradiation (20 min, 2, 5, 10 mW/cm2). 14C-20:4-labeled stromal cells were also stimulated with UVC and radiolabeled phospholipids, neutral lipids and eicosanoids were measured. Synthesis of cell-associated and secreted PAF from corneal stromal cells was increased by UV irradiation. UV irradiation (254 nm, 5mW/cm2) enhanced 20:4 release from triacylglycerols, phosphatidylinositol, phosphatidylserine and phosphatidylethanolamine, and increased levels of 20:4-diacylglycerol and unesterified 20:4. The released 20:4 entered both the cyclooxygenase and lipoxygenase pathways after UVC irradiation. The PAF antagonist, BN52021 (10 microM) reduced UVC irradiation-induced stimulation of prostaglandin production, but failed to inhibit UVC-induced 20:4 release and synthesis of lipoxygenase products. Furthermore, exogenous PAF (1 microM) stimulated prostaglandin production, but did not increase the synthesis of lipoxygenase products from radiolabeled 20:4. The effects of PAF on prostaglandin synthesis were inhibited by BN52021. These findings indicate that responses to injury in cultured corneal stromal cells include PAF synthesis, release of 20:4 from glycerolipids, accumulation of diacylglycerol and synthesis of eicosanoids. The data further suggest that during UVC irradiation in vitro, PAF is not a primary or initial mediator of 20:4 release and synthesis of lipoxygenase products, but may mediate UVC-induced prostaglandin synthesis.

In situ immunohistochemical analysis of cell adhesion molecules on human corneal endothelial cells

Interaction of leucocytes with human corneal endothelial cells (HCECs) can be observed in several clinicopathological conditions, such as uveitis, keratitis, and corneal graft rejection. Since leucocyte-endothelial cell interactions involve various adhesion receptors we have analysed the expression and distribution pattern of the neural cell adhesion molecule (NCAM), the intercellular adhesion molecule-1 (ICAM-1), the vascular cell adhesion molecule-1 (VCAM-1), the endothelial leucocyte adhesion molecule-1 (ELAM-1), and the cluster of differentiation antigen-44 (CD44) on flat preparations of normal and organ-cultured HCECs. NCAM and ICAM were constitutively expressed on HCECs whereas VCAM-1, ELAM-1, and CD44 were absent from normal HCECs. However flat mounts of HCECs from organ-culture preserved corneas showed a mosaic-like distribution pattern of VCAM-1 and ELAM-1 positive cells and garland-like clusters of CD44 positive cells. We suggest that modulation of ELAM-1, VCAM-1, and CD44 expression on HCECs may contribute to the regulation of leucocytes-HCECs interaction in the case of anterior segment inflammation.

Fluid transport across cultured bovine corneal endothelial cell monolayers

The mammalian corneal endothelium is known to transport fluid from the stromal compartment to the aqueous humor, thereby maintaining corneal transparency. Corneal endothelial cells have been cultured for some years now, but whether they preserve their in vivo ability to actively transport fluid is not known. We have now grown bovine corneal endothelial cell monolayers (BCECM) on permeable substrates (Transwell) and report that, just like their counterparts in vivo, these cultured cells pump fluid from the basal to the apical compartment and display measurable electrical resistance and potential difference across the monolayer. BCECM were grown on collagen-treated permeable supports using Dulbecco's modified Eagle's medium (DMEM)/20% fetal bovine serum with antibiotics. Cells grew to confluence in 5-7 days and displayed polygonal shape. Only cells from passages 1-3 were utilized. Inserts were fitted directly into Lucite chambers specially built. The rate of fluid pumping by BCECM was 3.96 +/- 0.49 (SE) microliter.h-1.cm-2 (n = 13) and could be measured continuously for several hours; fluid pumping was inhibited by 0.2 mM amiloride. The specific electrical resistance of the monolayers was 180 +/- 22 omega.cm2 (n = 11). A mean electrical potential difference of 63.8 +/- 3.7 microV (n = 15, range 40-100 microV, apical side negative) was recorded across the monolayers in DMEM. The availability of the commercial inserts makes this procedure practical; as a consequence, the rate of fluid transport by cultured corneal endothelium has been quantitated for the first time. This method can now be extended to other cultured layers.(ABSTRACT TRUNCATED AT 250 WORDS)

In vitro corneal pathogenicity of Acanthamoeba

The comparative cytopathic effects of a keratitis and an environmental isolate of Acanthamoeba were studied on confluent monolayers of human and rabbit corneal cells grown in culture. The presence of cells in culture induced excystment of amoebae to the active trophozoite form. Total destruction of cell monolayers was observed to be similar for both isolates, and dependent on incubation time and amoebic concentration. The relevance of these findings to human and experimental Acanthamoeba keratitis is discussed.

The use of cultured epithelial and endothelial cells for drug transport and metabolism studies

In an effort to develop novel strategies for delivery of drug candidates arising from rational drug design and recombinant DNA technology, pharmaceutical scientists have begun to employ the techniques of cell culture to study drug transport and metabolism at specific biological barriers. This review describes some of the general factors that should be considered in developing a cell culture model for transport studies and metabolism studies. In addition, we review in detail the recent progress that has been made in establishing, validating, and using cell cultures of epithelial barriers (e.g., cells that constitute the intestinal, rectal, buccal, sublingual, nasal, and ophthalmic mucosa as well as the epidermis of the skin) and the endothelial barriers (e.g., brain microvessel endothelial cells).

Optimization of culture conditions for human corneal endothelial cells

Long-term cultivation of human corneal endothelial cells (HCEC) was optimized with respect to different components of the culture system: 25 different nutrient media, different sera, 6 mitogens and various substrates were tested in their ability to influence clonal growth and morphology of HCEC. F99, a 1:1 mixture of the two media M199 and Ham's F12, was the most effective basal medium in promoting clonal growth of HCEC. Among various sera, human serum and fetal bovine serum showed optimal growth promoting activities in combination with F99, whereas newborn bovine serum (NBS) was by far superior for the development of a typically corneal endothelial morphology. Crude fibroblast growth factor (FGF), or alternatively endothelial cell growth supplement, was absolutely essential for clonal growth of HCEC at low serum concentrations, for example 5% NBS. Formation of a monolayer with a morphology similar to corneal endothelium in vivo was observed only on culture dishes coated with basal membrane components such as collagen type IV, laminin, or fibronectin. The most pronounced effect on morphologic appearance was obtained by culturing the cells on the extracellular matrix (ECM) produced by bovine corneal endothelial cells. Moreover, ECM could substitute for crude FGF in clonal growth assays.

Corneal preservation

Significant advances in corneal preservation have been made over the past decade. The introduction of chondroitin sulfate-containing media for use at 4 degrees C allows storage of corneas for up to ten days prior to transplantation. Organ culture techniques have also been improved with the addition of chondroitin sulfate. There has been an increase in our understanding of preservation using McCarey-Kaufman medium. Studies have been published that compare these methods and help the clinician decide which method to use based on objective data. Also, our understanding of existing methods of evaluating endothelial viability has increased and new methods have been developed. Finally, the acquired immunodeficiency syndrome is having an increasing effect on eye-banking and the supply of donor corneas.

A simplified technique for the short-term tissue culture of rabbit corneal cells

A technique for the short-term culture of pure populations of rabbit corneal endothelial and epithelial cells has been developed. Rabbit corneas were placed on concave agarose surfaces, treated briefly with a solution of trypsin and ethylenediamine tetracetic acid, and transferred, either epithelial cell surface or endothelial cell surface down, to microscope slide culture chambers. Within 6 to 12 h the epithelial cells or endothelial cells attached to the slide chamber surface and the cornea was removed, leaving behind a pure population of cells which spread out and grew to fill the surface of the slide chamber. This technique provides a simple and economic means for the reproducible initiation of primary cultures of rabbit corneal epithelial and endothelial cells for us in a variety of experiments.

Growth and characterization of rabbit corneal cells in vitro

Primary organ cultures of rabbit corneal epithelium, stroma, and endothelium were established by microdissection. Secondary cultures of epithelial cells, keratocytes, and endothelial cells were established by serial passage. The doubling time for epithelial cells and keratocytes was 18 h, and endothelial cells doubled their number in 5 days. Ultrastructural studies demonstrated characteristic morphological, nuclear, and cytoplasmic features of corneal epithelial cells, keratocytes, and endothelial cells and confirmed the identity of the cell lines. The purity of the three distinct cell types was ascertained by indirect immunofluorescence techniques, using antibodies against keratin, which identified epithelial cells, and fibronectin, which identified keratocytes and endothelial cells. The indirect immunofluorescence technique represents a simple method to screen an aliquot of a cell suspension and determine the purity of corneal cells grown in vitro.

Removal of corneal crystals by topical cysteamine in nephropathic cystinosis

In patients with nephropathic cystinosis, corneal crystals develop by one year of age; they progressively accumulate and eventually cause recurrent corneal erosions and photophobia. After an in vitro study of cystinotic corneal stromal cells showed cystine depletion by cysteamine and after topical cysteamine was determined to be nontoxic in rabbits, we performed a controlled double-blind clinical trial of 10 mM cysteamine eyedrops in young patients with cystinosis, using one eye for treatment and the other as the control. Two children begun on the protocol before two years of age had a striking decrease in the number of corneal crystals in the cysteamine-treated eye within four to five months of entering the study. Cysteamine eyedrops appear to be safe and efficacious in the short-term treatment of patients with cystinosis who are under two years of age. The long-term value of such treatment and its effectiveness in older patients remain to be determined.

Effect of neuraminidase on Fc and C3b receptors on rabbit corneal cells infected with herpes simplex virus

The effect of neuraminidase on Fc receptors (FcR) and C3b receptors (C3bR) was studied in epithelial, stromal and endothelial cells of the rabbit cornea infected with type 1 (HSV-1) and type 2 herpes simplex virus (HSV-2) in vitro. FcR were induced on epithelial, stromal and endothelial cells of the rabbit cornea by both HSV-1 and HSV-2, but their activities were not enhanced by neuraminidase. On the other hand, the treatment of HSV-infected corneal cells with neuraminidase resulted in the enhancement of C3bR activities on epithelial, stromal and endothelial cells infected with HSV-1, and the enhancing effect of neuraminidase was more pronounced on corneal endothelial cells. A similar neuraminidase treatment had no significant effect on C3bR activities on the corneal cells infected with HSV-2.

Herpes simplex virus type 1 persistence and latency in cultured rabbit corneal epithelial cells, keratocytes, and endothelial cells

Cell cultures of rabbit corneal epithelium, keratocytes, and endothelium were used to determine the lytic cycle of herpes simplex virus type 1. Viral growth was fastest in epithelial cells. A novel HSV-1 in-vitro latency system was established in the three distinct cell types. Cell cultures were inoculated at low multiplicities of infection with HSV-1. Temperature manipulation alone was used to induce and reactivate latent HSV-1 infections. The presence of cellular stress proteins was demonstrated at supraoptimal temperatures. All cell types were capable of maintaining latent viral infections under these conditions. Viral persistence was present in 20% of epithelial cell cultures at supraoptimal temperatures, but not in keratocyte cultures or endothelial cell cultures.

Clinical cryobiology of tissues: preservation of corneas

It is well recognized that the clarity of the cornea is a function of its hydration, and that this hydration is controlled by a "pump-and-leak" mechanism operating across the posterior monolayer of cells called the endothelium. A breakdown of the endothelium through disease or injury causes a marked increase in corneal thickness as the stroma imbibes fluid from the aqueous humor in the anterior chamber of the eye. This thickened, edematous condition of the stroma results in a cloudy cornea with an associated marked decrease in visual acuity. Treatment for this condition is usually by full-thickness corneal transplantation (penetrating keratoplasty), the success of which is dependent upon the donor cornea having an intact and healthy endothelium. It is essential, therefore, that any method of corneal storage for penetrating keratoplasty should protect and preserve the endothelium in a viable state. Current clinical practice relies upon short-term methods of preservation by two principal methods. Moist Chamber Storage is the time-honored corneal preservation method; it consists of keeping enucleated eyes at 0-4 degrees C in a sealed jar containing a pad of cotton gauze soaked in saline to provide a humid environment. The time limit placed upon this method of storage is 24-48 hr after which the viability of the endothelium deteriorates rapidly. Storage in M-K (McCarey-Kaufman) Medium involves excision of the corneoscleral segment from the donor eye and immersing it, endothelial side uppermost, in a medium consisting of tissue culture medium, 5% Dextran 40, and antibiotics. Laboratory and clinical studies indicate that storage in M-K medium at 4 degrees C preserves human endothelial cells for up to 4 days when the eye has been removed from the cadaver in less than 10 hr postmortem. Long-term preservation of corneas by freezing has long been a major goal in eye banking because indefinite storage by cryopreservation offers significant advantages for the quality and the quantity of material for use in keratoplasty, as well as for its distribution. However, procedures that have been developed for the cryopreservation of corneas have not been widely used, and a number of studies have shown that these procedures are inadequate for maintaining the integrity of the corneal endothelium. Not surprisingly, clinicians are now reluctant to accept corneas that have been frozen by these methods, though the clinical need is now greater than ever.(ABSTRACT TRUNCATED AT 400 WORDS)

Lysosomal enzyme activities of the bovine corneal endothelium

A quantitative study was carried out on the lysosomal enzyme activities of the bovine corneal endothelium-Descemet's membrane preparation. The corneal endothelium and Descemet's membrane were peeled off together. Cathepsin D was assayed using hemoglobin as substrate; N-acetyl-beta-D-glucosaminidase, beta-glucuronidase, acid phosphatase, and alpha-mannosidase were also examined using p-nitrophenyl derivatives as substrate. The proportions of N-acetyl-beta-D-glucosaminidase, cathepsin D, and beta-glucuronidase of the Descemet's membrane-endothelium complex were particularly high: 11.5%, 12.6%, and 12.5% of the whole cornea, respectively. Corneal endothelial cells also showed high activities of acid phosphatase and alpha-mannosidase (3.8%, and 5.0% of the whole cornea, respectively), while the protein and DNA contents were 0.5% and 0.5% in the complex. Lysosomal enzyme activities in the complex were also compared with those in other ocular tissues and were determined by the same methods at the same time.

Angiogenic activity of the corneal epithelium

A homogenate of corneal epithelial cells was tested to determine if it could induce vascularization in the cornea. Both fresh and cultured cells were used as sources of the homogenate which was evaluated in vivo using a self-contained system to perfuse it constantly into the corneal stroma. A vigorous growth of vessels resulted when the concentration of the homogenate exceeded a threshold of 20 micrograms per ml total protein. This capacity was not destroyed when the homogenate was heated. As a preliminary refinement of this response white blood cells were eliminated from the model with whole body X-irradiation. Vascularization occurred in leukopenic animals but was less than that observed in normal animals. It is concluded that the epithelial homogenate is able to provoke corneal vascularization in the absence of leukocytes.

Cyclic nucleotides in muscarinic regulation of DNA and RNA polymerase activity in cultured corneal epithelial cells of the rabbit

DNA and RNA polymerase activities in the purified nuclear fraction from cultured rabbit corneal epithelial cells were assayed over a range of substrate (labeled dTTP or UTP) concentrations using calf thymus DNA as template. Effects of carbamylcholine on polymerase activities were evaluated over a range of drug concentrations including those saturating muscarinic receptors. Carbamylcholine significantly (p less than 0.001) enhanced activity of both polymerases, both in nuclei incubated with the drug during assay and in nuclei from carbamylcholine-treated cells. Drug effects were blocked by atropine. Regression analysis of Hill plots for variation of polymerase activity with carbamylcholine concentration indicated half-maximal activity of both polymerases at approximately 1 microM carbamylcholine. Mechanisms by which carbamylcholine may alter polymerase activities are discussed in relation to effects of the drug on nuclear enzymes of cyclic nucleotide metabolism and on cyclic nucleotide-dependent protein phosphorylation.

Antibody-dependent cellular cytotoxicity mediated by polymorphonuclear leukocytes and mononuclear cells against HSV-1 infected primary cultures of rabbit corneal epithelium

The roles of rabbit polymorphonuclear leukocytes (PMNL) and mononuclear cells (MC) for the regulation of ocular herpes simplex virus-1 (HSV-1) infection were studied. The antibody-dependent cellular cytotoxicity (ADCC) mediated by PMNL and MC from normal rabbit peripheral blood was assessed kinetically employing a specific 51Cr release assay. The HSV-1 infected primary cultures of rabbit corneal epithelium (PRCE) were used as the target cells to obtain a homologous assay system. The PRCE was prepared by an epithelium outgrowth technique and identified by electron microscopy. The expression of the surface HSV-1 antigens on PRCE was examined by indirect immunofluorescent staining; the cell population stained by fluorescein increased from 40% at 3 hr postinfection (PI) to 100% at 8 hr PI. To determine how early the cytotoxicity occurs, PRCE were infected with HSV-1 for 2 hrs. After 2 hrs, the ADCC was checked every 10 min for the first 40 min and then at 1, 2 and 4 hr of incubation. The cytotoxicity was apparent at 10 min postincubation and reached 46% by PMNL and 40% by MC at 4 hr postincubation (6 hr PI). Significant cytotoxic effect (26% by PMNL and 16% by MC) occurred as early as 3 hr PI. When the one-step growth cycle of HSV-1 was studied in the PRCE, HSV-1 had an eclipse period of 4 hr and a rise period of 8 hr. This suggests that rabbit PMNL and MC have the potential to eliminate the HSV-1 infected rabbit corneal epithelium before HSV matures in the cells.

In vitro cytotoxicity assays. Potential alternatives to the Draize ocular allergy test

A short-term cytotoxicity assay carried out in multiwell test plates and a supplementary colony forming assay are both useful for screening and range finding of toxic concentrations of test agents. The highest tolerated dose (HTD), a concentration at which only minimal morphological changes were observed, was designated as endpoint in the assay. Epithelial rabbit cornea cells, murine fibroblasts, Chinese hamster lung cells, human hepatoma cells and mouse macrophage cultures were used as targets. Several of the alcohols tested at HTD in the colony forming assay were found to inhibit colony formation. An ID50 of colony formation was used as a quantitative corroborating test. The ranking of 34 toxicants was found to be virtually the same with all cell types examined. This easily reproducible, rapid in vitro test is cost-effective and can be used for preliminary large scale screening of potential toxicants.

Isolation and culture of corneal cells and their interactions with dissociated trigeminal neurons

The three cell types of rabbit cornea (epithelium, stromal fibroblasts and endothelium) were isolated by an improved method using both microdissection and selective enzyme treatment. This technique reproducibly resulted in an almost total recovery of each cell type from a given cornea. When maintained in culture, the three cell types showed different morphologic characteristics, each resembling the in vivo counterpart. The epithelial culture consisted of both attached and floating cells. The attached cells located at the marginal area of a colony were irregular in shape and possessed pseudopodia, while those in the confluent area were polygonal. Floating cells were typically vacuolated, curve-shaped and joined in groups of 2-4 cells as a spherical body enclosing a lucent interior. Comparison of mitotic rates, ultrastructure, keratin levels and other cytologic evidence suggested that the attached cells may correspond to the basal cells and less differentiated wing cells, while the floating cells may be analogous to the more differentiated wing cells and superficial cells. Neurons dissociated from neonatal rabbit trigeminal (Gasserian) ganglia were plated into multiwells partially covered with a given corneal cell type. The percentages of viable and neurite-bearing neurons were evaluated on the first three days. When neurons were grown in contact with each of the corneal cell types, neurites were extended in every case. However, when neurons were not in contact with the corneal cells in the coculture, only epithelial cells permitted neurite outgrowth. The data suggested two types of cellular interactions between corneal cells and sensory neurons, one of which may be the specific release of a neuronotrophic factor by epithelial cells. This culture system represents the first step towards developing an in vitro model for studying various cornea-trigeminal interactions.

Activity of rabbit monocytes, macrophages, and neutrophils in antibody-dependent cellular cytotoxicity of herpes simplex virus-infected corneal cells

Rabbit phagocytes were examined for their ability to kill target cells infected with herpes simplex virus by antibody-dependent cellular cytotoxicity. Two sources of rabbit corneal cells were used as targets: Staaten Seruminstitut Rabbit Cornea (a continuous cell line) and stromal keratocytes from the middle layer of corneas excised from New Zealand white rabbits. Peritoneal exudate and alveolar macrophages were found to be the most active effector cells, followed by blood neutrophils and monocytes. Peritoneal exudate and alveolar macrophages killed target cells with dilutions of antibody as high as 1:10,000. Monocyte antibody-dependent cellular cytotoxicity activity was absent in over one-third of the rabbits tested and was only weakly active in positive rabbits. In vitro aging of monocytes did not enhance activity. Antibody reactive with peritoneal exudate macrophages, alveolar macrophages, and blood neutrophil effector cells appeared 7 days after intracorneal injection of infectious herpes simplex virus. Results of these studies show that in vitro assays with a complete rabbit system (effectors cells, antibody, and target cells) can be developed to monitor herpetic disease and suggest an active role for rabbit phagocytes in cytotoxicity of herpes simplex virus-infected cells.

Antibody mediated destruction of keratocytes infected with herpes simplex virus

The ability of complement-dependent antibody to mediate destruction of rabbit corneal cells infected with the RE and KOS strains of HSV-1 was examined. Three sources of target cells were used: BHK-21, a monitor of surface antigen expression and cell lysis; SIRC, a continuous cell line of rabbit corneal origin; stromal keratocytes, grown from the middle layer of excised rabbit corneas. Rabbits were infected intrastromally with the RE strain, which causes a high incidence of stromal keratitis, and sacrificed at designated times. Results showed that all three cell types, infected with either HSV strain, expressed surface antigens and were susceptible to lysis by antibody produced during the course of stromal disease. 51Cr-release assays showed that cytolytic antibody appeared as early as 5 days postinfection and reached maximum by day 14-20. Findings suggest a direct relationship between the in vitro demonstration of complement dependent cytolysis of infected stromal cells and the disease process in experimental animals.

Corneal endothelial transplantation

Patients with visually significant corneal edema, secondary to endothelial dysfunction, before the development of scaring or vascularization, need only have the corneal endothelium replaced to restore corneal clarity. This fact, plus the lack of consistently available donor material, prompted us to evaluate tissue cultured corneal endothelim (TCCE) as a donor source. We have shown that TCCE, when transplanted, can regain normal morphology and physiologic function. To accomplish practical use of autologous stroma, a transparent gelatin membrane which can serve as substrate for endothelial growth in tissue culture, has been developed. This cellular membrane has been transplanted successfully in rabbits with good functional results. It is hoped that ultimately this technique can be developed for routine use in man.

Stimulation of endothelia cell proliferation by precursors of thymidylate

Bovine corneal and aortal endothelial cell cultures were established from primary explants and subcultured for at least 40 passages. With both cell lines, exogenous thymidine, folate or folinate markedly increased the proliferation of these cells and decreased their serum requirement in Medium 199. Medium 199 supplemented with thymidine was particularly useful for cell survival at low densities; cones were readily produced when single cells were plated as low as 0.07 cells. cm-2. In contrast to the results of others, neither fibroblast growth factor nor epidermal growth factor were necessary for cell proliferation or survival at low densities.

Effects of intraocular miotics on cultured bovine corneal endothelium

Two cases of severe corneal oedema occurred after the use of intraocular pilocarpine. Experimental investigations were conducted with cultured bovine corneal endothelial cells exposed for 5 minutes to 1% pilocarpine solutions of varying composition. Cells were destroyed in solutions not isotonic with aqueous humour, and calcium-free ionic solutions caused loss of cell adhesion without loss of viability. Low pH or the presence of 1% pilocarpine had no detectable effects; 1% acetylcholine chloride in 5% mannitol (Miochol) also caused cell destruction, and this preparation was found to be considerably hypertonic. The minimum requirements for the formulation of intraocular miotics are discussed.