Proteoglycans on normal and migrating human corneal endothelium.

Proteoglycans are of fundamental importance to the normal functioning of the cornea. They consist of a core protein to which one or more glycosaminoglycan chains are attached. Cell surface proteoglycans are known to mediate many aspects of cell behaviour including cell adhesion, control of extracellular matrix deposition, cell proliferation, cell migration, leukocyte adhesion and modulation of growth factor activity. This paper describes the first investigation into the distribution and function of the three main classes of proteoglycans on human corneal endothelium. Immuno-gold labelling techniques were used at the light, scanning and transmission electron microscope level to localise heparan sulphate, chondroitin sulphate and keratan sulphate proteoglycans on human corneal endothelium. Human corneas were freeze-wounded and kept in organ culture for 3 days in order to study the distribution of proteoglycans on migrating corneal endothelium. An Optimas image analysis system was used to quantify the change in proteoglycan labelling during cell migration. Labelling for chondroitin sulphate and heparan sulphate was at very low levels on normal corneal endothelium while keratan sulphate labelling was at high levels. The wound healing experiments showed that migrating cells had increased labelling for heparan sulphate and chondroitin sulphate with greatly decreased labelling for keratan sulphate. Statistical analysis showed these changes were highly significant (P<0.001). Transmission electron microscopy revealed that chondroitin sulphate and keratan sulphate were present throughout Descemet's membrane while heparan sulphate was concentrated at the interface of Descemet's membrane and the migrating corneal endothelial cells. The pattern of occurrence of chondroitin sulphate, heparan sulphate and keratan sulphate on the human endothelium in normal and wounded cornea suggests that these proteoglycans are linked to the process of cell migration.

An ultrastructural investigation into proteoglycan distribution in human corneas.

PURPOSE: We report an investigation into the distribution of proteoglycans (PGs) in normal, organ-cultured and dextran-treated human corneas. METHODS: Immunogold labeling was carried out at the electron microscope level to localize keratan sulphate (KS), chondroitin sulphate (CS), and heparan sulphate (HS) PGs. RESULTS: High levels of labeling for CS was found in the epithelium, endothelium, and keratocytes, with light labelling present in the basement membranes and the corneal stroma. Labeling for HS was present in the epithelium, endothelium, and keratocytes, with intense labeling present at the endothelium/Descemet's membrane interface and the epithelium/Bowman's layer interface. Large filaments were also observed in these regions in cuprolinic blue-stained specimens. Keratan sulphate was present at high levels in the stroma and the basement membranes with low levels present within the keratocytes, epithelium, and endothelium. The pattern of KS labeling along the collagen fibrils in the stroma sometimes showed evidence of periodicity. Organ-cultured corneas had extensive collagen-free "lakes," the interior of which immunolabeled positively for KS and showed staining with cuprolinic blue. The lakes were greatly reduced in the dextran-treated samples. CONCLUSION: This investigation determined the ultrastructural distribution of KS, CS, and HS PGs in human cornea and showed that organ culture is associated with a change in distribution of stromal PGs.

Endothelial viability of organ-cultured corneas following penetrating keratoplasty.

PURPOSE: To assess endothelial integrity following corneal transplantation using human corneas stored in organ culture in the Manchester Eye Bank. METHODS: A prospective study was undertaken on 24 patients who had received full-thickness corneal grafts using corneas stored in organ culture. The donor corneal endothelium was photographed prior to transplantation using light microscopy. Specular microscopy and ultrasonic pachometry were performed at 30 days (+/- 3 days), 12 weeks (+/- 1 week), 26 weeks (+/- 2 weeks) and 52 weeks (+/- 4 weeks) following corneal transplantation. The following cell parameters were measured: density, area, coefficient of variation (CV) for area, perimeter, diameter, form factor and corneal thickness. RESULTS: One year after corneal transplantation there was a statistically significant decrease in endothelial cell density (-39.4%), and a statistically significant increase in endothelial cell area (+94.4%), perimeter (+36.1%), diameter (+57.1%) and form factor (+5.8%). However, no significant changes were seen in CV or corneal thickness with respect to time after transplantation. (There were no significant changes in endothelial cell parameters and corneal thickness for 12 control subjects.) CONCLUSIONS: Endothelial cell loss occurs at an accelerated rate from corneal transplants. This highlights the need for improving corneal endothelial viability during and after storage in order to improve the chances of longer-term survival of the transplanted cornea.

Keratan sulphate in the trabecular meshwork and cornea.

PURPOSE: A study was made of the distribution of keratan sulphate in the human anterior chamber. METHODS: The monoclonal antibody, 5-D-4, was used in immuno-electron microscopy to visualise keratan sulphate distribution in the anterior chamber of 16 normal eyes, 7 Fuchs' dystrophy corneas, and a macular dystrophy cornea. RESULTS: Keratan sulphate was detected in normal human aqueous humour and also on the surface of trabecular cells in the uveal meshwork. Normal corneal stroma showed an increase in keratan sulphate labelling from anterior to posterior, with marked labelling in the posterior region of Descemet's membrane. The apical surface of the corneal endothelium labelled positively, but showed considerable variation in the level of labelling from cell to cell. The macular dystrophy cornea had the classic histopathological features of a type I case, including a highly abnormal Descemet's membrane. No keratan sulphate was detected in the macular dystrophy patient's corneal stroma or serum. The Fuchs' endothelial dystrophy corneas showed a normal distribution of keratan sulphate labelling in the stroma. The Fuchs' endothelial cells labelled for keratan sulphate but were highly abnormal in appearance, often exhibiting long filopodia and appearing to be actively migrating. CONCLUSIONS: This work has shown that keratan sulphate has a much wider distribution than was previously believed. The detection of keratan sulphate on the trabecular and endothelial cell surfaces also suggests a possible role for this molecule in cell adhesion and/or migration.

Diurnal variations in human corneal thickness.

AIM: To elucidate the diurnal variation in human corneal thickness over a 48 hour period. METHOD: Changes in central corneal thickness were monitored in eight healthy subjects (four male, four female) aged between 10 and 63 years using an ultrasonic pachymeter. Measurements were made over a 48 hour period-immediately before sleep, immediately upon waking and at 15, 30, 45 minutes, 1, 1.5, 2, 2.5, 3 hours, and at 2 hour intervals thereafter throughout the remainder of each day. RESULTS: The mean corneal thickness for the group (SD) was 546 (14) microns, with a mean overnight increase of 5.5% (2.9%) (range 1.9-12.6%) and a maximum diurnal increase of 7.2% (2.8%) (range 2.1-14.3%). Individual differences in the extent of diurnal and overnight variation occurred within the group. For three subjects, the first reading taken on waking was not the highest and corneal thickness continued to increase. CONCLUSION: These data confirm an increase of corneal thickness during sleep, but also reveal considerable variation during waking hours. Thus, the overnight changes in corneal thickness are not truly representative of diurnal variations in human corneal thickness and, in fact, much greater diurnal variation occurs than the 3.0-4.4% previously reported.

Prediction of refractive outcome in penetrating keratoplasty for keratoconus.

We carried out a retrospective study to evaluate the relationship between vitreous cavity length, graft size, and final spherical equivalent refraction after penetrating keratoplasty (PK) for keratoconus. We found a straight-line relationship between vitreous cavity length and spherical equivalent for a one-surgeon series using the same technique throughout. The use of 0.25-mm undersized grafts shifted the results an average of 2.2 Dioptres in a more hypermetropic direction (p = 0.07 for the whole group, p < 0.01 for paired eyes). Hence the final spherical equivalent following PK for keratoconus can be predicted. Also, by altering the size of the donor graft button, the final refraction can be manipulated to some extent towards acceptable ametropia or to match the refraction of the fellow eye.

Detection of herpes simplex virus DNA in donor cornea culture medium by polymerase chain reaction.

AIMS/BACKGROUND: Herpes simplex virus (HSV) may establish latent infection in the cornea and therefore be transmissible by corneal transplantation. Monitoring of donor cornea culture medium was evaluated for HSV infection. METHODS: HSV was sought using virus isolation in cell culture, and its DNA was amplified to detectable levels using the polymerase chain reaction (PCR). RESULTS: Virus isolation in cell culture was negative on neat, cell pellet, and cell free supernatant prepared from the spent culture media of 80 corneas. Three cell pellets (3.8%) were positive for HSV DNA. The PCR positive culture negative results might have reflected latent rather than active HSV infection of the cornea. Post transplant follow up of the three recipients of corneas with HSV PCR positive organ culture media revealed no evidence of HSV induced eye disease or primary graft failure. CONCLUSION: Screening of corneal culture medium for HSV by virus culture or for HSV DNA by PCR could not be recommended.

Cell surface-associated keratan sulfate on normal and migrating corneal endothelium.

PURPOSE: To investigate cell surface-associated keratan sulfate on the corneal endothelium. METHODS: Immunolabeling techniques were used at the light, scanning, and transmission electron microscopic level to localize keratan sulfate on the corneal endothelium. The investigation included human, bovine, and rabbit corneal endothelia. A quantitative study of the relationship between cell size and keratan sulfate levels was conducted on normal bovine corneal endothelium. Changes in the distribution of keratan sulfate and chondroitin sulfate on endothelial cell surfaces were investigated on organ cultured bovine corneas during endothelial wound healing. Changes in the levels of keratan sulfate during endothelial wound healing were investigated in organ cultured human corneas and in vivo in rabbit corneas. Inhibition-enzyme-linked immunosorbent assay also was used to detect keratan sulfate in the aqueous humor. RESULTS: A variegated distribution of keratan sulfate was revealed on normal human, bovine, and rabbit corneal endothelia. Some cells had high levels of keratan sulfate on their surfaces whereas others, sometimes immediately adjacent, had little or none. Wound healing experiments resulted in changes of keratan sulfate levels on the migrating endothelial cells in bovine, human, and rabbit. In wounded organ cultured bovine corneas, there was a decrease in keratan sulfate levels and an increase in chondroitin sulfate levels on migrating endothelial cells. Keratan sulfate was detected in bovine aqueous humor. CONCLUSIONS: The pattern of occurrence of keratan sulfate and chondroitin sulfate on the corneal endothelial cells in normal and wounded cornea suggests that these glycosaminoglycans have differing roles in endothelial adhesion and migration.

Corneal donor infection by herpes simplex virus: herpes simplex virus DNA in donor corneas.

Three corneoscleral discs (from two donors) underwent subtotal endothelial loss during routine "long-term" organ culture storage. Laboratory studies of these corneas revealed evidence of herpes simplex virus (HSV) infection. The fellow cornea from one of the donors had been issued for transplant to a patient with keratoconus. Deterioration of the graft was noted 5 days after surgery; the disc was removed at 2 months and was shown to be infected with HSV. In an experiment designed to simulate initial "cleansing" of donor globes, 0.1% polyvinylpyrolidone-iodine protected cells from infection with HSV. It was concluded that the detection of HSV in these corneas could not be explained by external contamination of the ocular surface. Furthermore, culture of conjunctival and pharangeal swabs taken from 47 consecutive donors confirmed that HSV is rarely isolated at or around the time of death. Five pairs of donor corneas destined for use in transplantation were selected at random and investigated for the presence of HSV. HSV DNA was detected by polymerase chain reaction (PCR) in tissue from two of the corneal donors. Sequential stepwise sectioning suggested that HSV DNA when present was distributed in discrete foci within the cornea. These observations suggest that HSV infection may be a cause of severe endothelial loss during corneal organ culture and possibly provide an explanation for some "failures" of corneal grafting.

Regional variation and age-related changes of lysosomal enzymes in the human retinal pigment epithelium.

In this study the activities of two lysosomal enzymes in retinal pigment epithelial (RPE) cells isolated from three regions of the human fundus were examined: the macula, the nasal midzone, and the periphery. The results obtained showed that the activities of acid phosphatase and cathepsin D were significantly higher in the RPE cells derived from the macular region when compared with those in the periphery. The values for the midzone appeared to be intermediate between the other two regions. Furthermore, the overall activity of both enzymes increased as a function of age.

The topography of nerve cell loss from the locus caeruleus in elderly persons.

A topographical analysis of nerve cell loss from the locus caeruleus in "mentally normal" old people shows cell loss to be uniformly diffuse throughout the whole nucleus with no preferential involvement of any one particular area. Such findings contrast with those of ours on Alzheimer's disease and suggest differing mechanisms underlying the cell loss of old age and Alzheimer's disease. Cell loss in Alzheimer's disease is thought to relate to primary pathogenetic events in terminal fields of cerebral cortex. In "normal" old age, cell loss may be determined by changes occurring at perikaryal level possibly in respect of the cytotoxic effects of noradrenaline degradation and neuromelanin accumulation.

The progression of the pathological changes of Alzheimer's disease in frontal and temporal neocortex examined both at biopsy and at autopsy.

Brains were obtained at autopsy from five patients with Alzheimer's disease, each of whom had undergone diagnostic craniotomy 3-7 years previously. It was possible, therefore, to examine the number (density) and nucleolar volume of pyramidal nerve cells, and the density of senile plaques and neurofibrillary tangles within the cerebral cortex on two occasions during the progression of their illness, and to assess how these measures might have changed during the period between biopsy and death. In all five patients, at biopsy, the density and the nucleolar volume of pyramidal nerve cells was significantly less than controls and, in general, values for both these measures fell significantly further from biopsy to death. By contrast, in none of the five patients did senile plaque density consistently change from biopsy to death; neurofibrillary tangle density either did not change, or indeed sometimes decreased from biopsy to death. These data show that both the clinical and the pathological progression of Alzheimer's disease is marked by a continuing loss of pyramidal cells from frontal and temporal cortex, although the densities of plaques and tangles within the cortex do not, per se, correlate with the stage of the illness. The usefulness of measurement of plaque and tangle densities as pathological criteria by which the clinical and neurochemical deficits of Alzheimer's disease can be compared in different patients is clearly questionable.

Saccharides of senile plaques and neurofibrillary tangles in Alzheimer's disease.

The contribution that oligosaccharides might make to the structure of the senile plaque and the neurofibrillary tangle was investigated using lectin histochemistry in 9 patients with Alzheimer's disease. One group of 4 lectins diffusely stained the neurites of senile plaques whereas two groups of 6 different lectins stained neurofibrillary tangles within neuronal perikarya and plaque neurites. Neuraminidase pretreatment abolished staining of tangles by one of these latter groups, but did not affect staining by the other group. Senile plaque core amyloid failed to stain with any lectin. It is concluded that oligosaccharides may contribute, but in different ways, to glycoprotein or glycolipid residues that form an integral part of the structure of the senile plaque and the neurofibrillary tangle.

Topography of nerve cell loss from the locus caeruleus in middle aged persons with Down's syndrome.

A topographical analysis of nerve cell loss from the locus caeruleus in middle aged patients with Down's syndrome (whose brains show the pathological changes of Alzheimer's disease), has shown that cell loss is confined to dorsal areas, being least most rostrally and greatest caudally. By contrast, there is no significant cell loss from ventral parts of the locus, at any point along its rostrocaudal length. Dorsally located neurones of the locus project to cerebral cortex; ventrally located neurones to non-cortical areas such as basal ganglia, cerebellum and spinal cord. These data suggest that the damage to nerve cells of the locus caeruleus in Down's syndrome at middle age, like that seen in Alzheimer's disease itself, relates to primary pathological events within the cortical projection fields of affected cells with perikaryal loss following on as a later change.

Loss of neurones from cortical and subcortical areas in Down's syndrome patients at middle age. Quantitative comparisons with younger Down's patients and patients with Alzheimer's disease.

Brains were examined after autopsy from 12 patients over 53 years of age with Down's syndrome (in whose brains plaques and tangles were numerous in many areas of cortex and subcortex), 3 patients under 53 years of age with Down's syndrome (in whose brains plaques and tangles were minimal or absent), 10 patients, of age range similar to the older Down's group but with Alzheimer's disease and 5 control patients of age range similar to the younger Down's group. The number of plaques and tangles in the hippocampus and their density within the temporal cortex, the thickness of the temporal cortex, the cross-sectional area of the hippocampus and the relative number and mean nucleolar volume of nerve cells in these cortical and in some subcortical areas were estimated and compared in each of the 4 groups. The relative loss of nerve cells and the decrease in mean nucleolar volume were calculated in percentage terms for the older Down's syndrome patients by reference to data from the younger Down's syndrome patients, whereas such losses in Alzheimer's disease were calculated by reference to the younger control patients. While in qualitative terms, all areas of brain found to be damaged in Alzheimer's disease were also damaged in Down's syndrome at middle age, quantitative differences emerged with the reductions in relative nerve cell number and mean nucleolar volume being significantly less in many areas in Down's syndrome. Conversely plaques and tangles were more numerous in the hippocampus in Down's syndrome though in the temporal cortex plaques were less numerous. It seems, therefore, that although the same pathological process is likely to operate in the two conditions, additional biological and mortality differences between Down's syndrome and the general population may account for the observed quantitative variations.

Dopaminergic neurotransmitter systems in Alzheimer's disease and in Down's syndrome at middle age.

In 15 patients with Alzheimer's disease and in 10 with Down's syndrome at middle age, there was severe atrophy, neurofibrillary degeneration and loss of pigmented dopaminergic nerve cells from ventral tegmental area (A10) whereas nerve cells in neighbouring substantia nigra (A9) were much less affected in all three respects. It is suggested that these findings may represent different patterns of damage within the two systems in these conditions which may relate to the presence of Alzheimer type changes (senile plaques) within their respective projection fields.

The topography of cell loss from locus caeruleus in Alzheimer's disease.

A topographical analysis of nerve cell loss from locus caeruleus in Alzheimer's disease has shown that cell loss is confined to the dorsal areas and occurs uniformly throughout the rostrocaudal length of the locus. By contrast there is no significant cell loss from ventral parts of the locus, at any point along its rostrocaudal length. Dorsally located neurones of the locus project to cerebral cortex; ventrally located neurones to non-cortical areas such as basal ganglia, cerebellum and spinal cord. These data suggest that damage to nerve cells of locus caeruleus in Alzheimer's disease relates primarily to pathological events within their terminal fields, with perikaryal loss following as a secondary retrograde change. The senile plaque may represent the actual site of the damage to nerve terminals.