A C T A O P H T H A L M O L O G I C A VOL. 56 1 9 7 8
Department of Ophthalmology, Arhus Kommunehospital (Head: N . EhEers), Uniuesity of Aarhus, Denmark
EARLY MORPHOLOGICAL CHANGES IN ORGAN CULTURED HUMAN CORNEAL ENDOTHELIUM BY STEFFEN SPERLING
Nineteen human cadaver corneas with few damaged endothelial cells were incubated under tissue culture conditions for time periods ranging from five min to 48 h. Morphological alterations of the endothelial cells were studied in whole wet mounts stained by alizarine red-alkohol-trypane blue and by scanning electron microscopy. Joint meetings of three cells are characterisic for normal corneal endothelium. After 15-60 min of incubation, damaged cells were expelled from the coherent cell sheet by expanding neighbouring cells. Joint meetings of 5-8 expanding cells were formed. After 24 h of incubation, joint meetings of four cells were the dominating morphological abnormality. Morphological changes during reduction of the numbers of cells in joint meetings are described.
Key words: human cornea - endothelium - morphology - reorganization alizarine red - trypane blue - scanning electron microscopy - organ culture.
In normal human corneas the membrane of Descemet is covered by a single layer of coherent endothelial cells. Each cell is surrounded by four to eight neighbouring cells (Sperling 1978). The cells form joint meetings of three (Fig. 1). Although the number of viable endothelial cells decreases post mortem, a coherent sheet of normal endothelial cells has been observed on whole corneas obtained more than 24 h post mortem, after storage in organ culture (DoughReceived April 5 , 1978.
785
Steffen Sperling m a n e t al. 1973; Lindstrom et al. 1976). T h i s indicates that a coherent cell layer must h a v e been reformed in vitro. In the present report e a r ly morphological changes in t h e endothelium of whole h u m a n corneas incubated u n d e r tissue culture conditions a r e described.
Material and Methods Twenty-nine human eyes were obtained 8-36 h post mortem from cadavers stored at 15-22OC for 8-10 h, and later at 4OC. Corneas in situ were rinsed in tap water. Later whole eyes were submerged in 20 ml of a n isotonic solution of Bacitracin (500 IU/ml), Polymixin B (5000 IU/ml), and Neomycinsulfate (2.5 mg/ml) for 15 min at 25OC. Corneas were excised with a scleral rim and the endothelium was stained for one min by 0.3 per cent trypane blue in 0.9 per cent NaC1, sterilized through a 0.2 p filter. After excision from a cadaver bulbus, the intercellular spaces between intact endothelial cells swell in 0.9 per cent NaCl (Schrerder & Sperling 1977). Swollen intercellular spaces are visible in a n ordinary light microscope. By staining with trypane blue in NaCI, the relative number of damaged cells can be estimated. Trypane blue stain nuclei of damaged cells deep blue in uiuo (Stocker et al. 1966, 1970; Van Horn et al. 1975). The nuclei of undamaged cells are not stained. Uncovered parts of the membrane of Descemet stain blue. Living cells are not damaged by trpane blue (Stocker et al. 1970; Norn 1971; Van Horn 1971). An exact measure of endothelial cell density before incubation could not be obtained by staining of cell borders with alizarine red, as this stain kills the cells (Fig. 2). Nineteen corneas, in which less than one per cent of damaged endothelial cells was found in unfolded areas, were suspended in sterilized 100 ml glass infusion bottles by
Figs. 1 and 2. Fig. 1. Whole wet mount of normal corneal endothelial cells. Stained by alizarine red treated with alcohol and stained trypane blue Magnification: 138 x. Fig. 2. Coherent dead cells killed by alizarine red staining. Stained by trypane blue, after 12 h of incubation in tissue culture medium. The alizarine red has diffused away in the aqueous medium. Magnification: 138 x.
786
O r g a n Cultured H u m a n Cornea
Figs. 3, 4 and 5. Fig. 3. Red dots heavily stained by alizarine red after 25 min of incubation. Stained by alizarine red-alcohol-trypane blue. Magnification: 138 x. Figs. 4 a n d 5. Joint meetings of live to eight cells after 25 and 35 min of incubation. Stained by alizarine red-alcohol-trypane blue. Magnification: 138 x .
a suture tied to the scleral rim and incubated in tissue culture medium at 37OC. The medium was composed of Eagles minimum essential medium, L-glutamine (2 md ml ), 20 O/o foetal calf serum, amikacin (0.1 mg/ml). carbenicillin (0.5 mgiml) and mycostatin (50 IU/ml). I n the bottles the atmospheric air was replaced by a mixture of 95 O/o C02. I n four bottles pH between 7.31 and 7.35 and osmotic pressures air and 5 between 290 and 296 Mcsm were found. Corneas were incubated for time periods ranging from five min to 48 h. After incubation the endothelium was stained by alizarine red-alkohol-trypane blue and whole wet monts of corneal endothelium were prepared (Sperling 1977). Alizarine red-alcohol-trypane blue stain cell borders purple and all nuclei blue. Corneas prepared for scanning elctron microscopy were fixed in 3 (I/o glutaraldehyde in 0.15 M phosphate buffer, pH 7.4 for six h at 25OC. After fixation, the tissue was transferred to pure buffer, dehydrated in graded ethanol, critical point dried from carbon dioxide, vacuum coated with gold and examined in a Cambridge Stereoscan 600 scanning electron microscope.
Results Human corneas were exceedingly sensitive to trauma from tear or bend 8-36 h post mortem. Damage due to handling appeared as converging o r concentric rows of blue nuclei in trypane blue staining. I n markedly hypotonic eyes with corneal infolding, concentric rows of celhlar debris were stained red by alizarine red and blue by trypane blue. In atraumatically removed corneas, a few damaged endothelial cells were evenly distributed in unfolded areas. Damaged cells appeared singly or in small groups of three to seven cells. On the crests of corneal folds rows of damaged cells were often found. No blue endothelial areas could be discerned by the naked eye after trypane blue staining. 787
S t e f f e n Sperling
Figs. 6, 7 , 8 a n d 9. Figs. 6 and 7. Scanning electron microscopic pictures of the posterior cell membrane after 25 rnin of incubation. Magnification: 238 x. (Photo: Steen Roj Jacobsen). Figs. S a n d 9. Scanning electron microscopic pictures of cellular remnants on normal converging endothelial cells after 45 min of incubation. Magnification: 238 x. (Photo: Steen R o j Jacobsen).
Damaged cells were polygonal after five min of incubation. After 15-60 min of incubation, single rounded cells with a blue stained nucleus and large oval spots heavily stained by alizarine red appeared (Fig. 3). Alteration from the polygonal cell shape was not observed in damaged cells surrounded by other damaged cells. Joint meetings of groups of five to eight cells with or without central spots were aIso observed (Figs. 4 and 5 ) . Scanning electron microscopy of endothelium incubated for 25 and 45 rnin showed that the red spots observed in alizarine red stain were cellular remnants or cells with disorganization ranging from slight posterior membrane changes to expulsion of intercellular organelles. The damaged cells were overlying neighbouring cells with normal posterior surfaces and elongated borders converging towards the centre of the damaged ceIls (Figs. 6-9). After two h of incubation the phenomena described above were rare. Joint 788
Organ Cultwed Hiimaz Cornea
meetings of groups of four cells and cells showing marked deviations from a regular polygonal shape were the dominating morphological abnormalities. Formation and reformation of a joint meeting of six originally hexagonal cells is shown in Fig. 10. Six hexagonal cells form a joint meeting by sliding after loss of a central seventh cell (Figs. 10 a-c). The configuration shown in Fig. 10 d appears by retraction of one cell from the centre. One of the constellations in Fig. 10 e, f or g is formed by retraction of one more cell. Restoration of a pattern with joint meetings of no more than three cells will lead to one of the constellations in Fig. 10 h, i, j or k. After 48 h of incubation, either one of the morphological constellations shown in Fig. 10 appeared. Joint meetings of groups of four cells were the most common abnormality.
Comments Dark round spots observed by specular microscopy were reported by Leibowitz et al. (1974), Sherrard (1974) and Coles (1975) after exposure of rabbit endothelium to air, antibiotics or refrigerator storage. These spots are seen as the two damaged cells in Figs. 6 and 7. Time lapse cinematography showed that spots developed from single hexagonal cells and that the spreading of the spots
Fig. 10.
Formation of a joint meeting of six originally hexagonal cells. d-k. Reformation of joint meetings of three cells. a-c.
7 89
Steffen Sperling
Fig. 11. a-b. Death of a hexagonal cell in a pleomorph population and reformation of joint meetings of three cells as indicated in Fig. 10 i.
over intact neighbouring cells occurred suddenly (Coles 1975). Sherrard (1976) showed that the spots were dead cells. Joint meetings of more than three cells in rabbit endothelium were originally described as rosette formation by Oh & Evans (1960) who considered the rosette to be a specific lesion related in some way to virus infection of the endothelial cells. Recently Sherrard (1976) showed that the rosette represented the initial stage of cellular repair after provoked cell death in rabbit epithelium. Comparison of th results of this study to the studies mentioned above indicates that the initial process of repair after death of one cell surrounded by live neighbouring cells is identical in rabbit and in human corneal endothelium. On rabbits under Nembutal anaesthesia Sherrard (1976) found that most rosettes were reorganized within one h. T h e time sequence of the dominating morphological abnormalities found in this study suggests a similar rate of rosette reformation in human endothelium in organ culture. The principles of rosette formation and reformation are shown in Fig. 10. The cells do not necessarily go through all the stages shown, and reformation by simultaneous movement of more than one of the cells in a joint meeting is more likely than movement of one cell at the time. In a pure population of hexagons, cell death leads to pleomorphism (Fig. 10). In a mixed population of penta-, hexa-, septa- and octagons, cell death does not necessarily lead to increased pleomorphism. In Fig. 11 a and b the pattern is reformed as shown principally in Fig. 10 i. One pentagon and one septagon is still left after death of a hexagon. A joint meeting of more than three cells is rarely observed in normal human endothelium. The common occurrence of joint meetings of four cells in organ cultured endothelium may be due to a decreased tendency of the cells to reform a normal pattern or to continuous cell death. T h e observation of rosettes after 48 h of incubation suggests a continuous process of cell death in organ culture.
790
Organ Cultured H u m a n Cornea
Acknowledgments The author wishes to express his gratitude to F. Jsrgensen, M. D., Tissue Typing Laboratory, Arhus Kommunehospital, and to A. Jepsen, D. D. S., Institute of oral pathology, Royal Dental College, Arhus, for advise on the organ culture technique, and to B. Olesen, Department of Ophthalmology, Arhus Kommunehospital, for fruitful discussions and technical assistance. This study was supported by a grant from the Danish Committee for the Prevention of Blindness.
References Coles W. H. (1975) Effects of antibiotics on the in vitro rabbit corneal endothelium. Invest. Ophthal. 14, 246-250. Doughman D. J., Van Horn D., Harris J. D., Miller G. E., Summerlin W. & Good R. A. (1973) Endothelium of the human organ cultured cornea: An electron microscopic study. Trans. amer. ophthal. Soc. 71, 304-328. Leibowitz H. M., Laing R. A. & Sandstrom M. (1974) Corneal Endotheliuni. The effect of air in the anterior chamber. Arch. Ophthul. (Chicago) 92, 227-230. Lindstrom R. L., Doughman D. J., Van Horn D. L., Dancil D. & Harris J. E. (1976) A metabolic and electron microscopic study of human organ-cultured cornea. Amer. /. Ophthal. 82, 72-82. Norn M. S. (1971) Vital staining of corneal endothelium in cataract extraction. Acta ophthal. (Kbh.) 49, 725-733. Oh J. 0. & Evans C. S. (1960) Suppressive effects of pyrilamine maleate and d-lysergic acid diethylamide (LSD-25) on early corneal lesions produced in uitro by Newcastle Disease Virus (NDV) and compound 48/80. Virology 10, 127-143. Schrsder H. D. & Sperling S. (1977) Polysaccharide coating of human corneal endothelium. Acta ophthal. (Kbh.) 55, 819-826. Sherrard E. S. (1974) The corneal endothelium in uitro: Its survival during banking at 4'C. Trans. Ophthalmol. Soc. U.K. 94, 80-93. Sherrard E. S. (1976) The corneal endothelium in vivo: Its response to mild trauma. E x p Eye Res. 22, 347-357. Sperling S. (1977) Combined staining of corneal endothelium by alizarine red and trypane blue. Acta ophthal. (Kbh.) 55, 1-8. Sperling S. (1978) Indirect evaluation of corneal endothelial cell density. Acta ophthal. (Kbh.) 56, 445-454. Stocker F. W., King E. H., Lucas D. 0. & Georgiade N. A. (1966) A comparison of two different staining methods for evaluating corneal endothelial viability. Arch. Ophthul. (Chicago) 76, 833-835. Stocker F. W., King E. H., 1,ucas D. 0. &: Georgiade N. A. (1970) Clinical t a t for evaluating donor corneas. Arch. Ophthal. (Chicago) 84, 2-7.
791
Steffen Sperling
Van Horn D. L. (1971) Evaluation of trypane blue staining of human corneal endothelium. In: Capella J. A,, Edelhauser H. F. and Van Horn D. L., Eds. Corneal Preservation. Thomas, Springfield. Van Horn D. L., Schultz R. 0. & De Bruin J. (1975) Endothelial survival in corneal tissue stored in M-K medium. Amer. J . Ophthaf. 80,612-647.
Author’s address: Steffen Sperling, Department of Ophthalmology, Arhus Kommunehospital, 8000 Arhus C, Denmark.
792
Department of Ophthalmology, Arhus Kommunehospital (Head: N . EhEers), Uniuesity of Aarhus, Denmark
EARLY MORPHOLOGICAL CHANGES IN ORGAN CULTURED HUMAN CORNEAL ENDOTHELIUM BY STEFFEN SPERLING
Nineteen human cadaver corneas with few damaged endothelial cells were incubated under tissue culture conditions for time periods ranging from five min to 48 h. Morphological alterations of the endothelial cells were studied in whole wet mounts stained by alizarine red-alkohol-trypane blue and by scanning electron microscopy. Joint meetings of three cells are characterisic for normal corneal endothelium. After 15-60 min of incubation, damaged cells were expelled from the coherent cell sheet by expanding neighbouring cells. Joint meetings of 5-8 expanding cells were formed. After 24 h of incubation, joint meetings of four cells were the dominating morphological abnormality. Morphological changes during reduction of the numbers of cells in joint meetings are described.
Key words: human cornea - endothelium - morphology - reorganization alizarine red - trypane blue - scanning electron microscopy - organ culture.
In normal human corneas the membrane of Descemet is covered by a single layer of coherent endothelial cells. Each cell is surrounded by four to eight neighbouring cells (Sperling 1978). The cells form joint meetings of three (Fig. 1). Although the number of viable endothelial cells decreases post mortem, a coherent sheet of normal endothelial cells has been observed on whole corneas obtained more than 24 h post mortem, after storage in organ culture (DoughReceived April 5 , 1978.
785
Steffen Sperling m a n e t al. 1973; Lindstrom et al. 1976). T h i s indicates that a coherent cell layer must h a v e been reformed in vitro. In the present report e a r ly morphological changes in t h e endothelium of whole h u m a n corneas incubated u n d e r tissue culture conditions a r e described.
Material and Methods Twenty-nine human eyes were obtained 8-36 h post mortem from cadavers stored at 15-22OC for 8-10 h, and later at 4OC. Corneas in situ were rinsed in tap water. Later whole eyes were submerged in 20 ml of a n isotonic solution of Bacitracin (500 IU/ml), Polymixin B (5000 IU/ml), and Neomycinsulfate (2.5 mg/ml) for 15 min at 25OC. Corneas were excised with a scleral rim and the endothelium was stained for one min by 0.3 per cent trypane blue in 0.9 per cent NaC1, sterilized through a 0.2 p filter. After excision from a cadaver bulbus, the intercellular spaces between intact endothelial cells swell in 0.9 per cent NaCl (Schrerder & Sperling 1977). Swollen intercellular spaces are visible in a n ordinary light microscope. By staining with trypane blue in NaCI, the relative number of damaged cells can be estimated. Trypane blue stain nuclei of damaged cells deep blue in uiuo (Stocker et al. 1966, 1970; Van Horn et al. 1975). The nuclei of undamaged cells are not stained. Uncovered parts of the membrane of Descemet stain blue. Living cells are not damaged by trpane blue (Stocker et al. 1970; Norn 1971; Van Horn 1971). An exact measure of endothelial cell density before incubation could not be obtained by staining of cell borders with alizarine red, as this stain kills the cells (Fig. 2). Nineteen corneas, in which less than one per cent of damaged endothelial cells was found in unfolded areas, were suspended in sterilized 100 ml glass infusion bottles by
Figs. 1 and 2. Fig. 1. Whole wet mount of normal corneal endothelial cells. Stained by alizarine red treated with alcohol and stained trypane blue Magnification: 138 x. Fig. 2. Coherent dead cells killed by alizarine red staining. Stained by trypane blue, after 12 h of incubation in tissue culture medium. The alizarine red has diffused away in the aqueous medium. Magnification: 138 x.
786
O r g a n Cultured H u m a n Cornea
Figs. 3, 4 and 5. Fig. 3. Red dots heavily stained by alizarine red after 25 min of incubation. Stained by alizarine red-alcohol-trypane blue. Magnification: 138 x. Figs. 4 a n d 5. Joint meetings of live to eight cells after 25 and 35 min of incubation. Stained by alizarine red-alcohol-trypane blue. Magnification: 138 x .
a suture tied to the scleral rim and incubated in tissue culture medium at 37OC. The medium was composed of Eagles minimum essential medium, L-glutamine (2 md ml ), 20 O/o foetal calf serum, amikacin (0.1 mg/ml). carbenicillin (0.5 mgiml) and mycostatin (50 IU/ml). I n the bottles the atmospheric air was replaced by a mixture of 95 O/o C02. I n four bottles pH between 7.31 and 7.35 and osmotic pressures air and 5 between 290 and 296 Mcsm were found. Corneas were incubated for time periods ranging from five min to 48 h. After incubation the endothelium was stained by alizarine red-alkohol-trypane blue and whole wet monts of corneal endothelium were prepared (Sperling 1977). Alizarine red-alcohol-trypane blue stain cell borders purple and all nuclei blue. Corneas prepared for scanning elctron microscopy were fixed in 3 (I/o glutaraldehyde in 0.15 M phosphate buffer, pH 7.4 for six h at 25OC. After fixation, the tissue was transferred to pure buffer, dehydrated in graded ethanol, critical point dried from carbon dioxide, vacuum coated with gold and examined in a Cambridge Stereoscan 600 scanning electron microscope.
Results Human corneas were exceedingly sensitive to trauma from tear or bend 8-36 h post mortem. Damage due to handling appeared as converging o r concentric rows of blue nuclei in trypane blue staining. I n markedly hypotonic eyes with corneal infolding, concentric rows of celhlar debris were stained red by alizarine red and blue by trypane blue. In atraumatically removed corneas, a few damaged endothelial cells were evenly distributed in unfolded areas. Damaged cells appeared singly or in small groups of three to seven cells. On the crests of corneal folds rows of damaged cells were often found. No blue endothelial areas could be discerned by the naked eye after trypane blue staining. 787
S t e f f e n Sperling
Figs. 6, 7 , 8 a n d 9. Figs. 6 and 7. Scanning electron microscopic pictures of the posterior cell membrane after 25 rnin of incubation. Magnification: 238 x. (Photo: Steen Roj Jacobsen). Figs. S a n d 9. Scanning electron microscopic pictures of cellular remnants on normal converging endothelial cells after 45 min of incubation. Magnification: 238 x. (Photo: Steen R o j Jacobsen).
Damaged cells were polygonal after five min of incubation. After 15-60 min of incubation, single rounded cells with a blue stained nucleus and large oval spots heavily stained by alizarine red appeared (Fig. 3). Alteration from the polygonal cell shape was not observed in damaged cells surrounded by other damaged cells. Joint meetings of groups of five to eight cells with or without central spots were aIso observed (Figs. 4 and 5 ) . Scanning electron microscopy of endothelium incubated for 25 and 45 rnin showed that the red spots observed in alizarine red stain were cellular remnants or cells with disorganization ranging from slight posterior membrane changes to expulsion of intercellular organelles. The damaged cells were overlying neighbouring cells with normal posterior surfaces and elongated borders converging towards the centre of the damaged ceIls (Figs. 6-9). After two h of incubation the phenomena described above were rare. Joint 788
Organ Cultwed Hiimaz Cornea
meetings of groups of four cells and cells showing marked deviations from a regular polygonal shape were the dominating morphological abnormalities. Formation and reformation of a joint meeting of six originally hexagonal cells is shown in Fig. 10. Six hexagonal cells form a joint meeting by sliding after loss of a central seventh cell (Figs. 10 a-c). The configuration shown in Fig. 10 d appears by retraction of one cell from the centre. One of the constellations in Fig. 10 e, f or g is formed by retraction of one more cell. Restoration of a pattern with joint meetings of no more than three cells will lead to one of the constellations in Fig. 10 h, i, j or k. After 48 h of incubation, either one of the morphological constellations shown in Fig. 10 appeared. Joint meetings of groups of four cells were the most common abnormality.
Comments Dark round spots observed by specular microscopy were reported by Leibowitz et al. (1974), Sherrard (1974) and Coles (1975) after exposure of rabbit endothelium to air, antibiotics or refrigerator storage. These spots are seen as the two damaged cells in Figs. 6 and 7. Time lapse cinematography showed that spots developed from single hexagonal cells and that the spreading of the spots
Fig. 10.
Formation of a joint meeting of six originally hexagonal cells. d-k. Reformation of joint meetings of three cells. a-c.
7 89
Steffen Sperling
Fig. 11. a-b. Death of a hexagonal cell in a pleomorph population and reformation of joint meetings of three cells as indicated in Fig. 10 i.
over intact neighbouring cells occurred suddenly (Coles 1975). Sherrard (1976) showed that the spots were dead cells. Joint meetings of more than three cells in rabbit endothelium were originally described as rosette formation by Oh & Evans (1960) who considered the rosette to be a specific lesion related in some way to virus infection of the endothelial cells. Recently Sherrard (1976) showed that the rosette represented the initial stage of cellular repair after provoked cell death in rabbit epithelium. Comparison of th results of this study to the studies mentioned above indicates that the initial process of repair after death of one cell surrounded by live neighbouring cells is identical in rabbit and in human corneal endothelium. On rabbits under Nembutal anaesthesia Sherrard (1976) found that most rosettes were reorganized within one h. T h e time sequence of the dominating morphological abnormalities found in this study suggests a similar rate of rosette reformation in human endothelium in organ culture. The principles of rosette formation and reformation are shown in Fig. 10. The cells do not necessarily go through all the stages shown, and reformation by simultaneous movement of more than one of the cells in a joint meeting is more likely than movement of one cell at the time. In a pure population of hexagons, cell death leads to pleomorphism (Fig. 10). In a mixed population of penta-, hexa-, septa- and octagons, cell death does not necessarily lead to increased pleomorphism. In Fig. 11 a and b the pattern is reformed as shown principally in Fig. 10 i. One pentagon and one septagon is still left after death of a hexagon. A joint meeting of more than three cells is rarely observed in normal human endothelium. The common occurrence of joint meetings of four cells in organ cultured endothelium may be due to a decreased tendency of the cells to reform a normal pattern or to continuous cell death. T h e observation of rosettes after 48 h of incubation suggests a continuous process of cell death in organ culture.
790
Organ Cultured H u m a n Cornea
Acknowledgments The author wishes to express his gratitude to F. Jsrgensen, M. D., Tissue Typing Laboratory, Arhus Kommunehospital, and to A. Jepsen, D. D. S., Institute of oral pathology, Royal Dental College, Arhus, for advise on the organ culture technique, and to B. Olesen, Department of Ophthalmology, Arhus Kommunehospital, for fruitful discussions and technical assistance. This study was supported by a grant from the Danish Committee for the Prevention of Blindness.
References Coles W. H. (1975) Effects of antibiotics on the in vitro rabbit corneal endothelium. Invest. Ophthal. 14, 246-250. Doughman D. J., Van Horn D., Harris J. D., Miller G. E., Summerlin W. & Good R. A. (1973) Endothelium of the human organ cultured cornea: An electron microscopic study. Trans. amer. ophthal. Soc. 71, 304-328. Leibowitz H. M., Laing R. A. & Sandstrom M. (1974) Corneal Endotheliuni. The effect of air in the anterior chamber. Arch. Ophthul. (Chicago) 92, 227-230. Lindstrom R. L., Doughman D. J., Van Horn D. L., Dancil D. & Harris J. E. (1976) A metabolic and electron microscopic study of human organ-cultured cornea. Amer. /. Ophthal. 82, 72-82. Norn M. S. (1971) Vital staining of corneal endothelium in cataract extraction. Acta ophthal. (Kbh.) 49, 725-733. Oh J. 0. & Evans C. S. (1960) Suppressive effects of pyrilamine maleate and d-lysergic acid diethylamide (LSD-25) on early corneal lesions produced in uitro by Newcastle Disease Virus (NDV) and compound 48/80. Virology 10, 127-143. Schrsder H. D. & Sperling S. (1977) Polysaccharide coating of human corneal endothelium. Acta ophthal. (Kbh.) 55, 819-826. Sherrard E. S. (1974) The corneal endothelium in uitro: Its survival during banking at 4'C. Trans. Ophthalmol. Soc. U.K. 94, 80-93. Sherrard E. S. (1976) The corneal endothelium in vivo: Its response to mild trauma. E x p Eye Res. 22, 347-357. Sperling S. (1977) Combined staining of corneal endothelium by alizarine red and trypane blue. Acta ophthal. (Kbh.) 55, 1-8. Sperling S. (1978) Indirect evaluation of corneal endothelial cell density. Acta ophthal. (Kbh.) 56, 445-454. Stocker F. W., King E. H., Lucas D. 0. & Georgiade N. A. (1966) A comparison of two different staining methods for evaluating corneal endothelial viability. Arch. Ophthul. (Chicago) 76, 833-835. Stocker F. W., King E. H., 1,ucas D. 0. &: Georgiade N. A. (1970) Clinical t a t for evaluating donor corneas. Arch. Ophthal. (Chicago) 84, 2-7.
791
Steffen Sperling
Van Horn D. L. (1971) Evaluation of trypane blue staining of human corneal endothelium. In: Capella J. A,, Edelhauser H. F. and Van Horn D. L., Eds. Corneal Preservation. Thomas, Springfield. Van Horn D. L., Schultz R. 0. & De Bruin J. (1975) Endothelial survival in corneal tissue stored in M-K medium. Amer. J . Ophthaf. 80,612-647.
Author’s address: Steffen Sperling, Department of Ophthalmology, Arhus Kommunehospital, 8000 Arhus C, Denmark.
792
