A C T A O P H T H A L M O L O G I C A VOL. 53 1 9 7 5
The Universily Eye Deparlment (Head: Th. L. Thomassen, M . D . ) and the Inslitute o j Pathology, Electron Microscopic Laboratory (Head: T . Hoaig, M . D . ) , Rikshospitalet, Oslo, Norway
EFFECTS OF BENZALKONIUM CHLORIDE UPON THE CORNEAL EPITHELIUM STUDIED WITH SCANNING ELECTRON MICROSCOPY BY
ASBJ0RN M. T0NJUM
The corneal surface of rabbits and vervet monkeys was studied by means of scanning electron microscopy after the influence of the cationic surfactant, benzalkonium chloride. The concentration of the drug was 0.02 O/o or 0.01 O/o and the exposure time was 2 or 4 min. T h e corneas treated with the drug had characteristic changes with small holes or more extensive lesions of the superficial cell membranes, and loss of microvilli or microplicae. The damage was considered to be due to the lytic effect of benzalkonium chloride upon the plasma membranes.
K e y words: benzalkonium chloride - surfactant - detergent - cornea epithelium - scanning electron microscopy - permeability.
The epithelium is a barrier which prevents water, solutes, and large molecules entering the cornea. T h e most important location of this barrier has been shown to be the anterior surface of the epithelium, the superficial plasma membranes and the intercellular junctional complexes between the most superficial cells (Tanjum 1974). Benzalkonium chloride, when applied to the anterior corneal surface; breaks down this barrier, presumably by producing defects of the plasma membranes Received January 22, 1975.
358
Benialkonium and Scunning Microscopy
Fig. I . Surface of normal rabbit cornea showing excrescences. Intercellular borders are present. A: x 3,000. B: x 10,000.
359
Asbjern
M.Tenjtcm
as indicated by transmission electron microscopic studies (Tsnjum 1975). Since benzalkonium chloride is a widely used preservative in eye drops, it is important to elucidate the effect of this drug upon the corneal epithelium. The purpose of the present work was to study the probable defects of the corneal surface after the influence of benzalkonium chloride (BA) by means of scanning electron microscopy (SEM).
Material and Methods Albino rabbits weighing 3-3.5 kg and vervet monkeys (Cercopithecus aethiops) weighing 4-4.5 kg were used. Immediately after killing the animals with an overdose of sodium pentobarbital the eyes were enucleated and submersed in toto in the BA solution for 2 or 4 min, washed in a buffered solution, and then submersed in 2&.0/0 glutaraldehyde for 1 hour. The corneas, which were now sufficiently fixed to maintain their shape, were excised and left in glutaraldehyde for 2 to 48 hours. KrebsRinger bicarbonate solution with 5 mg glucose per ml was used as the basic solution for washing and dissolving BA. The concentration of BA was 0.01 O/o or 0.020/0. T h e tissues were dehydrated in acetone (Merck) at increasing csncentrations up to 100 O/o, and were dried in the Sorvall carbon dioxide critical point drying system. Immediately afterwards the specimens were coated with molecular carbon and platinum in the Speedivac coating unit (Edwards High Vacuum Ltd.). Th e micrographs were taken with a Jeol JSM-JXA-50A Scanning Electron Microscope.
Observations a. The corneal surface treated with buffered solution only. The corneal surface had a velour-like appearance (Figs. 1 A, 1 B, 4 A), and the intercellular borders were easily identified. Even in the corneas untreated with BA there were holes or penetrating defects in many of the cell membranes, particularly in rabbit corneas. At high magnifications the cell surfaces were seen to be covered with excrescences in the form of microvilli and microplicae. Minor species differences were recognized. T he main findings were in accordance with Pfister (1973) and others who have studied the normal corneal surface.
360
Benzalkonium und Scanning Microscopy
Fig. 2. A: Surface of rabbit cornea exposed to 0.02 O/o benzalkonium chloride for 4 min. Loss of normal surface excrescences. Penetrating defects. x 3,000. B: The same specimen as above. The left cell has penetrating holes, and the right one has destruction of the entire outer plasma membrane. x 10,000.
361
A s b j s r n M. Tsnjum
Fig. 3. Surface of rabbit cornea exposed to 0.01 O/o benzalkonium chloride for 2 min. A: Different degrees of damage in the different cells. Loss of surface excrescences. One cell is apparently undamaged. x 3,000. B: The same specimen as above with small defects of the plasma membrane. x 10,000.
362
Benialkonium and Scanning Microscopy
b. The corneal surface after exposure to BA 0.02 010 f o r 4 min. (Figs. 2 A, 2 B, 5). At low magnifications the surface was smooth. At higher magnifications the surface of all cells were seen to be devoid of excrescences and to have defects of varying degrees. In some cells the entire anterior plasma membrane was disintegrated, whereas others had smaller defects. At high magnifications the smallest holes were circular and had a diameter of about 20 nm (Fig. 5). c. The corneal surface after exjlosure to BA 0.01 010 f o r 2 min (Figs. 3 A , 3 B , B). In principle, the damage was similar to that of the previous experi-
4 A, 4
Fig. 4 . Surface of monkey cornea. A: Normal epithelium. B: Cornea exposed to 0.01 O/o benzalkonium chloride for 2 min. Smooth surface with loss of surface excrescences. x 3,000.
363
Asbjorn M . Tonjztm
Fig. 5 .
Surface of rabbit corneal epithelium exposcd to 0.02°/0 benzalkonium chloride for 4 min. Penetrating defects of the plasma membrane. Some holes are isolated, the smallest ones having a diameter of about 20 nm, others are confluencing. x 30,000.
ment, although less pronounced. One particular feature was the difference in damage of the different cells; from severe disintegration to almost non-detectable damage (Figs. 3 A, 3 B). T h e presence of small holes, however, was a consistent finding in all cells. I n some places these appeared to have confluenced to larger defects. In the monkey eye the damage seemed more severe than in the rabbit.
Comments The effects of BA upon the corneal epithelium have been studied by means of scanning electron microscopy. The main findings were: a) loss of surface excrescences, b) penetrating defects of the superficial cell membrane. The degree 364
Benzalkonicim and Scanning Microscopy
of damage was dependent upon the concentration of BA and the duration of the exposure of this solution to the corneal surface. There was a difference in the response of the cells to the influence of BA, this was particularly seen in the series of experiments when the corneas were exposed to 0.01 O / O BA for 2 min. In the series when the corneas were treated with BA 0.02 " / o for 4 min all cells were severely damaged. The turnover rate of epithelial cells is high, and the differences of vulnerability may reflect the different stages in the life cycle of the cells. hiany surface cells of the corneas not treated with BA had defects of the plasma membrane, too. However, these cells did not have the conspicuous loss of surface excrescences. The surface coating by the tear film most likely influenced the appearance, at least of those corneas not treated with BA. This cationic surfactant exerts a n effect both upon the lipid and the mucinous components of the tear film. Possibly the tear film is partly or totally lost by the BA treatment. However, it was beyond the scope of the present work to rule out this effect. The penetrating holes of the plasma membranes are most likely a primary effect of BA. They were fairly easily differentiated from the defects of the surface cells of corneas not treated with BA. Sites of the plasma membranes might be particularly sensitive to the effect of BA. The intracellular disintegration (Tnnjum 1975) may well be secondary to lysis of the plasma membranes, and to effects upon the membranes surrounding the intracellular organelles as well. Release of lysosomal enzymes may, however, upgrade the process of tissue destruction. Furthermore, Bangham & Horne (1964) demonstrated that surfactants produced holes in phospholipid globules. Evidence of an early effect upon the membranes is also given by Saladino, Hawkins & Trump (197 1) who showed an increase of ion leakage in the toad bladder after the influence of another cationic surfactant, cetyl pyridinium chloride. Hodes, Palmer & Warren (1960) demonstrated entrance of the dye nigrosin into Ehrlich ascites cells after the influence of surface active agents. The possibility is therefore unlikely that the defects of the plasma membranes are secondary to intracellular damage. T h e loss of surface excrescences was a characteristic finding after BA treatment. Most likely BA affects the appearance of the surface partly by acting upon the tear film. Furthermore, the defects of the plasma membrane may result in a reorganization of this structure at an early stage during the experimentation, even though critical point drying gives a n optimal preservation of the architecture of the cells. The loss of surface excrescences could also be caused by the lytic effect of BA at the base of the excrescences. In this case the holes of the plasma membranes could represent the previous sites of the excrescences. The observations by SEM that BA produces penetrating defects of the plasma membranes of the corneal epithelial cells and loss of surface excrescences support
365
Asbjern M . Tsnjum
the studies of the increased corneal permeability given by BA (O’Brien & S w a n 1942, G r e e n & T s n j u m 1971, G r e e n & D o w n s 1974, T s n j u m 1975). The holes of t h e cell membranes, o r the more extensive lesions, o ffe r an e x p l a n a t i o n f o r the mechanism of this permeability increase
Acknowledgements The valuable advice in the preparation of the specimens and the use of the scanning electron microscope given by Mrs. Sigrid Lystad, and the financial support from the Norwegian Research Council for Science and Humanities is gratefully acknowledged.
References Bangham, A. D. & Horne, R. W. (1964) Negative staining of phospholipids and their structural modification by surface-active agents as observed in the electron microscope. J. Mol. Biol. 8, 660-668. Green, K. & Downs, S. J. (1974) Prednisolone phosphate penetration into and through the cornea. Invest. Ophthal. 1.7, 316-319. Green, K . & Tenjum, A. M. (1971) Influence of various agents on corneal permeability. Amer. J. Ophthal. 72, 897-905. Hodes, M. E., Palmer, C. G. & Warren, A. (1960) The effect of surface active agents on the permeability to dye of the plasma membrane of Ehrlich ascites cells. Ex!). Cell Re.$.21, 161-169. O’Brien, C. S. & Swan, K. C. (1942) Carbaminoylcholine chloride in the treatment of glaucoma simplex. Arch. Ophlhal. 27, 253-263. Pfister, R. R. (1973) The normal surface of the corneal epithelium: a scanning electron microscopic study. Invest. Ophthal. 12, 654-668. Saladino, A. J., Hawkins, H. K. & Trump, B. F. (1971) Ion movements in cell injury. Effects of the cationic detergent cetyl pyridinium chloride on the ultrastructure and function of the toad bladder. Amer. J. Path. 64, 271-286. Tenjum, A. M. (1974) Permeability of horseradish peroxidase in the rabbit corneal epithelium. Acta ophthal. (Kbh.) 52, 650-658. Tenjum, A. M. (1975) Permeability of the rabbit corneal epithelium to horseradish peroxidase after the influence of benzalkonium chloride. Acta oplzthal. (Kbh.) 53, 335347.
Author’s address: Dr. Asbjern M. Tenjum, University Eye Department, Rikshospi talet,
Oslo 1 , Norway.
366
The Universily Eye Deparlment (Head: Th. L. Thomassen, M . D . ) and the Inslitute o j Pathology, Electron Microscopic Laboratory (Head: T . Hoaig, M . D . ) , Rikshospitalet, Oslo, Norway
EFFECTS OF BENZALKONIUM CHLORIDE UPON THE CORNEAL EPITHELIUM STUDIED WITH SCANNING ELECTRON MICROSCOPY BY
ASBJ0RN M. T0NJUM
The corneal surface of rabbits and vervet monkeys was studied by means of scanning electron microscopy after the influence of the cationic surfactant, benzalkonium chloride. The concentration of the drug was 0.02 O/o or 0.01 O/o and the exposure time was 2 or 4 min. T h e corneas treated with the drug had characteristic changes with small holes or more extensive lesions of the superficial cell membranes, and loss of microvilli or microplicae. The damage was considered to be due to the lytic effect of benzalkonium chloride upon the plasma membranes.
K e y words: benzalkonium chloride - surfactant - detergent - cornea epithelium - scanning electron microscopy - permeability.
The epithelium is a barrier which prevents water, solutes, and large molecules entering the cornea. T h e most important location of this barrier has been shown to be the anterior surface of the epithelium, the superficial plasma membranes and the intercellular junctional complexes between the most superficial cells (Tanjum 1974). Benzalkonium chloride, when applied to the anterior corneal surface; breaks down this barrier, presumably by producing defects of the plasma membranes Received January 22, 1975.
358
Benialkonium and Scunning Microscopy
Fig. I . Surface of normal rabbit cornea showing excrescences. Intercellular borders are present. A: x 3,000. B: x 10,000.
359
Asbjern
M.Tenjtcm
as indicated by transmission electron microscopic studies (Tsnjum 1975). Since benzalkonium chloride is a widely used preservative in eye drops, it is important to elucidate the effect of this drug upon the corneal epithelium. The purpose of the present work was to study the probable defects of the corneal surface after the influence of benzalkonium chloride (BA) by means of scanning electron microscopy (SEM).
Material and Methods Albino rabbits weighing 3-3.5 kg and vervet monkeys (Cercopithecus aethiops) weighing 4-4.5 kg were used. Immediately after killing the animals with an overdose of sodium pentobarbital the eyes were enucleated and submersed in toto in the BA solution for 2 or 4 min, washed in a buffered solution, and then submersed in 2&.0/0 glutaraldehyde for 1 hour. The corneas, which were now sufficiently fixed to maintain their shape, were excised and left in glutaraldehyde for 2 to 48 hours. KrebsRinger bicarbonate solution with 5 mg glucose per ml was used as the basic solution for washing and dissolving BA. The concentration of BA was 0.01 O/o or 0.020/0. T h e tissues were dehydrated in acetone (Merck) at increasing csncentrations up to 100 O/o, and were dried in the Sorvall carbon dioxide critical point drying system. Immediately afterwards the specimens were coated with molecular carbon and platinum in the Speedivac coating unit (Edwards High Vacuum Ltd.). Th e micrographs were taken with a Jeol JSM-JXA-50A Scanning Electron Microscope.
Observations a. The corneal surface treated with buffered solution only. The corneal surface had a velour-like appearance (Figs. 1 A, 1 B, 4 A), and the intercellular borders were easily identified. Even in the corneas untreated with BA there were holes or penetrating defects in many of the cell membranes, particularly in rabbit corneas. At high magnifications the cell surfaces were seen to be covered with excrescences in the form of microvilli and microplicae. Minor species differences were recognized. T he main findings were in accordance with Pfister (1973) and others who have studied the normal corneal surface.
360
Benzalkonium und Scanning Microscopy
Fig. 2. A: Surface of rabbit cornea exposed to 0.02 O/o benzalkonium chloride for 4 min. Loss of normal surface excrescences. Penetrating defects. x 3,000. B: The same specimen as above. The left cell has penetrating holes, and the right one has destruction of the entire outer plasma membrane. x 10,000.
361
A s b j s r n M. Tsnjum
Fig. 3. Surface of rabbit cornea exposed to 0.01 O/o benzalkonium chloride for 2 min. A: Different degrees of damage in the different cells. Loss of surface excrescences. One cell is apparently undamaged. x 3,000. B: The same specimen as above with small defects of the plasma membrane. x 10,000.
362
Benialkonium and Scanning Microscopy
b. The corneal surface after exposure to BA 0.02 010 f o r 4 min. (Figs. 2 A, 2 B, 5). At low magnifications the surface was smooth. At higher magnifications the surface of all cells were seen to be devoid of excrescences and to have defects of varying degrees. In some cells the entire anterior plasma membrane was disintegrated, whereas others had smaller defects. At high magnifications the smallest holes were circular and had a diameter of about 20 nm (Fig. 5). c. The corneal surface after exjlosure to BA 0.01 010 f o r 2 min (Figs. 3 A , 3 B , B). In principle, the damage was similar to that of the previous experi-
4 A, 4
Fig. 4 . Surface of monkey cornea. A: Normal epithelium. B: Cornea exposed to 0.01 O/o benzalkonium chloride for 2 min. Smooth surface with loss of surface excrescences. x 3,000.
363
Asbjorn M . Tonjztm
Fig. 5 .
Surface of rabbit corneal epithelium exposcd to 0.02°/0 benzalkonium chloride for 4 min. Penetrating defects of the plasma membrane. Some holes are isolated, the smallest ones having a diameter of about 20 nm, others are confluencing. x 30,000.
ment, although less pronounced. One particular feature was the difference in damage of the different cells; from severe disintegration to almost non-detectable damage (Figs. 3 A, 3 B). T h e presence of small holes, however, was a consistent finding in all cells. I n some places these appeared to have confluenced to larger defects. In the monkey eye the damage seemed more severe than in the rabbit.
Comments The effects of BA upon the corneal epithelium have been studied by means of scanning electron microscopy. The main findings were: a) loss of surface excrescences, b) penetrating defects of the superficial cell membrane. The degree 364
Benzalkonicim and Scanning Microscopy
of damage was dependent upon the concentration of BA and the duration of the exposure of this solution to the corneal surface. There was a difference in the response of the cells to the influence of BA, this was particularly seen in the series of experiments when the corneas were exposed to 0.01 O / O BA for 2 min. In the series when the corneas were treated with BA 0.02 " / o for 4 min all cells were severely damaged. The turnover rate of epithelial cells is high, and the differences of vulnerability may reflect the different stages in the life cycle of the cells. hiany surface cells of the corneas not treated with BA had defects of the plasma membrane, too. However, these cells did not have the conspicuous loss of surface excrescences. The surface coating by the tear film most likely influenced the appearance, at least of those corneas not treated with BA. This cationic surfactant exerts a n effect both upon the lipid and the mucinous components of the tear film. Possibly the tear film is partly or totally lost by the BA treatment. However, it was beyond the scope of the present work to rule out this effect. The penetrating holes of the plasma membranes are most likely a primary effect of BA. They were fairly easily differentiated from the defects of the surface cells of corneas not treated with BA. Sites of the plasma membranes might be particularly sensitive to the effect of BA. The intracellular disintegration (Tnnjum 1975) may well be secondary to lysis of the plasma membranes, and to effects upon the membranes surrounding the intracellular organelles as well. Release of lysosomal enzymes may, however, upgrade the process of tissue destruction. Furthermore, Bangham & Horne (1964) demonstrated that surfactants produced holes in phospholipid globules. Evidence of an early effect upon the membranes is also given by Saladino, Hawkins & Trump (197 1) who showed an increase of ion leakage in the toad bladder after the influence of another cationic surfactant, cetyl pyridinium chloride. Hodes, Palmer & Warren (1960) demonstrated entrance of the dye nigrosin into Ehrlich ascites cells after the influence of surface active agents. The possibility is therefore unlikely that the defects of the plasma membranes are secondary to intracellular damage. T h e loss of surface excrescences was a characteristic finding after BA treatment. Most likely BA affects the appearance of the surface partly by acting upon the tear film. Furthermore, the defects of the plasma membrane may result in a reorganization of this structure at an early stage during the experimentation, even though critical point drying gives a n optimal preservation of the architecture of the cells. The loss of surface excrescences could also be caused by the lytic effect of BA at the base of the excrescences. In this case the holes of the plasma membranes could represent the previous sites of the excrescences. The observations by SEM that BA produces penetrating defects of the plasma membranes of the corneal epithelial cells and loss of surface excrescences support
365
Asbjern M . Tsnjum
the studies of the increased corneal permeability given by BA (O’Brien & S w a n 1942, G r e e n & T s n j u m 1971, G r e e n & D o w n s 1974, T s n j u m 1975). The holes of t h e cell membranes, o r the more extensive lesions, o ffe r an e x p l a n a t i o n f o r the mechanism of this permeability increase
Acknowledgements The valuable advice in the preparation of the specimens and the use of the scanning electron microscope given by Mrs. Sigrid Lystad, and the financial support from the Norwegian Research Council for Science and Humanities is gratefully acknowledged.
References Bangham, A. D. & Horne, R. W. (1964) Negative staining of phospholipids and their structural modification by surface-active agents as observed in the electron microscope. J. Mol. Biol. 8, 660-668. Green, K. & Downs, S. J. (1974) Prednisolone phosphate penetration into and through the cornea. Invest. Ophthal. 1.7, 316-319. Green, K . & Tenjum, A. M. (1971) Influence of various agents on corneal permeability. Amer. J. Ophthal. 72, 897-905. Hodes, M. E., Palmer, C. G. & Warren, A. (1960) The effect of surface active agents on the permeability to dye of the plasma membrane of Ehrlich ascites cells. Ex!). Cell Re.$.21, 161-169. O’Brien, C. S. & Swan, K. C. (1942) Carbaminoylcholine chloride in the treatment of glaucoma simplex. Arch. Ophlhal. 27, 253-263. Pfister, R. R. (1973) The normal surface of the corneal epithelium: a scanning electron microscopic study. Invest. Ophthal. 12, 654-668. Saladino, A. J., Hawkins, H. K. & Trump, B. F. (1971) Ion movements in cell injury. Effects of the cationic detergent cetyl pyridinium chloride on the ultrastructure and function of the toad bladder. Amer. J. Path. 64, 271-286. Tenjum, A. M. (1974) Permeability of horseradish peroxidase in the rabbit corneal epithelium. Acta ophthal. (Kbh.) 52, 650-658. Tenjum, A. M. (1975) Permeability of the rabbit corneal epithelium to horseradish peroxidase after the influence of benzalkonium chloride. Acta oplzthal. (Kbh.) 53, 335347.
Author’s address: Dr. Asbjern M. Tenjum, University Eye Department, Rikshospi talet,
Oslo 1 , Norway.
366
