Effects of Eosinophil Granule Proteins on Human Corneal Epithelial Cell Viability and Morphology

Transcription

1 Effects of Eosinophil Granule Proteins on Human Corneal Epithelial Cell Viability and Morphology Stefan D. Trocme* Csilla K. Hallberg* Kuljit S. Gill,* Gerald J. Gleich,^ Stephen K. Tyring,% and Miriam M. Brysk,% Purpose. There is mounting evidence that eosinophil granule proteins may cause tissue injury during allergic inflammation of the eye. Therefore, the authors investigated the in vitro effects of human eosinophil major basic protein (MBP), eosinophil cationic protein (ECP), eosinophil peroxidase (EPO), and eosinophil-derived neurotoxin (EDN) on cultured human corneal epithelial cell viability and morphology. Methods. Confluent primary human corneal epithelial cell cultures were exposed to each of the four human eosinophil cationic granule proteins at concentrations ranging from 0 to 100 //g/ml (0, 12.5, 25, 50, and 100 /Ltg/ml) for up to 48 hours in serum-free media. Morphologic changes were assessed by light microscopy at 1, 6, 24, and 48 hours; cell viability was measured using the MTT cell viability assay at 24 hours. Results. Cells treated with MBP and ECP induced a dose-dependent gradual increase in morphologic changes; in contrast, EPO and EDN induced minimal changes in cell morphology. At 24 hours, both MBP and ECP induced statistically significant (P < 0.05) decreases in cell viability at a concentration of 100 //g/ml; EPO induced a significant (P < 0.05) decrease in cell viability at all concentrations tested, and EDN showed no significant reduction of cell viability at any of the concentrations tested. Conclusions. The current study suggests that the human eosinophil granule proteins MBP and ECP affect human corneal epithelial cell viability and morphology in vitro, whereas the protein EPO affects cell viability only. EDN had no significant effect on cell viability or morphology. Hence, MBP, ECP, and EPO perturb the corneal epithelium differentially and may contribute to keratopathy associated with severe ocular allergy. Invest Ophthalmol Vis Sci. 1997; 38: Kecent studies have highlighted the role of the eosinophil as an effector cell actively contributing to the inflammatory process in ocular allergy. The cytoplasmic eosinophil granule contains toxic proteins, including major basic protein (MBP), eosinophil cationic protein (ECP), eosinophil peroxidase (EPO), From the * Department of Ophthalmology and Visual Sciences, Cornea Sendee and Eye Research Laboratory; the \Department of Dermatology, University of Texas Medical Branch, School of Medicine, Galveston, Texas; and the f Department of Immunology and the Allergic Diseases Research Laboratory, Mayo Clinic, Rochester, Minnesota. Supported in part by grants from the John Sealy Memorial Fund (Galveston, TX); Lions Eye Hank (San Antonio, Houston, Austin, TX); from the National Institutes of Health (AI0972S and All 5231); and from Research to Prevent Blindness (Neiu York, NY). Submitted for publication April 4, 1996; revised October IS, 1996; accepted October 22, Proprietary interest category: N. Reprint requests: Stefan D. Trocme, Department of Ophthalmology and Visual Sciences, University of Texas Medical. School, 326 Clinical Sciences Building, Rt. 0787, Room 340, Galveston, TX and eosinophil-derived neurotoxin (EDN), that are released into the extracellular space after degranulation during allergic inflammation. 1 ' 2 The principal granule protein, MBP, may participate in the inflammatory process by exerting cytotoxic effects, 3 ' 4 enhancing mast cell degranulation, 0 and by activating neutrophils. 6 In addition, ECP induces mast cell degranulation 7 and exerts cytotoxic effects. 8 " 10 Similarly, EPO in the presence of a H 2 O 2 -generating system and a halide, exerts cytotoxic effects, 11 " 13 promotes mast cell secretion, 14 and induces neutrophil accumulation at sites of inflammation. 15 Even in the absence of H 2 O 2, EPO demonstrates cytotoxic activity, although these effects are less pronounced EDN exhibits helminthotoxic, 1819 neurotoxic, 20 ' 21 and ribonucleolytic 99 9S acuvity. ^ Both MBP and ECP have been identified at sites of allergic lung tissue injury in patients with asthma. 24 "" 20 Investigative Ophthalmology & Visual Science, March 1997, Vol. 38, No. 3 Copyright. Association for Research in Vision and Ophthalmology 593

2 594 Investigative Ophthalmology 8c Visual Science, March 1997, Vol. 38, No. 3 In vernal keratoconjunctivitis, a severe form of ocular allergy, MBP and ECP have been identified in conjunctiva; the reported tear concentration was 0.22 to 28 /ig/ml for MBP and <2 ^g/ml for ECP. 27 " 29 Furthermore, MBP has been identified in sterile corneal ulcers associated with vernal keratoconjunctivitis. 30 To the best of our knowledge, no attempt has been made to identify EPO or EDN in the tears or the conjunctiva of patients with ocular allergy. In vitro investigations involving organ-cultured rat corneas have demonstrated MBP retardation of corneal epithelial wound healing at concentrations identified in the tears of patients with vernal keratoconjunctivitis. 31 Although ECP slows corneal epithelial wound healing, this only occurs at concentrations considerably higher those identified in tears. 32 The objective of the current study was to investigate the effects of the four human eosinophil granule proteins (MBP, ECP, EPO, and EDN) in vitro on cultured primary human corneal epithelial cell morphology and viability, with special emphasis on MBP, the principle granule protein, to gain further insight into the possible pathogenetic contribution of these eosinophil granule proteins in epithelial keratopathy associated with severe ocular allergy. MATERIALS AND METHODS Purification of Human Eosinophil Cationic Proteins Eosinophils were obtained from patients with hypereosinophilic syndrome by cytapheresis using continuous-flow or semicontinuous-flow cell processors with 6% hydroxyethyl starch in the anticoagulant, as described by Pineda et al. 33 Purification of human MBP, ECP, EPO, and EDN was performed as described elsewhere. 1? ' 34 ' 35 Human Corneal Epithelial Culture Human corneal tissue for research purposes was obtained from the Lions Eye Bank (Houston, San Antonio, Austin, TX). A 9 mm disposable trephine was used to punch central corneal buttons from Optisol- GS (Chiron Vision, Irvine, CA) preserved human corneal tissue. Corneal buttons were incubated in Ca 2+ and Mg 2+ -free Dubecco's phosphate-buffered saline containing 0.25% pronase and 0.25% trypsin for 1 hour at 37 C; then fetal bovine serum was added to neutralize the enzymes. Detached epithelial cells were harvested, washed twice with medium, and cultured in serum-free keratinocyte growth medium in tissue flasks previously coated with a collagen-fibronectin mix (Biological Research Faculty and Facility, Ijamsville, MD). Keratinocyte growth medium is keratinocyte basal medium supplemented with bovine pituitary extract, gentamicin sulfate, amphotericin-b, hydrocortisone, human epidermal growth factor, and bovine insulin. Once epithelial cells grew to confluency, they were detached with 0.025% trypsin -I- EDTA, washed, and recultured in 24-well cell culture plates (2 X 10 5 cells/well) coated with the collagen-fibronectin mix. The epithelial origin of the cells was confirmed by immunohistochemistry with monoclonal anti-cytokeratin hybridoma culture supernatants AE1/AE3 (Boehringer Mannheim, Indianapolis, IN). Culture supernatant AE1 recognizes acidic keratins, whereas AE3 reacts with basic keratins. Evaluation of Cellular Morphology After reaching confluency, the cells were treated with MBP, ECP, EPO, or EDN added to keratinocyte growth medium medium at 0, 12.5, 25, 50, and 100 //g/ml concentrations for up to 48 hours at 37 C. Effects of the eosinophil granule proteins on cell morphology were observed using an inverted light microscope (TMS; Nikon, Tokyo, Japan) and were photographed before treatment, and at various times after treatment, to document possible changes in morphology, such as vacuolation, bleb formation, and granulation of the cytoplasm MTT Colorimetric Cell Viability Microassay The MTT cell viability assay was chosen to assess changes in cell viability in response to the eosinophil granule proteins. 38 The assay was initiated at 19.5 hours by adding 20 //I from a 5 mg/ml MTT (3-(4,5-dimethylthiazol-2-yl)- 2,5-diphenyltetrazolium bromide) stock solution to each well. Cells were incubated for an additional 4.5 hours at 37 C. Cell viability was measured at 24 hours. The mitochondrial enzymes in the viable corneal epithelial cells reduced MTT to blue crystals of formazan. Formazan crystals were dissolved with 100 //I of 10% sodium dodecyl sulfate. Absorbency was measured in an enzyme-linked immunosorbent assay reader at 590 nm. Viable cell numbers correlated directly to the concentration of formazan. 39 Statistical Analysis Student's Mest of two independent groups was used to compare treated corneas to the control. A statistically significant difference in the MTT cell viability assay was considered to be P < RESULTS Morphologic Changes Direct observation by light microscopy was used to assess morphologic changes of human corneal epithelial cell cultures after exposure to MBP and the various other granule proteins. Figure 1 shows the morphol-

3 Eosinophil Protein Toxicity on Corneal Epithelium 1 HR. 24 HR. ECP EPO 48 HR. EDN l. The effects of the human eosinophil cationic proteins MBP, ECP, EPO, and EDN on human corneal epithelial cell morphology. The morphology of the cells treated with the 100 /ig/ml concentration of the various granule proteins is presented, (lop two rows) MBP affected cell morphology in a time- and dose-dependent manner. A minimal degree of cytoplasmic granularity was noted at 1 hour that gradually evolved to such changes as vacuolation, bleb formation, and loss of cell adhesion at 24 hours. Extensive change in cell morphology was noted at 48 hours after treatment, (bottom row) Cells treated with ECP for 24 hours demonstrated changes in morphology similar to 24-hour MBP-treated cells; loss of cell adhesion was more pronounced in the ECP group. The cellular architecture was essentially intact compared to control for the 24 hour EPO- and EDN-treated cells. MBP = human eosinophil major basic protein; ECP = eosinophil cationic protein; EPO = eosinophil peroxidase; EDN = eosinophil-derived neurotoxin. FIGURE ogy of cells treated with 100 /-tg/ml of the different granule proteins. Cells treated with MBP were photographed at 1, 24, and 48 hours after treatment and demonstrated time- and dose-related changes in morphology (Fig, 1A). A minimal degree of cytoplasmic granularity was first noted at 1 hour, and it increased slightly at 6 hours with 100 //g MBP/ml. At 24 hours, a modest degree of cytoplasmic granularity was evident at the lowest MBP concentration (12.5 /zg/ml). At the highest MBP concentration (100 /xg/ml), morphologic changes such as vacuolation, bleb formation, and loss of cell adhesion were identified. At 48 hours, a further change in cellular morphology was observed for all MBP concentrations tested. Cell cultures treated with ECP, EPO, or EDN were micrographed before treatment and at 24 hours after treatment. After treatment, micrography was limited to the 24 hour time point because preliminary observations, similar to what was noted for MBP, revealed no obvious changes in cellular morphology at the earlier time points. Changes in morphology, similar to those observed for MBP, were noted in the ECP-

4 596 Investigative Ophthalmology & Visual Science, March 1997, Vol. 38, No CATIONIC PROTEIN (ng/ml) FIGURE 2. The effects of the human eosinophil cationic proteins MBP, ECP, EPO, and EDN on human corneal epithelial cell viability. (A) Corneal epithelial cells subjected to 100 yiig/ ml concentrations of MBP demonstrated a statistically significant (P < 0.05) 20% decrease in cell viability. Corneal epithelial cells subjected to 100 //g/ml concentrations of ECP demonstrated a statistically significant (P < 0.05) 30% decrease in cell viability. (B) Corneal epithelial cells demonstrated a significant (P < 0.05) loss in cell viability at all EPO concentrations tested (12.5 //g/ml = 25%; 25 ^g/ml = 32%; 50 /xg/ml = 48%; 100 //g/ml = 40%). No significant change in corneal epithelial cell viability was detected at any concentration of EDN. Asterisks denote significant values (P < 0.05). MBP = human eosinophil major basic protein; ECP = eosinophil cationic protein; EPO = eosinophil peroxidase; EDN = eosinophil-derived neurotoxin. 100 B CATIONIC PROTEIN (ng/ml) 100 treated cells. However, loss of cell adhesion was more pronounced (Fig. IB). In contrast, EPO-treated cells demonstrated only minimal cytoplasmic granularity at varying times and concentrations (Fig. IB). Cells treated with EDN did not show any visible morphologic changes at any time point or concentration tested (Fig. IB). Cell Viability Because substantial changes in cellular morphology could be detected readily by 24 hours after treatment with two of the eosinophil granule proteins (MBP and ECP), we chose this time point to determine whether there was any associated change in cell viability for all four granule proteins at various concentrations using the MTT cell viability assay. Corneal epithelial cells treated with 100 /Lxg/ml concentrations of MBP or ECP demonstrated a significant (P < 0.05) decrease in cell viability (20% to 30%) compared to controls (Fig. 2, Table 1). At all concentrations tested (12.5 ig/ml = 25%, 25 fj.g/m\ = 32%, 50 /xg/ml = 48%, 100 //g/ml = 40%), EPO demonstrated a significant (P < 0.05) loss in cell viability. On the other hand, EDN showed no significant reduction of cell viability at any of the concentrations tested. DISCUSSION The current study investigated the in vitro effects of the human eosinophil cationic proteins MBP, ECP, EPO, and EDN on cultured human corneal epithelial cell morphology and viability. Results suggest that TABLE l. Percent Human Corneal Epithelial Cell Viability After Treatment With Human Eosinophil Cationic Proteins Mean (% of control) ± SD Concentration (/xg/ml) MBP ECP EPO EDN Control ± ± ± 3.4* ± ± ± ± 2.8* 87.9 ± ± ± ± 5.8* 93.2 ± ± ± ± 6.6* 91.4 ± ± 4.6* 69.5 ± 9.0* 60.4 ± 5.1* ± 7.9 P ^ MBP = human eosinophil major basic protein; ECP = eosinophil cationic protein; EPO = eosinophil peroxidase; EDN = eosinophil-derived neurotoxin.

5 Eosinophil Protein Toxicity on Corneal Epithelium 597 MBP and ECP affect human corneal epithelial cell morphology and viability in culture, whereas EPO affects cell viability but not morphology. However, EDN did not affect cell viability or morphology under the conditions chosen for this study. Other investigations, 8 ' 10 ' 18 including studies on respiratory epithelium, have documented that EDN exhibits minimal, if any, toxicity, and hence may play only a limited role in the propagation of disease. Although the mechanisms these proteins use to perturb corneal epithelial cells are unclear, the formation of pores in the target cell membrane has been explored as one of the mechanisms for toxicity. It has been shown that the cytotoxic properties of ECP are attributable to its ability to form pores in cellular membranes.' 10 By contrast, studies with MBP and synthetic lipid bilayers failed to show any evidence for penetration of the protein into the membrane bilayer or channel formation.' 11 The most recently proposed model for MBP interaction with lipid bilayers suggests binding of MBP to the surface of the bilayer by electrostatic and hydrophobic interactions, causing clustering of the surface-charged molecules, aggregation of the liposomes, and destabilization of the bilayer, leading to bilayer fusion and subsequent lysis.' 11 It is apparent from this model for MBP toxicity that electrostatic interactions between MBP and target membranes play a vital role. Hence, the canonic properties of MBP and ECP may be a contributing factor in their mechanism of toxicity. The ability of negatively charged molecules, such as heparin, to neutralize the toxic effects of MBP on human corneal epithelium further supports the role of the cationic groups of these granule proteins in corneal disease.' 12 In other experimental systems involving parasites, similar cationic molecules, namely protamine sulfate and polyarginine, have toxic effects comparable to MBP at equimolar concentrations.' 13 However, this effect was not universal. Other cations, such as polylysine, histone, lysozyme, and cytochrome C, did not induce cellular damage.' 13 In studies on corneal epithelial wound healing, MBP and ECP significantly inhibited the rate of wound closure However, the cations lysozyme and protamine did not affect wound healing rates, 31 indicating that a positive charge is only partially necessary for toxicity and that other properties of the granule proteins may be of importance. Cellular morphologic changes observed in this study after eosinophil granule protein exposure may be explained somewhat by the effects on the epithelial actin cytoskeleton. Indeed, actin filaments have been implicated in the control of cell shape, cell adhesion to substratum, and epithelial migration. 44 Previous work' 15 has demonstrated convincingly the redistribution of actin cytoskeletal patterns in migratory corneal epithelium, suggesting that it may have a role to play in the migratory process. Therefore, one could speculate that compounds that slow corneal epithelial migration, such as MBP and ECP, may do so by interfering with the actin cytoskeleton. Toxicity by local anesthetics disrupts cytoskeletal organization in corneal epithelial and BALB/3T3 cells and produces changes in cellular morphology similar to those seen in the current study. 30 ' 16 ' 47 Moreover, local anesthetics seem to have different effects on the cells, depending on concentration. Isolated morphologic changes occur at low concentrations, and decreased viability occurs only at higher concentrations, 36 similar to what was seen with MBP, ECP, and EPO in the current study. The current findings suggest that the eosinophil cationic proteins MBP, ECP, EPO, and EDN differentially affect human corneal epithelial cell viability and morphology. For example, lower concentrations of MBP and ECP influenced cell morphology but not viability, whereas higher concentrations affected both. ECP appeared to produce a greater loss of cell adhesion than the other tested cationic proteins. All tested concentrations of EPO generated loss of cell viability but minimal morphologic changes. These differential effects of the eosinophil cationic proteins were seen with other mammalian cell types, including guinea pig tracheal epithelium 12 and rat type II pneumocytes. 48 Similar observations have been documented in parasite models. 17 ' 18. In summary, MBP, ECP, and EPO may exert toxic effects on corneal epithelium and may contribute to the epithelial keratopathy seen in severe ocular allergy. Key Words eosinophil cationic protein, eosinophil major basic protein, eosinophil peroxidase, eosinophil-derived neurotoxin, vernal corneal ulcerations Acknowledgments The authors thank Dr. Judith Yannariello-Brown and Mrs. Cheryl Adolphson for their critical review of the manuscript. References 1. Trocme SD, Aldave AJ. The eye and the eosinophil. Suru Ophthalmol. 1994; 39: Trocme SD, Raizman MB, Bartley GB. Medical therapy for ocular allergy. Mayo Clin Proc. 1992; 67: Gleich GJ, Frigas E, Loegering DA, et al. Cytotoxic properties of the eosinophil major basic protein. / Immunol. 1979; 123: Gleich GJ, Kita H, Adolphson CR. Eosinophils. In: Frank MM, Austen KF, Claman HN, Unanue ER, eds. Senator's Immunologic Diseases. 5th ed. Boston: Little, Brown; 1995: Thomas LL, Zheutlin LM, Gleich GJ. Pharmacological control of human basophil histamine release stimu-

6 598 Investigative Ophthalmology & Visual Science, March 1997, Vol. 38, No. 3 lated by eosinophil granule major basic protein. Immunology. 1989; 66: Moy JN, Gleich GJ, Thomas LL. Noncytotoxic activation of neutrophils by eosinophil granule major basic protein. Effect on superoxide anion generation and lysosomal enzyme release. /Immunol. 1990; 145: Zheutlin LM, Ackerman SJ, Gleich GJ, Thomas LL. Stimulation of basophil and rat mast cell histamine release by eosinophil granule-derived cationic proteins. J Immunol. 1984; 133: Motojima S, Frigas E, Loegering DA, Gleich GJ. Toxicity of eosinophil cationic proteins for guinea pig tracheal epithelium in vitro. Am Rev Respir Dis. 1989; 139: Tai PC, Ackerman SJ, Spry CJ, Dunnette S, Olsen EG, Gleich GJ. Deposits of eosinophil granule proteins in cardiac tissues of patients with eosinophilic endomyocardial disease. Lancet. 1987; 1: Ackerman SJ, Gleich GJ, Loegering DA, Richardson BA, Butterworth AE. Comparative toxicity of purified human eosinophil granule cationic proteins for schistosomula of Schistosoma mansoni. AmJ Trop Med Hyg. 1985; 34: Hamann KJ, Gleich GJ, Checkel JL. In vitro killing of microfilariae of Brugia pahangi and Brugia malayi by eosinophil granule proteins. J Immunol. 1990; 144: Jong EC, Klebanoff SJ. Eosinophil-mediated mammalian tumor cell cytotoxicity: Role of the peroxidase system. / Immunol 1980; 124: Jong EC, Henderson WR, Klebanoff SJ. Bactericidal activity of eosinophil peroxidase. / Immunol. 1980; 124: Henderson WR, Chi EY, Klebanoff SJ. Eosinophil peroxidase-induced mast cell secretion. J Exp Med. 1980; 152: Zabucchi G, Menegazzi R, Cramer R, et al. Mutual influence between eosinophil peroxidase (EPO) and neutrophils: Neutrophils reversibly inhibit EPO enzymatic activity and EPO increases neutrophil adhesiveness. Immunology. 1990; 69: Nakajima H, Loegering DA and Gleich GJ. Cytotoxicity of eosinophil granule proteins for tumor cells. FA- SEBJ. 1988;2:A Hamann KJ, Barker RL, Checkel JL, McCall JW, Gleich GJ. Eosinophil peroxidase-mediated toxicity to microfilariae of Brugia pahangi and Brugia malayi. Fed Proc. 1987; 46: Hamann KJ, Barker RL, Loegering DA, Gleich GJ. Comparative toxicity of purified human eosinophil granule proteins for newborn larvae of Trichinella spiralis. J Parasitol. 1987; 73: McLaren DJ, Peterson CG, Venge P. Schistosoma mansoni: Further studies of the interaction between schistosomula and granulocyte-derived cationic proteins in vitro. Parasitology. 1984;88: Durack DT, Ackerman SJ, Loegering DA, Gleich GJ. Purification of human eosinophil-derived neurotoxin. Proc Natl Acad Sci USA. 1981;78: Fredens K, Dahl R, Venge P. The Gordon phenomenon induced by the eosinophil cationic protein and eosinophil protein X. J Allergy Clin Immunol. 1982; 70: Gullberg U, Widegren B, Arnason U, Egesten A, Olsson I. The cytotoxic eosinophil cationic protein (ECP) has ribonuclease activity. Biochem Biophys Res Commun. 1986; 139: Slifman NR, Loegering DA, McKean DJ, Gleich GJ. Ribonuclease activity associated with human eosinophil-derived neurotoxin and eosinophil cationic protein. J Immunol. 1986; 137: Filley WV, Kephart GM, Holley KE, Gleich GJ. Identification by immunofluorescence of eosinophil granule major basic protein in lung tissues of patients with bronchial asthma. Lancet. 1982; 2: Gleich GJ. The late phase of the immunoglobulin E- mediated reaction: A link between anaphylaxis and common allergic disease? / Allergy Clin Immunol. 1982;70: Venge P, Dahl R, Fredens K, Peterson CG. Epithelial injury by human eosinophils. Am Rev Respir Dis. 1988;138(6Pt2):S Trocme SD, Kephart GM, Allansmith MR, Bourne WM, Gleich GJ. Conjunctival deposition of eosinophil granule major basic protein in vernal keratoconjunctivitis and contact lens-associated giant papillary conjunctivitis. Am J Ophthalmol 1989; 108: Udell IJ, Gleich GJ, Allansmith MR, Ackerman SJ, Abelson MB. Eosinophil granule major basic protein and Charcot-Leyden crystal protein in human tears. AmJ Ophthalmol. 1981; 92: Saiga T, Shimizu Y. Activated eosinophils in vernal keratoconjunctivitis. Nippon Rinsho-Jap J Clin Med. 1993;51: Trocme SD, Kephart GM, Bourne WM, Buckley RJ, Gleich GJ. Eosinophil granule major basic protein deposition in corneal ulcers associated with vernal keratoconjunctivitis. Am J Ophthalmol. 1993; 115: Trocme SD, Gleich GJ, Kephart GM, Zieske JD. Eosinophil granule major basic protein inhibition of corneal epithelial wound healing. Invest Ophthalmol Vis Sci. 1994; 35: Mack KA, Hallberg CK, Gleich GJ, Trocme SD. Eosinophil cationic protein inhibits corneal epithelial migration in the rat. ARVO Abstracts. Invest Ophthalmol Vis Sci. 1994; 35: Pineda AA, Brzica SM Jr., Taswell HF. Continuousand semicontinuous-flow blood centrifugation systems: Therapeutic applications with plasma-, platelet-, lympha-, and eosinapheresis. Transfusion. 1977; 17: Gleich GJ, Loegering DA, Mann KG, Maldonado JE. Comparitive properties of the Charcot-Leyden crystal protein and the major basic protein from human eosinophils. / Clin Invest. 1976; 57: Ten RM, Pease LR, McKean DJ, Bell MP, Gleich GJ. Molecular cloning of the human eosinophil peroxidase./ */> Med. 1989; 169: Grant RL, Acosta D. Comparative toxicity of tetracaine, proparacaine and cocaine evaluated with pri-

7 Eosinophil Protein Toxicity on Corneal Epithelium 599 mary cultures of rabbit corneal epithelial cells. ExpEye Res. 1994; 58: Acosta D, Chappell R. Cardiotoxicity of diazepam in cultured heart cells. Toxicology. 1977;8: Kotnik V, Fleischmann WR Jr. A simple and rapid method to determine hematopoietic growth factor activity. J Immunol Methods. 1990; 129: Tada H, Shiho O, Kuroshima K, Koyama M, Tsukamoto K. An improved colorimetric assay for interleukin 2. j Immunol Methods. 1986;93: Young JD, Peterson CG, Venge P, Cohn ZA. Mechanism of membrane damage mediated by human eosinophil cationic protein. Nature. 1986;321: Abu-Ghazaleh RI, Gleich GJ, Prendergast FG. Interaction of eosinophil granule major basic protein with synthetic lipid bilayers: a mechanism for toxicity. / MembrBiol. 1992; 128: Trocme SD, Hallberg C. Heparin prevents eosinophil granule major basic protein inhibition of corneal epithelial wound healing in the rat. ARVO Abstracts. Invest Ophthalmol Vis Sci. 1993; 34: Butterworth AE, Wassom DL, Gleich GJ, Loegering DA, David JR. Damage to schistosomula of Schistosoma mansoni induced directly by eosinophil major basic protein. /Immunol. 1979; 122: Marinovich M. The role of actin in the transduction of toxic effect. Pharmacol Res. 1991; 24: Anderson RA. Actin filaments in normal and migrating corneal epithelial cells. Invest Ophthalmol Vis Sci. 1977;16: Dass BA, Soong HK, Lee B. Effects of proparacaine on actin cytoskeleton of corneal epithelium./ Ocular Phamiacol Ther. 1988;4: Nicolson GL, Smith JR, Poste G. Effects of local anesthetics on cell morpholog)' and membrane-associated cytoskeletal organization in BALB/3T3 cells. / Cell Biol. 1976; 68: Ayars GH, Altman LC, Gleich GJ, Loegering DA, Baker CB. Eosinophil- and eosinophil granule-mediated pneumocyte injury. / Allergy Clin Immunol. 1985; 76: