Extracellular Matrix Changes in Human Corneas After Radial Keratotomy

Exp. Eye Res. (1998), 67, 265–272 Article Number : ey980511

Extracellular Matrix Changes in Human Corneas After Radial Keratotomy* A L E X A N D E R V. L J U B I M O Va, †, S A U L A. A L B Aa, R O B E R T E. B U R G E S O Nb, Y O S H I F U M I N I N O M I Y Ac, Y O S H I K A Z U S A D Od, T U N G-T I E N S U Ne, A N T H O N Y B. N E S B U RNa, M. C R I S T I N A K E N N E Ya    E Z R A M A G U ENa a

Ophthalmology Research Laboratories, Burns & Allen Research Institute, Cedars-Sinai Medical Center, and Jules Stein Eye Institute, UCLA School of Medicine, Los Angeles, CA, U.S.A., b Massachusetts General Hospital East/Harvard Medical School, Charlestown, MA, U.S.A., c Okayama University Medical School, and d Shigei Medical Research Institute, Okayama, Japan, e Epithelial Biology Unit, Depts. of Dermatology, Pharmacology and Urology, New York University Medical School, New York, NY, U.S.A. (Received Cleveland 14 January 1998 and accepted in revised form 26 March 1998) Extracellular matrix and basement membrane alterations were identified in human corneas after radial keratotomy. Ten normal and five radial keratotomy autopsy corneas (two at 6 months post surgery, and three at 3 years post surgery) were studied by immunofluorescence with antibodies to 28 extracellular matrix and basement membrane components. Outside of radial keratotomy scars, all studied components had a normal distribution. Of stromal extracellular matrix, only type III collagen accumulated around the scars. The basement membrane around epithelial plugs had a normal composition except for type IV collagen. Its α1-α2 chains, normally present only in the limbal basement membrane, appeared around all plugs. α3 and α4 chains were very weak or absent in these areas, contrary to nonscarred areas. This basement membrane pattern was similar to the normal limbal but not to the central corneal pattern. Keratin 3 also had a limbal-like, suprabasal expression in the plug epithelium. The stroma around the scars accumulated tenascin-C, fibrillin-1, types VIII and XIV collagen, all of which were absent from normal corneal basement membrane and extracellular matrix. Only tenascin-C showed less staining in anterior scars 3 years post surgery than 6 months post surgery, but still persisted in posterior scars. Incomplete scar healing was evident even 3 years post radial keratotomy. It was manifested by the accumulation of abnormal extracellular matrix in the anterior and posterior scars and by the limbal-like pattern of type IV collagen isoforms in the basement membrane around epithelial plugs. # 1998 Academic Press Key words : human cornea ; extracellular matrix ; basement membrane ; radial keratotomy ; refractive surgery ; type IV collagen chains ; tenascin-C ; fibrillin-1, immunofluorescence.

1. Introduction Refractive surgery has been used over several decades to correct mild to moderate myopia, hyperopia, presbyopia and astigmatism (Binder, 1994 ; Parmley et al., 1995). A common refractive surgical procedure, radial keratotomy (RK), has shown success in correcting these visual defects (Waring et al., 1991 ; Jester et al., 1992). RK usually consists of performing 4 to 8 radial incisions on the cornea [sometimes, 2 incisions (Suarez et al., 1997)] at close to maximum depth with a resultant central corneal flattening. Infrequently, RK may result in over- and undercorrection, glare, fluctuation of vision, epithelial erosion, map-dot-fingerprint epithelial dystrophy, epithelial edema, ectasia and delayed bacterial keratitis (Nelson et al., 1985 ; Jester et al., 1992 ; Wellish et al., * Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology (ARVO), Fort Lauderdale, FL, U.S.A., May 1997. † Address all correspondence to : Alexander Ljubimov, Ophthalmology Research Laboratories, Burns & Allen Research Institute, Cedars-Sinai Medical Center, Davis-5069, 8700 Beverly Boulevard, Los Angeles, CA 90048, U.S.A.

0014–4835}98}090265­08 $30.00}0

1994 ; Parmley et al., 1995 ; Venter, 1997). These complications may be due to a delayed and incomplete wound healing, although molecular mechanisms of their development are not fully understood (Binder et al., 1987 ; Glasgow et al., 1988 ; Jester et al., 1992 ; Assil and Quantock, 1993 ; Binder, 1994 ; Melles, Binder and Anderson, 1994). Wound healing at sites of RK incisions is an important determinant of the outcome of refractive surgery and was an object of studies in animal models and human tissue. Modulated keratocytes (myofibroblasts) may play a major role in the development and subsequent contraction of scars at sites of RK incisions (Jester, Rodrigues and Herman, 1987 ; Garana et al., 1992 ; Melles et al., 1995a). This process involves an extensive remodeling of corneal extracellular matrix (ECM) and epithelial basement membrane (BM) that is apparently due to the activity of myofibroblasts. Transient changes of several ECM components and of β integrin (epithelial adhesion complex component) % during corneal wound healing have been documented (Malley et al., 1990 ; SundarRaj et al., 1990 ; Kahle et al., 1991 ; Rowe, Tuft and Meek, 1992 ; Stock et al., 1992 ; van Setten et al., 1992 ; Molander et al., 1993 ; # 1998 Academic Press

266

Balestrazzi et al., 1995 ; Latvala et al., 1995a ; Latvala, Tervo and Tervo, 1995b ; Melles et al., 1995b ; Anderson et al., 1996), mostly after photorefractive keratectomy (PRK). This procedure involves mechanical removal of corneal epithelium, and shaving off Bowman’s layer and anterior part of the corneal stroma with an ultraviolet 193 nm excimer laser. Similar to RK, PRK results in central corneal flattening. Normal stromal ECM composition and epithelial BM continuity are generally restored within about 2 years after PRK. In RK corneas, however, wound healing can be incomplete for up to 79 months (Binder et al., 1987 ; Glasgow et al., 1988). This finding emphasizes the importance of studies of RK corneas several years following refractive surgery. Such studies have been, however, scarce. In this report, significant abnormalities are described in the organization of stromal ECM and corneal epithelial BM of human corneas at various time points following RK, ranging from 0±5 to 3 years. 2. Materials Autopsy corneas from three patients with uneventful RK (1 from a 34 year old male, and 4 from two 50 year old females, Fig. 1) and 10 age-matched

A. V. L J U B I M O V E T A L.

individuals without ocular history were received within 30 hr after death from the National Disease Research Interchange (Philadelphia, PA, U.S.A.), in chilled Optisol. Well characterized antibodies were used to α1-α6 chains of type IV collagen, to α1, α2, α5, β1, β2, β3 and γ1 chains of laminin, to entactin}nidogen, to fibrillin-1, tenascin-C, cellular and total fibronectin, to types I, III, V, VI, VII, VIII, XII and XIV collagen, to perlecan, decorin and bamacan core proteins, to von Willebrand factor, to keratin 3 and α-smooth muscle actin (Ljubimov et al., 1995 ; 1996 ; Kenney et al., 1997 ; Maguen et al., 1997 ; Tiger et al., 1997). Antibodies to α (clone NKI-GoH3) ' an β (clone 3E1) integrin subunits, and cross-species % adsorbed fluorescein- and rhodamine-conjugated secondary antibodies were from Chemicon International (Temecula, CA, U.S.A.).

3. Methods All corneas were embedded in OCT compound (Miles, Elkhart, IN, U.S.A.), then 6 µm cryostat sections were prepared and stained by indirect immunofluorescence for ECM, cytoskeletal components, and integrins, as previously described

F. 1. A schematic of RK corneas under study showing the patterns of incisions. The corneas at 6 months post RK have 8 (OD) or 4 (OS) radial incisions. One cornea at 3 years post RK (left, lower row) has 8 radial incisions and 2 T-cuts that do not cross radial incisions. Two other corneas at 3 years post RK (center and right, lower row) have 7 radial incisions and 1 non-crossing T-cut. OD, right eye ; OS, left eye.

EXTRACELLULAR MATRIX IN RADIAL KERATOTOMY

(Ljubimov et al., 1995). To fully reveal type IV collagen epitopes, sections were air-dried for 5 min. and denatured in 6  urea at pH 3±5 for 1 hr at 4°C before staining (Ljubimov et al., 1995). Routine specificity controls (without primary or secondary antibodies) were negative. Sections of normal and RK corneas were exposed to the same dilutions of antibodies simultaneously. Hybridoma supernatants were used undiluted, and polyclonal antibodies were used to 20 µg ml−".

4. Results The distribution of all studied ECM and BM components in normal adult human corneas has been described previously (Ljubimov et al., 1995 ; 1996 ; 1998 ; Kenney et al., 1997 ; Maguen et al., 1997). In RK corneas outside of the scar areas, these components were also distributed in a normal way. Differences from the normal patterns were observed only at sites of RK incisions. Most of these incisions had epithelial plugs present. Generally, ECM and BM patterns in RK corneas were similar at both 6 months and 3 years post RK. Neovascularization (by von Willebrand factor staining) or myofibroblast involvement (by α-smooth muscle actin staining) were not seen in any of the RK corneas (not shown here). The patterns of RK incisions in the studied corneas are shown in Fig. 1. Staining patterns did not show any differences which could be attributed to different number or geometry of the incisions. In the corneal epithelial BM, many components displayed patchy}discontinuous distribution around the plugs at 6 months post RK [see also (Winter et al., 1997)], but at 3 years post RK their distribution was usually linear as in normal corneas. Most of the

267

epithelial BM components were distributed in a normal way around epithelial plugs (Fig. 2). Also, the hemidesmosomal component, α β integrin, had a ' % normal distribution pattern around the scars and epithelial plugs (not shown). However, striking abnormalities in the distribution of specific chains of type IV collagen were observed in these areas. α3 and α4 chains that in normal corneas and in nonscarred regions of RK corneas stained very strongly, were absent (6 months post RK) or very weak (3 years post RK) around all plugs [Fig. 3(A), (B)]. In contrast, α1α2 chains of type IV collagen that were normally present only in the limbal BM, appeared around the plugs [Fig. 3(C), (D)]. α5 and α6 chains of type IV collagen retained normal distribution around epithelial plugs (Fig. 2). The pattern of type IV collagen chain distribution around the plugs was similar to that seen in the normal limbus but not to the central corneal pattern. When RK corneas were stained for keratin 3, mostly suprabasal cells of the epithelial plugs were positive at 6 months post RK [Fig. 4(A)], again as in normal limbus (Schermer, Galvin and Sun, 1986 ; Kolega, Manabe and Sun, 1989). Even at 3 years post RK, some basal epithelial cells of the plugs were barely positive for keratin 3 [Fig. 4(B) open arrow]. The stroma around and beneath the RK scars and epithelial plugs had increased deposition of fibronectin and type III collagen (not shown). Abnormal deposits of tenascin-C, fibrillin-1, types VIII (not shown) and XIV collagen (Fig. 5) that were absent from normal corneal epithelial BM and stromal ECM were also found. Of these abnormally accumulated components, only tenascin-C showed less staining in anterior scars 3 years post RK than 6 months post RK. However, it persisted in posterior scars even 3 years post RK [Fig. 5(F)].

F. 2. Normal linear distribution of type IV collagen α6 chain around epithelial plugs at sites of RK incisions both at 6 months (A) and 3 years (B) post RK. e, epithelium ; s, stroma. Bar ¯ 40 µm.

268

A. V. L J U B I M O V E T A L.

F. 3. Altered distribution of other type IV collagen chains around epithelial plugs at sites of RK incisions. α3(IV) chain is absent from the plug’s epithelial BM at 6 months (A) and is very weak to absent at 3 years post RK (B). α1(IV) chain, in contrast, appears in the same locations at both time points (C), (D). e, epithelium ; s, stroma. Bar ¯ 40 µm.

5. Discussion The healing of nonsutured corneal wounds (such as caused by RK) involves extensive remodeling of the corneal ECM and BM. First, a provisional ECM is formed and it is then gradually replaced by permanent ECM of a healed scar (Garana et al., 1992 ; Jester et al., 1992 ; Stock et al., 1992 ; Assil and Quantock, 1993 ; Melles et al., 1995b). ECM and BM components including fibronectin, tenascin-C, laminin, types III, IV, VII collagen, hyaluronan and stromal proteoglycans, undergo transient changes in distribution during wound healing. In the first week post wounding, abnormal deposits of type III collagen, proteoglycans, fibronectin, tenascin-C and laminin appear in the corneal stroma at the wound sites. The staining for epithelial BM components becomes discontinuous. The abnormal ECM deposition in the

stroma continues for about a month, and then the deposits begin to disappear (SundarRaj et al., 1990 ; Kahle et al., 1991 ; Rowe et al., 1992 ; van Setten et al., 1992 ; Latvala et al., 1995a ; Melles et al., 1995b ; Anderson et al., 1996). By 12 months post refractive surgery, normal ECM and BM pattern is usually restored (SundarRaj et al., 1990 ; Balestrazzi et al., 1995 ; Latvala et al., 1995a ; 1995b ; Anderson et al., 1996). Type III collagen and fibronectin, however, can still be found in abnormal stromal deposits 16 months postoperatively (Anderson et al., 1996). Human corneas were shown to restore normal ECM patterns more slowly than monkey or rabbit corneas (van Setten et al., 1992 ; Melles et al., 1994). Some data concerning type IV collagen (Balestrazzi et al., 1995 ; Anderson et al., 1996) should be interpreted with caution because of the use of polyclonal antibodies that may react with several different isoforms.

EXTRACELLULAR MATRIX IN RADIAL KERATOTOMY

269

F. 4. Staining of epithelial plugs at sites of RK incisions for keratin 3 (K3). A, 6 months post RK ; arrows show the confines of the plug. Basal epithelial cells in the plug are keratin 3-negative. B, 3 years post RK ; some basal cells are keratin 3-positive, whereas some are still negative (open arrow). e, epithelium ; s, stroma. Bar ¯ 40 µm.

F. 5. Deposition of ECM proteins around epithelial plugs at sites of RK incisions. Staining for fibrillin-1 (FIB) is the same at 6 months (A) and 3 years (B) post RK. The same is true for type XIV collagen (C XIV) depicted in (C) (6 months) and (D) (3 years). The staining for tenascin-C (TN) is strong at 6 months post RK (E) and becomes weaker at 3 years (F). The figures D and F are composites showing most of the scar depth. e, epithelium ; s, stroma. Bar ¯ 40 µm.

270

Most of the previous data on corneal healing after refractive surgery concern excimer laser PRK, and much less information is available on corneal changes after RK or hexagonal keratotomy. Recently, longterm ECM changes in a cornea after combined keratomileusis and hexagonal keratotomy have been described (Maguen et al., 1997). The present study provides the first detailed account of long-term human corneal ECM and BM alterations following uncomplicated RK performed 6 months to three years before analysis. The major findings were the following : (1) Outside the RK scars, there were no significant ECM or BM changes. (2) In the corneal epithelial BM around the RK epithelial plugs, a shift of type IV collagen isoforms was observed. The central corneal pattern of these isoforms (Ljubimov et al., 1995 ; 1996) (α3-α4-α5-α6 chains) was shifted to the limbal pattern (α1-α2-α5-α6 chains). Concurrently, the plug epithelium overlying this limbal-like BM also had a predominantly limballike suprabasal expression of cornea-specific keratin 3. (3) The stroma around and along the scars and epithelial plugs contained local deposits of types VIII and XIV collagen, tenascin-C and fibrillin-1 that were absent from normal corneal stroma (Ljubimov et al., 1995 ; 1996 ; Wheatley et al., 1995 ; Kenney et al., 1997 ; Maguen et al., 1997). With increasing time post RK, only tenascin-C immunoreactivity diminished. The shift of type IV collagen in the BM around the epithelial plugs was also found in the hexagonal keratotomy scars (Maguen et al., 1997), and therefore, may be a common phenomenon accompanying healing of unsutured incisional wounds in human cornea. The persistent expression of limbal type IV collagen isoforms in the plug BM may play a role in the incomplete differentiation of corneal epithelium in the plugs evidenced by a limbal-like pattern of keratin 3 expression. In central cornea, keratin 3 is abundant in all epithelial layers, whereas in the limbus it is mostly present in suprabasal cells (Schermer et al., 1986 ; Kolega et al., 1989 ; Gipson, 1992). We have also shown earlier that type IV collagen α1-α2 chains were expressed in the limbus, and α3-α4 chains, in the central cornea (Ljubimov et al., 1995 ; 1996). It can be suggested that α1-α2 type IV collagen may inhibit keratin 3 expression in the basal limbal epithelial cells that are in direct contact with the BM. The present data lend strong support to the notion that in RK plugs, the same inhibition may occur. Although the plug epithelium assumes a limbal keratin 3 pattern, it seems unlikely that this central cornea-derived epithelium can return to a limbal stem cell status. Rather, the suprabasal keratin 3 staining is only a phenotypic trait related to an epithelial differentiation stage or mode (Jester, Rodrigues and Sun, 1985 ; SundarRaj et al., 1992). One can also suggest that such a phenotype would be a marker of incomplete wound healing at sites of RK incisions. Alternatively, it could result from

A. V. L J U B I M O V E T A L.

lack of Bowman’s layer and direct contact of plug epithelium with corneal stroma, similar to the situation in the limbus. Another indication of incomplete healing of RK incisions may be the persistent abnormal ECM deposits around these incisions. Fibronectin, type III collagen and tenascin-C are typically found in scars. They may be deposited by myofibroblasts, and are gradually removed from scar areas with time (Balestrazzi et al., 1995 ; Latvala et al., 1995a). Fibrillin-1 and type XIV collagen deposition may be part of fibrotic process leading to the scar formation, since they were also seen in fibrosed areas of bullous keratopathy (Ljubimov et al., 1996) and keratoconus (Zhou et al., 1996 ; Kenney et al., 1997 ; Tuori et al., 1997) corneas. These results on RK scars are very similar to those recently reported on a cornea that underwent a similar procedure, hexagonal keratotomy (Maguen et al., 1997). They show that even after 3 years following RK, the BM around the scars and epithelial plugs failed to deposit specific chains of type IV collagen and basal epithelial cells did not stain well for keratin 3. The data are consistent with the limbal-like phenotype of epithelial cells in the plugs. The permanent absence of Bowman’s layer in these regions could hamper the normal differentiation process of corneal epithelium and thus could contribute to such a considerable delay in wound healing. Failure to complete wound healing may also be reflected by persistent abnormal deposits of several ECM components in the stroma around the RK scars.

Acknowledgements We thank Drs E. Engvall (The Burnham Institute, La Jolla, CA, U.S.A.), J. R. Couchman (University of Alabama, Birmingham, AL, U.S.A.) and D. Gullberg (University of Uppsala, Uppsala, Sweden) for providing antibodies. Antibodies to laminin β2 chain produced by Dr J. Sanes and to type IV collagen α1-α2 chains produced by Dr H. Furthmayr were obtained from the Developmental Studies Hybridoma Bank, Department of Biology, University of Iowa (Iowa City, IA, U.S.A.), under contract N01-HD-3144 from the NICHD. Supported by a grant from the Discovery Fund for Eye Research (EM), and by the Skirball Program in Molecular Ophthalmology.

References Anderson, J. A., Binder, P. S., Rock, M. E. and Vrabec, M. P. (1996). Human excimer laser keratectomy. Immunohistochemical analysis of healing. Arch. Ophthalmol. 114, 54–60. Assil, K. K. and Quantock, A. J. (1993). Wound healing in response to keratorefractive surgery. Surv. Ophthalmol. 38, 289–302. Balestrazzi, E., De Molfetta, V., Spadea, L., Vinciguerra, P., Palmieri, G., Santeusanio, G. and Spagnoli, L. (1995). Histological, immunohistochemical, and ultrastructural findings in human corneas after photorefractive keratectomy. J. Refract. Surg. 11, 181–7.

EXTRACELLULAR MATRIX IN RADIAL KERATOTOMY

Binder, P. S. (1994). Radial keratotomy and excimer laser photorefractive keratectomy for the correction of myopia. J. Refract. Corneal Surg. 10, 443–64. Binder, P. S., Nayak, S. K., Deg, J. K., Zavala, E. Y. and Sugar, J. (1987). An ultrastructural and histochemical study of long-term wound healing after radial keratotomy. Am. J. Ophthalmol. 103, 432–40. Garana, R. M., Petroll, W. M., Chen, W. T., Herman, I. M., Barry, P., Andrews, P., Cavanagh, H. D. and Jester, J. V. (1992). Radial keratotomy. II. Role of the myofibroblast in corneal wound contraction. Invest. Ophthalmol. Vis. Sci. 33, 3271–82. Gipson, I. K. (1992). Adhesive mechanisms of the corneal epithelium. Acta Ophthalmol. (Suppl.) 202, 13–17. Glasgow, B. J., Brown, H. H., Aizuss, D. H., Mondino, B. J. and Foos, R. Y. (1988). Traumatic dehiscence of incisions seven years after radial keratotomy. Am. J. Ophthalmol. 106, 703–7. Jester, J. V., Rodrigues, M. M. and Sun, T. T. (1985). Change in epithelial keratin expression during healing of rabbit corneal wounds. Invest. Ophthalmol. Vis. Sci. 26, 828–37. Jester, J. V., Rodrigues, M. M. and Herman, I. M. (1987). Characterization of avascular corneal wound healing fibroblasts. New insights into the myofibroblast. Am. J. Pathol. 127, 140–8. Jester, J. V., Villasen4 or, R. A., Schanzlin, D. J. and Cavanagh, H. D. (1992). Variations in corneal wound healing after radial keratotomy : Possible insights into mechanisms of clinical complications and refractive effects. Cornea 11, 191–9. Kahle, G., Daqun, X., Seiler, T., Schro$ ter-Kermani, C. and Wollensak, J. (1991). Wundheilung der kornea von neuweltaffen nach fla$ chiger ketatektomie : Er : YAGexcimerlaser. Fortschr. Ophthalmol. 88, 380–5. Kenney, M. C., Nesburn, A. B., Burgeson, R. E., Butkowski, R. J. and Ljubimov, A. V. (1997). Abnormalities of the extracellular matrix in keratoconus corneas. Cornea 16, 345–51. Kolega, J., Manabe, M. and Sun, T. T. (1989). Basement membrane heterogeneity and variation in corneal epithelial differentiation. Differentiation 42, 54–63. Latvala, T., Tervo, K., Mustonen, R. and Tervo, T. (1995a). Expression of cellular fibronectin and tenascin in the rabbit cornea after excimer laser photorefractive keratectomy : A 12 month study. Br. J. Opthalmol. 79, 65–9. Latvala, T., Tervo, K. and Tervo, T. (1995b). Reassembly of the α β integrin and laminin in the rabbit corneal ' % basement membrane after excimer laser surgery. A twelve-month follow-up. CLAO J. 21, 125–9. Ljubimov, A. V., Burgeson, R. E., Butkowski, R. J., Michael, A. F., Sun, T. T. and Kenney, M. C. (1995). Human corneal basement membrane heterogeneity : Topographical differences in the expression of type IV collagen and laminin isoforms. Lab. Invest. 72, 461–73. Ljubimov, A. V., Burgeson, R. E., Butkowski, R. J., Couchman, J. R., Wu, R. R., Ninomiya, Y., Sado, Y., Maguen, E., Nesburn, A. B. and Kenney, M. C. (1996). Extracellular matrix alterations in human corneas with bullous keratopathy. Invest. Ophthalmol. Vis. Sci. 37, 997–1007. Ljubimov, A. V., Huang, Z., Huang, G. H., Burgeson, R. E., Gullberg, D., Miner, J. H., Ninomiya, Y., Sado, Y. and Kenney, M. C. (1998). Human corneal epithelial basement membrane and integrin alterations in diabetes and diabetic retinopathy. J. Histochem. Cytochem. 46, 1033–41. Maguen, E., Alba, S. A., Burgeson, R. E., Butkowski, R. J., Michael, A. F., Kenney, M. C., Nesburn, A. B. and Ljubimov, A. V. (1997). Alterations of corneal extra-

271

cellular matrix following multiple refractive procedures : A clinical and immunohistochemical study. Cornea 16, 675–82. Malley, D. S., Steinert, R. F., Puliafito, C. A. and Dobi, E. T. (1990). Immunofluorescence study of corneal wound healing after excimer laser anterior keratectomy in the monkey eye. Arch. Ophthalmol. 108, 1316–22. Melles, G. R. J., Binder, P. S. and Anderson, J. A. (1994). Variation of healing throughout the depth of long-term, unsutured, corneal wounds in human autopsy specimens and monkeys. Arch. Ophthalmol. 112, 100–9. Melles, G. R. J., Binder, P. S., Moore, M. N. and Anderson, J. A. (1995a). Epithelial-stromal interactions in human keratotomy wound healing. Arch. Ophthalmol. 113, 1124–30. Melles, G. R. J., SundarRaj, N., Binder, P. S., van der Weiden, M. M., Wijdh, R. H., Beekhuis, W. H. and Anderson, J. A. (1995b). Immunohistochemical analysis of unsutured and sutured corneal wound healing. Curr. Eye Res. 14, 809–17. Molander, N., Lindquist, U., Stenevi, U., von Mamborg, A. and Ehinger, B. (1993). Influence of radial keratotomy on endogenous hyaluronan in cornea and aqueous humour. J. Refract. Corneal Surg. 9, 358–65. Nelson, J. D., Williams, P., Lindstrom, R. L. and Doughman, D. J. (1985). Map-fingerprint-dot changes in the corneal epithelial basement membrane following radial keratotomy. Ophthalmology 92, 199–205. Parmley, V., Ng, J., Gee, B., Rotkis, W. and Mader, T. (1995). Penetrating keratopasty after radial keratotomy. A report of six patients. Ophthalmology 102, 947–50. Rowe, I. M., Tuft, S. J. and Meek, K. M. (1992). Proteoglycan and collagen morphology in superficially scarred rabbit cornea. Histochem. J. 24, 311–18. Schermer, A., Galvin, S. and Sun, T. T. (1986). Differentiation-related expression of a major 64K corneal keratin in vivo and in culture suggests limbal location of corneal epithelial stem cells. J. Cell Biol. 103, 49–62. Stock, E. L., Kurpakus, M. A., Sambol, B. and Jones, J. C. (1992). Adhesion complex formation after small keratectomy wounds in the cornea. Invest. Ophthalmol. Vis. Sci. 33, 304–13. Suarez, E. Verity, S. M., Torres, F. Y. and Assil, K. K. (1997). Two-incision radial keratotomy for low myopia with astigmatism. J. Cataract Refract. Surg. 23, 502–8. SunderRaj, N., Geiss, M. J., 3d, Fantes, F., Hanna, K., Anderson, S. C., Thompson, K. P., Thoft, R. A. and Waring, G. O., 3d. (1990). Healing of excimer laser ablated monkey corneas. An immunohistochemical evaluation. Arch. Ophthalmol. 108, 1604–10. SundarRaj, N., Rizzo, J. D., Anderson, S. C. and Gesiotto, J. P. (1992). Expression of vimentin by rabbit corneal epithelial cells during wound repair. Cell Tissue Res. 267, 347–56. Tiger, C.-F., Champliaud, M.-F., Pedros-Donello, F., Thornell, L.-E., Ekblom, P. and Gullberg, D. (1997). Presence of laminin α5 chain and lack of laminin α1 chain during human muscle development and in muscular dystrophies. J. Biol. Chem. 272, 28590–5. Tuori, A., Virtanen, I., Aine, E. and Uusitalo, H. (1997). The expression of tenascin and fibronectin in keratoconus, scarred and normal human cornea. Graefe’s Arch. Clin. Exp. Ophthalmol. 235, 222–9. van Setten, G. B., Koch, J. W., Tervo, K., Lang, G. K., Tervo, T., Naumann, G. O., Kolkmeier, J., Virtanen, I. and Tarkkanen, A. (1992). Expression of tenascin and fibronectin in the rabbit cornea after excimer laser surgery. Graefe’s Arch. Clin. Exp. Ophthalmol. 230, 178–83.

272

Venter, J. A. (1997). Photorefractive keratectomy for hyperopia after radial keratotomy. J. Refract. Surg. 13, S456. Waring, G. O., 3d, Lynn, M. J., Nizam, A., Kutner, M. H., Cowden, J. W., Culbertson, W., Laibson, P. R., McDonald, M. B., Nelson, J. D., Obstbaum, S. A., Rowsey, J. J., Salz, J. J., Bourque, L. B. and The PERK Study Group. (1991). Results of the prospective evaluation of radial keratotomy (PERK) study five years after surgery. Ophthalmology 98, 1164–76. Wellish, K. L., Glasgow, B. J., Beltran, F. and Maloney, R. K. (1994). Corneal ectasia as a complication of repeated keratotomy surgery. J. Refract. Corneal Surg. 10, 360–4.

A. V. L J U B I M O V E T A L.

Wheatley, H. M., Traboulsi, E. I., Flowers, B. E., Maumenee, I. H., Azar, D., Pyeritz, R. E. and Whittum-Hudson, J. A. (1995). Immunohistochemical localization of fibrillin in human ocular tissues. Relevance to the Marfan syndrome. Arch. Ophthalmol. 113, 103–9. Winter, M., Behrendt, S., Binder, P. S. and Duncker, G. I. W. (1997). Ultrastructural and immunohistochemical findings after linear excimer laser keratectomy. J. Refract. Surg. 13, 60–8. Zhou, L., Yue, B. Y., Twining, S. S., Sugar, J. and Feder, R. S. (1996). Expression of wound healing and stress-related proteins in keratoconus corneas. Curr. Eye Res. 15, 1124–31.