expression of laminin-5 with amniotic membrane transplantation in excimer laser ablated rat corneas

Expression of laminin-5 with amniotic membranetransplantation in excimer laser ablated rat corneasJun Haeng Lee, PhD, Hyung Keun Lee, MD, Jin Kuk Kim, MD, PhD,Jee Ho Chang, MD, Sung Eun Kim, MD, Eung Kweon Kim, MD, PhD,Gong Je Seong, MD, SoonWon Hong, MD

Purpose: To investigate the expression of laminin-5 during epithelial healing and eval-uate its expression in vivo using rat corneas on which amniotic membrane was ap-plied to cover the wound after excimer laser photoablation.

Setting: Department of Ophthalmology, Yonsei University College of Medicine,Seoul, Korea.

Methods: Myopic photorefractive keratectomy (PRK) with a 100 �m deep abla-tion was performed in Sprague Dawley rats killed 6, 12, 24, 48, 72, and 96 hoursafter the procedure. In the first group of 30 rats, the excimer laser-ablated corneawas covered with amniotic membrane after PRK. Thirty other rats in which no am-niotic membrane treatment was used served as controls. Immunohistochemicaland immunofluorescein techniques were used to monitor the expression of lami-nin-5, �2, and �1 in the rat corneas. Immunoblotting was used to compare the ex-pression of laminin between the amniotic membrane group and the control group.

Results: In the immunoblotting study, laminin-5, �3, and �2 increased 24 hoursafter amniotic membrane treatment compared to the control group. At 12 hours,in vivo immunostaining of the corneas in both groups expressed laminin, butlaminin-5 and �2 were more intensely expressed in the amniotic membranegroup. This continued until reepithelialization. Expression of the �1 chain was notdifferent between the 2 groups.

Conclusion: With the use of amniotic membrane, the expression of laminin-5 and�2 was faster and more intense than in a control group during reepithelializationof excimer laser-ablated rat corneas.

J Cataract Refract Surg 2004; 30:2192–2199 2004 ASCRS and ESCRS

Epithelial wound healing is considered to result from Laminin, a major component of the basementa highly organized series of events in which the cells membrane, is part of a family of at least 12 isoforms

initially adhere to the substrate and then proliferate and based on the variable heterotrimeric assembly of peptidemigrate to cover the wounded area.1 During corneal epithe- chains.4,5 Of these 12 isoforms, laminin-1 and laminin-5lial wound healing, the basement membrane plays an essen- are major components of the corneal basement mem-tial role in reepithelialization because it separates epithelial brane.6,7 Laminin-1 is the classic laminin and is ubiqui-cells from the underlying stroma and has many functions tously distributed in the basement membrane of allthat help maintain the normal corneal epithelium.2,3 tissue types. Laminin-5 has a more restricted distribu-

tion across tissues than laminin-1.6,8 Laminin-5 has beenreported in the corneal basement membrane and has amore important role than other basement components

Accepted for publication February 4, 2004.in cell adhesion and migration of corneal epithelial

Reprint requests to Hyung Keun Lee, MD, Institute of Vision Research, cells.9,10

Department of Ophthalmology, Yonsei University College of Medicine,Amniotic membrane transplantation is useful inSeoul, Korea, 146-92 Yongdong Severance Hospital, Dogok-dong,

Kangnam-gu, Seoul, Korea. E-mail: [email protected]. promoting epithelialization for persistent epithelial de-

2004 ASCRS and ESCRS 0886-3350/04/$–see front matterPublished by Elsevier Inc. doi:10.1016/j.jcrs.2004.02.021

LABORATORY SCIENCE: EXPRESSION OF LAMININ-5 WITH AMNIOTIC MEMBRANE TRANSPLANTATION IN RAT CORNEAS

fects and sterile ulceration when more conservative in- Rat Corneal Limbal Cell CultureRat corneal limbal epithelial cells were cultured followingterventions fail.11–13 However, the exact mechanism of

Geerling et al.’s16 method with modifications. To obtainamniotic membrane in promoting epithelialization hasprimary rat corneal epithelial cells, the cornea was removednot been revealed, and the underlying mechanism offrom the eye with sharp scissors under the microscope. Afterepithelial reconstruction by amniotic membrane is anincubation at 37oC for 1 hour with 1.2 IU/mL neutral prote-

active area of research. Recently, Kurpakus-Wheater14ase (Dispase II, Boehringer-Mannheim), the epithelium was

stated that laminin-5 is a component of preserved amni- stripped by gently scraping it from the limbus to the centerinto PBS. The sample was centrifuged at 100g for 5 minutes,otic membrane and hypothesized that amniotic membraneand the cells were suspended in keratinocyte-serum-free me-facilitates corneal resurfacing, promoting epithelial celldium (K-SFM, Gibco) supplemented with 100 IU/mL peni-adhesion via laminin-5.14 To our knowledge, no studycillin, 100 �g/mL streptomycin, 5 ng/mL epidermal growthhas evaluated the expression of laminin-5 during cornealfactor (EGF), 2.5 mg/mL bovine pituitary extract (Gibco),

epithelial healing in vivo. and 0.03 mM calcium chloride. The cells were then culturedIn this study, we investigated the pattern of at 37oC with 5% carbon dioxide in 95% humidified air

laminin-5 expression during epithelial regeneration after until 80% confluent and expanded using routine cell culturetechniques. From the second passage, corneal epithelial cellsphotorefractive keratectomy (PRK) and compared lami-were cultured using a 24-well transwell chamber (Corningnin expression between the eyes treated with amnioticInc.). The upper and lower chambers of the transwell weremembrane eye and a control group in vivo. Using im-separated by an 8.0 �m pore filter.

munoblotting, we also evaluated whether the expression Corneal epithelial cells were seeded with a density ofpattern of the �3 and �2 chains of laminin-5 change 2.0 � 105 cell on the upper wells of the chamber. After cellswhen amniotic membrane is used in cultured rat corneal were 50% to 60% confluent on the dish, the 1 cm � 1 cm

amniotic membrane was inserted into the bottom chamberepithelial cells.for 24 hours in 3 of the 6 wells in the experimental group.

Materials and Methods Sodium Dodecyl Sulfate–Polyacrylamide GelElectrophoresis and Western BlottingExperimental Animals

After 24 hours, the amniotic membrane insertion andThe principles of Proper Laboratory Animal Care (NIHcultured corneal epithelial sheets were washed with PBS 3publication No. 85-23, revised 1985) were followed. Sixtytimes. With the cell lifter (Costar, Corning Inc.), the epithelialSprague Dawley rats, 6 to 8 weeks of age (200 g each), werecells were collected into a 1.5 mL microtube and homogenizedused in this study. They were housed in a 12-hour-light/12-in 2� sample buffer (Bio-Rad) with 5% �-mercaptoethanolhour-dark cycle and fed ad libitum. All animals were treated(reducing sample buffer). Samples were heated to 95oC foraccording to the Association for Research in Vision and5 minutes, separated by 6% gradient sodium dodecyl sulfate–Ophthalmology Statement for Use of Animals in Ophthalmicpolyacrylamide gel electrophoresis, and transferred onto Hy-and Vision Research.bond-C nitrocellulose membrane (Amersham PharmaciaBiotech). Unbound membrane was blocked with 5% nonfatAmniotic Membrane Preparationmilk solution and incubated in the primary antibodies (lami-In accordance with the tenets of the Declaration ofnin �3, 1:1000 dilution, H 187; laminin �2, 1:1000 dilution,Helsinki and proper informed consent, human amnioticClone C20, polyclonal) for 1 hour at room temperature.membrane was obtained at the time of Caesarean section.After extensive washing, blots were incubated for 1 hour inThe amniotic membranes were collected and kept accordinghorseradish peroxidase–conjugated secondary antibodies, andto the method of Kim and Tseng15 with modifications. Thebound protein was detected using enhanced chemilumines-amniotic membrane was separated from the chorion of thecence reagents (Amersham Pharmacia Biotech).placenta and obtained immediately after Caesarean delivery.

Under a lamellar flow hood, the amniotic membrane wasLaser Surgerythoroughly washed with phosphate-buffered solution (PBS)

A 193 nm excimer laser (Keratome II, Schwind) with a(pH 7.2) containing gentamicin 8 �g/mL and cefaxolinefluence of 160 mJ/cm2 and a pulse rate of 5 Hz was used to4 �g/mL. It was then attached to a nitrocellulose membraneperform myopic PRK on the rat corneas. The ablation hadby placing the epithelial side on the top. The amniotic mem-a central diameter of 3.0 mm and was 100 �m deep. Thebrane was stored at �20oC in Dulbecco’s modified Eaglemean diameter of the normal rat corneas was 6.5 mm andmedium containing glycerin (Life Technology Inc.) in a ratiothe mean cornea stromal thickness, approximately 200 �m.of 1:1. Immediately before its use, the amniotic membrane

was thawed and washed 3 times with sterile PBS. In a group of 30 rats, the excimer laser-ablated cornea was

J CATARACT REFRACT SURG—VOL 30, OCTOBER 2004 2193


covered with amniotic membrane after PRK. Cryopreservedamniotic membrane was washed with PBS 3 times beforeuse and cut into 8.0 mm � 8.0 mm square pieces. Amnioticmembranes were sutured to bulbar conjunctiva with 10-0nylon. Thirty other rats in which no amniotic membranewas used served as controls. Tarsorrhaphy was performed inall eyes using a 7-0 polyglactin (Vicryl�) suture.

Postoperative follow-up indicated that the rats were notin apparent distress and the corneas were clear by gross eyeobservation. The rats were killed 6, 12, 18, 24, 48, and 96hours after PRK.

Preparation of Tissue Section for Immunostaining Figure 1. (Lee) Western-blot analysis comparing the expressionA minimum of 4 animals from each group were killed of laminin-5 �3 and �2 chains in cultivating rat corneal cells. Top:

Processed (160 kDa) and unprocessed (190 kDa) form of the �3at each follow-up time. The eyes were harvested and preparedchain was found in both samples. However, the cells cultured withfor histology and immunohistochemistry. The enucleatedamniotic membrane show a more intense band than the control cells.eyes were embedded in OCT compound (Miles), snap frozenBottom: The unprocessed (155 kDa) form of the laminin �2 chain isin a methylbutane–dry ice slurry, and stored at �70oC untilpresent. A more intense band was found in the amniotic membrane

sectioning. Frozen sections (8 �m thick) through the papil-epithelial cell culture.

lary–optic nerve planes were placed on gelatinized slides.

ResultsImmunohistochemical StainingImmunohistochemical staining was performed using both Western blot analysis of a corneal epithelial cell

avidin-biotin-peroxidase complex techniques. Antibodies sample determined that the �3 and �2 chains of lami-against laminin-5 (1:1000 dilution, Clone P3H9-2, mono- nin-5 were present in the cells (Figure 1). From theclonal) and laminin �2 (1:1000 dilution, Clone C20, poly- antibodies to �3 chains, the unprocessed (190 kDa)clonal) were used. Laminin �1 antibody (1:500 dilution,

and processed forms of this protein and the intensityClone D18, monoclonal, NeoMarkers) was also used. The

of the bands showed that the epithelial cells culturedappropriate dilutions for the primary antibodies were titratedwith amniotic membrane induced more laminin �3.in a pilot study. The duration of color development wasThe band intensity of the processed and unprocessedthe same in all staining procedures. For the avidin-biotin-

peroxidase complex technique, the frozen sections were fixed form appeared to be equal in the control and amnioticin cold acetone for 10 minutes and rinsed in PBS (pH 7.4). membrane samples. Like the �3 chain, the laminin �2The sections were treated with 3% hydrogen peroxide in PBS chain also showed more intense band activity in thefor 15 minutes to block the endogenous peroxidase activity and amniotic membrane sample than in the control sample.then rinsed in PBS. They were then incubated with normal

With immunostaining of rat cornea, antibodies forgoat serum for 20 minutes at room temperature to avoid

laminin-5, �1, and �2 generated 2 linear fluorescencenonspecific binding of the antibodies. The sections were incu-activities. One was along the interface between the epithe-bated with the primary antibodies overnight at 4oC. Immuno-lium and the corneal stroma. The other line was alongreactivity was detected by a Histofine SAB-PO kit (Nichirei)

according to the manufacturer’s protocol; 3,3�-diamino-ben- the interface between Descemet’s membrane and thezidine tetrahydrochloride was used as a chromogen, and the endothelium. These staining patterns were characteristicsections were counterstained with Mayer’s hematoxylin. findings for basement membrane molecules of the cornea.

Before the excimer laser treatment, Descemet’s mem-Indirect Immunofluorescence Staining brane and the endothelial interface showed more intense

For the indirect immunofluorescein staining, the corneal fluorescence activity than the epithelial side for all 3section was exposed to primary antibodies for 1 hour at room antibodies in the study (Figure 2).temperature and fluorescein isothiocyanate-conjugated goat

Six hours after excimer laser treatment, active immu-antimouse immunoglobulin G (IgG). Rabbit antigoat IgGnoreactivity for laminin-5 and the �2 and �1 chains waswas the second antibody used for immunofluorescence mi-found on the denuded corneal bed in the amniotic mem-croscopy. The primary antibodies were the same as those

used for immunohistochemical staining. brane group but not in the control group. Laminin-5

J CATARACT REFRACT SURG—VOL 30, OCTOBER 20042194


Figure 2. (Lee) Laminin-5 (a ) and laminin �1 (b ) are present in the normal (wound-free) corneal epithelial basement membrane (arrow) andendothelial surface (arrowhead) (magnification �200).

and laminin �2 formed a linear and well-localized im- epithelial cells both showed laminin-5 immunoreactiv-munoactivity on the ablated cornea. However, there ity (Figure 5, c).were no difference in linear fluorescein activity of lami-nin �1 between the 2 groups; the immunoactivity was Discussionincomplete and segmented.

In this study, we proved that laminin-5 intensifiedTwelve hours after PRK, the amniotic membraneits expression during corneal epithelial wound healinggroup showed more intense expression of laminin-5and abruptly decreased its expression at the basementand �2 than the control group (Figure 3). The differencemembrane when epithelialization was complete. Thein intensity continued for 6 to 24 hours after the excimerexpression of laminin-5 was more intense in eyes withlaser treatment (Figure 3, a to d ). However, there wasamniotic membrane transplantation on the excimer-no difference between the 2 groups in the immunoreac-laser-ablated rat cornea than in the control group. How-tivity of laminin �1 at 12 hours (Figures 3, e and f ).ever, expression of the �1 chain during epithelial healingDuring epithelialization, the immunoreactivity ofwas not affected by the use of amniotic membrane.laminin-5, �2, and �1 was seen on the denuded stromal

Among the laminin isoforms, only laminin-5 andbed and in the untreated area, which was not woundedlaminin-12 have a �2 chain; however, the other lamininby the excimer laser (Figure 4, a and b). The epithelialsubtypes have a �1 chain as their trimeric components.6cells on the untreated cornea showed more intense im-Thus, we used antibodies for laminin-5 and the �2munoreactivity than the regenerated epithelium. In ad-chain in immunostaining study as a double check. Indition to the epithelial cells, some keratocytes showedthe immunoblotting study, we used an antibody againstimmunoreactivity for laminin during reepithelializationthe �3 subchain, which only laminin-5 has, to distin-(Figure 4, c).guish laminin-5 from laminin-12.The immunoreactivity of laminin-5 and laminin

Previously, laminin-5 was named epiligrin, kalinin,�2 was intensified in the denuded stromal surface duringand nicein. It is well known that laminin-5 promotesepithelial healing in both groups; however, the immuno-cell adhesion and spreading, formation of hemidesmo-reactivity of both laminins significantly decreased whensomes, and cell migration in some human cell lines.17–20reepithelialization was complete. Forty-eight hours andQin and Kurpakus9 report that primary cultured corneal72 hours after treatment, the increased immunoreactiv-epithelial cells are stimulated by transforming growthity for laminin-5 and laminin �2 had almost disappearedfactor-� (TGF-�) or EGF to become highly motile;at the basement membrane and the intensity was similarthis is associated with endogenously secreted laminin-5.to that preoperatively (Figure 5, a and b). LamininHowever, O’Toole et al.21 and Ebihara et al.10 report�1, however, increased in intensity after 48 hours andthat exogenously added laminin-5 inhibits human kera-reached peak intensity at 96 hours, the last examination

in this study. Keratocytes with anterior stroma and tinocyte and keratocyte migration, respectively. Above



Figure 3. (Lee) Twelve hours after excimer laser PRK treatment of a rat cornea. a: The control group showed faint and coarse expressionof laminin-5 (arrow). b: The amniotic membrane group showed linear, well-formed immunoreactivity of laminin-5 on the wounded bed (arrowhead).c: Laminin �2 was not expressed on the wound bed in the control group until 12 hours after PRK. d: The amniotic membrane group hadlinear staining of laminin �2 on the ablation surface (arrow). e and f: There was no difference in immunofluorescein intensity in the laminin �1chain between the groups until 12 hours (arrowhead) (magnification �200). Ep � epithelial side; Endo � endothelial side.

all, laminin-5 was thought to participate in cell motility decided that if it did, we would examine whether amni-otic membrane intensifies the expression of laminin-5and adhesion during wound healing. Despite this evi-

dence, to our knowledge there have been no reports of and its components. Laminin-5 is a known componentof preserved human amniotic membrane.14 The exis-in vivo studies proving the expression of laminin-5

during epithelial wound healing, including in the cor- tence of laminin-5 in amniotic membrane might suggestthat laminin-5 from the amniotic membrane partici-neal tissue.

At the beginning of the study, we tried to determine pates in corneal epithelial cell motility and adhesionduring amniotic-membrane-induced corneal epithelialwhether laminin-5 is expressed during reepithelializa-

tion of the excimer laser ablation on the cornea. We healing. If this is true, amniotic membrane could be



Figure 4. (Lee) Laminin-5 expressions on the reepithelialized area (a ), denuded stromal bed (b ), and untreated area (c ) 12 hours afterPRK. a: The regenerated epithelial area showed intense laminin-5 expression (arrow). b: The denuded stromal bed also showed immunoreactivityof laminin-5 (arrowhead). c: The stromal keratocytes on the untreated area showed immunoreactivity of laminin-5 (arrow), and the epithelialcells showed immunoreactivity of laminin-5 (magnification �200). Ep � epithelial side; Endo � endothelial side.

used as a substrate in ocular surgery. In vivo and in vitro cells with EGF and TGF-� is related to increased lami-nin-5 production in vitro. Transforming growth factor-�,studies found laminin-5 was expressed during corneal

epithelial wound healing and participated in the corneal TGF-�, and EGF are known components of amnioticmembrane. Thus, it is possible these cytokines led toepithelial wound-healing process. Moreover, amniotic

membrane induced more laminin-5 in the epithelial laminin-5 production in the rat corneal epithelial cells,as in other types of cells.wound, which explains how amniotic membrane con-

tributes to the reepithelialization process. The other possibility is that soluble or insolublelaminin-5 in amniotic membrane was diffused on de-We have 2 hypotheses to explain the increased lami-

nin-5 expression during corneal epithelial healing with nuded stroma. The increased immunoactivity of lami-nin-5 on deepithelialized stromal bed in the amnioticamniotic membrane transplantation. The first is that

amniotic membrane actively induces laminin-5 expres- membrane group, therefore, originated from the amni-otic membrane itself and was not indirectly producedsion as a direct result of the membrane’s characteristics.

In our study, the epithelial and stromal cells not treated by corneal epithelial cells. However, in our study, wedid not remove the epithelium of the amniotic mem-with the excimer laser also showed active immunoreac-

tivity for laminin. Moreover, the amniotic membrane brane. In addition, the stromal side of amniotic mem-brane was in contact with the ablated corneal stromalgroup had more intense immunoreactivity of laminin-5

than the control group. This could mean that some bed. We thought it difficult to diffuse laminin-5 in theamniotic membrane through the epithelium or underly-factor of amniotic membrane stimulates the cells to

produce more laminin-5. Some cytokines, including ing amniotic membrane stroma on the excimer-laser-ablated corneal bed. Moreover, considering the increasedTGF-�, are known to regulate the laminin-5 gene in

human keratinocyte and fibroblast.8,9 Qin and Kurpa- immunoactivity of laminin-5 on the limbal area in theamniotic membrane group, it is hard to believe that thekus9 report that the faster healing of corneal epithelial



Figure 5. (Lee) Immunofluorescein staining of laminin-5 �2, and �1 subchains after reepithelialization in a cornea treated with amnioticmembrane. a: Laminin-5 was still faintly present on the epithelial basement membrane side (arrow) 48 hours after PRK (magnification �200).b: The �2 chain could barely be seen on the reepithelialized stromal bed 72 hours after PRK (magnification �300). c: The �1 chains remainedimmunoactive (arrow) 96 hours after the excimer laser ablation (magnification �300).

amniotic membrane laminin components were released especially laminin-5, during wound healing. Furtherduring reepithelialization. To identify the origin of in- studies to determine the exact mechanism of increasedcreased laminin-5 expression in the stromal bed, we laminin-5 production by amniotic membrane trans-tried to immunostain with different antibodies for hu- plantation are needed.mans and rats. For the cross-reactivity of the antibodyfor laminin-5, however, we could not differentiate the Referencesorigin of the increased laminin-5 in the amniotic mem- 1. Shapiro MS, Friend J, Thoft RA. Corneal re-epitheliali-brane cases. It is important to establish the origin of zation from the conjunctiva. Invest Ophthalmol Vis Scilaminin-5 in the amniotic membrane used to treat the 1981; 21:135–142

2. Yurchenco PD, O’Rear JJ. Basal lamina assembly. Currdeepithelialized cornea to uncover the mechanism ofOpin Cell Biol 1994; 6:674–681how the amniotic membrane helps reepithelialization.

3. Khodadoust AA, Silverstein AM, Kenyon KR, DowlingWith the molecular biology technique, the origin ofJE. Adhesion of regenerating corneal epithelium; the role

the increased laminin-5 in amniotic membrane canof basement membrane. Am J Ophthalmol 1968; 65:

be proved. 339–348In conclusion, laminin-5 was produced by corneal 4. Burgeson RE, Chiquet M, Deutzmann R, et al. A new

epithelial cells and keratocytes during epithelial wound nomenclature for the laminins. Matrix Biol 1994; 14:209–211healing. Because it is expressed during reepithelializa-

5. Jones JCR, Dehart GW, Gonzales M, Goldfinger LE.tion, laminin-5 probably actively participates in reepi-Laminins: an overview. Microsc Res Tech 2000; 51:211–thelialization. Amniotic membrane transplantation after213

PRK can increase laminin-5 expression in healing cor- 6. Ljubimov AV, Burgeson RE, Butkowski RJ, et al. Hu-neal tissue. These results indicate that the mechanism man corneal basement membrane heterogeneity: topo-of amniotic-membrane-facilitated epithelial healing graphical differences in the expression of type IV collagen

and laminin isoforms. Lab Invest 1995; 72:461–473may be related to the increased accumulation of laminin,



7. Qin P, Piechocki M, Lu S, Kurpakus MA. Localization of 16. Geerling G, Daniels JT, Dart JKG, et al. Toxicity ofnatural tear substitutes in a fully defined culture modelbasement membrane-associated protein isoforms duringof human corneal epithelial cells. Invest Ophthalmol Visdevelopment of the ocular surface of mouse eye. DevSci 2001; 42:948–956Dyn 1997; 209:367–376

17. Kikkawa Y, Umeda M, Miyazaki K. Marked stimulation8. Tunggal P, Smyth N, Paulsson M, Ott M-C. Laminins:of cell adhesion and motility by ladsin, a laminin-likestructure and genetic regulation. Microsc Res Tech 2000;scatter factor. J Biochem 1994; 116:862–86951:214–227

18. Rousselle P, Aumailley M. Kalinin is more efficient than9. Qin P, Kurpakus MA. The role of laminin-5 in TGF�/laminin in promoting adhesion of primary keratinocytesEGF-mediated corneal epithelial cell motility. Exp Eyeand some other epithelial cells and has a different require-Res 1998; 66:569–579ment for integrin receptors. J Cell Biol 1994; 125:205–10. Ebihara N, Mizushima H, Miyazaki K, et al. The func-214tions of exogenous and endogenous laminin-5 on corneal

19. Hormia M, Falk-Marzillier J, Plopper G, et al. Rapidepithelial cells. Exp Eye Res 2000; 71:69–79spreading and mature hemidesmosome formation in Ha-

11. Azuara-Blanco A, Pillai CT, Dua HS. Amniotic mem-CaT keratinocytes induced by incubation with soluble

brane transplantation for ocular surface reconstruction. laminin-5r. J Invest Dermatol 1995; 105:557–561Br J Ophthalmol 1999; 83:399–402 20. Baker SE, Dipasquale AP, Stock EL, et al. Morphoge-

12. Tseng SCG, Prabhasawat P, Barton K, et al. Amniotic netic effects of soluble laminin-5 on cultured epithelialmembrane transplantation with or without limbal allo- cells and tissue explants. Exp Cell Res 1996; 228:262–grafts for corneal surface reconstruction in patients with 270limbal stem cell deficiency. Arch Ophthalmol 1998; 21. O’Toole EA, Marinkovich MP, Hoeffler WK, et al.116:431–441 Laminin-5 inhibits human keratinocyte migration. Exp

13. Tsubota K, Satake Y, Ohyama M, et al. Surgical recon- Cell Res 1997; 233:330–339struction of the ocular surface in advanced ocular cicatri-cial pemphigoid and Stevens-Johnson syndrome. Am J From the Institute of Vision Research, Department of Ophthalmology,

Yonsei University College of Medicine (J.H. Lee, H.K. Lee, Chang,Ophthalmol 1996; 122:38–52E.K. Kim, Seong), Balgeunsesang Ophthalmology Clinic (J.K. Kim),14. Kurpakus-Wheater M. Laminin-5 is a component ofDepartment of Pathology, Yonsei University College of Medicine (S.E.preserved amniotic membrane. Curr Eye Res 2001;Kim, Hong), and Project of Brain Korea 21 (E.K. Kim), Seoul, Korea.22:353–357Supported by a grant from the Korea Health 21 R&D Project, Ministry15. Kim JC, Tseng SCG. Transplantation of preserved hu-of Health and Welfare, Republic of Korea (02-PJ1-PG1-CH02-0003).man amniotic membrane for surface reconstruction in

severely damaged rabbit corneas. Cornea 1995; 14:473– None of the authors has a financial or proprietary interest in anymaterial or method mentioned.484


expression of laminin-5 with amniotic membrane transplantation in excimer laser ablated rat corneas

Documents