- Email: [email protected]
The Role of Limbal Stem Cells in Corneal Epithelial Maintenance
The Role of Limbal Stem Cells in Corneal Epithelial Maintenance Testing the Dogma Harminder S. Dua, MD, PhD,1 Ammar Miri, MD,1 Thaer Alomar, MD,1 Aaron M. Yeung, MD,1 Dalia G. Said, MD1,2 Objective: To study and characterize the epithelial cells in patients with a central “island” of normal epithelial cells surrounded with 360° of clinically apparent limbal stem cell (SC) deficiency with conjunctivalization of the limbus and peripheral cornea. Design: Observational, prospective, consecutive case series. Participants: Five human subjects (8 eyes) who presented with total limbal SC deficiency in 1 or both eyes with a central area of normal corneal epithelial cells. Methods: Clinical slit-lamp examination, aided with fluorescein staining, for evidence of conjunctivalization and in vivo confocal microscopy (IVCM) of the conjunctivalized limbus and peripheral cornea and the normal central corneal epithelium. Main Outcome Measure: Long term survival of normal stratified corneal epithelial cell sheet in the presence of total limbal SC deficiency. Results: In all 8 eyes the diagnosis of limbal SC deficiency was confirmed by clinical and IVCM examination. The conjunctivalized area extended circumferentially along the entire limbus, seen clinically by the presence of fluorescein staining cells, epithelial irregularity, and vascularization and by IVCM showing bright conjunctival epithelial cells, superficial and deep blood vessels, and goblet cells. The central corneal epithelial cells had a normal appearance with polygonal superficial cells, well-defined wing cells, and smaller basal cells. The central “islands” of normal epithelial cells remained unchanged over the mean follow-up period of 60 months (range, 8 –12 years). Conclusions: The existence and survival of a healthy sheet of corneal epithelial cells over the follow-up period, in the presence of clinically apparent total limbal SC deficiency, suggests a limited role of limbal epithelial SC in physiologic homeostasis of the corneal epithelium. Financial Disclosure(s): The authors have no proprietary or commercial interest in any materials discussed on this article. Ophthalmology 2009;116:856 – 863 © 2009 by the American Academy of Ophthalmology.
Evidence accumulated through clinical research amply demonstrates that the corneoscleral limbus is the repository of corneal epithelial stem cells (SC).1–5 This evidence has been comprehensively reviewed in several publications.6 – 8 In partial limbal epithelial defects, a preferential circumferential migration of cells from both ends of the remaining intact epithelium until limbal epithelial integrity is restored. This precedes reepithelization of any central corneal defect. Thus, circumferentially migrating cells probably represent, in part, the healing response of limbal SC.9,10 Significant evidence from molecular studies and immunophenotypic characterization of limbal epithelial cells also point to the limbus as being the repository of epithelial SC. Important examples include the demonstration of the SC transporter protein ABCG2 in a specific population of limbal epithelial cells.11,12 A specific isoform of P63 (⌬Np63␣), a transcriptional factor believed to correlate with SC, has also been shown in a small population of limbal cells13 and the presence of limbal epithelial crypts (LEC), presenting characteristics of putative SC niches, along the human limbus have recently been demonstrated.14,15 Similarly, the adhesion molecule desmoglein, the absence of which correlates with increased prolifer-
856
© 2009 by the American Academy of Ophthalmology Published by Elsevier Inc.
ative potential, is downregulated in LEC.15 The extracellular matrix of the limbus too is different from that of the central cornea in that there is specific expression of 1 integrin and tenacin C, among others.15 Connexin 43, a gap junction protein, is normally downregulated in SC, but its expression is increased in niche cells during proliferation of SC.16,17 It is generally absent in limbal basal cells, but is expressed in a proportion of LEC and in limbal basal cells adjacent to LEC, which may be functioning as niche cells.14,15 The XYZ hypothesis (X ⫹ Y ⫽ Z), where X is the anterior migration of the cells from the basal epithelium, Y is the centripetal migration of peripheral cells from the limbus, and Z is the loss of cells from the surface, proposed by Richard Thoft, encompasses the limbus as the source for cell renewal.18 Damage to the limbus by disease or injury, especially chemical and thermal burns, results in an altered and abnormal corneal surface with serious visual consequences, amply illustrating the clinical importance of the limbus in contributing to the optical qualities of the cornea.7,8,19 –22 This evidence contributes to the widely accepted belief that the limbus is essential for maintenance of the central corneal epithelium and damage to it would result in abnorISSN 0161-6420/09/$–see front matter doi:10.1016/j.ophtha.2008.12.017
Dua et al 䡠 The Limbus in Corneal Epithelial Maintenance Table 1. Clinical Details of the Patients Patient
Eye
Vision (Snellen acuity)
Age
1
Right Left Right Left Right Left Left Right
6/36 6/60 6/5 6/9 6/18 6/18 6/12 CF
43
Multiple surgeries
19 months
18
Polyglandular autoimmune disease
10 years
57
Contact lens
74 41
Chronic inflammation (rheumatoid keratitis) Aniridia
2 3 4 5
Predisposing Factors
Duration of Follow-up
8 months 1 year 12 years
CF ⫽ Counting fingers at 2 meters distance. In vivo confocal microscopy was carried out in both eyes of patients 1 and 3 and in the right eye of patient 2.
Figure 1. Diffuse slit-lamp images of the cornea. The corneae are stained with 2% sodium fluorescein dye and the images taken with a cobalt blue filter (490 nm). In all images, the central corneal epithelial cell sheet has a smooth surface, which contrasts with the surrounding irregular conjunctivalized epithelium that stains variably with fluorescein dye. A, Patient number 4 in Table 1. Note the pooling of dye (tears) at the junction of the conjunctivalized (vascularized) epithelium and the normal central corneal epithelium. The pupil is partly covered with conjunctivalized epithelium. B, Patient number 5 in Table 1. A sheet of normal central corneal epithelium is seen in a patient with aniridia. The limbus and peripheral cornea shows fluorescein staining and vascularization (conjunctivalization). C, Patient number 2 in Table 1. The surviving central corneal epithelial sheet is distinctly demarcated from the surrounding conjunctivalized epithelium. D, Patient number 3 in Table 1. A circular sheet of normal corneal epithelium is seen surrounded with fluorescein stained conjunctivalized epithelium in the left eye. The appearance of the peripheral and limbal epithelium did not change despite discontinuation of contact lens wear for several months, but the central punctuate keratitis resolved. E, Patient number 1 in Table 1. The patient had multiple surgeries on both sides (bilateral retinal detachment repair, pars plana vitrectomy with silicone oil, bilateral lens extraction with implant, left cryocycloablation). The central surviving corneal epithelial sheet is covering the visual axis. It is surrounded by conjunctivalized epithelium.
857
Ophthalmology Volume 116, Number 5, May 2009
Figure 2. In vivo confocal microscopy images of the clinically normal central corneal epithelium. A, Polygonal flat cells with well-defined borders and distinct nuclei represent the normal superficial cells of the central corneal epithelium. B, The dark cytoplasm and well-defined bright borders represent normal intermediate cells of the central corneal epithelium. C, The small cell outlines seen in the central corneal epithelium are typical of the basal (transient amplifying cells) of the normal epithelium. D, Bright (hyperreflective) nuclei of normal keratocytes in the superficial corneal stroma of the central cornea underlying the normal corneal epithelial layers described in A–C. E, Dendritic (Langerhans) cells are seen among the basal epithelial cells of the central corneal epithelium.
858
Dua et al 䡠 The Limbus in Corneal Epithelial Maintenance mal corneal epithelium. In this paper, we present evidence from patients with clinically diagnosed total SC deficiency wherein the central corneal epithelium was preserved over a long period of time, raising questions about the role of the limbus in normal corneal epithelial homeostasis.
Methods A total of 8 eyes of 5 patients, of which 4 were males, who had clinically diagnosed total limbal SC deficiency were included in this study. All patients attended the Cornea Unit of the Department of Ophthalmology, Queens Medical Centre, University Hospital, Nottingham. Data were collected by a review of the case notes and clinical examination of the patients within the last couple of months under a protocol approved by the local Ethics Committee, No. 06/Q203/46.
Diagnosis of Stem Cell Deficiency The clinical diagnosis of limbal SC deficiency was made on the basis of the following parameters that have been previously described7,8: loss of normal limbal anatomy; conjunctivalization of the cornea with conjunctival/metaplastic cells being highlighted with fluorescein stain; and the corresponding corneal surface (epithelium) being thinner and irregular compared with normal corneal epithelium and superficial and deep vascularization in the conjunctivalized cornea. In vivo confocal microscopy (IVCM) characterization of the conjunctivalized epithelium on the cornea and along the circumference of the limbus was also performed. In 1 case, impression cytology was performed to demonstrate goblet cells in the impression specimen. We performed IVCM in 5 eyes of 3 patients (Table 1) using the Heidelberg Retina Tomograph II Rostock Corneal Module (Heidelberg Engineering GmBH, Dossenheim, Germany). This microscope utilizes a 670-nm red wavelength diode laser (class 1) source that captures 400 ⫻ 400 m digital images23 of the area scanned. In addition to use of coupling agent between the objective lens and the disposable cap, viscotears gel (Viscotears; polyacrylic acid, Novartis Ophthalmic Ltd, Hettingen, Switzerland) was applied to the contact surface of the cap to afford lubrication during movement and some protection from friction, to the corneal epithelial cells. The central corneal epithelium and the surrounding “abnormal epithelium” were scanned.
Results The age of the patients ranged from 28 years to 74 years with a mean of 47 years. There were 4 males and 1 female. Follow-up of patients was from 8 months to 12 years (mean, 60 months; median, 19 months) in our clinic; however, all had a diagnosis of an “abnormallooking limbus” for 3 to 6 months before being referred to us. The primary pathology included aniridia, long-term contact lens wear (SoftPerm Lens, CIBA Vision, Duluth, GA), multiple operative procedures, and rheumatoid and polyglandular autoimmune disease (Table 1). Patients’ symptoms included photophobia, watering of the eye, and chronic irritation with redness and blepharospasm (caused by photophobia in both eyes of patients 1 and 2 in Table 1).
Clinical Examination All eyes had stable and normal central corneal epithelium, with occasional transient superficial punctate keratitis in 5 eyes (3
patients) over the duration of the follow-up. All patients had loss of limbal anatomy, superficial vessels and altered limbal and peripheral corneal epithelium, which was thinner and stained with fluorescein extending from 2 to 5 mm from the limbus on to the corneal surface. These changes were present along the entire circumference of the limbus and cornea. The junction of the abnormal peripheral epithelium and the normal central epithelium was clearly demarcated when examined with fluorescein stain and cobalt blue filter illumination on the slit lamp (Fig 1A–E). There was pooling of fluorescein-stained tears along the junction of the thinner conjunctivalized corneal surface and the thicker corneal epithelium. The tear film breakup time over the abnormal epithelium was short (⬍5 seconds) in some areas. The central corneal epithelium was normal, with a uniform spread of the tear film. Central corneal sensations were normal. The Snellen visual acuity, measured in meters, was between counting fingers (2/60 approximate) and 6/5. Over the follow-up period, the best-corrected visual acuity remained within 1 line of the baseline. All patients received topical treatment with artificial tears, which included carboxymethylcellulose sodium protective gel (Celluvisc; Allergan, Mougins, France), polyacrylic acid ophthalmic gel (Viscotears, Novartis Ophthalmic Ltd.), and hydroxypropyl methylcellulose (Hypermellose; American Polymer Standards Corp, Mentor, OH). Two patients (1L and 5R in Table 1) also received antiglaucoma medication to control the intraocular pressure. Medication for 1L was dorzolamide 2% and timolol 0.5% combination (Cosopt; Merck Frosst Canada Ltd., Quebec, Canada) and bimatoprost ophthalmic solution (Lumigan, Allergan Inc., Irvine, CA). Medication for 5R was dorzolamide hydrochloride 2% (Trusopt, Merck Frosst Canada Ltd.), latanoprost ophthalmic solution 0.005% (Xalatan; Pharmacia, Uppsala, Sweden), brimonidine tartrate 0.2% (Alphagan; Allergan Inc.), and timolol 0.5% (Timoptol; Banyu Pharmaceutical Co, Ltd, Osaka, Japan). All patients had also received steroid drops (prednisolone acetate 0.5%, predsol minims; Bausch & Lomb, UK Ltd, London, UK) intermittently.
In Vivo Confocal Microscopy of the Central Cornea Scanning of the central cornea covered by clinically normal epithelium confirmed all features of normal epithelium and stroma on confocal imaging. The superficial cells of the corneal epithelium showed a diameter of 35 to 50 m, with bright cytoplasm and bright nuclei, which were 10 m in size. The wing cells seemed smaller than the superficial cells; their diameter was 15 to 25 m, with well-defined, bright borders. In comparison with the superficial cells, their reflectivity was more homogenous. The basal cell layer represented a honeycomb pattern of cells. Their diameter measured 10 to 20 m. The cells appeared dark with bright borders. Unlike the superficial cells, the nuclei in the wing cells and the basal cells were not seen on IVCM (Fig 2A–C). In the corneal stroma, the keratocyte nuclei were visible as separate bright structures with a dark background (Fig 2D). All these features were as described in normal corneae.24 Among basal epithelial cells, hyperreflective dendritic cells25 were demonstrated in both eyes of 2 patients (Fig 2E).
In Vivo Confocal Microscopy of the Transitional Area At the transitional zone where the conjunctivalized epithelium meets the normal corneal epithelium, the conjunctivalized epithelium could easily be distinguished from the corneal epithelium because the cells were hyperreflective with bright nuclei
859
Ophthalmology Volume 116, Number 5, May 2009
Figure 3. In vivo confocal microscopy images of the limbus and peripheral cornea. A, The superficial cells of the transitional zone. Conjunctival cells are bright and hyperreflective with ill defined margins. The normal corneal epithelial cells are dark, with well-defined margins. Arrow heads mark the junction of the 2 phenotypes of cells. *An area of conjunctivalized cornea; ⫹The adjacent area with normal corneal epithelial cells. B, The subsuperficial cells of the transitional zone. The conjunctivalized area (*) demonstrates prominent cell nuclei with no cell borders. The adjoining area (⫹) of normal corneal epithelial cells with bright, well-defined borders is clearly visible. Arrow heads mark the junction. C, Cystic changes (arrowhead) and goblet cells, at times arranged in rosettes (arrow), are seen over conjunctivalized limbs and cornea (*). D, A loose network of hyper reflective fibers is seen in the superficial stroma of conjunctivalized cornea and limbus. E, Blood vessels (arrowhead) are seen in the deep layers of conjunctivalized epithelium. An adjoining area of normal corneal epithelium (⫹) is also seen.
and ill-defined borders between cells. Conversely, the superficial and subsuperficial corneal epithelial cells presented as dark structures with very well-defined margins and bright nuclei (Fig 3A, B).
860
Goblet cells were visible throughout the conjunctivalized epithelium. These cells were between 10 and 23 m in size, hyperreflective, round to oval in shape, and at times visualized in characteristic rosettes (Fig 3C). Furthermore, unlike the corneal
Dua et al 䡠 The Limbus in Corneal Epithelial Maintenance stroma, the conjunctival stromal collagen could be visualized clearly by IVCM as a dense, irregular, hyperreflective network (Fig 3D). Other components of the conjunctiva which were visualized by IVCM included cystic changes and superficial and deep vessels (Fig 3C, E).
Discussion The basal epithelial cells of the palisades of Vogt together with the cell mass in the limbal epithelial crypts are the source of epithelial regeneration after extensive loss of the corneal epithelium. The epithelial SC of the limbus are slow cycling and are not known to respond to injury by increasing their rate of proliferation or their numbers. The progeny of these SC, termed the transient amplifying cells (TAC) have increased proliferative potential and respond immediately to epithelial injury by sliding and centripetal migration to cover denuded areas and cell division to generate postmitotic suprabasal wing cells and superficial cells.8 In partial limbal and corneal epithelial insults in humans, the damaged limbus heals by circumferential migration of cells from the surviving epithelium and, when completed, the central cornea retains a healthy sheet of normal corneal epithelium.10 However, when this process is halted by the encroachment of conjunctival epithelium, it results in partial limbal deficiency with part of the cornea and limbus being covered by normal corneal epithelial cells and part conjunctivalized, with both phenotypes coexisting for prolonged periods, almost indefinitely.26,27 This has also been demonstrated experimentally in rabbit eyes, although with a much shorter follow-up period.28,29 In rabbit eyes, experimentally created partial thickness and partial extent limbal deficiency were associated with conjunctivalization, which was more apparent when the limbus was challenged by removal of central corneal epithelium.28,29 The limbus is attributed with a “barrier” function, preventing encroachment of the conjunctival epithelium onto the cornea during normal homeostasis.1,30,31 When this “barrier” function is damaged, conjunctival epithelium together with blood vessels and fibrous tissue encroach onto the cornea. The effects of limbal deficiency mentioned provide compelling evidence to assume that the limbus plays a similar role during normal corneal epithelial turnover. However, the observations noted in the 8 eyes with clinically diagnosed total limbal SC deficiency presented in this study seem to challenge this assumption. The central islands of corneal epithelium are apparently capable of sustaining themselves in a healthy state, compatible with corneal visual function, despite complete loss of visible limbus. Two obvious explanations emerge: (1) that some limbal SC or SC niches continue to survive though not clinically visible, and contribute to the maintenance of the central epithelium or (2) that the basal TAC of the central surviving epithelium are independently capable of maintaining the overlying cell mass for a long period of time, suggesting that the limbus may not have a critical role in physiologic corneal epithelial homeostasis. In all the patients described in this study, the limbus was affected along its entire circumference. This was not only clinically evident, but also confirmed by detailed in vivo confocal microscopic examination of 5 of the 8 eyes.
All layers of epithelium of the affected limbus were examined and no evidence of normal limbal or corneal phenotype of cells was detected. This does not entirely rule out the possibility that some limbal cells may have survived, although this seems less likely. What would happen to the corneal surface should the central islands of healthy epithelium be damaged or destroyed by injury or disease remains an important and interesting albeit unanswered question. If some limbal SC are surviving, normal epithelium should in theory regenerate. The more likely scenario is that the conjunctivalization would progress to cover the denuded area and present as classical total SC deficiency. In experimentally created partial limbal SC deficiency, removal of the central corneal epithelium resulted in conjunctivalization of the cornea.28 Life-term follow-up of these patients may provide an opportunity to test this hypothesis, although every precaution is being employed to prevent such an eventuality. Some elegant animal experiments performed by Barrandon (Lausanne, Switzerland) lend support to the hypothesis that the limbal SC may not play an important role in normal epithelial turnover. When limbal explants from mice constitutively expressing -galactosidase (cells stained blue) or the green fluorescent protein (all cells fluoresce green) were transplanted on to corresponding limbal defects created in recipient nude mice, the labeled cells stayed in the limbal tissue and did not migrate on the to cornea over the period of the study. However, when central epithelial defects were created in the recipient mice, rapid healing was observed in which labeled cells migrated from the transplanted limbal explants onto the corneal surface. This suggests that the transplanted limbal tissue contributed to epithelial regeneration, together with the rest of the limbus, only when the central corneal epithelium was stressed by injury. In another study by Barrandon, 360° of limbal defects (total limbal deficiency) were made and the cornea remained transparent. Furthermore, when labeled central corneal mouse epitheliumbearing tissue was serially transplanted to the limbus, the central epithelial explants successfully regenerated the entire denuded corneal epithelial surface, which led him to postulate that the central basal epithelial cells could acquire SC characteristics (unpublished data under peer review and personal communication). An alternative explanation could be that the basal cells, given their proliferative potential as TAC, are independently capable of regenerating and maintaining an epithelial sheet. These cells may represent a heterogenous group of cells with some retaining a progenitor cell potential. In a recent study32 on human whole cornea mounts maintained in organ culture, it has been shown that total destruction to the limbus (with excimer laser ablation to a depth of 80 m) was followed by healing of the cornea from cells of the central epithelium, which in theory could not have been SC. Experiments on rabbits, where the limbus was totally destroyed, showed survival of the central epithelium, which, however, failed to regenerate when it too was mechanically removed.31 In recent years, successful ocular surface reconstruction has been carried out using ex vivo expanded sheets of limbal epithelial cells without substrate33 or with fibrin or
861
Ophthalmology Volume 116, Number 5, May 2009 amniotic membrane as substrate.34,35 This approach has yielded successful outcomes over a long follow-up period, leading some to postulate that the amniotic membrane substrate is capable of sustaining SC.36 Evidence for this is not convincing and does not reconcile with the widely accepted notion that SC retain their stemness within defined niches and tend to follow a path of differentiation when outside their niche. Any SC in ex vivo expanded sheets would therefore not remain so for long, yet the ocular surface remains healthy, again suggesting that it is the TAC and not SC in the sheets that account for the successful outcome. Alternatively, it is possible that within the TAC pool there are some cells that have characteristics closer to the parent SC than to the TAC progeny (we have termed these conceptual cells “transient cells”). The observations from the 8 eyes examined in this study and the other evidence presented support the hypothesis that physiologic corneal epithelial homeostasis can be maintained in the absence of clinically detectable limbal epithelial SC.
16. 17.
18. 19. 20. 21. 22.
References
23.
1. Tseng SC. Concept and application of limbal stem cells. Eye 1989;3:141–57. 2. Davanger M, Evensen A. Role of the pericorneal papillary structure in renewal of corneal epithelium. Nature 1971;229:560–1. 3. Townsend WM. The limbal palisades of Vogt. Trans Am Ophthalmol Soc 1991;89:721–56. 4. Matsuda M, Ubels JL, Edelhauser HF. A larger corneal epithelial wound closes at a faster rate. Invest Ophthalmol Vis Sci 1985;26:897–900. 5. Ebato B, Friend J, Thoft RA. Comparison of central and peripheral human corneal epithelium in tissue culture. Invest Ophthalmol Vis Sci 1987;28:1450 – 6. 6. Dua HS, Azuara-Blanco A. Limbal stem cells of the corneal epithelium. Surv Ophthalmol 2000;44:415–25. 7. Dua HS, Joseph A, Shanmuganathan VA, Jones RE. Stem cell differentiation and the effects of deficiency. Eye 2003;17: 877– 85. 8. Dua HS. Transplantation of limbal stem cells. In: Rienhard T, Larkin F, eds. Cornea and External Eye Disease. Berlin: Springer; 2006:35– 64. Essentials of Ophthalmology. Krieglstein GK, Weinreb RN, series eds. 9. Dua HS, Forrester JV. Clinical patterns of corneal epithelial wound healing. Am J Ophthalmol 1987;104:481–9. 10. Dua HS, Forrester JV. The corneoscleral limbus in human corneal epithelial wound healing. Am J Ophthalmol 1990;110:646–56. 11. Watanabe K, Nishida K, Yamato M, et al. Human limbal epithelium contains side population cells expressing the ATP-binding cassette transporter ABCG2. FEBS Lett 2004;565:6 –10. 12. Zhou S, Schuetz JD, Bunting KD, et al. The ABC transporter Bcrp1/ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat Med 2001;7:1028 –34. 13. Kawasaki S, Tanioka H, Yamasaki K, et al. Expression and tissue distribution of p63 isoforms in human ocular surface epithelia. Exp Eye Res 2006;82:293–9. 14. Shanmuganathan VA, Foster T, Kulkarni BB, et al. Morphological characteristics of the limbal epithelial crypt. Br J Ophthalmol 2007;91:514 –9. 15. Yeung AM, Schlotzer-Schrehardt U, Kulkarni B, et al. Limbal epithelial crypt: a model for corneal epithelial maintenance
24.
862
25. 26. 27. 28. 29. 30. 31. 32.
33. 34. 35.
36.
and novel limbal regional variations. Arch Ophthalmol 2008;126:665–9. Rosendaal M, Green CR, Rahman A, Morgan D. Up-regulation of the connexin43⫹ gap junction network in haemopoietic tissue before the growth of stem cells. J Cell Sci 1994;107:29–37. Cancelas JA, Koevoet WL, de Koning AE, et al. Connexin-43 gap junctions are involved in multiconnexin-expressing stromal support of hemopoietic progenitors and stem cells. Blood 2000;96:498 –505. Thoft RA, Friend J. The X, Y, Z hypothesis of corneal epithelial maintenance [letter]. Invest Ophthalmol Vis Sci 1983;24:1442–3. Lopez-Garcia JS, Rivas Jara L, Garcia-Lozano I, Murube J. Histopathologic limbus evolution after alkaline burns. Cornea 2007;26:1043– 8. Wagoner MD. Chemical injuries of the eye: current concepts in pathophysiology and therapy. Surv Ophthalmol 1997;41:275–313. Kinoshita S, Kiorpes TC, Friend J, Thoft RA. Limbal epithelium in ocular surface wound healing. Invest Ophthalmol Vis Sci 1982;23:73– 80. Holland EJ, Schwartz GS. The evolution of epithelial transplantation for severe ocular surface disease and a proposed classification system. Cornea 1996;15:549 –56. Patel DV, Sherwin T, McGhee CN. Laser scanning in vivo confocal microscopy of the normal human corneoscleral limbus. Invest Ophthalmol Vis Sci 2006;47:2823–7. Guthoff RF, Baudouin C, Stave J. Atlas of Confocal Laser Scanning In-vivo Microscopy in Ophthalmology. Berlin: Springer; 2006:31–52. Mastropasqua L, Nubile M, Lanzini M, et al. Epithelial dendritic cell distribution in normal and inflamed human cornea: in vivo confocal microscopy study. Am J Ophthalmol 2006;142:736–44. Dua HS, Gomes JA, Singh A. Corneal epithelial wound healing. Br J Ophthalmol 1994;78:401– 8. Dua HS. The conjunctiva in corneal epithelial wound healing. Br J Ophthalmol 1998;82:1407–11. Chen JJ, Tseng SC. Corneal epithelial wound healing in partial limbal deficiency. Invest Ophthalmol Vis Sci 1990;31:1301–14. Chen JJ, Tseng SC. Abnormal corneal epithelial wound healing in partial-thickness removal of limbal epithelium. Invest Ophthalmol Vis Sci 1991;32:2219 –33. Thoft RA, Friend J, Murphy HS. Ocular surface epithelium and corneal vascularization in rabbits. I. The role of wounding. Invest Ophthalmol Vis Sci 1979;18:85–92. Huang AJ, Tseng SC. Corneal epithelial wound healing in the absence of limbal epithelium. Invest Ophthalmol Vis Sci 1991;32:96 –105. Chang CY, Green CR, McGhee CN, Sherwin T. Acute wound healing in the human central corneal epithelium appears to be independent of limbal stem cell influence. Invest Ophthalmol Vis Sci 2008;49:5279 – 86. Daya SM, Watson A, Sharpe JR, et al. Outcomes and DNA analysis of ex vivo expanded stem cell allograft for ocular surface reconstruction. Ophthalmology 2005;112:470 –7. Ishino Y, Sano Y, Nakamura T, et al. Amniotic membrane as a carrier for cultivated human corneal endothelial cell transplantation. Invest Ophthalmol Vis Sci 2004;45:800 – 6. Rama P, Bonini S, Lambiase A, et al. Autologous fibrincultured limbal stem cells permanently restore the corneal surface of patients with total limbal stem cell deficiency. Transplantation 2001;72:1478 – 85. Grueterich M, Espana EM, Tseng SC. Modulation of keratin and connexin expression in limbal epithelium expanded on denuded amniotic membrane with and without a 3T3 fibroblast feeder layer. Invest Ophthalmol Vis Sci 2003;44: 4230 – 6.
Dua et al 䡠 The Limbus in Corneal Epithelial Maintenance
Footnotes and Financial Disclosures Originally received: July 14, 2008. Final revision: November 7, 2008. Accepted: December 4, 2008.
The authors have no proprietary or commercial interest in any materials discussed in this article. Manuscript no. 2008-842.
1
Division of Ophthalmology and Visual Sciences, University of Nottingham, England. 2
Research Institute of Ophthalmology, Cairo, Egypt. Financial Disclosure(s):
Correspondence: Harminder S. Dua, Division of Ophthalmology and Visual Sciences, B Floor, Eye ENT Centre, Queens Medical Centre, University Hospital, Derby Road, Nottingham. NG7 2UH, UK. E-mail: Harminder.dua@ nottingham.ac.uk.
863
Evidence accumulated through clinical research amply demonstrates that the corneoscleral limbus is the repository of corneal epithelial stem cells (SC).1–5 This evidence has been comprehensively reviewed in several publications.6 – 8 In partial limbal epithelial defects, a preferential circumferential migration of cells from both ends of the remaining intact epithelium until limbal epithelial integrity is restored. This precedes reepithelization of any central corneal defect. Thus, circumferentially migrating cells probably represent, in part, the healing response of limbal SC.9,10 Significant evidence from molecular studies and immunophenotypic characterization of limbal epithelial cells also point to the limbus as being the repository of epithelial SC. Important examples include the demonstration of the SC transporter protein ABCG2 in a specific population of limbal epithelial cells.11,12 A specific isoform of P63 (⌬Np63␣), a transcriptional factor believed to correlate with SC, has also been shown in a small population of limbal cells13 and the presence of limbal epithelial crypts (LEC), presenting characteristics of putative SC niches, along the human limbus have recently been demonstrated.14,15 Similarly, the adhesion molecule desmoglein, the absence of which correlates with increased prolifer-
856
© 2009 by the American Academy of Ophthalmology Published by Elsevier Inc.
ative potential, is downregulated in LEC.15 The extracellular matrix of the limbus too is different from that of the central cornea in that there is specific expression of 1 integrin and tenacin C, among others.15 Connexin 43, a gap junction protein, is normally downregulated in SC, but its expression is increased in niche cells during proliferation of SC.16,17 It is generally absent in limbal basal cells, but is expressed in a proportion of LEC and in limbal basal cells adjacent to LEC, which may be functioning as niche cells.14,15 The XYZ hypothesis (X ⫹ Y ⫽ Z), where X is the anterior migration of the cells from the basal epithelium, Y is the centripetal migration of peripheral cells from the limbus, and Z is the loss of cells from the surface, proposed by Richard Thoft, encompasses the limbus as the source for cell renewal.18 Damage to the limbus by disease or injury, especially chemical and thermal burns, results in an altered and abnormal corneal surface with serious visual consequences, amply illustrating the clinical importance of the limbus in contributing to the optical qualities of the cornea.7,8,19 –22 This evidence contributes to the widely accepted belief that the limbus is essential for maintenance of the central corneal epithelium and damage to it would result in abnorISSN 0161-6420/09/$–see front matter doi:10.1016/j.ophtha.2008.12.017
Dua et al 䡠 The Limbus in Corneal Epithelial Maintenance Table 1. Clinical Details of the Patients Patient
Eye
Vision (Snellen acuity)
Age
1
Right Left Right Left Right Left Left Right
6/36 6/60 6/5 6/9 6/18 6/18 6/12 CF
43
Multiple surgeries
19 months
18
Polyglandular autoimmune disease
10 years
57
Contact lens
74 41
Chronic inflammation (rheumatoid keratitis) Aniridia
2 3 4 5
Predisposing Factors
Duration of Follow-up
8 months 1 year 12 years
CF ⫽ Counting fingers at 2 meters distance. In vivo confocal microscopy was carried out in both eyes of patients 1 and 3 and in the right eye of patient 2.
Figure 1. Diffuse slit-lamp images of the cornea. The corneae are stained with 2% sodium fluorescein dye and the images taken with a cobalt blue filter (490 nm). In all images, the central corneal epithelial cell sheet has a smooth surface, which contrasts with the surrounding irregular conjunctivalized epithelium that stains variably with fluorescein dye. A, Patient number 4 in Table 1. Note the pooling of dye (tears) at the junction of the conjunctivalized (vascularized) epithelium and the normal central corneal epithelium. The pupil is partly covered with conjunctivalized epithelium. B, Patient number 5 in Table 1. A sheet of normal central corneal epithelium is seen in a patient with aniridia. The limbus and peripheral cornea shows fluorescein staining and vascularization (conjunctivalization). C, Patient number 2 in Table 1. The surviving central corneal epithelial sheet is distinctly demarcated from the surrounding conjunctivalized epithelium. D, Patient number 3 in Table 1. A circular sheet of normal corneal epithelium is seen surrounded with fluorescein stained conjunctivalized epithelium in the left eye. The appearance of the peripheral and limbal epithelium did not change despite discontinuation of contact lens wear for several months, but the central punctuate keratitis resolved. E, Patient number 1 in Table 1. The patient had multiple surgeries on both sides (bilateral retinal detachment repair, pars plana vitrectomy with silicone oil, bilateral lens extraction with implant, left cryocycloablation). The central surviving corneal epithelial sheet is covering the visual axis. It is surrounded by conjunctivalized epithelium.
857
Ophthalmology Volume 116, Number 5, May 2009
Figure 2. In vivo confocal microscopy images of the clinically normal central corneal epithelium. A, Polygonal flat cells with well-defined borders and distinct nuclei represent the normal superficial cells of the central corneal epithelium. B, The dark cytoplasm and well-defined bright borders represent normal intermediate cells of the central corneal epithelium. C, The small cell outlines seen in the central corneal epithelium are typical of the basal (transient amplifying cells) of the normal epithelium. D, Bright (hyperreflective) nuclei of normal keratocytes in the superficial corneal stroma of the central cornea underlying the normal corneal epithelial layers described in A–C. E, Dendritic (Langerhans) cells are seen among the basal epithelial cells of the central corneal epithelium.
858
Dua et al 䡠 The Limbus in Corneal Epithelial Maintenance mal corneal epithelium. In this paper, we present evidence from patients with clinically diagnosed total SC deficiency wherein the central corneal epithelium was preserved over a long period of time, raising questions about the role of the limbus in normal corneal epithelial homeostasis.
Methods A total of 8 eyes of 5 patients, of which 4 were males, who had clinically diagnosed total limbal SC deficiency were included in this study. All patients attended the Cornea Unit of the Department of Ophthalmology, Queens Medical Centre, University Hospital, Nottingham. Data were collected by a review of the case notes and clinical examination of the patients within the last couple of months under a protocol approved by the local Ethics Committee, No. 06/Q203/46.
Diagnosis of Stem Cell Deficiency The clinical diagnosis of limbal SC deficiency was made on the basis of the following parameters that have been previously described7,8: loss of normal limbal anatomy; conjunctivalization of the cornea with conjunctival/metaplastic cells being highlighted with fluorescein stain; and the corresponding corneal surface (epithelium) being thinner and irregular compared with normal corneal epithelium and superficial and deep vascularization in the conjunctivalized cornea. In vivo confocal microscopy (IVCM) characterization of the conjunctivalized epithelium on the cornea and along the circumference of the limbus was also performed. In 1 case, impression cytology was performed to demonstrate goblet cells in the impression specimen. We performed IVCM in 5 eyes of 3 patients (Table 1) using the Heidelberg Retina Tomograph II Rostock Corneal Module (Heidelberg Engineering GmBH, Dossenheim, Germany). This microscope utilizes a 670-nm red wavelength diode laser (class 1) source that captures 400 ⫻ 400 m digital images23 of the area scanned. In addition to use of coupling agent between the objective lens and the disposable cap, viscotears gel (Viscotears; polyacrylic acid, Novartis Ophthalmic Ltd, Hettingen, Switzerland) was applied to the contact surface of the cap to afford lubrication during movement and some protection from friction, to the corneal epithelial cells. The central corneal epithelium and the surrounding “abnormal epithelium” were scanned.
Results The age of the patients ranged from 28 years to 74 years with a mean of 47 years. There were 4 males and 1 female. Follow-up of patients was from 8 months to 12 years (mean, 60 months; median, 19 months) in our clinic; however, all had a diagnosis of an “abnormallooking limbus” for 3 to 6 months before being referred to us. The primary pathology included aniridia, long-term contact lens wear (SoftPerm Lens, CIBA Vision, Duluth, GA), multiple operative procedures, and rheumatoid and polyglandular autoimmune disease (Table 1). Patients’ symptoms included photophobia, watering of the eye, and chronic irritation with redness and blepharospasm (caused by photophobia in both eyes of patients 1 and 2 in Table 1).
Clinical Examination All eyes had stable and normal central corneal epithelium, with occasional transient superficial punctate keratitis in 5 eyes (3
patients) over the duration of the follow-up. All patients had loss of limbal anatomy, superficial vessels and altered limbal and peripheral corneal epithelium, which was thinner and stained with fluorescein extending from 2 to 5 mm from the limbus on to the corneal surface. These changes were present along the entire circumference of the limbus and cornea. The junction of the abnormal peripheral epithelium and the normal central epithelium was clearly demarcated when examined with fluorescein stain and cobalt blue filter illumination on the slit lamp (Fig 1A–E). There was pooling of fluorescein-stained tears along the junction of the thinner conjunctivalized corneal surface and the thicker corneal epithelium. The tear film breakup time over the abnormal epithelium was short (⬍5 seconds) in some areas. The central corneal epithelium was normal, with a uniform spread of the tear film. Central corneal sensations were normal. The Snellen visual acuity, measured in meters, was between counting fingers (2/60 approximate) and 6/5. Over the follow-up period, the best-corrected visual acuity remained within 1 line of the baseline. All patients received topical treatment with artificial tears, which included carboxymethylcellulose sodium protective gel (Celluvisc; Allergan, Mougins, France), polyacrylic acid ophthalmic gel (Viscotears, Novartis Ophthalmic Ltd.), and hydroxypropyl methylcellulose (Hypermellose; American Polymer Standards Corp, Mentor, OH). Two patients (1L and 5R in Table 1) also received antiglaucoma medication to control the intraocular pressure. Medication for 1L was dorzolamide 2% and timolol 0.5% combination (Cosopt; Merck Frosst Canada Ltd., Quebec, Canada) and bimatoprost ophthalmic solution (Lumigan, Allergan Inc., Irvine, CA). Medication for 5R was dorzolamide hydrochloride 2% (Trusopt, Merck Frosst Canada Ltd.), latanoprost ophthalmic solution 0.005% (Xalatan; Pharmacia, Uppsala, Sweden), brimonidine tartrate 0.2% (Alphagan; Allergan Inc.), and timolol 0.5% (Timoptol; Banyu Pharmaceutical Co, Ltd, Osaka, Japan). All patients had also received steroid drops (prednisolone acetate 0.5%, predsol minims; Bausch & Lomb, UK Ltd, London, UK) intermittently.
In Vivo Confocal Microscopy of the Central Cornea Scanning of the central cornea covered by clinically normal epithelium confirmed all features of normal epithelium and stroma on confocal imaging. The superficial cells of the corneal epithelium showed a diameter of 35 to 50 m, with bright cytoplasm and bright nuclei, which were 10 m in size. The wing cells seemed smaller than the superficial cells; their diameter was 15 to 25 m, with well-defined, bright borders. In comparison with the superficial cells, their reflectivity was more homogenous. The basal cell layer represented a honeycomb pattern of cells. Their diameter measured 10 to 20 m. The cells appeared dark with bright borders. Unlike the superficial cells, the nuclei in the wing cells and the basal cells were not seen on IVCM (Fig 2A–C). In the corneal stroma, the keratocyte nuclei were visible as separate bright structures with a dark background (Fig 2D). All these features were as described in normal corneae.24 Among basal epithelial cells, hyperreflective dendritic cells25 were demonstrated in both eyes of 2 patients (Fig 2E).
In Vivo Confocal Microscopy of the Transitional Area At the transitional zone where the conjunctivalized epithelium meets the normal corneal epithelium, the conjunctivalized epithelium could easily be distinguished from the corneal epithelium because the cells were hyperreflective with bright nuclei
859
Ophthalmology Volume 116, Number 5, May 2009
Figure 3. In vivo confocal microscopy images of the limbus and peripheral cornea. A, The superficial cells of the transitional zone. Conjunctival cells are bright and hyperreflective with ill defined margins. The normal corneal epithelial cells are dark, with well-defined margins. Arrow heads mark the junction of the 2 phenotypes of cells. *An area of conjunctivalized cornea; ⫹The adjacent area with normal corneal epithelial cells. B, The subsuperficial cells of the transitional zone. The conjunctivalized area (*) demonstrates prominent cell nuclei with no cell borders. The adjoining area (⫹) of normal corneal epithelial cells with bright, well-defined borders is clearly visible. Arrow heads mark the junction. C, Cystic changes (arrowhead) and goblet cells, at times arranged in rosettes (arrow), are seen over conjunctivalized limbs and cornea (*). D, A loose network of hyper reflective fibers is seen in the superficial stroma of conjunctivalized cornea and limbus. E, Blood vessels (arrowhead) are seen in the deep layers of conjunctivalized epithelium. An adjoining area of normal corneal epithelium (⫹) is also seen.
and ill-defined borders between cells. Conversely, the superficial and subsuperficial corneal epithelial cells presented as dark structures with very well-defined margins and bright nuclei (Fig 3A, B).
860
Goblet cells were visible throughout the conjunctivalized epithelium. These cells were between 10 and 23 m in size, hyperreflective, round to oval in shape, and at times visualized in characteristic rosettes (Fig 3C). Furthermore, unlike the corneal
Dua et al 䡠 The Limbus in Corneal Epithelial Maintenance stroma, the conjunctival stromal collagen could be visualized clearly by IVCM as a dense, irregular, hyperreflective network (Fig 3D). Other components of the conjunctiva which were visualized by IVCM included cystic changes and superficial and deep vessels (Fig 3C, E).
Discussion The basal epithelial cells of the palisades of Vogt together with the cell mass in the limbal epithelial crypts are the source of epithelial regeneration after extensive loss of the corneal epithelium. The epithelial SC of the limbus are slow cycling and are not known to respond to injury by increasing their rate of proliferation or their numbers. The progeny of these SC, termed the transient amplifying cells (TAC) have increased proliferative potential and respond immediately to epithelial injury by sliding and centripetal migration to cover denuded areas and cell division to generate postmitotic suprabasal wing cells and superficial cells.8 In partial limbal and corneal epithelial insults in humans, the damaged limbus heals by circumferential migration of cells from the surviving epithelium and, when completed, the central cornea retains a healthy sheet of normal corneal epithelium.10 However, when this process is halted by the encroachment of conjunctival epithelium, it results in partial limbal deficiency with part of the cornea and limbus being covered by normal corneal epithelial cells and part conjunctivalized, with both phenotypes coexisting for prolonged periods, almost indefinitely.26,27 This has also been demonstrated experimentally in rabbit eyes, although with a much shorter follow-up period.28,29 In rabbit eyes, experimentally created partial thickness and partial extent limbal deficiency were associated with conjunctivalization, which was more apparent when the limbus was challenged by removal of central corneal epithelium.28,29 The limbus is attributed with a “barrier” function, preventing encroachment of the conjunctival epithelium onto the cornea during normal homeostasis.1,30,31 When this “barrier” function is damaged, conjunctival epithelium together with blood vessels and fibrous tissue encroach onto the cornea. The effects of limbal deficiency mentioned provide compelling evidence to assume that the limbus plays a similar role during normal corneal epithelial turnover. However, the observations noted in the 8 eyes with clinically diagnosed total limbal SC deficiency presented in this study seem to challenge this assumption. The central islands of corneal epithelium are apparently capable of sustaining themselves in a healthy state, compatible with corneal visual function, despite complete loss of visible limbus. Two obvious explanations emerge: (1) that some limbal SC or SC niches continue to survive though not clinically visible, and contribute to the maintenance of the central epithelium or (2) that the basal TAC of the central surviving epithelium are independently capable of maintaining the overlying cell mass for a long period of time, suggesting that the limbus may not have a critical role in physiologic corneal epithelial homeostasis. In all the patients described in this study, the limbus was affected along its entire circumference. This was not only clinically evident, but also confirmed by detailed in vivo confocal microscopic examination of 5 of the 8 eyes.
All layers of epithelium of the affected limbus were examined and no evidence of normal limbal or corneal phenotype of cells was detected. This does not entirely rule out the possibility that some limbal cells may have survived, although this seems less likely. What would happen to the corneal surface should the central islands of healthy epithelium be damaged or destroyed by injury or disease remains an important and interesting albeit unanswered question. If some limbal SC are surviving, normal epithelium should in theory regenerate. The more likely scenario is that the conjunctivalization would progress to cover the denuded area and present as classical total SC deficiency. In experimentally created partial limbal SC deficiency, removal of the central corneal epithelium resulted in conjunctivalization of the cornea.28 Life-term follow-up of these patients may provide an opportunity to test this hypothesis, although every precaution is being employed to prevent such an eventuality. Some elegant animal experiments performed by Barrandon (Lausanne, Switzerland) lend support to the hypothesis that the limbal SC may not play an important role in normal epithelial turnover. When limbal explants from mice constitutively expressing -galactosidase (cells stained blue) or the green fluorescent protein (all cells fluoresce green) were transplanted on to corresponding limbal defects created in recipient nude mice, the labeled cells stayed in the limbal tissue and did not migrate on the to cornea over the period of the study. However, when central epithelial defects were created in the recipient mice, rapid healing was observed in which labeled cells migrated from the transplanted limbal explants onto the corneal surface. This suggests that the transplanted limbal tissue contributed to epithelial regeneration, together with the rest of the limbus, only when the central corneal epithelium was stressed by injury. In another study by Barrandon, 360° of limbal defects (total limbal deficiency) were made and the cornea remained transparent. Furthermore, when labeled central corneal mouse epitheliumbearing tissue was serially transplanted to the limbus, the central epithelial explants successfully regenerated the entire denuded corneal epithelial surface, which led him to postulate that the central basal epithelial cells could acquire SC characteristics (unpublished data under peer review and personal communication). An alternative explanation could be that the basal cells, given their proliferative potential as TAC, are independently capable of regenerating and maintaining an epithelial sheet. These cells may represent a heterogenous group of cells with some retaining a progenitor cell potential. In a recent study32 on human whole cornea mounts maintained in organ culture, it has been shown that total destruction to the limbus (with excimer laser ablation to a depth of 80 m) was followed by healing of the cornea from cells of the central epithelium, which in theory could not have been SC. Experiments on rabbits, where the limbus was totally destroyed, showed survival of the central epithelium, which, however, failed to regenerate when it too was mechanically removed.31 In recent years, successful ocular surface reconstruction has been carried out using ex vivo expanded sheets of limbal epithelial cells without substrate33 or with fibrin or
861
Ophthalmology Volume 116, Number 5, May 2009 amniotic membrane as substrate.34,35 This approach has yielded successful outcomes over a long follow-up period, leading some to postulate that the amniotic membrane substrate is capable of sustaining SC.36 Evidence for this is not convincing and does not reconcile with the widely accepted notion that SC retain their stemness within defined niches and tend to follow a path of differentiation when outside their niche. Any SC in ex vivo expanded sheets would therefore not remain so for long, yet the ocular surface remains healthy, again suggesting that it is the TAC and not SC in the sheets that account for the successful outcome. Alternatively, it is possible that within the TAC pool there are some cells that have characteristics closer to the parent SC than to the TAC progeny (we have termed these conceptual cells “transient cells”). The observations from the 8 eyes examined in this study and the other evidence presented support the hypothesis that physiologic corneal epithelial homeostasis can be maintained in the absence of clinically detectable limbal epithelial SC.
16. 17.
18. 19. 20. 21. 22.
References
23.
1. Tseng SC. Concept and application of limbal stem cells. Eye 1989;3:141–57. 2. Davanger M, Evensen A. Role of the pericorneal papillary structure in renewal of corneal epithelium. Nature 1971;229:560–1. 3. Townsend WM. The limbal palisades of Vogt. Trans Am Ophthalmol Soc 1991;89:721–56. 4. Matsuda M, Ubels JL, Edelhauser HF. A larger corneal epithelial wound closes at a faster rate. Invest Ophthalmol Vis Sci 1985;26:897–900. 5. Ebato B, Friend J, Thoft RA. Comparison of central and peripheral human corneal epithelium in tissue culture. Invest Ophthalmol Vis Sci 1987;28:1450 – 6. 6. Dua HS, Azuara-Blanco A. Limbal stem cells of the corneal epithelium. Surv Ophthalmol 2000;44:415–25. 7. Dua HS, Joseph A, Shanmuganathan VA, Jones RE. Stem cell differentiation and the effects of deficiency. Eye 2003;17: 877– 85. 8. Dua HS. Transplantation of limbal stem cells. In: Rienhard T, Larkin F, eds. Cornea and External Eye Disease. Berlin: Springer; 2006:35– 64. Essentials of Ophthalmology. Krieglstein GK, Weinreb RN, series eds. 9. Dua HS, Forrester JV. Clinical patterns of corneal epithelial wound healing. Am J Ophthalmol 1987;104:481–9. 10. Dua HS, Forrester JV. The corneoscleral limbus in human corneal epithelial wound healing. Am J Ophthalmol 1990;110:646–56. 11. Watanabe K, Nishida K, Yamato M, et al. Human limbal epithelium contains side population cells expressing the ATP-binding cassette transporter ABCG2. FEBS Lett 2004;565:6 –10. 12. Zhou S, Schuetz JD, Bunting KD, et al. The ABC transporter Bcrp1/ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat Med 2001;7:1028 –34. 13. Kawasaki S, Tanioka H, Yamasaki K, et al. Expression and tissue distribution of p63 isoforms in human ocular surface epithelia. Exp Eye Res 2006;82:293–9. 14. Shanmuganathan VA, Foster T, Kulkarni BB, et al. Morphological characteristics of the limbal epithelial crypt. Br J Ophthalmol 2007;91:514 –9. 15. Yeung AM, Schlotzer-Schrehardt U, Kulkarni B, et al. Limbal epithelial crypt: a model for corneal epithelial maintenance
24.
862
25. 26. 27. 28. 29. 30. 31. 32.
33. 34. 35.
36.
and novel limbal regional variations. Arch Ophthalmol 2008;126:665–9. Rosendaal M, Green CR, Rahman A, Morgan D. Up-regulation of the connexin43⫹ gap junction network in haemopoietic tissue before the growth of stem cells. J Cell Sci 1994;107:29–37. Cancelas JA, Koevoet WL, de Koning AE, et al. Connexin-43 gap junctions are involved in multiconnexin-expressing stromal support of hemopoietic progenitors and stem cells. Blood 2000;96:498 –505. Thoft RA, Friend J. The X, Y, Z hypothesis of corneal epithelial maintenance [letter]. Invest Ophthalmol Vis Sci 1983;24:1442–3. Lopez-Garcia JS, Rivas Jara L, Garcia-Lozano I, Murube J. Histopathologic limbus evolution after alkaline burns. Cornea 2007;26:1043– 8. Wagoner MD. Chemical injuries of the eye: current concepts in pathophysiology and therapy. Surv Ophthalmol 1997;41:275–313. Kinoshita S, Kiorpes TC, Friend J, Thoft RA. Limbal epithelium in ocular surface wound healing. Invest Ophthalmol Vis Sci 1982;23:73– 80. Holland EJ, Schwartz GS. The evolution of epithelial transplantation for severe ocular surface disease and a proposed classification system. Cornea 1996;15:549 –56. Patel DV, Sherwin T, McGhee CN. Laser scanning in vivo confocal microscopy of the normal human corneoscleral limbus. Invest Ophthalmol Vis Sci 2006;47:2823–7. Guthoff RF, Baudouin C, Stave J. Atlas of Confocal Laser Scanning In-vivo Microscopy in Ophthalmology. Berlin: Springer; 2006:31–52. Mastropasqua L, Nubile M, Lanzini M, et al. Epithelial dendritic cell distribution in normal and inflamed human cornea: in vivo confocal microscopy study. Am J Ophthalmol 2006;142:736–44. Dua HS, Gomes JA, Singh A. Corneal epithelial wound healing. Br J Ophthalmol 1994;78:401– 8. Dua HS. The conjunctiva in corneal epithelial wound healing. Br J Ophthalmol 1998;82:1407–11. Chen JJ, Tseng SC. Corneal epithelial wound healing in partial limbal deficiency. Invest Ophthalmol Vis Sci 1990;31:1301–14. Chen JJ, Tseng SC. Abnormal corneal epithelial wound healing in partial-thickness removal of limbal epithelium. Invest Ophthalmol Vis Sci 1991;32:2219 –33. Thoft RA, Friend J, Murphy HS. Ocular surface epithelium and corneal vascularization in rabbits. I. The role of wounding. Invest Ophthalmol Vis Sci 1979;18:85–92. Huang AJ, Tseng SC. Corneal epithelial wound healing in the absence of limbal epithelium. Invest Ophthalmol Vis Sci 1991;32:96 –105. Chang CY, Green CR, McGhee CN, Sherwin T. Acute wound healing in the human central corneal epithelium appears to be independent of limbal stem cell influence. Invest Ophthalmol Vis Sci 2008;49:5279 – 86. Daya SM, Watson A, Sharpe JR, et al. Outcomes and DNA analysis of ex vivo expanded stem cell allograft for ocular surface reconstruction. Ophthalmology 2005;112:470 –7. Ishino Y, Sano Y, Nakamura T, et al. Amniotic membrane as a carrier for cultivated human corneal endothelial cell transplantation. Invest Ophthalmol Vis Sci 2004;45:800 – 6. Rama P, Bonini S, Lambiase A, et al. Autologous fibrincultured limbal stem cells permanently restore the corneal surface of patients with total limbal stem cell deficiency. Transplantation 2001;72:1478 – 85. Grueterich M, Espana EM, Tseng SC. Modulation of keratin and connexin expression in limbal epithelium expanded on denuded amniotic membrane with and without a 3T3 fibroblast feeder layer. Invest Ophthalmol Vis Sci 2003;44: 4230 – 6.
Dua et al 䡠 The Limbus in Corneal Epithelial Maintenance
Footnotes and Financial Disclosures Originally received: July 14, 2008. Final revision: November 7, 2008. Accepted: December 4, 2008.
The authors have no proprietary or commercial interest in any materials discussed in this article. Manuscript no. 2008-842.
1
Division of Ophthalmology and Visual Sciences, University of Nottingham, England. 2
Research Institute of Ophthalmology, Cairo, Egypt. Financial Disclosure(s):
Correspondence: Harminder S. Dua, Division of Ophthalmology and Visual Sciences, B Floor, Eye ENT Centre, Queens Medical Centre, University Hospital, Derby Road, Nottingham. NG7 2UH, UK. E-mail: Harminder.dua@ nottingham.ac.uk.
863