Keratocytes, the major cell type occupying the corneal stroma, typically lie between stromal lamellae and occupy 2% to 5% of the stromal volume.1 Although keratocytes may share numerous cytological and functional characteristics with fibroblasts, research into the biology of this cell type has been minimal. Embryologically, keratocytes appear to be derived from neural crest-derived mesenchymal cells.2 During ocular development, the primary corneal stroma is invaded by keratocytes, which synthesize and secrete many of the stromal constituents, including collagene, proteoglycans (glycosaminoglycans), and glycoproteins.3 The corneal stroma does not become transparent until after the keratocytes have populated the primary stroma. In the avian cornea, corneal transparency appears to correlate with the degree of sulfation of specific stromal glycosaminoglycans which are produced by keratocytes.4 Furthermore, the relative degree of glycosaminoglycan sulfation may be controlled by an unidentified mechanism, which is dependent on keratocyte density.5
Studies from a number of laboratories suggest that the primary function of the keratocyte in the mature cornea is to maintain homeostasis of the corneal stroma by regulating the synthesis and turnover of coilagens, proteoglycans, glycosaminoglycans, and neutral proteases.6 The keratocyte secretes collagen fibrils, which group together to form fibers. These fibers are embedded within a dense coat of glycosaminoglycan and aggregate to form larger bundles of lamellae. Maintenance of this array of tightly-spaced collagen fibers is necessary for corneal transparency.7 The major mechanism for turnover of these corneal macromolecules appears to be extracellular degradation with subsequent phagocytosis and intracellular digestion by keratocytes.8 For example, the half-life of collagen turnover in normal corneas of young rabbits has been estimated to be 24 to 58 hours, although some fibrils persist substantially longer.9
In several studies, it has been demonstrated that the return of corneal clarity following cryorefractive surgery correlates with repopulation of the graft tissue by keratocytes.10"12 The restoration of corneal clarity is most likely related to remodeling of the extracellular matrix by keratocytes which have repopulated the graft. Our laboratory has developed an animal model of lamellar keratoplasty which offers an opportunity to study keratocyte repopulation of the graft stroma following freeze-injury.13 In the absence of a cryopreservative, frozen corneal tissue has a keratocyte survival of 0% to 5%.14 Based on histologic examination of corneas following freeze injury, keratocytes become "activated" at the wound margin of the lamellar keratoplasty incision and migrate into the gran within 3 days.12 However, because of the problems with distinguishing dead from viable keratocytes in routinely stained light microscopic sections, it is difficult to determine accurately the origin and migration rates of keratocytes located within the lamellar graft.12
We describe herein a novel technique which was developed to document the time course of keratocyte migration into the stromal matrix of corneal grafts following freeze injury. The fluorochrome 4', 6diamidino-2-phenylindole (DAPI), a fluorescent compound which intercalates specifically into adeninethymine base pairs of DNA,15 has been utilized to label the nuclei of keratocytes in a select region of the corneal lamellar bed prior to replacement of the anterior lamellar graft. We monitored DAPI-labeled keratocyte migration into the lamellar graft with the use of epifluorescence light microscopic observation of frozen corneal sections.
MATERIALS AND METHODS
Animal Surgery/DAPI Administration
New Zealand White rabbits (Doe Valley Farms, Benton ville, Ark) of random sex and weighing 3.0 to 3.5 kg, were used in this study. Animals were treated in accordance with the ARVO Kesolution on the Use of Animals in Research, the "Guide for Care and Use of Laboratory Animals" (NIH Publication No. 86-23, revised 1985) and the guidelines established by the St Louis University Animal Care and Use Committee. The rabbits were anesthetized with an intramuscular injection of xylazine (7 mg/kg) and ketamine (40 mg/kg). The operative eye received a retrobulbar injection of 1% lidocaine (neural block) prior to lamellar keratoplasty surgery. A radial abrasion was made in the corneal epithelium to aid in proper alignment of the lenticule during surgery.
Using a pediatric Barraquer microkeratome (Steinway Instruments, San Diego, Calif), we removed a lamellar section (lenticule) of the anterior cornea, comprising approximately 80% of the corneal thickness.13 For DAPI administration, a Wheaton micropipet was used to dispense 5 mL of a 5 µL/mL solution of DAPI (Sigma Chemical Co, St Louis, Mo) in balanced salt solution (BSS, Alcon Surgical, Fort Worth, Tex) along the microkeratome wound edge of the host bed stroma at the 6 o'clock position. The solution was left in contact with the stroma for 5 minutes. The excess dye was absorbed with a micro eye sponge and the stromal surface rinsed thoroughly with BSS.
The lenticule was frozen, without the use of a cryoprotectant, on a Barraquer cryolathe (Steinway Instruments, San Diego, Calif) for approximately 90 seconds, and then thawed in BSS at 370C. The lenticule was subsequently sutured onto the host with a 16-hite, running 10-0 nylon suture. Gentamicin sulfate (12 mg) and dexamethasone (1.2 mg) were injected into the subconjunctival tissue postoperatively. A single 4-0 silk suture was used to close the lids for 3 days, to aid reepithelialization and maintain normal hydration of the lenticule. Slit-lamp examinations with photographs and ultrasonic pachometry were performed on days 3, 7, 14, and 21 following lamellar keratoplasty surgery.
Tissue Collection and Preparation
Animals were sedated with acepromazine maleate (2 mg/kg) on the operative day or at postsurgical days 3, 7, 14, or 21, and euthanized with Sleepaway solution (sodium pentobarbital in 10% isopropyl alcohol and 20% propylene glycol; Fort Dodge Labs, Fort Dodge, Iowa). The anterior chamber was perfused with 4% formaldehyde (freshly generated from paraformaldehyde) in 100 mmol/L sodium cacodylate buffer (pH 7.4) for 15 minutes. Corneas were then excised with a 2-millimeter scleral rim, immersed in fresh, room temperature fixative for 2 hours, and subsequently rinsed in three changes of 100 mmol/L sodium cacodylate buffer (pH 7.4) for a total of 1 hour. The corneas were infiltrated overnight in acrylamide embedding solution (Bethesda Research Laboratories, Gaithersburg, Md), at 4°C, and subsequently embedded in acrylamide.16 Acrylamide-embedded tissue was placed in Tissue Tek O.C.T. (Miles Scientific, Naperville, 111) and frozen in liquid nitrogen. Sections, approximately 6 to 8 µ thick, were cut on an AO Scientific Instruments HistoStat Cryostat (Buffalo, NY) at -2O0C. Immunomount (Shandon, Pittsburgh, Pa) was used as a mounting media for coverslipping. The slides were observed immediately and photographed with an Olympus BH-2 fluorescence light microscope.
Figure 1: Fluorescence light micrograph depicting DAPIlabeled cell nuclei from grafted animals killed within 1 day following DAPI application. Keratocytes within the graft are not labeled (dashed line represents host-graft interface), indicating that DAPI does not diffuse from the host stroma into the graft stroma following replacement of the graft. Arrowheads indicate autofluorescing keratocyte nuclei.
Clinical examinations were not performed on rabbits that were killed immediately following surgery. The first postoperative clinical exam was conducted on postoperative day 3. At that time, the lenticules were approximately 20% reepithelialized and exhibited moderate stromal haze. By postoperative day 7, the grafts were 80% reepithelialized. Overall, the eyes were quiet, with no apparent reaction to the fluorochrome. By day 14, the corneal grafts were completely reepithelialized and no stromal haze was observed.
Fluorescence fight microscopic examination of corneal sections from immediate postoperative corneas revealed DAPI-labeled stromal keratocytes within the host stroma, epithelial cells along the microkeratome wound margin, and endothelial cells in the region of DAPI application. No DAPI-labeled keratocyte nuclei were observed in the grafted lenticules at this time point (Fig 1). This indicates that the initial rinsing step following DAPI aolministration sufficiently removed unincorporated dye, and that DAPI did not diffuse into the graft stroma and label DNA within nonviable keratocytes.
In corneas from postoperative day 7 animals, DAPI-labeled keratocytes were prominent in the stromal bed underlying the graft. DAPI-labeled keratocytes were also observed within the graft, and were highly concentrated along the graft-host interface, especially at the peripheral margin and within the central region of the graft. Few keratocytes, however, were observed within the anterior-most portion of the lenticule. DAPI-labeled epithelial cell nuclei were also observed in the single layer of newly-generated epithelium, which nearly covered the entire grafted lenticule. The fluorescent signal associated with epithelial cells was most intense at the wound edge and decreased in intensity toward the center of the graft, due to mitosis-associated diminution of DAPI concentration in these cells (Fig 2).
Although DAPI was originally developed as a chemotherapeutic drug for the treatment of trypanosomiasis,17 it has been subsequently utilized in a broad spectrum of applications, including: 1) the detection of mycoplasma and viruses in cell culture,14 2) as a probe for cytogenetic analyses,18 and 3) as a cell marker for migrating neural cells. 19 DAPI has a strong binding affinity for both adeninethymine (A-T) and guanine-cytosine (G-C) base pairs of DNA, and demonstrates its highest binding affinity for A-T base pairs.20 Once bound, DAPI becomes highly fluorescent, when excited at wavelengths of 365 nm, and emits a blue-white light at 450 nm.21 It is this ability to fluorescently label cellular DNA, without disrupting DNA or cellular function, that makes DAPI a useful tool for studies of keratocyte migration into corneal grafts.
Our previous histological investigations of corneal grafts following lamellar keratoplasty failed to identify the source of keratocytes present within the grafts, since it was difficult to distinguish dead keratocytes remaining in the frozen lenticule from "activated" keratocytes, which had migrated into the lenticule subsequent to surger}'.12 In the present study, the application of DAPI allowed us to label keratocytes in a select region of the host cornea and to follow their (and their progeny's) migration into lamellar grafts, over time. Since DAPI remains bound to DNA through multiple cell divisions, the progeny of a specific population of cells can be visualized and their migration followed.
Figure 2: Keratocytes (arrows) throughout the host cornea are DAPI-labeled (Zone 5, FIg 2D)1 as are the epithelium (EP) and endothelium (EN). Nofe keratocytes Juxtaposed to the microkeratome wound (dotted line) at the periphery of the graft (Zones 4 and 5, Fig 2D and 2C) before migration into the graft (arrowheads; Zones 3 and 2, Fig 2B and 2A).
These studies suggest that DAPI can be utilized in vivo to further characterize the migratory pattern of keratocytes as they repopulate a lamellar graft following keratoplasty. Furthermore, the use of this probe will allow us to determine whether keratocytes, which repopulate lenticules, originate from existing keratocytes that migrate into the graft, or from daughter cells derived from host keratocytes, or both. Although the studies described herein employed cryostat sections of DAPI-labeled corneas, it is anticipated that, with the development of fluorescence confocal microscopy, one should be able to study keratocyte dynamics in vivo. The technique described herein, in conjunction with confocal microscopy, should enable us to determine conclusively the time course and migration pattern of keratocytes as they repopulate corneal grafts following cryorefractive lamellar keratoplasty.
1. Klyce SD, Beuerman RW. Structure and function of the cornea. In: Kaufman HE, Barron BA, McDonald MB, Waltman SR, eds. The Cornea. New York, NY: Churchill Livingstone Ine; 1988:3.
2. Johnston MC. Origins of avian ocular and periocular tissues. Exp Eye Res. 1979;29:27-43.
3. Hay ED. Matrix-cytoskeletal interactions in the developing eye. J Cell Biochem. 1985;27:143-156.
4. Hart GW. Biosynthesis of glycoaminoglycans during corneal development. J Biol Chem. 1976;251:6513-6521.
5. Coleman JR, Herrmann H, Bess B. Biosynthesis of collagen and non-collagen protein during development of the chick cornea. J Cell Biol. 1965;25:69-78.
6. Mishima H, Yasumoto K, Nishida T, Otori T. Fibronectin enhances the phagocytic activity of cultured rabbit keratocytes. Invest Ophthalmol Vis Sci. 1987;28:1521-1526.
7. Maurice DM. The cornea and sclera. In: Duke-Elder S, ed. Physiology of the Eye and Vision. St Louis, Mo: CV Mosby Co; 1968:337.
8. Lande MA, Birk DE, Nagpal ML, Rader RL. Phagocytic properties of human keratocyte cultures. Invest Ophthalmol Vis Sci. 1981;20:481-489.
9. Lee RE, Davison PF. Collagen composition and turnover in ocular tissues of the rabbit. Exp Eye Res. 1981;32:737-745.
10. Lee TJ, Wan WL, Kash RL, Kratz KL, Schanzlin DJ. Keratocyte survival following a controlled-rate freeze. Invest Ophthalmol Vis Sci. 1985;26:1210-1215.
11. Zavala EY, Binder PS, Deg JK, Baumgartner SD. Refractive keratoplasty: lathing and cryopreservation. CLAO J. 1985;11:155-163.
12. Schanzlin DJ, Jester JV, Kay ED. Cryolathe corneal injury. Cornea. 1983;2:57-68.
13. Robin JB, Regis-Pacheco LF, Ward D, Schanzlin DJ. Microkeratome lamellar corneal resection in the rabbit: a model for experimental study of the effects of Barraquer cryolathing techniques. CLAO J. 1985;11:79-81.
14. Kownacki JJ, Kratz KL, Huff JW, Schanzlin DJ. Long term storage of frozen lenticles for cryorefractive surgery. Invest Ophthalmol Vis Sci. 1989;30(suppl):187.
15. Russell WC. A simple cytochemical technique for demonstration of DNA in cells infected with mycoplasmas and viruses. Nature. 1975;253:461-462.
16. Johnson LV, Blanks JC. Application of acrylamide as an embedding medium in studies of lectin and antibody binding in the vertebrate retina. CurrEye Res. 1984;3:969-974.
17. Dann O, Bergen G, Demant E, VoIz G. Trypanocide diamidine des 2-phenyl-benzofurans, 2-phenyl-indens und 2-phenylindols. Justus Liebigs Ann Chem. 1971;749:68-89.
18. Koller U, Haas OA, Kornmuller R, Majdic 0, Gadner H, Knapp W. Immunfluoreszena und immunzytochemische Farbemethoden zur simultanen zytogenetischen und phänotypischen Charakterisierung mitotischer Zellen. Wien Klin Wochenschr. 1989;101:111-117.
19. McConnell SK. Fates of visual cortical neurons in the ferret after isochronic and heterochronic transplantation. J Neurosa. 1988;8:945-974.
20. Lin MS, Comings DE, Alfi OS. Optical studies of the interaction of 4', 6-diamidino-2-phenyhndole with DNA and metaphase chromosomes. Chromosoma. 1977;60:15-25.
21. Williamson DH, Fennell DJ. The use of fluorescent DNAbinding agent for detecting and separating yeast mitochondrial DNA. In: Prescott D, ed. Methods in Cell Biology. New York, NY: Academic; 1974:335.