|
|
||||||||
1 From the Cooperative Research Centre for Eye Research and Technology, University of New South Wales, Sydney; 2 Commonwealth Scientific Industrial Research Organization (CSIRO) Molecular Science, Sydney, New South Wales; the 3 Department of Clinical Ophthalmology, Sydney Eye Hospital, New South Wales; and 4 CSIRO Molecular Science, Clayton, Victoria, Australia.
| Abstract |
|---|
|
|
|---|
METHODS. A perfluoropolyether-based polymer was cast into lenticules that were coated with collagen I and implanted in four feline corneas. Epithelial growth onto the lenticules was monitored clinically for 6 weeks, after which time the animals were killed, and three corneas were evaluated histologically. Immunohistochemistry was used to identify proteins associated with the formation of a basement membrane (laminin) and adhesion complexes (bullous pemphigoid antigen and collagen VII). Electron microscopy was used to examine the tissuepolymer interface for evidence of the assembly of these adhesive structures.
RESULTS. Postoperative epithelial growth began on days 2 to 3, and lenticules were fully epithelialized by days 5 to 9. Lenticules were clinically well tolerated and histology showed epithelium consisting of multiple layers adherent to the lenticules surface. Laminin, bullous pemphigoid antigen and collagen VII were identified at the tissuepolymer interface using immunohistochemistry. Ultrastructural examination showed evidence of assembly of these proteins into a recognizable basement membrane and hemidesmosomal plaques.
CONCLUSIONS. A perfluoropolyether-based polymer coated with collagen I was implanted in the feline cornea and supported epithelial growth that showed signs of persistent adhesion, both clinically and histologically. This polymer shows potential for ophthalmic applications that require sustained epithelialization.
| Introduction |
|---|
|
|
|---|
Previous devices designed for corneal applications that require epithelialization have used polymers of either biologic or synthetic origin. The biologically derived materials have shown some promising results in vivo, although their success has been limited by lenticule remodeling and epithelial tissue abnormalities.9 10 11 12 13 Epikeratophakia lenticules of donor stromal tissue supported epithelial growth and were retained in the primate cornea for 22 to 25 months despite a concomitant build up of cellular debris, pigmented cells, and keratocytes around the grafted tissue.14 Results from a nationwide epikeratophakia study,9 in which human patients received lyophilized donor lenticules to correct aphakia, myopia, and keratoconus, identified persistent epithelial defects as an early problem. Abnormal morphologic characteristics were reported in epithelial tissue covering donor epikeratophakia lenticules in human eyes for up to 16 months after surgery.10 Histologic evaluation of donor human epikeratoplasty lenticules that maintained epithelial cover for 8 months in patients revealed fewer hemidesmosomes and basement membrane defects at the epitheliumlenticule interface.11 Epikeratophakia lenticules made from purified human placental type IV collagen (Laboratoires Domilens, St. Just Chaleyssin, France) maintained epithelial cover when tested in primate corneas over a 6-month period.3 Similar lenticules showed good clinical performance when used in an epikeratoplasty procedure in rhesus monkeys for up to 30 months, although postmortem histology revealed epithelial thinning on the lenticules surface and proteolytic degradation of the collagen lenticule.12 More recently, a two-part full-thickness corneal graft made from solubilized rabbit dermal collagen was tested in rabbit corneas.13 Although partial epithelialization of the fibrous peripheral zone was achieved in that study, the central optical zone of these grafts failed to support any epithelial growth after 24 days. This response was attributed to the chemistry of the modified collagen. Biologically derived polymers have also been tested in combination with synthetic cross-linkers. Collagen epikeratoplasty lenticules that were polymerized in situ in primates maintained multilayered epithelial cover for more than 1 year with evidence of neovascularization and remodeling of the cross-linked collagen during this time.15
Synthetic polymers and copolymers offer certain advantages in the making of devices for epikeratoplasty (epikeratophakia) and keratoprosthesis applications. These include the reduced risk of transmission of infectious agents, an unlimited supply, improved optical properties and avoidance of postoperative remodeling.4 The design of polymers for these applications has been directed primarily at anchorage of the peripheral portion of the device in the corneal stroma. Results from in vivo studies show some success in terms of this aspect of the design, although optimization of a polymer to support epithelial growth over the central optical portion of these devices remains a challenge.16 17 18 A high-water-content contact lens material, lidofilcon A, has been tested as a synthetic epikeratophakia lenticule in rabbit corneas, but the lenses failed to support epithelial growth onto the polymer surface.3 A plasma-modified poly(vinyl alcohol) material showed only partial epithelial cover when implanted into rabbits as the optical component of a synthetic cornea,16 even though it supported epithelial growth in vitro during a 21-day period.19 Recent clinical trials of two different designs for a keratoprosthesis (KPro), the Chirila KPro17 (Lions Eye Institute, Perth, Australia) and the Biokpro II,18 (Department of Ophthalmology, Hotel-Dieu de Paris, Paris, France) have demonstrated that a full-thickness synthetic device can be maintained in the cornea for some months. These KPro devices were successfully anchored into the cornea by fibrovascular ingrowth into the peripheral skirt, but the polymers used in the optical core did not epithelialize in either case. Both groups17 18 have acknowledged that the long-term stability of such KPros would be improved by the development of a polymer that supported epithelial growth.
We are interested in the design of a synthetic polymer for use as a corneal onlay in the correction of corneal trauma and/or scarring, keratoconus, and high-level refractive errors. Previous work in our group has confirmed the requirement for porosity in a synthetic lenticule implanted in the cornea.20 In addition to this, collagen I coated on the surface of a microporous synthetic material was found to promote epithelialization in vivo.21 The stable epithelial growth observed on those lenticules was attributed to the formation of adhesive structures (basement membrane and hemidesmosomes) at the epitheliumlenticule interface.22 Concurrent with these findings, a perfluoropolyether (PFPE)-based polymer has been tested for use as a synthetic onlay material. The high oxygen permeability of this polymer type led to its original testing in the ophthalmic field as a contact lens material.23 Other characteristics of this family of polymers that are suited to ophthalmic applications include transparency, low refractive index, and high oxidative and thermal stability. PFPEs are hydrophobic and yet have been shown to support the growth of corneal epithelial cells in vitro.24 Recently, microporous PFPE-based polymers have been developed25 26 that allow for the diffusion of nutrients including glucose and high-molecular-weight proteins such as albumin. In the present study, we report the performance of lenticules made from porous PFPE that were coated with collagen I and implanted in the anterior stroma of a feline model. The performance of these implants was monitored clinically for a 6-week period, at which time the epitheliumlenticule interface was examined histologically for evidence of structures associated with stable epithelial growth.
| Methods |
|---|
|
|
|---|
![]() | (1) |
Briefly, the PFPE-dimethacrylate was dispersed in a mixture of alcohol, surfactant and 0.3% (wt/wt) of the photoinitiator Darocur 1173 (Ciba Speciality Chemicals, Basel, Switzerland). After mixing, the homogeneous PFPE mixture was dispensed in polypropylene lenticule molds (CibaVision, Duluth, GA) and polymerized over 1 mW/cm-2 UV lamps. The resultant porous PFPE lenticules were extracted in the fluorinated solvent PF-5060 (3M, St. Paul, MN), isopropyl acetate, and isopropanol to ensure the removal of residual surfactant and any unreacted PFPE.
Lenticule Design
The PFPE lenticules used in this study were 4 mm in diameter and
were planar, with an overall thickness of 80 µm tapering to 70 µm
with square edges. The lenticules had a water content of 52%,
representing the relative mass of water required to fill the porous
voids within the polymer and providing a measure of the degree of
porosity. The permeability of the lenticules to glucose, inulin, and
human serum albumin was measured to be 47, 14, and 3.6 x
10-5 cm/sec, respectively.
Surface Coating
A thin layer of collagen I (approximately 510 nm thick) was
covalently immobilized on the anterior surface of each lenticule by
reductive amination that involved functionalization of the surfaces by
exposure to acetaldehyde monomer in a radio frequency glow discharge
(RFGD).27
Immediately after RFGD treatment, lenticules
were placed in a solution of 50 µg/ml collagen I (Vitrogen; Collagen
Corporation, Palo Alto, CA) in phosphate-buffered saline (pH 7.4) on
ice and an excess of sodium cyanoborahydride added to act as a reducing
agent.28
Surgery
The animal research in this study was approved by the Animal Care
and Ethics Committee of the University of New South Wales. All
procedures were performed in compliance with the ARVO Statement for the
Use of Animals in Ophthalmic and Vision Research. The cats (n
= 4) used in this study were 3 to 5 years of age, were of
different sexes, and weighed 4 to 5.5 kg. The surgical procedure used
has been previously described29
and is shown in Figure 1 . Briefly, animals were anesthetized using an intramuscular injection of
10 to 15 mg/kg bodyweight ketamine and 1 mg/kg bodyweight xylazine to a
depth of stage 3, plane 2.21
A beaver blade was used to
remove the epithelium of the central cornea in a 4-mm diameter area. A
circular keratotomy 150 µm deep was made in the débrided area
using a 2-mm diameter trephine, and the stromal lamella within the
trephined area was removed. A sharp corneal dissector was used to make
a circular pocket 1 mm wide toward the limbus at the base of the
keratotomy. The superior portion of the pocket was further extended
until the superior edge of the pocket was approximately 5 mm from the
pockets mouth. A 5-mm arcuate incision was made along the superior
edge of the pocket that was used to deliver a 4-mm diameter lenticule
using a spatula. The lenticule was tucked into the circular stromal
pocket, and the superior edge incision secured with sutures. Clinical
examination of the cornea was performed daily using a slit lamp
biomicroscope and fluorescein to assess the speed and extent of
epithelial growth on the lenticules. On termination, corneas were
processed for evaluation by either immunohistochemistry or transmission
electron microscopy (TEM).
|
Ultrastructure
TEM was used to assess the assembly of adhesive structures
(basement membrane and hemidesmosomal plaques) at the
epitheliumlenticule interface. Tissue was blocked from the central
region of one feline cornea implanted with a collagen-coated PFPE
lenticule (cat 3). Samples were fixed in 2.5% (vol/vol) glutaraldehyde
and 2.5% paraformaldehyde in sodium cacodylate buffer (pH 7.2) with
1% sucrose and 1% calcium chloride for 3 hours and washed overnight
in buffer. Samples were postfixed in 1% osmium tetroxide supplemented
with 1.5% potassium ferrocyanide and stained en bloc with 2% aqueous
uranyl acetate. Samples were dehydrated through a graded series of
ethanol to propylene oxide, infiltrated, and embedded in Epon Araldite.
Sections 80 to 90 nm thick were stained en grid, using uranyl acetate
and Reynolds lead citrate, and were examined at 75 kV in a
transmission electron microscope (model 7100 Hitachi, Tokyo, Japan).
| Results |
|---|
|
|
|---|
|
|
|
| Discussion |
|---|
|
|
|---|
PFPE lenticules that were biostable, transparent, and permeable to nutrient flow were used in the present study. These lenticules were coated with collagen I, a biologic signal that we have previously shown to allow for rapid and sustained epithelialization in vivo when covalently immobilized onto the surface of a porous synthetic material.21 The PFPE lenticule was implanted in the anterior stroma of the feline cornea using a surgical model that exposed the anterior surface of the lenticule for epithelial overgrowth. Clinically, the collagen-coated PFPE-based polymer supported a steady rate of epithelial growth over the lenticules, with wound closure occurring between days 5 and 9. This was longer than the 3-day time before closure of an identical wound in the absence of a lenticule. The reason for the longer period is likely to be multifactorial and involve an interplay between the chemistry, permeability, and topography of the PFPE-based polymer and the biologic signal coated on the surface. However, three collagen-coated PFPE lenticules in the present study maintained stable epithelial cover for a period of 34 to 39 days, when the animals were killed for histologic evaluation.
Histology showed that the epithelium covering the PFPE lenticules comprised three to five cell layers; considerably less than the 10 to 12 cell layers seen in normal, intact feline corneal epithelium. This may have been a function of time and/or may have been related to an absence of proliferative cells on the lenticules surface. Despite this, the epithelial tissue covering the lenticules surface had some characteristics of intact tissue. Microvilli were present on the anterior surface of the epithelium and most likely supported the normal tear film observed in the clinical examinations. Desmosomes were present between the constituent cells and would account for the good overall integrity of this epithelial tissue. The cells of the basal layer, which were in direct contact with the lenticules surface, could be distinguished by the light staining nature of the cells rather than by their columnar morphology as in normal corneal epithelium. Immunohistochemistry results showed the polarized deposition of ECM proteins associated with the formation of the basement membrane and adhesion complexes at the tissuelenticule interface. Ultrastructurally, there was evidence of some assembly of this ECM at the interface, as exemplified by the presence of hemidesmosomal plaques along the basal cell membrane and fragments of basement membrane subjacent to this. The substantial accumulation of ECM material between the epithelial tissue and the lenticules surface appears to have been the result of continuing deposition of basement membrane and adhesion complex proteins by the epithelial cells. Our immunohistochemistry results showed almost linear staining for collagen VII at the tissuelenticule interface, although the assembly of this collagen VII into the cross-banded anchoring fibrils of the adhesion complexes was not observed ultrastructurally. This may have been a function of time, because the reformation of anchoring fibrils is known to be a slow process in some species.6 The accumulation of ECM material in the basement membrane zone observed ultrastructurally in our study was similar to that reported for by other investigators in slightly different circumstances.6 30 Massive duplication in the basement membrane zone was reported during the healing of a 7-mm diameter superficial keratectomy wound to the rabbit cornea examined over a 12-month period.6 Accumulations of basement membrane proteins were also identified in the unwounded corneas of diabetic humans, and this was thought to have decreased the penetration of anchoring fibrils into the anterior stroma and compromised epithelialstromal adhesion in those patients.30
The present study has demonstrated epithelialization of a synthetic material in vivo for the first time. PFPE-based lenticules, with a biologic signal on the anterior surface, supported stable epithelial growth and were clinically well tolerated when implanted in the feline cornea over a 6-week period. The epithelial tissue that covered the lenticules stratified, although at this early stage after implantation, the number of cell layers and the morphology of constituent cells was less than in normal feline corneal epithelium. Despite this, the epithelium showed signs of persistent adhesion, both clinically and histologically. Further refinements in the design of this PFPE-based polymer, the surface coating, and the surgical procedure used to implant the lenticules may optimize the performance of this polymer for ophthalmic applications that require sustained epithelialization.
| Acknowledgements |
|---|
| Footnotes |
|---|
Submitted for publication May 14, 1999; revised September 21 and December 20, 1999; accepted January 18, 2000.
Commercial relationships policy: C5 (MDME, RZX, MF, HC, GJB, GFM, HJG, JGS, DFS); N (MCM).
Corresponding author: Margaret D. M. Evans, CSIRO Molecular Science, PO Box 184, North Ryde, Sydney NSW 1670, Australia. meg.evans{at}molsci.csiro.au
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. Z. Xie, M. D. M. Evans, B. Bojarski, T. C. Hughes, G. Y. Chan, X. Nguyen, J. S. Wilkie, K. M. McLean, A. Vannas, and D. F. Sweeney Two-Year Preclinical Testing of Perfluoropolyether Polymer as a Corneal Inlay Invest. Ophthalmol. Vis. Sci., February 1, 2006; 47(2): 574 - 581. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. F. Sweeney, R. Z. Xie, M. D. M. Evans, A. Vannas, S. D. Tout, H. J. Griesser, G. Johnson, and J. G. Steele A Comparison of Biological Coatings for the Promotion of Corneal Epithelialization of Synthetic Surface In Vivo Invest. Ophthalmol. Vis. Sci., August 1, 2003; 44(8): 3301 - 3309. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. D. M. Evans, R. Z. Xie, M. Fabbri, B. Bojarski, H. Chaouk, J. S. Wilkie, K. M. McLean, H. Y. Cheng, A. Vannas, and D. F. Sweeney Progress in the Development of a Synthetic Corneal Onlay Invest. Ophthalmol. Vis. Sci., October 1, 2002; 43(10): 3196 - 3201. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |