Tissue Engineering for Reconstruction of the Corneal Epithelium

Published on 08/03/2015 by admin

Filed under Opthalmology

Last modified 08/03/2015

Print this page

This article have been viewed 3040 times

Tissue Engineering for Reconstruction of the Corneal Epithelium

Ursula Schlötzer-Schrehardt, Naresh Polisetti, Johannes Menzel-Severing and Friedrich E. Kruse

Introduction

Over the past several years tissue engineering has become a rapidly growing field of research with strong translational potential into clinical practice through application of adult stem cells (SC) and biomaterial sciences. It is believed that SC, with their inherent plasticity, could be utilized to regenerate the natural complexity present in native tissue, while providing factors required for lineage maintenance and differentiation. Thus, the basic strategy of tissue engineering is the construction of a biocompatible scaffold that in combination with SC and bioactive molecules replaces and regenerates damaged cells or tissues.¹

The cornea is composed of three layers, an outer stratified, rapidly regenerating epithelium, the underlying stroma, and an inner single-layered endothelium. Homeostasis of corneal epithelial cells is an important prerequisite not only for the integrity of the ocular surface, but also for corneal transparency and visual function (Fig. 43.1A,C). The continuous renewal of the corneal epithelium is provided by a population of adult SC located in the basal epithelium of the transitional zone between cornea and conjunctiva, the limbus (Fig. 43.1B).² Stem cell maintenance and function are controlled by various intrinsic and extrinsic factors provided by a unique local microenvironment or niche.³ Limbal stem cells (LSC) and their progeny reside within small clusters in the basal epithelium in close relationship with specific extracellular matrix components, stromal fibroblasts, blood vessels, and nerves providing increased levels of growth and survival factors. Limbal SC can divide both symmetrically to self-renew and asymmetrically to produce daughter transiently amplifying cells that migrate centripetally to populate the basal layer of the corneal epithelium and eventually become post-mitotic terminally differentiated epithelial cells (Fig. 43.1B). Limbal SC can be identified by positive expression of putative stem/progenitor cell markers, including ABCG2, ΔNp63α, Bmi1, CEBPδ, OCT4, Lgr5, cytokeratin K15, and N-Cadherin, and the absence of corneal epithelial differentiation markers, such as cytokeratins K3 and K12.⁴

Figure 43.1 The ocular surface. (A) Light microscopic image of a section through a human cornea showing its composition of a stratified surface epithelium, stroma, and single-layered endothelium (periodic acid–Schiff staining). (B) Schematic concept of limbal location of corneal stem cells and transient amplifying cells. (C) Clinical appearance of a healthy human cornea (arrows indicate corneal limbus). (D) Clinical appearance of a patient with severe limbal stem cell insufficiency showing conjunctivalization and vascularization of the corneal surface. (Fig. 1B: Reproduced from: Schlötzer-Schrehardt U, Kruse FE. Identification and characterization of limbal stem cells. Exp Eye Res 2005;81:247–64. With permission from Elsevier.)

Dysfunction or loss of the LSC population in combination with an impairment of their niche, due to different inherited or acquired conditions may result in partial or total limbal stem cell deficiency (LSCD), which has severe consequences for ocular surface integrity and visual function (Fig. 43.1D). A range of inflammatory eye conditions (e.g. Stevens – Johnson syndrome, mucous membrane pemphigoid), degenerative processes (e.g. recurrent pterygia), hereditary causes (e.g. congenital aniridia), and chemical or thermal trauma, can lead to chronic epithelial healing defects associated with conjunctival epithelial ingrowth, neovascularization, inflammation, ulceration and scarring, and ultimately to functional blindness.^5,6 These conditions are contraindications for conventional corneal transplantation, i.e., a full-thickness corneal graft, because the ocular surface will not re-epithelialize properly without replenishment of the depleted stem cell pool. Therefore, corneal repair may be possible only by addressing the epithelial disorder through layer-specific stem cell-based approaches to corneal surface reconstruction. Reconstruction of the stratified ocular surface epithelium, particularly in patients with bilateral LSCD, is critical to restore vision and represents one of the most challenging problems in clinical ophthalmology.

Current tissue engineering approaches for reconstruction of the corneal epithelium utilize adult SC, usually LSC derived from a small tissue biopsy from either the patient (autologous) or a donor (allogeneic), followed by their ex vivo expansion in culture on a natural scaffold, usually human amniotic membrane, and generation of three-dimensional epithelial constructs for transplantation.⁷ Differences in culture techniques include the use of explant or single-cell suspension systems, the presence or absence of a mouse 3T3 fibroblast feeder layer, the type of substrate, and the optional application of airlifting to promote epithelial differentiation and stratification. In their attempt to improve and standardize this therapeutic approach, researchers have focused on the optimization of culture conditions in order to replicate the in vivo stem cell niche and to preserve stemness, on the introduction of safer culture procedures to avoid the use of xenobiotics, on the exploration of alternative autologous stem cell sources for treatment of bilateral surface disorders, and on the evaluation of novel scaffolds to aid SC expansion and enhance transplantation efficacy. The challenges to create a tissue-engineered corneal epithelial equivalent include generating a biocompatible, mechanically stable, and optically transparent construct, which supports SC growth and maintenance in culture and after transplantation.

Stem Cell Sources for Corneal Epithelial Reconstruction

Limbal SC have been extensively investigated for their ex vivo culture and subsequent transplantation efficacy in clinical application. In addition, cultured oral mucosal epithelial SC have been used for the treatment of LSCD in patients with bilateral ocular surface disease.⁸ Cell types other than LSC or oral mucosal epithelial SC, including conjunctival epithelial SC,^9–11 hair follicle SC,^12,13 mesenchymal SC,^14–16 immature dental pulp SC,¹⁷ umbilical cord SC,¹⁸ and embryonic SC¹⁹ have been additionally proposed as alternative autologous SC sources for tissue engineering and already provided promising results in vitro and in vivo preclinical animal studies. It is clearly desirable to use autologous cells for tissue engineering, as this avoids the risk of allogeneic immune rejection and the need for immunosuppression.

Conditions mimicking the in vivo LSC niche, such as limbal fibroblast conditioned media, have been used to induce a corneal epithelial-like phenotype in constructs derived from non-corneal SC sources.12,16,20 The suitability of epithelial constructs as a corneal epithelium replacement is usually evaluated by expression of typical corneal epithelial differentiation markers (cytokeratins K3 and K12) and the ability to reconstruct the damaged ocular surface by providing sufficient mechanical stability and optical transparency.

Limbal Epithelial Stem/Progenitor Cells

A major tissue engineering strategy for reconstruction of the corneal epithelium is based on ex vivo expansion of autologous SC taken from the limbus of the patient’s contralateral healthy eye in unilaterally affected patients.21,22 This cultured limbal epithelial transplantation (CLET) appears to be a promising treatment modality for LSCD with an overall success rate of 75% up to 119 months follow-up.^23–26 The most widely used expansion method is the explant culture system, in which a small limbal biopsy (1–2 mm²) is placed on a carrier, usually amniotic membrane, and the limbal epithelial cells then migrate out of the biopsy and proliferate to form an epithelial sheet (Fig. 43.2). However, outgrowths from human limbal explants show a rapid decline in proliferative potential,²⁷ and use of limbal epithelial cell suspensions appears to increase the proportion of SC in the culture system.²⁸

Figure 43.2 Ex vivo expansion of human limbal stem cells on human amniotic membrane. (A) Harvest of a 1×2-mm biopsy from the superior limbus of the healthy eye from a patient with unilateral limbal stem cell deficiency. (B) Explant culture of a limbal biopsy (arrow) on human amniotic membrane. (C) Outgrowth of an epithelial monolayer from the explant. (D) Multilayered epithelial sheet growing on amniotic membrane after 2 weeks in culture.

An essential prerequisite of epithelial grafts for long-term restoration of the ocular surface is the presence of an adequate number of stem cells.4,29,30 Since the pioneering work by Rheinwald and Green,³¹ studies have confirmed that long-term survival of epithelial SC is possible if co-cultured with embryonic fibroblast feeder cells, which appear to reproduce aspects of the SC niche in vitro. This strategy led to the development of a culture system that involves enrichment of LSC by clonal growth on an inactivated mouse 3T3 fibroblast feeder layer before being seeded onto transplantable carriers to produce epithelial sheets.^22,32 Preoperative demonstration of ΔNp63-positive cells within epithelial grafts has been considered as a means of quality control for transplantation.^25,33–35

Over the past years there have been various modifications of culture techniques. Important parameters that influence cell proliferation and differentiation include growth factors, serum and calcium concentrations, type of matrix coating, and substrate stiffness as indicated by the elastic modulus.32,36 Culture of epithelial cells at the air – liquid interface (airlifting) has addressed the in situ environment of the cornea thereby promoting epithelial cell differentiation and stratification.³⁷ Hypoxic conditions coupled with air exposure have been shown to further enhance LSC proliferation.³⁸ Moreover, co-culture of LSC and adherent mesenchymal niche cells expressing embryonic SC markers has been reported to promote the preservation of a LSC phenotype.³⁹ Recent approaches aim at establishing genetically modified cells, such as telomerase immortalized corneal epithelial cells,⁴⁰ and at replacing potentially hazardous xenobiotics, such as fetal calf serum and murine feeder cells, by materials of human origin^41–43 or using serum- and feeder-free culture systems.⁴⁴

Nevertheless, at present, the most common method for clinical application uses LSC, human amniotic membrane or fibrin as substrates, and mouse 3T3 fibroblast feeder layers. Alternative sources of non-corneal adult SC and novel scaffolds are currently under extensive investigation.

Non-Corneal Epithelial Stem/Progenitor Cells

For treatment of patients with bilateral ocular surface disease, strategies may utilize autologous SC from stratified epithelia of other areas of the body. Recent progress in this field suggests that conjunctival epithelium,9–11 oral mucosal epithelium,^8,45,46 epidermis,⁴⁷ and hair follicle^12,32 may serve as alternative sources of autologous adult SC of related lineage, which can be used to construct an artificial corneal epithelium and to reconstruct the ocular surface in animal models and patients with LSCD.

Oral mucosal epithelium has attracted much attention as an autologous epithelial stem cell source and cultured oral mucosal epithelial transplantation (COMET) has already been used for restoration of the corneal surface in patients with bilateral LSCD.8,45,46,47,48 A recent study by Nakamura and coworkers showed promising long-term clinical results of COMET with an improved visual acuity in 10 eyes (53%) at 36 months postoperatively.⁴⁹ However, peripheral neovascularization is commonly seen in COMET, and the long-term clinical success of this technique still requires detailed investigation. It has also been reported that oral mucosal epithelial cell grafts rarely transdifferentiate to a corneal epithelial phenotype, as indicated by lacking expression of cytokeratin K12.⁸

The discovery of various, not yet completely defined SC populations in the epithelial and mesenchymal compartment of the hair follicle, has encouraged research into utilizing the hair follicle as a source of adult multipotent SC for regenerative medicine.⁵⁰ The hair follicle bulge region represents a major repository of multipotent keratinocyte SC, which have the potential to differentiate into hair follicle, sebaceous gland, and epidermis (Fig. 43.3A,B). The potential of murine hair follicle-derived adult SC to transdifferentiate into corneal epithelial-like cells has been recently shown by Blazejewska et al.¹² When exposed to a limbus-specific environment using conditioned medium from limbal stromal fibroblasts and laminin V-coated culture dishes, hair follicle-derived SC could be reprogrammed in vitro into a corneal epithelial phenotype expressing corneal epithelial differentiation markers K12 and Pax6 (Fig. 43.3C–F). Using a transgenic reporter mouse model, which allowed for the detection of K12 expression in vivo, the transplanted corneal epithelial constructs provided evidence of corneal epithelial transdifferentiation and were able to reconstruct the ocular surface of LSCD mice (Fig. 43.4).¹³ These data highlight the promising therapeutic potential of these plastic and readily accessible SC to treat bilateral LSCD.

Figure 43.3 Murine hair follicle bulge-derived stem cells for reconstruction of the corneal epithelium in vitro. (A) Isolated murine vibrissa hair follicle with bulge region harboring the epithelial stem cells. (B) Expression of cytokeratin K15, a putative epithelial stem and progenitor cell marker, in the hair follicle bulge region. (C–F) Phenotypic appearance of hair follicle-derived, fibrin-based epithelial constructs showing positive expression of α6 integrin, a putative stem and progenitor cell marker, in the basal layer (C), expression of the stem cell marker K15 (arrows) in few single basal cells (D), expression of the corneal epithelial differentiation marker K12 throughout all epithelial layers (E), and expression of the epidermal differentiation marker K10 in the most superficial cell layer (F). (Fig 3A–F: Reproduced from: Blazejewska EA, Schlötzer-Schrehardt U, Zenkel M, et al. Corneal limbal microenvironment can induce transdifferentiation of hair follicle stem cells into corneal epithelial-like cells. Stem Cells 2009;27:642–52. With permission from AlphaMed Press.)

Figure 43.4 Murine hair follicle bulge-derived stem cells for ocular surface reconstruction in a rabbit model of limbal stem cell deficiency in vitro (A–C, transplanted eye; D–F, untransplanted control eye). (A,B) Transplantation of the epithelial construct to the limbal stem cell deficient eye resulted in re-establishment of a transparent ocular surface (A) and epithelial barrier function as revealed by a decrease in fluorescein uptake in the ocular surface (B) when compared with untransplanted controls (D,E). (C,F) Periodic acid Schiff staining of excised corneas demonstrated reconstruction of a regular corneal epithelial surface 4 weeks after transplantation with little to no conjunctival goblet cells present (C); however, without transplantation, the ocular surface is entirely covered by goblet cells (arrows) indicating conjunctival epithelial ingrowth (F). (Fig. 4A–F: Reproduced from Meyer-Blazejewska EA, Call MK, Yamanaka O, et al. From hair to cornea: toward the therapeutic use of hair follicle-derived stem cells in the treatment of limbal stem cell deficiency. Stem Cells 2011;29:57–66. With permission from AlphaMed Press.)

Mesenchymal Stem Cells

Other proposed sources of SC for reconstruction of the corneal epithelium include bone marrow-derived mesenchymal SC,14,15,51,52 adipose tissue-derived mesenchymal SC⁵³ and dental pulp SC.¹⁷

Studies by Ye et al.¹⁵ have implied that locally recruited mesenchymal SC may play a role in corneal epithelial healing following alkali injury in a rabbit model and justified the idea that transplantation of mesenchymal SC could be used to treat corneal epithelial defects. Whereas injection of human mesenchymal SC under transplanted amniotic membrane did not provide an improved corneal surface, compared to controls,⁵² the ocular surface of chemically burned rat eyes was repaired when mesenchymal SC were transplanted as an intact sheet.¹⁴ However, their differentiation into corneal epithelial cells was not confirmed, leading these and other authors to suggest that the therapeutic effect of mesenchymal SC transplantation was rather associated with the inhibition of inflammation and angiogenesis than with mesenchymal to epithelial transdifferentiation.⁵⁴ Though co-culture with corneal fibroblasts stimulated mesenchymal SC to express K3 and K12 and to transdifferentiate into a corneal epithelial phenotype in vitro and in vivo.^16,51 Adipose tissue-derived mesenchymal SC, isolated from human orbital fat tissue, were also shown to differentiate into a corneal epithelial lineage when exposed to proper environmental stimuli, such as co-culture with corneal epithelial cells.⁵³ Finally, human dental pulp stem cells, which represent a type of mesenchymal SC from dental pulp of deciduous teeth, have been shown to share key features with LSC and to have the capacity to differentiate into and reconstruct the corneal epithelium in rabbits after chemical burn.¹⁷

Further studies are definitely needed to confirm the value of mesenchymal SC as a cell source for corneal epithelial repair.

Embryonic Stem Cells

Embryonic SC remain largely unexplored for corneal regenerative applications. In one study, mouse embryonic SC cultured on collagen type IV were found to transdifferentiate into corneal epithelial cells and to re-epithelialize the chemically injured corneas of mice within 24 hours of application.¹⁹ In a follow-up study, Ueno et al.⁵⁵ transplanted Pax6-transfected embryonic SC to damaged corneas with even better efficacy. Corneal restoration was achieved and no teratomas were observed. The group also used embryonic SC obtained from cynomolgous monkeys, which are more similar to human cells.⁵⁶ Similarly, human embryonic SC cultured on type IV collagen using limbal fibroblast conditioned medium were found to express corneal epithelial markers, such as K3 and K12, in vitro.19 These studies provide initial data to support further investigation of embryonic SC in corneal epithelial tissue engineering.

Scaffolds for Corneal Epithelial Reconstruction

Function and survival of the expanded SC and the successful establishment of a tissue-engineered corneal epithelium are highly dependent on the structural and biochemical support from the underlying substrate. Although human amniotic membrane has been widely used for LSC expansion and transplantation, a range of alternative biological, biosynthetic or synthetic carriers suitable for SC culture and transplantation has been tested for corneal epithelial tissue engineering in preclinical or clinical applications. A suitable scaffold for corneal epithelial reconstruction should have non-immunogenic and non-inflammatory properties, provide optical transparency and mechanical stability, and promote cell attachment and proliferation.

Biological Scaffolds

Human Amniotic Membrane

The most widely used substrate for corneal epithelial reconstruction is the human amniotic membrane (HAM), i.e., the innermost membrane of the fetal sac, which can be obtained from healthy volunteers during routine caesarean sections (Fig. 43.5A).⁵⁷ It is composed of a single-layered epithelium, which rests on a thick basement membrane and an avascular stroma (Fig. 43.5B). In addition to its structural stability and elasticity, it has anti-immunogenic, antiangiogenic and anti-inflammatory properties, and contains a variety of growth factors and cytokines promoting epithelialization.⁵⁸ HAM basement membrane composition shows extensive similarities with that of the human cornea and limbus (Fig. 43.5C,D),⁵⁹ supporting the notion of HAM serving as a surrogate niche for ex vivo expansion of LSC.^35,60 However, the preparation of HAM is not standardized, and it has been used fresh or frozen, and as an intact or epithelially denuded membrane. Epithelially denuded HAM exposing its BM appears to provide a superior niche for LSC proliferation and phenotypic maintenance.⁶¹ On the other hand, a progressive decline in the number of epithelial progenitor cells was observed during ex vivo expansion on HAM, suggesting limitations in its suitability as a surrogate limbal niche.³