2024 Volume 73 Issue 1 Pages 1-7
Regenerative medicine is a highly anticipated field with hopes to provide cures for previously uncurable diseases such as spinal cord injuries and retinal blindness. Most regenerative medical products use either autologous or allogeneic stem cells, which may or may not be genetically modified. The introduction of induced-pluripotent stem cells (iPSCs) has fueled research in the field, and several iPSC-derived cells/tissues are currently undergoing clinical trials. The cornea is one of the pioneering areas of regenerative medicine, and already four cell therapy products are approved for clinical use in Japan. There is one other government-approved cell therapy product approved in Europe, but none are approved in the USA at present. The cornea is transparent and avascular, making it unique as a target for stem cell therapy. This review discusses the unique properties of the cornea and ongoing research in the field.
The cornea is a unique connective tissue that covers the surface of the eye. It acts as both a physical barrier as well as a lens that functions to provide useful vision. The cornea is approximately 11 mm in diameter, and 500 to 600 µm in thickness at the apex. The cornea is continuous with the sclera, which is opaque and is identified by the white tissue surrounding the cornea. Whereas the sclera is covered by the vascularized conjunctival epithelium, the cornea is directly exposed to the atmosphere with a thin tear film covering the surface. The cornea epithelium is a stratified layer of cells derived from the ectoderm, and unlike the conjunctival epithelium, the cornea is completely avascular. Corneal epithelial cells express cytokeratin 12 (CK12), which is often used as a marker of corneal epithelial cells along with CK3,1 and stem cells of the corneal epithelium lie in the limbus (the basal epithelial layer surrounding the cornea).2 Corneal epithelial stem cells (limbal stem cells) express markers such as CK15 and ΔnP63 and are believed to be unipotent, although there are reports that limbal stem cells have the potential to differentiate into cell types other than the corneal epithelium.3,4,5
The transparency and shape of the cornea is maintained by the corneal stroma, which comprises approximately 90% of the total thickness of the cornea. The stroma consists of collagen fibrils (mostly type I collagen) and proteoglycans interspersed with neural crest-derived corneal stromal cells (keratocytes) and bone marrow-derived immune cells. The highly regular diameter and distribution of collagen fibrils are responsible for the transparency of the cornea, which is in contrast with the more random distribution of collagen fibrils in the stroma.6
The innermost layer of the cornea is covered by the corneal endothelium, which consists of a single layer of hexagonal corneal endothelial cells. These cells are also of neural crest origin and are terminally differentiated with a very limited proliferative capacity. The basement membrane of the corneal endothelium, which is known as the Descemet membrane, plays a crucial role during surgery of the cornea including cornea transplantation. The Descemet membrane is also expected to be critical to the success of cell injection therapy in the near future.7 The cornea endothelial cells express high levels of Na+- and K+-dependent ATPase, which along with other transport channels causes water to pass from the corneal stroma back into the anterior chamber.8 Maintaining an optimal water content in the stroma is also crucial for corneal transparency, because edema of the stroma results in opacification of the cornea. Given that corneal endothelial cells do not proliferate in humans, loss of cell density because of disease or trauma results in irreversible edema of the cornea. Replacing the damaged corneal endothelium with donor tissue through various transplantation techniques is the only viable option to restore useful vision.
Conventional transplantation techniques encompass penetrating keratoplasty (PK), a procedure that involves the replacement of all three layers of the cornea, and deep anterior lamellar keratoplasty (DALK), which focuses on replacing the corneal stroma while preserving the host endothelium.9,10 In contrast, Descemet stripping automated endothelial keratoplasty (DSAEK) and Descemet membrane endothelial keratoplasty (DMEK) are more contemporary approaches that specifically address the diseased endothelial layer while leaving the host epithelium and stroma undisturbed.11 Such advancements in transplantation techniques have significantly enhanced surgical outcomes, and the emerging field of regenerative medicine is poised to address unmet needs, including stem cell replacement and the scarcity of donors.
The corneal epithelium undergoes constant renewal by cell proliferation and differentiation originating from stem cells in the cornea limbus surrounding the transparent cornea. Cells also migrate towards the central cornea in an organized manner, and eventually desquamate from the ocular surface. This homeostasis orchestrated by limbal stem cells was first proposed by Thoft and Friend,12 although the precise mechanisms involved are still under vigorous study. There is a consensus among corneal surgeons that the limbus contains stem cells that can repopulate the cornea when depleted of stem cells because of disease or trauma. Transplanting limbal tissue from an autologous source was first reported by Kenyon et al.,13 and this method has since become a standardized therapy for patients with unilateral disease such as chemical and thermal burns. Current surgical techniques include harvesting limbal tissue (2 to 4 mm in size) from the healthy fellow eye and suturing the tissue on the diseased eye. A modified version of this technique was reported as simple limbal epithelial transplantation (SLET) in 2012 by Sangwan et al.14 SLET is a sutureless technique where the donor limbal tissue is cut into 10 to 15 pieces and applied to the diseased cornea using fibrin glue.
A further advance in the transplanting of autologous stem cells was reported in the form of cultivated limbal epithelial transplantation (CLET).15 Unlike the transplantation techniques described above, CLET involves cultivation of cells in vitro. This introduces an entirely different set of regulatory rules that require strict measures to guarantee the safety of cells to be transplanted in patients. Although the use of an autologous cell source can avoid immunological rejection, the logistics involved can become complicated. In Japan, an autologous CLET product was approved by the Pharmaceuticals and Medical Devices Agency (PMDA) as a regenerative medicine product. Similarly, autologous cultivated oral mucosal epithelial transplantation (COMET) products were also approved for clinical use to regenerate the ocular surface.16,17 Currently, there are two different COMET products with different properties and indications.
All three approved products cost more than 10 million yen, yet most of the cost is incurred through production and safety measures. The successful launch of such regenerative medicine products will depend on the dissemination of the technique among clinics in all parts of the country. To lower the cost of regenerative products, the use of an allogeneic source and the development of cryopreservation techniques are vital. An allogeneic source of cells is currently sought using allogeneic donor tissue and induced pluripotent stem cells (iPSCs).
Using autologous stem cells is only possible in unilateral disease where the healthy fellow eye can be used as a source of stem cells. For bilateral disease such as autoimmune or congenital conditions, an allogeneic source of cells is required. The transplantation of allogeneic tissue containing epithelial stem cells (limbal tissue) is well documented.18 The technique involves the harvest of limbal tissue from donor corneas, usually after the central cornea has been used for conventional transplantation. The procured donor tissue can then be sutured onto the diseased host after all diseased epithelial tissue has been removed. Given that HLA matching is not feasible, allogeneic limbal stem cell transplantation (kerato-limbal allograft: KLAL) requires long-term local and systemic immunosuppression. Although short- to mid-term results are promising for patients who would otherwise be blind, the long-term results are still not ideal. There is only limited evidence showing that transplanted donor stem cells survive in the long term, although there are reports showing donor-derived cells 1 year post surgery.19
Using cultivated allogeneic epithelial cell sheets is a promising alternative to autologous CLET and COMET. Clinical studies have shown that allogeneic cultivated epithelial cell sheets can regenerate a healthy ocular surface to provide useful vision. However, the timing of cultivation as well as the period required to obtain a robust stratified epithelial sheet while performing required safety tests can be challenging. To circumvent these issues, we previously reported a protocol to cultivate and maintain corneal epithelial cell sheets containing stem cells for over a year.20 To our surprise, these long-term cell sheets maintained a robust stratified epithelial layer with correct polarity, complete with basal cells with slow cycling properties. This long-term culture protocol will provide a longer shelf-life, the ability to complete all required safety tests, and a stock of ready-to-use transplantable sheets for emergency cases such as chemical or thermal burns.
The availability of donor tissue is also a limiting step for regenerative products. For example, if one donor cornea is required for every cultivated epithelial sheet, then there will be little advantage of cultivated sheets over standard limbal transplantation. We therefore devised a protocol to cultivate limbal stem cells as limbal organoids, with each organoid capable of forming a transplantable sheet. Our latest data show that approximately 50 organoids can be manufactured from one donor cornea.21 However, the ability to produce a homogeneous stock of epithelial sheets needs to be confirmed prior to commercialization of this method.
iPSCs are another source of allogeneic cells for regenerative medicine. Hayashi et al.22succeeded in engineering transplantable corneal epithelial cell sheets from iPSCs using their “SEAM” protocol. SEAM stands for “self-formed ectodermal autonomous multi-zone” and uses the pluripotent nature of iPSCs to form zones of committed ocular cells in two-dimensional culture dishes. Not all iPSC clones appear to form the SEAM structure, and therefore choosing an optimal clone is essential. Clinical trials for SEAM-derived corneal epithelial sheet transplantation are currently underway. Further refinements of this technique may include the use of iPSCs that have been modified to reduce immunological rejection.
Compared with advances in corneal therapy, the ability to regenerate the lacrimal gland still needs to be achieved to restore a healthy ocular surface in patients lacking tears. Recent progress in reconstructing the lacrimal gland using stem cells is an area that has attracted much attention. We demonstrated that three-dimensional bioengineered lacrimal glands regenerated in vitro can improve tear secretion in mice.23 More recently, Hayashi et al.24reported iPSC-derived lacrimal gland organoids from human ocular epithelium derived from their SEAM technique. Future research on how to transplant functionally regenerated lacrimal glands should pave the way for clinical application.
The loss of corneal endothelial function results in edema of the cornea, which causes opacification because of the scattering of light entering the cornea. Currently, the only means available to restoring the clarity of the cornea is to transplant donor tissue with healthy endothelial cells. Historically, penetrating keratoplasty, or full thickness transplantation, is the standardized technique of choice and is still performed currently. However, more refined techniques to transplant only a thin layer of tissue containing endothelial cells have appeared in the past decade. Endothelial keratoplasty is the term that is used currently to describe transplantation of corneal endothelial cells attached to a thin layer of stromal tissue or endothelial cells attached to the Descemet membrane (basement membrane). Endothelial keratoplasty has become the first choice of therapy over the past decade.25,26
More recently, a group from Kyoto Prefectural University reported the use of cultivated endothelial cells that can be injected into the anterior chamber of the eye to repopulate the diseased host cornea with healthy endothelial cells.27 A key aspect of this technique involves the use a rho kinase inhibitor (ROCK inhibitor) to improve cell attachment and cell viability. Clinical data presented to date show that cell injection therapy may become a viable option. Surprisingly, the topical use of ROCK inhibitor alone may effectively treat corneal endothelial dystrophy patients with a healthy reserve of endothelial cells in the peripheral area. This technique is reported as Descemet stripping only (DSO) and involves removal of central diseased cells (4.0 mm diameter) followed by ROCK inhibitor application to allow peripheral healthy cells to migrate and even possibly to proliferate.28
The use of other autologous or allogeneic sources of stem cells for endothelial regeneration have been reported, including studies from our group that attempted to induce functional corneal endothelial cells from corneal stromal progenitor cells (COPs),29 skin-derived progenitor cells (SKPs),30 and umbilical cord mesenchymal stem cells.31 However, the most realistic approach for clinical application was the use of allogeneic iPSCs as a source of corneal endothelial cell induction (Fig. 1). Derivation of corneal endothelial cells from iPSCs was first reported by Zhao and Afshari.32 Their protocol involves the induction of eye-field stem cells, which are then further cultured to obtain corneal endothelial cells with characteristic morphology and maker expression. Our group reported a more direct approach to induce corneal endothelial-like cells, which we termed “cornea endothelial substitutes”.29 The term “substitutes” was used because we did not recapitulate the developmental process of corneal endothelial development, and there are no definitive markers to identify cornea endothelial cells. These cells showed typical hexagonal morphology with expression of Na,K-ATPase alpha 1 subunit, ZO-1 tight junction protein, N-cadherin adherens junction, and nuclear PITX2. Following rigorous pre-clinical studies in monkeys and tumorigenicity tests in rats,33 a clinical study is currently underway.
Previous attempts to induce corneal endothelial cells.
Methods include the use of corneal stroma-derived precursor cells (neural crest origin), skin-derived precursor cells (neural crest origin), mesenchymal stem cells, and cells derived from induced-pluripotent stem cells (iPSCs) (closest to clinical use).
The use of mesenchymal stem/stromal cells (MSCs) for acute graft-versus-host disease (GVHD) was first reported in 2008 by Le Blanc et al.34 Currently, an MSC product that was recently approved by the PMDA is one of the most widely used cell therapy products in Japan. Although the complete mode of action of MSCs is not entirely clear, part of the pharmacological effect comes from suppressing the proliferation of CD4+ helper T cells and the subsequent induction of regulatory T cells (Tregs) from CD4+ T cells. These effects are reported to be a result of both direct cell–cell interactions as well as indirect actions through soluble factors.35,36
Several groups have reported the use of MSCs for inflammatory disease of the cornea.37,38 While systemic intra-venous infusion of MSCs is required for GVHD therapy, a more local application in the form of eyedrops or injections is possible in the case of the cornea. This will allow a more concentrated number of cells to be applied to the area of inflammation, while being safer than systemic injection. Subconjunctival injection of MSCs was shown to be protective in a murine model of GVHD.38 Another group suggested that MSCs can modulate macrophage phenotype in a corneal injury model to produce more anti-angiogenic and anti-inflammatory factors.37 We are currently preparing for our own clinical trial of sub-conjunctival injection of clinical-grade human adipocyte-derived MSCs in the treatment of acute inflammatory diseases of the ocular surface (Fig. 2). Because the cornea needs to maintain transparency for useful vision, prompt control of inflammation to reduce secondary neovascularization will help reduce the need for transplantation.
Protocol for treatment of inflammatory ocular surface disease.
Our ongoing clinical study involved the use of cryopreserved mesenchymal stem/stromal cells that are injected into the subconjunctival space near the area of inflammation.
Regenerative medicine is still in its early stages as a standardized modality of medical care. Further innovation is required in both the technical field as well as in regulatory science. Clarification of the numerous possible modes of action is required to produce a more standardized cell therapy product. Cost is also an issue, because currently available products are still very expensive for widespread use. However, such issues will probably be resolved over time as with other innovative products in other fields of consumer products. Cell products may become more customized with enhanced functions because of gene modification. It is our hope that innovation will exponentially enhance the potential of regenerative medicine once current global endeavors reach a certain threshold.
Some of the research on iPSCs discussed in this review was funded by the New Energy and Industrial Technology Development Organization (NEDO) and the Japan Agency for Medical Research and Development (AMED). Figures created with BioRender.com.
S.S., E.I., and S.H. have patents relevant to technology reported in this review. S.H. is employed by Cellusion Inc. M.H. declares no conflict of interest.
