Free
Review  |   February 2015
Understanding the Process of Corneal Endothelial Morphological Change In Vitro
Author Affiliations & Notes
  • Olivier Roy
    Centre de Recherche du Centre Hospitalier Universitaire (CHU) de Québec, Axe Médecine Régénératrice, Hôpital du Saint-Sacrement, Québec, Québec, Canada
  • Véronique Beaulieu Leclerc
    Centre de Recherche du Centre Hospitalier Universitaire (CHU) de Québec, Axe Médecine Régénératrice, Hôpital du Saint-Sacrement, Québec, Québec, Canada
  • Jean-Michel Bourget
    Centre de Recherche de l'Hôpital Maisonneuve-Rosemont (HMR), Montréal, Québec, Canada
    Département d'Ophtalmologie, Faculté de Médecine, Université de Montréal, Montréal, Québec, Canada
  • Mathieu Thériault
    Centre de Recherche du Centre Hospitalier Universitaire (CHU) de Québec, Axe Médecine Régénératrice, Hôpital du Saint-Sacrement, Québec, Québec, Canada
  • Stéphanie Proulx
    Centre de Recherche du Centre Hospitalier Universitaire (CHU) de Québec, Axe Médecine Régénératrice, Hôpital du Saint-Sacrement, Québec, Québec, Canada
  • Correspondence: Stéphanie Proulx, Centre de recherche du CHU de Québec, Hôpital du Saint-Sacrement, 1050 chemin Ste-Foy, Québec, QC, Canada G1S 4L8; Stephanie.Proulx@fmed.ulaval.ca
Investigative Ophthalmology & Visual Science February 2015, Vol.56, 1228-1237. doi:10.1167/iovs.14-16166
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Olivier Roy, Véronique Beaulieu Leclerc, Jean-Michel Bourget, Mathieu Thériault, Stéphanie Proulx; Understanding the Process of Corneal Endothelial Morphological Change In Vitro. Invest. Ophthalmol. Vis. Sci. 2015;56(2):1228-1237. doi: 10.1167/iovs.14-16166.

      Download citation file:


      © 2016 Association for Research in Vision and Ophthalmology.

      ×
  • Supplements
Abstract

Corneal endothelial cells often adopt a fibroblastic-like morphology in culture, a process that has been attributed to epithelial- or endothelial-to-mesenchymal transition (EMT or EndMT). Although being extensively studied in other cell types, this transition is less well characterized in the corneal endothelium. Because of their neuroectodermal origin and their in vivo mitotic arrest, corneal endothelial cells represent a particular tissue that deserves more attention. This review article presents the basic principles underlying EMT/EndMT, with emphasis on the current knowledge regarding the corneal endothelium. Furthermore, this review discusses cell culture conditions and major cell signaling pathways that have been identified as EndMT-triggering factors. Finally, it summarizes strategies that have been developed to inhibit EndMT in corneal endothelial cell culture. The review of current studies on corneal and classical EndMT highlights some research avenues to pursue in the future and underscores the need to extend our knowledge of this process in order to optimize usage of these cells in regenerative medicine.

The corneal endothelium is a monolayer of hexagonal cells located on the posterior surface of the cornea, lying on a specialized basement membrane called Descemet's membrane. It acts as a leaky barrier between the corneal stroma and the aqueous humor of the anterior chamber. The corneal stroma has a natural tendency to imbibe fluid from the anterior chamber; however, stromal edema decreases corneal transparency. The main role of the corneal endothelium is to regulate stromal hydration by pumping excess fluids out of the stroma, which is achieved through ionic pumps and cotransporters.1 Thus, the corneal endothelium is responsible for the maintenance of corneal transparency by controlling stromal deturgescence through a “pump-leak” mechanism. 
In vivo, human corneal endothelial cells (CECs) are arrested in the G1 phase of the cell cycle2 and thus remain in a nonproliferative state. A high endothelial cell density is necessary in order to maintain their pump-leak equilibrium.3 When endothelial density diminishes to under a critical value (around 400–700 cells/mm2)4 following diseases or injury, the normal barrier and pump functions of the endothelium are no longer sufficient to maintain the transparency of the cornea. This irreversible decompensation results in corneal edema and vision loss. To this day, the only way to restore vision after endothelial decompensation is to replace the endothelium with an allogeneic graft by either penetrating keratoplasty or endothelial keratoplasty. Endothelial dysfunction represented the first indication for corneal transplantation in the United States as 26,509 procedures (40% of all transplantations) were performed in 2013.5 For many years researchers have turned toward the use of cultured CECs as an alternative for the replacement of the corneal endothelium.68 
Even though CECs are in a nonproliferative state in vivo, they still possess regenerative capacities and can be expanded in vitro.9 It has been demonstrated that disruption of cell contacts and addition of growth factors to the culture medium stimulate replication of CECs. Specific culture media have been developed for human,10 porcine,11 and feline12 endothelial cells. They combine a variety of supplements such as insulin, growth factors (nerve growth factor [NGF], epithelial growth factor [EGF], fibroblast growth factor [FGF]), bovine pituitary extract (BPE), ascorbic acid, serum, and various cytokines and mitogens, all aimed at increasing cell proliferation. The use of pharmacologic inhibitors of Rho-kinase (ROCK) and type I TGF-β receptor (TβRI) as well as a medium conditioned by mesenchymal stem cells or irradiated 3T3 cells was also reported.13,14 However, cell culture of CECs can result in a phenotypic switch that changes their morphology from endothelial to fibroblastic.1518 Endothelial-to-mesenchymal transition (EndMT), a variant of the well-known epithelial-to-mesenchymal transition (EMT),19,20 has been postulated to play a role in this loss of endothelial phenotype.15,21 This loss is problematic since it limits the use of cultured CECs in tissue engineering. The diversity of factors that could cause this phenomenon highlights an increasing need to review what is known about this new concept of corneal EndMT and ways to antagonize it in order to improve culture methods, as well as to progress toward the use of tissue-engineered corneal endothelium for clinical applications. 
Epithelial- and Endothelial-to-Mesenchymal Transition
The EMT or EndMT, in a few words, is the process by which epithelial or endothelial cells lose their specific markers and acquire mesenchymal characteristics. The EMT was first described in the early 1980s by Elizabeth Hay, who observed this phenomenon in the primitive streak of chick embryos. The authors were able to recapitulate this process in vitro using collagen gels.22,23 This transition was then shown to play an important role in normal physiological processes such as embryonic development, wound healing, and stem cell behavior. However, EMT is also implicated in pathologic processes such as fibrosis and cancer.24,25 
The EMT and EndMT are characterized by key events that are either causes or consequences of the transition. Those include the disassembly of cell–cell junctions, loss of apical–basal polarity, actin cytoskeleton reorganization, changes in cell shape, increased cell motility, increase in the production of extracellular matrix proteins, and changes in gene expression. Cells can also acquire extracellular matrix (ECM) degradation capabilities (expression of matrix metalloproteinase [MMP-2 and MMP-9]) as well as resistance to apoptosis and senescence.19 Studies in cancer and other cell types, mostly of epithelial nature, have provided insight on how classical EMT occurs (extensively reviewed in Ref. 19).19,20 The EndMT has been studied on vascular endothelial cells.2628 However, the corneal endothelium, a transporting endothelium of neuroectodermal origin, is distinct from both the epithelium and the vascular endothelium. Thus, the loss of corneal endothelial phenotype following cell culture may be specific to this structure and regulated by different mechanisms, and deserves particular attention. 
Cell Junction Destabilization and Loss of Apical–Basal Polarity
Epithelia and endothelia act as selective barriers between two compartments, for either protection (e.g., epidermis), secretion (gland), absorption (intestine), or selective exchange (endothelium). This barrier function is dependent upon the capability of the cells to regulate passage of certain substances from one compartment to the other. This would not be possible without cells being tightly bound together by numerous tight and adherens junctions. Tight junctions, composed of claudins and occludins, seal the gap between cells and connect to the cytoskeleton with the help of membrane-associated proteins such as zonula occludens (ZO)-1.29 Adherens junctions also play a vital role in barrier function by holding cells together into a cohesive epithelium. These junctions are formed by epithelial cadherin (E-cadherin) that binds adjacent cell cadherin to the actin cytoskeleton by means of α- and β-catenins.30 
In EMT, tight and adherens junctions are destabilized. Zonula occludens-1 is relocated to the cytoplasm, and the expression of claudins and occludins is abrogated. E-cadherin is cleaved and degraded,31 leaving p120-catenin free to act as transcription factors32,33 or to be degraded. Moreover, desmosomes and gap junctions are disrupted and, as the EMT progresses, gene expression of those proteins decreases.34 
One important epithelial cell marker known to be downregulated in EMT is E-cadherin. Loss of this cell–cell junction protein reduces the adherence with epithelial cells. Moreover, the subsequent upregulation of N-cadherin leads to an increased adherence to mesenchymal cells. This phenomenon is known as the cadherin switch.31,35 Due to its neuroectodermal origin, the corneal endothelium represents a special case regarding the expression of cadherins. In vivo, the contact-inhibited human corneal endothelium expresses all three cadherins (E-, N-, and VE-cadherin) but only N- and VE-cadherin are properly located at the cell membrane while E-cadherin is present only in the cytoplasm.36 It thus appears that N-cadherin and not E-cadherin plays a central role as an adherens junction protein in human CECs. Therefore, this raises the question whether N-cadherin should be used as an EndMT marker in the corneal endothelium. 
Junctions also define the apical and basal compartment of polarized epithelial cells. The loss of apical–basal polarity is another characteristic found in EMT. The apical compartment is defined by the presence of partitioning-defective (PAR) and Crumbs complexes that associate with tight junctions. The basolateral compartment comprises the Scribble complexes.37 The PAR, Crumbs, and Scribble complexes are destabilized by dissolution of tight junctions, and their expression is repressed as EMT progresses.38 
Cell isolation and subsequent passaging of CECs is inevitably associated with the dissociation of cell junctions. Success in avoiding EndMT of cultured CECs could reside in the development of culture conditions in which confluency and gain of apical–basal polarity is rapidly achieved. For example, increasing cell seeding density has been shown to diminish the occurrence of fibroblastic transformation within human CEC culture. Indeed, Peh and colleagues39 demonstrated that a high CEC seeding density (20,000 cells/cm2) resulted in more compact and contact-inhibited CECs when compared to lower cell densities (2500–5000 cells/cm2), the latter resulting in a culture composed mostly of pleiomorphic fibroblastic-like cells.39 These results replicate the comparable findings of an earlier study of primate CEC cultures.40 
Cytoskeleton Reorganization: Changes in Cell Shape and Motility
The EMT is also characterized by a Rho guanosine triphosphatase (Rho GTPase) family-dependent actin cytoskeleton reorganization, where RAC1 and CDC42 induce the formation of cell projections (lamelipodia and filipodia) and where RhoA is responsible for stress fiber formation. The rearrangement from an epithelial-like cortical organization to a stress fiber mesenchymal phenotype allows for an increased directional motility and contractility and leads to cell shape modification from round to elongated cells.4144 Accordingly, Rho GTPase activity is deleterious to the apical–basal polarity.45,46 
Cells in EMT and EndMT can express a particular form of actin, the α-smooth muscle actin (α-SMA). Alpha-smooth muscle actin is often used as an EndMT marker, in both vascular cells47 and CECs.4850 The expression of α-SMA by CECs was shown to be dependent upon the addition of TGF-β, a cytokine that induces EndMT.49 
Extracellular Matrix Protein Secretion
Cells that undergo EMT show an increase in the secretion of ECM proteins, such as collagens and fibronectin.51 Expression of integrins and proteases is also modified in EMT.19 Some cells that undergo EMT can acquire ECM degradation abilities for invasion and migration, including the expression of MMPs.19 
This modulation of ECM secretion can be seen in cultured CECs. Recently, Okumura and colleagues15 observed that cultured simian CECs that had a fibroblastic morphology expressed higher levels of fibronectin and type I and IV collagens as well as α5 and β1 integrins. In vitro models of endothelial pathologies led to the discovery of some factors implicated in the EndMT that caused abnormal ECM secretion and fibrosis by CECs. Among them, FGF-2 has been extensively studied. Corneal endothelial cells cultured with FGF-2 acquire the ability, through PI3K signaling, to stabilize type I collagen mRNA, thus leading to its secretion outside of the cell.52 Studies on an ex vivo model of cat corneal endothelial fibrosis showed that TGF-β1-2-3 are implicated in the expression of fibronectin and α-SMA by CECs and in the loss of their endothelial phenotype.53,54 Addition of TGF-β1-2 to the culture medium of bovine CECs also induced the production of laminin and fibronectin in a dose-dependent manner, via TβRI and Smad2/3 and 4.55 Corneal endothelial cells from organ-cultured models of retrocorneal fibrous membrane were also shown to express type I collagen and fibronectin.56,57 
It is thus clear that cultured CECs that have undergone EndMT will not be of use in regenerative medicine. In order to control EndMT, we first need to review the current literature on signals that generate EMT. The information contained in the next paragraphs is summed up in the Figure, along with the other hallmarks of corneal EndMT discussed previously. 
Figure
 
Summary of the current knowledge on the molecular mechanisms that regulate endothelial-to-mesenchymal transition in cultured corneal endothelial cells. Multiple extracellular signals (TGF-β1-2-3, IL-1β, FGF-2, connexin 43) can influence cultured corneal endothelial cells. They act as trigger factors that can impact critical intracellular pathways (Smad, p38, Erk1/2, Wnt/β-catenin, NF-κB, p27kip1) and induce endothelial-to-mesenchymal transition. On one hand, the endothelial phenotype is characterized by a contact-inhibited monolayer of polygonal-shaped cells that are arrested in the G1 phase of the cell cycle and express cell junction proteins such as ZO-1 and N-cadherin. On the other hand, cells with a fibroblastic phenotype undergo changes in cell shape, actin cytoskeleton reorganization, and disassembly of cell–cell junctions. They also lose their contact inhibition. Furthermore, they present an increased cell motility and extracellular matrix production as well as changes in gene expression. TGF, transforming growth factor; IL, interleukin; FGF, fibroblast growth factor; Erk, extracellular signal-regulated kinase; NF, nuclear factor; ZO, zonula occludens; ECM, extracellular matrix.
Figure
 
Summary of the current knowledge on the molecular mechanisms that regulate endothelial-to-mesenchymal transition in cultured corneal endothelial cells. Multiple extracellular signals (TGF-β1-2-3, IL-1β, FGF-2, connexin 43) can influence cultured corneal endothelial cells. They act as trigger factors that can impact critical intracellular pathways (Smad, p38, Erk1/2, Wnt/β-catenin, NF-κB, p27kip1) and induce endothelial-to-mesenchymal transition. On one hand, the endothelial phenotype is characterized by a contact-inhibited monolayer of polygonal-shaped cells that are arrested in the G1 phase of the cell cycle and express cell junction proteins such as ZO-1 and N-cadherin. On the other hand, cells with a fibroblastic phenotype undergo changes in cell shape, actin cytoskeleton reorganization, and disassembly of cell–cell junctions. They also lose their contact inhibition. Furthermore, they present an increased cell motility and extracellular matrix production as well as changes in gene expression. TGF, transforming growth factor; IL, interleukin; FGF, fibroblast growth factor; Erk, extracellular signal-regulated kinase; NF, nuclear factor; ZO, zonula occludens; ECM, extracellular matrix.
Extracellular Signals That Drive EMT
Among the numerous factors that trigger EMT, two types of extracellular signals and receptors were particularly studied: the transforming growth factor beta (TGF-β) superfamily (reviewed in Ref. 58) and the tyrosine kinase receptors. They activate many pathways that are closely related with each other. 
Transforming Growth Factor-β
TGF-β and the Smad Signaling Pathway.
The TGF-β superfamily includes TGF-βs, bone morphogenic proteins (BMP), and activins.59 These proteins are essential to fetal development, as they regulate genes related to developmental EMT and the reverse phenomenon, mesenchymal-to-epithelial transition (MET). They are also implicated postnatally in physiological and pathological EMT, such as in wound healing, pluripotent cell differentiation, fibrosis, and cancer (reviewed in Ref. 60). They transduce extracellular signals via transmembrane Ser/Thr kinase heterotetramer receptors and Tyr kinase receptors. The canonical pathway is activated by binding of a TGF-β protein to a type II TGF-β receptor (TβRII), a Ser/Thr kinase, which phosphorylates a type I receptor, leading to the phosphorylation of Smads.61 Smad2 or 3 is activated by TGF-β signaling and Smad1, 5, or 8 in response to BMPs. Inhibitory Smads, namely Smad6 and 7, can compete with the latter Smads for binding to the type I receptors.59 Smad pathways lead to activation of various transcription factors, including Snail transcription repressors, ZEB transcription factors, and basic helix-loop-helix (bHLH) factors.62 Notably, Snail1 and 2 repress transcription of E-cadherin.63 Snail1 is linked to disruption of the epithelial polarity64 and remodeling of the cytoskeleton as well as cellular migration and invasion.65 Likewise, Smads directly control many genes linked to the mesenchymal phenotype, like fibronectin, vimentin, and collagen α1.58 
ZEB1 and ZEB2 can repress or activate transcription, depending on the coactivators/repressors associated with them. These factors repress many genes linked to the epithelial phenotype, including E-cadherin (adherens and tight junctions, epithelial polarity), activate the expression of genes associated with a mesenchymal phenotype, such as α-SMA and vimentin, and have been associated with cell migration.58 
Other extracellular signals can also lead to activation of Snails and ZEBs. Indeed, HGF, FGF, or EGF signaling by the Ras-mitogen-activated protein kinase (MAPK) or PI3K-Akt pathways can cooperate with Snails in order to induce EMT.59 It has also been shown in some cancers that Ras,66 Notch, and Wnt58 pathways can cooperate with TGF-β to induce Snail. Ras-MAPK and Wnt/β-catenin pathways also activate ZEB transcription factors.62 
Non-Smad Signaling by TGF-βs.
Noncanonical signaling by the TGF-β family of proteins has also been linked to EMT, including extracellular signal-related kinase (Erk)-MAPK, Rho GTPases, and PI3K/Akt pathways.58 The activation of these pathways can be done either by bonding of TGF-β to the tyrosine kinases receptors or by interaction of the phosphorylated receptors with Smads. Cooperation between Ras-Erk MAPK and the TGF-β pathways has been shown in TGF-β–induced EMT, notably by downregulation of E-cadherin.6769 
p38-MAP kinase can be activated by TGF-β to induce EMT and apoptosis.70 It has also been demonstrated that blocking activation of p38 inhibits TGF-β–induced EMT.71 
c-Jun N-terminal kinase (JNK) MAPK, as well as PI3K/Akt pathway and its downstream target mechanistic target of rapamycin (mTOR), are also implicated in TGF-β–induced EMT, particularly in pathological contexts.7274 
Rho-like GTPases (Rho, Rac, Cdc42) are a Ras-related family of proteins that actively control the organization of the actin cytoskeleton. RhoA has been particularly studied in TGF-β–induced EMT. When activated by TGF-β, RhoA transduces its signal to its downstream effector, ROCK, leading to formation of actin stress fibers, as discussed previously.7577 Furthermore, inhibition of Rho-ROCK signaling has been shown to block or reverse TGF-β–induced EMT and fibrosis in some types of cells.76,78 
TGF-β in Human Corneal Endothelial Cells (HCECs).
Human corneal endothelium expresses all three types of TGF-β receptors in vivo.79,80 The normal human aqueous humor contains 1.2 to 4.9 ng/mL total TGF-β81,82 of which TGF-β2, the most abundant isoform, occupies 0.27 to 2.24 ng/mL in its inactive form and 20 to 830 pg/mL in its activated form.83,84 On one hand, TGF-β is mostly present in its latent form; it therefore needs to be activated in order to induce an effect upon the cells. On the other hand, human CECs also express thrombospondin-1, which is a known activator of TGF-β.8587 Additionally, Descemet's membrane also contains fibronectin,88,89 which is implicated in latent TGF-β activation.90,91 
Considering the abundance of TGF-β in the aqueous humor, it is not surprising that this cytokine has been shown early to modulate CECs in many ways. Indeed, studies in rat and rabbit cultured endothelium showed that TGF-β2 suppresses S-phase entry, therefore leading to the loss of proliferative capabilities of CECs.9294 Recently, the implication of TGF-β2 in corneal endothelial wound healing was clarified by the demonstration that migration, and not proliferation, was induced by TGF-β2 in corneal endothelial wound healing models via the activation of p38.80 These findings support previous work showing that, in ex vivo organ culture of human CECs, disruption of cell junctions by EDTA and TGF-β induced EndMT but not proliferation of the cells.49 Globally, this reinforces the idea that TGF-β represses proliferation, and proposes that its role in the physiologic wound healing process of human corneal endothelium is based on cell migration and not proliferation.95 
Recently, Okumura et al.15 showed that TGF-β1 generated EndMT in cultured human CECs. They also showed that in vitro, simian CECs naturally underwent EndMT following a pattern resembling the one induced by TGF-β (phosphorylation of Smad2 and p38). Contrary to the findings of Joko and colleagues,80 they demonstrated phosphorylation of Erk1/2 by TGF-β1.15 However, this discrepancy can be explained by the presence of FGF-2 (an Erk1/2 activator) in their growth medium, thus correlating with their peers' findings.80 Furthermore, blocking TGF-β signaling using SB431542, a TβRI kinase activity inhibitor, allowed for CEC function-associated proteins ZO-1 and Na+/K+-adenosine triphosphatase (Na+/K+-ATPase) expression and contact-inhibited cell morphology to be maintained as well as reducing collagen I and fibronectin expression. 
Interestingly, cell response to TGF-β is highly influenced by the context, which raises the concept of a duality of action known as the TGF-β paradox, commonly known to take place in invasive cancers.96 This phenomenon describes the discrepancy between the antiproliferative effect of TGF-β and its ability to resume and deregulate proliferation in later stages of the tumoral progression. While more than a few mechanisms are thought to be implied in the TGF-β paradox, one appears particularly relevant in the context of corneal endothelial EndMT. Indeed, the TβRII serine-threonine kinase receptor also possesses a tyrosine kinase activity that introduces a wide variety of TGF-β-MAPK–associated noncanonical signaling.97 
This noncanonical signaling reveals itself to be of paramount importance in the determination of the CECs' response to TGF-β. It has been shown that TβRII possesses a tyrosine kinase activity enabling the phosphorylation of MAPK elements such as p38 or Erk1/2.98 Moreover, the downstream specificity of this signaling seems to be governed by the expression ratio of TβRII/TβRI; a higher ratio favors Erk1/2 activation, whereas a lower ratio favors the canonical Smad signaling.99 In order to further characterize the CECs' response to TGF-β, defining this ratio in the corneal endothelium might lead to a better targeting of TGF-β signaling. 
Fibroblast Growth Factor-2
Kay and colleagues100 were one of the first groups to have underlined a link between growth factors and changes in the CEC morphology. The first factor identified by this team as an EndMT trigger was FGF-2 (or bFGF), a growth factor known to be present within Descemet's membrane.101,102 They first discovered that exposure to polymorphonuclear leukocytes (PMN) led the cells to become fibroblastic-like along with causing synthesis of undegraded type I collagen in rabbit CECs.56,100,103 They later identified IL-1β as the factor synthesized by PMN responsible in exerting such changes on CECs.104 They also showed that this factor induced the production of FGF-2, which in the meantime had been reported to induce the same phenomenon as exposure to PMN.102,104,105 Exposure to IL-1β led to the production of FGF-2 via PI3K induction of p38 by an unclear mechanism.104,106,107 Recently, they showed that PI3K also allowed for the activation of NF-κB, the latter being a direct activator of FGF-2 transcription and synthesis, which underlines a preponderant role for both PI3K and NF-κB in FGF-2–induced EndMT.108 
Fibroblast growth factor-2 has been shown to stimulate the proliferation of CECs by downregulating p27 via activation of Erk1/2 by PI3K.80,109111 It has also been demonstrated that FGF-2 can stabilize type I collagen α1 mRNA and thus upregulate its synthesis and secretion into the ECM by inhibiting its degradation.112 Moreover, FGF-2 stimulates migration of CECs113 by a multitude of mechanisms. For instance, FGF-2 promotes the activation of the small GTPase Cdc42 and inactivation of Rho via PI3K signaling pathway, which favors a migratory phenotype.114,115 Supporting this, Joko et al.80 recently demonstrated the implication of p38 as a downstream effector of FGF-2–induced migration of human CECs. These results are summarized in the Table, along with the other factors modulating CEC phenotype in cell culture. 
Table
 
Extracellular Factors and Signaling Pathways and Their Effects on CEC Phenotype in Culture
Table
 
Extracellular Factors and Signaling Pathways and Their Effects on CEC Phenotype in Culture
Authors, Ref. Numbers Factors/Pathways Effect Cell Type, Passage Culture Method
Joko et al.80 TGF-β2/p38 MAPK Migration Human, P5 Primary explant
Joko et al.80 FGF-2/p38 MAPK, Erk1/2 Proliferation/migration Human, P5 Primary explant
Zhu et al.49 Cell junction disruption, TGF-β, FGF-2/Wnt EndMT Human, P0 Collagenase
Petroll et al.53,54 TGF-β1-2-3 Loss of endothelial phenotype, fibrosis, myofibroblast transformation Feline, P0 Ex vivo corneas
Okumura et al.15 TGF-β1/Smad2, p38 EndMT Human, simian, P2-5 Collagenase
Kay et al.100,103; Lee et al.104; Lee and Kay106; Song et al.107; Lee et al.21; Lee and Kay108; IL-1β, FGF-2/PI3K, p38, NF-κB EndMT, proliferation, migration Human Trypsine and/or collagenase-hyluronidase
Nakano et al.48 Connexin43/p27kip1 EndMT Rat In vivo* (wound healing)
Other Signals
Wnt Signaling.
Although disruption of cell junctions is usually considered to be a consequence of EMT, the case of CEC EndMT might oppose this paradigm as evidence shows this event as a potential cause of EndMT.116,117 Indeed, Tseng's group49 recently unveiled a link between loss of contact inhibition by cell junction disruption and appearance of EndMT in human CECs. They showed that canonical Wnt signaling, through nuclear translocation of β-catenin following EDTA-induced cell dissociation, was directly implicated in inducing EndMT. Despite disruption of cell junctions, FGF-2 is needed to unlock the mitotic block, allowing resumption of proliferation. This effect of FGF-2 is not explained by its aforementioned EndMT-triggering capabilities per se, but by the upregulation of β-catenin it induces.49 Indeed, overexpression of stable β-catenin suppresses the need of growth factors for EndMT to occur via disruption of cell junctions and Wnt signaling, indicating the existence of a threshold in the level of β-catenin needed to induce EndMT. Though FGF-2 activates canonical Wnt signaling by the upregulation of β-catenin protein expression, Zhu and colleagues49 did not observe similar behavior when TGF-β was used instead of FGF-2. This may explain the absence of proliferation seen in the latter case, and it also indicates that canonical Wnt signaling is not the only pathway responsible for corneal EndMT since EDTA-TGF-β–treated cultures also underwent EndMT. However, the findings suggest a way to interpret the role of proliferation in EndMT and offer an explanation for distinguishing proliferative (i.e., FGF-2), migratory (i.e., TGF-β), or mixed EndMT (i.e., FGF-2 and TGF-β). 
This suggests that there would be a junctional environment-dependent effect of known EndMT-triggering factors, more specifically TGF-β. Indeed, there seems to be a relationship between the context in which CECs are exposed to EndMT-triggering factors and the actions exerted upon and within the cells. A striking case that exemplifies this concept has been shown in the study by Masszi and colleagues116 on TGF-β–induced EMT in a porcine renal proximal tubular cell line. By inducing the loss of epithelial integrity in confluent cultures, they found that the level of maturity of the cell junctions played a role in the TGF-β–induced EMT. More precisely, they showed that while TGF-β1 induced EMT in subconfluent or disrupted cultures, the cytokine did not induce EMT in confluent cultures. Though being associated with stress fiber formation, TGF-β treatment of confluent cultures did not induce α-SMA or fibronectin production or E-cadherin downregulation. By studying β-catenin implications in the phenomenon, they demonstrated that TGF-β1 treatment of subconfluent cultures protected β-catenin from degradation and allowed its signaling to take place. Once in the nucleus, β-catenin exerted a potentiating effect on the α-SMA promoter.116,117 Later, the same group showed that the mechanism responsible for the activation of α-SMA promoter in cells lacking contact was driven by Rho-ROCK myosin light chain (MLC) phosphorylation and cooperated with TGF-β1 signaling to enhance α-SMA expression.117 
Epidermal Growth Factor.
Epidermal growth factor has been shown to enhance human CECs proliferation in vitro.17 However, EGF is known to exert an influence in the EMT process. For example, in mammary epithelial cells, cyclooxygenase-2 (COX-2) is needed for TGF-β to induce EMT.118,119 However, there exists a synergy in the action of coupled COX-2 and TGF-β when EGF is present and activates the MAPK signaling.120 This demonstrates a possible synergic role of TGF-β and EGF in inducing EMT. Interestingly, COX-2 is expressed in the corneal endothelium under inflammatory conditions,121 and its expression is induced by FGF-2.122 Whether or not this phenomenon takes place in corneal EndMT remains to be elucidated. 
Of interest, EGF was shown to induce the expression of both TWIST and SNAIL1123,124 and, more importantly regarding corneal EndMT, EGF can induce translocation of β-catenin to the nucleus, therefore allowing the latter to fulfill its function as a transcription regulator.123,125 This effect of EGF on β-catenin signaling, taken together with FGF-2 and TGF-β involvement in similar processes, seems to point toward the Wnt/β-catenin signaling as an important regulator of corneal EndMT. The precise nature and mechanisms of Wnt/β-catenin signaling in CECs should be thoroughly investigated in order to achieve a better understanding of its interaction with the other EndMT triggers. 
Strategies Being Developed to Antagonize EndMT
In order to allow for the use of cultured human CECs in regenerative medicine, EndMT must be antagonized during the culture process. The EndMT appears to spontaneously take place in vitro, in a variable cell population-dependent manner, and seems to be related to the induction of CEC proliferation. Thus, strategies must be developed following the pursuit of maximal CEC expansion and minimal induction of corneal EndMT. In the next paragraphs, methods currently being developed to avoid EndMT in CEC cultures will be introduced with emphasis on its application to regenerative medicine. 
Inhibition of TGF-β Signaling
Since EndMT has been linked to pathway activation exerted by certain factors, the use of pathway-specific antagonists appears as a reasonable choice to inhibit undesired EndMT-induced loss of function of CECs in vitro. A recent study demonstrated the potential of SB431542, a specific TβRI (ALK5) kinase activity inhibitor, in abolishing the spontaneously occurring EndMT in in vitro human CECs.15 The authors demonstrated that SB431542 allowed the preservation of cell shape as well as the localization of both ZO-1 and Na+/K+-ATPase at the cell membrane from the second to the fifth passage of cell cultivation. Finally, they showed that SB431542 decreased both fibronectin and collagen I at the mRNA level when compared to control cultures and diminished collagen I protein production. 
In the same study, the authors evaluated the EndMT-inhibition potential of BMP-7, another member of the TGF-β superfamily. They showed that cell shape as well as localization of both ZO-1 and Na+/K+-ATPase at the cell membrane was preserved in human CECs treated with BMP-7.15 However, the mechanism of action remains unknown. 
Interestingly, corneal endothelial wound healing studies in in vitro and in vivo models have shown that addition of Smad7 could block TGF-β–induced fibrosis,126 hence showing the importance of the Smad2/3 activation in EMT/EndMT. Similar findings were obtained in the retinal pigment epithelium.127 
Unlocking the Mitotic Block
To isolate human CECs from donor corneas, many researchers follow the method first described by Chen and colleagues,128 which includes the stripping of Descemet's membrane followed by chemical disruption of cell junctions using EDTA and mechanical agitation. This process results in human CEC cultures exempt from stromal keratocyte contamination. This method of cell isolation, as was proposed by Tseng's team,49 could be implicated in the induction of EndMT in vitro because of the associated Wnt signaling activation. After having uncovered the link between contact-inhibition unlocking and proliferation with EndMT, Zhu and colleagues49 developed a way to resume human CEC proliferation in confluent explants, hence subtracting the need to disrupt cell junctions with EDTA. Indeed, they showed that treatment with p120-catenin (subsequently referred to as p120) small interfering ribonucleic acid (siRNA) resulted in reactivation of proliferation associated with conservation of a contact-inhibited phenotype.49 Moreover, even though addition of TGF-β alone or TGF-β and FGF-2 with p120 siRNA suppressed the proliferation induced by p120 siRNA alone, p120 siRNA abolished the EndMT whether TGF-β or FGF-2 was present or not. P120 siRNA achieved this protection from EndMT by the translocation of remnant p120 protein to the nucleus, leading to the expulsion of the transcriptional repressor Kaiso from the nucleus. More importantly, this treatment did not allow β-catenin to translocate to the nucleus. Thus, p120 siRNA acts as an inhibitor of Wnt canonical pathway and consequently prevents the activation of EndMT. 
The mechanism underlying the activity of p120 siRNA is related to RhoA activation since concomitant treatment with Y27632, a ROCK inhibitor, antagonized p120 siRNA-induced proliferation. Recently, it was reported that BMP noncanonical signaling was also involved in this process through demonstration of a causative link between p120 siRNA treatment and BMP2 and 4 upregulation. This led to translocation of phosphorylated (p65) NF-κB to the nucleus.129 Finally, the translocation of p120 to the nucleus caused by p120 siRNA treatment inhibited the Hippo signaling, a pathway in contact-inhibition and organ size regulation,130 by triggering the translocation of unphosphorylated YAP/TAZ by an unknown mechanism. 
ROCK Inhibitor
In the last years, a great deal of interest has been shown in the use of a Rho/ROCK pathway inhibitor (Y27632) in order to ameliorate CEC adhesion capabilities as well as preservation of contact-inhibited cell morphology. By inhibiting the Rho pathway with Y27632 in culture, Okumura and colleagues14 have shown that monkey CECs exhibited greater adhering capabilities, were less prone to apoptosis, and benefited from an increased proliferation. More recently, this team developed a model of postinjury rabbit and monkey CEC injection as a potential way to administer cells as a therapy.131 In this study, they showed that injection of Y27632 with the CECs in the anterior chamber of the eye allowed for better adhesion capabilities as well as preservation of cellular morphology (otherwise compromised without the use of Y27632), leading to recovery of a contact-inhibited corneal endothelium. Additionally, this study provided evidence that nontreated injected CECs might have undergone EndMT since they expressed α-SMA; this phenomenon was not reported with the injected cells treated with Y27632. Later, the same team showed that Y27632 applied topically as 10 mM drops on monkey corneas, whose endothelium had been injured by transcorneal freeze, enhanced the rate of wound closure and promoted contact-inhibited endothelial morphology compared to fellow PBS-treated injured cornea.132 To conclude, their work provides evidence that Y27632 might be useful in promoting survival of grafted as well as cultured CECs while maintaining contact-inhibited morphology, especially in the context of bioengineered corneal endothelial graft. 
Connexin 43
Connexin43 (Cx43) is a component of CEC gap junctions.133 It has been shown that its expression is dynamically regulated during wound healing of the rabbit corneal epithelium.134 In an attempt to accelerate injured corneal endothelium wound closure, Nakano and colleagues48 exploited this finding by inducing the knockdown of Cx43 in a rat corneal endothelial scrape injury. They found that Cx43 knockdown not only accelerated wound closure by inducing proliferation but also diminished the proportion of α-SMA–expressing cells, demonstrating a far less important EndMT when compared to control. They underlined that knockdown of Cx43 caused a decrease in p27kip1, a process implicated in the induction of proliferation. Despite this decreased p27kip1 expression, the general mechanism underlying the effect of Cx43 knockdown treatment remains unknown. Nonetheless, this finding seems to indicate a role played by Cx43 in corneal EndMT. 
Conclusions
Corneal endothelial cells that have lost their endothelial phenotype in vitro clearly exhibit many of the classical features of EndMT. It is possible to draw links between the enormous number of results that have been obtained using other cell types, as well as using CECs. While there is still some work to do in order to clarify corneal EndMT, studies conducted on the roles of some cytokines (TGF-β, FGF-2) have shown that it is possible to block many signaling pro-EndMT pathways in order to improve morphology, phenotype, and functionality of cultured CECs. The importance of cell–cell junctions in the maintenance of the endothelial phenotype is of high interest. The related activation of the Wnt/β-catenin pathway seems like a good candidate to explain many of the phenotypic events observed in in vitro CECs. 
There is an increasing need to find ways to block EndMT in vitro while promoting expansion of the CECs in order to use these cells in the engineering of a functional corneal endothelium. Moreover, these studies, while providing a better understanding of the in vitro phenomenon of corneal EndMT, could eventually help in the study and the treatment of related pathological events observed in vivo. 
Acknowledgments
Supported by Natural Sciences and Engineering Research Council of Canada (NSERC) Grant 418598-2012 (SP). VBL holds a studentship from the Fonds Wilbrod-Bhérer of Laval University. OR and J-MB hold studentships from the Fonds de la Recherche du Québec-Santé (FRQS). 
Disclosure: O. Roy, None; V. Beaulieu Leclerc, None; J.-M. Bourget, None; M. Thériault, None; S. Proulx, None 
References
Srinivas SP. Dynamic regulation of barrier integrity of the corneal endothelium. Optom Vis Sci. 2010; 87: E239–E254. [PubMed]
Joyce NC Meklir B Joyce SJ Zieske JD. Cell cycle protein expression and proliferative status in human corneal cells. Invest Ophthalmol Vis Sci. 1996; 37: 645–655. [PubMed]
Edelhauser HF. The balance between corneal transparency and edema: the Proctor Lecture. Invest Ophthalmol Vis Sci. 2006; 47: 1754–1767. [CrossRef] [PubMed]
Mishima S. Clinical investigations on the corneal endothelium-XXXVIII Edward Jackson Memorial Lecture. Am J Ophthalmol. 1982; 93: 1–29. [CrossRef] [PubMed]
2013 Eye Banking Statistical Report. Washington, DC: Eye Bank Association of America; 2014: 114.
Proulx S Brunette I. Methods being developed for preparation, delivery and transplantation of a tissue-engineered corneal endothelium. Exp Eye Res. 2012; 95: 68–75. [CrossRef] [PubMed]
Okumura N Kinoshita S Koizumi N. Cell-based approach for treatment of corneal endothelial dysfunction. Cornea. 2014; 33 (suppl 11): S37–S41. [CrossRef] [PubMed]
Mimura T Yamagami S Amano S. Corneal endothelial regeneration and tissue engineering. Prog Retin Eye Res. 2013; 35: 1–17. [CrossRef] [PubMed]
Joyce NC Zhu CC. Human corneal endothelial cell proliferation: potential for use in regenerative medicine. Cornea. 2004; 23: S8–S19. [CrossRef] [PubMed]
Peh GS Toh KP Wu FY Tan DT Mehta JS. Cultivation of human corneal endothelial cells isolated from paired donor corneas. PLoS One. 2011; 6: e28310. [CrossRef] [PubMed]
Proulx S Bourget JM Gagnon N Optimization of culture conditions for porcine corneal endothelial cells. Mol Vis. 2007; 13: 524–533. [PubMed]
Proulx S Audet C Uwamaliya J Tissue engineering of feline corneal endothelium using a devitalized human cornea as carrier. Tissue Eng Part A. 2009; 15: 1709–1718. [CrossRef] [PubMed]
Nakahara M Okumura N Kay EP Corneal endothelial expansion promoted by human bone marrow mesenchymal stem cell-derived conditioned medium. PLoS One. 2013; 8: e69009. [CrossRef] [PubMed]
Okumura N Ueno M Koizumi N Enhancement on primate corneal endothelial cell survival in vitro by a ROCK inhibitor. Invest Ophthalmol Vis Sci. 2009; 50: 3680–3687. [CrossRef] [PubMed]
Okumura N Kay EP Nakahara M Inhibition of TGF-beta signaling enables human corneal endothelial cell expansion in vitro for use in regenerative medicine. PLoS One. 2013; 8: e58000. [CrossRef] [PubMed]
Engelmann K Friedl P. Optimization of culture conditions for human corneal endothelial cells. In Vitro Cell Dev Biol. 1989; 25: 1065–1072. [CrossRef] [PubMed]
Zhu C Joyce NC. Proliferative response of corneal endothelial cells from young and older donors. Invest Ophthalmol Vis Sci. 2004; 45: 1743–1751. [CrossRef] [PubMed]
Engelmann K Bednarz J Valtink M. Prospects for endothelial transplantation. Exp Eye Res. 2004; 78: 573–578. [CrossRef] [PubMed]
Lamouille S Xu J Derynck R. Molecular mechanisms of epithelial-mesenchymal transition. Nat Rev Mol Cell Biol. 2014; 15: 178–196. [CrossRef] [PubMed]
Piera-Velazquez S Li Z Jimenez SA. Role of endothelial-mesenchymal transition (EndoMT) in the pathogenesis of fibrotic disorders. Am J Pathol. 2011; 179: 1074–1080. [CrossRef] [PubMed]
Lee JG Ko MK Kay EP. Endothelial mesenchymal transformation mediated by IL-1beta-induced FGF-2 in corneal endothelial cells. Exp Eye Res. 2012; 95: 35–39. [CrossRef] [PubMed]
Greenburg G Hay ED. Epithelia suspended in collagen gels can lose polarity and express characteristics of migrating mesenchymal cells. J Cell Biol. 1982; 95: 333–339. [CrossRef] [PubMed]
Hay ED. An overview of epithelio-mesenchymal transformation. Acta Anat (Basel). 1995; 154: 8–20. [CrossRef] [PubMed]
Gavert N Ben-Ze'ev A. Epithelial-mesenchymal transition and the invasive potential of tumors. Trends Mol Med. 2008; 14: 199–209. [CrossRef] [PubMed]
O'Connor JW Gomez EW. Biomechanics of TGFbeta-induced epithelial-mesenchymal transition: implications for fibrosis and cancer. Clin Transl Med. 2014; 3: 23. [CrossRef] [PubMed]
Zeisberg EM Potenta S Xie L Zeisberg M Kalluri R. Discovery of endothelial to mesenchymal transition as a source for carcinoma-associated fibroblasts. Cancer Res. 2007; 67: 10123–10128. [CrossRef] [PubMed]
van Meeteren LA ten Dijke P. Regulation of endothelial cell plasticity by TGF-beta. Cell Tissue Res. 2012; 347: 177–186. [CrossRef] [PubMed]
Potenta S Zeisberg E Kalluri R. The role of endothelial-to-mesenchymal transition in cancer progression. Br J Cancer. 2008; 99: 1375–1379. [CrossRef] [PubMed]
Huang RY Guilford P Thiery JP. Early events in cell adhesion and polarity during epithelial-mesenchymal transition. J Cell Sci. 2012; 125: 4417–4422. [CrossRef] [PubMed]
Gumbiner BM. Cell adhesion: the molecular basis of tissue architecture and morphogenesis. Cell. 1996; 84: 345–357. [CrossRef] [PubMed]
Yilmaz M Christofori G. EMT the cytoskeleton, and cancer cell invasion. Cancer Metastasis Rev. 2009; 28: 15–33. [CrossRef] [PubMed]
Kourtidis A Ngok SP Anastasiadis PZ. p120 catenin: an essential regulator of cadherin stability, adhesion-induced signaling, and cancer progression. Prog Mol Biol Transl Sci. 2013; 116: 409–432. [PubMed]
Niehrs C. The complex world of WNT receptor signalling. Nat Rev Mol Cell Biol. 2012; 13: 767–779. [CrossRef] [PubMed]
D'Souza-Schorey C. Disassembling adherens junctions: breaking up is hard to do. Trends Cell Biol. 2005; 15: 19–26. [CrossRef] [PubMed]
Wheelock MJ Shintani Y Maeda M Fukumoto Y Johnson KR. Cadherin switching. J Cell Sci. 2008; 121: 727–735. [CrossRef] [PubMed]
Zhu YT Hayashida Y Kheirkhah A Characterization and comparison of intercellular adherent junctions expressed by human corneal endothelial cells in vivo and in vitro. Invest Ophthalmol Vis Sci. 2008; 49: 3879–3886. [CrossRef] [PubMed]
St Johnston D Ahringer J. Cell polarity in eggs and epithelia: parallels and diversity. Cell. 2010; 141: 757–774. [CrossRef] [PubMed]
Moreno-Bueno G Portillo F Cano A. Transcriptional regulation of cell polarity in EMT and cancer. Oncogene. 2008; 27: 6958–6969. [CrossRef] [PubMed]
Peh GS Toh KP Ang HP Optimization of human corneal endothelial cell culture: density dependency of successful cultures in vitro. BMC Res Notes. 2013; 6: 176. [CrossRef] [PubMed]
Arita T Okamura R Kodama R Density dependent growth of corneal endothelial cells cultured in vitro. Cell Differ. 1987; 22: 61–69. [CrossRef] [PubMed]
McNiven MA. Breaking away: matrix remodeling from the leading edge. Trends Cell Biol. 2013; 23: 16–21. [CrossRef] [PubMed]
Ridley AJ. Life at the leading edge. Cell. 2011; 145: 1012–1022. [CrossRef] [PubMed]
Thiery JP Sleeman JP. Complex networks orchestrate epithelial-mesenchymal transitions. Nat Rev Mol Cell Biol. 2006; 7: 131–142. [CrossRef] [PubMed]
Yilmaz M Christofori G. Mechanisms of motility in metastasizing cells. Mol Cancer Res. 2010; 8: 629–642. [CrossRef] [PubMed]
Godde NJ Galea RC Elsum IA Humbert PO. Cell polarity in motion: redefining mammary tissue organization through EMT and cell polarity transitions. J Mammary Gland Biol Neoplasia. 2010; 15: 149–168. [CrossRef] [PubMed]
Nelson WJ. Remodeling epithelial cell organization: transitions between front-rear and apical-basal polarity. Cold Spring Harb Perspect Biol. 2009; 1: a000513. [CrossRef] [PubMed]
Ghosh AK Nagpal V Covington JW Michaels MA Vaughan DE. Molecular basis of cardiac endothelial-to-mesenchymal transition (EndMT): differential expression of microRNAs during EndMT. Cell Signal. 2012; 24: 1031–1036. [CrossRef] [PubMed]
Nakano Y Oyamada M Dai P Connexin43 knockdown accelerates wound healing but inhibits mesenchymal transition after corneal endothelial injury in vivo. Invest Ophthalmol Vis Sci. 2008; 49: 93–104. [CrossRef] [PubMed]
Zhu YT Chen HC Chen SY Tseng SC. Nuclear p120 catenin unlocks mitotic block of contact-inhibited human corneal endothelial monolayers without disrupting adherent junctions. J Cell Sci. 2012; 125: 3636–3648. [CrossRef] [PubMed]
Li C Dong F Jia Y Notch signal regulates corneal endothelial-to-mesenchymal transition. Am J Pathol. 2013; 183: 786–795. [CrossRef] [PubMed]
Thiery JP Acloque H Huang RY Nieto MA. Epithelial-mesenchymal transitions in development and disease. Cell. 2009; 139: 871–890. [CrossRef] [PubMed]
Lee JG Kay EP. FGF-2-mediated signal transduction during endothelial mesenchymal transformation in corneal endothelial cells. Exp Eye Res. 2006; 83: 1309–1316. [CrossRef] [PubMed]
Petroll WM Jester JV Bean JJ Cavanagh HD. Myofibroblast transformation of cat corneal endothelium by transforming growth factor-beta1, -beta2, and -beta3. Invest Ophthalmol Vis Sci. 1998; 39: 2018–2032. [PubMed]
Petroll WM Jester JV Barry-Lane PA Cavanagh HD. Effects of basic FGF and TGF beta 1 on F-actin and ZO-1 organization during cat endothelial wound healing. Cornea. 1996; 15: 525–532. [PubMed]
Usui T Takase M Kaji Y Extracellular matrix production regulation by TGF-beta in corneal endothelial cells. Invest Ophthalmol Vis Sci. 1998; 39: 1981–1989. [PubMed]
Kay ED Cheung CC Jester JV Nimni ME Smith RE. Type I collagen and fibronectin synthesis by retrocorneal fibrous membrane. Invest Ophthalmol Vis Sci. 1982; 22: 200–212. [PubMed]
Sabet MD Gordon SR. Ultrastructural immunocytochemical localization of fibronectin deposition during corneal endothelial wound repair. Evidence for cytoskeletal involvement. Biol Cell. 1989; 65: 171–179. [CrossRef] [PubMed]
Xu J Lamouille S Derynck R. TGF-beta-induced epithelial to mesenchymal transition. Cell Res. 2009; 19: 156–172. [CrossRef] [PubMed]
Derynck R Zhang YE. Smad-dependent and Smad-independent pathways in TGF-beta family signalling. Nature. 2003; 425: 577–584. [CrossRef] [PubMed]
Massague J. TGFbeta signalling in context. Nat Rev Mol Cell Biol. 2012; 13: 616–630. [CrossRef] [PubMed]
Wrighton KH Lin X Feng XH. Phospho-control of TGF-beta superfamily signaling. Cell Res. 2009; 19: 8–20. [CrossRef] [PubMed]
Peinado H Olmeda D Cano A. Snail, Zeb and bHLH factors in tumour progression: an alliance against the epithelial phenotype? Nat Rev Cancer. 2007; 7: 415–428. [CrossRef] [PubMed]
Cano A Perez-Moreno MA Rodrigo I The transcription factor snail controls epithelial-mesenchymal transitions by repressing E-cadherin expression. Nat Cell Biol. 2000; 2: 76–83. [CrossRef] [PubMed]
Whiteman EL Liu CJ Fearon ER Margolis B. The transcription factor snail represses Crumbs3 expression and disrupts apico-basal polarity complexes. Oncogene. 2008; 27: 3875–3879. [CrossRef] [PubMed]
De Craene B Gilbert B Stove C The transcription factor snail induces tumor cell invasion through modulation of the epithelial cell differentiation program. Cancer Res. 2005; 65: 6237–6244. [CrossRef] [PubMed]
Oft M Peli J Rudaz C TGF-beta1 and Ha-Ras collaborate in modulating the phenotypic plasticity and invasiveness of epithelial tumor cells. Genes Dev. 1996; 10: 2462–2477. [CrossRef] [PubMed]
Grande M Franzen A Karlsson JO Transforming growth factor-beta and epidermal growth factor synergistically stimulate epithelial to mesenchymal transition (EMT) through a MEK-dependent mechanism in primary cultured pig thyrocytes. J Cell Sci. 2002; 115: 4227–4236. [CrossRef] [PubMed]
Xie L Law BK Chytil AM Activation of the Erk pathway is required for TGF-beta1-induced EMT in vitro. Neoplasia. 2004; 6: 603–610. [CrossRef] [PubMed]
Uttamsingh S Bao X Nguyen KT Synergistic effect between EGF and TGF-beta1 in inducing oncogenic properties of intestinal epithelial cells. Oncogene. 2008; 27: 2626–2634. [CrossRef] [PubMed]
Yu L Hebert MC Zhang YE. TGF-beta receptor-activated p38 MAP kinase mediates Smad-independent TGF-beta responses. EMBO J. 2002; 21: 3749–3759. [CrossRef] [PubMed]
Bakin AV Rinehart C Tomlinson AK Arteaga CL. p38 mitogen-activated protein kinase is required for TGFbeta-mediated fibroblastic transdifferentiation and cell migration. J Cell Sci. 2002; 115: 3193–3206. [PubMed]
Santibanez JF. JNK mediates TGF-beta1-induced epithelial mesenchymal transdifferentiation of mouse transformed keratinocytes. FEBS Lett. 2006; 580: 5385–5391. [CrossRef] [PubMed]
Lamouille S Derynck R. Cell size and invasion in TGF-beta-induced epithelial to mesenchymal transition is regulated by activation of the mTOR pathway. J Cell Biol. 2007; 178: 437–451. [CrossRef] [PubMed]
Lamouille S Connolly E Smyth JW Akhurst RJ Derynck R. TGF-beta-induced activation of mTOR complex 2 drives epithelial-mesenchymal transition and cell invasion. J Cell Sci. 2012; 125: 1259–1273. [CrossRef] [PubMed]
Ridley AJ Hall A. The small GTP-binding protein rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors. Cell. 1992; 70: 389–399. [CrossRef] [PubMed]
Bhowmick NA Ghiassi M Bakin A Transforming growth factor-beta1 mediates epithelial to mesenchymal transdifferentiation through a RhoA-dependent mechanism. Mol Biol Cell. 2001; 12: 27–36. [CrossRef] [PubMed]
Burridge K Wennerberg K. Rho and Rac take center stage. Cell. 2004; 116: 167–179. [CrossRef] [PubMed]
Das S Becker BN Hoffmann FM Mertz JE. Complete reversal of epithelial to mesenchymal transition requires inhibition of both ZEB expression and the Rho pathway. BMC Cell Biol. 2009; 10: 94. [CrossRef] [PubMed]
Joyce NC Zieske JD. Transforming growth factor-beta receptor expression in human cornea. Invest Ophthalmol Vis Sci. 1997; 38: 1922–1928. [PubMed]
Joko T Shiraishi A Akune Y Involvement of P38MAPK in human corneal endothelial cell migration induced by TGF-beta(2). Exp Eye Res. 2013; 108: 23–32. [CrossRef] [PubMed]
Jampel HD Roche N Stark WJ Roberts AB. Transforming growth factor-beta in human aqueous humor. Curr Eye Res. 1990; 9: 963–969. [CrossRef] [PubMed]
Cousins SW McCabe MM Danielpour D Streilein JW. Identification of transforming growth factor-beta as an immunosuppressive factor in aqueous humor. Invest Ophthalmol Vis Sci. 1991; 32: 2201–2211. [PubMed]
Tripathi RC Li J Chan WF Tripathi BJ. Aqueous humor in glaucomatous eyes contains an increased level of TGF-beta 2. Exp Eye Res. 1994; 59: 723–727. [CrossRef] [PubMed]
Picht G Welge-Luessen U Grehn F Lutjen-Drecoll E. Transforming growth factor beta 2 levels in the aqueous humor in different types of glaucoma and the relation to filtering bleb development. Graefes Arch Clin Exp Ophthalmol. 2001; 239: 199–207. [CrossRef] [PubMed]
Schultz-Cherry S Ribeiro S Gentry L Murphy-Ullrich JE. Thrombospondin binds and activates the small and large forms of latent transforming growth factor-beta in a chemically defined system. J Biol Chem. 1994; 269: 26775–26782. [PubMed]
Hiscott P Seitz B Schlotzer-Schrehardt U Naumann GO. Immunolocalisation of thrombospondin 1 in human, bovine and rabbit cornea. Cell Tissue Res. 1997; 289: 307–310. [CrossRef] [PubMed]
Crawford SE Stellmach V Murphy-Ullrich JE Thrombospondin-1 is a major activator of TGF-beta1 in vivo. Cell. 1998; 93: 1159–1170. [CrossRef] [PubMed]
Tervo T Sulonen J Valtones S Vannas A Virtanen I. Distribution of fibronectin in human and rabbit corneas. Exp Eye Res. 1986; 42: 399–406. [CrossRef] [PubMed]
Gordon SR. Fibronectin antibody labels corneal stromal collagen fibrils in situ along their length and circumference and demonstrates distinct staining along the cell and stromal interfaces of Descemet's membrane. Curr Eye Res. 2014; 39: 312–316. [CrossRef] [PubMed]
Dallas SL Sivakumar P Jones CJ Fibronectin regulates latent transforming growth factor-beta (TGF beta) by controlling matrix assembly of latent TGF beta-binding protein-1. J Biol Chem. 2005; 280: 18871–18880. [CrossRef] [PubMed]
Fontana L Chen Y Prijatelj P Fibronectin is required for integrin alphavbeta6-mediated activation of latent TGF-beta complexes containing LTBP-1. FASEB J. 2005; 19: 1798–1808. [CrossRef] [PubMed]
Harris DL Joyce NC. Transforming growth factor-beta suppresses proliferation of rabbit corneal endothelial cells in vitro. J Interferon Cytokine Res. 1999; 19: 327–334. [CrossRef] [PubMed]
Chen KH Harris DL Joyce NC. TGF-beta2 in aqueous humor suppresses S-phase entry in cultured corneal endothelial cells. Invest Ophthalmol Vis Sci. 1999; 40: 2513–2519. [PubMed]
Kim TY Kim WI Smith RE Kay ED. Role of p27(Kip1) in cAMP- and TGF-beta2-mediated antiproliferation in rabbit corneal endothelial cells. Invest Ophthalmol Vis Sci. 2001; 42: 3142–3149. [PubMed]
Joyce NC. Proliferative capacity of the corneal endothelium. Prog Retin Eye Res. 2003; 22: 359–389. [CrossRef] [PubMed]
Wendt MK Tian M Schiemann WP. Deconstructing the mechanisms and consequences of TGF-beta-induced EMT during cancer progression. Cell Tissue Res. 2012; 347: 85–101. [CrossRef] [PubMed]
Lawler S Feng XH Chen RH The type II transforming growth factor-beta receptor autophosphorylates not only on serine and threonine but also on tyrosine residues. J Biol Chem. 1997; 272: 14850–14859. [CrossRef] [PubMed]
Zhang YE. Non-Smad pathways in TGF-beta signaling. Cell Res. 2009; 19: 128–139. [CrossRef] [PubMed]
Bandyopadhyay B Han A Dai J TbetaRI/Alk5-independent TbetaRII signaling to ERK1/2 in human skin cells according to distinct levels of TbetaRII expression. J Cell Sci. 2011; 124: 19–24. [CrossRef] [PubMed]
Kay EP Nimni ME Smith RE. Modulation of endothelial cell morphology and collagen synthesis by polymorphonuclear leukocytes. Invest Ophthalmol Vis Sci. 1984; 25: 502–512. [PubMed]
Folkman J Klagsbrun M Sasse J A heparin-binding angiogenic protein--basic fibroblast growth factor--is stored within basement membrane. Am J Pathol. 1988; 130: 393–400. [PubMed]
Kay EP Gu X Ninomiya Y Smith RE. Corneal endothelial modulation: a factor released by leukocytes induces basic fibroblast growth factor that modulates cell shape and collagen. Invest Ophthalmol Vis Sci. 1993; 34: 663–672. [PubMed]
Kay EP Nimni ME Smith RE. Stability of collagen phenotype in morphologically modulated rabbit corneal endothelial cells. Invest Ophthalmol Vis Sci. 1984; 25: 495–501. [PubMed]
Lee HT Lee JG Na M Kay EP. FGF-2 induced by interleukin-1 beta through the action of phosphatidylinositol 3-kinase mediates endothelial mesenchymal transformation in corneal endothelial cells. J Biol Chem. 2004; 279: 32325–32332. [CrossRef] [PubMed]
Kay EP Gu X Smith RE. Corneal endothelial modulation: bFGF as direct mediator and corneal endothelium modulation factor as inducer. Invest Ophthalmol Vis Sci. 1994; 35: 2427–2435. [PubMed]
Lee JG Kay EP. Common and distinct pathways for cellular activities in FGF-2 signaling induced by IL-1beta in corneal endothelial cells. Invest Ophthalmol Vis Sci. 2009; 50: 2067–2076. [CrossRef] [PubMed]
Song JS Lee JG Kay EP. Induction of FGF-2 synthesis by IL-1beta in aqueous humor through P13-kinase and p38 in rabbit corneal endothelium. Invest Ophthalmol Vis Sci. 2010; 51: 822–829. [CrossRef] [PubMed]
Lee JG Kay EP. NF-kappaB is the transcription factor for FGF-2 that causes endothelial mesenchymal transformation in cornea. Invest Ophthalmol Vis Sci. 2012; 53: 1530–1538. [CrossRef] [PubMed]
Lee JG Kay EP. Two populations of p27 use differential kinetics to phosphorylate Ser-10 and Thr-187 via phosphatidylinositol 3-Kinase in response to fibroblast growth factor-2 stimulation. J Biol Chem. 2007; 282: 6444–6454. [CrossRef] [PubMed]
Lee JG Kay EP. Involvement of two distinct ubiquitin E3 ligase systems for p27 degradation in corneal endothelial cells. Invest Ophthalmol Vis Sci. 2008; 49: 189–196. [CrossRef] [PubMed]
Lee JG Song JS Smith RE Kay EP. Human corneal endothelial cells employ phosphorylation of p27(Kip1) at both Ser10 and Thr187 sites for FGF-2-mediated cell proliferation via PI 3-kinase. Invest Ophthalmol Vis Sci. 2011; 52: 8216–8223. [CrossRef] [PubMed]
Ko MK Kay EP. Regulatory role of FGF-2 on type I collagen expression during endothelial mesenchymal transformation. Invest Ophthalmol Vis Sci. 2005; 46: 4495–4503. [CrossRef] [PubMed]
Grant MB Khaw PT Schultz GS Adams JL Shimizu RW. Effects of epidermal growth factor, fibroblast growth factor, and transforming growth factor-beta on corneal cell chemotaxis. Invest Ophthalmol Vis Sci. 1992; 33: 3292–3301. [PubMed]
Lee JG Kay EP. FGF-2-induced wound healing in corneal endothelial cells requires Cdc42 activation and Rho inactivation through the phosphatidylinositol 3-kinase pathway. Invest Ophthalmol Vis Sci. 2006; 47: 1376–1386. [CrossRef] [PubMed]
Lee JG Kay EP. Cross-talk among Rho GTPases acting downstream of PI 3-kinase induces mesenchymal transformation of corneal endothelial cells mediated by FGF-2. Invest Ophthalmol Vis Sci. 2006; 47: 2358–2368. [CrossRef] [PubMed]
Masszi A Fan L Rosivall L Integrity of cell-cell contacts is a critical regulator of TGF-beta 1-induced epithelial-to-myofibroblast transition: role for beta-catenin. Am J Pathol. 2004; 165: 1955–1967. [CrossRef] [PubMed]
Fan L Sebe A Peterfi Z Cell contact-dependent regulation of epithelial-myofibroblast transition via the rho-rho kinase-phospho-myosin pathway. Mol Biol Cell. 2007; 18: 1083–1097. [CrossRef] [PubMed]
Neil JR Johnson KM Nemenoff RA Schiemann WP. Cox-2 inactivates Smad signaling and enhances EMT stimulated by TGF-beta through a PGE2-dependent mechanisms. Carcinogenesis. 2008; 29: 2227–2235. [CrossRef] [PubMed]
Tian M Schiemann WP. PGE2 receptor EP2 mediates the antagonistic effect of COX-2 on TGF-beta signaling during mammary tumorigenesis. FASEB J. 2010; 24: 1105–1116. [CrossRef] [PubMed]
Saha D Datta PK Sheng H Synergistic induction of cyclooxygenase-2 by transforming growth factor-beta1 and epidermal growth factor inhibits apoptosis in epithelial cells. Neoplasia. 1999; 1: 508–517. [CrossRef] [PubMed]
Sellers RS Silverman L Khan KN. Cyclooxygenase-2 expression in the cornea of dogs with keratitis. Vet Pathol. 2004; 41: 116–121. [CrossRef] [PubMed]
Daniel TO Liu H Morrow JD Crews BC Marnett LJ. Thromboxane A2 is a mediator of cyclooxygenase-2-dependent endothelial migration and angiogenesis. Cancer Res. 1999; 59: 4574–4577. [PubMed]
Lu Z Ghosh S Wang Z Hunter T. Downregulation of caveolin-1 function by EGF leads to the loss of E-cadherin, increased transcriptional activity of beta-catenin, and enhanced tumor cell invasion. Cancer Cell. 2003; 4: 499–515. [CrossRef] [PubMed]
Lo HW Hsu SC Xia W Epidermal growth factor receptor cooperates with signal transducer and activator of transcription 3 to induce epithelial-mesenchymal transition in cancer cells via up-regulation of TWIST gene expression. Cancer Res. 2007; 67: 9066–9076. [CrossRef] [PubMed]
Li Y Zhang X Polakiewicz RD Yao TP Comb MJ. HDAC6 is required for epidermal growth factor-induced beta-catenin nuclear localization. J Biol Chem. 2008; 283: 12686–12690. [CrossRef] [PubMed]
Sumioka T Ikeda K Okada Y Inhibitory effect of blocking TGF-beta/Smad signal on injury-induced fibrosis of corneal endothelium. Mol Vis. 2008; 14: 2272–2281. [PubMed]
Saika S Yamanaka O Nishikawa-Ishida I Effect of Smad7 gene overexpression on transforming growth factor beta-induced retinal pigment fibrosis in a proliferative vitreoretinopathy mouse model. Arch Ophthalmol. 2007; 125: 647–654. [CrossRef] [PubMed]
Chen KH Azar D Joyce NC. Transplantation of adult human corneal endothelium ex vivo: a morphologic study. Cornea. 2001; 20: 731–737. [CrossRef] [PubMed]
Zhu YT Han B Li F Knockdown of both p120 catenin and Kaiso promotes expansion of human corneal endothelial monolayers via RhoA-ROCK-noncanonical BMP-NFkappaB pathway. Invest Ophthalmol Vis Sci. 2014; 55: 1509–1518. [CrossRef] [PubMed]
Zeng Q Hong W. The emerging role of the hippo pathway in cell contact inhibition, organ size control, and cancer development in mammals. Cancer Cell. 2008; 13: 188–192. [CrossRef] [PubMed]
Okumura N Koizumi N Ueno M ROCK inhibitor converts corneal endothelial cells into a phenotype capable of regenerating in vivo endothelial tissue. Am J Pathol. 2012; 181: 268–277. [CrossRef] [PubMed]
Okumura N Koizumi N Kay EP The ROCK inhibitor eye drop accelerates corneal endothelium wound healing. Invest Ophthalmol Vis Sci. 2013; 54: 2493–2502. [CrossRef] [PubMed]
Williams K Watsky M. Gap junctional communication in the human corneal endothelium and epithelium. Curr Eye Res. 2002; 25: 29–36. [CrossRef] [PubMed]
Matic M Petrov IN Rosenfeld T Wolosin JM. Alterations in connexin expression and cell communication in healing corneal epithelium. Invest Ophthalmol Vis Sci. 1997; 38: 600–609. [PubMed]
Footnotes
 OR and VBL contributed equally to the work presented here and should therefore be regarded as equivalent authors.
Figure
 
Summary of the current knowledge on the molecular mechanisms that regulate endothelial-to-mesenchymal transition in cultured corneal endothelial cells. Multiple extracellular signals (TGF-β1-2-3, IL-1β, FGF-2, connexin 43) can influence cultured corneal endothelial cells. They act as trigger factors that can impact critical intracellular pathways (Smad, p38, Erk1/2, Wnt/β-catenin, NF-κB, p27kip1) and induce endothelial-to-mesenchymal transition. On one hand, the endothelial phenotype is characterized by a contact-inhibited monolayer of polygonal-shaped cells that are arrested in the G1 phase of the cell cycle and express cell junction proteins such as ZO-1 and N-cadherin. On the other hand, cells with a fibroblastic phenotype undergo changes in cell shape, actin cytoskeleton reorganization, and disassembly of cell–cell junctions. They also lose their contact inhibition. Furthermore, they present an increased cell motility and extracellular matrix production as well as changes in gene expression. TGF, transforming growth factor; IL, interleukin; FGF, fibroblast growth factor; Erk, extracellular signal-regulated kinase; NF, nuclear factor; ZO, zonula occludens; ECM, extracellular matrix.
Figure
 
Summary of the current knowledge on the molecular mechanisms that regulate endothelial-to-mesenchymal transition in cultured corneal endothelial cells. Multiple extracellular signals (TGF-β1-2-3, IL-1β, FGF-2, connexin 43) can influence cultured corneal endothelial cells. They act as trigger factors that can impact critical intracellular pathways (Smad, p38, Erk1/2, Wnt/β-catenin, NF-κB, p27kip1) and induce endothelial-to-mesenchymal transition. On one hand, the endothelial phenotype is characterized by a contact-inhibited monolayer of polygonal-shaped cells that are arrested in the G1 phase of the cell cycle and express cell junction proteins such as ZO-1 and N-cadherin. On the other hand, cells with a fibroblastic phenotype undergo changes in cell shape, actin cytoskeleton reorganization, and disassembly of cell–cell junctions. They also lose their contact inhibition. Furthermore, they present an increased cell motility and extracellular matrix production as well as changes in gene expression. TGF, transforming growth factor; IL, interleukin; FGF, fibroblast growth factor; Erk, extracellular signal-regulated kinase; NF, nuclear factor; ZO, zonula occludens; ECM, extracellular matrix.
Table
 
Extracellular Factors and Signaling Pathways and Their Effects on CEC Phenotype in Culture
Table
 
Extracellular Factors and Signaling Pathways and Their Effects on CEC Phenotype in Culture
Authors, Ref. Numbers Factors/Pathways Effect Cell Type, Passage Culture Method
Joko et al.80 TGF-β2/p38 MAPK Migration Human, P5 Primary explant
Joko et al.80 FGF-2/p38 MAPK, Erk1/2 Proliferation/migration Human, P5 Primary explant
Zhu et al.49 Cell junction disruption, TGF-β, FGF-2/Wnt EndMT Human, P0 Collagenase
Petroll et al.53,54 TGF-β1-2-3 Loss of endothelial phenotype, fibrosis, myofibroblast transformation Feline, P0 Ex vivo corneas
Okumura et al.15 TGF-β1/Smad2, p38 EndMT Human, simian, P2-5 Collagenase
Kay et al.100,103; Lee et al.104; Lee and Kay106; Song et al.107; Lee et al.21; Lee and Kay108; IL-1β, FGF-2/PI3K, p38, NF-κB EndMT, proliferation, migration Human Trypsine and/or collagenase-hyluronidase
Nakano et al.48 Connexin43/p27kip1 EndMT Rat In vivo* (wound healing)
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×