Amelioration of Fibrosis via S1P Inhibition Is Regulated by Inactivation of TGF-ß and SPL Pathways in the Human Cornea.

Human corneal fibrosis can lead to opacity and ultimately partial or complete vision loss. Currently, corneal transplantation is the only treatment for severe corneal fibrosis and comes with the risk of rejection and donor shortages. Sphingolipids (SPLs) are known to modulate fibrosis in various tissues and organs, including the cornea. We previously reported that SPLs are tightly related to both, transforming growth factor beta (TGF-ß) signaling and corneal fibrogenesis. The aim of this study was to investigate the effects of sphingosine-1-phosphate (S1P) and S1P inhibition on specific TGF-ß and SPL family members in corneal fibrosis. Healthy human corneal fibroblasts (HCFs) were isolated and cultured in EMEM + FBS + VitC (construct medium) on 3D transwells for 4 weeks. The following treatments were prepared in a construct medium: 0.1 ng/mL TGF-ß1 (ß1), 1 µM sphingosine-1-phosphate (S1P), and 5 µM Sphingosine kinase inhibitor 2 (I2). Five groups were tested: (1) control (no treatment); rescue groups; (2) ß1/S1P; (3) ß1/I2; prevention groups; (4) S1P/ß1; and (5) I2/ß1. Each treatment was administered for 2 weeks with one treatment and switched to another for 2 weeks. Using Western blot analysis, the 3D constructs were examined for the expression of fibrotic markers, SPL, and TGF-ß signaling pathway members. Scratch assays from 2D cultures were also utilized to evaluate cell migration We observed reduced fibrotic expression and inactivation of latent TGF-ß binding proteins (LTBPs), TGF-ß receptors, Suppressor of Mothers Against Decapentaplegic homologs (SMADs), and SPL signaling following treatment with I2 prevention and rescue compared to S1P prevention and rescue, respectively. Furthermore, we observed increased cell migration following stimulation with I2 prevention and rescue groups, with decreased cell migration following stimulation with S1P prevention and rescue groups after 12 h and 18 h post-scratch. We have demonstrated that I2 treatment reduced fibrosis and modulated the inactivation of LTBPs, TGF-ß receptors, SPLs, and the canonical downstream SMAD pathway. Further investigations are warranted in order to fully uncover the potential of utilizing SphK I2 as a novel therapy for corneal fibrosis.

Potentiation of Sphingolipids and TGF-ß in the human corneal stroma reveals intricate signaling pathway crosstalks.

Corneal haze brought on by fibrosis due to insult can lead to partial or complete vision loss. Currently, corneal transplantation is the gold standard for treating severe corneal fibrosis, which comes with the risk of rejection and the issue of donor tissue shortages. Sphingolipids (SPLs) are known to be associated with fibrosis in various tissues and organs, including the cornea. We previously reported that SPLs are tightly related to Transforming Growth Factor ß (TGF-ß) signaling and corneal fibrogenesis. This study aimed to elucidate the interplay of SPLs, specifically sphingosine-1-phosphate (S1P) signaling, and its' interactions with TGF-ß signaling through detailed analyses of the corresponding downstream signaling targets in the context of corneal fibrosis, in vitro. Healthy human corneal fibroblasts (HCFs) were isolated, plated on polycarbonate membranes, and stimulated with a stable Vitamin C derivative. The 3D constructs were treated with either 5 µM sphingosine-1-phosphate (S1P), 5 µM SPHK I2 (I2; inhibitor of sphingosine kinase 1, one of the two enzymes responsible for generating S1P in mammalian cells), 0.1 ng/mL TGF-ß1, or 0.1 ng/mL TGF-ß3. Cultures with control medium-only served as controls. All 3D constructs were examined for protein expression of fibrotic markers, SPLs, TGF-ßs, and relevant downstream signaling pathways. This data revealed no significant changes in any LTBP (latent TGF-ß binding proteins) expression when stimulated with S1P or I2. However, LTBP1 was significantly upregulated via stimulation of TGF-ß1 and TGF-ß3, whereas LTBP2 was significantly upregulated only with TGF-ß3 stimulation. Significant downregulation of TGF-ß receptor II (TGF-ßRII) following S1P stimulation but significant upregulation following I2 stimulation was observed. Following TGF-ß1, S1P, and I2 stimulation, phospho-SMAD2 (pSMAD2) was significantly downregulated. Furthermore, I2 stimulation led to significant downregulation of SMAD4. Adhesion/proliferation/transcription regulation targets, SRC, FAK, and pERK 1/2 were all significantly downregulated by exogenous S1P, whereas I2 only significantly downregulated FAK. Exogenous TGF-ß3 caused significant upregulation of AKT. Interestingly, both I2 and TGF-ß3 caused significant downregulation of JNK expression. Lastly, TGF-ß1 led to significant upregulation of sphingosine kinase 1 (SphK1) and sphingosine-1-phosphate receptor 3 (S1PR3), whereas TGF-ß3 caused significant upregulation of only SphK1. Together with previously published work from our group and others, S1P inhibition exhibits great potential as an efficacious anti-fibrotic modality in human corneal stromal ECM. The current findings shed further light on a very complex and rather incompletely investigated mechanism, and cement the intricate crosstalk between SPLs and TGF-ß in corneal fibrogenesis. Future studies will dictate the potential of utilizing SPLs/TGF-ß signaling modulators as novel therapeutics in corneal fibrosis.

Characterization of C-X-C chemokine receptor type 5 in the cornea and role in the inflammatory response after corneal injury.

C-X-C chemokine receptor type 5 (CXCR5) regulates inflammatory responses in ocular and non-ocular tissues. However, its expression and role in the cornea are still unknown. Here, we report the expression of CXCR5 in human cornea in vitro and mouse corneas in vivo, and its functional role in corneal inflammation using C57BL/6J wild-type (CXCR5+/+) and CXCR5-deficient (CXCR5-/-) mice, topical alkali injury, clinical eye imaging, histology, immunofluorescence, PCR, qRT-PCR, and western blotting. Human corneal epithelial cells, stromal fibroblasts, and endothelial cells demonstrated CXCR5 mRNA and protein expression in PCR, and Western blot analyses, respectively. To study the functional role of CXCR5 in vivo, mice were divided into four groups: Group-1 (CXCR5+/+ alkali injured cornea; n = 30), Group-2 (CXCR5-/- alkali injured cornea; n = 30), Group-3 (CXCR5+/+ naïve cornea; n = 30), and Group-4 (CXCR5-/- naïve cornea; n = 30). Only one eye was wounded with alkali. Clinical corneal evaluation and imaging were performed before and after injury. Mice were euthanized 4 h, 3 days, or 7 days after injury, eyes were excised and used for histology, immunofluorescence, and qRT-PCR. In clinical eye examinations, CXCR5-/- mouse corneas showed ocular health akin to the naïve corneas. Alkali injured CXCR5+/+ mouse corneas showed significantly increased mRNA (p < 0.001) and protein (p < 0.01 or p < 0.0001) levels of the CXCR5 compared to the naïve corneas. Likewise, alkali injured CXCR5-/- mouse corneas showed remarkably amplified inflammation in clinical eye exams in live animals. The histological and molecular analyses of these corneas post euthanasia exhibited markedly augmented inflammatory cells in H&E staining and significant CD11b + cells in immunofluorescence (p < 0.01 or < 0.05); and tumor necrosis factor-alpha (TNFα; p < 0.05), cyclooxygenase 2 (COX-2; p < 0.0001), interleukin (IL)-1ß (p < 0.0001), and IL-6 (p < 0.0001 or < 0.01) mRNA expression compared to the CXCR5+/+ mouse corneas. Interestingly, CXCR5-/- alkali injured corneas also showed altered mRNA expression of fibrotic alpha smooth muscle actin (α-SMA; p > 0.05) and angiogenic vascular endothelial growth factor (VEGF; p < 0.01) compared to the CXCR5+/+ alkali injured corneas. In summary, the CXCR5 gene is expressed in all three major layers of the cornea and appears to influence corneal inflammatory and repair events post-injury in vivo. More studies are warranted to tease the mechanistic role of CXCR5 in corneal inflammation and wound healing.

Transforming growth factor beta-3 localization in the corneal response to epithelial-stromal injury and effects on corneal fibroblast transition to myofibroblasts.

The purpose of this study was to evaluate the localization of TGF beta-3 in situ in unwounded rabbit corneas and corneas that had epithelial-stromal injuries produced by photorefractive keratectomy (PRK) in rabbits and to evaluate the in vitro effects of TGF beta-3 compared to TGF beta-1 on alpha-smooth muscle actin (α-SMA) protein expression and myofibroblast development in corneal fibroblasts. Forty-eight New Zealand white rabbits underwent either -3 diopter (D) or -9D PRK and were studied from one to eight weeks (four corneas in each group at each time point) after surgery with immunohistochemistry for TGF beta-3, laminin alpha-5, and alpha-smooth muscle actin (α-SMA). Rabbit corneal fibroblasts were treated with activated TGF beta-1 and/or TGF beta-3 at different concentrations and duration of exposure and studied with immunocytochemistry for myofibroblast development and the expression of α-SMA using Jess automated Western blotting. TGF beta-3 was detected at high levels in the stroma of unwounded corneas and corneas at one to eight weeks after -3D or -9D PRK, as well as in the epithelium and epithelial basement membrane (EBM). No difference was noted between corneas that healed with and without myofibroblast-mediated fibrosis, although TGF beta-3 was commonly associated with myofibroblasts. TGF beta-3 effects on corneal fibroblasts in vitro were similar to TGF beta-1 in stimulating transition to α-SMA-positive myofibroblasts and promoting α-SMA protein expression. The corneal stromal localization pattern of TGF beta-3 protein in unwounded corneas and corneas after epithelial-stromal injury was found to be higher and different from TGF beta-1 and TGF beta-2 reported in previous studies. TGF beta-3 had similar effects to TGF beta-1 in driving myofibroblast development and α-SMA expression in corneal fibroblasts cultured in medium with 1% fetal bovine serum.

Corneal fibrosis abrogation by a localized AAV-mediated inhibitor of differentiation 3 (Id3) gene therapy in rabbit eyes in vivo.

Previously we found that inhibitor of differentiation 3 (Id3) gene, a transcriptional repressor, efficiently inhibits corneal keratocyte differentiation to myofibroblasts in vitro. This study evaluated the potential of adeno-associated virus 5 (AAV5)-mediated Id3 gene therapy to treat corneal scarring using an established rabbit in vivo disease model. Corneal scarring/fibrosis in rabbit eyes was induced by alkali trauma, and 24 h thereafter corneas were administered with either balanced salt solution AAV5-naked vector, or AAV5-Id3 vector (n = 6/group) via an optimized reported method. Therapeutic effects of AAV5-Id3 gene therapy on corneal pathology and ocular health were evaluated with clinical, histological, and molecular techniques. Localized AAV5-Id3 gene therapy significantly inhibited corneal fibrosis/haze clinically from 2.7 to 0.7 on the Fantes scale in live animals (AAV5-naked versus AAV5-Id3; p < 0.001). Furthermore, AAV5-Id3 treatment significantly reduced profibrotic gene mRNA levels: α-smooth muscle actin (α-SMA) (2.8-fold; p < 0.001), fibronectin (3.2-fold; p < 0.001), collagen I (0.8-fold; p < 0.001), and collagen III (1.4-fold; p < 0.001), as well as protein levels of α-SMA (23.8%; p < 0.001) and collagens (1.8-fold; p < 0.001). The anti-fibrotic activity of AAV5-Id3 is attributed to reduced myofibroblast formation by disrupting the binding of E-box proteins to the promoter of α-SMA, a transforming growth factor-ß signaling downstream target gene. In conclusion, these results indicate that localized AAV5-Id3 delivery in stroma caused no clinically relevant ocular symptoms or corneal cellular toxicity in the rabbit eyes.

Novel Correlation between TGF-ß1/-ß3 and Hormone Receptors in the Human Corneal Stroma.

This study investigated the interplay between transforming growth factor beta (TGF-ß1/T1 and TGF-ß3/T3), and sex hormone receptors using our 3D in vitro cornea stroma model. Primary human corneal fibroblasts (HCFs) from healthy donors were plated in transwells at 106 cells/well and cultured for four weeks. HCFs were supplemented with stable vitamin C (VitC) and stimulated with T1 or T3. 3D construct proteins were analyzed for the androgen receptor (AR), progesterone receptor (PR), estrogen receptor alpha (ERα) and beta (ERß), luteinizing hormone receptor (LHR), follicle-stimulating hormone receptor (FSHR), gonadotropin-releasing hormone receptor (GnRHR), KiSS1-derived peptide receptor (KiSS1R/GPR54), and follicle-stimulating hormone subunit beta (FSH-B). In female constructs, T1 significantly upregulated AR, PR, ERα, FSHR, GnRHR, and KiSS1R. In male constructs, T1 significantly downregulated FSHR and FSH-B and significantly upregulated ERα, ERß, and GnRHR. T3 caused significant upregulation in expressions PR, ERα, ERß, LHR, FSHR, and GNRHR in female constructs, and significant downregulation of AR, ERα, and FSHR in male constructs. Semi-quantitative Western blot findings present the interplay between sex hormone receptors and TGF-ß isoforms in the corneal stroma, which is influenced by sex as a biological variable (SABV). Additional studies are warranted to fully delineate their interactions and signaling mechanisms.

Evaluation of CRISPR/Cas9 mediated TGIF gene editing to inhibit corneal fibrosis in vitro.

Corneal wound healing is influenced by many factors including transcriptional co-repressors and co-activators. Interactions of co-activators and co-repressors with Smads influence mechanistic loop facilitating transcription of alpha-smooth muscle actin (α-SMA), a key profibrotic gene, in corneal repair. The role of a transcriptional repressor, 5'TG3'-interacting factor (TGIF), in the regulation of α-SMA and myofibroblast formation in the cornea was shown previously by our group. This study tested a hypothesis if TGIF1 gene editing via CRISPR/Cas9 can ease myofibroblast formation in the cornea using an in vitro model. Primary human corneal stromal fibroblasts (hCSFs) generated from donor corneas received gene-editing plasmid facilitating loss (CRISPR/Cas9 knockout) or gain (CRISPR activation) of TGIF function by UltraCruz transfection reagent. Phase-contrast microscopy, immunoblotting, immunocytochemistry and quantitative polymerase chain reaction (qPCR) were used to measure levels of myofibroblast profibrotic genes (α-SMA, fibronectin, Collagen-I, and Collagen-IV) in hCSFs lacking or overexpressing TGIF1 after growing them in± transforming growth factor beta1 (TGF-ß1) under serum-free conditions. The CRISPR-assisted TGIF1 activation (gain of function) in hCSFs demonstrated significantly decreased myofibroblast formation and messenger ribonucleic acid (mRNA) and protein levels of profibrotic genes. Conversely, CRISPR/Cas9-assisted TGIF knockdown (loss of function) in hCSFs demonstrated no significant change in the levels of myofibroblast formation or profibrotic genes under similar conditions. These results suggest that TGIF gene-editing approach can be employed to modulate the transcriptional activity of α-SMA in controlling pathological and promoting physiological wound healing in an injured cornea.

PCL scaffold combined with rat tail collagen type I to reduce keratocyte differentiation and prevent corneal stroma fibrosis after injury.

The cornea is one of the major refractive eye components and could be easily injured. An ineffective healing of corneal stromal wound may cause fibrosis and even loss of vision. Therefore, it is pivotal to prevent corneal fibrosis after injury. In this study, a poly (ε-caprolactone) (PCL) microfibrous scaffold infused with rat tail collagen type I was fabricated to obtain a 3D composite material. Physical and biological properties of PCL/collagen scaffold were evaluated, the effect of PCL/collagen scaffold on the proliferation and differentiation of limbal stromal stem cells (LSSCs) were detected in vitro, the differentiation of keratocytes as well as the expression and arrangement of extracellular matrix (ECM) influenced by PCL/collagen scaffold were investigated in vivo. RNA-sequencing on normal and injured corneas was carried out to find out the differential enriched pathways and gene expression. We discovered that the PCL/collagen scaffold simulated the stromal structure with properties that were most similar to the native cornea, the PCL/collagen scaffold exhibited good mechanical and biological properties. We also observed that the PCL/collagen scaffold reduced keratocyte differentiation. Injured corneas treated with PCL/collagen scaffold exhibited more regular collagen distribution and less fibroblasts and myofibroblasts distribution. By RNA-sequencing, we observed that in injured group, ECM-related pathway was enriched and several ECM-related genes were up-regulated. This study provides evidence that application of PCL/collagen scaffold could be a new therapeutic strategy for corneal injury.

Topical losartan inhibits corneal scarring fibrosis and collagen type IV deposition after Descemet's membrane-endothelial excision in rabbits.

The purpose of this study was to examine the effect of topical and/or oral angiotensin converting enzyme II inhibitor and TGF-beta signaling blocker losartan on corneal stromal fibrosis that developed in rabbit corneas after Descemetorhexis removal of central Descemet's membrane and corneal endothelium. Twenty-eight New Zealand white rabbits were included and either had 8 mm central Descemetorhexis or sham control surgery without Descemetorhexis in one eye. Groups of 4 eyes without Descemetorhexis were treated for one month with no medications, topical losartan or oral losartan. Groups of 4 eyes with Descemetorhexis were treated with topical and oral vehicle, topical losartan, oral losartan, or both topical losartan and oral losartan for one month. Standardized slit lamp photos were obtained with central opacity intensity measured with ImageJ. The posterior fibrotic zone of corneas was measured on immunohistochemistry for alpha-smooth muscle actin (SMA) and keratocan using QuPath analysis. Collagen type IV expression in the posterior cornea was quantitated with ImageJ and duplex immunohistochemistry for collagen type IV and TGF beta-1. After Descemetorhexis, topical, but not oral, losartan decreased the intensity of central stromal opacity, reduced peripheral corneal scarring, and decreased alpha-smooth muscle actin myofibroblast fibrosis area compared to corneas that had Descemetorhexis and treatment with vehicles alone. Topical losartan decreased posterior stromal cellular, non-Descemet's membrane, collagen type IV production, that is likely stimulated by TGF beta as part of a negative regulatory feedback mechanism, compared to vehicle treatment at one month after Descemetorhexis. Topical losartan is likely to be effective in reducing corneal scarring fibrosis produced by traumatic injury, microbial infection, and some corneal diseases and surgeries.

The Soluble Guanylate Cyclase Stimulator BAY 41-2272 Attenuates Transforming Growth Factor ß1-Induced Myofibroblast Differentiation of Human Corneal Keratocytes.

Corneal transparency, necessary for vision and depending on the high organization of stromal extracellular matrix, is maintained by keratocytes. Severe or continuous corneal injuries determine exaggerated healing responses resulting in the formation of irreversible fibrotic scars and vision impairment. Soluble guanylate cyclase (sGC) stimulation demonstrated antifibrotic effects in both experimental fibrosis and human lung and skin fibroblasts. Here, we assessed whether sGC stimulation with BAY 41-2272 could attenuate transforming growth factor ß1 (TGFß1)-induced myofibroblast differentiation of human corneal keratocytes. Cells were challenged with TGFß1, with/without BAY 41-2272 preincubation, and subsequently assessed for viability, proliferation, migration, chemoinvasion, as well for the expression of myofibroblast/fibroblast activation markers and contractile abilities. Treatment with BAY 41-2272 did not affect keratocyte viability, while preincubation of cells with the sGC stimulator was able to inhibit TGFß1-induced proliferation, wound healing capacity, and invasiveness. BAY 41-2272 was also able to attenuate TGFß1-induced myofibroblast-like profibrotic phenotype of keratocytes, as demonstrated by the significant decrease in ACTA2, COL1A1, COL1A2, FN1 and PDPN gene expression, as well as in α-smooth muscle actin, α-1 chain of type I collagen, podoplanin, vimentin and N-cadherin protein expression. Finally, BAY 41-2272 significantly counteracted the TGFß1-induced myofibroblast-like ability of keratocytes to contract collagen gels, reduced phosphorylated Smad3 protein levels, and attenuated gene expression of proinflammatory cytokines. Collectively, our data show for the first time that BAY 41-2272 is effective in counteracting keratocyte-to-myofibroblast transition, thus providing the rationale for the development of sGC stimulators as novel promising modulators of corneal scarring and fibrosis.

Descemet's membrane injury and regeneration, and posterior corneal fibrosis, in rabbits.

The purpose of this investigation was to study Descemet's membrane and corneal endothelial regeneration, myofibroblast generation and disappearance, and TGF beta-1 localization after Descemet's membrane-endothelial excision (Descemetorhexis) in rabbits. Thirty-six rabbits had 8 mm Descemetorhexis and standardized slit lamp photos at 1, 2 and 4 days, 1, 2 and 4 weeks, and 2, 4 and 6 months, as well as multiplex IHC for stromal cell markers keratocan, vimentin, and alpha-smooth muscle actin (SMA); basement membrane (BM) components perlecan, nidogen-1, laminin alpha-5, and collagen type IV; and corneal endothelial marker Na,K-ATPase ß1, and TGF beta-1, with ImageJ quantitation. Stromal transparency increased from the periphery beginning at two months after injury and progressed into the central cornea by six months. At six months, central transparency was primarily limited by persistent mid-stromal neovascularization. Stromal myofibroblast zone thickness in the posterior stroma peaked at one month after injury, and then progressively decreased until to six months when few myofibroblasts remained. The regeneration of a laminin alpha-5 and nidogen-1 Descemet's membrane "railroad track" structure was accompanied by corneal endothelial closure and stromal cell production of BM components in corneas from four to six months after injury. TGF beta-1 deposition at the posterior corneal surface from the aqueous humor peaked at one day after Descemetorhexis and diminished even before regeneration of the endothelium and Descemet's membrane. This decrease was associated with collagen type IV protein production by corneal fibroblasts, and possibly myofibroblasts, in the posterior stroma. Descemet's membrane and the corneal endothelium regenerated in the rabbit cornea by six months after eight mm Descemetorhexis. Real-time quantitative RT-PCR experiments in vitro with marker-verified rabbit corneal cells found that 5 ng/ml or 10 ng/ml TGF beta-1 upregulated col4a1 or col4a2 mRNA expression after 6 h or 12 h of exposure in corneal fibroblasts, but not in myofibroblasts. Stromal cells produced large amounts of collagen type IV that likely decreased TGF beta-1 penetration into the stroma and facilitated the resolution of myofibroblast-generated fibrosis.

Mitigating fibrosis-An impediment to corneal re-innervation following lamellar flap surgery.

Restoration of corneal sensitivity is of utmost importance to maintain corneal homeostasis following any injury or insult, for which, both corneal nerve regeneration and re-innervation are essential. Fibrosis poses a major impediment for re-innervation. We have in this study evaluated the influence of various nerve growth factors and corneal fibrosis on corneal nerve regeneration and reinnervation following lamellar flap surgery (LFS) and its modulation using antifibrotic drug pirfenidone. To achieve this, trigeminal ganglion cells were treated with pirfenidone, NGF, and NT-3 to evaluate their effect on trigeminal cell neurite growth. Following LFS, the gene expression of nerve growth factors NGF, BDNF and NT-3, Gap 43, Nogo-A and profibrotic factors Tenascin C, TGF-beta 1 were evaluated with and without pirfenidone. Wound fibrosis and corneal nerve regeneration using pirfenidone following LFS were evaluated by staining whole corneal mounts with α SMA and ß tubulin 3. Safety of NGF and pirfenidone topical drops in normal unoperated cornea and its efficacy in enhancing corneal healing was evaluated following LFS. Our study shows, pirfenidone did not influence trigeminal cell neurite elongation; NGF and NT-3 significantly enhanced trigeminal cell neurite elongation. NT-3 also significantly increased neurite branching. There was significant increase in the gene expression of NGF, BDNF, NT-3, Gap- 43, TGF beta-1, Tenascin C, Nogo-A genes in the operated cornea compared to normal cornea, treatment of operated corneas with pirfenidone prevented the increased expression of these genes except Gap 43 which remained unchanged. The treatment of operated eyes with combination of NGF and pirfenidone positively influenced corneal healing compared to treatment with NGF alone, and had no adverse influence on the cornea. Pirfenidone appreciably reduced corneal fibrosis which aided in re-innervation. Both NGF and NT3 positively influence trigeminal neurite elongation. NGF and pirfenidone have complementary influence on corneal wound healing.

Descemet's membrane development, structure, function and regeneration.

Basement membranes are layers of extracellular matrix which anchor the epithelium or endothelium to connective tissues in most organs. Descemet's membrane- which is the basement membrane for the corneal endothelium- is a dense, thick, relatively transparent and cell-free matrix that separates the posterior corneal stroma from the underlying endothelium. It was historically named Descemet's membrane after Jean Descemet, a French physician, but it is also known as the posterior limiting elastic lamina, lamina elastica posterior, and membrane of Demours. Normal Descemet's membrane ultrastructure in humans has been shown to consist of an interfacial matrix that attaches to the overlying corneal stroma, an anterior banded layer and a posterior non-banded layer-upon which corneal endothelial cells attach. These layers have been shown to have unique composition and morphology, and to contribute to corneal homeostasis and clarity, participate in the control of corneal hydration and to modulate TGF-ß-induced posterior corneal fibrosis. Pathophysiological alterations of Descemet's membrane are noted in ocular diseases such as Fuchs' dystrophy, bullous keratopathy, keratoconus, primary congenital glaucoma (Haab's striae), as well as in systemic conditions. Unrepaired extensive damage to Descemet's membrane results in severe corneal opacity and vision loss due to stromal fibrosis, which may require penetrating keratoplasty to restore corneal transparency. The purpose of this article is to highlight the current understanding of Descemet's membrane structure, function and potential for regeneration.

Corneal epithelial basement membrane: Structure, function and regeneration.

Basement membranes are highly specialized extracellular matrices. More than providing scaffolds, basement membranes are recognized as dynamic and versatile structures that modulate cellular responses to regulate tissue development, function, and repair. Increasing evidence suggests that, in addition to providing structural support to adjacent cells, basement membranes serve as reservoirs and modulators of growth factors that direct and fine-tune cellular functions. Since the corneal stroma is avascular and has a relatively low keratocyte density, it's likely that the corneal BM is different in composition from the BMs in other tissues. BMs are composed of a diverse assemblage of extracellular molecules, some of which are likely specific to the tissue where they function; but in general they are composed of four primary components-collagens, laminins, heparan sulfate proteoglycans, and nidogens-in addition to other components such as thrombospondin-1, matrilin-2, and matrilin-4 and fibronectin. Severe injuries to the cornea, including infection, surgery, and trauma, may trigger the development of myofibroblasts and fibrosis in the normally transparent connective tissue stroma. Ultrastructural studies have demonstrated that defective epithelial basement membrane (EBM) regeneration after injury to the cornea underlies the development of myofibroblasts from both bone marrow- and keratocyte-derived precursor cells. Defective EBM permits epithelium-derived and tear-derived transforming growth factor beta (TGF-ß), platelet-derived growth factor (PDGF), and possibly other modulators, to penetrate the stroma at sustained levels necessary to drive the development and persistence of vimentin + alpha-smooth muscle actin + desmin+ (V + A + D+) mature myofibroblasts. A recent discovery that has contributed to our understanding of haze development is that keratocytes and corneal fibroblasts produce critical EBM components, such as nidogen-1, nidogen-2 and perlecan, that are essential for complete regeneration of a normal EBM once laminin secreted by epithelial cells self-polymerizes into a nascent EBM. Mature myofibroblasts that become established in the anterior stroma are a barrier to keratocyte/corneal fibroblast contributions to the nascent EBM. These myofibroblasts, and the opacity they produce, often persist for months or years after the injury. Transparency is subsequently restored if the EBM is fully regenerated, myofibroblasts are deprived of TGF-ß and undergo apoptosis, and keratocytes reoccupy the anterior stroma and reabsorb the disordered extracellular matrix.

Desmin deficiency is not sufficient to prevent corneal fibrosis.

The type III intermediate filament (IF) proteins vimentin and desmin are sequentially overexpressed in stromal myofibroblasts over the period when fibrosis sets in after corneal injury. Prior findings have revealed vimentin-deficient mice are significantly protected from corneal fibrosis after alkali injury, which has implicated this IF protein as an important regulator of corneal fibrosis. It has remained as yet unproven whether desmin contributes in any significant manner to corneal fibrosis. Here we have employed desmin-deficient (Des KO) mice in the corneal alkali injury model and show that injured Des KO mice develop fibrosis and show similar levels of corneal opacity at 14 days post-injury as wild type (WT) mice and retain this phenotype even at 30d post injury. Des KO corneas from injured mice show upregulation of vimentin and alpha-smooth muscle actin expression to equivalent levels as WT corneas, illuminating that desmin deficiency does not interfere with myofibrobast differentiation. Employing the small molecule withaferin A (WFA), an inhibitor of vimentin, we show that WFA treatment causes the decrease in steady state levels of vimentin and serine 38 phosphorylated vimentin, the latter a biomarker associated with corneal fibrosis, and improved corneal clarity through blockade of myofibroblast differentiation. To investigate further the mechanism of fibrosis in desmin deficiency, we examined keratin 8 expression in the epithelium, and found reduced levels of this cytokeratin in injured Des KO corneas compared to WT corneas. This finding also corroborates the decrease of cell proliferation in injured Des KO corneas compared to that in WT corneas. The fibrotic phenotype of Des KO corneas also features abundant vascularization, further exemplifying the magnitude of corneal pathology. Together, these findings illuminate that desmin does not contribute significantly to corneal fibrosis in this injury model.

Sustained activation of ERK1/2 MAPK in Schwann cells causes corneal neurofibroma.

Recent studies have shown that constitutive activation of extracellular signal-regulated kinases 1 and 2 (ERK1/2) in Schwann cells (SCs) increases myelin thickness in transgenic mice. In this secondary analysis, we report that these transgenic mice develop a postnatal corneal neurofibroma with the loss of corneal transparency by age six months. We show that expansion of non-myelinating SCs, under the control of activated ERK1/2, also drive myofibroblast differentiation that derives from both SC precursors and resident corneal keratocytes. Further, these mice also harbor activated mast cells in the central cornea, which contributes to pathological corneal neovascularization and fibrosis. This breach of corneal avascularity and immune status is associated with the growth of the tumor pannus, resulting in a corneal stroma that is nearly four times its normal size. In corneas with advanced disease, some axons became ectopically myelinated, and the disruption of Remak bundles is evident. To determine whether myofibroblast differentiation was linked to vimentin, we examined the levels and phosphorylation status of this fibrotic biomarker. Concomitant with the early upregulation of vimentin, a serine 38-phosphorylated isoform of vimentin (pSer38vim) increased in SCs, which was attributed primarily to the soluble fraction of protein-not the cytoskeletal portion. However, the overexpressed pSer38vim became predominantly cytoskeletal with the growth of the corneal tumor. Our findings demonstrate an unrecognized function of ERK1/2 in the maintenance of corneal homeostasis, wherein its over-activation in SCs promotes corneal neurofibromas. This study is also the first report of a genetically engineered mouse that spontaneously develops a corneal tumor.

Complete metabolome and lipidome analysis reveals novel biomarkers in the human diabetic corneal stroma.

Prolonged hyperglycemia during diabetes mellitus can cause severe ophthalmic complications affecting both the anterior and posterior ocular segments leading to impaired vision or blindness. Diabetes-induced corneal pathologies are associated with decreased wound healing capacity, corneal edema, and altered epithelial basement membrane. The mechanism by which diabetes modulates structure and function within the corneal stroma are unknown. In our study, we characterized the effects of diabetes on extracellular matrix, lipid transport, and cellular metabolism by defining the entire metabolome and lipidome of Type 1 and Type 2 human diabetic corneal stroma. Significant increases in Collagen I and III were found in diabetic corneas suggesting that diabetes promotes defects in matrix structure leading to scarring. Furthermore, increased lipid content, including sphingosine-1-phosphate and dihydrosphingosine, in diabetic corneas compared to healthy controls were measured suggesting altered lipid retention. Metabolomics analysis identified elevated tryptophan metabolites, independent of glucose metabolism, which correlated with upregulation of the Kynurenine pathway in diabetic corneas. We also found significant upregulation of novel biomarkers aminoadipic acid, D,L-pipecolic acid, and dihydroorotate. Our study links aberrant tryptophan metabolism to end-stage pathologies associated with diabetes indicating the potential of the Kynurenine pathway as a therapeutic target for inhibiting diabetes-associated defects in the eye.

Development of a novel in vivo corneal fibrosis model in the dog.

The aim of this study was to develop a novel in vivo corneal model of fibrosis in dogs utilizing alkali burn and determine the ability of suberanilohydroxamic acid (SAHA) to inhibit corneal fibrosis using this large animal model. To accomplish this, we used seven research Beagle dogs. An axial corneal alkali burn in dogs was created using 1 N NaOH topically. Six dogs were randomly and equally assigned into 2 groups: A) vehicle (DMSO, 2 µL/mL); B) anti-fibrotic treatment (50 µM SAHA). The degree of corneal opacity, ocular health, and anti-fibrotic effects of SAHA were determined utilizing the Fantes grading scale, modified McDonald-Shadduck (mMS) scoring system, optical coherence tomography (OCT), corneal histopathology, immunohistochemistry (IHC), and transmission electron microscopy (TEM). The used alkali burn dose to produce corneal fibrosis was well tolerated as no significant difference in mMS scores between control and treatment groups (p = 0.89) were detected. The corneas of alkali burned dogs showed significantly greater levels of α-smooth muscle actin, the fibrotic marker, than the controls (p = 0.018). Total corneal thickness of all dogs post-burn was significantly greater than baseline OCT images irrespective of treatment (p = 0.004); TEM showed that alkali burned corneas had significantly greater minimum and maximum interfibrillar distances than the controls (p = 0.026, p = 0.018). The tested topical corneal alkali burn dose generated significant opacity and fibrosis in dog corneas without damaging the limbus as evidenced by histopathology, IHC, TEM, and OCT findings, and represents a viable large animal corneal fibrosis in vivo model. Additional in vivo SAHA dosing studies with larger sample size are warranted.

Characterization of Inhibitor of differentiation (Id) proteins in human cornea.

Inhibitor of differentiation (Id) proteins are DNA-binding transcription factors involved in cellular proliferation, migration, inflammation, angiogenesis and fibrosis. However, their expression and role in the cornea is unknown. The present study was undertaken to characterize the expression of Id proteins and their interactions with the pro-fibrotic cytokine Transforming Growth Factor ß1 (TGFß1) and anti-fibrotic cytokine, bone morphogenic protein 7 (BMP7) in human cornea. Human donor corneas procured from Eye Bank were used. Id proteins were localized in human corneal sections using immunofluorescence. Primary cultures of human corneal fibroblasts (HCF) were established and treated with either TGFß1 (5 ng/ml) or BMP7 (10 ng/ml) for 24 h in serum free medium. Expression of Id's in response to TGFß1, BMP7 and TGFß1 + BMP7 was analyzed by quantitative real time PCR (qRT-PCR) and western blot analysis. Id1 and Id2 proteins were ubiquitously expressed in the epithelial cells and stromal keratocytes in human cornea. The Id1 was localized to the basal epithelial cells as seen by immunohistochemistry. HCF expressed all known mammalian Id genes (Id1-Id4). In addition, Id1 and Id2 are selectively expressed in HCF. Treatment of human recombinant TGFß1 (5 ng/ml) to serum-starved HCF showed a significant increase in Id genes (Id1, Id2 and Id4) at 2 h time point compared to BMP7 treatment, which showed time dependent increase in the expression of Id1-Id3 at 24-48 h. Combined treatment with TGFß1 + BMP7 to HCF showed a significant increase in Id1 transcript and an increasing trend in Id3 and Id4 expression. The results of this study suggest that Id family of genes (Id1-Id4) are localized in the human cornea and expressed in the corneal fibroblasts. Also, Id's were differentially regulated with TGFß1 and/or BMP7 in a time dependent manner and might serve as a therapeutic target in corneal fibrosis.

Molecular mechanisms of suberoylanilide hydroxamic acid in the inhibition of TGF-ß1-mediated canine corneal fibrosis.

OBJECTIVE: To investigate molecular mechanisms mediating anti-fibrotic effect of SAHA in the canine cornea using an in vitro model. We hypothesized that SAHA attenuates corneal fibrosis by modulating Smad-dependent and, to a lesser extent, Smad-independent signaling pathways activated by TGF-ß1, as well as matrix metalloproteinase (MMP) activity. METHODS: Cultured canine corneal fibroblasts (CCF) were incubated in the presence/absence of TGF-ß1 (5 ng/mL) and SAHA (2.5 µm) for 24 h. Western blot analysis was used to quantify non-phosphorylated and phosphorylated isoforms of Smad2/3, p38 MAP kinase (MAPK), ERK1/2, and JNK1. Real-time PCR and zymography were utilized to quantify MMP1, MMP2, MMP8, and MMP9 mRNA expressions and MMP2 and MMP9 protein activities, respectively. RESULTS: TGF-ß1 treatment caused a significant increase in phospho-Smad2/3 and phospho-p38 MAPK. SAHA treatment reduced TGF-ß1-induced phosphorylation of Smad2/3 but not of p38 MAPK. TGF-ß1 did not modulate the phosphorylation of ERK1/2 or JNK1. SAHA caused a significant reduction in phospho-ERK1/2 expression regardless of concurrent TGF-ß1 treatment. Neither SAHA alone nor in combination with TGF-ß1 altered phospho-JNK1 expression. TGF-ß1 significantly increased MMP1 and MMP9 mRNA expressions but did not alter MMP2 mRNA. SAHA treatment attenuated TGF-ß1-induced MMP9 mRNA expression while significantly enhancing TGF-ß1-induced MMP1 mRNA expression. Zymography detected reduced expression of MMP2 and MMP9 proteins in untreated control CCF. TGF-ß1 treatment did not alter their expression, but SAHA treatment +/-TGF-ß1 significantly increased MMP2 and MMP9 protein expressions. CONCLUSIONS: The corneal anti-fibrotic effects of SAHA involve multiple mechanisms including modulation of canonical and non-canonical components of TGF-ß1 intracellular signaling and MMP activity.