Blocking Mitochondrial Pyruvate Transport Alters Corneal Myofibroblast Phenotype: A New Target for Treating Fibrosis.

Purpose: The purpose of this study was to critically test the hypothesis that mitochondrial pyruvate carrier (MPC) function is essential for maintenance of the corneal myofibroblast phenotype in vitro and in vivo. Methods: Protein and mRNA for canonical profibrotic markers were assessed in cultured cat corneal myofibroblasts generated via transforming growth factor (TGF)-ß1 stimulation and treated with either the thiazolidinedione (TZD) troglitazone or the MPC inhibitor alpha-cyano-beta-(1-phenylindol-3-yl) acrylate (UK-5099). RNA sequencing was used to gain insight into signaling modules related to instructive, permissive, or corollary changes in gene expression following treatment. A feline photorefractive keratectomy (PRK) model of corneal wounding was used to test the efficacy of topical troglitazone at reducing α-smooth muscle actin (SMA)-positive staining when applied 2 to 4 weeks postoperatively, during peak fibrosis. Results: Troglitazone caused cultured myofibroblasts to adopt a fibroblast-like phenotype through a noncanonical, peroxisome proliferator-activated receptor (PPAR)-γ-independent mechanism. Direct MPC inhibition using UK-5099 recapitulated this effect, but classic inhibitors of oxidative phosphorylation (OXPHOS) did not. Gene Set Enrichment Analysis (GSEA) of RNA sequencing data converged on energy substrate utilization and the Mitochondrial Permeability Transition pore as key players in myofibroblast maintenance. Finally, troglitazone applied onto an established zone of active fibrosis post-PRK significantly reduced stromal α-SMA expression. Conclusions: Our results provide empirical evidence that metabolic remodeling in myofibroblasts creates selective vulnerabilities beyond simply mitochondrial energy production, and that these are critical for maintenance of the myofibroblast phenotype. For the first time, we provide proof-of-concept data showing that this remodeling can be exploited to treat existing corneal fibrosis via inhibition of the MPC.

Post-DSAEK optical changes: a comprehensive prospective analysis on the role of ocular wavefront aberrations, haze, and corneal thickness.

PURPOSE: The aim was to assess the visual impact of ocular wavefront aberrations, corneal thickness, and corneal light scatter prospectively after performing a Descemet stripping automated endothelial keratoplasty (DSAEK) in humans. METHODS: Data were obtained prospectively from 20 eyes preoperatively and at 1, 3, 6, and 12 months post-DSAEK. At each visit, the best spectacle-corrected visual acuity and visual acuity with glare (brightness acuity testing) were recorded, and ocular wavefront measurements and corneal optical coherence tomography (OCT) were performed. The magnitude and the sign of individual Zernike terms [higher-order aberrations (HOAs)] were determined. Epithelial, host stromal, donor stromal, and total corneal thicknesses were quantified. The brightness and intensity profiles of OCT images were generated to quantify light scatter in the whole cornea, subepithelial region, anterior and posterior host stroma, interface, and donor stroma. RESULTS: The mean best spectacle-corrected visual acuity and glare disability at low light levels improved from 1 to 12 months post-DSAEK. All corneal thicknesses and ocular lower-order aberrations and HOAs were found to be stable from 1 to 12 months, whereas total corneal, host stromal, and interface brightness intensities decreased significantly over the same period. A repeated measures analysis of variance performed across the follow-up period revealed that the change in scatter, but not the change in the HOAs, could account for the variability occurring in the acuity from 1 to 12 months post-DSAEK. CONCLUSIONS: Although ocular HOAs and scatter are both elevated over normal values post-DSAEK, our results demonstrate that the improvements in visual performance occurring over the first year post-DSAEK are associated with decreasing light scatter. In contrast, there were no significant changes in the ocular HOAs during this time. Because corneal light scatter decreased between 1 and 12 months despite there being stable corneal thicknesses over the same period, we conclude that factors that induced light scatter, other than tissue thickness or swelling (corneal edema), significantly impacted the visual improvements that occurred over time post-DSAEK. A better understanding of the cellular and extracellular matrix changes of the subepithelial region and interface, incurred by the surgical creation of a lamellar host-graft interface, and the subsequent healing of these tissues, is warranted.

Keratocyte apoptosis and not myofibroblast differentiation mark the graft/host interface at early time-points post-DSAEK in a cat model.

PURPOSE: To evaluate myofibroblast differentiation as an etiology of haze at the graft-host interface in a cat model of Descemet's Stripping Automated Endothelial Keratoplasty (DSAEK). METHODS: DSAEK was performed on 10 eyes of 5 adult domestic short-hair cats. In vivo corneal imaging with slit lamp, confocal, and optical coherence tomography (OCT) were performed twice weekly. Cats were sacrificed and corneas harvested 4 hours, and 2, 4, 6, and 9 days post-DSAEK. Corneal sections were stained with the TUNEL method and immunohistochemistry was performed for α-smooth muscle actin (α-SMA) and fibronectin with DAPI counterstain. RESULTS: At all in vivo imaging time-points, corneal OCT revealed an increase in backscatter of light and confocal imaging revealed an acellular zone at the graft-host interface. At all post-mortem time-points, immunohistochemistry revealed a complete absence of α-SMA staining at the graft-host interface. At 4 hours, extracellular fibronectin staining was identified along the graft-host interface and both fibronectin and TUNEL assay were positive within adjacent cells extending into the host stroma. By day 2, fibronectin and TUNEL staining diminished and a distinct acellular zone was present in the region of previously TUNEL-positive cells. CONCLUSIONS: OCT imaging consistently showed increased reflectivity at the graft-host interface in cat corneas in the days post-DSAEK. This was not associated with myofibroblast differentiation at the graft-host interface, but rather with apoptosis and the development of a subsequent acellular zone. The roles of extracellular matrix changes and keratocyte cell death and repopulation should be investigated further as potential contributors to the interface optical changes.