Ex-vivo cultured human corneoscleral segment model to study the effects of glaucoma factors on trabecular meshwork

Abstract

Glaucoma is the second leading cause of irreversible blindness worldwide. Primary open angle glaucoma (POAG), the most common form of glaucoma, is often associated with elevation of intraocular pressure (IOP) due to the dysfunction of trabecular meshwork (TM) tissues. Currently, an ex vivo human anterior segment perfusion cultured system is widely used to study the effects of glaucoma factors and disease modifying drugs on physiological parameters like aqueous humor (AH) dynamics and IOP homeostasis. This system requires the use of freshly enucleated intact human eyes, which are sparsely available at very high cost. In this study, we explored the feasibility of using human donor corneoscleral segments for modeling morphological and biochemical changes associated with POAG. Among the number of corneas donated each year, many are deemed ineligible for transplantation due to stringent acceptance criteria. These ineligible corneoscleral segments were obtained from the Lions Eye Bank, Tampa, Florida. Each human donor anterior corneoscleral segment was dissected into four equal quadrants and cultured for 7 days by treating with the glaucoma factors dexamethasone (Dex) or recombinant transforming growth factor (TGF) β2 or transduced with lentiviral expression vectors containing wild type (WT) and mutant myocilin. Hematoxylin and Eosin (H&E) staining analysis revealed that the TM structural integrity is maintained after 7 days in culture. Increased TUNEL positive TM cells were observed in corneoscleral quadrants treated with glaucoma factors compared to their respective controls. However, these TUNEL positive cells were mainly confined to the scleral region adjacent to the TM. Treatment of corneoscleral quadrants with Dex or TGFβ2 resulted in glaucomatous changes at the TM, which included increased extracellular matrix (ECM) proteins and induction of endoplasmic reticulum (ER) stress. Western blot analysis of the conditioned medium showed an increase in ECM (fibronectin and collagen IV) levels in Dex- or TGFβ2-treated samples compared to control. Lentiviral transduction of quadrants resulted in expression of WT and mutant myocilin in TM tissues. Western blot analysis of conditioned medium revealed decreased secretion of mutant myocilin compared to WT myocilin. Moreover, increased ECM deposition and ER stress induction was observed in the TM of mutant myocilin transduced quadrants. Our findings suggest that the ex-vivo cultured human corneoscleral segment model is cost-effective and can be used as a pre-screening tool to study the effects of glaucoma factors and anti-glaucoma therapeutics on the TM.

Fig 1. Schematic showing the experimental design.

Human corneoscleral segments with an intact TM rim were dissected into 4 equal quadrants and each quadrant was cultured and treated with glaucoma-inducing factors for 7 days. Each quadrant was then either fixed, or TM tissue and conditioned medium was isolated. Fixed tissues were utilized for H&E and immunohistochemical analysis while TM tissue and conditioned medium was analyzed by Western blot.

Fig 2. TM structural integrity is maintained in an ex-vivo cultured corneoscleral segment.

H&E staining (upper panel) and TUNEL assay (lower panel) was performed in human corneoscleral quadrants cultured at zero days (A&D), 7 days with 10% FBS (B&E) and 7 days with 0.5% FBS (C&F) containing medium. H&E staining revealed no obvious morphological changes after 7-days of culture compared to zero-days of culture. The total number of TUNEL positive cells over TM cells (DAPI positive) in TM region were shown graphically (G). No difference in TUNEL-positive TM cells was observed between the zero- day and 7-day cultured group with 10% FBS. A significant increase in the TUNEL- positive cells was observed in quadrants cultured with 0.5% FBS. (n = 4 biological replicates per group, one-way ANOVA, *p<0.05), scale bar is 50μm. Arrows indicate the TM region; asterisks represent the Schlemm’s canal.

Fig 3. Dex- or TGFβ2-treatment leads to increased apoptosis of TM cells.

H&E staining (left panel) and TUNEL assay (right panel) of cultured corneoscleral segments treated with either Dex (100nM) (B&F) or TGFβ2 (5ng/ml) (D&H) is compared with corresponding quadrants treated with vehicle (A&E or C&G) for 7-days. The total number of TUNEL-positive cells in the TM region were shown graphically (I). Increased TUNEL-positive (green) cells at the TM region was observed in both Dex- and TGFβ2-treated groups. (n = 3 biological replicates per group), scale bar is 50μm. Arrows indicate the TM region; asterisks represent the Schlemm’s canal. N = 3 each; one-way ANOVA; *p<0.05, ***p<0.0001.

Fig 4. Increased ECM accumulation and ER stress induction in the TM of Dex-treated cultured corneoscleral segments.

A) Immunostaining for myocilin and fibronectin, B) GRP78 (ER stress marker) and collagen IV in quadrants cultured for 7 days treated with the vehicle (0.1% ethanol) and Dex (100nM). Dex treatment prominently increased myocilin, fibronectin, collagen IV and GRP78 staining in the TM region. (n = 4 biological replicates, scale bar is 50μm). Western blot and densitometric analysis for FN (ECM marker) and ATF4, CHOP and GRP78 (ER stress markers) in TM tissue lysates (C-D) of vehicle- and Dex- (100nM) treated cultured corneoscleral segments. Dex treatment led to a significant increase in the ECM marker, FN (n = 3 biological replicates) and ER stress markers CHOP and ATF4 but not GRP78 (n = 4 biological replicates). Note that densitometric analysis included only Dex responders. Similarly, conditioned medium (E-F) from Dex-treated corneoscleral segments showed a significant increase in ECM proteins FN and Col IV (n = 8); unpaired t-test, *P<0.05, **P<0.01, ***P<0.001). Arrows indicate the TM region.

Fig 5. Increased ECM accumulation in the TM of TGFβ2-treated cultured corneoscleral segments.

(A and B) Immunostaining for fibronectin (FN), collagen IV (Col-IV), αSMA and GRP78 in vehicle and TGFβ2 (5ng/ml) treated cultured corneoscleral segments. (n = 4 biological replicates, scale bar is 50μm). Western blot and densitometric analysis for FN, Col-IV (ECM markers), ATF4, CHOP, GRP78 in the TM tissue lysates (C-D) and conditioned medium (E-F) of vehicle and TGFβ2-treated cultured corneoscleral quadrants (n = 4 biological replicates for lysates and n = 8 for the conditioned medium), unpaired t-test, *P<0.05, **P<0.01, ***P<0.001. Arrows indicate the TM region.

Fig 6. Lentiviral expression of mutant myocilin induces ER stress and ECM changes in the TM of cultured corneoscleral segments.

The cultured corneoscleral quadrants were transduced with FTS tagged (FLAG & S tag) WT myocilin or mutant myocilin (Y437H) expressing lentiviral particles (1ml of lentivirus supernatant) for 7 days. A) H&E staining and (B&C) immunostaining for myocilin, FN, GRP78 and Col-IV in cultured corneoscleral segments transduced with WT or mutant myocilin. Increased myocilin staining was observed in the TM of mutant myocilin-transduced quadrants compared to WT myocilin. In addition, increased FN and Col-IV staining indicate more ECM accumulation in the TM of mutant myocilin-transduced quadrants (n = 3 biological replicates, scale bar is 50μm). Western blot and densitometric analysis of TM tissue lysates (D-E) and conditioned medium (F) obtained from cultured quadrants transduced with WT and mutant myocilin lentiviral expression vectors. A significant increase in the ECM marker FN (n = 6 biological replicates) and the ER stress marker CHOP (n = 3 biological replicates) was observed in mutant myocilin-transduced TM tissue lysates. Similarly, conditioned medium (F) from mutant myocilin-treated corneoscleral segments showed increases in ECM proteins FN and Col IV. Moreover, WT myocilin was detected in conditioned media of WT myocilin-transduced quadrants while no myocilin was detected in quadrants expressing mutant myocilin indicating that expression of mutant myocilin inhibits its secretion and accumulates in the TM cells. Unpaired t-test, *P<0.05, **P<0.01, ***P<0.001. Arrows indicate the TM region.