Excessive accumulation of ECM proteins is considered a major feature of glaucomatous TM tissues. The TM cells and surrounding ECM work like a syncytium complex called matrisome that regulates aqueous humor outflow in the anterior chamber [18]. The pathogenic mechanism of increased resistance to aqueous humor outflow in PXF is poorly understood, but excessive ECM deposition, aggravated cell death, and alterations in cytoskeletal organization [1,2,8] apart from mechanical obstruction of the TM spaces by exfoliative material, have been thought to be the key mechanisms. Experimental models to study these mechanisms are lacking with current animal models mimicking certain ocular features of the human disease without evident glaucoma [3,19,20]. One of the key feature of PXF is protein aggregate formation. The exact composition of the pseudoexfoliative material is still not known histological and biochemical approaches have shown that the aggregates in PXF have a protein core surrounded by glycoconjugates that are resistant to enzymatic degradation [19]. The protein core has been shown to be composed of ECM proteins, MMPs and cross-linking enzymes like LOXL1 [3,4,18,19]. Therefore, we tried to create an in vitro model on HTM cells that will mimic PXF conditions in the eye. In this in vitro model, TGF-β1 (10 ng/mL) exposure caused upregulated expression of ECM proteins, EMT morphological changes, paralleled by deregulated enzymatic activity of MMP9 and protein aggregate formation at 48–72 h of exposure, all of which mimic features of PXF disease in the human eye. We visualized the protein aggregates using antibodies that bind to amyloid fibrils, key component of misfolded proteins. Though transmission electron microscope (TEM) and a 3D model of HTM cell culture would have been ideal, nanometer size of the aggregates and disruption during processing prevented visualization of these fibrils. Here we found that chronic TGF-β1 (10 ng/mL) exposure induces EMT in TM cells at 48 h, which was also confirmed by increased α-SMA and vimentin expression. This paralleled increased IL-6 and IL-8 levels at the same time period, both of which are TGF-induced deleterious effects for ECM homeostasis [12,21,22]. TGF-β1 induced EMT was inhibited by Smad3 inhibition, while other proteins were differentially regulated by other pathways. TGFβ1 is known to orchestrate EMT as a mechanism of regulating ECM homeostasis and cell survival [23,24,25,26]. EMT is a normal process involved in tissue repair [27,28] but deregulated EMT can lead to fibrosis. EMT-associated disease pathogenesis has already been studied in a range of diseases including lung fibrosis and cancer [29]. In the eye, EMT has been associated with subretinal fibrosis leading to age-related macular degeneration (AMD) [25]. The canonical and non-canonical pathways differentially regulate ECM proteins and this maybe one of the mechanisms of the TGF paradox driving increased fibrosis and functional damage in eyes with glaucoma in pseudoexfoliation. Study of these effects in our model particular could give more insights into actual regulation and cross talk of other pathways modulating TGF induced fibrosis in HTM cells. It may also be used to identify putative molecules to reverse EMT, cell death or aggregate formation, which may form important targets for preventing glaucoma in pseudoexfoliation.

To maintain homeostasis, cells rely on protective mechanisms to help them cope with ER stress, pathways referred to collectively as the unfolded protein response (UPR) [30]. The relation between UPR signaling and fibrogenesis are not fully understood. Studies suggest that ER stress facilitates fibrotic remodeling through activation of pro-apoptotic pathways, induction of epithelial—mesenchymal transition, and promotion of inflammatory responses [30,31,32]. In our previous study, we found that the expression of UPR response genes like XBP1 and CCT4 were significantly downregulated in PXG [15]. Therefore, we wanted to see if the UPR pathway is also hampered in the in vitro model. We found reduced expression of UPR markers like XBP1 and CCT4 at delayed TGF-β1 exposures. This hints at the insufficient UPR activation hampering the process of protein clearance and the misfolded proteins forming insoluble aggregates. Overall, our data indicate that sustained TGF-β1 treatment used in this in vitro HTM cell lime model could be used to mimic PXF human disease. This may enable studying temporal molecular events at different time points and allow us to identify drug targets for preventing cell death or aggregate formation.