The main feature of myopia is axial length hyperextension[20]. The RPE cells forms a blood-retinal barrier through cellular connections and selectively transports biomolecules between the neural retina and the choriocapillaris [21]. Alterations in RPE barrier function can affect the transmission of chemical signals involved in the development of myopia. During the development of myopia, the vitreous chamber was elongated, but the number of RPE cells did not increase. From this phenomenon, it is easy to imagine that the morphology of the RPE cells change during myopia, and the tight junction proteins were susceptible to be influenced and changed. In our study, the expression of the tight junction protein ZO-1 decreased in RPE cells irradiated with blue light. After the intervention of atropine, the mRNA and protein expression of ZO-1 were reelevated. The expression of ZO-1 in myopic guinea pigs with form deprivation was reduced. ZO-1 mRNA and protein expression in the RPE-choroidal complex was increased after intervention with atropine in FDM guinea pigs. It is suggested that the tight junctions between RPE cells are disrupted in myopia, and atropine could attenuate this damage. Maintaining the structural integrity of tissues requires stable interactions between cells, while dynamic changes in cell adhesion are involved in the morphogenesis of developing tissues. Neighboring cells are connected mainly by Ca2+-dependent cell adhesion molecules (Cadherins). Adhesion mechanisms are highly regulated during tissue morphogenesis and are closely linked to cell motility and cell migration processes[22]. In our study, the expression of the adhere junction protein E-cadherin decreased in RPE cells irradiated with blue light. After the intervention of atropine, the mRNA and protein expression of E-cadherin were reelevated. In myopic models, E-cadherin as well as P-cadherin protein expression in the RPE-choroidal complex of guinea pigs in the FDM group was not statistically different from that of the control group. E-cadherin, P-cadherin protein expression in the RPE-choroidal complex was increased after intervention with atropine in FDM guinea pigs. mRNA expression of CDH-1 was decreased in myopic guinea pigs, and increased after atropine was used. Previous studies have found that all-trans retinoic acid (atRA) inhibits the adhesion function of ARPE-19 cells by decreasing MAPK phosphorylation and affecting the MAPK and FAK pathways[23]. atRA inhibits RPE contractility by inhibiting integrin β3 expression or blocking integrin β3-mediated signaling with tissue remodeling ability[15]. In vitro studies have found that atRA increases the trans-epithelial resistance of myopic RPE monolayers and decreases cell permeability. In atRA-treated myopic RPE cells, the expression levels of E-cadherin, Occludin, and Claudin-1 were increased, suggesting that atRA may promote the epithelial barrier function of RPE monolayers in myopic eyes by regulating the expression of intercellular adhesion proteins and junction-related proteins[24]. However, in the previous LIM-induced myopia guinea pig model for 14 days, the tight junction proteins ZO-1 and Occludin in the RPE-choroidal complex showed an up-regulation trend with the increase of myopia progression, on the contrary, intravitreal injection of the RA receptor antagonist LE540 in the guinea pigs reduced the expression level of RA tight junction-related proteins, so that the growth of myopic axes of guinea pigs was suppressed[25]. It is suggested that the RPE epithelial barrier function may show a dynamic change during the progression of myopia. In the early stage, there might be a negative feedback mechanism in the RPE cells during the early stages of myopia development, compensating for the RPE cell pulling caused by vitreous lengthening by enhancing the epithelial barrier function. The RPE epithelial barrier function is enhanced by secretion of tight junction proteins and adhesion proteins. In the later stage, with the prolongation of the ocular axis, the decrease of RPE cell density and the aggravation of damage, the RPE is unable to secrete enough proteins and cytokines to maintain the barrier function. As myopia progresses further, the eye axis lengthens and the integrity of the retinal pigment epithelium is lost, resulting in pathological myopic fundus changes such as retinal detachment and macular degeneration. In this study, it was proved that, by the late stage of myopia development, the intercellular connections of RPE cells were disrupted and were unable to compensate for intercellular tension over time. In addition to the difference cut-off points between studies, another possible reason for the variability of the results may be the difference in myopia modelling methodology, with different mechanisms of effect of LIM and FDM on myopic formation and progression.
Axial elongation of the eye must involve morphological changes in the scleral tissue and matrix remodeling. Collagen accounts for approximately 90% of scleral proteins, the predominant one being type I collagen. The majority of studies have shown that a decrease in type I collagen is responsible for scleral thinning and its abnormal and progressive elongation, partly due to decreased biosynthesis and partly due to increased degradation[26]. In this study, atropine increased COL1A1 protein and mRNA expression in RPE cells, and there was a consistent trend of increased COL1A1 mRNA expression in the RPE-choroid complex of guinea pigs in the FDM group. In this study, atropine was found to increase the expression of fibronectin protein in guinea pig sclera. This corresponds with previous reports that atropine can stimulate the production of fibronectin by human scleral fibroblasts[27]. However, after intervention with atropine in normal ARPE-19, both its fibronectin protein and mRNA expression decreased. Intervention with atropine in FDM guinea pigs also decreased fibronectin protein expression in the RPE-choroidal complex. Fibronectin, an important glycoprotein of the basement membrane, can regulate cell growth and survival signaling by interacting with specific integrin receptors on the cell membrane. Fibronectin is also found in Bruch's membrane, the basement membrane of the RPE layer, which is composed of collagen type IV, laminin, fibronectin, heparin sulphate, and chondroitin sulphate[28–32], all of which can be produced through RPE cell secretion. Clinical and experimental studies have found that Bruch's membrane functions as a biomechanical property to maintain the normal structure of the eyeball[33–36]. Clinical and animal experiments have shown that Bruch's membrane thickness does not change with increasing eye axis in myopic eyes[37, 38]. However, during orthogonalization or myopia, along with the lengthening of the ocular axis, the internal surface area of the eyeball increases, and it can be hypothesized that the volume of the Bruch's membrane increases with the lengthening of the ocular axis during myopia, and that it is actively produced and elongated during axial myopia[13]. Atropine's reduction of fibronectin expression in myopia may reduce the production of fibronectin by the RPE during myopia to form BM, which inhibits axial lengthening by reducing the increase in the volume of Bruch's membrane. Previous studies have also found that atropine reduces choroidal fibroblast ECM biosynthesis[19]. Reduced choroidal thickness and choroidal blood perfusion are associated with progressive myopia in guinea pig models and in humans[39]. It can be speculated that the reduced production of ECM proteins such as fibronectin triggered by atropine in the RPE-BM-choroidal complex may favor choroidal blood perfusion, thus sparing scleral tissue from hypoxia.
During the onset and progression of myopia, the strong elongation of the ocular axis and vitreous body leads to stretching of the RPE layer and disruption of the RPE barrier function. In our study, it was that atropine affected both RPE tight junction protein ZO-1 and adhesion protein E-cadherin in vivo and in vitro experiments. Further experiments are needed to refine the localization of such proteins in RPE tissues and cells and their specific functions. And other tight junction proteins and adhesion proteins (such as ICAM-1, Occludin, Claudins, etc.) still need to be further investigated and explored by subsequent animal and cell experiments. The in vivo sample consists of an RPE Choroid complex; thus, it still remains unclear if the in vitro results truly reflect what occurs in vivo. For accurate experimentation, it needs further studies to isolate RPE and perform experiments separately. In addition, the relevant mechanisms of the involved upstream cytokines and signaling pathways need to be further investigated in the future.
In conclusion, this study demonstrates that myopia could affect adhere and tight junction proteins in the RPE and atropine was used to intervene RPE cells to explore whether atropine has an effect on RPE cell barrier function as well. Our study may find out the possible theoretical basis for the pathogenesis of myopia in RPE-BM levels and provide new insights for myopia prevention and treatment.