Bone Morphogenetic Protein (BMP)4 But Not BMP2 Disrupts the Barrier Integrity of Retinal Pigment Epithelia and Induces Their Migration: A Potential Role in Neovascular Age-Related Macular Degeneration

Abstract

Disruption of retinal pigment epithelial (RPE) barrier integrity and RPE migration are hallmark features in neovascular age-related macular degeneration (nAMD), but the underlying causes and pathophysiology are not completely well-defined. Herein, we aimed to evaluate the effect of bone morphogenetic proteins (BMPs) on the barrier function and migration of RPE. In particular, we investigated the role of BMP2 and BMP4 in these processes as our analysis of RNA-sequencing (seq) data from human donor eyes demonstrated that they are highly differentially expressed BMP members in macular RPE/choroid versus macular retina. We used electrical cell-substrate impedance sensing (ECIS) system to monitor precisely in real time the barrier integrity and migration of ARPE-19 after treatment with various concentrations of BMP2 or BMP4. Immunofluorescence was also used to assess the changes in the expression and the organization of the key tight junction protein, zona occludens (ZO)-1, in ARPE-19 cells under BMP2 or BMP4 treatment. This was followed by measuring the activity of matrix metalloproteinases (MMPs). Finally, RNA-seq and ELISA were used to determine the local and circulating levels of BMP2 and BMP4 in retinas and serum samples from nAMD donors. Our ECIS results showed that BMP4 but not BMP2 decreased the transcellular electrical resistance (TER) of ARPE-19 and increased their migration in comparison with control (vehicle-treated cells). Furthermore, immunofluorescence showed a disorganization of ZO-1 in BMP4-treated ARPE-19 not in BMP2-treated cells or vehicle-treated controls. This effect of BMP4 was associated with significant increases in the activity of MMPs, specifically MMP2. Lastly, these results were corroborated by additional findings that circulating but not local BMP4 levels were significantly higher in nAMD donor samples compared to controls. Collectively, our results demonstrated unreported effects of BMP4 on inducing RPE dysfunction and suggest that BMP4 but not BMP2 may represent a potential therapeutic target in nAMD.

Expression of Pro-Angiogenic Markers Is Enhanced by Blue Light in Human RPE Cells

“…A2E-treated cells showed dysregulated expression of genes clustered within nectin, focal adhesion kinase, E-cadherin and N-cadherin adhesion pathways. Our results deal with those recently reported by Ibrahim et al [ 49 ]. The authors observed the disruption of the RPE barrier due to increased MMP2 metalloproteinase activity, following high BMP4 concentration.…”

Bone Morphogenetic Proteins and Diabetic Retinopathy

“…Therefore, our group suggested that circulating BMP4 but not BMP2 or local retinal BMP4 is implicated in the pathogenesis of neovascular AMD. This conclusion was confirmed by demonstrating that BMP4 but not BMP2 interrupted the RPE barrier function, induced RPE migration and increased MMPs’ activity in RPE [ 80 ].…”

“…Collectively, we are suggesting the presence of a differential impact of BMPs affecting retinal-blood barriers, with BMP2 dysregulation affecting mostly the inner barrier [ 54 , 72 ], and circulating BMP4 dysregulation affecting mostly the outer barrier [ 80 ] ( Figure 2 ). Retinal levels of BMP4 were detected by RNase protection assays in both non-diabetic RPE from human donors and ARPE-19 cells.…”

Real-Time Monitoring the Effect of Cytopathic Hypoxia on Retinal Pigment Epithelial Barrier Functionality Using Electric Cell-Substrate Impedance Sensing (ECIS) Biosensor Technology

“…Effects of CoCl 2 on barrier function of ARPE-19 were evaluated by monitoring changes in overall impedance (Z; ohms (Ω)). Normalized Z was recorded by Electric Cell-substrate Impedance Sensing (ECIS ® Zθ (theta)) biosensor technology (Applied Biophysics Inc) as previously described [ 22 ]. Briefly, a 96-wells arrays (96W20idf PET, Applied Biophysics Inc) were coated with 50 µL of 100 µM cysteine for 30 min followed by coating with 50 µL of 0.02% gelatin (Sigma; G1393) for 30 min.…”

“…ARPE-19 cells were stained with zonula occludens (ZO)-1 antibody according to our previous procedure [ 22 ]. Briefly, ARPE-19 cells were fixed in paraformaldehyde (4%, 10 min) followed by one-hour blockage in 0.2% gelatin in 0.3% tritonX-PBS.…”