Dielectrophoretic characterization of peroxidized retinal pigment epithelial cells as a model of age-related macular degeneration

Age-related macular degeneration (AMD) is a prevalent ocular pathology affecting mostly the elderly population. AMD is characterized by a progressive retinal pigment epithelial (RPE) cell degeneration, mainly caused by an impaired antioxidative defense. One of the AMD therapeutic procedures involves injecting healthy RPE cells into the subretinal space, necessitating pure, healthy RPE cell suspensions. This study aims to electrically characterize RPE cells to demonstrate a possibility using simulations to separate healthy RPE cells from a mixture of healthy/oxidized cells by dielectrophoresis. BPEI-1 rat RPE cells were exposed to hydrogen peroxide to create an in-vitro AMD cellular model. Cell viability was evaluated using various methods, including microscopic imaging, impedance-based real-time cell analysis, and the MTS assay. Healthy and oxidized cells were characterized by recording their dielectrophoretic spectra, and electric cell parameters (crossover frequency, membrane conductivity and permittivity, and cytoplasm conductivity) were computed. A COMSOL simulation was performed on a theoretical microfluidic-based dielectrophoretic separation chip using these parameters. Increasing the hydrogen peroxide concentration shifted the first crossover frequency toward lower values, and the cell membrane permittivity progressively increased. These changes were attributed to progressive membrane peroxidation, as they were diminished when measured on cells treated with the antioxidant N-acetylcysteine. The changes in the crossover frequency were sufficient for the efficient separation of healthy cells, as demonstrated by simulations. The study demonstrates that dielectrophoresis can be used to separate healthy RPE cells from oxidized ones based on their electrical properties. This method could be a viable approach for obtaining pure, healthy RPE cell suspensions for AMD therapeutic procedures.