Evaluation of neuroprotective and immunomodulatory properties of mesenchymal stem cells in an ex vivo retinal explant model.

Background: Glaucoma is a blinding degenerative neuropathy in which the death of retinal ganglion cells (RGCs) causes progressive loss of visual field and eventually vision. Neuroinflammation appears to be a key event in the progression and spread of this disease. Thus, microglial immunomodulation represents a promising therapeutic approach in which mesenchymal stem cells (MSCs) might play a crucial role. Their neuroprotective and regenerative potentials have already raised hope in animal models. Yet no definitive treatment has been developed, and some safety concerns have been reported in human trials. In the present study, we investigated the neuroprotective and immunomodulatory properties as well as the safety of MSCs in an ex vivo neuroretina explant model.

Methods: Labeled rat bone marrow MSCs were placed in coculture with rat retinal explants after optic nerve axotomy. We analyzed the neuroprotective effect of MSCs on RGC survival by immunofluorescence using RBPMS, Brn3a, and NeuN markers. Gliosis and retinal microglial activation were measured by using GFAP, CD68, and ITGAM mRNA quantification and GFAP, CD68, and Iba1 immunofluorescence stainings. We also analyzed the mRNA expression of both 'M1' or classically activated state inflammatory cytokines (TNFα, IL1β, and IL6), and 'M2' or alternatively activated state microglial markers (Arginase 1, IL10, CD163, and TNFAIP6).

Results: The number of RGCs was significantly higher in retinal explants cultured with MSCs compared to the control group at Day 7 following the optic nerve axotomy. Retinal explants cultured with MSCs showed a decrease in mRNA markers of gliosis and microglial activations, and immunostainings revealed that GFAP, Iba1, and CD68 were limited to the inner layers of the retina compared to controls in which microglial activation was observed throughout the retina. In addition, MSCs inhibited the M1 phenotype of the microglia. However, edema of the explants was observed in presence of MSCs, with an increase in fibronectin labeling at the surface of the explant corresponding to an epiretinal membrane-like phenotype.

Conclusion: Using an ex vivo neuroretina model, we demonstrated a neuroprotective and immunomodulatory effect of MSCs on RGCs. Unfortunately, the presence of MSCs also led to explant edema and epiretinal membrane formation, as described in human trials. Using the MSC secretome might offer the beneficial effects of MSCs without their potential adverse effects, through paracrine signaling.