Search

Retinal ganglion cells (RGCs) lack regenerative capacity in mammals, and their degeneration in glaucoma leads to irreversible blindness. The transplantation of stem cell-derived RGCs lacks clinically relevant effect due to insufficient survival and integration of donor cells. We hypothesize that the retinal microenvironment plays a critical role in this process, and we can engineer a more acceptable setting for transplantation. Since the adult mammalian retina does not have regenerative capacity, we turned to the developing human retina to reconstruct cell-cell interactions at a single-cell level. We established a human fetal retina atlas by integrating currently available single-cell RNA sequencing datasets of human fetal retinas into a unified resource. We align RGC transcriptomes in pseudotime to map RGC developmental fate trajectories against the broader timeline of retinal development. Through this analysis, we identified brain-derived neurotrophic factor (BDNF) and glial-derived neurotrophic factor (GDNF) as key factors in RGC survival, highly expressed during fetal development but significantly reduced in adulthood despite the persistence of their receptors. To demonstrate the practical application of these findings, we show that using a slow-release formulation of BDNF and GDNF enhances RGC differentiation, survival, and function in vitro and improves RGC transplantation outcomes in a mouse model. BNDF/GDNF co-treatment not only increased survival and coverage of donor RGCs within the retina but also showed neuroprotective effects on host RGCs, preserving retinal function in a model of optic neuropathy. Altogether, our findings suggest that manipulating the retinal microenvironment with slow-release neurotrophic factors holds promise in regenerative medicine for treating glaucoma and other optic neuropathies. This approach not only improves donor cell survival and integration but also provides a neuroprotective benefit to host cells, indicating a significant advancement for glaucoma therapies.

The limited regenerative potential of the optic nerve in adult mammals presents a major challenge for restoring vision after optic nerve trauma or disease. The mechanisms of this regenerative failure are not fully understood 1,2 . Here, through small-molecule and genetic screening for epigenetic modulators 3 , we identify DNA methyltransferase 3a (DNMT3a) as a potent inhibitor of axon regeneration in mouse and human retinal explants. Selective suppression of DNMT3a in retinal ganglion cells (RGCs) by gene targeting or delivery of shRNA leads to robust, full-length regeneration of RGC axons through the optic nerve and restoration of vision in adult mice after nerve crush injury. Genome-wide bisulfite and transcriptome profiling in combination with single nucleus RNA-sequencing of RGCs revealed selective DNA demethylation and reactivation of genetic programs supporting neuronal survival and axonal growth/regeneration by DNMT3a deficiency. This was accompanied by the suppression of gene networks associated with apoptosis and inflammation. Our results identify DNMT3a as the central orchestrator of an RGC-intrinsic mechanism that limits optic nerve regeneration. Suppressing DNMT3a expression in RGCs unlocks the epigenetic switch for optic nerve regeneration and presents a promising therapeutic avenue for effectively reversing vision loss resulted from optic nerve trauma or diseases., Competing Interests: Competing interests The other authors declare no competing interests.