A new gene therapy approach has enabled retinal regeneration in mice by preventing the transfer of a single inhibitory protein into Müller glia, offering a potential new direction for treating retinal degenerative diseases.
The findings, published in April in Nature Communications, come from a research team at the Korea Advanced Institute of Science and Technology (KAIST) led by Eun Jung Lee, Museong Kim, Sooyeon Park, Ji Hyeon Shim, and colleagues. The study identifies the transcription factor prospero-related homeobox 1 (Prox1) as a molecular barrier to regeneration and shows that blocking its intercellular transfer allows Müller glia to resume progenitor-like behavior after retinal injury.
The researchers demonstrated that after chemically induced retinal injury, Prox1 protein accumulates in Müller glia, despite the absence of Prox1 mRNA expression in these cells. The protein was traced to neighboring bipolar cells, suggesting it was transferred between cells rather than produced endogenously. In contrast, zebrafish—which are capable of spontaneous retinal regeneration—do not exhibit Prox1 accumulation in Müller glia after injury.
To test whether blocking this protein transfer could restore regenerative capacity in mammals, the researchers developed 2 experimental models. In one, they genetically deleted Prox1 in donor bipolar cells. In the other, they used an adeno-associated virus (AAV) to deliver an antibody fragment that binds and neutralizes Prox1 in the extracellular space. Both approaches successfully reduced Prox1 uptake by Müller glia.
In treated mice, Müller glia showed markers of cell cycle reentry and progenitor status after injury. Moreover, some animals demonstrated regeneration of retinal neurons and a delay in vision loss in a model of retinitis pigmentosa. Importantly, blocking Prox1 did not alter its expression levels in the source neurons, suggesting that the intervention specifically targets intercellular protein transfer without broader transcriptional effects.
The authors note that while mammalian Müller glia do not typically regenerate retinal neurons, this is not due to a lack of regenerative machinery but rather the presence of external suppressive signals like Prox1. By removing these inhibitory cues, they were able to unlock regenerative potential in otherwise quiescent glia.
“These findings establish Prox1 as a barrier to MG-mediated regeneration and highlight anti-Prox1 therapy as a promising strategy for restoring retinal regeneration in mammals,” the authors wrote. The team disclosed that several of the authors are co-founders or employees of Celliaz Inc, a company that is developing anti-Prox1 therapeutics based on these findings.
The findings also highlight a previously unrecognized mechanism of protein-based intercellular regulation that may influence regeneration in other tissues. The transfer of transcription factors between neurons and glia is a novel concept that could reshape understanding of cell fate control in the central nervous system.
Although the regenerative effects were modest and the durability of the AAV-based therapy may require further optimization, the authors suggest that this work represents a proof of concept for regenerative intervention in the retina. They propose follow-up studies to refine delivery vectors and explore combination strategies to enhance neuron replacement. RP