Back

Gelatinase activity is required for Muller glia morphogenesis and neuronal maintenance in the retina

Jaroszynska, N.; Kothurkar, A. A.; Jankovic, D.; Cameron-Pack, M.; MacDonald, R. B.

2025-12-19 neuroscience
10.64898/2025.12.17.694830 bioRxiv
Show abstract

Glial cells are indispensable support elements of the nervous system, yet the mechanisms by which they integrate into densely packed, extracellular matrix (ECM)-rich cellular environments remain poorly understood. In the retina, Muller glia (MG) serve as the principal support cells, performing roles analogous to astrocytes in the brain. As late-born retinal cells, MG undergo morphogenesis within a complex extracellular milieu composed of neighbouring neurons, synapses, and ECM components. Using the zebrafish retina, we show that developing MG dynamically interact with the ECM across all retinal layers while expressing the ECM-remodelling enzymes matrix metalloproteinases (Mmp) 2 and 9. Although Mmps are known to contribute to tissue remodelling and disease processes such as glioma invasion, their developmental roles in glia remain less defined. Here, we demonstrate that Mmp2 and Mmp9 are required for MG morphogenesis and glia-neuron integration. Pharmacological inhibition or genetic loss of mmp2 and mmp9 impaired MG outgrowth and reduced process complexity within synaptic layers, leading to decreased MG-synapse associations. Double mutants (mmp2-/-;mmp9-/-) displayed normal neuronal development and visual function at larval stages but exhibited gliosis, premature neurodegeneration, and visual deficits in adulthood. Together, these findings reveal that Mmp2 and Mmp9 are critical for establishing MG-synapse interactions during development and for maintaining neuronal integrity and visual function in the mature retina.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.