Phosphorylation Determines Whether Neuroligin-3 is at Excitatory or Inhibitory Synapses in Different Regions of the Brain

Abstract

Neuroligin-3 is a postsynaptic adhesion molecule involved in development, function, and pathologies of synapses in the brain. It is a genetic cause of autism and a potent component of the tumor microenvironment in gliomas. There are four Neuroligins that operate at distinct synapse types, selectively interacting with presynaptic adhesion and postsynaptic scaffold proteins. We investigated the subcellular localization and scaffold specificities of synaptic Neuroligin-3 and demonstrate an unexpected pattern of localization to excitatory synapses in cortical areas, and inhibitory synapses in subcortical areas. Using phosphoproteomics, we identify Neuroligin-3-specific serine phosphorylation in cortex and hippocampus that obstructs a key binding site for inhibitory synapse scaffolds. Using in utero CRISPR/Cas9 knockout and replacement with phosphomimetic mutants, we demonstrate that phosphorylation at this site determines excitatory versus inhibitory synapse localization of Neuroligin-3 in vivo. Our data reveal a mechanism that differentially regulates the balance of Neuroligin-3 between excitatory and inhibitory synapses, adding to our emerging understanding of their role in the development of brain connectivity and associated pathologies.

Insulin-like growth factor 2 (IGF-2) rescues social deficits in NLG3–/y mouse model of ASDs

“…Mutations/deletions of X-linked neuroligin 3 ( Nlg3 ) gene, have been found in non-syndromic cases of autism ( Jamain et al, 2003 ; Levy et al, 2011 ; Quartier et al, 2019 ). The localization of NLG3 at both excitatory and inhibitory synapses is region and circuit specific and depends on its phosporilation state ( Varoqueaux et al, 2006 ; Altas et al, 2022 bioRxiv). Mice carrying the human R451C mutation of the Nlg3 gene exhibit impaired social approach and enhanced spatial learning associated with alterations of glutamatergic and GABAergic signaling in selective neuronal circuits ( Tabuchi et al, 2007 ; Etherton et al, 2011 ; Pizzarelli and Cherubini, 2011 , 2013 ; Cellot and Cherubini, 2014 ).…”