First of all, we validated Pcdh10 as novel SUMO substrate in mammalian brain. Next, we aimed to understand the molecular mechanisms controlling Pcdh10 function in neurons. We demonstrated that Pcdh10 is a SUMO target in vitro and in vivo and that its sumoylation is activity-dependent and developmentally regulated. Using site-directed mutagenesis, we point-mutated the predicted K831 SUMO site into an arginine (K831R), generating a non-sumoylatable form of Pcdh10. Interestingly, preventing Pcdh10 sumoylation at the K831 residue does not perturb the stability of Pcdh10 and its subcellular localization. Prior to unraveling the functional consequences of sumoylation in regulating Pcdh10 function and dysfunction, we first investigated the physiological role of Pcdh10 in the assembly and function of cortical synapses. To this end, we silenced Pcdh10 expression in cortical neurons at 7 days-in-vitro (DIV7) using a specific shRNA (shRNA-Pcdh10) and evaluated the effects on synapse development and physiology at DIV15. Combining advanced microscopy and electrophysiology approaches we demonstrated that the loss of Pcdh10 expression leads to an increase in both dendritic spines and excitatory synapses density, which correlates with higher frequency and amplitude of miniature excitatory postsynaptic currents (mEPSCs). Taking advantage of super-resolution Stimulated emission depletion (STED) microscopy, we performed detailed measurements of dendritic spine architecture, including spine head and neck. Neurons lacking Pcdh10 displayed shorter spines characterized by a larger head, indicating that Pcdh10 is essential in maintaining the proper dendritic spine morphology. We confirmed these results in vivo using In Utero Electroporation (IUE), further supporting the role of Pcdh10 in controlling excitatory synapse formation and function.
Conversely, neither the number and architecture of inhibitory synapses nor the frequency and amplitude of miniature inhibitory postsynaptic currents (mIPSCs) were impaired in Pcdh10 silenced neurons, indicating that Pcdh10 is essential in the physiology of excitatory but not inhibitory synapses.
Interestingly, rescue experiments demonstrated that reintroducing the non sumoylatable form of Pcdh10 failed in rescuing the dendritic spine density in Pcdh10-silenced neurons. Conversely, the WT and K831R forms of Pcdh10 properly restored the mEPSCs frequency defects induced by Pcdh10 knockdown.
Altogether, our results demonstrate that sumoylation is an essential mechanism controlling the Pcdh10 physiological role in neurons and that perturbing such process compromises excitatory synapse formation.
