A biophysical hypothesis for compartment-specific conformational states of the IQSEC2 intrinsically disordered protein.

The postsynaptic density (PSD) of neuronal synapses is a crowded, viscous, membraneless compartment consisting of densely packed protein mixtures formed and maintained through liquid-liquid phase separation. A key regulator of synaptic function within the PSD is IQSEC2, an intrinsically disordered protein (IDP), which functions as a guanine nucleotide exchange factor, promoting the activation of the small GTPase ARF6 by catalyzing the exchange of ARF6-bound GDP for GTP. Experiments have shown that IQSEC2 is inactive in a folded state in the dendritic cytosol but can transiently adopt a catalytically active extended conformation in the PSD upon activation by neurotransmitter-mediated calcium influx. In this study, we performed accelerated molecular dynamics (aMD) simulations of the IQSEC2 folding process in aqueous solvent and applied the Gibbs ergodic hypothesis formulated in statistical ensembles for post-processing and analysis of the results. We compared the effects of two sets of parameters on folding: a force field and a water model. The optimal point charge water model optimized for proteins with disordered structure in the extended state in bulk aqueous solvent predicted an incorrect, distorted, 3D folded structure of IQSEC2. We propose that, due to fundamental biophysical differences between the dendritic cytosol and the PSD, there are two distinct classes of IDPs functioning in these different environments. These differences should be considered when optimizing force fields and parameters of water models for studying protein folding.