Ranking Single\nFluorescent Protein-Based Calcium Biosensor\nPerformance by Molecular Dynamics Simulations

Genetically encoded fluorescent biosensors (GEFBs) have\nbecome\nindispensable tools for visualizing biological processes <i>in\nvivo.</i> A typical GEFB is composed of a sensory domain (SD)\nthat undergoes a conformational change upon ligand binding or enzymatic\nreaction; the SD is genetically fused with a fluorescent protein (FP).\nThe changes in the SD allosterically modulate the chromophore environment\nwhose spectral properties are changed. Single fluorescent (FP)-based\nbiosensors, a subclass of GEFBs, offer a simple experimental setup;\nthey are easy to produce in living cells, structurally stable, and\nsimple to use due to their single-wavelength operation. However, they\npose a significant challenge for structure optimization, especially\nconcerning the length and residue content of linkers between the FP\nand SD, which affect how well the chromophore responds to conformational\nchange in the SD. In this work, we use all-atom molecular dynamics\nsimulations to analyze the dynamic properties of a series of calmodulin-based\ncalcium biosensors, all with different FP–SD interaction interfaces\nand varying degrees of calcium binding-dependent fluorescence change.\nOur results indicate that biosensor performance can be predicted based\non distribution of water molecules around the chromophore and shifts\nin hydrogen bond occupancies between the ligand-bound and ligand-free\nsensor structures.