Abstract
The thermodynamics of arginine-phosphate binding is key to cellular signaling, protein-nucleic acid interactions, and membrane protein dynamics. In biomolecules, monoester phosphates are typically employed as strong electrostatic anchors strategically placed in switch domains to mediate specific interactions. In the diester configuration, phosphate groups act as ubiquitous connectors in all nucleic acids and polar lipids, while also engaging in less specific but multiple electrostatic interactions. Here, we employ isothermal titration calorimetry and a set of small-molecule models and peptides to benchmark the ability of the CHARMM force field to accurately reproduce these interactions. We observe good agreement between isothermal titration calorimetry and computational results for methylguanidinium (MGUA) with glycerol and glucose phosphates (MGUA-Gly3P, MGUA-Glu6P), and for arginine-glycine-arginine peptide with inositol triphosphate (RGR-IP3) systems, with experimental binding energies of -3.30 ± 0.30, -3.89 ± 0.30, and -8.96 ± 0.17 kcal/mol, compared with computational values of -4.08 ± 0.00, -4.20 ± 0.00, and -9.17 ± 0.20 kcal/mol, respectively. However, the experimental binding energy of -2.24 ± 0.71 kcal/mol between MGUA and dimethylphosphate in a diester configuration was significantly underestimated in CHARMM computations (-0.51 ± 0.01 kcal/mol). The force field was, therefore, refined by reducing the Lennard-Jones R(min) parameter from 3.55 to 3.405 Å for a specific interaction involving nitrogen and oxygen atoms in MGUA-dimethylphosphate. Our study brings another experimental means for fine-tuning force field parameters for the phosphates in two distinct configurations and enhances the accuracy of modeling nucleic acids, lipids, and membrane proteins.