Improved GROMOS 54A7 Charge Sets for Phosphorylated Tyr, Ser, and Thr to Deal with pH-Dependent Binding Phenomena

Abstract

Phosphorylation is a ubiquitous post-translational modification in proteins, and the phosphate group is present constitutively or transiently in most biological building blocks. These phosphorylated biomolecules are involved in many highaffinity binding/unbinding events that rely predominantly on electrostatic interactions. To build accurate models of these molecules, we need an improved description of the atomic partial charges for all relevant protonation states. In this work, we showed that the commonly used protocols to derive atomic partial charges using well-solvated molecules are inadequate to model the protonation equilibria in binding events. We introduced a protocol based on PB/MC calculations with a single representative conformation (of both protonation states) and used the resulting pK a estimations to help manually curate the atomic partial charges. The final charge set, which is fully compatible with the GROMOS 54A7 force field, proved to be very effective in modeling the protonation equilibrium in different phosphorylated peptides in the free (tetrapeptides, pentapeptides, and pY1021) and protein-complexed forms (pY1021/PLC-γ1 complex). This was particularly important in the case of the pY1021 bound to the SH2 domain of PLC-γ1, where only our curated charge set captured the correct protonation equilibrium at the neutral to slightly acidic pH range. The binding/unbinding phenomena in that pH range are biologically relevant, and to improve our models, we need to go beyond the commonly used protocols and obtain revised force field parameters for these molecules.

Development and Validation of AMBER-FB15-Compatible Force Field Parameters for Phosphorylated Amino Acids

“…The AmberTools 20 distribution includes newly derived side-chain-specific torsion parameters for phosphorylated residues that are compatible with the ff14SB and ff19SB force fields, but the parameterization and validation approaches have not yet been published. Parameters for phosphorylated residues have also been developed for the GROMOS and CHARMM force fields, with parameters either derived from nucleic acids or parameterized similar to the AMBER force fields. − …”

Extending the Stochastic Titration CpHMD to CHARMM36m

“…However, these pH effects are often ignored in molecular dynamics (MD) simulations due to the difficulty of sampling correct protonation states. Over the last 28 years, many constant-pH MD (CpHMD) methods have been developed to address these limitations. − The different strategies employed can be distinguished mainly by (i) the type of protonation (continuous vs discrete); (ii) the force field (AMBER, CHARMM, GROMOS, OPLS, MARTINI, etc.) and its level of detail (all-atom, united-atom, and coarse grain); and (iii) the approximations used to deal with charge fluctuations in the simulation box, often related with the use of counterions and the long-range electrostatics treatment (reaction-field vs Ewald summation methods).…”

“…and its level of detail (all-atom, united-atom, and coarse grain); and (iii) the approximations used to deal with charge fluctuations in the simulation box, often related with the use of counterions and the long-range electrostatics treatment (reaction-field vs Ewald summation methods). In discrete CpHMD methods, ,,,− ,,,,− ,− , continuum electrostatics calculations are used to estimate the energies for the Monte Carlo (MC) move while the MD simulations are run in either implicit or fully explicit solvent. The most recent continuous CpHMD methods are based on the λ-dynamics approach for free-energy calculations, where an individual λ variable is assigned to each titratable site of the protein. ,− ,− ,,,− , All protonation states coordinates vary continuously between 0 and 1, representing the protonated and deprotonated states, respectively.…”

Increasing the Realism of in Silico pHLIP Peptide Models with a Novel pH Gradient CpHMD Method

“…Several studies focused on the structural stability, , (de-)­insertion kinetics, possible metastable states, and electrostatic interactions modulating the pH dependency. ,,, Most of them used 2-oleoyl-1-palmitoyl- sn -glycero-3-phosphocholine (POPC) membrane bilayers to mimic liposomal conditions, such as ionic strength, pH, and peptide state. ,, Particularly, constant-pH molecular dynamics (CpHMD) simulations are helpful, as the residues are titrating at specific pH values, allowing the conformational and protonation sampling to be coupled. These methods may treat the protonation either as discrete ,,− or continuous. − Consequently, they promote the study of complex peptide–membrane configurations and the impact of the key titrating residues positioning along the membrane normal on helical (un)­folding and side-chain interactions. The electrostatic environment around the key aspartate residues changes with the peptide movement, favoring either insertion or exiting processes through protonation or deprotonation events, respectively.…”