Assigning crystallographic electron densities with free energy calculations—The case of the fluoride channel Fluc

Abstract

Approximately 90% of the structures in the Protein Data Bank (PDB) were obtained by X-ray crystallography or electron microscopy. Whereas the overall quality of structure is considered high, thanks to a wide range of tools for structure validation, uncertainties may arise from density maps of small molecules, such as organic ligands, ions or water, which are non-covalently bound to the biomolecules. Even with some experience and chemical intuition, the assignment of such disconnected electron densities is often far from obvious. In this study, we suggest the use of molecular dynamics (MD) simulations and free energy calculations, which are well-established computational methods, to aid in the assignment of ambiguous disconnected electron densities. Specifically, estimates of (i) relative binding affinities, for instance between an ion and water, (ii) absolute binding free energies, i.e., free energies for transferring a solute from bulk solvent to a binding site, and (iii) stability assessments during equilibrium simulations may reveal the most plausible assignments. We illustrate this strategy using the crystal structure of the fluoride specific channel (Fluc), which contains five disconnected electron densities previously interpreted as four fluoride and one sodium ion. The simulations support the assignment of the sodium ion. In contrast, calculations of relative and absolute binding free energies as well as stability assessments during free MD simulations suggest that four of the densities represent water molecules instead of fluoride. The assignment of water is compatible with the loss of these densities in the non-conductive F82I/F85I mutant of Fluc. We critically discuss the role of the ion force fields for the calculations presented here. Overall, these findings indicate that MD simulations and free energy calculations are helpful tools for modeling water and ions into crystallographic density maps.

Ion Permeation, Selectivity, and Electronic Polarization in Fluoride Channels

“…S2 ) dominantly contributed by E self ( Fig. S3-S4 ), consistent with previous studies (98,99). This difference is due to a limited description of the ion-induced polarization of π electron in additive FFs (61), which E self accounts for in Drude FFs (31,79).…”

“…The same observation was reported by Ariz-Extreme et al . (98). In addition, it was curious that Na + is tetrahedrally coordinated in Fluc crystal structures (5,14,16), as Na + has a typical coordination number (N coord ) of 5 – 6.…”

Alchemical Free Energy Calculations on Membrane-Associated Proteins

“…The same approach was used to refine docking poses of the general anesthetic sevofluorane on the voltage-gated potassium channel Kv1.2 and obtain a 3D map highlighting the complexity of molecular recognition at membrane receptors with multiple binding sites with single or multiple occupancy states being in competition. , Last, both absolute and relative FEP calculations in ion channels were used to assign ambiguous disconnected electron densities in protein crystal structures. In the case of the fluoride channel Fluc, for instance, the higher affinities of water molecules relative to fluorine ions were demonstrated at several sites, regardless of previous speculations …”

Computational Studies of Small Molecule Permeation across Membrane Channels

“…Accordingly, it was identified as a structural Na + ion that unavoidably binds during the homodimer assembly process. The remaining set of four disconnected electron densities was provisionally identified as four F − ions which simultaneously occupy the channel (with occupancies of ∼ 80%) [85,91]. They are coordinated by residue N43 from TM2, residues S108 and S112 from TM4, and F82 and F85 from the surroundings of the TM3-break (Figure 3.2).…”

“…Wwe shifted our attention to Fluc-Bpe channel (see Chapter 3), whose novel crystal structure was analyzed by means of computational methods, such as MD, US, Thermodynamic Integration (TI), and Quantum Mechanics (QM) simulations [91]. In addition, we aimed to shed light onto the unanswered questions about Fluc: i) what is the permeation mechanism, and ii) what is the mechanism for the selec-…”