ALX/FPR2 Activation by Stereoisomers of D1 Resolvins Elucidating with Molecular Dynamics Simulation

Abstract

Chronic inflammation contributes to several diseases, but its resolution is driven by specialized pro-resolving mediators (SPM) such as resolvin D1 (RvD1) and its epimer aspirin-triggered resolvin D1 (AT-RvD1), both biosynthesized from ω-3 fatty docosahexaenoic acid (DHA). RvD1 and AT-RvD1 have anti-inflammatory and pro-resolution potentials, and their effects could be mediated by formyl peptide receptor type 2 receptor ALX/FPR2, a G-protein-coupled receptor (GPCR). In this work, we performed 44 μs of molecular dynamics simulations with two complexes: FPR2@AT-RvD1 and FPR2@RvD1. Our results show the following: (i) in the AT-RvD1 simulations, the ALX/FPR2 receptor remained in the active state in 62% of the frames, while in the RVD1 simulations, the receptor remained in the active state in 74% of the frames; (ii) two residues, R201 and R205, of ALX/FPR2 appear, establishing interactions with both resolvins in all simulations (22 in total); (iii) RvD1 hydrogen bonds with R201 and R205 presented higher frequency than AT-RvD1; and (iv) residues R201 and R205 are the two receptor hotspots, demonstrated by the binding free calculations. Such results show that the ALX/FPR2 receptor remained in the active state for longer in the FPR2@RvD1 simulations than in the FPR2@AT-RvD1 simulations.

Figure 1:

AT-RvD1 (green) and RvD1 (cyan) superposition. (A) Highlighted on carbon C17, the chiral carbon. In AT-RvD1 the carbon chain is turned to the right starting from carbon 17 (17R), while in RvD1 the carbon chain is turned to the left (17S). (B) AT-RvD1 and RvD1 molecular docking on 6OMM structure.

Figure 2:

(A) NPxxY motif's location on ALX/FPR2 receptor. (B) NPxxY motif’s RMSD variation for AT-RvD1 simulations. (C) NPxxY motif’s RMSD variation for RVD1 simulations. For the calculation of RMSD we use Cα atoms.

Figure 3:

(A) Distance TMH3-TMH6 variation, average I116-V247 and R123-L243 distances, analyzed in all simulations. (B) The frames' frequency where the average between the two distances was >=10.5Å. (C) and (D) Distance TMH3-TMH6 variation for AT-RvD1 simulations (C), and RvD1 simulations (D). Red line is the 10.5Å threshold.

Figure 4:

Hydrogen bond frequency in FPR2@AT-RvD1 simulations (A) and FPR2@RvD1 simulations (see Tables S2 and S3). Residues that presented hydrogen bond frequency >=0.20 in at least one simulation. Hydrogen bonds between D1 resolvins and hotspots residues: (C) FPR2@AT-RvD1, S9’s pose, and (D) FPR2@RvD1, S11’s pose.

Figure 5:

Average total ΔG_Bind and contributions (Non-Electrostatic and Electrostatic) on eleven simulations at (A) FPR2@AT-RvD1 and (B) FPR2@RvD1 (see Tables S4 and S5 ). Residue energy decomposition for FPR2@AT-RvD1 (C), and FPR2@RvD1 (D) simulations. The contribution per residue is the average in all eleven simulations (see Tables S6 and S7).