Insulin Analog Specific for IR-B: MD with Effective Electronic Polarization
Description
Molecular dynamics with effective electronic polarisation of insulin analogue 6 specific for insulin receptor isoform B (IR-B) isolated (folder 1-Isol-Ins-An) and in complex with IR-A and IR-B (folder 2-Ins-An-IR-Complex), compared to a closely related analogue 5 and human insulin. The isolated systems start from PDB codes 8ONP, 8ONR and 3W7Y for An6, An5, and HI, respectively. 3-An6-conf-An5 is analog 6 in the conformation of analog 5 and 5-An5-conf-An6 is analog 5 in the crystallographic conformation of analog 6. structures. The complexes start from PDB 6HN5 (IR-A/human insulin) and cryo-EM structures from this work. Rep1,2,3 stand for replicas.
Files
Steps to reproduce
For full details, see manuscript. Briefly, the starting structures were prepared in CHARMM-GUI. Charge-scaled prosECCo75 protein and ion force fields along with the TIP3P water model were employed. The systems were neutralized in a 150 mM NaCl solution using CHARMM-GUI solution builder tool. The MD simulations were performed using GROMACS versions 2022.3 and 2022.4. For the isolated systems, the production runs in two replicas of the cumulative length of 140 μs were collected. Due to the high flexibility of the analogs, only four stable parts of trajectories (assessed by backbone RMSD) of 2.5 μs length were used for further analyses. For the complexes, Only the IR ectodomain regions relevant to Site 1 engagement were included, namely the L1, CR, L2, and FnIII-1 domains of the first IR protomer, together with the L2′, FnIII-1′, and αCT′ elements of the second protomer. The newly formed N- and C- termini (including the C-terminal Thr719 of Analog 6-IRB asym complex) were capped by acetyl and N-methylamide, respectively, using CHARMM-GUI, to avoid introducing a non-natural charge. In Analog 6-IRA asym and Analog 6-IRA sym systems, the loop between residues 519 and 527 in FnIII-1, which is not resolved by cryo-EM, was added by ModLoop functionality of the MODELLER programme. It was then adjusted manually by VMD programme, version 1.9.3 to form the S-S bond between Cys524 in FnIII-1 and Cys524 in FnIII-1', then minimized and equilibrated by GROMACS. For a more realistic description, the GlcNAc residue glycosylation was retained on Asn16. The production runs in three replicas of the cumulative length of nearly 20 μs were collected. Stable 1 μs-long parts of trajectories (assessed by insulin and IR backbone RMSD) of each replica were used for further analyses. All trajectory processing and analyses were carried out using the GROMACS analysis tools and the CPPTRAJ module of the AMBER16 software suite. H-bonding was identified by two geometric criteria: (i) the donor (D) and acceptor (A) atom distance was not greater than 3.5 Å, and (ii) the D-H…A angle was in range of 180 ± 60°. H-bond occurrences were calculated as percentage of the frames with at least 1 H-bond between the two partners. The analysis of the intra-IR H-bonds was limited only to the IR residues within 14 Å of B26. Representative snapshots from MD were selected to experimentally verify and visualize the strongest native H-bond interactions.