CoChat

Assessment of the Topology and Oligomerisation States of Coiled Coils Using Metadynamics with Conformational Restraints

Open access
Notari, Evangelia, Wood, Christopher W., Michel, Julien · 2025 · Journal of Chemical Theory and Computation

The DOI matches its Crossref record; 133 of the 133 references it cites were found in a registry or an index; 32 of the 117 sources its claims cite were checked for support.

Abstract

protein design due to their established sequence-to-structure relationships and ability to switch conformations in response to external stimuli, such as changes in pH or temperature. However, the computational design of multistate coiled-coil protein assemblies is challenging, as it requires accurate estimates of the free energy differences between multiple alternative coiled-coil conformations. Here, we demonstrate how this challenge can be tackled using metadynamics simulations with orientational, positional and conformational restraints. We show that, even for subtle sequence variations, our protocol can predict the preferred topology of coiled-coil dimers and trimers, the preferred oligomerization states of coiled-coil dimers, trimers, and tetramers, as well as the switching behavior of a pH-dependent multistate system. Our approach provides a method for predicting the stability of coiled-coil designs and offers a new framework for computing binding free energies in protein-protein and multiprotein complexes.

Overview

CoChat Summary

TL;DR

The paper presents a metadynamics-based free-energy protocol with orientational, positional, helical, and RMSD restraints for predicting coiled-coil topology, oligomerization state, and pH-dependent switching. It reports that the method correctly recovers the preferred states of several designed coiled coils and outperforms unrestrained or funnel metadynamics, which often produce misfolded bound states.

Key findings

  • The restrained metadynamics protocol correctly predicted topology for two dimers: CC-Di as a parallel dimer and apCC-Di as an antiparallel dimer; it also recovered the preferred trimer topology of CC-Tri.
  • For oligomerization state, the method matched experiment for CC-Di, CC-Tri, and CC-Tet: dimer, trimer, and tetramer were each preferred in the corresponding native systems.
  • The pH-switchable CC-Hex*-L24E system was also captured: the parallel hexamer is more favorable at low pH, while the antiparallel tetramer is more favorable at high pH; the paper reports ΔGbind,total°/helix values of −5.3 ± 1.3 kcal mol–1 for the low-pH hexamer and −5.4 ± 1.1 kcal mol–1 for the high-pH tetramer.
  • For the stable coiled coils, the paper reports ΔGbind,total°/helix < −10 kcal mol–1 for CC-Di, CC-Tri, and CC-Tet, except for the dimeric CC-Tet at −9.5 ± 0.8 kcal mol–1; apCC-Di had a lower value of −5.3 ± 0.5 kcal mol–1.

Methods

  • The study used well-tempered metadynamics in OpenMM 7.5 with Boresch restraints, plus additional helical ϕ/ψ and RMSD restraints, to compute binding free energies for coiled-coil assemblies treated as fully folded α-helices.
  • Systems analyzed included CC-Di, apCC-Di, CC-Tri, CC-Tet, and CC-Hex*-L24E; structures were obtained from PDBe or designed in ISAMBARD, and CC-Hex*-L24E was modeled by fixed protonation states for low vs high pH.
  • Simulations were run in implicit solvent with the AMBER ff96/OBC2 force field at 298.15 K; unrestrained metadynamics, funnel metadynamics, and restrained Boresch-metadynamics protocols were compared, with three independent replicates used in several runs.

Limitations

  • The paper states that the protocol does not account for the folding free energy of isolated monomers into helices prior to assembly; this could affect the estimated total energetic cost of association.
  • The pH treatment is simplified: only low and high pH conditions with fixed protonation states were examined, and the authors say a constant-pH MD approach would be more realistic.
  • The text makes evident that the method depends on having plausible starting structures and that implicit solvent can struggle with hydrophobic length-scale effects, interfacial water, and interfaces with many salt bridges; the authors note explicit solvent may be needed in such cases.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1021/acs.jctc.4c01695

Cites

133 of 133 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Proposed pipeline for the calculation of binding free energies of de novo coiled coils. (A) Parametric design of coiled coils in ISAMBARD, with user-specified amino acid sequence, topology and oligomerization state. The radius, interface angle and pitch are optimized during design. The designed structures are scored with the BUDE force field. Each design is compared to the design with the lowest BUDE score by their RMSD. (B) Selection of anchor points for Boresch restraints. Three Cα atoms are chosen on each of the “ligand” and “receptor” moieties. Angle and torsion distributions are obtained from equilibrium simulations of the complex. (C) Metadynamics simulations are carried out sequentially in OpenMM, with one helix being removed from the assembly at a time. Torsional restraints are applied on each monomer helix to prevent unfolding upon dissociation. The orientation and position of the “ligand” relative to the “receptor” is restrained with Boresch restraints, and the “receptor” is additionally restrained with RMSD restraints to prevent its rearrangement upon dissociation of the “ligand”.
Figure 1 Proposed pipeline for the calculation of binding free energies of de novo coiled coils. (A) Parametric design of coiled coils in ISAMBARD, with user-specified amino acid sequence, topology and oligomerization state. The radius, interface angle and pitch are optimized during design. The designed structures are scored with the BUDE force field. Each design is compared to the design with the lowest BUDE score by their RMSD. (B) Selection of anchor points for Boresch restraints. Three Cα atoms are chosen on each of the “ligand” and “receptor” moieties. Angle and torsion distributions are obtained from equilibrium simulations of the complex. (C) Metadynamics simulations are carried out sequentially in OpenMM, with one helix being removed from the assembly at a time. Torsional restraints are applied on each monomer helix to prevent unfolding upon dissociation. The orientation and position of the “ligand” relative to the “receptor” is restrained with Boresch restraints, and the “receptor” is additionally restrained with RMSD restraints to prevent its rearrangement upon dissociation of the “ligand”.CC BY · PubMed Central
Figure 2 De novo coiled coils studied and their PDB accession codes. Blue denotes N-terminus and red denotes C-terminus. Rendered with PyMOL.91
Figure 2 De novo coiled coils studied and their PDB accession codes. Blue denotes N-terminus and red denotes C-terminus. Rendered with PyMOL.91CC BY · PubMed Central
Figure 3 Boresch restraints selection and PMF profiles for the apCC-Di dimer. (A) The selected anchor point residues are depicted as sticks. The Boresch restraints are imposed on the Cα atom of each residue, shown as a black sphere. The N-termini of the peptide are shown in blue and the C-termini in red. Rendered with PyMOL.91 (B) The distributions of the Boresch angles and dihedrals and the two unrestrained angles c–b–a and A–B–C. All angles should be away from 0° or 180°. (C) The PMF profiles of each CV, obtained from running simulations with the modified geometrical route.
Figure 3 Boresch restraints selection and PMF profiles for the apCC-Di dimer. (A) The selected anchor point residues are depicted as sticks. The Boresch restraints are imposed on the Cα atom of each residue, shown as a black sphere. The N-termini of the peptide are shown in blue and the C-termini in red. Rendered with PyMOL.91 (B) The distributions of the Boresch angles and dihedrals and the two unrestrained angles c–b–a and A–B–C. All angles should be away from 0° or 180°. (C) The PMF profiles of each CV, obtained from running simulations with the modified geometrical route.CC BY · PubMed Central
Figure 4 (A) Separation PMF profiles for CC-Di and apCC-Di in their parallel and antiparallel conformations. Representative structures of the bound and unbound states are shown for each case. The shaded areas represent standard error. The N-termini are shown in blue and the C-termini are shown in red. (B) The total binding free energy for each dimer, divided by the number of helices in the assembly. The bars represent standard error. ⧫ denotes the native structures. For comparison of dimers, it is not necessary to divide by the number of helices in the assembly, but this has been done here to facilitate comparisons with higher oligomerization states.
Figure 4 (A) Separation PMF profiles for CC-Di and apCC-Di in their parallel and antiparallel conformations. Representative structures of the bound and unbound states are shown for each case. The shaded areas represent standard error. The N-termini are shown in blue and the C-termini are shown in red. (B) The total binding free energy for each dimer, divided by the number of helices in the assembly. The bars represent standard error. ⧫ denotes the native structures. For comparison of dimers, it is not necessary to divide by the number of helices in the assembly, but this has been done here to facilitate comparisons with higher oligomerization states.CC BY · PubMed Central
Figure 5 Total binding free energy for the parallel and up–down–up CC-Tri coiled coils, divided by the number of helices in the assembly. Simulations of the parallel CC-Tri were carried out both in the absence and presence of RMSD restraints on the “receptor” helices. Simulations of the up–down–up CC-Tri were carried out with two different separation schemes, with a different order of helices separated from the assembly in each case. The bars represent standard error. ⧫ denotes the native structures. For comparison of trimers, it is not necessary to divide by the number of helices in the assembly, but this has been done here to facilitate comparisons with other oligomerization states.
Figure 5 Total binding free energy for the parallel and up–down–up CC-Tri coiled coils, divided by the number of helices in the assembly. Simulations of the parallel CC-Tri were carried out both in the absence and presence of RMSD restraints on the “receptor” helices. Simulations of the up–down–up CC-Tri were carried out with two different separation schemes, with a different order of helices separated from the assembly in each case. The bars represent standard error. ⧫ denotes the native structures. For comparison of trimers, it is not necessary to divide by the number of helices in the assembly, but this has been done here to facilitate comparisons with other oligomerization states.CC BY · PubMed Central
Figure 6 Total binding free energy for each dimeric, trimeric and tetrameric complex of the CC-Di, CC-Tri and CC-Tet sequences, divided by the number of helices in the assembly to allow comparisons between different oligomerization states. The bars represent standard error. ⧫ denotes the native structures.
Figure 6 Total binding free energy for each dimeric, trimeric and tetrameric complex of the CC-Di, CC-Tri and CC-Tet sequences, divided by the number of helices in the assembly to allow comparisons between different oligomerization states. The bars represent standard error. ⧫ denotes the native structures.CC BY · PubMed Central
Figure 7 Individual ΔGbind◦ values from each intermediate state calculation for (A) CC-Di, (B) CC-Tri and (C) CC-Tet, each simulated as a parallel dimer, trimer and tetramer. The structures above each ΔGbind◦ value depict the complex being dissociated at each free energy calculation step. The bars represent standard error.
Figure 7 Individual ΔGbind◦ values from each intermediate state calculation for (A) CC-Di, (B) CC-Tri and (C) CC-Tet, each simulated as a parallel dimer, trimer and tetramer. The structures above each ΔGbind◦ value depict the complex being dissociated at each free energy calculation step. The bars represent standard error.CC BY · PubMed Central
Figure 8 (A) The total binding free energies of the CC-Hex*-L24E hexamer and tetramer at low and high pH conditions, divided by the number of helices in the assembly. The gray spheres represent the hydrogens of the protonated Glu residues at the low pH condition. ⧫ denotes the native structures. (B,C) Individual ΔGbind◦ values from each intermediate state calculation for (B) the parallel hexamer and (C) the antiparallel tetramer. The structures above each ΔGbind◦ value depict the complex being dissociated at each free energy calculation step. The insets provide views from the N- to the C-terminus of the peptides and show the protonated Glu at the core (hydrogens depicted in gray spheres). The bars represent standard error.
Figure 8 (A) The total binding free energies of the CC-Hex*-L24E hexamer and tetramer at low and high pH conditions, divided by the number of helices in the assembly. The gray spheres represent the hydrogens of the protonated Glu residues at the low pH condition. ⧫ denotes the native structures. (B,C) Individual ΔGbind◦ values from each intermediate state calculation for (B) the parallel hexamer and (C) the antiparallel tetramer. The structures above each ΔGbind◦ value depict the complex being dissociated at each free energy calculation step. The insets provide views from the N- to the C-terminus of the peptides and show the protonated Glu at the core (hydrogens depicted in gray spheres). The bars represent standard error.CC BY · PubMed Central

Verification

Claims the paper makes

17 Valid · 1 Not valid · 36 Not checked · 104 Own results and interpretations

56 more claims

Create a free account to see all 158 claims and to verify the 30 references still open.

See all claims — free account
  • Probably not a substantial claim. 31
  • It may state two separate claims as one. 68
  • The quote only partly states this claim. 2

Create a free account to see these claims.

Citations

Cites

Cited by

Metadata

Authors
Notari, E., Wood, C. W., & Michel, J.
Affiliations
University of Edinburgh (GB)
Corresponding author
Michel, Julien
Venue
Journal of Chemical Theory and Computation
Published
2025-03-05
Volume
21
Issue
6
Pages
3260-3276
ISSN
  • 1549-9618 (ISSN-L)
  • 1549-9626
OpenAlex
W4408219584
Cited by
4
References
133
Citations by year
2026: 4
FWCI
1.19
Type
journal article
Language
en
Open access
Locations
  • Journal of Chemical Theory and Computation (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Edinburgh Research Explorer (University of Edinburgh) (submittedVersion, cc-by)
Funders
Biotechnology and Biological Sciences Research Council (2664327, BB/X003027/1)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
  • Physical Sciences › Chemistry › Physical and Theoretical Chemistry › Advanced Physical and Chemical Molecular Interactions
  • Physical Sciences › Chemistry › Organic Chemistry › Chemical Synthesis and Analysis
Keywords
coiled coil, protein design, metadynamics, conformational restraints, oligomeric state, binding free energy, protein-protein complex, protein complexes
MeSH terms
Hydrogen-Ion Concentration, Protein Conformation, Proteins, Thermodynamics, Protein Multimerization, Molecular Dynamics Simulation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.