Conformational footprints of agonism: Differences of inositol 1,4,5-trisphosphate (IP3) and adenophostin A on IP3 receptor’s N-terminal dynamics
Abstract
Inositol 1,4,5-trisphosphate receptor (IP3R) represents a major family of intracellular Ca2+-release channel. Distal to its pore-forming region lies its cytoplasmic N-terminus (NT) that harbours the agonist-binding pocket. Although inositol 1,4,5-trisphosphate (IP3) is the endogenous agonist, the fungal metabolite adenophostin A (AdA) is known to function as a “super-agonist”, displaying roughly ten-fold higher affinity on binding. Using all-atom molecular dynamics simulations of rat IP3R1 NT in apo, IP3-bound and AdA-bound states, we here show that both agonists alter NT’s flexibility, yet principal component analysis reveals that AdA drives the domain into broader and distinct conformational substates that are visited by neither the apo nor the IP3-bound form. AdA occupies a more spacious, hydrophobic pocket, engaging a wider spectrum of transient polar and non-polar contacts and favouring an entropy-dominated binding mode despite fewer enduring hydrogen bonds. Dynamic cross-correlation analysis further demonstrates that AdA enhances long-range correlated motions characteristic of a gate-ready ensemble. Taken together, the data indicate that AdA’s superior potency arises from its unique capacity to remodel NT dynamics more profoundly than IP3, thereby pre-configuring or priming IP3R for activation gating. These findings refine current models of ligand efficacy and offer a framework for the rational design of next-generation IP3R modulators.
Dataset DOI: 10.5061/dryad.98sf7m0wp
Overview
This dataset contains molecular dynamics (MD) simulation outputs and Boltz-2 structure prediction outputs used to investigate the structural dynamics and ligand binding of the IP3 receptor N-terminal region in three states:
- apo (unbound)
- IP3-bound
- adenophostin A (AdA)-bound
The MD data include reference structures and simulation trajectories. The Boltz-2 data include predicted structures, affinity outputs, confidence summaries, and model confidence/error files.
File structure
data.zip
Compressed archive containing the following folders:
MD_data/boltz_data/
MD_data/
This folder contains structures and trajectories for the apo, IP3-bound, and AdA-bound systems.
Reference structure files
apo.pdbIP3_bound.pdbAdA_bound.pdb
These are Protein Data Bank (.pdb) files containing atom-level coordinates for each system. They can be opened directly in molecular visualisation software such as VMD, PyMOL, or ChimeraX.
These files also serve as the reference structures for the corresponding trajectory files.
Trajectory files
[system]_RUN1.xtc[system]_RUN2.xtc[system]_RUN3.xtc
where [system] is one of:
apoIP3_boundAdA_bound
These are GROMACS compressed trajectory (.xtc) files from three independent MD simulation runs for each system.
The .xtc files are not only companion files; they can be viewed and analysed directly, but they should be loaded together with the matching .pdb file because the trajectory file contains coordinate frames only and relies on the reference structure for atom and residue definitions.
Recommended file pairings
apo.pdbwithapo_RUN1.xtc,apo_RUN2.xtc,apo_RUN3.xtcIP3_bound.pdbwithIP3_bound_RUN1.xtc,IP3_bound_RUN2.xtc,IP3_bound_RUN3.xtcAdA_bound.pdbwithAdA_bound_RUN1.xtc,AdA_bound_RUN2.xtc,AdA_bound_RUN3.xtc
Suggested use
- Use
.pdbfiles to inspect a single structure - Use
.pdband.xtcfiles together to view structural motion over time - Use the trajectories for downstream MD analyses such as RMSD, RMSF, distance measurements, and conformational comparison between systems and runs
Recommended software
- VMD: recommended for viewing
.pdband.xtcfiles - GROMACS: recommended for trajectory analysis
- MDAnalysis or MDTraj: suitable for Python-based reanalysis
boltz_data/
This folder contains Boltz-2 outputs for ligand-bound systems.
Subfolders
P29994_AdA/P29994_IP3/
Each folder contains output files for one ligand-bound system.
Files in each Boltz-2 folder
affinity_*.json
JSON files containing predicted binding affinity output and related metadata.
These files can be opened in any text editor or analysed in Python.
*_model_0.cif
Structure files in CIF format containing the predicted protein–ligand complex.
These can be opened in molecular visualisation software such as VMD, PyMOL, or ChimeraX.
confidence_*.json
JSON files containing model confidence information, including per-residue confidence scores on a 0–100 scale.
pae_*.npz, pde_*.npz, plddt_*.npz, pre_*.npz
Compressed NumPy archive files containing Boltz-2 confidence and error metrics.
These files are intended for computational reanalysis and are best accessed in Python using numpy.load().
Suggested use
- Use
.ciffiles to inspect predicted ligand-bound structures - Use
.jsonfiles to review affinity and confidence outputs - Use
.npzfiles for numerical reanalysis of Boltz-2 confidence and error metrics
Recommended software
- Text editor: for inspecting
.jsonfiles - Python + NumPy: for opening
.npzfiles - VMD, PyMOL, or ChimeraX: for viewing
.ciffiles
Guidance for accessing the files
Users who are new to MD data may wish to begin with the following workflow:
- Open one of the
.pdbfiles to view the structure of a system. - Load the matching
.xtcfile alongside that.pdbfile in VMD to view the simulation trajectory. - Open the Boltz-2
.ciffiles to inspect predicted ligand-bound structures. - Open the
.jsonand.npzfiles if numerical reanalysis of the Boltz-2 outputs is required.
In practical terms, the .pdb files are the main entry point for structure viewing, while the .xtc files are the main files for following the MD simulations over time.
Software used
- GROMACS 2023.3: used for MD simulations and trajectory analysis
- Boltz-2: used for structure prediction and binding analysis
Abbreviations
- apo: unbound receptor
- IP3: inositol 1,4,5-trisphosphate
- AdA: adenophostin A
- PDB: Protein Data Bank format
- CIF: Crystallographic Information File
- XTC: GROMACS compressed trajectory format
- JSON: text-based structured data format
- NPZ: compressed NumPy array format