The assembled catalytic triad of Ideonella sakaiensis PETase is metastable in apo molecular dynamics simulations across multiple control conditions
Abstract
Active-site geometry is routinely reported from molecular dynamics of engineered variants of Ideonella sakaiensis PETase (IsPETase) as evidence that a design preserves catalytic capability. We simulated wild-type IsPETase and six length-matched linker fusion constructs, four replicates each, for 350 ns at 343 K under a preregistered analysis plan. Twenty-six of 28 trajectories began with the Ser133 OG to His210 NE2 distance inside the competent range, below 0.35 nm, and two began marginally above it. All 28 ended outside that range without recovery. The median time to crossing, from the onset of free dynamics, was 4.06 ns, with a range spanning more than four orders of magnitude. Triad loss was not generally accompanied by global destabilisation. The loss persisted when the system was built from the crystal structure rather than a predicted model, when both native disulfide bonds were present, and at 300 K. Changing the catalytic histidine protonation state did not restore persistent catalytic geometry. A post hoc control under a second protein and water parameter set also left the competent range in all four replicates, but reached separations of 0.48 to 0.90 nm rather than the 1.9 to 2.1 nm seen under CHARMM36m. Because the endpoint failed in the wild-type control, the intended comparison of linker family and length could not be completed. Active-site distances from apo IsPETase simulations should not be interpreted as evidence of catalytic competence without a wild-type control and replicate sampling.
Related
- Whole-Enzyme Reactive Free-Energy Simulations with a Universal Machine Learning Interatomic Potential: Catalytic Mechanism and Mutational Effects of PETase
- Metastable phase switching and an effective h/2e flux scale in hydrated DNA
- A protein–foldamer supramolecular synthon for self-assembled hybrid architectures
- Broad Stability and pH-Mediated Reversible Disassembly of In Vitro Assembled Bacterial Microcompartment Shells.
- In-column Synthesis and Atomic-resolution Electron Microscopic Analysis of Metastable Reactive Nitrides
BibTeX
@misc{okorafor2026assembled,
title = {{The assembled catalytic triad of Ideonella sakaiensis PETase is metastable in apo molecular dynamics simulations across multiple control conditions}},
author = {Amarechi Okorafor and Aryan Atri and Ayushmaan Singh},
howpublished = {ChemRxiv},
year = {2026},
month = oct,
doi = {10.26434/chemrxiv.15009960/v1},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009960/v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - The assembled catalytic triad of Ideonella sakaiensis PETase is metastable in apo molecular dynamics simulations across multiple control conditions
AU - Amarechi Okorafor
AU - Aryan Atri
AU - Ayushmaan Singh
PY - 2026
DA - 2026/10/06
PB - ChemRxiv
DO - 10.26434/chemrxiv.15009960/v1
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009960/v1
AB - Active-site geometry is routinely reported from molecular dynamics of engineered variants of Ideonella sakaiensis PETase (IsPETase) as evidence that a design preserves catalytic capability. We simulated wild-type IsPETase and six length-matched linker fusion constructs, four replicates each, for 350 ns at 343 K under a preregistered analysis plan. Twenty-six of 28 trajectories began with the Ser133 OG to His210 NE2 distance inside the competent range, below 0.35 nm, and two began marginally above it. All 28 ended outside that range without recovery. The median time to crossing, from the onset of free dynamics, was 4.06 ns, with a range spanning more than four orders of magnitude. Triad loss was not generally accompanied by global destabilisation. The loss persisted when the system was built from the crystal structure rather than a predicted model, when both native disulfide bonds were present, and at 300 K. Changing the catalytic histidine protonation state did not restore persistent catalytic geometry. A post hoc control under a second protein and water parameter set also left the competent range in all four replicates, but reached separations of 0.48 to 0.90 nm rather than the 1.9 to 2.1 nm seen under CHARMM36m. Because the endpoint failed in the wild-type control, the intended comparison of linker family and length could not be completed. Active-site distances from apo IsPETase simulations should not be interpreted as evidence of catalytic competence without a wild-type control and replicate sampling.
ER -