An experimentally representative in-silico protocol for dynamical studies of lyophilised and weakly hydrated amorphous proteins

Elisa Bassotti; Sara Gabrielli; Gaio Paradossi; Ester Chiessi; Mark Telling

doi:10.1038/s42004-024-01167-6

An experimentally representative in-silico protocol for dynamical studies of lyophilised and weakly hydrated amorphous proteins

Commun Chem. 2024 Apr 12;7(1):83. doi: 10.1038/s42004-024-01167-6.

Authors

Elisa Bassotti¹, Sara Gabrielli¹, Gaio Paradossi¹, Ester Chiessi², Mark Telling^{3

4}

Affiliations

¹ Department of Chemical Science and Technologies, University of Rome Tor Vergata, Via della Ricerca Scientifica I, 00133, Rome, Italy.
² Department of Chemical Science and Technologies, University of Rome Tor Vergata, Via della Ricerca Scientifica I, 00133, Rome, Italy. ester.chiessi@uniroma2.it.
³ STFC, ISIS Facility, Rutherford Appleton Laboratory, Harwell Campus, Didcot, OX11OQX, UK. mark.telling@stfc.ac.uk.
⁴ Department of Materials, University of Oxford, Parks Road, Oxford, UK. mark.telling@stfc.ac.uk.

Abstract

Characterization of biopolymers in both dry and weakly hydrated amorphous states has implications for the pharmaceutical industry since it provides understanding of the effect of lyophilisation on stability and biological activity. Atomistic Molecular Dynamics (MD) simulations probe structural and dynamical features related to system functionality. However, while simulations in homogenous aqueous environments are routine, dehydrated model assemblies are a challenge with systems investigated in-silico needing careful consideration; simulated systems potentially differing markedly despite seemingly negligible changes in procedure. Here we propose an in-silico protocol to model proteins in lyophilised and weakly hydrated amorphous states that is both more experimentally representative and routinely applicable. Since the outputs from MD align directly with those accessed by neutron scattering, the efficacy of the simulation protocol proposed is shown by validating against experimental neutron data for apoferritin and insulin. This work also highlights that without cooperative experimental and simulative data, development of simulative procedures using MD alone would prove most challenging.