Molecular dynamics simulations of water in glassy carbohydrates
Title: Molecular dynamics simulations of water in glassy carbohydrates
SNIC Project: SNIC 2021/22-149
Project Type: SNIC Small Compute
Principal Investigator: Vitaly Kocherbitov <vitaly.kocherbitov@mau.se>
Affiliation: Malmö universitet
Duration: 2021-03-08 – 2022-04-01
Classification: 10402
Keywords:

Abstract

BACKGROUND Glassy state of carbohydrates is used to stabilize biological molecules such as proteins in solid-state pharmaceutical formulations. The state of residual water in such formulations, including its dynamics plays a decisive role in physical and chemical stability of the formulations. Recently, using advanced calorimetric methods we obtained a detailed thermodynamic data on heat capacities of water confined in glassy matrix, which shows rather unexpected differences between states of water in small and large carbohydrates (i.e. disaccharides vs polymers). AIM We propose to perform molecular dynamics simulations of water in glassy carbohydrates to understand the energy, dynamics and heat capacity of water in dynamical arrest conditions. In particular, three types of carbohydrates will be studied: trehalose, sucrose and maltodextrin. Two different types of temperature programs will be used. Firstly, cooling-heating cycles: the systems will be equilibrated at high temperature, then a cooling scan will be performed, followed by a heating scan to evaluate the main properties of the glass transition (Tg, Ea etc). Secondly, isothermal simulations in the glassy state obtained by cooling from an equilibrium liquid to monitor dynamics of water molecules in the glassy state will be done. PRELIMINARY RESULTS We performed test MD simulations sucrose on desktop computers and the results look very promising. Glass transitions in both volume and enthalpy curves are clearly seen and are in the reasonable temperature range (considering the difference in scan rates between simulations and calorimetric experiments). Nest step is to use longer simulations to get results closer to experimental conditions and also collect statistics to decrease the noise level of the data. For that, an access to a computer cluster is needed.