Coarse-Grained Simulation of Biosurfactants for Risk Assessment

Knowledge of a material's interactions with biological membranes is used in environmental risk assessment, but in silico predictive methods are unvalidated for novel, bio-based surfactants. In particular, there are currently there are almost no published coarse-grained simulations of rhamnolipids. This class of lipids contain one or two rhamnose moeties (a deoxy sugar) and are synthesised by certain bacteria. Experimental studies have begun to explore the effects of rhamnolipids on the physical properties of phospholipid membranes, such as their tendency to induce membrane curvature [1]. Rhamnolipids have also been reported to have antimicrobial benefits for the cells that produce them. Compared to conventional synthetic surfactants, biosurfactants have the major advantages of being possible to produce from renewable sources and showing low toxicity.

An ongoing collaboration between Unilever and Durham University has yielded new simulation methodology to map, parametrise and run simulations in the Martini coarse-grained framework [2,3] (including the new Martini 3 version [4]) to predict membrane-water partitioning (log KMW) for small organic molecules. This physical property is important for risk assessment in connection with bioaccumulation and baseline toxicity. Log KMW is a more accurate and pertinent physical quantity for such assessments than the more established but cruder octanol-water partitioning coefficient (log KOW), which does not directly account for the phospholipid bilayer of biological membranes. The Durham-Unilever collaboration is currently expanding the chemical scope of log KMW predictions to include some classes of conventional surfactants. This new project will extend and validate this methodology to predict physical properties of biosurfactants.

Goals:
(a) Develop simulation capability for better understanding of the interactions of biosurfactants in biological systems.
(b) Understand and quantify how variation within biosurfactant classes impacts their physico-chemical properties.
(c) Characterise the influence of biosurfactant structure and membrane composition on membrane shape effects.

[1] M Herzog, T Tiso, LM Blank and R Winter; BBA Biomembranes 1862, 183431 (2020)
[2] TD Potter, EL Barrett and MA Miller; J Chem Theor Comput 17 5777 (2021)
[3] TD Potter, N Haywod, A Teixeira, G Hodges, EL Barrett and MA Miller; Env Sci Proc Impacts 25 1082 (2023)
[4] PCT Souza et al.; Nature Methods 18 382 (2021)