Thèse Electrostatique de l'Eau dans les Structures Cristallines d'Amidon H/F - 59

Établissement : Université de Lille
École doctorale : Sciences de la Matière du Rayonnement et de l'Environnement
Laboratoire de recherche : UGSF - Unité de Glycobiologie Structurale et Fonctionnelle
Direction de la thèse : Ralf BLOSSEY ORCID 0000000248237037
Début de la thèse : 2026-10-01
Date limite de candidature : 2026-12-31T23:59:59

L'amidon, un polysaccharide, est la source d'hydrate de carbone la plus importante pour la nutrition humaine, présent sous forme de granules, structures semi-cristallines contenant des régions cristallines ordonnées. La cristallinité de l'amidon varie entre différentes plantes nutritionnelles et a une teneur en eau différente selon le type. La pression hydrostatique peut modifier la structure moléculaire des cristallites d'amidon, mais son influence n'est pas encore comprise. Notre projet vise à une compréhension quantitative du rôle de la teneur en eau sur les structures cristallines de l'amidon. Notre hypothèse clé est que l'influence de l'eau sur la structure moléculaire des cristallites d'amidon peut être comprise par une cartographie quantitative des interfaces internes avec l'aide du potentiel électrostatique. L'obtention d'une vue claire sur la façon dont les molécules d'eau actuelles modifient la structure de l'amidon au niveau moléculaire et atomique est liée à la compréhension de la manière dont l'humidité environnementale influence le type de cristal d'amidon, dont notre étude pourrait donner une première indication.
À cette fin, nous proposons une approche computationnelle multi-échelle, allant des échelles de longueur atomistique aux échelles structurelles. Une quantité centrale ici est le potentiel électrostatique, qui sera utilisé pour identifier les interactions et estimer leur force.

Starch, a polysaccharide, is the most important carbohydrate source for human nutrition, as it is produced by most green plants for energy storage. It consists of two types of molecules, the linear and helical amylose and the branched amylopectin. Depending on the plant, starch contains 20 to 25% amylose and 75 to 80% amylopectin by weight. In plant cells, starch is present in the form of granules, semi-crystalline structures containing ordered crystalline regions, formed by the starch polymers, and resulting in double-helical and lamellar structures. Starch crystallinity varies between different nutritional plants, of which three are considered canonical: maize (A-type), potato (B-type), and pea (C-type). This nomenclature reflects the water content in the crystallites: A-type crystals contain only 8 inter-helical water molecules per crystal unit, while B-type crystals have a more open packing with 36 water molecules per crystal unit1,2 . C-type starch consists of both A- and B-type polymorphs3. The water content of the crystallites is influenced by the environmental conditions. Earlier studies make it clear4,5,6 that environmental hydrostatic pressure modifies the starch structure7, but it is not understood how. On the other hand, the role water plays in the structure of cellular fibrils and crystallites is well-acknowledged in the literature, e.g. for cellulose8 and amyloid fibrils9. For nanocellulose, the significance of water in cellulose structures is reviewed8. But for starch granules, very little is known about how native granules transition between A-, B-, and C-type starch structures in response to hydration changes, particularly under climate-relevant stresses. Our study might give a first indication hereof, making this academic study towards the impact of water evaporation on crystalline structures very timely in the context of the current climate change.
Further, to tackle our research questions, we use a multidisciplinary approach at the frontline of the state of the art, hereby providing I. macromolecular electrostatics at the starch assembly level10 and II. insight at the atomistic level, using electrostatic potential calculations.

Our project aims at a quantitative understanding up to the molecular and atomic level of the role of water in starch crystallite structures. Our key hypothesis is that the influence of water on the structure of starch crystallites can be understood by a quantitative mapping of internal water interfaces with the help of electrostatic potential10,18. Our goal is to elucidate the fundamentals of the interactions making the water-interfaces in the different crystal types found in stored starch, up to the molecular and atomic level. Indirectly, this can give some understanding how environmental moisture modifies the starch crystal structure 6,7. This relates to identifying how altering climate conditions impact starch storage.

This project is designed around multiscale modeling, ranging from atomistic to structural length scales.
We combine 3 levels of modelling, going from a broad vision to a detailed description at the atomistic level. This multi-level system approach has recently been successfully used before in our team11,12, and consists of:
1. Continuum models for the electrostatic potential based on the Poisson-Boltzmann model and its variants, which have matured over the last years to include the role of water in recent years10 (cfr. amyloid fibril aggregates as example9). These models will be used to quantify the free energies of starch assemblies, hence allowing for an understanding of the formation of the different entities' structures.
2. Molecular dynamics simulations will reveal how structures are held together. They give an overview of details of the structural and dynamical aspects of the starch assemblies.
3. Quantum mechanical calculations will give information on the atomistic level. Hereby, the details of the interactions will be revealed. We will perform all-electron single molecule calculations giving more precise data for the interactions, but on smaller systems.

A central quantity at all three modelling levels will be the electrostatic potential. It will be calculated at the classical physics level and at the quantum mechanical level, hereby providing the student an overall training in different methodologies of molecular modelling.
At the classical physics level, a detailed comparison of continuum approaches with MD simulations, giving criteria for their applicability, was performed in our team13. These models will be used to quantify the free energies of starch assemblies, hence allowing for an understanding of the formation of the different entities' structures.
At the quantum mechanical level, the electrostatic potential will be calculated at the molecular surface to visualize and quantify the non-covalent water-glucan interactions. In using electrostatic potentials to analyze noncovalent interactions, V(r) is now generally computed on a molecular surface14. Our new contribution here is that we consider the full 3D nature of the electrostatic potential, not only at outer iso-density surfaces, as suggested in the 1990's14, but at contours much closer to the nuclei, a methodology established in work by recent our team15-17