Coarse-Grained Molecular Dynamics Modelling of Human Low-Density Lipoprotein

Abstract:

 

Cardiovascular disease is accounted for 32% of death worldwide, and is considered the leading cause of premature mortality. For its risk assessment and treatment, modern guidelines indicate serum low-density lipoprotein (LDL) cholesterol level as the primary target. These LDLs (~17-28 nm) [1] are complex amphiphilic nanoparticles that are instrumental in the metabolism and homeostasis of cholesterol and lipid transportation.
Recent developments in cryo-electron microscopy (cryoEM), and artificial intelligence-based algorithms opened new avenues towards solving the structure. In this current study, my task was to apply molecular dynamics simulations based on mid-resolution cryo-electron microscopy (cryoEM) data, AlphaFold predictive software, and Python-based model building and data processing. I utilized GROMACS.2024 and the Martini3 force field [2] for coarse-grained simulations. The missing force field parameters for glycans and lipids were generated to reproduce the CHARMM36 all-atom representation geometries based on Earthmovers’ distance and heavy-ball minimization. My modelling approach first divided the LDL particle into three main parts: a) the Apolipoprotein B100 (apo-B100, ~550kDa, with 17 N-glycans), b) the phospholipid membrane surface (phosphatidylcholine lipids, cholesterol), c) the lipid core (cholesterol, cholesterol esters, triglycerides). Structure predictions indicate the amphiphilic nature of apo-B100 originates from the well-tuned sequence of the 3-4 nm-wide β-belt motif with over 2000 residues that surrounds the LDL lipid core. The core forms three double-layers at lower temperatures, and is stabilized by the lipophilic side of the β-belt. With clustering analysis, I was able to reveale surface domains with specific compositions that suggests the formation of cholesterol-rich reservoirs in the LDL surface membrane. Finally, I combined the three parts into simplified and full LDL simulations. They clearly demonstrate the conformational preferences and dynamics of flexible protein segments. These findings allow us to further expand our understanding of this particle and pave out the road towards further investigations. The research is a joint project financed by the Austrian Science Fund and the Agence Nationale de la Recherche.
 
[1] a, Prassl, R. and P. Laggner, Eur Biophys J, 2009, 38(2), 145-58 b, Kuklenyik, Z., et al., PLoS One, 2018, 13(4), e0194797 c, Cisse, A., et al., Int J Biol Macromol, 2023, 252, 126345.
[2] Souza, P.C.T., Alessandri, R., Barnoud, J. et al. Nat Methods 2021, 18, 382–388.