MULTISCALE MOLECULAR SIMULATIONS OF COMPLETE LOWDENSITY LIPOPROTEIN PARTICLES

Low-density lipoprotein (LDL) plays a vital role in human health by carrying

cholesterol and other lipids through the bloodstream to tissues throughout the body.

However, its retention and oxidation within arterial walls can promote atherosclerosis

and downstream cardiovascular disease, the leading cause of death globally. LDL

particles are structurally heterogeneous, consisting of a hydrophobic core rich in

cholesteryl esters and triglycerides surrounded by a surface monolayer of

phospholipids, free cholesterol, and a single copy of apolipoprotein B-100 (apoB).

One of the largest monomeric proteins in the human proteome, apoB wraps around

the particle, stabilizing it and mediating clearance via the LDL receptor. Recently, we

reported the first structure of full-length apoB, determined using cryo-electron

microscopy (cryo-EM), AlphaFold, and molecular dynamics (MD) simulation

(Berndsen and Cassidy, Nature, 2025), and have since extended this framework to

particles of multiple sizes. While these structures clarify underlying conformational

changes in apoB, they provide limited insight into the dynamic organization of the lipid

particle itself. Multiscale MD simulations offer a powerful complementary approach for

resolving lipid structure and dynamics at high resolution. Here, we present a

computational pipeline for constructing atomic and near atomic models of complete

LDL particles guided by cryo-EM density maps. This multiscale approach enables

direct interrogation of lipid-species-specific binding, diffusion, and overall organization

using a Martini coarse-grained representation, while capturing local and global

changes in apoB conformation at all-atom resolution. Taken together, these results

establish a computational framework for systematically investigating how lipid

composition, particle size, and disease-associated mutations alter LDL function and

pathology.