Abstract
Cell membranes are one of the most fundamental structures in biology. In addition to the other constituents in the membrane, such as proteins and carbohydrates, the natural cell membrane is also composed of a diverse mixture of lipid types. Disturbance in the lipid composition has been associated with many brain-related disorders. In combination with a variety of experimental methods, molecular dynamics (MD) simulations have been used extensively to probe membrane-related systems. MD simulations are specifically advantageous for the extraction of structural, thermodynamic, and interaction details at the atomic level. Over the many decades that cell membranes have been studied using simulations, the membrane models that have been used are largely limited to lipid compositions that are homogeneous and/or symmetric. This thesis focuses on the all-atom MD simulations of brain cell membranes using heterogeneous lipid compositions. An introduction to the lipids that constitute the cell membrane and a brief background on MD simulations is provided in Chapter 1. Subsequently, the computational methodology behind atomistic MD simulations is provided in Chapter 2.Chapter 3 presents the results of atomistic MD simulations that were used to compare four membrane models of increasing complexity: a homogeneous 1-palmitoyl-2-oleoyl phosphatidylcholine (POPC) membrane, a bi-component POPC:cholesterol membrane (1:1 ratio), and the first realistic models of a healthy (BH) and an Alzheimer’s diseased (BAD) human brain cell membrane. The lipid compositions of BH and BAD were modelled based on a subset of their respective lipidomic profiles, which were obtained from experiment and comprise of 27 different lipid types. Results from the simulations reveal effective differences in the structure and dynamic properties between the simple and complex membranes, especially in the membrane dipole potential and the lateral diffusion of lipids.
Among the various lipid types, a primary lipid constituent in the human brain is the omega-3 fatty acid, docosahexaenoic acid (DHA). Chapter 4 presents an investigation on the effects of free protonated DHA (DHAP) that is included as part of the lipid composition in the complex healthy brain model (BH). Increasing molar concentrations of DHAP in the BH membrane at 0%, 17%, 30%, and 38%, notably, results in an increased disorder of the membrane packing, increased membrane fluidity, and increased flip-flop rates of DHAP. Interestingly, the MD simulations also revealed two distinct flip-flop mechanisms involving pairs of DHAP molecules in the events of double flip-flop or assisted flip-flop.
Then, Chapter 5 explores the interactions of a small library of drugs with a heterogeneous and asymmetric membrane. The library of drugs include clonidine, dopamine, ibuprofen, and salicylic acid. The membrane composition includes cholesterol, POPC, 1-palmitoyl-2-oleoyl phosphatidylethanolamine (POPE), 1-palmitoyl-2-oleoyl phosphatidylserine (POPS), and palmitoyl sphingomyelin (PSM), in ratios taken from existing literature in an effort to mimic the blood-brain barrier. The MD simulations show that all of the various drugs are able to diffuse from the water layers directly into the membrane without protein assistance. In addition, flip-flop events were observed for clonidine, ibuprofen, and salicylic acid, but not for dopamine. Generally, insertion of the various drugs into the bilayer causes a reduction in the bilayer thickness, disorders the membrane packing, and interestingly, has mixed effects on the lateral diffusion of the various lipid types.
Ultimately, the results presented in this thesis have contributed to an understanding on human cell membranes that involve a complex lipid mixture. Additionally, insight into the effects of fatty acids and small therapeutics on the structure and dynamic properties will be beneficial in the comprehension of how these complex membranes react to such molecules in their environment. Finally, knowledge on the transport mechanisms of the various small molecules and fatty acids will be particularly relevant to the pharmacokinetics of novel therapeutics, and thus, aid in processes, such as drug design and drug development.
| Date of Award | 1 Jul 2022 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Chris Lorenz (Supervisor) & Dylan Owen (Supervisor) |