High-Density Lipoproteins (HDL)

Discoidal HDL

Lipoproteins are protein-lipid particles which circulate in the blood collecting cholesterol, fatty acids and lipids. One such lipoprotein particle is called high-density lipoproteins (HDL) and is often regarded as "good cholesterol" due to its ability to remove cholesterol from artery walls and transport them to the liver for degradation. Low levels of HDL have been implicated in the increased risk of coronary heart disease, the single largest killer in the United States.

The reverse cholesterol transport pathway regulates the production, transformation, and degradation of HDL particles. Apolipoprotein A-1 (apo A-1), the primary protein component of HDL, is produced in the liver and intestines. Apo A-1 initially forms a lipid-free/poor particle in which little or no lipid is associated with the protein. The efflux of lipids and cholesterol from peripheral tissues to lipid-free/poor HDL particles results in the formation of nascent discoidal HDL particles. The free cholesterol that is located on the surface is then esterfied by lecithin cholesterol acyl transferase. The esterfied cholesterol migrates to the core of the nascent discoidal HDL causing the particles to change from discoidal to spherical. These mature spherical HDL will continue to grow when additional cholesterol from cells aggregate on the surface of the particles and are subsequently esterfied. The spherical HDL are eventually degraded and recycled in the liver, thus allowing for the removal of cholesterol from tissues.

Nanodiscs

Nanodisc with bR

Native HDL are heterogeneous particles with a variety of shapes thus precluding structural studies of apo A-1. However, reconstituted HDL, in which purified (and often truncated) apo A-1 is used to form HDL particles with a variety of lipids, with and without cholesterol, can be made into homogeneous particles. Nanodiscs are an engineered rHDL mimic being developed by our collaborator Stephen Sligar (U. Illinois at Urbana-Champaign) which can be self-assembled using a precise set of optimized conditions to form discoidal protein-lipid particles with homogeneous size and composition. These nanodiscs are being developed as platforms in which to embed membrane proteins in a native phospholipid bilayer environment. Various membrane proteins, such as cytochrome P450s, bacteriorhodopsin, rhodopsin, translocon, blood clotting factors, and bacterial chemotaxis receptors, have already been successfully incorporated into nanodiscs. We utilized the extensive characterization of nanodiscs in furthering our understanding of the structure of apo A-1 in HDL as well as on the assembly of such particles.

Structure of Apo A-1 and Assembly of Nanodiscs and HDL

Native discoidal HDLs are heterogeneous particles which can exist in a variety of sizes. There is evidence of particles with two, three, or four apo A-1 proteins per HDL. The most widely studied of these particles are those with two proteins per HDL particle, such as nanodiscs, for which three models exist for the orientation and structure of apo A-1 around the lipid bilayer: double-belt, helical-hairpin, and picket-fence models. A variety of evidence exists for each of these models, although recent experimental and theoretical data favor either a double-belt or a helical-hairpin, but not the picket-fence model.

Lipid binding domain of apolipoprotein A-1

Secondary structure analysis of apo A-1 suggested a 43 residue N-terminal globular domain and a 200 residue C-terminal lipid binding domain. However, experimentally, nanodiscs prepared with truncated apo A-1 proteins (&Delta1-43, &Delta1-54, &Delta1-65), called membrane scaffold proteins, resulted in discs of identical size (9.6 nm) and composition (2 proteins and 160 DPPC lipids per disc). All-atom molecular dynamics simulations were done on these systems in order to determine the actual lipid binding domain of apo A-1 in formation of these discoidal particles (Biophysical Journal, 88:548-556, 2005). Based on mean surface area per lipid calculations, comparison of small angle x-ray scattering curves, and the relatively planar shape of nanodiscs made from truncated scaffold proteins, we concluded that the first 17 to 18 residues of the 200-residue apo A-1 lipid-binding domain are not involved in formation of the protein belts surrounding the lipid bilayer. Subsequent, coarse grained molecular dynamics simulations also confirmed the lipid binding domain of apo A-I was shorter then originally predicted (Journal of Structural Biology, 157:579-592, 2007).

Apo A-I  Lipid Binding Domain

Assembly of Nanodiscs & HDL

Nanodisc assembly was optimized to produce homogeneous particles. The process uses cholate (a detergent) to solubilize lipids and membrane scaffold proteins (engineered truncated apo A-1). Upon removal of cholate using either dialysis or detergent removal beads, the lipids and proteins self-assemble into discoidal particles with discrete size and composition. Due to the long time scales of such assembly processes a coarse-grained model was used to allow molecular dynamics simulations into the microsecond time scale (all-atom molecular dynamics is limited to the nanosecond time scale).

Molecular dynamics simulations of pre-assembled nanodiscs using the coarse grained model accurately reproduced overall structural features, such as bilayer thickness and disc diameter. Increases in temperature resulted in a decrease in bilayer thickness and an increase in disc diameter (Journal of Physical Chemistry B, 110:3674-3684, 2006) in agreement with experimental data.

The coarse grained model was next used to simulate the assembly of nanodiscs starting from various configurations, including one simulation in which the two scaffold proteins are shaped as half-circles and 160 DPPC lipids randomly scattered (see movie below). Results of the 10 &mus simulations showed that the intial aggregation of proteins and lipids is driven by the hydrophobic effect and results in the formation of a single particle within several microseconds. This initial aggregation is followed by a much slower protein tertiary structure rearrangement which eventually leads to the formation of a double-belt discoidal HDL particle (Journal of Structural Biology, 157:579-592, 2007 and Journal of Physical Chemistry B, 111:11095-11104, 2007).

HDL Assembly

Disassembly of Nanodiscs with Cholate

The effect of cholate on pre-formed nanodisc particles was investigated using both coarse-grained molecular dynamics and experimental small-angle X-ray scattering measurements. Small-angle X-ray scattering provides low-resolution structural information of solution state nanodiscs, allowing one to capture the "disassembly" of nanodiscs. The addition of small concentrations of cholate (5 to 10 molecules) to nanodiscs was found to have very little effect on the discoidal shape and size of nanodiscs. Additions of larger amounts of cholate (50 to 100 molecules) resulted in an opening of the scaffold proteins encircling the lipid bilayer resulting in an increase in overall particle dimension. The scaffold proteins in the coarse-grained simulations, fluctuated at the edge of the discs, never falling off the edge to lay along the top of the nanodiscs. Experimental small-angle X-ray scattering evidence suggests that given enough time, the proteins do in fact fall move from the perimeter of the lipids to lay on top of the bilayer. Addition of cholate concentrations equivalent to starting ratios used in nanodisc assembly (i.e., 320 cholate molecules), resulted in the formation of a spherical particle with a water filled interior cavity and with the scaffold proteins wrapped around the outside. This particle upon reverse coarse-graining and simulating using all-atom molecular dynamics was observed to fall apart, as would be expected (Nano Letters, 7:1692-1696, 2007).

Nanodisc disassembly with cholate

Maturation of HDL

The maturation of HDL particles from discoidal to spherical involves the conversion of cholesterol to cholesterol ester by the enzyme LCAT and the absorption of the cholesterol ester molecules. A series of molecular dynamics simulations were performed which describe the maturation of discoidal HDL into spherical HDL upon absorption of cholesterol ester as well as the resulting atomic level structure of a mature circulating spherical HDL particle. Sixty cholesterol ester molecules were added to a discoidal HDL particle containing two apolipoproteins wrapped around a 160 DPPC lipid bilayer in a stepwise fashion. The resulting matured particle, captured first in a coarse-grained description, was then described in a consistent all-atom representation and analyzed in chemical detail. The simulations show that maturation results from the formation of a hydrophobic core comprised of cholesterol ester surrounded by phospholipid and protein; the two apolipoprotein strands remain in a belt-like conformation with flexible N- and C-terminal helices and a central region stabilized by salt-bridges. The juxtaposition of the stable central region, which has been proposed to bind lectithin cholesterol acyl transferase, with the flexible terminal helices provides an ideal location for incorporating cholesterol ester into HDL particles. (Journal of the Royal Society Interface, 6:863-871, 2009).

Maturation of HDL

Movies

Nanodisc Assembly with Lipids Initially in a Random Configuration (mpg, 4.3M)
Coarse-grained molecular dynamics simulation of nanodisc assembly with two semi-circular membrane scaffold proteins around which are randomly scatttered lipids. Results of the 10 &mus simulation show that the lipids form micelles very quickly some already aggregated to proteins. The fusion of these lipid micelles draws the two protein strands together eventually forming a single lipoprotein particle. This hydrophobic effect driven aggregation of lipids and proteins is followed by a much slower protein tertiary structure rearrangement. Eventually the protein strands rearrange themselves to form a double-belt nanodisc.

Nanodisc Disassembly with 100 Cholates (mpg, 2.3M)
Coarse-grained molecular dynamics simulation of the effect of the addition of 100 cholate molecules to a preformed nanodisc. The cholate molecules at this concentration aggregate with the nanodisc and insert themselves primarily at the interface of the lipid and scaffold proteins. This aggregation results in significant fluctuations in the nanodisc itself however in the simulations the scaffold protein remains encircling the lipid bilayer.

Publications

Publications Database

Maturation of high-density lipoproteins. Amy Y. Shih, Stephen G. Sligar, and Klaus Schulten. Journal of the Royal Society Interface, 6:863-871, 2009.

Molecular models need to be tested: the case of a solar flares discoidal HDL model. Amy Y. Shih, Stephen G. Sligar, and Klaus Schulten. Biophysical Journal, 94:L87-L89, 2008.

Molecular modeling of the structural properties and formation of high-density lipoprotein particles. Amy Y. Shih, Peter L. Freddolino, Anton Arkhipov, Stephen G. Sligar, and Klaus Schulten. In Scott Feller, editor, Current Topics in Membranes: Computational Modeling of Membrane Bilayers, chapter 11, pp. 313-342. Elsevier, 2008.

Application of residue-based and shape-based coarse graining to biomolecular simulations. Peter L. Freddolino, Anton Arkhipov, Amy Y. Shih, Ying Yin, Zhongzhou Chen, and Klaus Schulten. In Gregory A. Voth, editor, Coarse-Graining of Condensed Phase and Biomolecular Systems, chapter 20, pp. 299-315. Chapman and Hall/CRC Press, Taylor and Francis Group, 2008.

Computer modeling in biotechnology, a partner in development. Aleksei Aksimentiev, Robert Brunner, Jordi Cohen, Jeffrey Comer, Eduardo Cruz-Chu, David Hardy, Aruna Rajan, Amy Shih, Grigori Sigalov, Ying Yin, and Klaus Schulten. In Protocols in Nanostructure Design, Methods in Molecular Biology, pp. 181-234. Humana Press, 2008.

Disassembly of nanodiscs with cholate. Amy Y. Shih, Peter L. Freddolino, Stephen G. Sligar, and Klaus Schulten. Nano Letters, 7:1692-1696, 2007.

Assembly of lipids and proteins into lipoprotein particles. Amy Y. Shih, Anton Arkhipov, Peter L. Freddolino, Stephen G. Sligar, and Klaus Schulten. Journal of Physical Chemistry B, 111:11095-11104, 2007.

Assembly of lipoprotein particles revealed by coarse-grained molecular dynamics simulations. Amy Y. Shih, Peter L. Freddolino, Anton Arkhipov, and Klaus Schulten. Journal of Structural Biology, 157:579-592, 2007.

Coarse grained protein-lipid model with application to lipoprotein particles. Amy Y. Shih, Anton Arkhipov, Peter L. Freddolino, and Klaus Schulten. Journal of Physical Chemistry B, 110:3674-3684, 2006.

The role of molecular modeling in bionanotechnology. Deyu Lu, Aleksei Aksimentiev, Amy Y. Shih, Eduardo Cruz-Chu, Peter L. Freddolino, Anton Arkhipov, and Klaus Schulten. Physical Biology, 3:S40-S53, 2006.

Molecular dynamics simulations of discoidal bilayers assembled from truncated human lipoproteins. Amy Y. Shih, Ilia G. Denisov, James C. Phillips, Stephen G. Sligar, and Klaus Schulten. Biophysical Journal, 88:548-556, 2005.

Predicting the structure of apolipoprotein A-I in reconstituted high density lipoprotein disks. James C. Phillips, Willy Wriggers, Zhigang Li, Ana Jonas, and Klaus Schulten. Biophysical Journal, 73:2337-2346, 1997.

Investigators

Collaborator

Related TCB Group Projects

Page created and maintained by Amy Shih.
Last updated on August 30, 2007