Lipid Luminations: Translating Physical Studies of Lipoproteins into Clinical Lipidology

Physical chemistry is, in part, the study of macroscopic and particulate phenomena in terms of laws and concepts of physics, and it can include the physical concepts determining motion, energy, force, time, thermodynamics, light and equilibrium. Physical chemistry, in contrast to chemical physics, is a macroscopic science, because its concepts are relevant to bulk scales rather than molecular and/or atomic scales. Physical methods are applicable to studies of lipids and lipoproteins, with major areas of application including reaction kinetics, surface chemistry of lipoproteins and membranes, equilibria and thermodynamics, thermotropic phase transitions, and colligative properties. The methods of physical chemistry have informed our understanding of lipoprotein structure, apolipoprotein conformation, the dynamics of lipid exchange and transfer, and the configuration of lipids and proteins on lipoprotein surfaces. Moreover, the tools of physical chemistry have influenced the development of diagnostics for clinical lipidologists, and they have helped uncover the mechanistic relationships between lipoproteins, diabetes and atherogenesis for the academic lipidologist. Nearly every physical method has been applied to the study of lipoproteins, and those that have had the greatest impact on the field of lipoprotein pathobiology are reviewed here.

The Ultracentrifuge: The ultracentrifuge played a central role in clinical lipidology. According to their names, high-, low- and very-low-density lipoproteins (HDL, LDL and VLDL, respectively) have different densities and can be analyzed on the basis of density. W. Virgil Brown has called John W. Gofman the father of modern Clinical Lipidology. In a landmark 10-year study, Gofman and colleagues showed that men who developed ischemic heart disease had lower HDL_² and HDL₃, combined with higher levels of LDL, intermediate density lipoprotein (IDL) and small VLDL.¹ A 29-year follow-up to this study² showed that total incident coronary heart disease (CHD) was inversely related to HDL₂- and HDL₃-mass and directly related to LDLmass, IDL-mass and small and large VLDLmass concentrations.

Other technologies have been based on differential lipoprotein densities. The vertical spin method has been used to profile lipoproteins in a way that identifies various dyslipidemic states.³ Furthermore, recognizing that lipoproteins could be distinguished on the basis of differing sizes and densities, others developed preparative ultracentrifugation for large-scale isolation of individual lipoprotein classes.² This key technology was essential to defining the composition and properties of lipoproteins and their interactions with plasma lipidtransfer, lipolytic and acyltransferase proteins, as well as cell surface receptors and lipid transporters in the context of normal and pathological states.

Electrophoresis: Early studies showed that plasma proteins could be separated on the basis of charge by electrophoretic methods.⁴ With the discovery that albuminated buffers improved resolution,⁵ paper electrophoresis emerged as a key tool for phenotyping hyperlipoproteinemia.⁶ According to their mobilities with respect to α and ß globulins, HDL, LDL and VLDL were respectively α−, ß−, and pre-ß- lipoproteins. Gradient gel electrophoresis under non-denaturing conditions, which separates according to size, is another tool that identified small, dense LDL as a CVD risk factor,⁷ and subdivided HDL into two HDL₂ and three HDL₃ subfractions.⁸

Circular Dichroic (CD) Spectroscopy: CD spectroscopy reveals the chirality of molecules by measuring the difference in the absorption of left- and rightcircularly polarized light. An algorithm based on the CD spectra of a polypeptide containing a known α-helical content⁹ permitted laboratory lipidologists to conclude that HDL contained mainly α-helical structures,^10,11 whereas LDL comprised mostly β-sheet structures.^12,13 Subsequently, the extant theory of lipid-associating amphipathic helical apolipoproteins was based on modelbuilding and the α-helical content of just a few apolipoproteins revealed by their CD spectra.¹ Ultimately, all apolipoproteins in the gene family of exchangeable apolipoproteins were found to be rich in amphipathic α-helices, and synthetic apolipopeptides based on the amphipathic helical model were shown to be physiologically active.^15-17

Differential Scanning Calorimetry (DSC): DSC reveals thermotropic transitions (e.g., melting) and the energy quantity required to complete the transition. For example, DSC reveals that water melts at 0° C with an attendant heat of melting = 80 cal/g. According to the DSC method, the melting behavior of cholesteryl esters (CE) in LDL and in arterial atherosclerotic lesions are similar, exhibiting a transition from liquid crystal to an isotropic liquid phase.¹⁸ Moreover, increasing the triglyceride (TG) content of LDL reduces its CE melting temperature and its binding to the LDL receptor.¹⁹ According to cryoelectron microscopy, increasing the TG content of LDL also changes its structure from ellipsoidal to spherical.^20-22 The DSC method provided one of the most insightful observations showing that HDL is a uniquely labile lipoprotein. During the heating of HDL in a DSC experiment, its lipid components fuse into a large apo A-II-rich particle, while apo A-I is released in a lipid-free state.²³ These studies indicated that HDL resides in a kinetic trap from which labile apo A-I escapes to the surrounding aqueous phase, lipid-free.²³ This apo A-I lability also is seen when HDL interacts with lipid transfer proteins, lecithin-cholesterol acyltransferase, hepatic lipase, and streptococcal serum opacity factor,^24-28 all of which disrupt HDL structure and release lipid-free apo A-I.

Mass Spectrophotometric Analyses:Three powerful structural approaches are based on mass spectrometry (MS). They are apolipoprotein configuration analysis by cross-linking (CL) and MS; hydrogen-deuterium (H-D) exchange as assessed by MS; and proteomic analysis by MS. All three methods have been applied to studies of the structure of HDL and reconstituted HDL [(r)HDL]. MS determines the mass of molecules or fragments on the basis of their mass-tocharge ratios.

In the CL-MS application, the ε-amino groups of lysines in rHDL proteins and in HDL containing mainly apo A-I are crosslinked. Subsequent trypsinolysis releases cross-linked peptides that are analyzed by MS. Given that the molecular masses of the tryptic peptides of apo A-I and the cross-linker are known, the site of crosslinking is determined by matching the theoretical and actual molecular masses. With knowledge of the main CL, the configurations of apo A-I on the surface of rHDL and HDL can be determined.²⁹ According to the CL-MS method, HDL of different sizes and containing no apo A-II had the same CL profiles, suggesting that strong protein-protein interactions on the surface of HDL maintain this constant structure.³⁰

H-D exchange complements CD methods by identifying specific amino acid residues that are nonexchangeable because they are hidden within α-helices. Briefly, the protein protons are exchanged for deuterons by dilution into D₂O. The pH is reduced to quench additional exchange, and the protein is trypsinized. The liberated peptides are separated and their molecular masses determined. The additional molecular mass of nonexchanged deuterons and the known molecular masses of the tryptic peptides reveal which residues are nonexchangeable because they reside in α-helices.

The α-helix of lipid-free apo A-I as assessed by H-D and CD analysis agreed well; moreover, the location of the α-helices for the most part supported what was predicted by computational methods.³¹ However, when apo A-I in rHDL was similarly analyzed, only a few amino acids at each end of the protein contained exchangeable deuterons so that when associated with lipid, apo A-I is ~95% α-helical.³² H-D analysis also has revealed differences in the structural dynamics of apo A-I point variants. Apo A-I Iowa (G26R) is associated with systemic amyloidosis; apo A-I Milano (R173C) is associated with hypoalphalipoproteinemia and, apparently, cardioprotection. According to H-D exchange studies, the helical structures of both are less stable than those of native apo A-I.³³ Although they will not replace X-ray crystallography for structural analysis, H-D exchange studies have the advantage of giving information about proteins in aqueous solutions.

Proteomics, which is more a discovery-driven platform than a hypothesis-driven platform, is the large-scale study of structure, occurrence and function of proteins, especially by MS. According to shotgun proteomics, HDL contains multiple complement-regulatory proteins, serpins with serine-type endopeptidase inhibitor activity, and acute-phase response proteins, which suggests that HDL may play a role in regulating the complement system by protecting tissue from proteolysis and inflammation.³⁴ With improved methodologies, other studies have identified numerous HDL proteins associated with complement function, coagulation, neurogenesis, development, insulin signaling and collagen binding.³⁵ MS proteomics has advantages and disadvantages. Because of the sensitivity of MS, this approach can detect very low levels of proteins with a heretofore unsuspected role in lipoprotein metabolism. However, the importance of low copy-number proteins is uncertain, and it will be important to determine quantitative stoichiometric relationships between these low-level species in a way that leads to mechanistic insights.

The Future: Physical methods will continue to play an important role in identifying structure-function relationships among lipoproteins and in defining atherogenic vs. atheroprotective processes. As current methods are refined, new physical approaches will be applied to studies of lipoproteins.

Disclosure statement: Dr. Pownall has no disclosures to report. Dr. Gotto has received consulting fees from Aegerion Pharmaceuticals, Arisaph Pharmaceuticals, DuPont, Janssen, Kowa Pharmaceuticals America, Merck & Co., Pfizer Inc., Roche, and Vatera Capital.