| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Metabolic Laboratory, Department of Clinical Chemistry, VU University Medical Center, Amsterdam, The Netherlands
aAddress correspondence to this author at: P.O. Box 7057, 1007 MB Amsterdam, The Netherlands. Fax 31-2044-43895; e-mail t.teerlink{at}vumc.nl.
To the Editor:
The recent article by Ensign et al. (1), reporting on a disappointingly low agreement among 4 methods to assess LDL particle size and phenotypic classification, casts doubt on the utility of these techniques in clinical practice. A similar poor correlation (r = 0.39) between LDL size measurements by nuclear magnetic resonance (NMR) and gradient gel electrophoresis (GGE) has been reported by Witte et al. in a study with 324 individuals (2). Ensign et al.(1) advocate a standardization program to reduce the lack of concordance between methods. In addition to standardization problems, however, there is another reason for the observed discrepancy that is not addressed in the Ensign paper and accompanying editorial; the assumption that LDL size is adequately described by a single variable, diameter.
We would like to point out that this assumption may not be valid for LDL particle size. The situation is akin to the quantification of obesity, for which several measures are commonly used, such as body mass index, waist circumference, and waist-to-hip ratio. Although these measures are significantly correlated, they cannot be used interchangeably. If humans were spherical objects, the agreement between these measures of obesity would be perfect. But this is obviously not the case, because human beings come in all sorts of shapes. We think that to a certain extent this is also true for LDL particles. Although a spherical shape may seem intuitively right, several experimental approaches have not confirmed this characteristic of LDL particles but instead suggest that they are nonspherical. Nonspherical shape is not unique to LDL particles, but also occurs within the HDL class of lipoproteins. Although mature HDL particles are spherical, it is well accepted that nascent HDL particles are discoidal.
The fact that each LDL particle contains a single copy of apolipoprotein B-100, almost fully accounting for the protein content of LDL, allows straightforward calculation of average LDL particle volume from its chemical composition (3). Assuming a spherical particle shape, average diameter can then be calculated by simple arithmetic. In a study including 160 individuals, we observed that LDL diameters measured by high-performance gel-filtration chromatography correlated poorly (r = 0.60) with calculated diameters (3). This discrepancy could be reconciled by assuming that LDL particles are discoidal, with a mean diameter of 20.9 nm (range 19.6–21.6 nm) and a mean height of 12.1 nm (range 10.5–13.9 nm) (3). These values are in striking agreement with dimensions obtained by cryoelectron microscopy, which is a technique allowing visualization of single LDL particles from different angles (4). Furthermore, data obtained by crystallographic analysis are also indicative of a pseudocylindrical or discoidal particle shape (5).
An important consequence of the discoidal LDL model is that techniques that are currently used to assess LDL size are not equivalent. Techniques such as dynamic light scattering and NMR actually measure particle volume, from which diameter is calculated, assuming a spherical particle shape. This principle also applies to density gradient ultracentrifugation, because density is inversely proportional to particle volume. In contrast, measurements by high-performance gel-filtration chromatography and electrophoretic techniques such as GGE and tube gel electrophoresis probably reflect particle diameter more closely. A striking feature of the discoidal particle model is that diameter and height are not significantly correlated (3). Consequently, a flat particle of large diameter can have the same volume as a thick particle of small diameter, resulting in similar NMR readings but widely differing GGE results.
In conclusion, we suggest that the lack of agreement between various methods to assess LDL particle size or phenotype is partly due to the fact that LDL particles are not spherical and therefore their size cannot be described by a single variable. In addition to partly resolving the perceived discrepancy between LDL size measurements, the discoidal particle concept puts a new perspective on the notion that small, dense LDL are more atherogenic than their large counterparts. Clinically, discoidal particle shape raises the question of what measure of LDL size or shape—volume, diameter, height, or aspect ratio—is most closely related to cardiovascular disease, an evaluation process reminiscent of the ongoing discussion of whether body mass index or waist circumference is a better predictor of cardiovascular outcome. Unfortunately, in contrast to these anthropometric measures, which are readily performed on large numbers of individuals, measurement of LDL dimensions is not easily performed on a large scale. Nevertheless, we do think that it would shed more light on the relative atherogenicity of specific LDL subclasses.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
