THE LIPOPROTEINS

 THE LIPOPROTEINS

Introduction

The French physician-scientist Michel

Macheboeuf is acknowledged as the

father of plasma lipoproteins. His

seminal 1928 discovery fi ts the adage

that science usually precedes tech-

nology, in this case by several de-

cades. In his doctoral thesis, Recher-

ches sur les lipides, les stérols et les

protéides du sérum et du plasma san-

guinis, Macheboeuf described horse

serum lipoproteins, demonstrating

the association of lipids and proteins,

ìlipido-protéidiquesî, in plasma.1 It is

now recognized that plasma lipids

are transported by lipoproteins, which

are defi ned by the densities at which

they are isolated, that is, as the

high-, low-, intermediate-, and very-

low-density lipoproteins (HDLs, LDLs,

IDLs, and VLDLs, respectively); chy-

lomicrons, which are intestinally de-

rived, are composed mainly of dietary

lipids and small amounts of protein.

HDL appears in two subclasses, HDL2

and HDL3. Through a simple veni-

puncture, plasma lipoprotein levels,

which are arguably among the most

important risk factors for coronary

artery disease, provide clues about

the etiology of lipid disorders and

about their most prominent patho-

logic sequela, atherosclerosis. From

this window on life, a host of infor-

mative analyses has emerged. Corr-

elations between coronary artery

disease and the properties, composi-

tions, and plasma concentrations of

various analytes—lipids as well as

lipoproteins—have revealed mecha-

nisms that ultimately aided diagnosis

and provided new targets for pharma-

cologic management of dyslipidemia

and atherosclerosis.

LIPOPROTEIN PROPERTIES

The plasma lipoproteins are composed

of neutral lipids, polar lipids, and spe-

cialized proteins called apolipopro-

teins (apos). The major neutral lipids

are cholesteryl esters (CEs) and triglyc-

erides (TGs); the polar lipids are phos-

phatidylcholine, sphingomyelin, and

free cholesterol with small amounts of

phosphatidylethanolamine and traces

of other phospholipids (PLs). The com-

positions of the lipoproteins determine

their size and structures (Table 1-1).

Lipoprotein size and density are a di-

rect function of neutral lipid content,

with the largest lipoprotein particles

being the least dense and having the

highest ratio of neutral to polar lipids.

Surface charge as revealed by agarose

gel electrophoresis varies among lipo-

proteins according to the amount of

charged lipids and the conformations

of their apos. Lipoproteins have been

isolated according to size, charge, and

density by size exclusion chromatogra-

phy, ion exchange chromatography,

and ultracentrifugation, respectively,

with the latter technique being used for

preparative isolation.

The organization of the remainder of

this chapter proceeds from our view that

much of the currently unanswered lipid

and lipoprotein pathologies proceed from

dysregulated fatty acid metabolism,

which is important in numerous diseases

of public interest, including obesity,

diabetes, and atherosclerosis. Although many of the downstream effects of dysregulated fatty acid

metabolism are the targets of therapy, it remains important

to identify therapeutic modalities that might address un-

derlying causes.

Nonesterifi ed Fatty Acids

Some plasma nonesterifi ed fatty acids (NEFAs) are de-

rived from VLDL- and chylomicron-TG hydrolysis by li-

poprotein lipase (LPL), which is attached to the capillary

endothelium via proteoglycans. A small fraction of the

released NEFAs is released into the plasma, particularly

in the postprandial state, when rates of chylomicron-TG

hydrolysis are high. Most of the liberated NEFAs trans-

migrate the endothelium to adipose tissue (AT), where

they diffuse across the adipocyte plasma membrane and

esterify glycerol-3-phosphate via the Kennedy pathway

for glycerolipid synthesis. The TGs so formed associate

in TG droplets that are visible under light microscopy

(Fig. 1-1). The surfaces of the droplets are surrounded by

a mixed monomolecular layer of PLs and by specialized

proteins that are essential to normal TG storage and

hydrolysis, including perilipin, Comparative Gene

Identifi cation–58 (CGI–58), and microsomal transfer pro

tein (MTP)-B, a splicing variant of the canonical MTP-A

that is found in liver and is associated with protein disul-

fi de isomerase. Whereas MTB-B mediates fusion of small

fat droplets into larger ones,2 it has no direct effect on

lipolysis. In contrast, perilipin and CGI–58 are important

in the regulation of adipocyte lipolysis, and under fasting

conditions, AT TG is a major source of plasma NEFAs.

In the absence of stimulation, perilipin on the sur-

face of fat droplets blocks hormone-sensitive lipase (HSL)–mediated hydrolysis. With β-adrenergic recep-

tor activation, protein kinase A hyperphosphorylates

perilipin, thereby rapidly altering its conformation in a

way that exposes the TG to HSL. Ablation of the perili-

pin gene in mice is antiadipogenic and illustrates its

importance in energy distribution and storage.3 In the

absence of perilipin, β-oxidation is increased and he-

patic glucose production is reduced, whereas glucose

tolerance and peripheral tissue insulin resistance are

normal.4 According to gene array analyses, these effects

are associated with coordinated up-regulation of oxida-

tive pathways and down-regulation of lipid biosynthe-

sis.5 Adipose tissue expresses another lipase, AT TG

lipase,6 which is also associated with plasma NEFAs

and TGs in patients with type 2 diabetes.7 CGI-58, a

member of the subfamily hydrolase fold enzymes, acti-

vates (20-fold) adipose TG lipase, and variants of the

human CGI-58 are associated with Chanarin-Dorfman

syndrome, a disease characterized by ectopic fat depo-

sition.8 Like HSL, adipose TG lipase is essential to

normal lipid metabolism in adipocytes. The combined

activities of adipose TG lipase and HSL account for

more than 95% of the TG hydrolase activity present in

murine white AT. CGI-58 binds to perilipin A–coated

lipid droplets in a manner that is dependent on the

metabolic status of the adipocyte and the activity of

cAMP-dependent protein kinase.9

The Apolipoproteins

The distribution of the apos among plasma lipopro-

teins (see Table 1-1) determines some of their meta-

bolic effects. The apos, which can be classifi ed as

soluble and insoluble, are important directors of lipo-

protein metabolism. The soluble apos belong to the

same gene family in which the terminal exon IV

codes for the region of the apo that gives it its distin-

guishing biologic activities. All soluble apos are ex-

changeable and contain extended regions of am-

phipathic helices that mediate binding to lipid

surfaces. ApoA-I and apoC-I10,11 are activators of cho-

lesterol esterifi cation via lecithin:cholesterol acyl-

transferase (LCAT); apoC-II stimulates LPL-mediated

hydrolysis of chylomicrons and VLDL12,13; apoE is the

ligand for the cellular uptake of IDL and chylomicron

remnants.14–16 Mechanistic links between other apos

and lipid metabolism are more subtle but likely pres-

ent in ways that remain to be determined. Plasma

apoC-III correlates with plasma TG levels,17,18 whereas

apoA-V, which occurs at low levels in human plasma,

appears to be antilipemic.19 ApoB-100, an approxi-

mately 550-kDa nonexchangeable protein, is a major

protein of VLDL, IDL, and LDL, and it contains the

ligands for the cellular uptake of LDL via its receptor.

Chylomicrons contain a truncated form of apoB, that

is, apoB-48, which is a product of a novel mRNA edit-

ing mechanism wherein an amino acid codon is con-

verted to a stop codon, giving an expression product

that lacks the LDL receptor–binding domain.20,21

Lipoprotein (a) [Lp (a)] is another large lipoprotein, in

which the major protein, apo(a), associates with apoB

via a disulfi de bridge.22

LIPOPROTEIN PRODUCTION

Triglyceride-Rich Lipoproteins

The secreted lipoproteins VLDL and chylomicrons are

assembled and secreted by hepatocytes and enterocytes

in the liver and intestine, respectively. Their respective

assembly is driven by the TG synthesis from endoge-

nous and exogenous, that is, dietary, fatty acids. Thus,

an important determinant of fasting plasma TG concen-

tration is the plasma NEFA concentration that is avail-

able for hepatic uptake. Some of the NEFAs that are

liberated by the hydrolysis of chylomicrons following

an oral fat load are hepatically removed and used for

VLDL-TG production and secretion. Insulin resistance

in AT that impairs fatty acid storage also raises plasma

NEFAs, which are used for VLDL production. This

mechanism accounts in part for the association of dia-

betes and other insulin-resistant states with fasting

hypertriglyceridemia (HTG) and enhanced postpran-

dial lipemia. HTG is exacerbated by increased hepatic

lipase activity, which diverts TG-derived NEFAs to the

liver, where they cycle back to VLDL-TG. Although the

identifi cation of the mechanisms for protein folding is

usually diffi cult, identifi cation of the mechanism for

VLDL assembly, which involves protein folding and

the addition of specifi c amounts of polar and nonpolar

lipids, has been much more challenging. Nevertheless,

morphologic and subcellular fractionation studies of

hepatocytes23 have provided some support for a two-

step model of VLDL assembly. The fi rst step, partial

lipidation of apoB with TG, CE, and PL during its trans-

lation and translocation to the lumen of the rough en-

doplasmic reticulum by MTP-A, yields a pre-VLDL

that remains weakly associated with the endoplasmic

reticulum membrane. The pre-VLDL interacts with a

TG-rich particle from the smooth endoplasmic reticu-

lum. The molecular details for this step are not known

but may involve chaperones. In some hepatic cells,

inadequate lipidation leads to degradation of early

forms of VLDL via the ubiquitination-proteosome path-

way.24–26 Hepatic MTP-A is associated with protein di-

sulfi de isomerase, the endoplasmic reticulum retention

sequence of which keeps MTP within the endoplasmic

reticulum.27 Recent studies of a splicing variant of

MTP-B suggest that it is a fusogen2 and that in hepatic

cells it could mediate the fusion of pre-VLDL with TG-

rich particles. This remains to be demonstrated. Stud-

ies of chylomicron assembly have been sparse, but it is

presumed without much evidence to be similar to that

of VLDL. As discussed later, the HTG resulting from

insulin resistance in AT contributes to the complex

phenotype that presents in the metabolic syndrome

and obesity-linked diabetes.

Intermediate-Density Lipoproteins

and Low-Density Lipoproteins

Following their entry into the plasma compartment,

VLDLs are modifi ed by LPL. VLDL, which is the major

carrier of endogenous TG, contains apos B-100, E, C-I,

C-II, and C-III (see Table 1-1), which are segregated during lipolysis. Hahn fi rst reported that heparin administra-

tion released a factor that caused the clearing of human

plasma.28 This observation supported the extant view

that LPL associates with capillary proteoglycans and

that the main site for the uptake of the fatty acids re-

leased by LPL is peripheral tissues that are perfused via

the capillary bed. In a classic citation, Korn described

the properties of the clearing factor and determined that

it was a lipoprotein lipase.29 Havel and LaRosa further

established the importance of LPL in lipoprotein me-

tabolism by showing that LPL activity was stimulated

by apoC-II.30,31 Hydrolysis of TG by LPL converts VLDLs

to IDLs and chylomicrons to chylomicron remnants. As

expected, apoC-II and LPL defi ciency are associated

with severe HTG.32 Independent of mutations in apoC-II

and LPL, moderate HTG is associated with type 2 dia-

betes and atherosclerosis (see later discussion). Interest-

ingly, as VLDL is hydrolyzed by LPL, the C apos, in-

cluding apoC-II, transfer to HDL and lipolytic activity

via LPL is arrested, leaving IDL, which is not an LPL

substrate. However, in the liver, hepatic lipase, which

does not require apoC-II, continues the hydrolysis of

IDL to the mature apoB-100–containing product, LDL.

During this step, the particle loses most of its apoE.

Hepatic lipase remodels HDL through the hydrolysis of

PLs and TGs, an activity that is more profound in the

postprandial state.33

High-Density Lipoproteins

For a number of reasons, models for the structure, pro-

duction, remodeling, and catabolism of HDLs have been

more diffi cult to identify than those for the apoB-

containing lipoproteins. HDLs are small and hetero-

geneous with respect to size and composition (see

Table 1-1), so many conventional methods such as cryo-

electron microscopy, x-ray crystallography, and nuclear

magnetic resonance have limited value. Unlike the apoB-

containing lipoproteins, all the components of HDLs are

exchangeable. Thus, traditional kinetic methods cannot

be used to study their turnover. Lastly, although several

sources of HDL have been identifi ed on the basis of cell

studies, the quantitative importance of these sources in

human HDL metabolism is not known except in cases of

a natural ablation of a gene coding one of the proteins

that forms HDL.

There is evidence that some HDLs are secreted,

whereas others are a product of lipolysis in the plasma

compartment. The human hepatic cell line, HepG2,

secretes particles that have the properties of HDL and

contain its major apos.34,35 The perfusate from rat liver

contains small particles that have been described as

nascent HDL.36 Studies by Patsch and Tall have shown

that some HDL subclasses are formed by the lipolysis

of TG-rich lipoproteins.37,38 On the other hand, more

direct studies of HDL production by human hepato-

cytes are needed to better understand this important

process and its regulation. More recent studies have

focused on the role of HDL production and remodeling

in reverse cholesterol transport (RCT).

Unlike liver tissue, extrahepatic tissue can synthe-

size but cannot degrade cholesterol. Thus, cholesterol

accumulation in macrophages, a key cell type in ath-

erogenesis, produces a lipotoxic, pathologic state, un-

less there is a mechanism for its disposal; that mecha-

nism is RCT, and HDL is its central player.39,40 Within

the context of cardiovascular disease (CVD), RCT com-

prises three steps: cholesterol effl ux from monocyte-

derived macrophages within the arterial wall, esterifi -

cation and interaction with lipid transfer proteins in

the plasma compartment, and selective hepatic uptake

by HDL-CE by its receptor (Fig. 1-2). There are at least

four mechanisms for cholesterol effl ux:

• One mechanism is mediated by the microsolubiliza-

tion of membrane lipids by apoA-I via its interac-

tions with the ATP-binding cassette A1 (ABCA1)

transporter, which triggers unidirectional release of

cholesterol and PL, forming nascent HDL.41–43 Tang-

ier disease is a severe manifestation of an ABCA1

mutation in which plasma HDL-C levels are close to

nil and cellular transfer of cholesterol to lipid-free

apos is impaired.44,45

• ABCG1/G4 mediates effl ux to HDL2 and HDL3 but not

to lipid-free apoA-I.46–48 Effl ux increases with HDL-PL

content. ABCG1 is highly expressed in macrophages46–48

and might mediate effl ux from macrophage–foam

cells to HDL. ABCG1 might be the mechanistic link

between high HDL-C and low risk of CVD.

• A third mechanism is spontaneous cholesterol

desorption from the plasma membrane into the sur-

rounding aqueous phase, where it associates with

HDL. This process is driven by a cholesterol concen-

tration gradient from high (donor) to low (acceptor);

high relative levels of acceptor-sphingomyelin, which

is highly cholesterophilic, increase effl ux.49–51

FIGURE 1-2 Cholesterol transfers from macrophages to high-density lipoprotein

(HDL) via ATP-binding cassette transporter A1 (ABCA1), ABCG1/4, spontaneous

transfer, and Scavenger receptor class B type I (SR-BI) (1a–d), and is converted to

Cholesteryl ester (CE) via lecithin:cholesterol acyltransferase (LCAT) (2). With further

cholesterol accretion and esterifi cation, HDL grows to its mature forms from which

lipids are removed by hepatic SR-BI receptors (3). •Hepatic Scavenger receptor class B type I (SR-BI),

which selectively removes HDL-CE, HDL-TG,

and HDL-PL,52 also mediates cholesterol effl ux to

HDL, a process that is dose dependent with respect

to acceptor-PL content; acceptor-PL enrichment/

depletion increases/decreases effl ux via SR-BI; effl ux

is enhanced by addition of phosphatidylcholine53 and

by replacing acceptor-phosphatidylcholine with more

cholesterophilic PLs such as sphingomyelin.54–57

Lecithin:Cholesterol Acyltransferase

After cholesterol transfer to early forms of HDL, the par-

ticle undergoes a series of remodeling reactions involving

lipid transfer proteins and cholesterol esterifi cation. Al-

though Sperry identifi ed a plasma cholesterol-esterifying

activity in the 1930s,58 nearly three decades elapsed be-

fore studies of families with esterifi cation defi ciency re-

newed interest in this process because of the emerging

correlation between plasma cholesterol concentration

and CVD. Studies of LCAT, which catalyzes the transfer

of fatty acyl chains of phosphatidylcholine to choles-

terol, prompted Glomset to propose that HDL was the

vehicle for RCT, the transfer of cholesterol from periph-

eral tissue to the liver for disposal or recycling.59,60 LCAT

is central to RCT because it converts cholesterol to its

ester, which is not as readily transferred among mem-

branes and lipoproteins, and it converts HDL from a disc

to a sphere with a core containing mostly CEs. Additional

rounds of effl ux to HDL and esterifi cation produce the

mature form that is eventually removed by the liver. As

expected, patients with familial LCAT defi ciency have

very low plasma cholesteryl esters levels and an altered

lipoprotein profi le. The most profound effect of LCAT

defi ciency is corneal opacifi cation. In a milder form of

defi ciency—fi sh eye disease—corneal opacities occur

later in life and the reduction of plasma HDL-CE levels is

not as profound. In vitro expression of variants found in

LCAT defi ciency and fi sh eye disease has revealed that

the reduction in secretion and specifi c activity is more

severe in the former.61 Surprisingly, association of LCAT

defi ciency with CVD has not been fi rmly established,

perhaps because of the small number of patients, and the

studies of atherosclerosis in mice overexpressing LCAT

have been contradictory.62–66

Lipid Transfer Proteins

Human plasma contains two proteins—cholesteryl es-

ter transfer protein (CETP) and PL transfer protein

(PLTP)—that transfer lipids among lipoproteins. Among

the lipoproteins, the main donor–acceptor targets of

PLTP are HDLs, which PLTP remodels into large and

small particles with the concomitant dissociation of

lipid-free apoA-I from HDL.67 Studies in mice overex-

pressing PLTP suggest that PLTP is atherogenic because

it lowers plasma HDL levels.68 Indeed, some studies in

mice have shown that systemic PLTP expression cor-

relates positively with atherosclerotic lesion develop-

ment.69 Moreover, macrophage PLTP is an important

contributor to plasma PLTP activity, and its defi ciency

lowers lesion development in LDL receptor–knockout

mice on Western-type diet.70,71 However, similar

studies in LDL receptor–knockout mice suggest that

macrophage-derived PLTP is atheroprotective.72 These

fi ndings and the absence of natural PLTP mutants with

any associated pathology in humans make it diffi cult to

estimate the physiologic importance of PLTP.

In contrast, there is little ambiguity about the impor-

tance of CETP, and studies in humans with CETP defi -

ciency and in mice in which the CETP gene has been

inserted leave little doubt about its importance in lipid

metabolism. Current evidence, particularly in patients

with HTG, reveals CETP as an integrator of lipoprotein

remodeling that connects the metabolism of TG-rich

lipoproteins with those of HDL and LDL (Fig. 1-3).

Whereas PLTP transfers mainly PLs, CETP transfers

some PLs but has as its primary activity the exchange of

neutral lipids—CE and TG—between lipoproteins. In

normolipidemic subjects, CETP exchanges small

amounts of HDL-CE for VLDL-TG, thereby producing a

small increase in the TG content of HDL; effects on

VLDL are small. HTG profoundly alters the effects of

CETP on lipoprotein profi les, structure, and catabolism.

In the presence of HTG, the high VLDL levels provide a

large pool of TG for exchange with a much smaller pool

of HDL- and LDL-CE so that HDL and LDL are made

TG rich.73 According to cryoelectron microscopy,

enrichment of LDLs with TGs shifts their shape from

oval7 to spherical,74,75 reduces its binding to the fi bro-

blast LDL receptor, and lowers its stability as assessed

by enhanced PLTP-mediated release of apoA-I.76

The Special Role of Apolipoprotein A-I

in High-Density Lipoprotein Stability

and Metabolism

It has long been observed that HDL is much less stable

than other plasma lipoproteins. Early studies by Nichols77

showed that the chaotrope, guanidinium chloride, trig-

gered the release of apoA-I, but not apoA-II, TG-rich

from human HDL. More recently, Mehta and colleagues78

and Gursky79 showed that apoA-I in native HDL

FIGURE 1-3 Formation of small, triglyceride (TG)-rich high-density lipoprotein (HDL)

by the activities of cholesteryl ester transfer protein (CETP) and hepatic lipase (HL).

A, Under normolipidemic conditions, HDL is formed via multiple cycles of phospholipid

(PL) and cholesterol effl ux via ABCA1 or ABCG1 followed by lecithin:cholesterol acyl-

transferase (LCAT)–mediated esterifi cation that leads to a mixture of HDL2 and HDL3;

B, low-density lipoprotein (LDL) is formed via lipolysis of very-low-density lipoprotein

(VLDL) and intermediate-density lipoprotein (IDL) by lipoprotein lipase (LPL) and hepatic

lipase (HL), during which the C and E apolipoproteins (apos) are transferred to HDL.

C, In hypertriglyceridemia (HTG), CETP mediates the net transfer of TGs from a large pool

of VLDLs to HDLs, giving TG-rich HDLs that are hydrolyzed to small, TG-rich HDLs (HDLs). resided in a kinetic trap and that, when a mechanism for

its release was provided, a large fraction of the apoA-I

transferred to the aqueous phase, with the concomitant

fusion of the remaining apoA-II–rich species into larger

particles. The fusion product is highly stable, and treat-

ment with guanidinium chloride does not release its

remaining complement of apoA-I.80 These studies fur-

ther showed that large HDLs, including HDL2, are more

stable than small HDLs and that apoA-I and apoA-II are

equally lipophilic with respect to large HDLs. Detergent

perturbation,81 LCAT,82 CETP,83 and especially PLTP76

and serum opacity factor from Streptococcus pyo-

genes84,85 also catalyze desorption of apoA-I from HDL.

This effect is particularly important in cellular choles-

terol effl ux via ABCA1, which requires lipid-free apoA-I,

and in the terminal RCT step, uptake of HDL lipids

without apoA-I (see later discussion). Thus, the HDL

instability fi rst observed and characterized by physico-

chemical perturbations is relevant to plasma and cellu-

lar activities that alter HDL compositions in vivo.

Metabolic Syndrome

Metabolic syndrome (MetS) is a dyslipidemic state that

is associated with a cluster of risk factors. As defi ned

by the National Cholesterol Education Program Adult

Treatment Panel III (NCEP ATP III),86 a diagnosis of

metabolic syndrome can be made if three of fi ve

conditions are found. These are HTG, low plasma

HDL-C, hypertension, hyperglycemia, and a large waist

circumference. Many patients with MetS eventually

develop type 2 diabetes, with which it shares many

characteristics, including a link with obesity. There is

a growing opinion that MetS and diabetes might be bet-

ter viewed as a state of dysregulated lipid metabolism

with an attendant impaired rate of glucose disposal.87,88

The morphing of diabetes and/or MetS from glucocen-

tric to lipocentric has been driven by both clinical and

basic research and according to one model should in-

clude a cluster of abnormalities that goes beyond the

diagnostic criteria of NCEP ATP III (Table 1-2). As a

consequence, one can narrow the search for underlying

metabolic abnormalities to those that would give rise to

those shown in Table 1-2. It is crucial to acknowledge

that, in many cases, the abnormalities shown in  Table 1-2 taken one at a time are not atherogenic, but

taken together produce an atherogenic profi le for which

current therapies are inadequate.

Thus, one should search for treatments that address

the underlying cause, thereby correcting the entire

MetS cluster of abnormalities. The clustering of some

of these abnormalities, for example, HTG and low

HDL-C, has long been noted,73 and it is of interest that the

main therapeutic effect of gemfi brozil, a TG-lowering

drug, is achieved through increased HDL-C concentra-

tions, an indirect effect that is likely mediated by

CETP.89 CVD is increased in Japanese-American men

with increased HDL levels because of a variant of

CETP that does not readily exchange HDL-CE for

VLDL-TG,90 further underscoring the importance of

treating the cluster of abnormalities and not just one of

its components.

Two lifestyle practices—alcohol consumption and

exercise—are widely viewed as cardioprotective.

Regular consumption of moderate amounts of alcohol

reduces CVD mortality,91 an effect that is likely medi-

ated by increased HDL-C and HDL-PL.92 This occurs

despite the well-known alcohol-induced inhibition

of chylomicron lipolysis that leads to enhanced post-

prandial lipemia. Exercise has broad effects that ad-

dress many of the abnormalities of MetS, including

the reduction of waist circumference through weight

loss. Vigorous aerobic exercise reduces plasma TG and

postprandial lipemia while raising plasma HDL-C,

especially the more cardioprotective fraction, HDL2.

93

Thus, the identity of the mechanism that raises

HDL-C concentrations may be more important to ath-

eroprotection than the plasma HDL-C concentration.

This view is supported by studies in mice showing

that increasing SR-BI expression lowers plasma

HDL-C levels while increasing RCT. Hepatic SR-BI

overexpression decreases plasma HDL-C,94–96 in-

creases HDL-CE clearance,95–97 and increases biliary

cholesterol and its transport into bile.94,97,98 Mice

with ablated or attenuated hepatic SR-BI expression

exhibit elevated plasma HDL-C and reduced selective

HDL-CE clearance.99

Anthropomorphic Determinants

of Metabolic Syndrome

Most of the MetS abnormalities shown in Table 1-2

have been cited as having a genetic component, and

searches for one or more underlying atherogenic

genes have been reported. On the other hand, one

could postulate, based on the number of analytes in-

volved (see Table 1-2), that MetS has a highly poly-

genic origin. A third hypothetical view, presented

here, is that one or perhaps a few master genes that

control NEFA metabolism give rise to the MetS phe-

notype. One of the abnormalities that could explain

the occurrence of the remainder of the abnormalities

is dysregulated NEFA metabolism in AT that induces

systemic hyperNEFAemia. Given that circulating

plasma NEFAs can rapidly enter and exit cells, plasma

hyperNEFAemia could be diagnostic for systemic hy-

perNEFAemia, in which all tissue sites are challenged

by a NEFA overload. 

TABLE 1-2 Characteristics of the Metabolic Syndrome

High plasma NEFA Insulin resistance

Hypertriglyceridemia* Hyperinsulinemia

Profound postprandial lipemia Hyperglycemia*

Low HDL Pear–apple anthropomorphism*†

No HDL2 Low lipoprotein lipase

Small dense LDL High hepatic lipase

Elevated CET activity Hypertension*

*Metabolic syndrome according to National Cholesterol Education Program Adult

Treatment Panel III.86

†Large waist circumference.

CET, cholesteryl ester transfer HDL, high-density lipoprotein; LDL, low-density

lipoprotein; NEFA, nonesterifi ed fatty acid. In liver, the hyperNEFAemia provides the substrate

needed for TG overproduction and VLDL hyperse-

cretion that leads to HTG.

• The increased pool of VLDL is a source of additional

TG that exchanges for LDL-CE and HDL-CE via

CETP, thereby lowering LDL cholesterol and HDL

cholesterol by reducing the CE content of LDL and

HDL and forming TG-rich LDL and HDL.73

• The TG-rich LDL and HDL are substrates for hepatic

lipase, which removes some of the TG via lipolysis,

leaving small dense LDL and the smaller, less athe-

roprotective HDL3 at the expense of HDL2; both LDL

and HDL are still relatively TG rich.

• Postprandial lipemia is more profound in MetS

because VLDL-TG is a competitive inhibitor for

chylomicron hydrolysis, and NEFAs are a product

inhibitor of LPL activity.

100

• In skeletal muscle, hyperNEFAemia is lipotoxic and

its presence is associated with increased myocyte TG

and reduced glucose disposal, which triggers insulin

secretion that gives rise to hyperinsulinemia.101,102

According to this model, interventions that would

address the entire MetS risk cluster would have to im-

prove fatty acid storage in AT. The pear-to-apple an-

thropomorphism that is associated with MetS and re-

vealed as increased waist circumference provides a

clue to underlying causes and possible therapies.

Although obesity is associated with diabetes, insulin

resistance, and CVD, this effect is depot specifi c.103–108

Early studies showed a higher risk of diabetes and CVD

in patients with upper body (truncal, central, abdo-

minal, or visceral) obesity than in those with lower

body (femoral-gluteal or noncentral) obesity.109–111

Waist-to-hip circumference ratio is associated with hy-

perinsulinemia, impaired glucose tolerance, type 2 dia-

betes, and HTG112–116 and attendant CVD.117–122 In non-

diabetic, middle-aged men, subcutaneous abdominal

fat mass is a better predictor of insulin sensitivity than

intraperitoneal fat mass; the sum of truncal skinfold

thickness also better predicts insulin resistance than

intraperitoneal, retroperitoneal, or peripheral subcuta-

neous fat.123 Importantly, posterior subcutaneous ab-

dominal fat mass better predicts insulin sensitivity

than anterior subcutaneous abdominal fat mass.125 De-

spite some evidence that central fat contains the under-

lying cause of MetS, there is other evidence that the

dysregulated energy metabolism and attendant lipoat-

rophy in noncentral fat depots, particularly the

femoral-gluteal depots, is mechanistically linked to the

MetS cluster (see Table 1-2).126–128 Thiazolidinediones,

which target peroxisome-proliferator activated recep-

tors, apparently work through depot-specifi c effects

that improve global insulin sensitivity despite weight

gain.129–132

High-Density Lipoprotein Therapy

Although the current RCT model is essentially a refi ne-

ment of that originally described by Glomset,60 new

transporters and receptors that participate in RCT have

been identifi ed. These include cholesterol effl ux via

spontaneous transfer or via ABCG1, which depends on

PLs. PLs are the essential cholesterophilic component

of all lipoproteins, including HDL, and increased

HDL-PC would be expected to increase cholesterol

effl ux in a way that is therapeutic.

• Reconstituted HDLs are highly cholesterophilic133–136

and superior LCAT substrates.137–139

• Reconstituted HDL infusion into healthy men in-

creases plasma PL and effl ux of tissue cholesterol to

small pre-β-HDL where it is esterifi ed.140

• Small pre-β-HDLs cross endothelium into tissue

fl uid, collect free cholesterol, and transfer it to the

liver, where it is converted to bile acids.141

•Infusion of a microemulsion of 1-palmitoyl-2-oleoyl-

phosphatidylcholine (POPC) and apoA-IMilano, that

is, reconstituted HDL–A-IMilano, produced lesion

regression.142

• Phospholipidated HDL is more cholesterophilic than

native HDL and a better cholesterol effl ux acceptor.53

• The reaction catalyzed by serum opacity factor has

some therapeutic promise; serum opacity factor

transfers the CE of 100,000 HDL particles to a single

particle that contains apoE while forming a PL-rich

neo-HDL.85 Hepatic clearance of the apoE-containing

particles via the LDL receptor could greatly enhance

RCT while the neo-HDLs, which are potential accep-

tors of macrophage cholesterol effl ux, could initiate

new RCT cycles.

Thus, discovery of new ways to increase plasma PL,

particularly HDL-PL, is a promising avenue and a chal-

lenge that would complement the effects of the statin

class of lipid-lowering drugs.

Acknowledgment

Henry J. Pownall is supported by grants-in-aid from the

National Institutes of Health (HL-30914 and HL-

56865)