1574
ElResearch Advances Series
Focus on Molecular Biology
Recent Progress in Understanding
Apolipoprotein B
Stephen G. Young, MD
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
T he two isoforms of apolipoprotein (apo) B,
apo B-48 and apo B-100, are important
proteins in human lipoprotein metabolism.
Apo B-48, so named because it appears to be about
48% the size of apo B-100 on sodium dodecyl sulfate
(SDS)-polyacrylamide gels, is synthesized by the
intestine in humans.' Apo B-48 is necessary for the
assembly of chylomicrons and therefore has an obligatory role in the intestinal absorption of dietary fats.
Apo B-100, which is produced in the liver in humans,
is required for the synthesis and secretion of very low
density lipoproteins (VLDL). Low density lipoproteins (LDL), which contain about two thirds of the
cholesterol in human plasma, are metabolic products
of VLDL. Apo B-100 is virtually the only protein
component of LDL. Elevated concentrations of apo
B-100 and LDL cholesterol in plasma are recognized
risk factors for developing atherosclerotic coronary
artery disease.2
Because of the central roles of the two isoforms of
apo B in lipoprotein metabolism and the atherogenic
potential of apo B-containing lipoproteins, apo B has
been a prime target for study by investigators in the
lipoprotein and arteriosclerosis field. Over the past 4
years, significant progress has been made in understanding apo B, including the determination of the
primary structure of the two isoforms of apo B, the
organization of the apo B gene, and the delineation
of the important functional domains of apo B. A
substantial amount of heterogeneity in the apo B
gene has been documented, and multiple apo B gene
mutations that affect the plasma cholesterol level
have been reported. Recent developments in these
areas will be included in this review. However, not all
of the important developments within the apo B field
From the Gladstone Foundation Laboratories for Cardiovascular
Disease, Cardiovascular Research Institute, Department of Medicine, Cardiology Division, University of California, San Francisco.
Supported by grants HL-0162 and HL-41633 from the National
Institutes of Health and by a Syntex Scholars Award.
Address for correspondence: Stephen G. Young, MD, Gladstone Foundation Laboratories for Cardiovascular Disease, Cardiovascular Research Institute, Department of Medicine, Cardiology Division, University of California, San Francisco, San
Francisco, CA 94140-0608.
A glossary of molecular biology terminology can be found in
Circulation 1989;80:219-233.
can be reviewed within the space limitations of this
article; several of these topics deserve special attention. In the past few years, progress has been made in
understanding the kinetics of apo B synthesis and
secretion by cells and also the cellular pathways for
the catabolism of apo B-containing lipoproteins;
these topics have been recently reviewed.3-5 Additionally, in the past 5 years it has become apparent
that biological modifications of apo B-containing
lipoproteins may be important in understanding their
atherogenicity. This exciting area of apo B research
has been recently reviewed by Steinberg and coworkers.6 Another rapidly developing area of apo B
research has been the structure and functional significance of Lp(a), an apo B-100-containing lipoprotein that appears to be a strong risk factor for
coronary artery disease. Progress in understanding
Lp(a) has been recently reviewed by Utermann.7
Overview of Roles of Apo B-48 and Apo B-100 in
Lipoprotein Metabolism
Apo B-48 has an obligatory structural role in the
assembly of chylomicrons in the intestine (Figure
1A).l18 Existing evidence suggests that only one apo
B-48 molecule is integrally associated with each
chylomicron particle9; unlike the other apolipoprotein components of chylomicrons, it remains associated with the particle and does not exchange onto
other lipoprotein particles. Whether chylomicrons
from normal adult human subjects might occasionally
contain apo B-100 instead of apo B-48 is under
investigation in several laboratories. Recently, Dullaart et al10 presented evidence for some apo B-100
production in the intestine. However, in our own
immunocytochemical studies of the duodenal and
jejunal enterocytes of organ donor patients, we found
abundant amounts of intracellular apo B-48 but no
evidence of apo B-100 synthesis (J. Boyles, S. Young;
unpublished results). After secretion from enterocytes, chylomicrons enter the lymph and are delivered to the bloodstream through the thoracic duct. In
the circulation, chylomicrons are rapidly sequestered
along the capillary endothelium, principally that of
skeletal muscle and adipose tissue, where the enzyme
lipoprotein lipase hydrolyzes 80-90% of the particles' triglyceride mass. The resultant lipoproteins,
Young Advances in Understanding Apolipoprotein B
1575
A. Chylomicron/Apo-B48 Metabolism
Dietary Fats
Enterocyte
A-l
.A
B48
B48
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
FIGURE 1. Panel A: Schematic of metabolism of apolipoprotein (apo) B-48 -containing
lipoproteins. Panel B: Schematic of metabolism of apo B-100-containing lipoproteins. See
text for discussion of this figure and Reference 8
for overview of metabolism of apo B. ER,
endoplasmic reticulum; VLDL, very low density
lipoprotein; LDL, low density lipoprotein; IDL,
intermediate density lipoprotein.
B. VLDL/Apo-B100 Metabolism
Hepatocyte
B100, E
Cs
E
EB100
Lipoprotein
Lipase
B100
B100, E
(LDL)
Extrahepatic Tissue Cell
which are called chylomicron remnants, are then
transported to the liver, where they are probably
further metabolized by hepatic lipase and are ultimately bound and taken up by a remnant receptor
that recognizes apo E (Figure 1A). Whether the
receptor involved in chylomicron remnant uptake is
the LDL receptor,1" a recently described LDL
receptor-related protein,12 or another protein13 is a
matter for active investigation. The existing evidence
suggests that apo B-48 does not participate in the
particle's interaction with the hepatic remnant receptor.14 Once internalized into the hepatocyte, the
Receptor
particles are directed to the lysosome, where they are
digested. The cholesterol derived from remnant lipoproteins is esterified and stored as cholesteryl ester,
used for synthesis of bile acids, or used in the
synthesis of hepatic lipoproteins.8 Virtually all chylomicron remnants are removed from plasma as relatively large particles that have a density of less than
1.019 g/ml; consequently, apo B-48 is not normally
detectable in the LDL (d=1.019-1.063 g/ml).
The normal residence time of chylomicrons in the
plasma is only 5-10 minutes.15,'6 As a consequence of
this rapid catabolism, the concentration of apo B-48
1576
Circulation Vol 82 No 5, November 1990
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
in the plasma is very low, probably only a few
micrograms per milliliter, and perhaps only 0.1% that
of apo B-100. Despite the very low levels of apo B-48
in the plasma, it is invariably detectable on a stained
SDS-polyacrylamide gel of a VLDL fraction obtained
from a fasting donor. Apo B- 48 concentrations
within the chylomicron and VLDL fractions increase
after a fat-rich meal relative to that of apo B-100, as
determined by relative staining intensity of the two
bands on SDS-polyacrylamide gels. Recently, Simons
et a117 examined the amount of apo B-48 relative to
apo B-100 on stained gels of the postprandial lipoproteins of control subjects and subjects with coronary artery disease. They found that the apo B-48to-apo B-100 ratio was significantly increased in
patients with coronary artery disease, even when age
and other lipid discriminators, such as triglycerides
and cholesterol, were included in the analysis. The
metabolic basis of their observation on apo B-48
levels -whether the finding is due to increased secretion or decreased catabolism of apo B-48-containing
lipoproteins-is not known; however, the finding is
consistent with long-standing theories that postprandial lipoproteins may be atherogenic.18 No immunoassay for apo B-48 in plasma has yet been developed because no one has been able to develop an
antibody that is specific for apo B-48 (i.e., an antibody that would bind to apo B-48 but not to the more
abundant apo B-100).
Apo B-100-containing lipoproteins are assembled
within hepatocytes, secreted into the space of Disse,
and ultimately enter the circulation through the
hepatic vein (Figure 1B). Most of the apo B-100 is
secreted from the liver on triglyceride-rich VLDL
particles, which also contain apo E and various apo
Cs; kinetic data from turnover studies on normolipidemic subjects suggest that there is probably very
little independent secretion of apo B-100 on smaller,
denser particles of the LDL fraction.19 Unlike other
apolipoprotein constituents of the particle, the apo
B-100 of VLDL particles does not exchange between
lipoprotein particles, so radiolabels attached to apo
B-100 can be used to trace the metabolic fate of the
particle in plasma. After release of nascent VLDL
particles into the bloodstream, the initial phase of
their metabolism resembles that of chylomicrons.8
The triglyceride-rich core of VLDL particles is
hydrolyzed by lipoprotein lipase along the capillary
endothelium. As the particle's core is reduced in
volume, the apo B-100 and apo E on the surface of
the VLDL particle probably change their conformation in such a way that they can serve as ligands for
uptake of the particle by hepatic receptors.8 Probably
one half of all VLDL particles is removed from the
circulation in the liver through interaction with the
LDL receptor (also known as the apo B,E receptor);
the other half remains in the circulation and is
further metabolized to denser particles of the LDL
fraction.8 The half-life of VLDL particles in the
plasma is heterogeneous, ranging from minutes to
hours. Existing evidence suggests that large VLDL
particles, which may contain a substantial number of
apo E molecules, are quickly removed from the
circulation, whereas smaller VLDL particles tend to
be metabolized more slowly, with a greater likelihood
of being metabolized to the smaller, denser particles
of the intermediate density lipoprotein (IDL) and
LDL fractions. By the time the particles are metabolized to the size and density of the LDL range, they
have become enriched in cholesteryl esters, and
virtually the only remaining protein component is
apo B-100. The average residence time of LDL is 2-3
days8; about 80% of LDL is removed from the
circulation by the interaction of apo B-100 with the
LDL receptor and the remainder by nonreceptor
pathways.20 Approximately one half of the LDL is
removed from human plasma by the liver; the other
half is removed by extrahepatic tissues.21 Once LDL
is bound and internalized by cells, it is directed to the
lysosome, where both its protein and lipid components are digested (Figure 1B).
Apo B-100 concentrations can be measured by a
variety of immunoassays,22,23 and normal plasma levels range from 60 to 120 mg/dl. More than 90% of the
apo B-100 within the plasma of normolipidemic
individuals is contained within the LDL fraction22;
hence, there is normally little difference between the
total plasma apo B and LDL apo B levels. The LDL
size and chemical composition are known to be
heterogeneous in the population, and some of this
heterogeneity is probably due to genetic factors.24 A
variety of data indicate that the plasma of patients
with coronary artery disease tends to contain an
increased number of small, dense LDL particles that
have a decreased cholesteryl ester-to-apo B-100
ratio.25,26 The metabolic basis for this finding is
incompletely understood but is under active investigation.27 These patients may have relatively normal
LDL cholesterol concentrations yet elevated apo
B-100 levels. This observation has led some investigators to recommend measuring apo B levels in
addition to LDL cholesterol levels in patients at risk
for coronary artery disease.28
Apo B-100 can become associated with apo(a), a
large glycoprotein structurally related to plasminogen, to form a unique lipoprotein [Lp(a)]. Lp(a)
particles have about the same lipid composition as
LDL particles, but because of the addition of apo(a)
they are larger and denser than LDL particles. Apo
B-100 and apo(a) dissociate from each other when
subjected to reducing agents, and it is generally
assumed that a disulfide bond links the two proteins.
The molecular weight of apo(a) is quite heterogeneous in the population and is genetically determined. The plasma concentration of Lp(a) is strongly
controlled by genetic factors and markedly varied,
from virtually undetectable amounts to 100 mg/dl.
The concentration of Lp(a) in plasma is strongly
related to the risk of developing coronary artery
disease. The cause of its atherogenicity is under
intense investigation; the atherogenicity may relate
Young Advances in Understanding Apolipoprotein B
in part to the potential of Lp(a) in inhibiting thrombolytic mechanisms.7
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Structure of Apo B-100
In 1986 four groups reported the complete nucleotide sequence for the apo-B complementary DNA
(cDNA), along with the deduced amino acid
sequence.29-32 The cDNA is 14,121 nucleotides in
length, with 5' and 3' untranslated regions of 128 and
304 base pairs (bp), respectively. The cDNA codes
for a 27-amino-acid hydrophobic signal peptide,
which is cotranslationally cleaved, followed by a
mature protein of 4,536 amino acids. Boerwinkle and
Chan33 have confirmed the existence of a 3-aminoacid insertion/deletion polymorphism in the apo B
signal peptide, which is of no known functional
significance. The sequence of the apo B-100 polypeptide chain is unique, although computer-assisted
searches have identified remote but statistically significant internal amino acid repeats within apo
B-100.34 Computer sequence alignment programs
have also identified remote but statistically significant
amino acid homologies with vitellogenin and other
apolipoproteins.3435 The sequence homologies with
other apolipoproteins were almost always within
sequences predicted to be capable of forming
amphipathic helices. Apo B is known to be a glycoprotein, with about 4-9% of its mass as carbohydrate
linked to asparagine.36 There are 19 potential
N-linked glycosylation sites on apo B; by direct
sequencing, Yang et a137 found that all but three of
these are glycosylated. Apo B-100 is reportedly posttranslationally modified by covalently bound fatty
acids.38,39 Weisgraber and Rall40 have shown that
there are seven different regions of the apo B-100
molecule that strongly bind to heparin.40 Heparinbinding sites on apo B may serve to promote the
binding of triglyceride-rich lipoproteins to the capillary endothelium, where their lipid cores are digested
by lipoprotein lipase. Apo B contains 25 cysteine
residues, and these are distributed in an asymmetric
fashion within the molecule -12 occur within the first
500 residues. Sixteen of the 25 cysteines are known to
exist in disulfide form, including all 12 in the first 500
amino acids.41 Recently, Coleman et a142 reported
that there are two free cysteines in LDL-apo B-100
located at positions 3,734 and 4,190. Either of these
carboxyterminal cysteines (or both) could potentially
form a disulfide bond with apo(a) to form Lp(a).
While recently investigating a subject whose plasma
lipoproteins contained both apo B-100 and an amino
terminal-truncated isoform of apo B (apo B-46), we
found apo(a) exclusively on the apo B-100 particles
(A. Scanu, S. Young, unpublished observations).
These data are consistent with one of the carboxyterminal cysteines being involved in the formation of
Lp(a).
Apo B-100 contains numerous hydrophobic
domains throughout its length that are believed to be
important in lipid binding.3 In addition, De Loof et
a134 and others29 have identified nine amphipathic
1577
helices, each one 22 amino acids in length, that are
similar to amphipathic a-helices found in the putative lipid-binding domains of other apolipoproteins.
Nearly all of the apo B-100 sequences having the
capacity to form amphipathic a-helices are located in
the carboxyterminal half of the apo B-100 molecule.
In addition, apo B contains multiple proline-rich
sequences predicted to form amphipathic /-sheets
and P3-turns; these are located throughout the apo B
sequence except for the aminoterminal 1,000 amino
acids. Although amphipathic P3-sheets are not found
in other apolipoproteins, these structures are thought
to have high lipid-binding potential. Thus, apo B has
many potential lipid-binding regions throughout its
length, a result thought to be adequate for explanation of the fact that apo B never exchanges between
lipoprotein particles, in contrast to the other apolipoproteins, which have one or two putative lipidbinding domains and readily exchange between lipoproteins. Chen and coworkers have reported
experimental evidence that many different apo B
peptides are capable of binding to lipoprotein particles.43 They digested LDL particles with proteases in
the presence of SDS and then removed the SDS by
dialysis, allowing the apo B peptides to recombine
with lipid. Recombinant lipoproteins having the size
and physical characteristics of LDL formed spontaneously after removal of the detergent. Peptides
throughout the length of apo B-100 were identified in
these recombinant particles. Although these investigators did not rigorously exclude peptide-to-peptide
interactions among multiple hydrophobic fragments
of apo B, their study tends to confirm the prediction
of multiple lipid-binding domains. If many sequences
throughout the entire length of apo B are important
for the binding of lipid, then truncated apo B proteins would be predicted to form denser, lipid-poor
particles. Indeed, studies performed by Young et a144
and Graham et a145 have shown that shorter isoforms
of apo B seem capable of binding less lipid. Young
and coworkers found that a truncated apo B of 2,046
amino acids (apo B-46) was found primarily in the
VLDL and LDL fractions46; a truncated apo B of
1,728 amino acids (apo B-37) was also found in the
d<1.006 g/ml fraction of both fasting and postprandial plasma, but a large portion of the apo B-37 was
contained in the HDL fraction47; a truncated apo B
of 1,425 amino acids (apo B-31) was found only in the
dense subfractions of HDL and in the d>1.21 g/ml
fraction.44 So the shorter the apo B protein, the
denser and more lipid poor the particle. Furthermore, the discovery of apo B-37 but not apo B-31 in
the VLDL fraction suggests that apo B sequences
between 1,425 and 1,728 amino acids may be crucial
for forming buoyant, triglyceride-rich VLDL particles.44
A significant contribution to our understanding of
the disposition of apo B on lipoprotein particles has
been made by Yang and coworkers.37 These investigators have determined the regions of the apo B
molecule on LDL particles that were "accessible" to
1578
Circulation Vol 82, No 5, November 1990
FIGURE 2. Schematic of apolipoprotein (apo) B-100 strucdensity lipoprotein particle. The location of two
sites where thrombin cleaves apo B on low density lipoprotein
particles (residues 1,297 and 3,249) is indicated. Indicated are
the locations of 16 N-linked carbohydrates (CHO; o), cysteine
residues (o), and disulfide bridges (=). Additional data on
disulfide bridges are given in subsequent abstract.41 Trypsinreleasable regions of apo B are shown on outside ofparticle;
trypsin-nonreleasable regions are inside core ofparticle. Yang
and coworkers41 emphasized that surface versus core location
of apo B peptides shown in this illustration is hypothetical. The
five hypothetical domains of apo B discussed in text are
indicated by Roman numerals and separated by dashed lines.
Reproduced with permission.37
ture on low
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
trypsin and therefore readily releasable from LDL
particles [the trypsin-releasable (TR) peptides] and
the portions that were accessible only after the LDL
particles had been delipidated and retrypsinized (the
trypsin nonreleasable [TN] peptides). Some peptides
were contained in both the TR and TN fractions;
these were designated mixed (MX) peptides. Altogether, they sequenced more than 88% of the apo
B-100 protein, with many peptides being repetitively
sequenced. On the basis of the trypsin releasability
and nonreleasability of various regions of the molecule, they identified five broad domains of apo B on
LDL particles (Figure 2): domain I, amino acids
1-1,000, predominantly TR; domain II, amino acids
1,001-1,700, alternating regions of TR and TIN peptides, with a significant proportion of MX peptides;
domain III, amino acids 1,701-3,070, occasional TR
peptides but primarily a TN region; domain IV,
amino acids 3,071-4,100, mainly TR and MX peptides; and domain V, amino acids 4,101-4,536,
almost exclusively TN peptides. In general, the more
hydrophobic the region, the less likely it was to be
accessible to trypsin. The authors were careful to
point out that the factors governing trypsin releasability versus nonreleasability are not completely
understood, and they explicitly cautioned against the
assumption that the trypsin releasability versus nonreleasability implies a lipoprotein-surface versus
lipoprotein-core location for the peptide. However,
the latter assumption clearly represents a tempting
hypothesis. Recently, Yang and coworkers48 purified
and sequenced 13 tryptic peptides of apo B-100 that
spontaneously associated with three different phospholipids. All of the peptides were highly hydrophobic; predictive algorithms revealed no evidence for
amphipathic helices, and no spectroscopic evidence
for P-sheet structure was found. Five of the peptides
were "surface" ones (trypsin accessible), and these
were located throughout the polypeptide chain. The
remainder were core peptides (trypsin inaccessible),
and most of these were clustered in the carboxyterminal 33 amino acids of apo B-100.
Recently, Phillips and Schumaker49 examined the
apo B-100 of LDL particles by electron microscopy
(EM) after the particles had been treated with
glutaraldehyde and their lipid extracted; the images
thus produced strongly suggested that apo B-100 may
indeed encircle LDL particles, as shown in Yang et
al's schematic (Figure 2). New EM techniques have
recently been developed to visualize apo B on the
surface of LDL particles.50 The binding of apo
B-specific monoclonal antibodies whose epitopes
have been precisely localized can also be directly
visualized (V. Schumaker, personal communication).
These studies should improve our understanding of
the relative location of various regions of apo B on
LDL particles.
For several years, it has been observed that the
LDL receptor-binding region of apo B-100 is in the
carboxyterminal portion of the molecule. The
receptor-binding region of apo B-100 has been
recently studied in detail by Milne and coworkers.51
They determined the location of the epitopes of more
than 30 apo B-specific monoclonal antibodies and
which of the antibodies were capable of inhibiting the
binding of LDL particles to the LDL receptor. The
antibodies with epitopes located between amino
acids 2,980 and 3,780 completely block the specific
binding of LDL to the LDL receptor when bound to
LDL. Antibodies that bind to epitopes immediately
flanking this region partially block binding of LDL to
its receptor, whereas monoclonal antibodies binding
elsewhere in the molecule had for the most part little
or no receptor-blocking activity. A summary of the
efforts of Milne and coworkers is shown in Figure 3.
Other data also implicate the carboxyterminal region
of apo B-100 in binding to the LDL receptor. First, a
lipoprotein containing a truncated apo B isoform
lacking the carboxyterminal region does not bind to
the LDL receptor.52 Second, the carboxyterminal
domain of apo B implicated in the studies of Milne et
al contains three different heparin-binding regions40
that have clusters of positively charged amino acids;
sequences within one of the heparin-binding domains
(residues 3,359-3,367) have sequence homology with
the region of apo E known to interact with the LDL
receptor.29 Yang and coworkers53 have presented
evidence that a peptide containing these amino acids
but no other apo B peptide could bind to the LDL
receptor. The regions of apo B containing the clusters of positively charged amino acids are known to
be evolutionarily well conserved.51 Positively charged
Young Advances in Understanding Apolipoprotein B
3F5
4G3
(2835.2922) (2980-3084)
NH2
regions of apo B are thought to bind to the many
negatively charged amino acid residues within the
cysteine-rich repeats of the ligand-binding domain of
the LDL receptor.5 Chemical modification of the
positively charged residues of apo B block its binding
to the LDL receptor.54 Finally, as discussed below, a
missense (Arg -- Gln) mutation in the codon for apo
B-100 amino acid 3,500 is associated with defective
binding of LDL to the LDL receptor.
3167
-
-
5E11, 3A10, MB47
(3441-3569)
S,
3297
Arg/Gin
3500
-COOH
B1B3
(3665-3780)
MB43
(4027X4081)
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
FIGURE 3. Schematic ofproposed structure of the putative
binding domain of apolipoprotein (apo) B. Regions (specified
by position numbers of apo B-100 amino acid residues) that
contain the epitopes of monoclonal antibodies that totally or
partially block binding of low density lipoprotein (LDL) to the
LDL receptor are indicated by solid and open boxes above or
below polypeptide chain, respectively. The two short stretches
ofpositively charged amino acids are indicated by open boxes
within polypeptide chain; regions of amphipathic P-sheet
structure are depicted by zigzag line. The locations of cysteines
at residues 3,167 and 3,297, which are linked by disulfide
bond (S-S), are indicated, as is the location of the Arg---Gln
amino acid change at residue 3,500, a substitution that is
associated with hypercholesterolemia (see text). The cysteines
at residues 3,167 and 3,297 are not well conserved in apo B
sequences of other mammalian species.51,i54 The epitopes for
antibodies MB47 and B1B3 are shown in close proximity
because these antibodies have been shown to compete with
each other for binding to LDL. Reproduced, with slight
modifications, with permission.51
Apo B Gene and Control of Its Expression
The apo B gene, which is located on the p23-ter
region of chromosome 2,55-58 spans approximately 43
kb and contains 29 exons and 28 introns59; two of the
exons, exons 26 and 29, are extremely long (7,572 and
1,906 bp, respectively) (Figure 4). Exon 26, which
codes for amino acids 1,379-3,903, is threefold
longer than any previously reported exon of any
mammalian gene. Whether the two large exons arose
from fusion of smaller exons is unknown. Six repetitive sequences are present in the introns of the
gene.59 There is an AT-rich hypervariable region
within 200 bp of the polyadenylation site that consists
of a variable number of 15-bp hypervariable elements.6061 There are at least 14 different 3' hypervariable region alleles, and about 75% of the population is heterozygous for different alleles at this
site.62,63 Another tandem repetitive sequence, with at
least three commonly occurring alleles, has been
recently identified within intron 20 of the apo B gene
(H. Hobbs, personal communication). The number of
-43 kb
Apo-B Gene
5'
exon 26
1111111111111 II1 1I
1 11
CAA
GIn-2153
1
ver
Apo-B mRNA
1
Apo-B Protein
-3
A
CAA
GIn-2153
Asoforms
1
14.1 kb
5"
NH
NHa
Apo-Bl
00:
UAA
Termination
Codon
COOH
4536 amino acids
1579
exon 29
A
TAA
Intestine i
14.1 kb
5'
~~~A
UAA
3'
UAA
7-8 kb
5'1
3'A
UAA
NH2
COOH
Apo-B48: 2152 amino acids
FIGURE 4. Schematic of mechanism of apolipoprotein (apo) B-48 synthesis from the apo B gene. The apo B gene contains 29
which are illustrated by black lines and bars at top offigure. The location of the two large exons (26 and 29) are indicated.
Within the apo B gene and hepatic apo B messenger RNA (mRNA), codon 2,153 (CAA) specifies glutamine. However, in the
intestine, the first base of this codon (nucleotide 6,666) is edited to U, producing a UAA termination codon. A subset of human
apo B mRNAs are 7-8 kb in length because of the use of altemate polyadenylation signals, but virtually all human intestinal apo
B mRNAs, regardless of length, contain the edited base and therefore yield apo B-48.8081
exons,
1580
Circulation Vol 82, No 5, November 1990
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
repetitive elements at the hypervariable regions can
be rapidly and accurately assessed with polymerase
chain reaction-based techniques; therefore, the utility of these regions for epidemiological and family
studies is large.
The apo B gene contains a TATA box and a CAT
box within the 60 bp upstream from the transcriptional start site.59 Das and coworkers64 have reported
that 261 bp upstream from the transcriptional start
site was sufficient for liver-specific expression;
sequences located between -128 and -86 contained
a positive element acting to increase expression.64 In
the latter positive element, Levy-Wilson et a165
reported the existence of a DNase I-hypersensitive
site and a 13-bp sequence that is perfectly conserved
between the mouse and human promoters.65
Sequences between -86 and -70 contained another
strongly positive element; recent studies have shown
that a protein designated AF-1 binds to these
sequences. AF-1 has also been shown to bind to
promoters of other apolipoprotein genes.66 Mutations within this AF-1-binding sequence have been
shown to reduce transcription rates by 50-fold.
C/EBP, another hepatic transcription factor, binds to
an element between -69 and -52; site-directed
mutagenesis of this element reduced transcriptional
activity twofold to fourfold.66 These sequences are
not likely to be the only important protein-binding
sites; Levy-Wilson et a165 have documented the existence of DNase I-hypersensitive sites (which tend to
correlate with protein binding) as far as 700 bp
upstream from the transcriptional start site.
The factors and sequence elements that are
responsible for terminating apo B gene expression in
tissues other than liver and intestine are not fully
understood. One factor could be DNA methylation.
Levy-Wilson and Fortier67 have shown that the promoter of the apo B gene is undermethylated in
intestinal and liver cell lines but methylated in cells
not transcriptionally active. Whatever the mechanism
for terminating apo B transcription in other cell lines,
it is probably not invariably absolute. We have
recently amplified the apo B cDNA from fibroblast
RNA.68 Thus, there is probably low-level "illegitimate transcription"69 of the apo B gene in tissues
other than liver and intestine.
There is little evidence that the apo B messenger
RNA (mRNA) level is significantly regulated by
dietary factors in mammals. Long-term cholesterol
feeding has reportedly resulted in only minor
changes in apo B mRNA levels in mice, rats, and
rabbits.70-72 Recently, Sorci-Thomas et a173 investigated apo B mRNA levels in African Green monkeys
fed low- or high-cholesterol diets in the presence of
saturated or polyunsaturated fats. Ingestion of cholesterol and saturated fats markedly increased
plasma apo B and LDL cholesterol levels, but no
increases in hepatic apo B mRNA levels were found
in the animals. However, the high-cholesterol-diet
group did have a 50% decrease in mRNA for the
LDL receptor. The authors concluded that the
plasma levels of apo B-containing lipoproteins were
governed primarily by factors controlling the removal
of these lipoproteins from the plasma rather than by
factors controlling their synthesis.
No studies on human hepatic or intestinal apo B
mRNA levels in response to diet have been performed. However, Pullinger et a174 recently studied
the apo B mRNA levels and synthesis rate in a
human hepatoblastoma cell line under a variety of
metabolic conditions. Although they found that the
secretion of apo B into the medium could be modulated by insulin and fatty acids, they found little or no
evidence for acute regulation of apo B mRNA levels.
They concluded that the apo B gene was constitutively expressed in these cells and that differences in
apo B secretion rates under different conditions must
be due to cotranslational or posttranslational processes. This conclusion was supported by the observation that the apo B mRNA half-life was quite long
(16 hours). In a study of rat intestinal synthetic rates
and mRNA levels under different conditions, Davidson and coworkers75 reached similar conclusions.
The apo B mRNA level did not change under
conditions in which synthesis rates varied significantly. The cotranslational and posttranslational factors governing apo B synthesis and secretion rates are
currently not fully understood; however, several
investigators working with different systems have
reported evidence that intracellular lipid availability
may be an extremely important factor in the rate of
apo B secretion from cells.75-78 Several investigators
engaged in the study of the kinetics of apo B secretion from cells have recently reported evidence that a
portion of the intracellular apo B may be degraded
but not secreted and have suggested that metabolic
conditions (i.e., lipid availability) may influence the
portion of the intracellular apo B pool that is secreted.76'79 Whether the rate of apo B secretion is
physiologically regulated in vivo by this type of mechanism is an important issue that deserves intensive
investigation, as it could have a bearing on strategies
to control oversecretion of apo B.
Mechanism of Apo B48 Formation
In addition to the structural relation of apo B-48 to
apo B-100, the mechanism for apo B-48 formation
was almost simultaneously determined in 1987 by
Chen et a180 and Powell et a181 and later by Hospattankar et al.82 These investigators demonstrated that
apo B-48 is produced from the apo B-100 gene by a
novel mechanism involving mRNA editing. In
sequencing the apo B cDNA isolated from human
intestinal cDNA libraries, they found that the intestinal apo B cDNA contained a T at nucleotide 6,666,
in contrast to the C found at the position in the liver
apo B cDNA clones (Figure 4). The substitution of
the T for a C at nucleotide 6,666 yields an in-frame
stop codon (TAA) that replaces the CAA codon
specifying apo B-100 amino acid Gln at position
2,153. The location of the stop codon predicts that
apo B-48 found in plasma would contain the amino-
Young Advances in Understanding Apolipoprotein B
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
terminal 2,152 amino acids of apo B-100; this prediction was strongly supported by the proteinsequencing efforts of Chen et a180 as well as by the
studies of Hardman et al.83 This substitution at
nucleotide 6,666 was present in the intestinal
mRNA80 but not in intestinal genomic DNA,81 so it
was clearly the result of a specific form of posttranscriptional editing of the intestinal apo B mRNA.
Apo B mRNA editing to create an in-frame stop
codon has been shown to occur in rat and rabbit
intestinal apo B mRNA8184 and rat liver, which is
known to produce both apo B-48 and apo B-100.84,85
A fraction of the human intestinal apo B mRNA is
7-8 kb in length because of the use of alternate
polyadenylation signals after the edited nucleotide
(Figure 4). The process in humans appears to be
developmentally regulated; intestinal organ cultures
drawn from fetuses in early gestation primarily make
apo B-100, whereas the adult intestine makes apo
B-48 and very little, if any, apo B-100.86 The efficiency of apo B mRNA editing in rat liver has
recently demonstrated modulation by fasting/
refeeding87 and thyroid hormone.84
The exact mechanism for apo B mRNA editing
remains unclarified, but the most plausible hypothesis is that a tissue-specific enzyme recognizes a specific sequence within the apo B mRNA and deaminates position 4 of cytosine-6,666, creating a uracil
residue.88 A reasonable hypothesis suggests that a
specific RNA editing process would require conservation of the nucleotide sequence surrounding nucleotide 6,666 through evolution. Indeed, sequencing
studies of the apo B gene from several species have
shown that the sequence of the region flanking
nucleotide 6,666 is highly homologous (90%) among
the mouse, rat, rabbit, and human genes.89 However,
Chen et a190 have recently used site-directed mutagenesis techniques to introduce 22 different mutations into a nine-base region flanking nucleotide
6,666. Unexpectedly, they discovered that most of the
constructs were edited, implying that the editing
mechanism may be at least partially lax. Two groups
have reported data on the length of the apo B
sequence required for apo B editing. Bostrom and
coworkers91 inserted two lengths of the apo B cDNA
flanking nucleotide 6,666 into the protein-coding
sequence of an apo E expression vector. When a
354-bp segment of apo B cDNA flanking nucleotide
6,666 was inserted into the expression vector and the
expression vector transfected into an intestinal carcinoma cell line, editing of the mRNA at nucleotide
6,666 occurred. However, no editing was observed
when a 63 -bp portion of apo B cDNA flanking
nucleotide 6,666 was inserted into the expression
vector. Davies and coworkers89 have also examined
the question of sequences required for the apo B
editing process. They transfected constructs containing eight different lengths of apo B cDNA flanking
nucleotide 6,666 (ranging in size from 26 to 2,385
bases) into McArdle 7777 cells, a rat hepatoma cell
line that synthesizes and secretes apo B-48. They
1581
found that mRNA editing occurred with each of the
constructs, revealing that as few as 26 nucleotides
surrounding base 6,666 were sufficient for mRNA
editing to occur. The discrepancy between Davies's
and Bostrom's results could very well be due to the
confounding effects of the apo-E sequences in Bostrom's constructs. Davies et a189 reported that an
RNA-folding computer program predicted that the
edited nucleotide exists in a conserved eightnucleotide loop in an mRNA stem loop structure.
Driscoll and coworkers88 have demonstrated that
the RNA editing mechanism can be observed in vitro
with cytoplasmic extracts from McArdle 7777 cells.
They inserted various lengths of the apo B cDNA
surrounding base 6,666 into appropriate plasmid
vectors, and various lengths of apo B RNA were
synthesized in vitro. Editing of the synthetic RNAs
containing as few as 55 bases of the apo B sequence
was demonstrated with cytoplasmic extracts. Editing
was specific for RNA but not DNA and was
destroyed by proteinase K, suggesting that the editing
activity involves a protein. Results of gel retardation
assays using extracts from an apo B-48-producing
cell line have suggested the presence of a factor that
binds specifically to the appropriate region of the apo
B mRNA.92 Efforts to purify this activity are under
way in several laboratories. Scott and coworkers have
recently developed transgenic mice containing a
2-3-kb construct spanning the apo B-48 editing
site.93 Apo B mRNA was identified in multiple
tissues, and substantial amounts of editing activity
were documented in the intestine, spleen, and lung.
Lower levels of editing activity were documented in
the liver, brain, and heart. The physiological rationale for editing in this broad spectrum of tissues is
not known, but these data certainly present the
possibility that editing might occur for gene products
other than apo B.
Editing of the intestinal apo B mRNA to create an
in-frame stop codon was unprecedented in the
molecular biology of mammals, and its existence
raises many questions.80,81,88 Is the C at base 6,666
simply deaminated to produce a U, or is it possible
that the base is modified in some other way so that it
is read by the ribosome and reverse transcriptase as a
U? Outside the nine-base region examined by Chen
et al,90 what are the nucleotide specifications of the
editing process? What process signals the use of
alternate polyadenylation sites after the edited nucleotide in the human intestine? Is RNA editing the
result of a single protein or a complex comprising
several proteins (and perhaps RNAs)? Does the
RNA editing process exist only for the apo B mRNA,
or is it a mechanism that is used for other transcripts
in other tissues? These questions are important, and
at least some of the answers will probably be forthcoming in the next 2-3 years. However, a more
fundamental question for the lipoprotein metabolism
field should also be considered. Why (in mammalian
physiology) is there a need to make two forms of apo
B, and why has the process been so conserved
1582
Circulation Vol 82, No 5, November 1990
through evolution? Currently, there is no clear
answer to this question. Apo B-100 seems perfectly
capable of participating in the assembly of large,
triglyceride-rich particles. What physiological role
does apo B-48 have in the intestine that would not be
equally well served by apo B-100? Perhaps apo B-48
particles can accommodate more apo E molecules
than can apo B-100 particles, resulting in more
efficient and rapid delivery of dietary fats to the liver.
Scott et a193 recently postulated that apo B-48 may
have evolved for the purpose of efficient delivery of
antioxidants to the liver, where they may protect
nascent hepatic lipoproteins from free radical attack.
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Genetic Variation in Apo B Gene
Even before the cloning and sequencing of the
cDNA and gene for apo B, strong evidence existed
that apo B was a polymorphic protein. From the early
1960s, it has been observed that antibodies reacted
with human LDL in the sera of multiply transfused
thalassemic patients. Family studies showed that the
antigens bound by these antibodies were present in
the sera of some individuals but not others and that
the presence of the antigens was a trait that was
inherited in a simple Mendelian fashion. By 1974 five
different antithetical pairs of LDL antigens [Ag(x/y),
Ag(al/d), Ag(c/g), Ag(t/z), and Ag(h/i)] had been
described; no new antigens have since been
described.94 When clones for the apo B cDNA
became available, a family study demonstrated that
one of the Ag polymorphisms was in strong linkage
disequilibrium with a commonly occurring restriction
fragment length polymorphism (RFLP) in the apo B
gene,95 lending considerable support to the idea that
the Ag polymorphisms resulted from inherited structural variations in the apo B gene. Based on various
combinations of these five pairs of antigens, many
different LDL antigenic phenotypes are clearly possible; with the use of these antisera, a large number
of different LDL phenotypes has actually been documented in human populations. The large number of
LDL phenotypes has been shown to be due to at least
15 different LDL (or apo B) haplotypes.94 The frequency of different LDL phenotypes and haplotypes
differs in various ethnic populations.94
In the past 3 years, several investigators have
shown that several of the same Ag antigens originally
recognized by human polyclonal antisera could also
be recognized by apo B-specific murine monoclonal
antibodies. Young et a197 showed that monoclonal
antibody MB19, which was originally developed by
Curtiss and Edgington,96 detects a commonly occurring two-allele polymorphism in apo B; it binds to
apo B allotypes MB191 and MB192 with high and low
affinity, respectively. Subsequently, the polymorphism detected by antibody MB19 was shown to be
identical to the Ag(c/g) polymorphism.98,99 Antibody
MB19 has been a very useful reagent in determining
the concentrations of a particular apo B allotype
produced by a mutant apo B allele.100 In addition,
immunoaffinity columns using antibody MB19 have
been useful techniques for separating the apo B
allotypes in the plasma of MB19 heterozygotes (S.
Young, J. Witztum, L. Curtiss, T. Innerarity, unpublished observations). More recently, a mouse monoclonal antibody (H1163) that detects the Ag(h/i)
antigen pair has been developed and characterized.'01 DNA polymorphisms associated with the
MB19 and H1163 immunochemical polymorphisms
have been identified.102,103
With the cloning of the apo B cDNA and gene,
data demonstrating the genetic heterogeneity in apo
B have rapidly accumulated. Many RFLPs have been
defined (Table 1). Furthermore, many nucleotide
sequence differences have been documented. A list
of these sequence differences, originally compiled by
Ludwig et al,104 has been recently updated and
expanded by Yang et a137 to include 75 sequence
differences. As Yang and associates discussed, some
of these differences are undoubtedly the result of
DNA sequencing errors; however, 12 of the differences have been documented by more than one
laboratory and therefore probably represent bona
fide nucleotide sequence polymorphisms. Ten of
these 12 sequence changes result in amino acid
substitutions. Several of the amino acid substitutions
correlate with the Ag phenotypes and are easily
detectable because they change restriction endonuclease sites in the apo B gene (Table 1).
With the identification of apo B gene RFLPs, a
number of different investigators performed "association" studies, that is, studies designed to determine
whether some commonly occurring apo B sequence
polymorphisms were associated with a given clinical
phenotype, such as hypercholesterolemia, hypertriglyceridemia, or coronary artery disease. Such studies
have been based on the supposition that the DNA
sequence polymorphisms accounting for the RFLP
might change a crucial amino acid, alter the metabolism of apo B, and thereby cause a particular clinical
phenotype or on the more likely supposition that the
DNA sequence polymorphism would not cause the
phenotype but would be in linkage disequilibrium
with an as-yet undiscovered causative DNA polymorphism.105 The most widely evaluated RFLP has been
the XbaI polymorphism (Table 1). Interestingly, the
DNA sequence change altering this restriction site
does not result in an amino acid substitution and
therefore could not by itself account for a phenotype
change. Many investigators have found the XbaI
polymorphism to have significant clinical associations. In a study of nonfasting serum from men
attending a London heart study clinic, Law et al'06
found that individuals homozygous for the presence
of the XbaI site had slightly increased cholesterol and
triglyceride levels compared with men with normal
levels. Shortly afterward, the same positive association between the XbaI site and lipid concentrations
was observed by Talmud et al'07 in a random series of
normolipidemic men in London. An association
between elevated cholesterol and apo B levels and
the presence of the XbaI site was noted in Norwegian
Young Advances in Understanding Apolipoprotein B
1583
TABLE 1. Restriction Fragment Length Polymorphisms in the Apolipoprotein B Gene*
Frequency of rare allele
0.2 (site absent)
0.21 (site absent)
ApaLI
Location
4 kb upstream of exon 1
Apo B gene site promoter, -265
bp from cap site
Exon 4; cDNA nucleotide 417
HinclI
PvuII
AluI
Intron 4, 3,334 bp 3' to exon 3
Intron 4,523 bp 5' to exon 5
Exon 14; cDNA nucleotide 1,981
0.12 (site present)
0.08 (site present)
0.48 (site absent)
BalI
XbaI
MspI
Intron 20, 146 bp 5' to exon 21
Exon 26; cDNA nucleotide 7,674
Exon 26, cDNA nucleotide 11,040
0.50
0.4-0.5 (site present)
0.12 (site absent)
Reference 155
No associated amino acid substitution.159
Presence of site associated with Arg3,611 and Ag(i).
Absence of site associated with Gin3,611 and
EcoRI
Exon 26; cDNA nucleotide 12,670
0.20 (site absent)
Presence of site associated with Glu4,154 and Ag(t).
Absence of site associated with Lys4,154 and
Enzyme
AvaIl
MspI
0.36 (site absent)
Comments
Reference 155
Reference 156
Presence of site associated with Thr71, Ag(g),
low-affinity MB19 allotype. Absence of site
associated with I1e71, Ag(c), and high-affinity
MB19 allotype.97,O2,157
References 116, 155, and 158
References 116, 155, and 158
Presence of site associated with Ala591, Ag(a,), and
low-affinity H1163 allotype. Absence of site
associated with Vals,l, Ag(d), and high-affinity
H1163 allotype.103
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Ag(h).160,161
Ag(z).56.162
cDNA, complementary DNA.
cDNA, complementary DNA.
*A rare TaqI polymorphism,"12 rare EcoRV and Stul polymorphisms,'63 and uncommon RsaI polymorphismsM64 have also been reported.
Size of polymorphisms created by 3' hypervariable region can be detected with many different restriction enzymes.6061
students108; a follow-up study on hypercholesterolemic Norwegian men yielded similar results.109
In Finland, the same association between the XbaI
site and total cholesterol levels has been observed in
a survey of both healthy subjects'10 and those with
familial hypercholesterolemia.1"' However, the association between the Xbal site and elevated lipid
levels has not been universally observed. In Seattle,
Deeb and coworkers12 discovered a significant relation between the absence of the XbaI site and higher
triglyceride levels. In Japan, where the allele frequencies for the XbaI polymorphism are different, no
association between the XbaI polymorphism and
lipid values was observed."13 In both coronary artery
disease and control groups, Hegele and coworkers"14
found no statistically significant effect of the XbaI
polymorphism on lipid levels. Similarly, no association was observed in Sweden,"15 and no relation
between the XbaI polymorphisms and lipid values
was observed in a South Wales population.16 Finally,
in a study by Monsalve et al,1"7 association of the
presence of the Xbal site with higher cholesterol and
triglyceride levels was observed only within certain
subgroups of patients with atherosclerotic disease;
the relation was actually reversed in other subgroups
of patients (i.e., the presence of the XbaI site was
correlated with lower lipid levels). Conflicting data
also exist for the association of lipid levels with other
apo B polymorphisms, such as the one recognized by
monoclonal antibody MB19.1"0,118
Several groups have examined the association
between apo B RFLPs and the presence of coronary
artery disease. A significant relation between certain
apo B haplotypes and atherosclerotic disease was
observed in a South Wales population.1"6 In a casecontrol study, Hegele et al found that alleles lacking
the XbaI sites, alleles lacking the EcoRI site, and
alleles containing more repeats in the 3' hypervariable region were all found at higher frequencies in
patients with coronary artery disease compared with
that of controls.14 Monsalve et al1"7 found a similar
association between the XbaI and EcoRI polymorphisms and the presence of peripheral vascular disease. In a recent study of an Austrian population,
Friedl et al1"9 also found an increased frequency of
alleles with more repeats in the 3' hypervariable
region in a coronary artery disease population. However, in Seattle there appeared to be an increased
frequency of shorter 3' hypervariable region alleles in
a coronary artery disease population compared with
that of controls, and there was no difference in the
frequency of the XbaI and EcoRI RFLPs between
controls and coronary artery disease patients."12
Additionally, no association between the XbaI and
EcoRI polymorphisms and the presence of coronary
artery disease could be demonstrated in a small London population studied by Ferns and coworkers.'20
Several reasons are obvious for some of the different results obtained in these association studies.
First, several studies were quite small, involving less
than 100 and occasionally less than 50 subjects, and
the possibility of overlooking a significant relation is
increased in small studies. Second, in some studies,
the control group was poorly defined and probably
not comparable with the hyperlipidemic or athero-
1584
Circulation Vol 82, No 5, November 1990
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
sclerotic disease groups. Finally, different ethnic
groups were evaluated in the different studies, and
apo B haplotype frequencies are known to be different in different ethnic populations.94 Thus, it is
quite plausible that in Finland one of the apo B
haplotypes containing the XbaI site contains an
as-yet-undefined "mutation X" that causes high cholesterol levels. Linkage of mutation X to the XbaI site
could explain the fact that Finnish subjects possessing an apo B allele with the Xbal site have higher
cholesterol levels. However, mutation X may be
extremely rare in South Wales, and the association
between the XbaI polymorphism and cholesterol
levels would therefore be expected to be negative.
What have the apo B RFLP association studies
taught us? First, despite the aforementioned methodological concerns, the significant RFLP associations in multiple studies suggest that genetic variation at the apo B locus can influence lipids and
attendant coronary artery disease risk. Theoretically,
this phenomenon is interesting and informative.
However, because conflicting data on specific associations have been obtained in different populations, it
seems equally clear that determination of individual
RFLPs in heterogeneous populations for assessing
the genetic risk of coronary artery disease is not
worthwhile. Second, these studies have taught us that
future efforts to assess the effect of the apo B locus
on hyperlipidemia and atherosclerosis need to
involve larger numbers of patients as well as carefully
defined controls, and genetic variation at the apo B
locus will need a more sophisticated definition.
Rather than observing associations with a single
RFLP, future studies must use a combination of
genetic markers that include multiple RFLPs and a
precise assessment of the number of repeats in the 3'
hypervariable domain so as to define apo B haplotypes. Precise definition of haplotypes will require
many family studies. Because a substantial number of
apo B haplotypes are likely to be defined, large
population studies will be required to discern associations between an isolated apo B haplotype and a
clinical phenotype. If an association between a given
haplotype and a clinical phenotype is found, then the
observation should be confirmed in another subgroup
of the same population and also in family studies.
If future studies provide successful identification
of an apo B haplotype associated with a clinical
phenotype, every effort should be made to understand the observation on a biochemical level. For
example, when an apo B haplotype is associated with
altered levels of cholesterol, a key goal should be to
determine whether the apo B haplotype is associated
with an altered plasma concentration of the apo B
allotype. That certain apo B alleles may be associated
with an altered concentration of an apo B allotype
was initially demonstrated by Young et al in 198697
and more recently in studies by Gavish et al.121 If a
haplotype is found to be associated with altered
levels of a given apo B haplotype, every effort should
be made to identify the precise apo B gene mutation
accounting for this finding. Given the size of the apo
B gene, the magnitude of these studies would not be
trivial. However, new techniques for screening mutations may ultimately facilitate this process.
Mutations in Apo B Gene That Cause Low Blood
Cholesterol Levels
For many years, it has been recognized that a
phenotype characterized by low plasma levels of LDL
cholesterol can be inherited within families in a
simple Mendelian fashion. The phenotype was designated familial hypobetalipoproteinemia because of
the low plasma concentrations of LDL, a class of
lipoproteins that migrate in the /3-position on agarose
electrophoresis. In the past 3 years, a variety of
laboratories have shown that specific apo B gene
mutations interfering with the translation of a fulllength apo B-100 molecule (i.e., nonsense and frameshift mutations) can cause the syndrome of familial
hypobetalipoproteinemia (Table 2).
Subjects who inherit one normal and one mutant
apo-B allele, and are therefore heterozygous for
familial hypobetalipoproteinemia, are usually asymptomatic.123 It makes intuitive sense that these individuals, who possess only one normal apo B allele,
would have plasma LDL cholesterol and apo B
concentrations of about one half those found in
normal subjects. In fact, most hypobetalipoproteinemia heterozygotes have LDL cholesterol and apo B
concentrations closer to one quarter or one third
those of normal (-30-40 mg/dl). The explanation
for the less-than-half normal apo B and LDL cholesterol levels observed in many of these subjects is as
yet not fully understood. Although one could argue
that this finding could partially be the result of
selection bias, the presence of very low cholesterol
values in affected members of an extended kindred
argues against this possibility. It is possible that the
hepatic LDL receptor activity is upregulated in these
individuals, resulting in an enhanced clearance rate
for the VLDL and LDL particles produced by the
normal allele. Alternatively, it is possible that the
buoyant, triglyceride-rich, apo B-containing lipoproteins produced by the normal apo B allele contain the
full amount of apo E produced by two normal apo E
alleles. A double dose of apo E per lipoprotein
particle would be expected to increase the hepatic
clearance of VLDL and IDL particles derived from
the normal allele, thereby reducing LDL production
rates from the normal allele. These possible explanations for the very low cholesterol levels of familial
hypobetalipoproteinemia heterozygotes could potentially be tested by lipoprotein turnover studies and
biochemical analysis of the lipoproteins of affected
subjects.
Although the prevalence of atherosclerotic disease
in well-characterized familial hypobetalipoproteinemia heterozygotes has never been systematically
studied, it is reasonably presumed that because of
their low LDL cholesterol levels, they would be at
low risk for developing coronary heart disease.
Young Advances in Understanding Apolipoprotein B
1585
TABLE 2. Apolipoprotein B Gene Mutations Associated With Altered Levels of Blood Cholesterol
Low cholesterol levels -Familial hypobetalipoproteinemia
Mutation
Apo B-25: Deletion of 694 bp, including all of exon 21
Apo B-29: C->T transition at cDNA nucleotide 4,125
Apo B-31: Deletion of cDNA nucleotide 4,480
Apo B-37: Deletion of cDNA nucleotides 5,391-5,394
Apo B-39: Deletion of cDNA nucleotide 5,591
Apo B-40: Deletion of cDNA nucleotides 5,693 and
5,694
Apo B-46: C->T transition at cDNA nucleotide 6,381
Apo B-50: C->T transition at cDNA nucleotide 6,963
Apo B-86: Deletion of cDNA nucleotide 11,840
Apo B-87: Deletion of cDNA nucleotide 12,032
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Apo B-89: Deletion of cDNA nucleotide 12,309
Comments
Predicted to yield apo B isoform of 1,085 amino acids, but none
was detected in plasma.137
Predicted to yield apo B isoform of 1,305 amino acids, but none
was detected in plasma.128
Apo B-31 (1,425 amino acids) present in HDL and d> 1.21 g/ml
fraction.44
Apo B-37 (1,728 amino acids) present in VLDL, LDL, and
HDL.47,52,100,165
Apo B-39 (1,799 amino acids) present in VLDL and LDL.128
Apo B-40 (1,829 amino acids) present in VLDL, LDL, and
HDL.133,134
Apo B-46 (2,057 amino acids) present in VLDL, LDL, and
HDL.38
Apo B-50 (2,252 amino acids) present in VLDL.131,132
Apo B-86 (3,896 amino acids) present in VLDL and LDL. This
allele appears to yield normal amounts of apo B-48.68
Apo B-87 (3,978 amino acids) present in VLDL and LDL; apo
B-87 binds with increased affinity to LDL receptor.'66
Apo B-89 (4,039 amino acids) present in VLDL and LDL. Apo
B-89 binds with increased affinity to LDL receptor.133'34
High cholesterol levels-Familial defective apolipoprotein B-100
Mutation associated with hypercholesterolemia and LDL that is
G->A transition at cDNA nucleotide 10,708, resulting
in Arg-+Gln substitution at amino acid 3,500
defective in binding to LDL receptor.'45-147'50
cDNA, complementary DNA; HDL, high density lipoproteins; VLDL, veiy low density lipoproteins; LDL, low density lipoproteins.
Indeed, several investigators have reported data suggesting that heterozygotes may be protected from the
development of atherosclerosis and may therefore
have increased longevity.124'125 Thus, a single copy of
an apo B gene containing a premature stop codon
might be beneficial for a North American adult with
a diet rich in saturated fats and cholesterol!
It is rare that an individual inherits two copies of a
defective apo B allele, causing hypobetalipoproteinemia. The clinical phenotype of familial hypobetalipoproteinemia homozygotes (or compound heterozygotes) is quite variable, and the phenotype probably
depends on whether the mutant apo B alleles are
incapable of producing any apo B (null apo B alleles)
or whether one or both of the mutant alleles are
capable of producing small amounts of a full-length
or truncated apo B. When a subject inherits two null
apo B alleles, the clinical syndrome is often severe
and essentially indistinguishable from that of abetalipoproteinemia.123 Abetalipoproteinemia is a rare
autosomal recessive syndrome in which apo B is
apparently synthesized in the liver126 but in which
there is an apparent defect in the synthesis and
secretion of apo B-containing lipoproteins. The
molecular basis for abetalipoproteinemia is
unknown, but genetic studies have shown that it does
not result from a defect in the apo B gene.127 Familial
hypobetalipoproteinemia homozygotes that possess
two null alleles, as well as subjects with abetalipoproteinemia, have no conventional VLDL and LDL
particles in their plasma; they typically have very low
total cholesterol levels (<50 mg/dl) and triglyceride
levels of only a few milligrams per deciliter.123 These
subjects have a variety of significant clinical problems, including steatorrhea and mental retardation.123 Most of the medical problems are caused by
the inability to synthesize chylomicrons and failure to
absorb dietary fats and fat-soluble vitamins. An
appropriate diet and treatment with the fat-soluble
vitamins, in particular vitamin E, can favorably influence many of these clinical problems.123 Several
investigators have described familial hypobetalipoproteinemia homozygotes who had symptoms of
intestinal malabsorption despite the apparent ability
of at least one apo B allele to synthesize trace
amounts of apo B; in these cases, the small amount of
apo B synthesis was apparently inadequate for intestinal fat absorption.128,129
In several cases, familial hypobetalipoproteinemia
homozygotes or compound heterozygotes have been
essentially asymptomatic. One such case was identified and characterized by Steinberg and coworkers130
in 1979. They investigated a 67-year-old man who had
extremely low levels of LDL cholesterol (<3 mg/dl)
but normal triglyceride levels. Notably, the triglyceride levels increased normally after a fat-rich meal.
Small amounts of apo B were detected in the man's
triglyceride-rich lipoproteins. He had no symptoms of
fat malabsorption, although metabolic ward studies
did document slightly increased amounts of fat in the
stool. This constellation of findings was recognized as
unique; it was quite obvious that he did not have the
clinically severe form of the aforementioned homozygous hypobetalipoproteinemia. In 1986 Young, Witz-
1586
Circulation Vol 82, No 5, November 1990
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
tum, and coworkers47100 showed that the patient and
two of his siblings were compound heterozygotes for
familial hypobetalipoproteinemia, possessing one
mutant apo B allele yielding small amounts of a
truncated apo B protein (apo B-37) and a second
mutant apo B allele producing small amounts of apo
B-100.47100 Recent studies have strongly suggested
that the latter allele yields small amounts of apo
B-100, small amounts of a truncated apo B species
(apo B-86), and normal amounts of apo B-48.68 Apo
B-100, apo B-48, and apo B-37 were all prominent
protein components of the patient's d<1.006 g/ml
fraction, both after a fat-rich meal and after an
overnight fast. The ability of both the patient's
mutant apo B alleles to produce apo B, thereby
participating in the formation of triglyceride-rich
lipoproteins, explains the normotriglyceridemic phenotype and the absence of clinically significant fat
malabsorption. In 1981 Malloy and coworkers131
reported on and extensively characterized another
subject with normotriglyceridemic hypobetalipoproteinemia. This subject was homozygous for a mutant
apo B allele that yielded a truncated apo B species
(apo B-50).132 That mutant allele also produced apo
B-48. Both apo B-48 and apo B-50 were present in
the subject's VLDL fraction, and the subject had no
fat malabsorption.132 Krul and coworkers133,134 have
recently described a family in which three siblings
were compound heterozygotes for familial hypobetalipoproteinemia. These subjects were normotriglyceridemic and had no symptoms of fat malabsorption.
In these subjects, one mutant apo B allele produced
a truncated apo B species (apo B-40), and the second
yielded another truncated apo-B species (apo B-89).
Thus, the generative capacity of the mutant apo B
alleles to produce small amounts of apo B (even a
truncated isoform) that can be assembled into VLDL
and chylomicron particles seems to be critical for the
asymptomatic, normotriglyceridemic phenotype.
The causes of low plasma levels of LDL cholesterol
in nonsense and frameshift mutations of the apo B
gene are incompletely understood, and the explanation may prove to be different for the different apo B
mutations. A consistent finding with the mutations
yielding truncated apo B isoforms is that the absolute
concentration of the truncated apo B protein in
plasma is very low, in the range of 2-10% of the
concentration of apo B-100 produced by a normal
apo B allele. Apo B-37 is present in very low levels
(<3 mg/dl) in the plasma of affected subjects, despite
a demonstration that it is incapable of binding to the
LDL receptor.52 Similarly, apo B-31 and apo B-46
are present in low levels in plasma, although they
lack the region of apo B that binds to the LDL
receptor. The inability of the mutant lipoproteins to
bind to the LDL receptor might be expected to cause
them to accumulate in plasma and have a high rather
than low concentration. This logic has produced a
hypothesis that the low plasma levels of the truncated
apo B species may be explained by low rates of
synthesis and secretion from hepatocytes and entero-
cytes. Recent studies by our laboratory on the postprandial lipoproteins of apo B-46 heterozygotes
could be interpreted as supporting this hypothesis.
After a fat-rich meal, there was a significant increase
in the amount of apo B-48 in the plasma chylomicrons and VLDL subfractions of apo B-46 heterozygotes, but there was no detectable increase in the
amount of apo B-46 (S. Young and D. Chappell,
unpublished observations). In an earlier study involving normal volunteers, our laboratory showed that
both apo B alleles are normally expressed in the
intestine and that both equally contribute to the
production of chylomicrons and VLDL found in the
blood after a fat-rich meal.'35 Failure to observe an
increase in the amount of apo B-46 in plasma after a
fat-rich meal strongly implies the possibility of
decreased secretion of apo B-46 by the intestine.
However, this evidence is not definitive because one
could propose that the low levels of apo B-46 are a
consequence of extremely rapid catabolism of apo
B-46-containing chylomicrons. Ultimately, lipoprotein turnover studies and careful study of the synthesis and secretion of truncated apo B isoforms in cell
culture are necessary to understand the metabolism
of the truncated apo B species and why these apo B
mutations cause low plasma cholesterol levels. Graham and coworkers45 have recently reported cell
culture expression of apo B-39, and we have
expressed two other truncated apo B species (apo
B-31 and apo B-37) found in families with hypobetalipoproteinemia (Z. Yao, M. Linton, S. Young, B.
Blackhart, B. McCarthy, unpublished observations).
However, the efficiency of synthesis, intracellular
transport, and secretion of these truncated proteins
compared with that of the full-length proteins have
yet to be studied. The mechanism for low LDL and
apo B concentrations in the apo B-87 and apo B-89
mutations almost certainly involves another factor. In
vitro fibroblast binding studies have demonstrated
that unlabeled lipoprotein particles containing these
longer truncated isoforms compete more effectively
with 125I-LDL for binding to the LDL receptor compared with unlabeled apo B-100-containing LDL
particles.134136 Thus, the low level of these truncated
isoforms in plasma may be explained in part by
increased LDL receptor-mediated clearance of the
mutant lipoproteins.
Mutations occurring within the portion of the apo
B gene encoding the aminoterminal 30% of the
protein have not been associated with detectable
levels of a truncated apo B protein in plasma in two
reported cases.'28'137 The explanation for this finding
has yet to be investigated, but the finding is consistent
with two different hypotheses: 1) The length of these
very short apo B isoforms falls below the threshold
length required for synthesis and secretion of apo B
from cells, and 2) these isoforms are normally
secreted from cells, yet are removed very rapidly
from plasma. Recently, Graham and coworkers45 as
well as Blackhart and coworkers (B. Blackhart, Z.
Yao, B. McCarthy, personal communication) have
Young Advances in Understanding Apolipoprotein B
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
observed synthesis and secretion of very short apo B
isoforms (apo B-18) in cultured cells that had been
transfected with eukaryotic expression vectors containing 5' fragments of the apo B cDNA. The secretion of short apo B isoforms in cell culture suggests
that the absence of very short apo B isoforms (apo
B-25 and apo B-29) in patients' plasma may not
result from decreased synthesis and secretion of
these proteins. However, the cell culture studies
performed to date are certainly not conclusive.
Recently, Baserga and coworkers138 have shown that
5' nonsense mutations in the ,B-globin gene are
associated with low levels of ,-globin mRNA.
Although the basis of this interesting observation is
incompletely understood, it is conceivable that some
5' nonsense mutations in the apo B gene are associated with low levels of apo B mRNA, resulting in very
low synthetic rates for very short apo B species.
To date, the apo B mutations causing hypobetalipoproteinemia have been nonsense mutations or
deletions that interrupt the reading frame of the apo
B message. However, it is likely that other types of
apo B gene mutations will be discovered that reduce
the synthesis of a full-length apo B-100. As noted
above, several investigators have reported subjects
with hypobetalipoproteinemia who possess an apo B
allele associated with markedly reduced amounts of a
full-length apo B-100.100128,'29 Gavish and coworkers12' have recently examined an abnormal apo B
allele associated with low plasma cholesterol levels
and low levels of the apo B-100 allotype produced by
that allele. They characterized this type of abnormal
apo B allele by using immunoassays with monoclonal
antibody MB19, an antibody that binds to apo B
allotypes MB191 and MB192 with markedly different
affinities.97"135 They identified hypocholesterolemic
individuals who were heterozygotes for the MB19
polymorphism yet had a preponderance of the MB192
allotype in their plasma. Ordinarily, the amount of
allotypes MB191 and MB192 in the plasma of MB19
heterozygotes is roughly equal.97 The metabolic basis
for the low cholesterol levels and the relatively lower
amount of allotype MB191 in the plasma of the
Gavish et al subjects is yet to be determined; their
study did not address the issue of whether the
allotype MB191 was synthesized at a reduced rate or
the allotype was removed from plasma more rapidly.
It is probable that this phenotype might ultimately be
the result of an allele that yields reduced levels of
mRNA and therefore reduced levels of the apo B
allotype produced by that allele. This type of abnormal apo B allele might prove to be analogous to
mutant ,B-globin alleles in P+thalassemia, a disease
in which f-globin is made but in reduced amounts.
,3+Thalassemia has been shown to result from a
variety of mutation types, including intron-exon splicing errors, which result in low 83-globin mRNA
levels.'39
The frequency of mutant apo B alleles that cause
hypobetalipoproteinemia in the general population is
unknown, and
an
exact frequency will be difficult to
1587
determine. Several investigators' estimation of the
frequency of the familial "low cholesterol" phenotype is around 0.1-0.8% of the population,14014' but
it is possible that occasional subjects with the phenotype have defects in other gene products.'42 Another
difficulty in the estimation of the frequency of apo B
defects that cause low cholesterol levels is that some
subjects with heterozygous hypobetalipoproteinemia
do not have abnormal total cholesterol levels. For
example, two apo B-46 heterozygotes had total cholesterol levels of 179 and 186 mg/dl, respectively,
levels that are within the normal range.46 Familial
hypobetalipoproteinemia heterozygotes are
undoubtedly susceptible to the same "polygenic"
factors that result in broad variation of cholesterol
levels in the general population. In rare cases, a
hypobetalipoproteinemia heterozygote can inherit
another gene that raises plasma cholesterol levels.
Hegele and coworkers'43 reported a subject who
inherited single copies of defective apo B and LDL
receptor genes. Her lipoprotein profile was in the
normal range because the hypercholesterolemic
effects of the defective LDL receptor gene were
compensated for by the hypocholesterolemic effects
of the defective apo B allele. In preliminary studies,
Gavish et al'44 have examined a number of MB19
heterozygotes and found that apo B alleles associated
with a reduced plasma level of apo B allotype MB191
might be relatively common in the general population. If apo B alleles significantly affecting the
amount of an apo B allotype in the plasma are
common, then they might help explain the broad
variation of plasma cholesterol levels in the general
population, helping to make some physiological sense
out of the observations that certain apo B RFLPs are
associated with elevated or reduced cholesterol
levels.
Mutation in Apo B Gene Associated With High
Blood Cholesterol Levels
As reviewed above, the majority of LDL particles
are removed from plasma by the interaction of apo B
with the cellular LDL receptor. Goldstein, Brown,
and coworkers5 demonstrated that the gene for mutations in the LDL receptor protein can retard the
clearance of LDL and lead to an accumulation of
LDL particles in the plasma. A defect in the LDL
receptor protein presents a dual difficulty for LDL
metabolism. Not only is there delayed receptormediated clearance of LDL particles from plasma,
but the rate of LDL particle production is increased.
LDL production rate is increased because the apo
E-mediated hepatic uptake of LDL precursors (e.g.,
VLDL particles and IDL particles) is delayed as a
result of decreased LDL receptor activity; consequently, an increased fraction of these precursor
particles is metabolized to LDL. Most patients with
familial hypercholesterolemia have total cholesterol
levels well above 300 mg/dl. For years lipoprotein
investigators speculated that a specific inherited
defect in apo B could abolish its ability to bind to the
LDL receptor and therefore cause delayed LDL
1588
Circulation Vol 82, No 5, November 1990
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
clearance and an accumulation of LDL particles in
the plasma. Such a defect would not be expected to
produce the same degree of increase in the plasma
LDL concentration as that observed in familial
hypercholesterolemia. Although the removal of LDL
from the plasma would be retarded because of the
defect in the apo B protein on LDL particles, the
predicted production rate of LDL would be essentially normal because the apo E-mediated removal of
LDL precursor particles would be predicted to proceed at a normal rate.
Over the past 3 years, a genetic defect in apo B
that interferes with its capacity to bind to the LDL
receptor has been characterized. The seminal observations were made by Vega and Grundy145 during
lipoprotein turnover studies that were designed to
understand the metabolic basis for moderate hypercholesterolemia.145 For their study, they selected
subjects with total cholesterol levels of 250-300
mg/dl, deliberately excluding subjects with overt clinical signs of familial hypercholesterolemia. In metabolic ward studies, they compared the fractional
catabolic rate of each patient's autologous LDL with
that of homologous LDL obtained from normolipidemic control subjects. In five subjects, the clearance
of the patient's own LDL was significantly slower
than that of the control LDL, resulting in the
authors' speculation that these patients had abnormal LDL that was not removed at a normal rate by
the LDL receptor. Later, in collaboration with Vega
and Grundy, Innerarity and coworkers146 examined
the ability of these five subjects' LDL to bind to the
LDL receptor of cultured fibroblasts. They found
that the LDL of one of the five subjects showed
defective binding to the LDL receptor, having only
about 32% of the normal receptor-binding activity.
Studies with partially delipidated LDL showed an
identical binding defect, strongly suggesting that the
defect resided in the protein moiety of LDL particles
(apo B). An identical binding defect was identified in
the LDL of several hypercholesterolemic family
members; Innerarity and coworkers proposed that all
of the affected family members were heterozygotes
for a defect in apo B that prevented its interaction
with the LDL receptor, and they designated the
disorder familial defective apo B-100.
Subsequently, Soria and coworkers147 cloned and
sequenced a large portion of both apo B alleles from
the patient in the above studies and demonstrated
that the DNA sequence of the two alleles differed by
only a few nucleotides. One of the nucleotide substitutions, which had not been previously reported in
any of the prior apo B sequencing efforts, resulted in
a substitution of Gln for Arg at apo B-100 amino acid
residue 3,500 (Table 2). Additional studies revealed
the same mutation in all of the patient's affected
family members but not in any of those unaffected.
The amino acid 3,500 substitution has subsequently
been identified in 10 additional hypercholesterolemic
unrelated subjects whose LDL was defective in binding to the LDL receptor in fibroblast binding studies.
The substitution has never been identified in subjects
whose LDL binds normally to the LDL receptor.148
These data alone strongly suggest that the observed
amino acid substitution at residue 3,500 causes the
binding defect. The location of the amino acid substitution certainly adds weight to the argument that
the mutation is causative; the substitution occurs
close to the epitopes for many monoclonal antibodies
that block binding of LDL to the LDL receptor
(Figure 3). In fact, recent studies have revealed that
one of these antibodies (MB47)149 binds to the LDL
of familial defective apo B-100 heterozygotes with
altered affinity compared with its binding to normal
LDL.150 Thus, the current data strongly implicate the
amino acid 3,500 mutation in the defective binding of
the phenotype. Definitive proof of this will ultimately
require the characterization of the apo B-containing
lipoproteins produced by cells transfected with a
human apo B-100 expression vector and the lipoproteins produced by cells transfected with an apo B
expression vector differing only in the codon for
amino acid residue 3,500.
To date, all subjects with familial defective apo
B-100 have been heterozygous for the mutation;
typical total cholesterol levels for these subjects are
250-300 mg/dl-elevated but lower than that of
familial hypercholesterolemia heterozygotes.148 However, a recent study from Europe has demonstrated
that some affected subjects may have the extremely
elevated cholesterol levels characteristic of patients
with familial hypercholesterolemia.151 As a result of
the high cholesterol levels, it is most probable that
these individuals will have an increased incidence of
atherosclerotic disease; however, the exact frequency
of premature atherosclerosis in subjects with this
disorder relative to the frequency in normal subjects
and other hypercholesterolemic populations remains
to be defined. The frequency of the amino acid
substitution at the residue 3,500 mutation is unknown,
but the screening of miscellaneous populations has
suggested that the frequency might be in the same
range as that reported for the sum of all the LDL
receptor gene mutations - approximately one in every
500 people.148
The amino acid 3,500 mutation, as in several of the
mutations causing hypobetalipoproteinemia, is
caused by a mutation in a CG dinucleotide. CG
dinucleotides are mutational hot spots in higher
animals, and base substitutions in CG dinucleotides
are important in the pathogenesis of many inherited
diseases in humans.46,147 An important question
about the amino acid 3,500 mutation is whether this
mutation occurred independently in different human
populations, as has been documented with specific
thalassemia mutations,139 or only once in a single
"founder" and then subsequently propagated among
subgroups in the general population. Ludwig and
McCarthy152 have performed apo B haplotyping studies with the amino acid 3,500 mutation on nine
unrelated subjects and their families and have determined that this mutation is invariably found on the
Young Advances in Understanding Apolipoprotein B
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
same apo B haplotype. These data suggest that the
mutation has probably occurred only once. If there
was only one founder for the amino acid 3,500
mutation, then it is probable that future studies will
reveal that this mutation will occur with greater
regularity in selected countries or ethnic populations.
The region of the apo B molecule interacting with
the LDL receptor may be quite large (Figure 3); thus,
it seems probable that other apo B amino acid
substitutions could interfere with its ability to bind to
the LDL receptor, thereby producing the familial
defective apo B-100 phenotype. However, to date no
other apo B mutation except the amino acid 3,500
mutation has been reported associated with hypercholesterolemia and defective LDL binding.
Recently, Ladias and coworkers described a subject
with elevated apo B levels and coronary artery disease who had a single nucleotide substitution in the
apo B gene that resulted in a substitution of Trp for
Arg at apo B-100 amino acid residue 4,019.153 However, in a family study, the mutation did not segregate
with elevated apo B levels; thus, the mutation could
not account for the clinical findings. Furthermore,
the LDL from the proband tended to have slightly
increased uptake by fibroblasts rather than the
diminished uptake characteristic of the LDL isolated
from subjects with familial defective apo B-100.
Summary
For the past 5 years, investigators from many
different laboratories have contributed to a greatly
increased understanding of two very important lipidcarrying proteins in plasma-apo B-100 and apo
B-48. Apo B-100, an extremely large protein composed of 4,536 amino acids, is synthesized by the liver
and is crucial for the assembly of triglyceride-rich
VLDL particles. Apo B-100 is virtually the only
protein of LDL, a cholesteryl ester-enriched class of
lipoproteins that are metabolic products of VLDL.
The apo B-100 of LDL serves as a ligand for the LDL
receptor-mediated uptake of LDL particles by the
liver and extrahepatic tissues. The LDL receptorbinding region of apo B-100 is located in the carboxyterminal portion of the molecule, whereas its lipidbinding regions appear to be broadly dispersed
throughout its length. Apo B-48 contains the aminoterminal 2,152 amino acids of apo B-100 and is
produced by the intestine as a result of editing of a
single nucleotide of the apo B mRNA, which changes
the codon specifying apo B-100 amino acid 2,153 to a
premature stop codon. Apo B-48 has an obligatory
structural role in the formation of chylomicrons;
therefore, its synthesis is essential for absorption of
dietary fats and fat-soluble vitamins. Both apo B-48
and apo B-100 are encoded on chromosome 2 by a
single gene that contains 29 exons and 28 introns.
An elevated level of apo B-100 in the plasma is a
potent risk factor for developing premature atherosclerotic disease. In the past 3 years, many different
apo B gene mutations that affect the concentrations
of both apo B and cholesterol in the plasma have
1589
been characterized. A missense mutation in the
codon for apo B-100 amino aid 3,500 is associated
with hypercholesterolemia. This mutation results in
poor binding of apo B-100 to the LDL receptor,
thereby causing the cholesteryl ester-enriched LDL
particles to accumulate in the plasma. This disorder
is called familial defective apo B-100, and it is
probably a cause of premature atherosclerotic disease. Familial hypobetalipoproteinemia is a condition associated with abnormally low levels of apo B
and cholesterol; affected individuals may actually
have a reduced risk of atherosclerotic disease. Hypobetalipoproteinemia can be caused by a number of
different apo B gene mutations that interfere with
the translation of a full-length apo B-100; the syndrome can also result from apo B alleles that yield
reduced amounts of full-length apo B-100. Existing
evidence suggests that neither familial defective apo
B-100 nor familial hypobetalipoproteinemia is particularly rare; thus, apo B mutations may help explain a
small portion of the variation in serum cholesterol
levels in the general population. Future research will
undoubtedly uncover many more apo B gene mutations affecting cholesterol levels.
Acknowledgments
I thank the many friends and collaborators in the
lipoprotein metabolism and arteriosclerosis field for
their support and encouragement, C. Yang and R.
Milne for permission to use their figures in this
review, T. Gridley for manuscript preparation, T.
Rolain for graphics, and A. Averbach and S. Gullatt
Seehafer for the provision of editorial comments on
the completed manuscript.
References
1. Kane JP, Hardman DA, Paulus HE: Heterogeneity of apolipoprotein B: Isolation of a new species from human chylomicrons. Proc Natl Acad Sci USA 1980;77:2465-2469
2. Consensus Conference: Lowering blood cholesterol to prevent heart disease. JAMA 1985;253:2080-2086
3. Olofsson SO, Bjursell G, Bostrom K, Carlsson P, Elovson J,
Protter AA, Reuben MA, Bondjers G: Apolipoprotein B:
Structure, biosynthesis and role in the lipoprotein assembly
process. Atherosclerosis 1987;68:1-17
4. Havel RJ, Hamilton RL: Hepatocytic lipoprotein receptors
and intracellular lipoprotein catabolism. Hepatology 1988;8:
1689-1704
5. Brown MS, Goldstein JL: A receptor-mediated pathway for
cholesterol homeostasis. Science 1986;232:34-47
6. Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum
JL: Beyond cholesterol: Modifications of low-density lipoprotein that increase its atherogenicity. N Engl J Med 1989;
320:915-924
7. Utermann G: The mysteries of lipoprotein(a). Science 1989;
246:904-910
8. Havel RJ, Kane JP: Structure and metabolism of plasma
lipoproteins, in Scriver CR, Beaudet AL, Sly WS, Valle D
(eds): The Metabolic Basis of Inherited Disease, ed 6. New
York, McGraw-Hill Book Co, 1989, pp 1129-1138
9. Elovson J, Chatterton JE, Bell GT, Schumaker VN, Reuben
MA, Puppione DL, Reeve JR Jr, Young NL: Plasma very low
density lipoproteins contain a single molecule of apolipoprotein B. J Lipid Res 1988;29:1461-1473
10. Dullaart RPF, Speelberg B, Schuurman HJ, Milne RW,
Havekes LM, Marcel YL, Geuze HJ, Hulshof MM, Erkelens
1590
11.
12.
13.
14.
15.
16.
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
Circulation Vol 82, No 5, November 1990
DW: Epitopes of apolipoprotein B-100 and B-48 in both liver
and intestine: Expression and evidence for local synthesis in
recessive abetalipoproteinemia. J Clin Invest 1986;78:
1397-1404
Nagata Y, Chen J, Cooper AD: Role of low density lipoprotein receptor-dependent and -independent sites in binding
and uptake of chylomicron remnants in rat liver. J Biol Chem
1988;263:15151-15158
Kowal RC, Herz J, Goldstein JL, Esser V, Brown MS: Low
density lipoprotein receptor-related protein mediates uptake
of cholesteryl esters derived from apoprotein E-enriched
lipoproteins. Proc NatlAcad Sci USA 1989;86:5810-5814
Mahley RW, Hui DY, Innerarity TL, Beisiegel U: Chylomicron remnant metabolism: Role of hepatic lipoprotein receptors in mediating uptake. Arteriosclerosis 1989;9(suppl I):
I-14-1-18
Hui DY, Innerarity TL, Milne RW, Marcel YL, Mahley RW:
Binding of chylomicron remnants and ,8-very low density
lipoproteins to hepatic and extrahepatic lipoprotein receptors: A process independent of apolipoprotein B48. J Biol
Chem 1984;259:15060-15068
Grundy SM, Mok HYI: Chylomicron clearance in normal and
hyperlipidemic man. Metabolism 1976;25:1225-1239
Stalenhoef AFH, Malloy MJ, Kane JP, Havel RJ: Metabolism of apolipoproteins B-48 and B-100 of triglyceride-rich
lipoproteins in normal and lipoprotein lipase-deficient
humans. Proc Natl Acad Sci USA 1984;81:1839-1843
Simons LA, Dwyer T, Simons J, Bernstein L, Mock P, Poonia
NS, Balasubramaniam S, Baron D, Branson J, Morgan J, et
al: Chylomicrons and chylomicron remnants in coronary
artery disease: A case-control study. Atherosclerosis 1987;
65:181-189
Zilversmit DB: Atherogenesis: A Postprandial phenomenon.
Circulation 1979;60:473-485
Janus ED, Nicoll AM, Turner PR, Magill P, Lewis B: Kinetic
bases of the primary hyperlipidaemias: Studies of apolipoprotein B turnover in genetically defined subjects. Eur J Clin
Invest 1980;10:161-172
Kesaniemi YA, Witztum JL, Steinbrecher UP: Receptormediated catabolism of low density lipoprotein in man:
Quantitation using glucosylated low density lipoprotein. J
Clin Invest 1983;71:950-959
Steinberg D: Lipoproteins and atherosclerosis: A look back
and a look ahead. Arteriosclerosis 1983;3:283-301
Schonfeld G, Lees RS, George PK, Pfleger B: Assay of total
plasma apolipoprotein B concentration in human subjects. J
Clin Invest 1974;53:1458-1467
Young SG, Smith RS, Hogle DM, Curtiss LK, Witztum JL:
Two new monoclonal antibody-based enzyme-linked assays
of apolipoprotein B. Clin Chem 1986;32:1484-1490
Fisher WR, Hammond MG, Mengel MC, Warmke GL: A
genetic determinant of the phenotypic variance of the molecular weight of low density lipoprotein. Proc Natl Acad Sci
USA 1975;72:2347-2351
Sniderman A. Shapiro S, Marpole D, Skinner B, Teng B,
Kwiterovich PO Jr: Association of coronary atherosclerosis
with hyperapobetalipoproteinemia [increased protein but
normal cholesterol levels in human plasma low density (,1)
lipoproteins]. Proc Natl Acad Sci USA 1980;77:604-608
Teng B, Thompson GR, Sniderman AD, Forte TM, Krauss
RM, Kwiterovich PO Jr: Composition and distribution of low
density lipoprotein fractions in hyperapobetalipoproteinemia, normolipidemia, and familial hypercholesterolemia.
Proc Natl Acad Sci USA 1983;80:6662-6666
Cianflone KM, Sniderman AD, Walsh MJ, Vu HT, Gagnon J,
Rodriguez MA: Purification and characterization of acylation
stimulating protein. J Biol Chem 1989;264:426-430
Sniderman AD, Wolfson C, Teng B, Franklin FA, Bachorik
PS, Kwiterovich PO Jr: Association of hyperapobetalipoproteinemia with endogenous hypertriglyceridemia and atherosclerosis. Ann Intem Med 1982;97:833-839
Knott TJ, Pease RJ, Powell LM, Wallis SC, Rall SC Jr,
Innerarity TL, Blackhart B, Taylor WH, Marcel Y, Milne R,
et al: Complete protein sequence and identification of struc-
30.
31.
32.
33.
34.
tural domains of human apolipoprotein B. Nature 1986;
323:734-738
Chen SH, Yang CY, Chen PF, Setzer D, Tanimura M, Li
WH, Gotto AM Jr, Chan L: The complete cDNA and amino
acid sequence of human apolipoprotein B-100. J Biol Chem
1986;261:12918-12921
Cladaras C, Hadzopoulou-Cladaras M, Nolte RT, Atkinson
D, Zannis VI: The complete sequence and structural analysis
of human apolipoprotein B-100: Relationship between apoB100 and apoB-48 forms. EMBO J 1986;5:3495-3507
Law SW, Grant SM, Higuchi K, Hospattankar A, Lackner K,
Lee N, Brewer HB Jr: Human liver apolipoprotein B-100
cDNA: Complete nucleic acid and derived amino acid
sequence. Proc Natl Acad Sci USA 1986;83:8142-8146
Boerwinkle E, Chan L: A three codon insertion/deletion
polymorphism in the signal peptide region of the human
apolipoprotein B (APOB) gene directly typed by the polymerase chain reaction. Nucleic Acids Res 1989;17:4003
De Loof H, Rosseneu M, Yang CY, Li WH, Gotto AM Jr,
Chan L: Human apolipoprotein B: Analysis of internal
repeats and homology with other apolipoproteins. J Lipid Res
1987;28:1455-1465
35. Baker ME: Is vitellogenin an ancestor of apolipoprotein
B-100 of human low-density lipoprotein and human lipoprotein lipase? Biochem J 1988;255:1057-1060
36. Vauhkonen M: Complex-type carbohydrates of apolipoprotein-B of human plasma low-density lipoproteins. Glycoconjugate J 1986;3:35-43
37. Yang CY, Gu ZW, Weng SA, Kim TW, Chen SH, Pownall
HJ, Sharp PM, Liu SW, Li WH, Gotto AM Jr, et al: Structure
of apolipoprotein B-100 of human low density lipoproteins.
Arteriosclerosis 1989;9:96-108
38. Hoeg JM, Meng MS, Ronan R, Demosky SJ Jr, Fairwell T,
Brewer HB Jr: Apolipoprotein B synthesized by Hep G2 cells
undergoes fatty acid acylation. J Lipid Res 1988;29:1215-1220
39. Huang G, Lee DM, Singh S: Identification of the thiol ester
linked lipids in apolipoprotein B. Biochemistry 1988;
27:1395-1400
40. Weisgraber KH, Rall SC Jr: Human apolipoprotein B100
heparin-binding sites. J Biol Chem 1987;262:11097-11103
41. Yang C, Kim TW, Weng S, Lee B, Yang M, Gotto AM Jr:
Isolation and characterization of disulfide linkages of apolipoprotein B-100 in low density lipoprotein. Circulation
1989;80(suppl II):II-180
42. Coleman RD, Kim TW, Gotto AM Jr, Yang C: Determination of cysteine on low-density lipoproteins using the fluorescent probe, 5-iodoacetamidofluoresceine. Biochem Biophys
Acta 1990;1037:129-132
43. Chen GC, Hardman DA, Hamilton RL, Mendel CM, Schilling JW, Zhu S, Lau K, Wong JS, Kane JP: Distribution of
lipid-binding regions in human apolipoprotein B-100. Biochemistry 1989;28:2477-2484
44. Young SG, Hubl ST, Smith RS, Snyder SM, Terdiman JF:
Familial hypobetalipoproteinemia caused by a mutation in
the apolipoprotein B gene that results in a truncated species
of apolipoprotein B (B-31): A unique mutation that helps to
define the portion of the apolipoprotein B molecule required
for the formation of buoyant, triglyceride-rich lipoproteins. J
Clin Invest 1990;85:933-942
45. Graham LD, Pullinger CR, Jones TC, Knott TJ, Scott J:
Expression and secretion of apolipoprotein (apo) B constructs by HepG2 cells. Circulation 1989;80(suppl II):II-466
46. Young SG, Hubl ST, Chappell DA, Smith RS, Claiborne F,
Snyder SM, Terdiman JF: Familial hypobetalipoproteinemia
associated with a mutant species of apolipoprotein B (B46).
N Engl J Med 1989;320:1604-1610
47. Young SG, Bertics SJ, Curtiss LK, Witztum JL: Characterization of an abnormal species of apolipoprotein B, apolipoprotein B-37, associated with familial hypobetalipoproteinemia. J Clin Invest 1987;79:1831-1841
48. Yang C, Kim TW, Pao Q, Chan L, Knapp RD, Gotto AM Jr,
Pownall HJ: Structure and conformational analysis of lipidassociating peptides of apolipoprotein B-100 produced by
trypsinolysis. J Prot Chem 1989;8:689- 699
Young Advances in Understanding Apolipoprotein B
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
49. Phillips ML, Schumaker VN: Conformation of apolipoprotein B after lipid extraction of low density lipoproteins
attached to an electron microscope grid. J Lipid Res 1989;
30:415-422
50. Atkinson D: Electron microscopy of unstained frozen
hydrated low density lipoprotein (LDL). Circulation 1988;
78(suppl II):II-287
51. Milne R, Th6olis R Jr, Maurice R, Pease RJ, Weech PK,
Rassart E, Fruchart JC, Scott J, Marcel YL: The use of
monoclonal antibodies to localize the low density lipoprotein
receptor-binding domain of apolipoprotein B. J Biol Chem
1989;264:19754-19760
52. Young SG, Peralta FP, Dubois BW, Curtiss LK, Boyles JK,
Witztum JL: Lipoprotein B37, a naturally occurring lipoprotein containing the amino-terminal portion of apolipoprotein
B100, does not bind to the apolipoprotein B,E (low density
lipoprotein) receptor. J Biol Chem 1987;262:16604-16611
53. Yang CY, Chen SH, Gianturco SH, Bradley WA, Sparrow JT,
Tanimura M, Li WH, Sparrow DA, DeLoof H, Rosseneu M,
et al: Sequence, structure, receptor-binding domains and
internal repeats of human apolipoprotein B-100. Nature 1986;
323:738-742
54. Weisgraber KH, Innerarity TL, Mahley RW: Role of the
lysine residues of plasma lipoproteins in high affinity binding
to cell surface receptors on human fibroblasts. J Biol Chem
1978;253:9053-9062
55. Mehrabian M, Sparkes RS, Mohandas T, Klisak IJ, Schumaker VN, Heinzmann C, Zollnan S, Ma Y: Human apolipoprotein B: Chromosomal mapping and DNA polymorphisms of hepatic and intestinal species. Somat Cell Mol
Genet 1986;12:245-254
56. Huang LS, Miller DA, Bruns GAP, Breslow JL: Mapping of
the humanAPOB gene to chromosome 2p and demonstration
of a two-allele restriction fragment length polymorphism.
Proc Natl Acad Sci USA 1986;83:644-648
57. Law SW, Lee N, Monge JC, Brewer HB Jr, Sakaguchi AY,
Naylor SL: Human apoB-00 gene resides in the p23->pter
region of chromosome 2. Biochem Biophys Res Commun
1985;131:1003-1012
58. Knott TJ, Rall SC Jr, Innerarity TL, Jacobson SF, Urdea MS,
Levy-Wilson B, Powell LM, Pease RJ, Eddy R, Nakai H, et
al: Human apolipoprotein B: Structure of carboxylterminal
domains, sites of gene expression, and chromosomal localization. Science 1985;230:37-43
59. Blackhart BD, Ludwig EH, Pierotti VR, Caiati L, Onasch
MA, Wallis SC, Powell L, Pease R, Knott TJ, Chu ML, et al:
Structure of the human apolipoprotein B gene. J Biol Chem
1986;261:15364-15367
60. Knott TJ, Wallis SC, Pease RJ, Powell LM, Scott J: A
hypervariable region 3' to the human apolipoprotein B gene.
Nucleic Acids Res 1986;14:9215-9216
61. Huang LS, Breslow JL: A unique AT-rich hypervariable
minisatellite 3' to the apoB gene defines a high information
restriction fragment length polymorphism. J Biol Chem 1987;
262:8952-8955
62. Ludwig EH, Friedl W, McCarthy BJ: High-resolution analysis of a hypervariable region in the human apolipoprotein B
gene. Am J Hum Genet 1989;45:458-464
63. Boerwinkle E, Xiong W, Fourest E, Chan L: Rapid typing of
tandemly repeated hypervariable loci by the polymerase
chain reaction: Application to the apolipoprotein B 3' hypervariable region. Proc Natl Acad Sci USA 1989;86:212-216
64. Das HK, Leff T, Breslow JL: Cell type-specific expression of
the human apoB gene is controlled by two cis-acting regulatory regions. J Biol Chem 1988;263:11452-11458
65. Levy-Wilson B, Fortier C, Blackhart B, McCarthy BJ: DNase
I- and micrococcal nuclease-hypersensitive sites in the human
apolipoprotein B gene are tissue specific. Mol Cell Biol
1988;8:71-80
66. Metzger S, Leff T, Breslow JL: Nuclear factors AF-1 and
C/EBP modulate apo B transcriptional activity in hepatic
cells. Circulation 1989;80(suppl II):II-265
1591
67. Levy-Wilson B, Fortier C: Tissue-specific undermethylation
of DNA sequences at the 5' end of the human apolipoprotein
B gene. J Biol Chem 1989;264:9891-9896
68. Linton MF, Pierotti VR, Hubl ST, Young SG: An apo-B gene
mutation causing familial hypobetalipoproteinemia analyzed
by examining the apo-B cDNA amplified from the fibroblast
RNA of an affected subject. Clin Res 1990;38:286A
69. Chelly J, Condordet JP, Kaplan JC, Kahn A: Illegitimate
transcription: Transcription of any gene in any cell type. Proc
Natl Acad Sci USA 1989;86:2617-2621
70. Lusis AJ, Taylor BA, Quon D, Zollman S, LeBoeuf RC:
Genetic factors controlling structure and expression of apolipoproteins B and E in mice. J Biol Chem 1987;262:
7594-7604
71. Kroon PA, DeMartino JA, Thompson GM, Chao YS: Molecular cloning of partial cDNAs for rabbit liver apolipoprotein
B and the regulation of its mRNA levels by dietary cholesterol. Proc NatlAcad Sci USA 1986;83:5071-5075
72. Matsumoto A, Aburatani H, Shibasaki Y, Kodama T, Takaku
F, Itakura H: Cloning and regulation of rat apolipoprotein B
mRNA. Biochem Biophys Res Commun 1987;142:92-99
73. Sorci-Thomas M, Wilson MD, Johnson FL, Williams DL,
Rudel LL: Studies on the expression of genes encoding
apolipoproteins B100 and B48 and the low density lipoprotein receptor in nonhuman primates: Comparison of dietary
fat and cholesterol. J Biol Chem 1989;264:9039-9045
74. Pullinger CR, North JD, Teng BB, Rifici VA, Ronhild de
Brito AE, Scott J: The apolipoprotein B gene is constitutively
expressed in HepG2 cells: Regulation of secretion by oleic
acid, albumin, and insulin, and measurement of the mRNA
half-life. J Lipid Res 1989;30:1065-1077
75. Davidson NO, Drewek MJ, Gordon JI, Elovson J: Rat
intestinal apolipoprotein B gene expression: Evidence for
integrated regulation by bile salt, fatty acid, and phospholipid
flux. J Clin Invest 1988;82:300-308
76. Bostrom K, Boren J, Wettesten M, Sjdberg A, Bondjers G,
Wiklund 0, Carlsson P, Olofsson SO: Studies on the assembly of apo B-100-containing lipoproteins in HepG2 cells. J
Biol Chem 1988;263:4434-4442
77. Yao Z, Vance DE: The active synthesis of phosphatidylcholine is required for very low density lipoprotein secretion
from rat hepatocytes. J Biol Chem 1988;263:2998-3004
78. Craig WY, Nutik R, Cooper AD: Regulation of apoprotein
synthesis and secretion in the human hepatoma Hep G2: The
effect of exogenous lipoprotein. J Biol Chem 1988;
263:13880-13890
79. Borchardt RA, Davis RA: Intrahepatic assembly of very low
density lipoproteins: Rate of transport out of the endoplasmic reticulum determines rate of secretion. J Biol Chem
1987;262:16394-16402
80. Chen SH, Habib G, Yang CY, Gu ZW, Lee BR, Weng SA,
Silberman SR, Cai SJ, Deslypere JP, Rosseneu M, et al:
Apolipoprotein B-48 is the product of a messenger RNA
with an organ-specific in-frame stop codon. Science 1987;
238:363-366
81. Powell LM, Wallis SC, Pease RJ, Edwards YH, Knott TJ,
Scott J: A novel form of tissue-specific RNA processing
produces apolipoprotein-B48 in intestine. Cell 1987;
50:831-840
82. Hospattankar AV, Higuchi K, Law SW, Meglin N, Brewer
HB Jr: Identification of a novel in-frame translational stop
codon in human intestine apoB mRNA. Biochem Biophys Res
Commun 1987;148:279-285
83. Hardman DA, Protter AA, Schilling JW, Kane JP: Carboxylterminal analysis of human B-48 protein confirms the novel
mechanism proposed for chain termination. Biochem Biophys
Res Commun 1987;149:1214-1219
84. Davidson NO, Powell LM, Wallis SC, Scott J: Thyroid
hormone modulates the introduction of a stop codon in rat
liver apolipoprotein B messenger RNA. J Biol Chem 1988;
263:13482-13485
85. Tennyson GE, Sabatos CA, Higuchi K, Meglin N, Brewer HB
Jr: Expression of apolipoprotein B mRNAs encoding higher-
1592
Circulation Vol 82, No 5, November 1990
and lower-molecular weight isoproteins in rat liver and
intestine. Proc NatlAcad Sci USA 1989;86:500-504
86. Glickman RM, Rogers M, Glickman JN: Apolipoprotein B
synthesis by human liver and intestine in vitro. Proc NatlAcad
Sci USA 1986;83:5296-5300
87. Baum CL, Teng B, Davidson N: Differential regulation of
hepatic apolipoprotein gene expression and apo B100 mRNA
editing following alterations in hepatic lipogenesis. Clin Res
1990;38:482A
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
88. Driscoll DM, Wynne JK, Wallis SC, Scott J: An in vitro
system for the editing of apolipoprotein B mRNA. Cell
1989;58:519-525
89. Davies MS, Wallis SC, Driscoll DM, Wynne JK, Williams
GW, Powell LM, Scott J: Sequence requirements for apolipoprotein B RNA editing in transfected rat hepatoma cells.
JBiol Chem 1989;264:13395-13398
90. Chen SH, Li X, Liao WSL, Wu JH, Chan L: RNA editing of
apolipoprotein B mRNA: Sequence specificity determined by
in vitro coupled transcription editing. J Biol Chem 1990;
265:6811-6816
91. Bostrom K, Lauer SJ, Poksay KS, Garcia Z, Taylor JM,
Innerarity TL: Apolipoprotein B48 RNA editing in chimeric
apolipoprotein EB mRNA. J Biol Chem 1989;
264:15701-15708
92. Davies MS, Shah RR, Wallis SC, Driscoll DM, Wynne JK,
Powell LM, Scott J: Sequences required for apolipoprotein
(apo)-B mRNA editing bind specific proteins in rat hepatoma extracts. Circulation 1989;80(suppl II):II-466
93. Scott J, Wallis SC, Davies MS, Wynne JK, Powell LM,
Driscoll DM: RNA editing: A novel mechanism for regulating lipid transport from the intestine. Gut Festschrift 1989;
30:35-43
94. Butler R: The Ag-system, a genetic polymorphism of human
low density lipoprotein, and its application in forensic serology, in Lee HC, Gaensslen RE (eds): Advances in Forensic
Science. Chicago, Mosby Yearbook, vol III, 1990, pp 135-159
95. Berg K, Powell LM, Wallis SC, Pease R, Knott TJ, Scott J:
Genetic linkage between the antigenic group (Ag) variation
and the apolipoprotein B gene: Assignment of the Ag locus.
Proc Natl Acad Sci USA 1986;83:7367-7370
96. Curtiss LK, Edgington TS: Immunochemical heterogeneity of
human plasma apolipoprotein B: I. Apolipoprotein B binding
of mouse hybridoma antibodies. J Biol Chem 1982;
257:15213-15221
97. Young SG, Bertics SJ, Curtiss LK, Casal DC, Witztum JL:
Monoclonal antibody MB19 detects genetic polymorphism in
human apolipoprotein B. Proc Natl Acad Sci USA 1986;
83:1101-1105
98. Tikkanen MJ, Ehnholm C, Kovanen PT, Butler R, Young
SG, Curtiss LK, Witztum JL: Detection of two apolipoprotein B species (apo B, and Bg) by a monoclonal antibody.
Atherosclerosis 1987;65:247-256
99. Robinson MT, Schumaker VN, Butler R, Berg K, Curtiss LK:
Ag(c): Recognition by a monoclonal antibody. Arteriosclerosis
1986;6:341-344
100. Young SG, Bertics SJ, Curtiss LK, Dubois BW, Witztum JL:
Genetic analysis of a kindred with familial hypobetalipoproteinemia. Evidence for two separate gene defects: One
associated with an abnormal apolipoprotein B species, apolipoprotein B-37; and a second associated with low plasma
concentrations of apolipoprotein B-100. J Clin Invest 1987;
79:1842-1851
101. Schlapfer P, Nydegger T, Butler-Brunner E, Morgenthaler
JJ, Butler R, Blaser K: Two monoclonal antibodies that
discriminate between allelic variants of human low density
lipoprotein. Hybridoma 1987;6:575-588
102. Young SG, Hubl ST: AnApaLI restriction site polymorphism
is associated with the MB19 polymorphism in apolipoprotein
B. J Lipid Res 1989;30:443-449
103. Wang X, Schlapfer P, Ma Y, Butler R, Elovson J, Schumaker
VN: Apolipoprotein B: The Ag(al/d) immunogenetic polymorphism coincides with a T-to-C substitution at nucleotide
1981, creating an Alu I restriction site. Arteriosclerosis 1988;
8:429-435
104. Ludwig EH, Blackhart BD, Pierotti VR, Caiati L, Fortier C,
Knott T, Scott J, Mahley RW, Levy-Wilson B, McCarthy BJ:
DNA sequence of the human apolipoprotein B gene. DNA
1987;6:363-372
105. Lusis AJ: Genetic factors affecting blood lipoproteins: The
candidate gene approach. J Lipid Res 1988;29:397-429
106. Law A, Powell LM, Brunt H, Knott TJ, Altman DG, Rajput
J, Wallis SC, Pease RJ, Priestley LM, Scott J, et al: Common
DNA polymorphism within coding sequence of apolipoprotein B gene associated with altered lipid levels. Lancet
1986;1:1301-1303
107. Talmud PJ, Barni N, Kessling AM, Carlsson P, Darnfors C,
Bjursell G, Galton D, Kirk VWAH, Hayden MR, Humphries
SE: Apolipoprotein B gene variants are involved in the
determination of serum cholesterol levels: A study in normoand hyperlipidemic individuals. Atherosclerosis 1987;67:81-89
108. Berg K: DNA polymorphism at the apolipoprotein B locus is
associated with lipoprotein level. Clin Genet 1986;30:515-520
109. Leren TP, Berg K, Hjermann I, Leren P: Further evidence
for an association between the XbaI polymorphism at the
apolipoprotein B locus and lipoprotein level. Clin Genet
1988;34:347-351
110. Aalto-Setala K, Tikkanen MJ, Taskinen MR, Nieminen M,
Holmberg P, Kontula K: XbaI and c/g polymorphisms of the
apolipoprotein B gene locus are associated with serum
cholesterol and LDL-cholesterol levels in Finland. Atherosclerosis 1988;74:47-54
111. Aalto-Setala K, Gylling H, Helve E, Kovanen P, Miettinen
TA, Turtola H, Kontula K: Genetic polymorphism of the
apolipoprotein B gene locus influences serum LDL cholesterol level in familial hypercholesterolemia. Hum Genet 1989;
82:305-307
112. Deeb S, Failor A, Brown BG, Brunzell JD, Albers JJ,
Motulsky AG: Molecular genetics of apolipoproteins and
coronary heart disease. Cold Spring Harb Symp Quant Biol
1986;51:403-409
113. Aburatani H, Matsumoto A, Itoh H, Yamada N, Murase T,
Takaku F, Itakura H: A study of DNA polymorphism in the
apolipoprotein B gene in a Japanese population. Atherosclerosis 1988;72:71-76
114. Hegele RA, Huang LS, Herbert PN, Blum CB, Buring JE,
Hennekens CH, Breslow JL: Apolipoprotein B-gene DNA
polymorphisms associated with myocardial infarction. N Engl
J Med 1986;315:1509-1515
115. Darnfors C, Wiklund 0, Nilsson J, Gerard B, Carlsson P,
Johansson S, Bondjers G, Bjursell G: Lack of correlation
between the apolipoprotein B XbaI polymorphism and blood
lipid levels in a Swedish population. Atherosclerosis 1989;
75:183-188
116. Rajput-Williams J, Knott TJ, Wallis SC, Sweetnam P, Yarnell
J, Cox N, Bell GI, Miller NE, Scott J: Variation of
apolipoprotein-B gene is associated with obesity, high blood
cholesterol levels, and increased risk of coronary heart
disease. Lancet 1988;2:1442-1446
117. Monsalve MV, Young R, Jobsis J, Wiseman SA, Dhamu S,
Powell JT, Greenhalgh RM, Humphries SE: DNA polymorphisms of the gene for apolipoprotein B in patients with
peripheral arterial disease. Atherosclerosis 1988;70:123-129
118. Young SG, Bertics SJ, Scott TM, Dubois BW, Beltz WF,
Curtiss LK, Witztum JL: Apolipoprotein B allotypes MB191
and MB192 in subjects with coronary artery disease and
hypercholesterolemia. Arteriosclerosis 1987;7:61-65
119. Friedl W, Ludwig EH, Paulweber B, Sandhofer F, McCarthy
BJ: Hypervariability in a minisatellite 3' of the apolipoprotein B gene in patients with coronary heart disease compared
with normal controls. J Lipid Res 1990;31:659-665
120. Ferns GAA, Robinson D, Galton DJ: DNA haplotypes of the
human apoprotein B gene in coronary atherosclerosis. Hum
Genet 1988;81:76-80
121. Gavish D, Brinton EA, Breslow JL: Heritable allele-specific
differences in amounts of apoB and low-density lipoproteins
in plasma. Science 1989;244:72-76
Young Advances in Understanding Apolipoprotein B
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
122. Cai SP, Kan YW: Identification of the multiple ,B-thalassemia
mutations by denaturing gradient gel electrophoresis. J Clin
Invest 1990;85:550-553
123. Malloy MJ, Kane JP: Hypolipidemia. Med Clin North Am
1982;66:469-484
124. Kahn JA, Glueck CJ: Familial hypobetalipoproteinemia:
Absence of atherosclerosis in a postmortem study. JAMA
1978;240:47-48
125. Glueck CJ, Gartside PS, Mellies MJ, Steiner PM: Familial
hypobeta-lipoproteinemia: Studies in 13 kindreds. Trans
Assoc Am Physicians 1977;90:184-203
126. Ross RS, Gregg RE, Law SW, Monge JC, Grant SM, Higuchi
K, Triche TJ, Jefferson J, Brewer HB Jr: Homozygous
hypobetalipoproteinemia: A disease distinct from abetalipoproteinemia at the molecular level. J Clin Invest 1988;
81:590-595
127. Talmud PJ, Lloyd JK, Muller DPR, Collins DR, Scott J,
Humphries S: Genetic evidence from two families that the
apolipoprotein B gene is not involved in abetalipoproteinemia. J Clin Invest 1988;82:1803-1806
128. Collins DR, Knott TJ, Pease RJ, Powell LM, Wallis SC,
Robertson S, Pullinger CR, Milne RM, Marcel YL,
Humphries SE, et al: Truncated variants of apolipoprotein B
cause hypobetalipoproteinaemia. Nucleic Acids Res 1988;
16:8361-8375
129. Berger GMB, Brown G, Henderson HE, Bonnici F: Apolipoprotein B detected in the plasma of a patient with homozygous hypobetalipoproteinaemia: Implications for aetiology. J
Med Genet 1983;20:189-195
130. Steinberg D, Grundy SM, Mok HYI, Turner JD, Weinstein
DB, Brown WV, Albers JJ: Metabolic studies in an unusual
case of asymptomatic familial hypobetalipoproteinemia with
hypoalphalipoproteinemia and fasting chylomicronemia. J
ClGn Invest 1979;64:292-301
131. Malloy MJ, Kane JP, Hardman DA, Hamilton RL, Dalal KB:
Normotriglyceridemic abetalipoproteinemia: Absence of the
B-100 apolipoprotein. J Clin Invest 1981;67:1441-1450
132. Hardman DA, Pullinger CR, Kane JP, Malloy MJ: Molecular
defect in normotriglyceridemic abetalipoproteinemia. Circulation 1989;80(suppl II):II-466
133. Talmud P, King-Underwood L, Krul E, Schonfeld G,
Humphries S: The molecular basis of truncated forms of
apolipoprotein B in a kindred with compound heterozygous
hypobetalipoproteinemia. J Lipid Res 1989;30:1773-1779
134. Krul ES, Kinoshita M, Talmud P, Humphries SE, Turner S,
Goldberg AC, Cook K, Boerwinkle E, Schonfeld G: Two
distinct truncated apolipoprotein B species in a kindred with
hypobetalipoproteinemia. Arteriosclerosis 1989;9:856-868
135. Young SG, Bertics SJ, Scott TM, Dubois BW, Curtiss LK,
Witztum JL: Parallel expression of the MB19 genetic polymorphism in apoprotein B-100 and apoprotein B-48: Evidence that both apoproteins are products of the same gene.
J Biol Chem 1986;261:2995-2998
136. Gabelli C, Baggio G, Bilato C, Martini S, Previato L, Marcel
Y, Corsini A, Crepaldi G: Identification of a new apoB
variant associated with hypobetalipoproteinemia. Circulation
1989;80(suppl II):II-276
137. Huang LS, Ripps ME, Korman SH, Deckelbaum RJ, Breslow
JL: Hypobetalipoproteinemia due to an apolipoprotein B
gene exon 21 deletion derived by Alu-Alu recombination. J
Biol Chem 1989;264:11394-11400
138. Baserga SJ, Benz EJ Jr: Nonsense mutations in the human
(3-globin gene affect mRNA metabolism. Proc Natl Acad Sci
USA 1988;85:2056-2060
139. Weatherall DJ: The thalassemias: Molecular pathogenesis, in
Bunn H, Forget BG (eds): Hemoglobin: Molecular, Genetic,
and Chemical Aspects. Philadelphia, WB Saunders Co, 1986,
pp 223-321
140. Andersen GE, Brokhattingen K, Lous P: Familial hypobetalipoproteinaemia in 9 children diagnosed as the result of cord
blood screening for hypolipoproteinaemia in 10,000 Danish
newborns. Arch Dis Child 1979;54:691-694
141. Laskarzewski PM, Khoury P, Morrison JA, Kelly K, Mellies
MJ, Glueck CJ: Prevalence of familiar hyper- and hypoli-
1593
poproteinemias: The Princeton School District Family Study.
Metabolism 1982;31:558-577
142. Hobbs HH, Leitersdorf E, Leffert CC, Cryer DR, Brown MS,
Goldstein JL: Evidence for a dominant gene that suppresses
hypercholesterolemia in a family with defective low density
lipoprotein receptors. J Clin Invest 1989;84:656-664
143. Hegele RA, Emi M, Wu L, Hopkins PN, Lalouel JM,
Williams RR: An apolipoprotein B gene mutation normalizes
the lipid profile in a patient with a gene for familial hypercholesterolemia. Clin Res 1989;37:129A
144. Gavish D, Nance WE, Schieken RM, Breslow JL: Frequency
of allele-specific differences in the amount of apo B in
plasma. Clin Res 1989;37:374A
145. Vega GL, Grundy SM: In vivo evidence for reduced binding
of low density lipoproteins to receptors as a cause of primary
moderate hypercholesterolemia. J Clin Invest 1986;
78:1410-1414
146. Innerarity TL, Weisgraber KH, Arnold KS, Mahley RW,
Krauss RM, Vega GL, Grundy SM: Familial defective apolipoprotein B-100: Low density lipoproteins with abnormal
receptor binding. Proc NatlAcad Sci USA 1987;84:6919-6923
147. Soria LF, Ludwig EH, Clarke HRG, Vega GL, Grundy SM,
McCarthy BJ: Association between a specific apolipoprotein
B mutation and familial defective apolipoprotein B-100. Proc
NatlAcad Sci USA 1989;86:587-591
148. Innerarity TL, Mahley RW, Weisgraber KH, Bersot TP,
Krauss RM, Vega GL, Grundy SM, Friedl W, Davignon J,
McCarthy BJ: Familial defective apolipoprotein B100: A
mutation of apolipoprotein B that causes hypercholesterolemia. J Lipid Res 1990;3:1337-1349
149. Young SG, Witztum JL, Casal DC, Curtiss LK, Bernstein S:
Conservation of the low density lipoprotein receptor-binding
domain of apoprotein B: Demonstration by a new monoclonal antibody, MB47. Arteriosclerosis 1986;6:178-188
150. Weisgraber KH, Innerarity TL, Newhouse YM, Young SG,
Arnold KS, Krauss RM, Vega GL, Grundy SM, Mahley RW:
Familial defective apolipoprotein B-100: Enhanced binding
of monoclonal antibody MB47 to abnormal low density
lipoproteins. Proc NatlAcad Sci USA 1988;85:9758-9762
151. Tybjaerg-Hansen A, Gallagher J, Vincent J, Houlston R,
Talmud P, Dunning AM, Seed M, Hamsten A, Humphries
SE, Myant NB: Familial defective apolipoprotein B-100:
Detection in the United Kingdom and Scandinavia, and
clinical characteristics of ten cases. Atherosclerosis 1990;
80:235-242
152. Ludwig EH, McCarthy BJ: Haplotype analysis of the human
apolipoprotein B mutation associated with familial defective
apolipoprotein B100. Am J Hum Genet 1990;47:712-720
153. Ladias JAA, Kwiterovich PO Jr, Smith HH, Miller M,
Bachorik PS, Forte T, Lusis AJ, Antonarakis SE: Apolipoprotein B-100 Hopkins (arginine4019-tryptophan): A new
apolipoprotein B-100 variant in a family with premature
atherosclerosis and hyperapobetalipoproteinemia. JAMA4
1989;262:1980-1988
154. Maeda N, Ebert DL, Doers TM, Newman M, Hasler-Rapacz
J, Attie AD, Rapacz J, Smithies 0: Molecular genetics of the
apolipoprotein B gene in pigs in relation to atherosclerosis.
Gene 1988;70:213-229
155. Huang LS, Ripps ME, Breslow JL: Molecular basis of five
apolipoprotein B gene polymorphisms in noncoding regions.
J Lipid Res 1990;31:71-77
156. Jones T, Rajput-Williams J, Knott TJ, Scott J: An MspI
RFLP in the APOB promoter. Nucleic Acids Res 1989;17:472
157. Ma Y, Wang X, Butler R, Schumaker VN: Bsp 12861
restriction fragment length polymorphism detects Ag(c/g)
locus of human apolipoprotein B in all 17 persons studied.
Arteriosclerosis 1989;9:242-246
158. Darnfors C, Nilsson J, Protter AA, Carlsson P, Talmud PJ,
Humphries SE, Whalstrdm J, Wiklund 0, Bjursell G: RFLPs
for the human apolipoprotein B gene: HincIl and PvuIl.
Nucleic Acids Res 1986;14:7135
159. Priestley L, Knott T, Wallis S, Powell L, Pease R, Brunt H,
Scott J: RFLP for the human apolipoprotein B gene: V;
XbaI. Nucleic Acids Res 1985;13:6793
1594
Circulation Vol 82, No 5, November 1990
160. Huang LS, de Graaf J, Breslow JL: ApoB gene MspI RFLP
in exon 26 changes amino acid 3611 from Arg to Gln. J Lipid
Res 1988;29:63-67
161. Xu C, Nanjee N, Tikkanen MJ, Huttunen JK, Pietinen P,
Butler R, Angelico F, Del Ben M, Mazzarella B, Antonio R,
et al: Apolipoprotein B amino acid 3611 substitution from
arginine to glutamine creates the Ag (h/i) epitope: The
polymorphism is not associated with differences in serum
cholesterol and apolipoprotein B levels. Hum Genet 1989;
82:322-326
162. Ma Y, Schumaker VN, Butler R, Sparkes RS: Two DNA
restriction fragment length polymorphisms associated with
Ag(t/z) and Ag(g/c) antigenic sites of human apolipoprotein
B. Arteriosclerosis 1987;7:301-305
163. Protter AA, Coleman RT, Sato KY, Dillan NA, Lim D,
Frossard PM: Very low nucleotide diversity in the 5' portion
of the human apolipoprotein B gene. Am J Hum Genet
1986;39(suppl):A216
164. Genest JJ Jr, Ordovas JM, Robbins AH, King DC, Frossard
PM, Schaefer EJ: Two new apoB gene polymorphisms: Rsl
and Rs2. Nucleic Acids Res 1988;16:8746
165. Young SG, Northey ST, McCarthy BJ: Low plasma cholesterol levels caused by a short deletion in the apolipoprotein B
gene. Science 1988;241:591-593
166. Tennyson GE, Gabelli C, Baggio G, Bilato C, Brewer HB Jr:
Molecular defect in the apolipoprotein B gene in a patient
with hypobetalipoproteinemia and three distinct apoB species. Clin Res 1990;38:482A
KEY WoRDs lipoproteins * hypercholesterolemia * cholesterol
* lipid metabolism * hypolipidemia
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Recent progress in understanding apolipoprotein B.
S G Young
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Circulation. 1990;82:1574-1594
doi: 10.1161/01.CIR.82.5.1574
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1990 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/82/5/1574
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further
information about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/