Arteriosclerosis, Thrombosis, and Vascular Biology. 2000;20:e63-e67
(Arteriosclerosis, Thrombosis, and Vascular Biology. 2000;20:e63.)
© 2000 American Heart Association, Inc.
|
Atherosclerosis and Lipoproteins |
Oxidation of Apolipoprotein B-100 in Circulating LDL Is Related to LDL Residence Time
In Vivo Insights From Stable-Isotope Studies
Jens Pietzsch;
Peter Lattke;
Ulrich Julius
From the Institute and Polyclinic of Clinical Metabolic Research (J.P.,
U.J.) and the Institute of Clinical Chemistry and Laboratory Medicine (P.L.),
Medical Faculty, Technical University Dresden, Dresden, Germany.
Correspondence to Dr Jens Pietzsch, Institute and Polyclinic of Clinical Metabolic Research, Medical Faculty ‘Carl Gustav Carus,’ Technical University, Fetscherstrasse 74, D-01307 Dresden, Germany. E-mail julius{at}rcs.urz.tu-dresden.de
 |
Abstract
|
|---|
Abstract—5-Hydroxy-2-aminovaleric
acid (HAVA) has been
suggested to be a specific marker of
oxidation of apolipoprotein
(apo) B-100 proline (Pro) and arginine
(Arg) side-chain residues
in low density lipoprotein (LDL) in vitro.
Here we describe
the application of sensitive mass spectrometric
techniques to
the characterization of Pro/Arg-modified apoB-100 in
LDL
1 (S
f 7 to 12) and
LDL
2 (S
f 0 to 7) in vivo.
We studied 7 subjects
with familial defective apoB-100 (FDB) and 8
normolipidemic
controls. In FDB subjects, the presence of a mutant
apoB-100
(FDB
3500Q) in LDL markedly reduced its
affinity for the LDL
receptor, leading to increased residence times
(RTs) of LDL
1 (65±21 versus 32±12 hours,
P<0.005) and LDL
2 (230±40 versus
53±7 hours,
P<0.001) when compared
with controls, as
determined by stable-isotope turnover studies.
LDL
1 HAVA content was not different between the
groups (FDB, 0.004±0.001
mol/mol apoB-100 versus controls,
0.003±0.001 mol/mol
apoB-100,
P=0.200).
LDL
2 HAVA content was higher in FDB subjects
(0.374±0.088
versus 0.013±0.002 mol/mol apoB-100,
P<0.001). In
both groups, LDL
2 HAVA
was positively associated with LDL
2 RT
(FDB,
r=0.893,
P=0.003; controls,
r=0.976,
P=0.000) and negatively
correlated with
LDL
2 
-tocopherol content (FDB,
r=-0.929,
P=0.003;
controls,
r=-0.903,
P=0.002). No significant correlations
could
be found between LDL
1 HAVA,
LDL
1 RT, and
-tocopherol,
respectively.
The low LDL
1 HAVA content observed
in both FDB and control groups
was thought to be due to the relatively
lower RT as well as
the higher
-tocopherol content of
these lipoproteins. In contrast,
LDL
2 seemed to
be strongly prone to direct oxidation of apoB-100
in vivo. The longer
these particles linger in the circulation,
the more apoB-100 Pro/Arg
residues become modified.
Key Words: familial defective apolipoprotein B-100 • lipoproteins • residence time • oxidation • atherosclerosis
|
Introduction
|
|---|
Oxidative modification of LDL lipids and apoB-100 by
reactive
oxygen species (ROS) is widely regarded as a crucial event in
atherogenesis.
1 2 ApoB-100 modifications, eg, binding of
lipid peroxidation
products or direct oxidation of amino acid
side-chain residues,
are thought to finally result in the formation of
new epitopes
that are specifically recognized by scavenger
receptors.
1 2 However, despite convincing evidence from in
vitro and animal
experiments, data concerning the role of oxidized
apoB-100 in
the development of atherosclerosis in
humans in vivo are scarce.
One reason for this is the shortage of
sensitive and specific
methods for direct measurement of oxidized
apoB-100 in circulating
LDL. Recent studies have described the
determination of circulating
autoantibodies to oxidized LDL, the
measurement of specific
stable oxidation products of LDL apoB-100
isolated from human
atherosclerotic tissue (eg, 3-chlorotyrosine or
nitrotyrosine),
or the measurement of stable oxidation products of
tyrosine,
valine, lysine, and phenylalanine side-chain residues of
total
LDL protein obtained from human plasma.
3 4 5 More
recently,
the oxidation of LDL apoB-100 proline (Pro) and arginine
(Arg)
residues primarily to
-glutamyl semialdehyde, which by
reduction
forms 5-hydroxy-2-aminovaleric acid (HAVA), has been measured
in
vitro and in normolipidemic subjects in vivo.
6 However,
the
value of HAVA as a specific marker of LDL apoB-100 oxidation
under
pathophysiological conditions such as
hypercholesterolemia
has not been established.
Oxidative damage of LDL apoB-100 is
believed to substantially occur in
the subendothelial space
of the vessel
wall.
1 During circulation, LDL particles enter
and
reemerge from the subendothelium.
2 7 Under
hypercholesterolemic
conditions, eg, in subjects with
familial defective apoB-100
(FDB), the time needed to remove LDL from
the circulation is
drastically increased.
8 This should
favor an increase in the
number of LDL particles that are exposed to
the subendothelium
and in the duration of LDL apoB-100
exposure to ROS. Furthermore,
compositional changes in LDL particles
possibly result in different
exposures of both lipids and apoB-100 to
ROS.
1 7 9 Thus, individuals
with
hypercholesterolemia not only possess more
circulating
LDL but also have older, modified LDL. In this context, we
hypothesized
that the longer LDL particles linger in the circulation,
the
more Pro and Arg residues of apoB-100 should be modified. To
prove
this hypothesis, the present study combined specific and
sensitive
gas chromatography–mass spectrometry (GC-MS)
methodologies
to measure both HAVA formation and retention times of
native
LDL in vivo in FDB subjects and normolipidemic controls.
|
Methods
|
|---|
Subjects
Seven subjects with heterozygous FDB (3 men and 4 women; 21
to
63 years old) and 8 normolipidemic control subjects (4 men
and 4 women;
20 to 51 years old) volunteered for the study.
All subjects were
nonsmokers and were free of renal, hepatic,
hematological, and thyroid
abnormalities, and all medications
known to affect lipid levels were
discontinued at least 6 weeks
before the study. No subject was taking
antioxidants such as
probucol and vitamins A or E. All subjects were
normoglycemic.
All subjects gave informed, written consent; ethical
approval
was granted by the local ethics committee. Fasting blood
samples
were collected into tubes containing EDTA at a final
concentration
of 0.1%. The blood was centrifuged at 4°C
(2000
g for 10
minutes) to separate cells from plasma. Blood
plasma and LDL
samples were all processed in subdued light to prevent
any photooxidation
of LDL. All buffers and solutions were degassed and
stored under
argon. Buoyant LDL
1
(S
f 7 to 12) and small, dense
LDL
2 (S
f 0
to 7) were
isolated from plasma by a combination of both cumulative
and sequential
gradient ultracentrifugation techniques as previously
described.
8 Total LDL protein was determined by the
bicinchoninic acid
protein assay (Pierce) with bovine serum
albumin as the protein
standard. LDL apoB-100 was measured by
immunoelectrophoresis
(Sebia). LDL cholesterol was
determined enzymatically by using
CHOD-PAP test kits (Roche). The total
carbonyl group content
in LDL apoB-100 and LDL
-tocopherol content were measured as
described elsewhere
and are expressed as mol/mol of apoB-100.
6 9 10
Determination of HAVA
Delipidation of LDL, formation of HAVA by reduction of
-glutamyl semialdehyde with NaBH4, and
enzymatic hydrolysis of apoB-100 with nonspecific bacterial protease
type XIV (Sigma Chemical Co) were performed as previously
described.6 The free amino acids were isolated from
protein hydrolysates, derivatized to their
N(O)-ethoxycarbonyl ethyl ester derivatives, and
analyzed by electron-impact ionization GC-MS by following the
protocol described elsewhere.6 11 LDL HAVA content is
expressed as mol/mol of apoB-100. The intra-assay coefficient of
variation was <4.5%,and the interassay coefficient of variation,
<6.1%.
Determination of LDL ApoB-100 Residence Time
The determination of LDL apoB-100 residence time (RT) in FDB and
control subjects has been described elsewhere.8 In brief,
the stable-isotope tracers used were
L-[ring-13C6]phenylalanine
or L-[5,5,5-2H3]leucine.
After administration of a priming bolus of 550 µg/kg of
[13C6]phenylalanine
([2H3]leucine data in
parentheses; 655 µg/kg), a constant infusion of 12 µg ·
kg-1 · min-1 (16
µg · kg-1 ·
min-1) was continued for 12 hours. Blood samples
were obtained before the priming bolus; at 10-minute intervals for 2
hours; and after 2, 2.5, 3, 3.5, 4, 5, 6, 9, 10, 11, 12, 24, 48, and 72
hours. ApoB-100 of LDL subfractions was separated by preparative
polyacrylamide gel electrophoresis. The stained apoB-100 bands
were excised from gels and hydrolyzed. The free amino acids were
isolated from apoB-100 hydrolysates by cation-exchange
chromatography and then derivatized, and isotopic
enrichment was determined by GC-MS. The kinetic parameters
of LDL apoB-100 metabolism were estimated by
multicompartmental analysis using the
SAAM (simulation analysis and modeling,
version 31) software package as previously
published.8 After fitting the model to the tracer
data, LDL apoB-100 fractional catabolic rates and RTs were determined
with reasonable certainty on the basis of the fractional standard
deviations of the model parameter estimates.8
In our studies, LDLs were fractionated into 2 subclasses of particles:
"buoyant" LDL1 and smaller, more dense
LDL2. Here, the enrichment curves clearly
indicated that the labeling of LDL1 preceded that
of LDL2; hence, they were modeled as precursor
and product, respectively.8
Statistical Analysis
Descriptive data were expressed as arithmetic means±SDs.
Statistical analyses (Mann-Whitney tests, Spearman rank
correlation analysis) were calculated by using the
SPSS 9.0 software package.
|
Results
|
|---|
FDB patients had significantly higher levels of plasma total
cholesterol,
plasma apoB-100, and cholesterol
and apoB-100 levels of LDL
1 and
LDL
2 (Tables I
and II
), whereas the concentrations of plasma
triglycerides
and lipoprotein constituents of VLDL, IDL,
and HDL particles
were not different between the 2 groups (statistical
results
not shown in detail).
8 The mean lipoprotein mass
composition
of LDL
1 and
LDL
2 particles, ie, the mass percentage of
cholesterol
and apoB-100 of total lipoprotein mass, did not
differ between
the 2 groups. Furthermore, the
-tocopherol content in LDL
1 and
LDL
2 particles was similar in the 2 groups.
However, in both groups,
the
-tocopherol content in
LDL
1 was significantly higher when
compared with
that in LDL
2 (
P=0.000). The
cholesterol-rich,
buoyant LDL
1
particles contained
5 molecules of
-tocopherol
per
particle, whereas the cholesterol-poor, dense
LDL
2 particles
contained
1 molecule of
-tocopherol per particle (Table II
).
The HAVA content in total LDL
(LDL
1 plus LDL
2) was
significantly
higher in FDB subjects when compared with controls
(0.063±0.020
versus 0.004±0.001 mol/mol apoB-100,
P<0.001). LDL
1 HAVA content
showed no differences between the 2 groups. LDL
2
particles
in FDB contained significantly more HAVA molecules than did
LDL
2 in controls. Furthermore, the HAVA content
in LDL
2 was higher
when compared with
LDL
1 within and between the 2 groups (Table
III
). The apoB-100 carbonyl group content in total LDL was
somewhat
but not significantly higher in FDB (0.21±0.08 versus
0.18±0.03
mol/mol apoB-100,
P=0.064). The
LDL
1 apoB-100 carbonyl group
content showed no
differences between the 2 groups (0.05±0.02
versus 0.06±0.07 mol/mol
apoB-100,
P=0.090), whereas
LDL
2
particles in FDB contained significantly more carbonyl
groups than did
LDL
2 in controls (0.16±0.04 versus 0.12±0.03,
P<0.01).
The mean RT of LDL subfractions (reciprocal of
fractional catabolic
rate at steady state) derived by
multicompartmental analysis
is presented in Table III
. The other kinetic parameters of
apoB-containing
lipoproteins in the subjects studied have been
published previously
and are not shown in detail.
8
The mean RT of total LDL (LDL
1 plus
LDL
2) apoB-100 was >3-fold higher in FDB when
compared
with controls (147.2±19.3 versus 42.2±3.7 hours;
P<0.001).
In FDB, the mean RT of LDL
1
and LDL
2 particles was increased
2-fold and
>4-fold, respectively, when compared with controls
(Table III
). In both groups, significant, positive correlations
were
found between LDL
2 HAVA and
LDL
2 RT (Figure
I).
In addition,
in both groups, LDL
2 HAVA content
was significantly correlated
in a negative manner with
LDL
2 -tocopherol (Figure
II) and positively
with
LDL
2 apoB-100 carbonyl group content (FDB,
r=0.764,
P=0.012;
controls,
r=0.508,
P=0.030). No significant correlations were
found between
LDL
1 HAVA, LDL
1 RT,
LDL
1 -tocopherol, and
LDL
1 apoB-100 carbonyl group content,
respectively.
View this table:
[in this window]
[in a new window]
|
Table 3. Residence Time and HAVA Content of ApoB-100
in LDL Subfractions of FDB Patients and Normolipidemic Controls
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Figure 1. Figure I. Relationship between LDL2 apoB-100 RT
and LDL2 HAVA content in 7 FDB patients (triangles;
r=0.893, P=0.003) and 8 normolipidemic
controls (squares; r=0.976,
P=0.000).
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Figure 2. Figure II. Relationship between LDL2 apoB-100
-tocopherol content and LDL2 HAVA content in
7 FDB patients (triangles; r= -0.929,
P=0.003) and 8 normolipidemic controls (squares;
r= -0.903, P=0.002).
|
|
|
Discussion
|
|---|
-Glutamyl semialdehyde is a primary oxidation product of
both
Pro and Arg side-chain residues.
5 12 13 By reduction
with NaBH
4,
-glutamyl semialdehyde forms
HAVA.
6 12 13 Recently, HAVA has
been shown to be a
specific marker for apoB-100 oxidation in
vitro and in normolipidemic
subjects in vivo.
6 In that former
study, the level of HAVA
demonstrated in native total LDL obtained
from 10 normolipidemic, young
male volunteers was 0.012±0.004
mol/mol apoB-100 (0.4:10 000,
Pro/Arg). The present work for
the first time reports experiments
with HAVA as a highly specific
and sensitive marker of direct apoB-100
oxidation of circulating
human LDL subfractions under
pathophysiological conditions.
Therefore, as a
model, patients with heterozygous FDB showing
a moderate to severe
hypercholesterolemia were studied. The
risk for
the development of premature ischemic heart disease
is strongly
increased in FDB.
14 The apoB-100 defect (Arg3500Gln)
primarily
affects the fractional catabolism of LDL.
8 15
The in vivo consequence
of this has been shown to be a 2-fold higher RT
of buoyant LDL
1 and a >4-fold higher RT of small
LDL
2 in FDB when compared
with
controls.
8 The longer the RT of LDL, the longer is the
exposure
of its apoB-100 moiety to the attack of ROS. ApoB-100 consists
of
4563 amino acids and has a molecular weight of 516 000 (minus
the
carbohydrate content).
16 ApoB-100 contains 170 Pro and
148
Arg residues that are partially susceptible to direct, oxidative
damage
in vivo.
6 HAVA is suggested to be formed by
metal-catalyzed
oxidation processes.
6 Apparently,
Cu
2+ or Fe
2+ bind to
discrete
sites of apoB-100 and form centers for repeated radical
production.
The exact number of such binding sites is not
known, and values
ranging from 3 to
12 have been reported by
others.
9 17 However,
participation of other
ROS-generating processes, eg, myeloperoxidase
reaction, in the
formation of HAVA is unknown. The present study
shows higher HAVA
levels in total LDL in FDB subjects when compared
with controls
(1.97:10 000 versus 0.13:10 000 Pro/Arg,
P<0.01).
In
controls, LDL
2 HAVA was higher when compared with
LDL
1, but
in total it did not exceed the range
demonstrated in healthy
men elsewhere.
6 In FDB,
apoB-100 of circulating LDL
2 contained
significantly
higher amounts of modified Pro/Arg residues when compared
with
controls. Here, the level of HAVA amounted to
12:10 000
Pro/Arg
residues. There exists a strong association between the
extremely
higher RT of LDL
2 and the oxidative
modification of apoB-100
Pro/Arg in FDB. These findings are
consistent with the increment
in nonspecific carbonyl group
content in LDL
2 apoB-100 in FDB.
In contrast,
HAVA levels in LDL
1 particles were not increased
in
FDB, and no associations could be found with the RT of
LDL
1.
In addition, qualitative changes in the LDL
particles could
render them more or less prone to oxidation. For
instance, LDL
-tocopherol molecules are supposed to be
good competitive substrates
for oxidative attack.
7 9 In
hypercholesterolemic subjects,
LDL
-tocopherol is suggested to be a predictor of LDL
oxidizability.
9 18 In the present study, LDL
-tocopherol levels were not different
between the groups
but were significantly higher in LDL
1 particles
when
compared with LDL
2 particles.
LDL
2 -tocopherol showed a strong,
negative
association with LDL
2 HAVA. Although
there is no significant
relationship between LDL
1
apoB-100 modification and LDL
1
-tocopherol,
the higher content of the antioxidant could
provide an explanation
for the lower extent of HAVA formation in
LDL
1. This is consistent
with data
published by others.
7 9 17 18 In conclusion, because
HAVA
is not a normal constituent of human apolipoproteins, the
overall yield
of HAVA that has been found in LDL
2 apoB-100 is
remarkably
high in FDB and indicates that LDL apoB-100 Pro/Arg residues
are
good targets for oxidative attack under present
pathophysiological
conditions. Our data suggest
that oxidative damage of a particularly
small, more dense,
-tocopherol–poor, and "aged" LDL entity
both in
blood and in the subendothelium may be an important
mechanism
underlying the premature ischemic heart disease in
FDB. However,
additional work is needed to understand the specific
consequences
of
-glutamyl semialdehyde formation for the
metabolic fate of
apoB-containing lipoproteins in vivo.
|
Acknowledgments
|
|---|
The published work is part of the thesis of J.P. for his
postgraduate
study of toxicology and environmental medicine at the
Institute
of Legal Medicine at the Leipzig University,
Germany.
Received June 28, 2000;
accepted August 16, 2000.
|
References
|
|---|
-
Berliner JA, Heinecke JW. The role of oxidized
lipoproteins in atherogenesis. Free Radic Biol Med. 1996;20:707–727.[Medline]
[Order article via Infotrieve]
-
Ross R. Atherosclerosis: an
inflammatory disease. N Engl J Med. 1999;340:115–126.[Free Full Text]
-
Ylä-Herttuala S. Is oxidized low-density
lipoprotein present in vivo? Curr Opin
Lipidol. 1998;9:337–344.[Medline]
[Order article via Infotrieve]
-
Hazen SL, Heinecke JW. 3-Chlorotyrosine, a specific
marker of myeloperoxidase-catalyzed oxidation, is markedly elevated in
low density lipoprotein isolated from human atherosclerotic intima.
J Clin Invest. 1997;99:2075–2081.[Medline]
[Order article via Infotrieve]
-
Davies MJ, Fu S, Wang H, Dean RT. Stable markers of
oxidant damage to proteins and their application in the study of human
disease. Free Radic Biol Med. 1999;27:1151–1163.[Medline]
[Order article via Infotrieve]
-
Pietzsch J. Measurement of 5-hydroxy-2-aminovaleric
acid as a specific marker of iron-mediated oxidation of proline and
arginine side chain residues of low density lipoprotein apolipoprotein
B-100. Biochem Biophys Res Commun. 2000;270:852–857.[Medline]
[Order article via Infotrieve]
-
Walzem RL, Watkins S, Frankel EN, Hansen RJ, German
JB. Older plasma lipoproteins are more susceptible to oxidation: a
linking mechanism for the lipid and oxidation theories of
atherosclerotic cardiovascular disease. Proc Natl
Acad Sci U S A.. 1995;92:7460–7464.[Abstract/Free Full Text]
-
Pietzsch J, Wiedemann B, Julius U, Nitzsche S,
Gehrisch S, Bergmann S, Leonhardt W, Jaross W, Hanefeld M. Increased
clearance of low density lipoprotein precursors in patients with
heterozygous familial defective apolipoprotein B-100: a stable isotope
approach. J Lipid Res. 1996;37:2074–2087.[Abstract]
-
Leonhardt W, Hanefeld M, Schaper F. Diminished
susceptibility to in vitro oxidation of low-density lipoproteins in
hypercholesterolemia: key role of
-tocopherol content.
Atherosclerosis. 1999;144:103–107.[Medline]
[Order article via Infotrieve]
-
Cao G, Cutler RG. Protein oxidation and aging, I:
difficulties in measuring reactive protein carbonyls in tissues using
2,4-dinitrophenylhydrazine. Arch Biochem Biophys. 1995;320:106–114.[Medline]
[Order article via Infotrieve]
-
Pietzsch J, Nitzsche S, Wiedemann B, Julius U,
Leonhardt W, Hanefeld M. Stable isotope ratio analysis of amino
acids: the use of N(O)-ethoxycarbonyl ethyl ester
derivatives and gas chromatography/mass spectrometry.
J Mass Spectrom. 1995;30:S129–S135.
-
Ayala A, Cutler RG. The utilization of
5-hydroxyl-2-aminovaleric acid as a specific marker of oxidized
arginine and proline residues in proteins. Free Radic Biol
Med. 1996;21:65–80.[Medline]
[Order article via Infotrieve]
-
Amici A, Levine RL, Tsai L, Stadtman ER. Conversion of
amino acid residues in proteins and amino acid homopolymers to carbonyl
derivatives by metal-catalyzed oxidation reactions. J Biol
Chem. 1989;264:3341–3346.[Abstract/Free Full Text]
-
Tybjaerg-Hansen A, Steffensen R, Meinertz H, Schnohr P,
Nordestgaard BG. Association of mutations in the apolipoprotein B gene
with hypercholesterolemia and the risk of
ischemic heart disease. N Engl J Med. 1998;338:1577–1584.[Abstract/Free Full Text]
-
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
Natl Acad Sci U S A. 1987;84:6919–6923.[Abstract/Free Full Text]
-
Scott J. The molecular and cell biology of
apolipoprotein-B. Mol Biol Med.. 1989;6:65–80. (SWISS-PROT
data bank entry
P04114—http://www.expasy.ch/cgi-bin/get-sprot-entry?P04114).[Medline]
[Order article via Infotrieve]
-
Esterbauer H, Gebicky J, Puhl H, Jürgens G. The
role of lipid peroxidation and antioxidants in oxidative modification
of LDL. Free Radic Biol Med. 1992;13:341–390.[Medline]
[Order article via Infotrieve]
-
Stalenhoef AFH, Defesche JC, Kleinveld HA, Demacker
PNM, Kastelein JJP. Decreased resistance against in vitro oxidation of
LDL from patients with familial defective apolipoprotein B-100.
Arterioscler Thromb. 1994;14:489–493.[Abstract/Free Full Text]
Top
Abstract
Introduction