1700
Hemin: A Possible Physiological Mediator of
Low Density Lipoprotein Oxidation
and Endothelial Injury
G. Balla, H.S. Jacob, J.W. Eaton, J.D. Belcher, and G.M. Vercellotti
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Oxidized low density lipoprotein (LDL), formed in vivo from presently unknown reactions, may
play a role in atherogenesis. In vitro, transition metals such as iron and copper will facilitate
LDL oxidation, but these metals are unlikely to exist in free form in normal body fluids. We
have explored the possibility that LDL oxidation may be promoted by heme, a physiologically
ubiquitous, hydrophobic, iron-containing compound. Indeed, during several-hour incubation,
heme caused extensive oxidative modification of LDL; however, such modification requires only
minutes in the presence of small amounts of HjO2 or preformed lipid hydroperoxides within the
LDL. Oxidative interactions between heme, LDL, and peroxides lead to degradation of the heme
ring and consequent release of heme iron, which further accelerates heme degradation. Coupled
(evidently iron-catalyzed) heme degradation and LDL oxidation are both effectively inhibited
by hydrophobic antioxidants and iron chelators. That such hemin-induced LDL oxidation may
be involved in atherogenesis is supported by the finding that LDL oxidized by hemin is
extremely cytotoxic to cultured aortic endothelial cells. Overall, these investigations not only
lend support to the idea that LDL oxidation by physiological substances such as heme may play
a role in the process of atherogenesis but also may have broader implications, as similar
oxidative reactions between heme and unsaturated fatty acids may occur consequent to
hemorrhagic injury. (Arteriosclerosis and Thrombosis 1991;ll:1700-1711)
O
xidized low density lipoprotein (LDL) exerts
several potentially atherogenic effects.1-3
Thus, oxidized LDL 1) is chemotactic for
monocytes (but not neutrophils); 2) is rapidly taken up
by macrophages through the "scavenger" receptor on
these cells, reproducing in vitro the appearance of
atherosclerotic plaque foam cells; and 3) is cytotoxic to
endothelial cells. Furthermore, antioxidants that inhibit LDL oxidation prevent fatty streak formation in
rabbits and are associated with protection against
coronary artery disease in population studies. 4 " 6 However, the mechanisms responsible for LDL oxidation
in vivo are by no means clear. Most in vitro studies
demonstrating presumed atherogenic effects of oxidized LDL have employed prolonged ( > 24-hour)
From the Departments of Medicine (G.B., H.SJ., G.M.V.) and
Laboratory Medicine and Pathology (J.W.E.), and the School of
Public Health (J.D.B.), University of Minnesota, Minneapolis,
Minn.
Supported by National Institutes of Health grants RO1 HL33793 and R32 HL-28935, Army research grant DAMD 17-91-Z1009, and Fogarty International Award 1FO5-TW 04106-01.
Address for correspondence: G.M. Vercellotti, MD, University
of Minnesota, Department of Medicine, Division of Hematology,
516 Delaware Street SE, Box 480 UMHC, Minneapolis, MN
55455.
Received May 6, 1991; revision accepted Jury 16, 1991.
incubations of native LDL with distinctly nonphysiological oxidizing systems to produce the altered LDL
capable of provoking foam cell production and chemotaxis of monocytes.
One maneuver often employed for in vitro oxidation
of LDL is exposure to added transition metals, which
will amplify LDL oxidation in serum-free medium.
However, in vivo the two most abundant transition
metals, iron and copper, are unlikely to exist in free
form in normal plasma or interstitial fluids. In the
present work, we have explored the possibility that
hemin may represent a more likely physiological mediator of LDL oxidation in vivo. Hemin is a ubiquitous
iron-containing compound that will spontaneously assort into hydrophobic matrixes. For example, free
hemin rapidly accumulates in the membranes of intact
cells. That is, we have found that the plasma membranes of intact cultured endothelial cells rapidly accumulate hemin and become extraordinarily susceptible
to oxidant damage caused by activated inflammatory
cells or soluble H2O2.7 Moreover, the iron moiety of
hemin is important in this injury; iron-free protoporphyrin or tin-substituted protoporphyrin does not promote endothelial killing. Furthermore, lipid-soluble
iron chelators, such as the 21-amino steroid "lazaroids," completely prevent oxidant injury catalyzed by
hemin incorporation.7
Balla et al Hemin Augments LDL Oxidation
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
In light of these prior observations, we hypothesized that hemin, a physiologically ubiquitous iron
compound, might similarly augment the oxidation
of LDL. In a partial test of this hypothesis, we
found in the present studies not only that hemin
directly promoted oxidative modification of LDL
but also that such oxidation was greatly accelerated
by small amounts of added H2O2 or by endogenous
preformed LDL lipid hydroperoxides. During these
oxidative reactions between hemin, LDL, and peroxides, the heme ring was degraded, with the
resultant release of free iron. This "coupled oxidation" likely occurs within lipid domains because
lipid-soluble antioxidants or iron chelators are
more inhibitory of the process than are watersoluble ones. Perhaps most importantly, hemininduced oxidation produces an LDL molecule that
is cytotoxic for vascular endothelium. Overall, our
results lend support to the idea that LDL oxidation
by more physiological substances such as hemin
may play a role in the process of atherogenesis.
Methods
Materials
Dulbecco's modified Eagle's medium (DMEM),
fetal calf serum, and Hanks' balanced salt solution
(HBSS) were obtained from GIBCO (Grand Island,
N.Y.); minimum essential medium was from Hazleton
(Lenexa, Kan.); collagenase type I was from Worthington (Freehold, NJ.); heparin was from LyphoMed (Melrose Park, 111.); H2O2 was from Fischer
Scientific (Pittsburgh, Pa.); 6% hetastarch (hydroxyethylstarch) was from DuPont (Wilmington, Del.);
deferoxamine was from CIBA-GEIGY (Basel, Switzerland); chromium-51-labeled CrO4 (as the sodium
salt) was from Amersham (Arlington Heights, 111.);
cadmium acetate and cyclohexane were from Aldrich
Chemical Co., Inc. (Milwaukee, Wis.); chloroform was
from EM Science (Gibbstown, N.J.); KBr was from
Fisher Scientific (Fair Lawn, NJ.); FeCl3 and FeSO4
were from Mallinckrodt Inc. (Paris, Ky.); Ar was from
Northern Cryogenics (St. Paul, Minn.); protoporphyrin DC was from Porphyrin Products (Logan, Utah);
bovine hemin type I, butylated hydroxytoluene (BHT),
catalase (EC 1.11.1.6; 14,100 units/mg protein),
cumene hydroperoxide, dimethyl sulfoxide, essentially
fatty acid-free human and bovine serum albumin,
EDTA, ferrozine, fluorescamine, Folin and Ciocalteu's phenol reagents, Af-2-hydroxyethylpiperazineW-2-ethanesulfonic acid (HEPES), human haptoglobin, lauryl sulfate, L-rysine, Percoll, 4/3-phorbol-12myristate-13-acetate (PMA), KI, K 3 Fe(CN) 6 ,
pyridine, 2-thiobarbituric acid, trichloroacetic acid,
Trypan blue, and trypsin type IX were from Sigma
Chemical Company (St. Louis, Mo.); acetic acid was
from Spectrum Chemical Mfg. Corp. (Gardena, Calif.); U74500A was a gift from The Upjohn Company
(Kalamazoo, Mich.); probucol was from Merrell Dow
Pharmaceuticals Inc. (Cincinnati, Ohio); and rat hemopexin was generously provided by Ursula Muller-
1701
Eberhard, The New York Hospital-Cornell Medical
Center, New York, N.Y.
Methods
Preparation of human low density lipoprotein.
Plasma LDL (1.025-1.050 g/ml density) was prepared from EDTA (1 mg/ml)-anticoagulated venous
blood of healthy humans after a 2,000g centrifugation
of blood at 4°C for 20 minutes.8 The plasma density
was adjusted to 1.025 g/ml with a KBr solution of
1.346 g/ml density and ultracentrifuged at 100,000g at
4°C for 24 hours. The centrifuged tubes were sliced
with a rube slicer, and the bottom fraction was saved.
The density of this fraction was then increased from
1.025 g/ml to 1.050 g/ml with a KBr solution. After
another 24-hour ultracentrifugation at 100,000g at
4°C, the tubes were sliced again, and the top fraction
was saved (usually 20 ml from 100 ml blood) and
diafyzed in dialysis tubing (molecular weight, 12,00014,000) against 4 1 of cold, degassed, N2-saturated
HEPES-NaCl (10-150 mM) buffer (pH 7.4) in the
dark for 8 hours, repeated three times. The dialyzed
LDL solution was sterilized by passage through a
Millipore filter (0.2 ^m), stored in a sterile plastic
tube under Ar in the dark at 4°C, and used within 2
weeks. The protein content was determined by the
Peterson-Lowry method with bovine serum albumin
as a standard.9
Oxidation of low density lipoprotein. LDL was oxi-
datively modified in HBSS (supplemented with 4 mM
NaHCO3 to provide physiological pH) for the neutrophil (polymorphonuclear leukocyte [PMN])-mediated oxidation and the porcine aortic endothelial cell
(PAEC) killing experiments. HEPES-NaCl (10-150
mM) buffer (pH 7.4) was used in all other studies.
The oxidation was initiated by H2O2 or by PMAactivated PMNs. Alternatively, 1 mM FeSO4, FeCl3,
and CuSO4 stock solutions were prepared in distilled
water just before the experiments and were used
immediately to catalyze the LDL oxidation. A 1 mM
hemin stock solution was made fresh in 20 mM
NaOH and then diluted to the desired concentrations with the reaction buffer. The protoporphyrin IX
was prepared in the same way as hemin. The LDL
oxidation was performed at 37°C for 3 hours, but in
some experiments different incubation times and
room temperatures were used. The hydrophobic antioxidants, U74500A (a 21-amino steroid lazaroid),
BHT, probucol, and vitamin E, were dissolved in
ethanol to a concentration of 50 mM. For U74500A,
pH 3 was used to improve solubility. These freshly
prepared stock solutions were further diluted to the
desired final concentrations with the reaction buffers.
The final concentration of ethanol was kept below
0.2%, a concentration without any effect in the
system.
Lipid peroxidation assays.
MEASUREMENT OF THIOBARBITURIC A C I D - R E A C TIVE SUBSTANCES. After oxidative modification of
LDL (usually 200 /ig/ml LDL protein concentration), 300 fi\ of the incubation mixture was added to
1702
Arteriosclerosis and Thrombosis
Vol 11, No 6 November/December 1991
TABLE 1. TBABS, Total Lipid Hydroperoxide, and Conjugated Diene Prodnction After 24-Hour Low Density Upoprotein OiidaUon
Parameter
TBARS
(nmol/mg LDL)
Total LOOH
(nmol/mg LDL)
Conjugated dienes
(OD at 234 nm)
LDL (under Ar)
LDL+buffer
LDL+H2O2
LDL+hemin
LDL+hemin+H2O2
LDL+copper
0.31 ±0.10
0.92±0.05*
1.22±0.05t
56.10±0.78*
40.00±1.01*
55.0±5.12$
0.001+0.89
1.24±1.24§
0.89±0.56
151.0±18.86$
209.4±10.94t
209.8± 16.64$
0.165±0.002
1.748±0.0084:
1.907±0.06*
1.840±0.07t
0.160±0.006
0.167±0.006
Results represent mean±SEM of four experiments.
)
Low density lipoprotein (LDL) protein (500 ti%lm\) in HEPES-NaCl (10-150 mM) buffer (pH 7.4) was oxidized with H2O2 (100
hemin (10 nM), or CuSO4 (10 tiM) for 24 hours at 37°C. Lipid peroxidation was analyzed by measuring thiobarbituric acid-reactive
substances (TBARS), total lipid hydroperoxide (LOOH) (iodometric method), and conjugated dienes (optical density [OD] at 234 nm in
cyclohexane after lipid extraction).
HEPES, A^-hydroxyethylpiperazine-AT^-ethanesulfonic acid.
"p<0.01, §p<0.001 compared with LDL under Ar.
tp<0.01, 4p<0.001 compared with buffer-treated LDL.
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
600 /A thiobarbituric acid reagent (0.375 g 2-thiobarbituric acid, 2.08 ml 12N HC1, 15 ml 100% trichloroacetic acid, and distilled water to a final volume of
100 ml). After heating at 100°C for 15 minutes, the
samples were cooled to room temperature and were
centrifuged at 10,000g for 10 minutes. The clear
supernatants were analyzed spectrophotometrically
at 532 nm, using an extinction coefficient of 1.56 x 1$
M"1 • cm"1,10 and the results are presented as nanomoles thiobarbituric acid-reactive substances
(TBARS) per milligram LDL protein.
MEASUREMENT OF TOTAL LIPID HYDROPEROXIDES
BY IODOMETRIC METHOD. LDL samples (1 ml) (usu-
ally 500 Mg/ml LDL protein concentration) were
mixed with 0.1 mM EDTA and vortexed with 3 ml
chilled chloroform/methanol (2:1, vol/vol). A 1-ml
aliquot of the organic layer was evaporated to dryness under N2, layered with Ar, and kept in the dark.
Deaerated acetic acid/chloroform (0.6 ml; 3:2, vol/
vol) was added to each sample at 25°C for dissolving
the extract, and then 40 /A KI (1.2 g/ml in deaerated
distilled water) was added. After a 5-minute incubation in the dark at room temperature, the reaction
was stopped by adding 1.8 ml 20 mM cadmium
acetate. The clear aqueous phase was spectrophotometrically analyzed at 353 nm after centrifugation of
samples at 2,000g for 10 minutes. All readings were
corrected for LDL-free blanks. Results were calculated by using an extinction coefficient of 2.19X104
M"1 • cm"1 for Lf.11-12
MEASUREMENT OF CONJUGATED DIENE FORMATION.
A 0.5-ml aliquot of the organic layer of 500 /Ag/ml
LDL protein of the chloroform/methanol extract was
evaporated to dryness under N2. One milliliter of
cyclohexane was added to each sample, and the
optical density was read at 234 nm. The absorbance
was corrected for LDL-free blanks.13 The conjugated
diene formation in unextracted LDL was measured
by directly reading the optical density at 234 nm. In
these series of experiments the LDL concentration
was 100 Mg/ml in HEPES-NaCl buffer.
Measurement offluorescamine-reactiveamino group
content of low density Upoprotein. Reactive amino
groups in LDL were estimated with fluorescamine
with the use of lysine as a standard.14 After LDL
oxidation, 200 /xl 1 mM fluorescamine in acetone was
added to 800 ^1 LDL sample in HEPES-NaCl buffer
during vortexing. The final LDL protein concentration was 50 /xg/ml. The fluorescence intensity was
determined on a Perkin-Elmer spectrofluorometer at
390 nm excitation and 475 nm emission wavelengths
after 30 minutes' room-temperature incubation in
the dark. LDL-free samples were used as blanks. The
results are expressed as moles reactive amino group
per mole LDL protein. The molecular weight of LDL
protein (apoprotein B-100) was assumed to be
550,000 d.15
Electrophoretic methods. Electrophoretic mobility
was determined by 0.6% agarose gel electrophoresis
in 50 mM barbital buffer (pH 8.6).16 Lipoproteins
were visualized by staining with 0.1% (wt/vol) fat red
7B in 95% methanol.
Hemin and iron determinations.
HEMIN DETERMINATION. Hemin was measured
spectrophotometrically as a hemin-pyridine complex.
Pyridine (0.35 ml) and 0.15 ml 1.0N NaOH were
added to 1.25 ml 200 p,g/ml LDL samples. After
vortexing, the samples were divided into two equal
parts. The first one was oxidized by 25 /xl 3 mM
K3Fe(CN)6, and the second was reduced by 1 mg
dithionite. The absorbances were measured at 541
and 557 nm, respectively, using the oxidized samples
as blanks. For calculation of the results, the differences between optical densities at 541 and 557 nm
were used, with the extinction coefficient of 2.07 x 104
M"1 • cm"1.17-18 The results are given as micromoles
per liter hemin concentration.
IRON DETERMINATION. Iron was measured spectrophotometrically as a ferrozine-iron complex in a
reducing environment. One half milliliter of 200
/ig/ml LDL sample was added to 2 ml iron buffer
reagent (1 M acetate buffer, pH 4.5, contains 3%
lauryl sulfate, 170 mM ascorbic acid, and 5.3 mM
Na2S2O5). After vortexing and a 15-minute incubation at 37°C, the absorbance at 562 nm was measured
for the calculation of the blank value.19 During the
second phase of the procedure, 100 fil ferrozine
reagent (9 mM ferrozine and 328 mM thiourea in
Balla et al
Hemin Augments LDL Oxidation
1703
35-1
30-
FIGURE 1. Line plot of time course of
Hfij-hemin-mediated low density lipoprolan (LDL) lipidperooddation. Oxidation of an LDL sample at 100 uglml
protein concentration in HEPES-NaCl
(10-150 mM) pH 7.4 buffer by 100 uM
HjO2 tn the presence of 5 fiM hemin was
followed at 37°C for 3 hours. At each
time point, TEARS quantity was determined as nanomoles TEARS per milligram LDL protein as described in
"Methods." HEPES, N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid;
TEARS, thiobarbituric acid-reactive
substances.
25-
£
a
20-
I
o
0)
E
<
LDL+Hemlr>+H2O2
15-
LDL+Hemtn
LDL+H2O2
10-
5-
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
20
40
60
80
100
TIME
(mln)
120
140
distilled water) was added to the 2.5-ml samples, and
after another 15 -minute incubation time at 37°C, the
optical density at 562 nm was measured. The iron
concentration was calculated using the extinction
coefficient 2.79X104 M"1 • cm"1.20 The results are
given as micromoles per liter iron concentration.
Hemin degradation as a measure of the induction
intervaltime(lagtime)of low density lipoprotein oxidation. The kinetics of oxidation of LDL were continuously monitored by the absorbance of hemin at an
optical density at 405 nm in a 96-well microtiter plate
reader (Molecular Devices Corp., Menlo Park, Calif.) at room temperature for 4 hours. The LDL
protein concentration was 100 jtg/ml in HEPESNaCl buffer, pH 7.4. The oxidizing agent was 50 fiM
H2O2, and the initial hemin concentration was 2.5
^.M. The lag time for LDL oxidation was measured
as the time required after hemin addition for rapid
hemin degradation to occur (as measured by loss of
absorbance at 405 nm). This lag time represents the
induction phase of LDL oxidation. The results are
expressed as minutes of induction phase. For the
LDL oxidation-inhibition studies, concentrations of
exogenous antioxidants or iron chelators, which were
necessary to increase the interval time of the induction phase (lag time) by twofold or fourfold, were
assessed. The exogenous antioxidants were added to
the LDL samples at 37°C for 15 minutes before
hemin administration.
Preparation of human neutrophils. PMNs were isolated from human volunteers after informed consent
(following the guidelines of the Committee on the
Use of Human Subjects in Research of the University
of Minnesota). Briefly, 40-ml blood samples were
160
180
drawn into plastic syringes containing 20 ml 6%
hetastarch in 0.9% NaCl and 200 units of preservative-free heparin.21 The mixture was allowed to sediment at room temperature for 30-45 minutes, and
the supernatant was collected and centrifuged (400g
for 5 minutes at 4°C). The pellet was resuspended in
• 15 ml ice-cold water, and after 25 seconds, isotonicity
was restored by addition of 3.6% NaCl. This suspension was centrifuged at 400g for 5 minutes, the pellet
was resuspended in 5 ml HBSS, layered on top of
Percoll at 1.075 g/ml specific gravity, and then centrifuged at 20,00% for 30 minutes at 4°C. The resulting PMNs were washed twice again, counted, and
assessed by Trypan blue exclusion for viability.
Greater than 98% viability and less than 1% platelet
contamination were uniformly achieved.
Isolation and culture ofporcine aortic endothelial cells.
PAECs were isolated enzymatically from porcine aorta
by use of type I collagenase (0.2%) for 15 minutes at
3TC.72 Primary cells were grown in DMEM containing
10% (vol/vol) heat-inactivated fetal calf serum, penicillin (100 units/ml), and streptomycin (100 units/ml)
supplemented with L-glutamine at 37°C in 5% CO2.
Endothelial cells were identified by their morphology
and ability to take up acetylated LDL.23 The cells were
subcultured after resuspension by brief exposure to
0.05% trypsin-0.53 mM EDTA and then grown to
confluence in 24-well (2-cm2 wells) culture plates (CoStar, Cambridge, Mass.). The cultures were used from
passages four to 10 and studied within 48 hours of
reaching confluence.
Porcine aortic endothelial cell cytotoxicity assays.
Confluent endothelial cells grown in 24-well (2-cm2/
1704
Arteriosclerosis and Thrombosis
Vol 11, No 6 November/December 1991
HlOt (100pH)
£
I
9
f
BUFFER
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
HEUIN (M<)
B
I
HEWN (SpM)
FIGURE 2. Hemin and Hfi2 dose-response
curves measured by low density lipoprotein
(LDL) TEARS formation. Panel A: LDL (200
fig/ml) in HEPES-NaCl (10-150 mM), pH
7.4, was incubated in the presence of different
hemin concentrations. Controls did not contain
HjO2 (buffer). At the end of a 3-hour incubation, TBARS expressed as nanomoles TEARS
per milligram LDL protein were measured as
described in "Methods."
Results are
mean±SEM of three experiments performed in
duplicate. Panel B: Oxidation of 200 figlml
LDL in HEPES-NaCl (10-150 mM), pH 7.4,
catalyzed by 5 \iM hemin at 37"C in the
presence of different H/D2 concentrations. Controls did not contain hemin (buffer). At the end
of a 3-hour incubation, TBARS were determined as described in "Methods." Results are
mean±SEM of three experiments performed in
duplicate. HEPES, N-2-hydroxyethylpiperaacid; TBARS,
zine-N'-2-ethanesulfonic
thiobarbituric acid-reactive substances.
S
BUFFER
0
10
10
10
40
10
10
70
t0
tO
well) tissue-culture plates were radiolabeled with 2
/iCi/well of [51Cr]Na2CrO4 for 8 hours in cell culture
medium and then washed three times with HBSS.
The control cells were incubated with 500 /il HBSS,
and the others were incubated with normal or oxidatively modified LDL in HBSS. The 8-hour cytotoxicity assays were performed at 37°C in a humidified
atmosphere of 95% air, 5% CO2. At the end of the
incubation the reaction solution was removed, the
monolayer was washed two times with 0.5 ml HBSS,
and the combined washes from each well were centrifuged at l,000g for 10 minutes. The radioactivity of
the supernatants, the pellets, and the NaOH-lysed
(adherent) endothelial cells were measured separately in a gamma counter, generating specific cytotoxicity values, calculated as previously described24;
spontaneous 51Cr release was less than 30% in all
experiments.
100
In the cytotoxicity inhibition experiments, LDL
oxidation was inhibited with different agents during
the 3-hour hemin-H2O2 modification, and the endothelial cell killing assays were performed with
these LDL samples.
Statistics
Significance of differences was determined by
Student's t test.
Results
Hemin Catalyzes Low Density Lipoprotein Oxidation
During long-term incubations, hemin greatly enhanced the oxidation of LDL. As shown in Table 1,
the addition of small amounts of hemin (10 fiM) led
to large increases in the formation of TBARS, total
lipid hydroperoxides, and conjugated dienes after 24
Balla et al
LDL
LDL
F»(I+)
LDL
Fa<3+)
PP-IX
LOL
Hwnln
LDL
Cu
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
FIGURE 3. Bar graph of catalytic effects of iron (Fe) protoporphyrin DC (PP-IX), hemin, and copper (Cu) on low density
lipoprotein (LDL) oxidation. LDL (200 vglml) (asprotein) in
HEPES^laCl (10-150 mM), pH 7.4, was incubated with 100
IxM Hfi2 or control buffer at 37 °C in the presence of 5 pM
FeSOt, FeClh PP-IX, hemin, or CuSO4. At the end of a
3-hour incubation, TBARS were determined as described in
"Methods." Results are mean±SEM of four experiments
performed in duplicate. HEPES, N-2-hydroxyethylpiperazineN'-2-ethanesidfonic acid; TBARS, thiobarbituric acid-reactive substances.
hours of incubation. In fact, in most respects this
hemin-catalyzed LDL oxidation was approximately
equal to that caused by added copper (also in a
concentration of 10 pM). In both cases, near-maximal oxidation of LDL may be occurring; the simultaneous addition of both hemin and H2O2 caused
little increase in LDL oxidation (Table 1). In contrast
with the substantial LDL oxidation caused by addition of either hemin or copper, similar incubations of
the starting material with no additives or with H2O2
alone generated insignificant oxidation (Table 1).
This was probably due to the fact that the LDL used
in these studies was prepared under conditions designed to minimize spontaneous oxidation.
As shown in Figure 1, the kinetics of LDL oxidation by hemin alone were characterized by a lag
phase lasting more than 3 hours (open circles) (as
measured by TBARS formation). In contrast, hemincatalyzed LDL oxidation proceeded quite rapidly
(with a lag time of only 20 minutes) with the coaddition of 100 fiM H2O2 (closed circles). Indeed, in such
short-term incubations the combination of hemin and
H2O2 caused at least 50% as much LDL oxidation as
did the addition of copper (data not shown). This
shortening of the lag time by added H2O2 likely
reflects the coupled nature of the oxidation (LDL
plus hemin plus oxidant). Hemin-H2O2-dependent
LDL oxidation was also dependent on the hemin
concentration (Figure 2A). As little as 1 /xM hemin
significantly increased TBARS production in the
presence of 100 juM H2O2. Similarly, even smaller
concentrations of H2O2 (10 fiM)—in the presence of
5 fiM hemin—were strikingly effective in accelerating
LDL oxidation (Figure 2B).
Mechanism of Hemin-Mediated Low Density
Lipoprotein Oxidation
Although in other systems transition metals markedly enhance lipid peroxidation, chelates of these
Hemin Augments LDL Oxidation
1705
metals are often far less effective. In the present
instance, we suspected that iron, while still bound to
hemin, might not be as active as the free metal in
catalyzing LDL oxidation. Surprisingly, however, as
shown in Figure 3, "free" iron (added as FeSO4 or
FeQ 3 , 5 fiM) did not induce LDL oxidation even in
the presence of H2O2. Furthermore, iron-EDTA or
iron-ADP in similar concentrations did not promote
LDL oxidation (data not shown). These observations
indicate that although free iron is probably an important catalyst of hemin-induced LDL oxidation because iron-deficient protoporphyrin DC is ineffective,
the catalytically active iron is likely delivered by the
hemin into hydrophobic compartments normally inaccessible to externally added free iron or hydrophilic iron chelates. Indeed, subsequent experiments
provided evidence for the breakdown of LDL-associated hemin and the release of heme iron (this
study).
During the oxidation of LDL with hemin and
H2O2, the brownish color typical of hemin gradually
disappeared, reflecting coincident hemin destruction.
In fact, it was previously shown by others that lipid
hydroperoxides can mediate the disintegration of the
hemin ring.25 In the present system, the disintegration of the hemin moiety was made evident by the
decrease in hemin absorption at 412 nm, as shown in
Figure 4A. Practically all hemin disappeared within
30 minutes in incubations containing added H2O2
(solid circles). This rapid destruction of hemin was
not seen in samples containing hemin without added
H2O2. Furthermore, the oxidative destruction of
hemin depended on the simultaneous presence of
LDL and H2O2; in the absence of LDL, hemin was
only slowly destroyed by H2O2 (100 fiM), with only
10% being degraded over 3 hours (data not shown).
These observations imply that H2O2 may promote
hemin-dependent LDL oxidation, at least in part, by
facilitating the oxidative destruction of the hemin
group, leading perhaps to the release of free catalytically active iron. The importance of this degradation
is emphasized by the fact that conjugated diene
formation occurred in parallel with the destruction of
hemin (Figure 4A).
Hemin-dependent LDL oxidation was not absolutely dependent on added H2O2. As shown in Figure
4B, when LDL was minimally modified by storage
(under atmospheric O2) at 4°C, added hemin readily
promoted LDL oxidation even in the absence of
H2O2. In this case, hemin disappeared within minutes, presumably via reaction with preformed LDL
lipid hydroperoxides, and the loss of intact hemin
again coincided with conjugated diene formation.
As might be expected, the (probably oxidative)
destruction of intact hemin results in the release of
free iron (Figure 5). Both hemin destruction and the
attendant release of ferrozine-trappable free iron
evidently involved LDL oxidation because the addition of the antioxidant BHT stabilized the heme ring
and prevented the release of free iron.
1706
Arteriosclerosis and Thrombosis
Vol 11, No 6 November/December 1991
A
r
0.4
-
0.1
I
- e.a
ffi
<
140
1(0
1(0
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
TIME (mln)
B
0J
1.0
ai
FIGURE 4. Panel A: Plot of low density
lipoprotein (LDL) conjugated diene formation
and hemin degradation. Simultaneous hemin
degradation and oxidation of an LDL sample
at 100 ng/ml protein concentration in
HEPES-NaCl (10-150 mM), pH 7.4, was
followed at 37 °C for 3 hours; when appropriate, hemin (5 yJA) and Hfi2 (100 IJM) were
also added At each time point, conjugated
diene formation and hemin disappearance
were determined as described in "Methods."
Panel B: Plot of minimally modified low
density lipoprotein (LDL) conjugated diene
formation and hemin degradation. Hemin (5
yM) was added to minimally modified LDL,
produced by storage of the lipoprotein at 4°C in
HEPES-NaCl (10-150 mM), pH 7.4, under
air for 1 month. During this time, spontaneous
LDL lipid peroxidation caused 4.22±0.9 nmol
TBARS/mg LDL protein and 49.4+8.6 nmol
lipid hydroperoxide/mg LDL protein. Results
represent mean±SEM of seven different LDL
samples. At each time point, conjugated diene
formation and hemin (5 pM) disappearance
were determined as described in "Methods."
HEPES, N-2-hydroxyethylpiperazine- N' -2ethanesulfonic acid; TBARS, thiobarbituric
acid-reactive substances.
T1UI (mln)
The importance of iron released from LDL-associated hemin is further emphasized by the findings
that deferoxamine (a potent hydrophilic iron chelator) and the lazaroid U74500A,26-27 a hydrophobic
antioxidant and weak iron chelator, retarded heminH2O2-mediated lipid peroxidation (Figure 6); consistent with its ability to insert into the hydrophobic
milieu of the LDL particle, the hydrophobic "lazaroid" compound was 10-fold more effective than
deferoxamine in retarding lipid peroxidation.
Because antioxidants and iron chelators seemingly
play an important role in limiting LDL oxidation, a
variety of other drugs and natural antioxidants were
tested for their ability to retard the induction of LDL
oxidation by hemin and hydrogen peroxide. As shown
in Table 2, the lag time necessary for hemin disappearance, closely associated with the induction phase
of LDL lipid peroxidation, was used as a measure of
LDL resistance to oxidation. With this assay, U74500
is by far the most potent protectant, whereas probucol, an antioxidant used in patients with hypercholesterolemia, is modestly so. Moreover, both bilirubin
and vitamin C also prolonged the induction phase for
hemin degradation and did so at physiological concentrations. Vitamin E, on the other hand, was not
nearly so effective, nor was it as effective as the iron
chelator, deferoxamine, or the antioxidant BHT in
prolonging the induction phase of lipid peroxidation.
The requirement for intimate association between
LDL and hemin in LDL oxidation is supported by
experiments employing hemopexin. This serum protein, present in remarkably high plasma concentrations (=75 mg/dl),28 binds hemin with extraordinary
avidity and is likely to prevent insertion of hemin into
Balla et al Hemin Augments LDL Oxidation
10.0-1
10.0
7.5-
7.5
5.0-
5.0
FIGURE 5. Bar graph of hemin degradation and
release of hemin and iron during Hfirmediated
hemin-cataiyzed low density lipoprotein (LDL) lipid
peroxidation. LDL (200 ug/ml) in HEPES-NaCl
(10-150 mM) pH 7.4 buffer was oxidized by 100 yM
Hfi2 at 37°C for 3 hours in the presence of 10 fiM
hemin. At the end of the incubation, hemin and iron
concentrations were determined as micromoles per liter
as described in "Methods." BHT concentration was 10
fimol/L Results are mean±SEM of two experiments
performed in duplicate. H2O2 in LDL-free buffer (not
shown) released less than 10% of the total hemin iron.
HEPES,
N-2-hydroxyethylpiperazine-N'-2-ethanesuifonic acid; BHT, butylated hydroxytohiene.
O
2.5
LDL
Hainln
H2O2
LDL
Hamln
H2O2+BHT
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
LDL. Indeed, as shown in Figure 7, hemopexin
added in stoichiometric amounts with hemin almost
completely inhibited TBARS production from LDL
induced by H2O2-hemin. The specificity of this effect
is indicated by the observations that equimolar concentrations of haptoglobin (specific for binding hemoglobin) and albumin (which has only weak affinity
for free hemin) do not protect LDL from hemincatalyzed peroxidation.
Consequences of Hemin-Catatyzed Low Density
Lipoprotein Oxidation
Hemin and H2O2 not only modify the lipid moiety
of LDL but also denature its protein as well. LDL
exposed to various combinations of hemin, H2O2, or
copper underwent considerable alteration in mobility
when assayed by agarose gel electrophoresis (Figure
8). LDL incubated with hemin and H2O2 for 3 hours
manifested increased anodal mobility, which increased markedly after incubation for 24 hours to
that noted with copper-treated LDL. The increased
mobility suggests a loss of net positive charge, which
can also be assayed independently by measurement
of free amino groups on the LDL particle. As shown
in Figure 9, fluorescamine-titratable free amino
groups progressively decreased in LDL incubated
with hemin and H2O2. Moreover, LDL that was
already slightly peroxidized reacted with hemin even
more rapidly and lost free amino groups without
added exogenous H2O2 (open squares).
In vivo, inflammatory cells may play a role in LDL
oxidation and vascular injury. We have previously
shown7 that exposure of endothelial cells to free
hemin potentiates damage mediated by granulocytes
and toxic O2 species. Oxidant-producing PMA-stimulated neutrophils were therefore tested for their
ability to modify LDL. Hemin plus PMA-activated
PMNs potently provoked LDL oxidation (Figure 10).
Conversely, hemin or activated PMNs alone did not
cause LDL lipid peroxidation over 3 hours.
Also potentially relevant to in vivo vascular damage are studies demonstrating that copper or celloxidized LDL can be directly cytotoxic to endothelium. As shown in Figure 11, hemin-exposed LDL
also damaged endothelial cells (Figure 11 A), and
0.0
LDL oxidation was required for this toxicity. That is,
if LDL oxidation was inhibited by diverse reagents
including U74500A, BHT, probucol, or catalase, endothelial cells were no longer damaged (Figure 11B).
Discussion
The present studies demonstrate that small
amounts of free hemin in the presence of H2O2 or
activated PMNs catalyze the rapid peroxidation of
LDL to substances highly cytotoxic for endothelial
cells. Previous studies of LDL oxidation in vitro have
generally used prolonged (24 hours or more) periods
of aerobic incubation to provoke LDL oxidation.1-3
Hemin strikingly shortens the time required for potentially pathophysiological alterations in the LDL
particle to occur. In fact, studies shown in Figure 1
suggest that maximal lipid peroxidation, which is also
70-1
rot
LDL
Hamln
Buffw
1707
a.
a
O
60-
"1
40-
E
•5
30-
E
c
20-
CO
r
10-
D
00
1
10
DRUG CONCENTRATION (fill)
100
FIGURE 6. Semilog plot of inhibition of hemin-HiOrmediated low density lipoprotein (LDL) lipid peroxidation by
U74500A "lazaroid" and deferoxamine (DF). LDL (200
tiglmt) (asprotein) in HEPES-NaCl (10-150 mM), pH 7.4,
was oxidized by 100 uM H2O2 at 37°C for 3 hours in the
presence of 5 pM hemin. Before addition of HjO2 to the test
tubes, LDL was preincubated with different concentrations of
U74500A or DF at 37 °C for 30 minutes. At the end of the
incubation, TBARS were determined as nanomoles TBARS
per milligram LDL protein as described in "Methods." Results
are mean±SEM of three experiments performed in duplicate.
HEPES, N-2-hydroxyethylpiperazine- N ' -2-ethanesulfonic
acid; TBARS, thiobarbitunc acid-reactive substances.
1708
Arteriosclerosis and Thrombosis
Vol 11, No 6 November/December 1991
TABLE 2. Effect of Antioxidants and Iron Chelators on the
Induction (Lag) Phase of Hemin and Hydrogen Peroxide-Mediated Low Density Lipoprotein Peroxidation
a
b
c
d
e
f
Prolongation of induction phase
By 2 times
By 4 times
Antioxidant/chelator
U74500A
BHT
Deferoxamine
Probucol
Bilirubin
Vitamin C
Vitamin E
(Required drug concentration,
/xmol/1)
0.51+0.03
1.49±0.35
4.90±0.15
7.15±1.79
11.76+0.75
22.00±4.02
47.00+10.63
1.38±0.08
3.63±0.83
12.35 ±1.46
18.50±2.47
22.16±1.76
51.75±10.55
112.50±6.85
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Results represent mean±SEM of four experiments performed
in duplicate.
Low density lipoprotein (LDL) (100 fig/m\) was oxidized by 2.5
fiM hemin and 50 AM H2O2 in HEPES-NaCI (10-150 mM)
buffer, pH 7.4, at room temperature, and peroxidation was followed by measuring hemin absorbance optical density at 405 nm.
Duration of the induction phase was derived from the intersection
time point of the linear slopes of the induction phase and the
propagation phase. Control value was 31.25±2.39 minutes
(mean±SEM, four experiments).
HEPES, Af-2-hydroxyethylpiperazine-/v"-2-ethanesulfonic acid;
BHT, butylated hydroxytoluene.
accompanied by hemin degradation and free inorganic iron release, may occur within 20-60 minutes
of coincubation.
The results of several independent assays of LDL
oxidation all support the conclusion that hemin very
efficiently promotes LDL oxidation. Thus, the "traditional" assay of lipid peroxidation—production of
TBARS—was correlated closely with other measures
of lipid peroxidation including conjugated diene formation, iodometrically assayable lipid hydroperoxide generation, and loss of reactive amino groups. This latter
alteration may have direct pathophysiological significance in atherogenesis, as the binding of short-chain
fatty aldehydes (released during lipid peroxidation) to
lysine e-amino groups of LDL has been shown to foster
LDL uptake by scavenger receptors of macrophages,
with formation of foam cells.1 In addition, these aldehyde-LDL adducts have been shown to be highly
immunogenic,29 suggesting that uptake of such altered
c
1
Q.
a
CO
K
ffl
LDL
LDL+H2O2 LDL+H2O2 LDL+H2O2 LDL+H2O2
Hemin
Hemin
Hemin
Hemin
HPX
HPT
ALB
B
FIGURE 8. Agarose gel electrophoretogram showing mobility
after 3- or 24-hour low density lipoprotein (LDL) oxidation
caused by H2O2-hemin or copper. Figure is of fat red-stained
1% agarose gel LDL electrophoresis. For oxidative modification of LDL (500 iiglml), 100 pMHfi^, 10 \M hemin, or 10
\M CuSO4 was added at 37 "C for 3 (panel A) or 24 (panel
B) hours as discussed in Table 1. Lane a, incubated under Ar;
lane b, incubated aerobically with buffer alone; lane c,
incubated with Hfi^ lane d, incubated with hemin; lane e,
incubated with hemin and H2O2; and lane f incubated with
copper.
LDL by macrophages could also occur through macrophage immunoprotein receptors.
The importance of hemin-derived iron in promoting LDL oxidation is supported by the observation
that iron-deficient "heme" (protoporphyrin IX) has
no amplifying effect (Figure 3). Furthermore, release
of inorganic iron from hemin, as detected by ferrozine binding, occurs concomitantly with LDL oxidation (Figure 5). Such iron release is prevented by
the potent antioxidant BHT, which concomitantly
prevents LDL oxidation as well as generation of
material cytotoxic for endothelium. Interestingly,
FeSO4 and FeCl3 are unable to oxidize LDL (Figure
3), suggesting the need for a hydrophobic "trafficking" molecule to insert the iron into the lipid milieu
FIGURE 7. Bar graph showing that hemopexin (HPX)
inhibits catalytic activity of hemin on H-flr mediated low
density lipoprotein (LDL) oxidation. LDL (200 /Jg/ml) (as
protein) in HEPES (HEPES) -Nad (10-150 mM), pH 7.4,
was oxidized by 100 yM H^ at 37"C for 3 hours in the
presence of 5 \JM hemin. In HPX, haptoglobin (HPT) and
human albumin (ALB) groups, hemin, and proteins were
preincubated at equimolar concentration in HEPES buffer at
room temperature for 20 minutes. Final concentration of
proteins was 5 /xM during LDL oxidation. At the end of the
incubation, TBARS were determined as nanomoles TBARS
per milligram LDL protein as described in "Methods."
Results are mean±SEM of two experiments performed in
duplicate. TBARS, thiobarbituric acid-reactive substances.
Balla et al
Hemln Augments LDL OxidaUon
1709
fKD.01
LOL
LOL
HEMIN
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
FIGURE 9. Line plot of primary amino group content of
hemin-H-flj-coddaed low density lipoprotein (LDL). LDL
(200 iiglml) (asprotein) in HEPES-NaO (10-150 mM), pH
7.4, was incubated at 37 "C for 3 hours in the presence of
various concentrations ofhemin. Hfi2 (100 tiM) was added
to freshly prepared (solid symbols) but not to stored (open
symbols) LDL. At the end of incubation, fluorescaminereactive amino group content was determined as moles per
mole LDL protein as described in "Methods." HEPES,
N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid.
of the LDL particle. It seems likely that hemin and/or
its released iron act in the hydrophobic domains of
the LDL molecule to promote oxidation, as lipidsoluble BHT as well as the 21-amino steroid lazaroid,
U74500A, are much more potent inhibitors of heminmediated LDL oxidation than the highly avid but
relatively lipid-insoluble iron chelator, deferoxamine
(Figure 6 and Table 2). Similarly, others have demonstrated that probucol, which is carried in the LDL
particle, inhibits LDL oxidation during prolonged
aerobic incubation in vitro and inhibits development
of atherosclerosis in LDL receptor-deficient (Watanabe) rabbits as well.4 In the present studies, we
report that probucol is also effective in inhibiting
hemin-stimulated LDL oxidation but is less potent in
this regard than the lipid-soluble lazaroids. This
suggests that the latter might be worth testing in
animal models of atherogenesis.
The small amounts of H2O2 (10 jtM) (Figure 2B),
which were shown in the present studies to efficiently
promote rapid LDL oxidation in the presence but not
absence of hemin, are far less than those previously
calculated to exist in the interspace between a marginated activated neutrophil and the endothelial surface.30 In fact, in the presence of hemin, activated
PMNs themselves are highly effective in promoting
LDL oxidation, possibly mimicking in vivo events at
vascular wall surfaces. We emphasize that the level of
LDL oxidation provoked by these modest amounts of
soluble or PMN-generated oxidants in the presence
of hemin is equivalent to that catalyzed by addition of
high concentrations of inorganic copper (Tables 1
and 2), which are far in excess of those extant in vivo.
Moreover, it seems evident that plasma copper,
because of its avid binding to ceruloplasmin, would
unlikely act as a catalyst of LDL oxidation in vivo.
LDL
PUN+PMA
LDL
HEUIN
PUN+PUA
FIGURE 10. Bar graph showing that activated PMNs oxidize
low density lipoprotein (LDL) in the presence ofhemin. LDL
(200 fig/ml) in Hanks' balanced salt solution was oxidized by
5 xltf/ml PMNs with PMA (100 ng/ml) as the activating agent
at 37 °C for 3 hours in the presence of 5 \JM hemin. Hemin
was added to LDL before PMNs. At the end of the incubation
period, samples were centrifuged at 400%, 4°C for 7 minutes,
and TBARS in the supematants were determined as nanomoles per milligram LDL protein as described in "Methods."
Results are mean±SEM of two experiments performed in
duplicate. PMNs, pofymorphonuclear leukocytes; PMA, 4B12-myristate-13-acetate; TBARS, thiobarbituric acid-reactive
substances.
Instead, we suggest that the initial intercalation of
hemin into the lipid domain of LDL, with subsequent
release of its free iron in that environment, might
avert iron chelation by its binding protein, transferrin, thereby permitting efficient catalysis of lipid
oxidation.
Nevertheless, the process of hemin-mediated
LDL oxidation is clearly not a simple one and
probably involves presently uncharacterized coupled oxidative interactions between LDL, hemin,
exogenous oxidants, hemin-derived iron, and endogenous antioxidants. These complex reactions
may include 1) spontaneous insertion of the hemin
into LDL, 2) subsequent oxidative scission of the
porphyrin ring, 3) release of free iron from the
heme ring, and 4) iron catalysis of oxidation of
further hemin groups, LDL fatty acids, and proteins
(Figures 8 and 9). The precise nature of the iron
catalysis of LDL fatty acid peroxidation is not
presently clear. Certainly, Fe2+ reacts very readily
with lipid hydroperoxides at a rate several orders of
magnitude greater than the reaction between Fe2+
and H2O2 (1.5 x 103 M"1 • sec"1 versus 76 M"1 • sec"1,
respectively),31 yielding alkoxyl radicals (R-O • )
and Fe 3+ . Fe3+, although less reactive, can catalyze
the formation of both alkoxyl and peroxy (RO 2 •)
radicals and the coupled formation of Fe . The
continued reduction of iron may be important because maximal rates of lipid peroxidation require
the presence of both ferrous and ferric species.32
Indeed, on exposure of neuronal tissues to free
hemoglobin, brisk lipid peroxidation ensues (ac-
1710
Arteriosclerosis and Thrombosis
LDL
LDL
Ham In
LDL
H2O2
Vol 11, No 6 November/December 1991
LDL
H*mln
H2O2
FIGURE 11. Bar graphs showing that oxidized low density
lipopmtein (LDL) is cytotoodc to porcine aortic endotheUal
cells. Panel A: LDL (200 fig/ml) in Hanks' balanced salt
solution was incubated at 37 "C for 3 hours with HjO2 (25
fiM), hemin (5 fiM), or CuSO4 (5 yM). Chromium-51loaded endotheUal cells were incubated with the LDL samples
for 8 hours, and percentage specific cytotoxicity was determined as described in "Methods." Hemin, Hfi2, CuSO4
alone, and hemin plus H2O2 together at the above concentrations did not cause cytotoxicity. Results are mean±SEM
of three experiments performed in duplicate. Panel B:
Inhibition of hemin-plus Hflrinduced
LDL oxidation by
antioxidants and iron chelators protects against subsequent
LDL-mediated endotheUal cell injury. Inhibitors used during
the 3-hour LDL modification period were U7450QA, 25 \iM;
BHT, 25 fiM; probucoL, 25 pM; and catalase, 20 y%lml
Results are mean±SEM of three experiments performed in
duplicate. BHT, butylated hydroxytoluene.
LDL
Cu
B
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
X
I
u
u
c
o
ui
Q.
CO
LDL
+U74500A
H»mln+H2O2
•BHT
+PROBUCOL+CATALA3E
companied by an oxidative destruction of hemoglobin heme) and is catalyzed largely by Fe2+ formation
engendered by endogenous neuronal ascorbate.33
Once the process is initiated, our results demonstrate that LDL oxidation is autocatalytic. That is,
minimally modified LDL produced by simple in vitro
storage becomes markedly susceptible to oxidation by
hemin alone without the addition of other exogenous
oxidants such as H2O2 or PMNs (Figures 4B and 9).
Previous studies have demonstrated that the LDL in
the artery walls of patients with atherosclerosis is
already peroxidized.34-35 Hemorrhage, which may occur with plaque fracture, may provide hemin that
could markedly accelerate further vessel wall injury.
These studies may have broader implications when
one considers the possibility that similar heminmediated reactions may occur during 1) hemorrhagic
injury, especially to the lipid-rich central nervous
system; 2) acute intravascular hemolysis, which may
lead to renal damage and failure; and 3) various
inflammatory processes, such as arthritic joint damage due to recurrent hemorrhage into joint tissues, as
is seen in hemophiliacs.
We conclude from the present studies (together
with our previous observation7 that hemin is rapidly
incorporated into and oxidatively damages endothelium) that hemin constitutes a significant danger to
vascular integrity. Not only may hemin directly render the vascular endothelium hypersusceptible to
oxidant injury, but by its ability to intercalate into
LDL particles and catalyze their oxidation, it may
also promote foam cell production and thereby the
formation of the early fatty streak lesions of atherosclerosis. It is tempting to speculate that local accumulation of hemin from trivial hemolysis in areas of
turbulent red blood flow in major blood vessels—for
instance, at vessel bifurcations — may provoke both
enhanced endothelial damage and oxidation of LDL
with its hypothesized ramifications for atherogenesis;
together, these phenomena may help explain the
propensity of atherosclerotic lesions to be found at
these sites. It seems ironic that although heme-based
proteins undoubtedly evolved as a significant advance
in the control of O2 delivery to tissues, they also can
provoke oxidant sensitivity of endothelium and LDL;
by doing so, they may worsen and may even engender
atherosclerotic damage. The remarkably large concentrations of plasma hemopexin (approximately 75
mg/dl) may be an evolutionary tribute to the clear
and present danger posed by free hemin.
References
1. 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
2. Esterbauer H, Dieber-Rotheneder M, Waeg G, Striegl G,
Jurgens G: Biochemical, structural, and functional properties
Balla et al Hemin Augments LDL Oxidation
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
of oxidized low-density lipoprotein. Chem Res Taxicol 1990;3:
77-92
3. Montgomery RR, Nathan CF, Cohn ZA: Effects of reagent
and cell-generated hydrogen peroxide on the properties of low
density lipoprotein. Pmc Natl Acad Sci USA
1986;83:
6631-6635
4. Carew TE, Schwenke DC, Steinberg D: Antiatherogenic effect
of probucol unrelated to its hypocholesterolemic effect: Evidence that antioxidants in vivo can selectively inhibit low
density lipoprotein degradation in macrophage-rich fatty
streaks and slow the progression of atherosclerosis in the
Watanabe heritable hyperlipidemic rabbit. Pmc Natl Acad Sci
USA 1987;84:7725-7729
5. Bjorkhem I, Henriksson-Freyschuss A, Breuer O, Diczfalusy
U, Berglund L, Henriksson P: The antioxidant butylated
hydroxytoluene protects against atherosclerosis. Arterioscler
Thromb 1991;ll:15-22
6. Gey KF, Puska P, Jordan P, Moser UK: Inverse correlation
between plasma vitamin E and mortality from ischemic heart
disease in cross-cultural epidemiology. Am J Clin Nutr 1991;
53(suppl):326S-334S
7. Balla G, Vercellotti GM, Muller-Eberhard U, Eaton J, Jacob
HS: Exposure of endothelial cells to free herae potentiates
damage mediated by granulocytes and toxic oxygen species.
Lab Invest 1991;64:648-655
8. Havel RJ, Eder HA, Bragdon JH: The distribution and
chemical composition of ultracentrifugally separated lipoproteins in human serum. / Clin Invest 1955^4:1345-1353
9. Peterson GL: A simplification of the protein assay method of
Lowry et al which is more generally applicable. Anal Biochem
1977;83:346-356
10. Beuge JA, Aust SD: Microsomal lipid peroxidation. Methods
Enzymol 1978;52:302-310
11. Aust SD: Lipid Peroxidation. Boca Raton, Fla, CRC Press, Inc,
1985, pp 203-207
12. Thomas JP, Maiorino M, Ursini F, Girotti AW: Protective
action of phospholipid hydroperoxide glutathione peroxidase
against membrane-damaging lipid peroxidation. / Biol Chem
1990;265:454-461
13. Braughler JM: Lipid peroxidation-induced inhibition of
•y-aminobutyric acid uptake in rat brain synaptosomes: Protection by glucocorticoids. J Neumchem 1985;44:1282-1288
14. Undenfriend S, Stein S, Bohlen P, Dairman W, Leimgruber
W, Weigele M: Fluorescamine: A reagent for assay of amino
acids, peptides, proteins, and primary amines in the picomole
range. Science 1972;178:871-872
15. 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
16. Hatch FT, Lees RS: Practical methods for plasma lipoprotein
analysis. Adv Lipid Res 1968;6:l-68
17. Omura T, Sato R: The carbon monoxide-binding pigment of
liver microsomes. J Biol Chem 1964;239:2370-2378
18. Berry EA, Trumpower BL: Simultaneous determination of
hemes a, b, and c from pyridine hemochrome1 spectra. Clin
Biochem 1987;161:1-15
19. Carter P: Spectrophotometric determination of serum iron at
the submicrogram level with a new reagent (ferrozine). Anal
Biochem 1971;40:450-458
1711
20. Artiss JD, Vinogradov S, Zak B: Spectrophotometric study of
several sensitive reagents for serum iron. Clin Biochem 1981;
14:311-315
21. Vercellotti GM, Yin HQ, Gustafson KS, Nelson RD, Jacob
HS: Platelet-activating factor primes neutrophil responses to
agonists: Role in promoting neutrophil-mediated endothelial
damage. Blood 1988;71:1100-1107
22. Whorton AR, Montgomery ME, Kent RS: Effect of hydrogen
peroxide on prostaglandin production and cellular integrity in
cultured porcine aortic endothelial cells. / Clin Invest 1985;76:
295-302
23. Nagelkerke J, Barto KP, van Berkel TJC: In vivo and in vitro
uptake and degradation of acetylated low density lipoprotein
by rat liver endothelial, kupffer, and parenchymal cells. / Biol
Chem 1983;258:12221-12227
24. Balla G, Vercellotti GM, Eaton JW, Jacob HS: Iron loading of
endothelial cells augments oxidant damage. / Lab Clin Med
1990-.116:546-554
25. Haurowitz F, Schwerin P, Yenson MM: Destruction of hemin
and hemoglobin by the action of unsaturated fatty acids and
oxygen. / Biol Chem 1941;140:353-359
26. Braughler JM, PTegenzer JF, Chase RL, Duncan LA, Jacobsen EJ, McCall JM: Novel 21-amino steroids as potent inhibitors of iron-dependent lipid peroxidation. / Biol Chem 1987;
262:10438-10440
27. Braughler JM, PTegenzer JF: The 21-aminosteroid inhibitors
of lipid peroxidation: Reactions with lipid peroxyl and phenoxy radicals. Free Radio Biol Med 1989;7:125-130
28. Muller-Eberhard U, Javid J, Liem HH, Manstein A, Hanna
M: Plasma concentrations of hemopexin, haptoglobin, and
heme in patients with various hemorytic diseases. Blood 1968;
32:811-815
29. Steinbrecher UP, Fisher M, Witztum JL, Curtiss LK: Immunogenicity of homologous low density lipoprotein after methylation, ethylation, acetylation, or carbamylation: Generation of
antibodies specific for derivatized lysine. J Lipid Res 1984;25:
1109-1116
30. Schraufstatter IU, Cochrane CG: Oxidants: Types, sources,
and mechanisms of injury, in Crystal RG, West JB (eds): The
Lung: Scientific Foundations. New York, Raven Press, Publishers, 1991, pp 1803-1810
31. Halliwell B, Gutteridge JMC: Free Radicals in Biology and
Medicine. Oxford, UK, Clarendon Press, 1989, pp 201-202
32. Minotti G, Aust SD: The requirement for ferric in the
initiation of lipid peroxidation by ferrous and hydrogen peroxide. J Biol Chem 1987;262:1098-1104
33. Sadrzadeh SMH, Eaton JW: Hemoglobin-mediated oxidant
damage to the central nervous system requires endogenous
ascorbate. / Clin Invest 1988;82:1510-1515
34. Shaikh M, Martini S, Quiney JR, Baskerville P, LaVille AE,
Browse NL, Duffield R, Turner PR, Lewis B: Modified
plasma-derived lipoproteins in human atherosclerotic plaques.
Atherosclerosis 1988;69:165-172
35. Yla-Herttuala S, Palinski W, Rosenfeld ME, Parthasarathy S,
Carew TE, Butler S, Witztum JL, Steinberg D: Evidence for
the presence of oxidatively modified low density lipoprotein in
atherosclerotic lesions of rabbit and man. / Clin Invest 1989;
84:1086-1095
KEY WORDS • hemin • endothelium
oxidants • low density lipoproteins
• atherosclerosis •
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Hemin: a possible physiological mediator of low density lipoprotein oxidation and
endothelial injury.
G Balla, H S Jacob, J W Eaton, J D Belcher and G M Vercellotti
Arterioscler Thromb Vasc Biol. 1991;11:1700-1711
doi: 10.1161/01.ATV.11.6.1700
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272 Greenville
Avenue, Dallas, TX 75231
Copyright © 1991 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/11/6/1700
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Arteriosclerosis, Thrombosis, and Vascular Biology 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 Answerdocument.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/