| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Original Contributions |
From the Cell Biology Group, Heart Research Institute, Camperdown, Australia.
Correspondence to Roger T. Dean, Cell Biology Group, Heart Research Institute, 145 Missenden Rd, Camperdown NSW 2050, Australia. E-mail r.dean{at}hri.org.au
| Abstract |
|---|
|
|
|---|
Key Words: low-density lipoprotein lipid peroxidation macrophage transition metal
| Introduction |
|---|
|
|
|---|
The various cell types found in an atheromatous plaque have all been shown to promote LDL oxidation in vitro,6 7 8 9 and this requires the presence of transition metals, both Fe and Cu.10 Indeed, we have recently shown11 that cell-mediated reduction of transition metal may be an important component of cell-mediated oxidation, providing a continued flux of the reduced form of the metal, which cleaves hydroperoxides most efficiently. Ham's F-10 (F-10), which is supplemented with Cu and Fe (0.01 and 3 µmol/L, respectively), is a commonly used medium in these studies.
We have previously demonstrated an inhibition of LDL oxidation by macrophages in HBSS supplemented with either Fe or Cu.10 The present study has directly addressed the inhibition of LDL oxidation in more enriched media and provides observations regarding the mechanism(s) involved. In comparison with F-10, RPMI had to be supplemented with substantially higher concentrations of Cu and Fe to support comparable LDL oxidation; only at these high concentrations did murine macrophages promote LDL oxidation in RPMI. At lower concentrations of metal, these cells effectively blocked LDL oxidation in dual-metalsupplemented RPMI by mechanisms which included metal sequestration.
| Methods |
|---|
|
|
|---|
Media
At the time of use, 2 mmol/L L-glutamine, 100
U/mL penicillin, and 100 µg/mL streptomycin (Sigma) were added to
both media. Where indicated, RPMI was supplemented with Fe (as
FeCl3 · 6H2O) and Cu
(as CuSO4 · 5H2O)
from 100x stock solutions freshly prepared in 0.9% (wt/vol) NaCl for
each experiment. A Cu:Fe ratio of 1:10 was chosen as routine because
this ratio was shown to be optimal for Cu2+
catalysis of Fe2+ oxidation12 and
for cell-mediated oxidation of LDL in HBSS.10 Other metal
ratios were used and are specified.
Cells
Murine resident macrophages were isolated from adult QS
mice as described previously.10 Cells were plated in
tissue culture dishes (Falcon) at an initial concentration of
4x106 cells/24 mm diameter well or
9.6x106 cells/35 mm well.
LDL
Plasma was drawn from healthy, fasted adults. LDL (
=1.02 to
1.05) was isolated by discontinuous ultracentrifugation
as described previously,10 and filter-sterilized
(0.45 µm). LDL was stored in the presence of 1 mg/mL EDTA and
100 µg/mL chloramphenicol under nitrogen, kept in the dark at 4°C,
and used within 7 days of isolation. Before use, LDL was dialyzed (in
the dark at 4°C for at least 16 hours) against 4x1 L of
deoxygenated PBS, containing 100 µg/L chloramphenicol and
4 g/L of Chelex (prewashed before use),13 to remove the
EDTA. Chelex was included to prevent adventitious LDL oxidation. After
dialysis, LDL protein was assayed using the bicinchoninic acid assay
with a BSA standard as previously described.14 Suppression
of oxidation in LDL isolated by this procedure was assessed by
measurement of
-tocopherol and cholesteryl ester
hydroperoxides at the time of use. The tocopherol content
of LDL varies among individuals15 ; from our pool of donors
LDL routinely contained 12.3±3.6 nmol
-tocopherol/mg
cell protein (n=15), which closely matches that previously reported in
a larger study 11.58±3.4 (n=85).15 The LDL contained
undetectable amounts of cholesteryl ester hydroperoxides (detection
limit 10 pmol/mg LDL protein). On this evidence, it is concluded that
no significant oxidation of LDL occurs during isolation or
storage.
Cell-Mediated LDL Oxidation
Cells (4x106/24 mm well) were
incubated in 1 mL of either F-10 or RPMI (metals added where indicated)
containing LDL (50 µg/mL) at 37°C and 5%
CO2/95% air. Cell-free incubations were
performed in parallel. At timed intervals, the supernatants were
collected and centrifuged for 30 seconds at 16 000g
to sediment any cell debris. 800 µL of the resultant supernatant was
added to 200 µL PBS containing 2 mmol/L EDTA and 20
µmol/L butylated hydroxytoluene; this solution was extracted into 2.5
mL of methanol and 5 mL of hexane. We have previously observed a 100%
recovery of sterol into the hexane phase.16 17
Samples were stored at -80°C. Four-milliliter samples of the hexane
phases were evaporated and redissolved in appropriate mobile phases for
HPLC analysis (see below). Separate samples of supernatants
were analysed immediately for flourescence and electrophoretic
mobility.
Cell-Conditioned Medium Experiments
Cells (9.6x106/35 mm well) were
incubated in 3 mL of medium for 24 hours. Cell-free incubations were
performed in parallel. Supernatants from 2 to 3 wells of the same
treatment were pooled and then centrifuged at 750g
for 5 minutes at 4°C to remove any detached cells. LDL was added to
the resultant supernatants for a protein concentration of 50 µg/mL
and incubated (cell-free) in 24 mm wells at 37°C and 5%
CO2 for 24 hours. The media were then extracted
into methanol/hexane as described above.
HPLC Analysis
Analysis of LDL oxidation was by reverse-phase HPLC with
UV detection for free cholesterol and cholesteryl esters
(isopropanol:acetonitrile, 70:30;
=210 nm) and two of their
oxidation products: 7-ketocholesterol and cholesteryl
linoleate hydroperoxide
(isopropanol:acetonitrile:H2O, 54:44:2;
=234
nm) using a 25-cm Supelcosil C18 column as described
previously.16 17
Indirect Measurements of LDL Oxidation
The electrophoretic mobility of LDL samples was measured as
previously described17 18 using 1% Universal agarose gels
(Ciba-Corning, Palo Alto, CA) in Tris-Barbitone buffer (pH 8.6) at 90 V
for 45 minutes. Bands were visualized using Fat Red 7B. LDL which had
not been incubated in metal-containing media was used as reference.
Mobilities of modified LDLs were calculated by dividing their distance
migrated by that traveled by the reference LDL.
The generation of fluorescent products during LDL oxidation was measured in a Perkin-Elmer LS50B luminescence spectrophotometer with 360 nm excitation and 430 nm emission wavelengths and 5 nm slit widths, as described previously.19 The fluorescence of relevant LDL-free media were subtracted from the readings.
Ascorbate Oxidation
Cells (4x106/24 mm well or
9.6x106/35 mm well) were incubated up to 24
hours in metal-supplemented RPMI or in F-10. Cell-free incubations were
performed in parallel. Supernatants were collected in the same manner
as the cell conditioned medium experiments. 700 µL of supernatant was
incubated with 10 µmol/L ascorbate at room temperature;
consumption of ascorbate was determined by a decrease in absorbance at
265 nm using a Hitachi U-3210 spectrophotometer.20
Ascorbate-containing samples were compared against an ascorbate-free
sample blank.
Atomic Absorption Spectrometry
Dissolved metal concentrations were determined on the media
samples by graphite furnace atomic absorption spectrometry following
micro-solvent extraction,21 using a Perkin-Elmer 4100ZL
spectrometer. Analyses were performed at the Center for
Advanced Analytical Chemistry, CSIRO Division of Coal and Energy
Technology, Lucas Heights Science and Technology Center, Lucas Heights,
Sydney, Australia.
Fe Determination by Ferrozine Assay
Cell supernatants and media from parallel cell-free treatments
were adjusted to pH 3 to 4 with HCl. The Fe was reduced with 100
µmol/L ascorbate and chelated with 150 µmol/L ferrozine.
Samples were incubated for up to 120 minutes at room temperature and
formation of the Fe(II)-ferrozine complex was measured at 562 nm using
a Hitachi U-1100 spectrophotometer. In preliminary experiments standard
curves of Fe in the media of interest were prepared and, although a
linear increase was observed, the absorbance values were slightly lower
than those estimated using the published extinction coefficient
(2.86x104 M-1
cm-1).22 For this reason the loss of
ferrozine-detectable Fe in the cell conditioned media was expressed as
a percentage of that detected in the cell-free control.
Thiol Determination
Cell supernatants and media from parallel cell-free treatments
were assayed for thiols by the dithionitrobenzoic (DTNB)
reaction23 using L-cysteine standards prepared
in the appropriate medium with 20 mmol/L EDTA. Aliquots (200 µL)
of sample or standard were mixed with 750 µL of 200 mmol/L
Na2HPO4 ·
12H2O and 20 mmol/L EDTA (pH 9.0). To each
sample was added 50 µL DTNB (4 mmol/L) in 50 mmol/L PBS (pH
7.0). After 30 minutes at 37°C, the absorbance was measured at 412 nm
against a DTNB-free blank.
| Results |
|---|
|
|
|---|
In cell-free RPMI there was a small degree of LDL oxidation. After 24 hours in vitro, CLOOH was 7.3±1.2 nmol/mg LDL protein (n=6), equivalent to 3 peroxide molecules per particle. This was accelerated when the medium was supplemented with 0.1 µmol/L Cu and 1 µmol/L Fe (31.2±15.0 nmol/mg LDL protein; n=10).
In agreement with previous studies, murine macrophages in F-10
promoted LDL oxidation (Figure 1B
and 1D
). Compared with cell-free conditions, the rates of cholesteryl ester
consumption and formation of CLOOH and 7-KC were more rapid. Protein
modification, as measured by changes in both relative electrophoretic
mobility and fluorescence, also developed at an accelerated
rate in the presence of cells (Figure 1F
). Changes in relative
electrophoretic mobility and fluorescence of LDL during
metal-catalyzed oxidation are considered to predominantly
represent changes in apoB composition secondary to lipid
peroxidation,24 25 although they may additionally reflect
direct oxidation of some amino acid residues.26 As such,
they are little affected at early stages of oxidation when lipid
peroxides are first identified; but, like 7-KC, increase progressively
as oxidation advances beyond the peak of hydroperoxide
generation17 18 24 (Figure 1
).
|
LDL oxidation in RPMI supplemented with Cu and Fe at 0.1 and 1
µmol/L was markedly lower than that seen in F-10 (Figure 1A
cf
1B, 1C cf 1D, and 1E cf 1F) despite the concentration of Cu in
metal-supplemented RPMI being 10-fold greater than that in F-10.
Macrophages effectively blocked LDL oxidation in such
(moderately) metal-supplemented RPMI. Thus the modest cholesteryl
linoleate consumption (Figure 1A
) and CLOOH generation that
occurred in this medium was effectively blocked in the presence of
macrophages (Figure 1C
; note however the difference in
scale compared with Figure 1D
).
The degree of LDL oxidation in metal-supplemented RPMI under cell-free
conditions was dependent on the concentration of metals supplied
(Figure 2
). At the highest concentrations
(1 µmol/L Cu and 30 µmol/L Fe; being respectively 100-
and 10-fold higher than the nominal concentrations in F-10) of the
metals studied, oxidation was almost comparable with that in F-10, with
hydroperoxides past their peak, and substantial 7-KC levels in the
supplemented RPMI conditions. As the concentrations of metal added to
RPMI increased, the cells changed from being antioxidant to prooxidant.
At the highest metal concentrations, the cells promoted LDL oxidation
to almost the same extent as they did in F-10 (Figure 2
). To
appreciate this, it is necessary to keep in mind that hydroperoxide
levels reach a peak and then decline, whereas 7-KC increases relatively
steadily, as oxidation progresses.18 24 Table 1
provides additional measures of LDL
oxidation in F-10 and RPMI (1 µmol/L Cu and 30 µmol/L
Fe), which further demonstrate the prooxidant activities of
macrophages in these media. Thus at 24 hours, increases in both
relative electrophoretic mobility (R.E.M.) and fluorescence of
LDL samples paralleled changes in 7-KC content (Table 1
),
further indicating that in both media, cell-mediated LDL oxidation
exceeded that of cell-free controls.
|
|
Although RPMI has an approximately 4-fold higher concentration of phenol red than F-10, this was apparently not important. We found cell-free oxidation and cell-mediated antioxidation of LDL in metal-supplemented RPMI with or without phenol red were indistinguishable (data not shown).
Mechanisms that may be involved in the inhibition of LDL oxidation in
metal-supplemented RPMI include metal sequestration,27 28
removal of cholesteryl ester hydroperoxides from LDL,29 or
release of inhibitory species into the extracellular
environment.30 31 To further investigate these
mechanism(s), F-10, or RPMI with 0.1 µmol/L Cu and 1
µmol/L Fe were preincubated with macrophages for 24 hours,
then the capacity of these media to support subsequent LDL oxidation in
the absence of cells was measured. Cellular "conditioning" either
significantly (P<0.01) reduced (in the case of F-10) or
blocked (in the case of metal-supplemented RPMI) the ability of these
media to support cell-free oxidation (Table 2
).
|
We assessed the importance of metal sequestration by determining
whether the capacity of conditioned medium to oxidize added ascorbate
was decreased in comparison with that of unconditioned medium.
Ascorbate oxidation is a sensitive assay for measuring redox-active
transition metals in buffers.20 Ascorbate oxidation by
metal-supplemented RPMI (control rate 0.12±0.04 µmol ·
L1 · min1) was
decreased after treatment with macrophages (Figure 3
). This decrease was apparent after 6
hours preincubation with cells but maximal after 24 hours in vitro.
This indicates a selective loss of redox active metal in the
cell-conditioned medium. No such clear difference was seen with
cell-conditioned F-10 (control rate of ascorbate oxidation,
0.18±0.05 µmol · L1 ·
min1), probably because of its higher capacity
to oxidize ascorbate.
|
Cu and Fe sequestration from either F-10 or RPMI supplemented with Cu and Fe at 0.1 and 1 µmol/L, respectively, was assessed using atomic absorption spectroscopy. However, we found that quantification of both Cu and Fe in such complex matrices by atomic absorption spectroscopy was unreliable. We therefore used a colorimetric method for determination of Fe in a chelated complex as an alternative approach. When macrophages were incubated in RPMI supplemented with 30 µmol/L Fe for 24 hours, there was a clear loss of metal as determined spectrophotometrically using the ferrozine assay (only 8±2% of that measured in the parallel cell-free incubated medium for 3 independent experiments).
Sequestration of metals was not the sole mechanism by which
macrophages inhibited LDL oxidation. Resupplementing
cell-conditioned metal-supplemented RPMI with Cu and Fe at the original
concentrations did not restore its capacity to support LDL oxidation
(Table 3
).
|
Release of thiols was considered as a possible additional
inhibitory action, because at relatively high
concentrations, several thiols can inhibit metal-catalyzed lipid
peroxidation.32 Both F-10 and RPMI contain thiols (RPMI as
disodium cystine and glutathione:
400 µmol/L available SH;
and F-10 as cysteine
200 µmol/L available SH) which in the
presence of metals would all be oxidized. Cells in vitro are capable of
re-reducing these disulfides.9 11 33 In the absence of
cells, thiol levels in metal-supplemented RPMI and F-10 were below the
level of detection (
1 µmol/L). The levels in cell-conditioned
metal-supplemented RPMI were not significantly higher than for the
cell-free condition (Table 4
), although
significant (P<0.05) amounts of thiol were detected in
cell-conditioned F-10 (Table 4
). Thus there was no indication
that cell-associated increases in thiol levels were important in the
inhibitory effects of medium conditioning in the case of
metal-supplemented RPMI.
|
| Discussion |
|---|
|
|
|---|
It is only at high concentrations of both Cu and Fe that the degree of LDL oxidation in cell-free metal-supplemented RPMI was comparable to that seen in F-10. Thus, apparently the redox availability of the metals in RPMI is less than that in F-10. Apart from containing both Cu and Fe, F-10 differs from RPMI in being relatively deficient in a number of components including: amino acids (the exceptions being arginine and histidine); total phosphates; total available thiols; and phenol red. These differences may explain the relative inability of RPMI to support oxidation, and hence cell-associated antioxidative processes may override simultaneous cellular prooxidant activities.
Several mechanisms could be involved in the inhibition of LDL
oxidation. Metal sequestration by macrophages is known to
occur. We have shown a loss of ability to support oxidation in
metal-supplemented media after preincubation with cells; we have also
directly measured the loss of Fe from the same media. Thus metal
sequestration probably contributes to the cell-mediated inhibition of
LDL oxidation. However, metal sequestration alone cannot fully explain
the results. The degree of LDL oxidation was still reduced after
resupplementation of RPMI with metals and the capacity of conditioned
F-10 to support subsequent cell-free LDL oxidation was reduced even
though there was not detectable loss of ascorbate-detectable
redox-active metal. Thus it is likely that the cells additionally
modified the medium. A major role of thiols in this inhibition was
eliminated, because cell-conditioned metal-supplemented RPMI contained
low or undetectable thiols; this finding is in agreement with our other
recent studies.11 The cells might also inhibit oxidation
by removing hydroperoxides (by nonradical reduction), as may be
indicated because the maximum levels of CLOOH are higher in the absence
of cells than in the presence of cells (Figure 2
).
In conclusion, we have demonstrated that for the nutrient-rich medium RPMI to support LDL oxidation in the absence or presence of cells, relatively high levels of metal supplementation are required. At lower metal concentrations, cells can effectively block LDL oxidation for an extended period of incubation. This is likely to involve sequestration of metals as well as other modifications of the extracellular environment. In physiological conditions (where free metal concentrations are low), the sequestration of metals by macrophages will thus probably contribute to antioxidant function. However, the prooxidant activity of macrophages may be expressed if free metal concentrations rise, as apparently occurs, for example, during atherosclerosis.2
| Acknowledgments |
|---|
Received October 20, 1997; accepted October 28, 1998.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
H. Fujimoto, J.-i. Taguchi, Y. Imai, S. Ayabe, H. Hashimoto, H. Kobayashi, K. Ogasawara, T. Aizawa, M. Yamakado, R. Nagai, et al. Manganese superoxide dismutase polymorphism affects the oxidized low-density lipoprotein-induced apoptosis of macrophages and coronary artery disease Eur. Heart J., May 2, 2008; 29(10): 1267 - 1274. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Baoutina, R. T. Dean, and W. Jessup Macrophages Can Decrease the Level of Cholesteryl Ester Hydroperoxides in Low Density Lipoprotein J. Biol. Chem., January 21, 2000; 275(3): 1635 - 1644. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1999 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |