Effects of Iron- and Hemoglobin-Loaded Human Monocyte–Derived Macrophages on Oxidation and Uptake of LDL
Abstract It is generally accepted that transition metals are required for cellular LDL oxidation. LDL may also be oxidized by iron and reducing agents in cell-free systems. We hypothesized that lysosomal iron may be exocytosed from macrophages that have been iron loaded by phagocytosis and degradation of iron-rich structures, eg, erythrocytes, and that such released iron may promote LDL oxidation and uptake by macrophages. Human monocyte–derived macrophages (HMDMs) were isolated and cultured for 7 days and then exposed to FeCl3, Fe-ADP, or Fe-EDTA (100 μmol/L) or hemoglobin (25 or 50 μg/mL) for 24 hours. After rinsing, LDL (50 to 150 μg/mL) was added in fresh culture medium without serum. After another 24 hours the media concentrations of iron and thiobarbituric acid–reacting substances as well as the electrophoretic mobility of LDL were increased, while the cells showed only minimal signs of decreased viability. Lipofuscin, neutral lipids, and phospholipids accumulated in a granular, lysosome-like pattern, and the cells acquired a foam cell–like morphology. There was a strong correlation (r=.87, P=.005) between the amount of iron added during the pre-exposure period and lipofuscin accumulation during the ensuing exposure to LDL in fresh, serum-free medium. Our results support our hypothesis and indicate that lysosomal iron may be exocytosed from HMDMs and promote oxidation and uptake of LDL and thus induce foam cell formation.
- Received December 20, 1994.
- Accepted May 9, 1995.
Oxidation of LDL is considered to be an important step in the alteration of its apoB into a ligand for the scavenger receptors of macrophages that temporarily reside in the arterial intima. The ensuing LDL uptake would result in the conversion of macrophages into foam cells, thus starting the formation of the early atheromatous lesion.1
Although this chain of events is generally accepted, there are many unsolved questions concerning several fundamental parts of the process, eg, how and where the LDL particles become oxidized. Most authors favor the opinion that LDL oxidation occurs in the arterial intima under the influence of oxygen-derived reactive metabolites and the catalytic activity of transition metals in a reduced state.2 3
It has been convincingly shown that both copper and iron in reduced form can oxidize LDL into a form that is recognized by the macrophage scavenger receptor.4 5 This type of LDL oxidation is further stimulated by a variety of cell types, eg, monocytes, macrophages, smooth muscle cells, and endothelial cells, which are all known to produce superoxide anion radicals and hydrogen peroxide.1 2
Both copper and iron are present in atheromatous lesions, but neither the origin of these metals nor their relation to macrophages and smooth muscle cells under the formation of early atherosclerotic lesions has been determined.6 7 8
In vitro oxidation of LDL is often accomplished by exposing it in serum-free cell culture medium to reduced transition metals in nanomole or micromole per liter concentrations, eg, Ham’s F10 medium, which contains about 3 μmol/L iron. However, in vivo the two most abundant catalytically active transition metals, iron and copper, are unlikely to exist in free form in plasma or interstitial fluids.6
In this study we suggest that migrating HMDMs that temporarily reside in vessel walls may, under certain conditions, eg, as a result of phagocytotic uptake of cellular debris or erythrocytes rich in iron-containing metalloproteins, for a limited time contain secondary lysosomes with a high concentration of iron. Such cells may, in a microenvironment, expose LDL to a locally high concentration of iron in a catalytically active form due to exocytosis of lysosomal contents and, moreover, to superoxide, hydrogen peroxide, and hydroxyl radicals (as a consequence of Fenton chemistry). We here present experimental support for this hypothesis by using an in vitro model system of HMDMs exposed in culture to simple iron compounds and Hb.
Chemicals and Culture Media
TBA, butylated hydroxytoluene, trichloroacetic acid, EDTA, ADP, Ficoll-Paque, Nile red, cytochrome c (horse-heart type VI), NADH, pyruvate, and Hb were obtained from Sigma Chemical Co. SOD and Triton X-100 were from Boehringer-Mannheim. Iron (III) chloride (FeCl3 · 6H2O) was purchased from E. Merck AG. RPMI-1640 medium with and without phenol red, FCS, and HBSS were from GIBCO. ADP-Fe and EDTA-Fe (100:1 and 3:1, respectively) were freshly prepared from stock solutions of FeCl3, ADP, and EDTA.
LDL (1.025<d<1.050 g/mL) was freshly isolated by sequential ultracentrifugation9 from sera of normolipidemic donors who had fasted overnight. LDL was prepared in the presence of EDTA (1.4 mg/mL) to inhibit lipid peroxidation. LDL was finally dialyzed under nitrogen against 0.01 mol/L phosphate-buffered saline (0.16 mol/L NaCl), pH 7.4, for 24 hours at 4°C before incubation with HMDMs.
Isolation and Culture of HMDMs
Mixed mononuclear cells were separated from the buffy coat of fresh type A Rh+ donor blood by using the density-gradient centrifugation method of Böyum.10 The buffy coat (20 mL) was layered over Ficoll-Paque (15 mL) and centrifuged at 400g for 30 minutes at room temperature. The mixed mononuclear cell band was removed by aspiration, and the cells were washed three times and recentrifuged at 200g for 10 minutes in HBSS at room temperature to remove platelets.
The platelet-free mononuclear cells were finally suspended in RPMI-1640 with 10% (vol/vol) FCS or homologous serum and plated in FCS-coated 35-mm tissue-culture Petri dishes (Costar) prepared according to the method of Kumagai et al.11 The cells were allowed to adhere for 3 hours at 37°C. Nonadherent cells were removed by gentle washing with complete growth medium.
This isolation procedure yielded monocyte preparations of more than 95% purity as judged by immunocytochemistry using the monoclonal antibody CD14. The monocytes matured to macrophages in 6 to 8 days. The medium was changed every 48 hours.
After isolation the monocytes were cultured in RPMI-1640 medium (claimed to be almost free of iron) with 10% FCS for 7 to 8 days in a humidified 5% CO2–air incubator at 37°C.
These cells were counted (5×106/mL) and seeded into multiwell plates (Costar) and exposed for 24 hours to iron compounds and Hb in complete culture medium. The medium was then removed, and the cells were rinsed twice in warm RPMI-1640. Fresh medium with LDL but without serum was then added to the wells. At the same time conditioned medium was collected from the iron- and Hb-exposed HMDM cultures. Iron content, superoxide production, TBARS, and LDL electrophoretic mobility were periodically measured during the incubation period. Cells on coverslips were removed from the wells 48 hours after the addition of medium with LDL, and autofluorescence (lipofuscin), neutral lipid, and phospholipid levels were measured.
Iron Content of Growth Media
After culture in complete growth media with FeCl3, Fe-ADP, or Fe-EDTA (100 μmol/L iron) or Hb (25 or 50 μg/mL) for 24 hours, the HMDMs were washed twice with RPMI-1640. Cells were then exposed to LDL (100 μg/mL) for another 24 hours. The iron concentrations of the media were determined by atomic absorption spectrophotometry (Z-8270 Polarized Zeeman, Hitachi) and compared with concentrations in the media of control HMDMs.
Lactic dehydrogenase activities in both the incubation media and cellular pellet (homogenized in 0.1% Triton X-100) from iron- and Hb-loaded HMDMs were determined at 37°C by measuring the oxidation rate of lactate to pyruvate as the NAD+/NADH conversion, which was monitored spectrophotometrically at 340 nm.12 The viability of experimental cells was also assayed morphologically and by a Trypan Blue dye (0.1%) exclusion test.13 14
HMDM Superoxide Production
To examine the role of O2< ARRANGE="STAGGER">−• in the experimental system, superoxide production was measured as SOD-inhibitible reduction of ferric cytochrome c.15 HMDMs were exposed for 24 hours to iron compounds or Hb in complete growth media. The cells were then rinsed in HBSS and incubated under ordinary culture conditions in RPMI-1640 (without phenol red and serum) together with LDL (100 μg/mL) and cytochrome c (1 mg/mL) with (50 μg/mL) or without SOD. Medium was removed for analysis of reduced cytochrome c at regular intervals. The absorbance was read at 550 nm against an RPMI-1640 blank. Reduction of cytochrome c was calculated by using an extinction coefficient of 21.0 mmol/L · cm−1 and was expressed as the difference between medium with and without SOD.16
LDL Oxidation (TBARS)
TBARS levels produced during LDL oxidation were determined according to the method of Buege and Aust.17 Trichloroacetic acid (15% wt/vol) and TBA (0.375% wt/vol) were dissolved in 0.25N hydrochloric acid. Butylated hydroxytoluene in ethanol was added to the TBA reagent (final concentration, 0.01%). Samples (0.5 mL) were added to 1 mL TBA reagent, and the mixed solution was heated for 30 minutes at 98°C, cooled for 10 minutes, and centrifuged at 1000g for 10 minutes. The absorbance was determined at 532 nm in a Perkin-Elmer Lambda 2 UV/Vis Spectrometer against a blank of TBA reagents added to RPMI-1640 medium without phenol red. The TBARS concentration of the sample was calculated by using an extinction coefficient of 156 mmol/L · cm−1.
Cells grown on coverslips were exposed to iron compounds and Hb as described above, rinsed, and incubated in RPMI-1640 with LDL but without phenol red and serum for 48 hours.
To evaluate the accumulation of lipids within HMDMs, a Nile red stock solution (100 μg/mL) in acetone was diluted 20 times with PBS and used to stain unfixed cells for 10 minutes at 37°C. The staining was followed by a rinse in PBS. After inverting the coverslips on excavated microculture slides in a drop of PBS, the stained cells were examined in an MPV-III Leitz fluorescence microscope integrated with an ABC 800 computer (Nokia AB) at two spectral settings: yellow-gold fluorescence, representing neutral lipids, was measured by using a filter combination (a 450- to 490-nm band-pass excitation filter, a 510-nm center wavelength chromatic beam splitter, and a 515-nm long-pass barrier filter) for fluorescein fluorescence; red fluorescence, representing phospholipids, was measured by using a filter combination (a 530- to 560-nm band-pass excitation filter, a 580-nm center wavelength chromatic beam splitter, and a 580-nm long-pass barrier filter) for rhodamine fluorescence.18 The microscope was calibrated by using a uranyl standard. Data, which are expressed in AUs, were stored and processed by the computer.
Before the cells were stained with Nile red, HMDM lipofuscin-specific autofluorescence was measured at the same spectral settings that were used for neutral lipids.19
Agarose Gel Electrophoresis
The electrophoretic mobility of LDL was determined by using 1.2% agarose gel electrophoresis in 50 mmol/L barbital buffer. Lipoproteins were visualized by staining with 0.2% Sudan blue in 60% ethanol.20
Significance of differences was calculated by using Student’s t test and ANOVA.
Iron Content of Media and Cellular Viability
To determine whether endocytosed iron could be exocytosed, HMDMs were exposed for a 24-hour period to FeCl3, Fe-ADP, or Fe-EDTA (100 μmol/L Fe3+) or Hb (25 or 50 μg/mL) under ordinary culture conditions, rinsed, and cultivated for another 24 hours in RPMI-1640 without serum.
Initially the iron contents of media from the experimental groups were almost the same as from control media. However, 24 hours later there were significantly increased iron concentrations in the media of most experimental groups (Table 1⇓). No exocytosed iron was detected after the exposure of HMDMs to the higher concentration (50 μg/mL) of Hb, which may be due to cytotoxic effects of Hb.
Although it is generally stated that RPMI-1640 does not contain iron, we found a minute concentration (30.73±2.11 μg/L) of iron in fresh, untreated RPMI-1640 (GIBCO, cat No. 041-02404).
The addition of iron or Hb to HMDMs during culture for 24 hours resulted in only a minor loss of viability as estimated by lactic dehydrogenase retention and the Trypan Blue dye exclusion test (Table 2⇓).
The production of superoxide by HMDMs that were or were not pre-exposed to iron compounds or Hb and then incubated with LDL in serum-free medium was determined and compared with control production. The results, based on SOD-sensitive reduction of cytochrome c, are shown in Fig 1⇓. The rates of cytochrome c reduction in the media of cells pre-exposed to iron or Hb were somewhat lower than that in control media. The O2< ARRANGE="STAGGER">− • release by HMDMs (control and pre-exposed) in the presence of LDL was gradual, with the highest cumulative value obtained after 8 hours.
The TBARS concentrations of the LDL-containing media increased significantly during the 24 hours of incubation with the HMDMs pre-exposed to iron or Hb (Fig 2⇓). No significant formation of TBARS took place in the absence of LDL. TBARS did not increase in parallel with the amount of LDL, possibly due to toxic effects of LDL at high concentrations or a cellular inability to oxidize more than a certain amount per unit of time (Fig 3⇓).
The capacity of media conditioned for 24 hours by HMDM cultures that were or were not pre-exposed to iron or Hb was also studied with respect to the induction of TBARS after the addition of LDL. Media from iron- or Hb-exposed cells induced considerably more TBARS than those of control cells (data not shown).
Agarose Gel Electrophoresis
After 24 hours of incubation with HMDM cultures pre-exposed to iron compounds or Hb, the mobility of LDL was much increased compared with native LDL. LDL exposed for 24 hours to conditioned medium from cells previously exposed to iron or Hb showed increased electrophoretic mobility (Fig 4⇓).
When HMDM cultures were pre-exposed for 24 hours to FeCl3, Fe-ADP, Fe-EDTA, or Hb and then incubated with LDL in fresh medium without serum for 48 hours, large accumulations of lipid droplets were apparent in the cells after they were stained with Nile red. The HMDMs adopted a foam cell–like morphology (Fig 5⇓), and their content of neutral lipids and phospholipids increased significantly compared with the controls (Fig 6⇓). The accumulation of lipid droplets was enhanced by increasing the concentrations of LDL.
Control HMDMs and HMDMs exposed to FeCl3, Fe-ADP, or Fe-EDTA (100 μmol/L Fe3+) or Hb (25 or 50 μg/mL) were incubated for 48 hours with LDL. Lipofuscin-specific autofluorescence was markedly increased in all LDL-exposed cells, although cultures exposed to iron or Hb showed significantly higher autofluorescence after LDL incubation than cells that were not (Fig 7⇓). A linear correlation was found between lipofuscin formation and the concentration of FeCl3 to which the HMDMs had been pre-exposed (r=.87, P=.0054) (Fig 8⇓).
We wanted to learn how iron is involved in the process of LDL oxidation by human macrophages, whether iron compounds may be exocytosed after endocytotic uptake, and if such exocytosis would affect LDL oxidation and uptake by macrophages. We demonstrated that HMDMs that are initially exposed to simple iron compounds or Hb in nontoxic concentrations under standard culture conditions for 24 hours can exocytose iron into their surrounding medium and enhance cell-mediated LDL oxidation and that the HMDMs themselves show morphological transformation into foam cells.
HMDMs were found to continuously produce superoxide anion radicals for a prolonged time when exposed to LDL. The finding of a somewhat decreased production of superoxide by the iron- or Hb-exposed cells compared with unexposed cells may reflect a cellular adaptation mechanism secondary to iron sequestration, as suggested by Olakanmi et al.21 The combined production of superoxide and exocytosis of iron would allow the reduction of the iron from the ferric to the ferrous form with the possibility of further iron-catalyzed oxidative reactions, including Fenton chemistry, that might lead to LDL oxidation.
LDL is modified into a high-uptake form by human monocytes and HMDMs,22 23 24 although its subsequent uptake by macrophages has seldom been reported. The mechanisms behind cell-mediated LDL modification are not entirely clear, but they seem to involve lipid peroxidation. A range of cell types, particularly macrophages/monocytes, are capable of generating superoxide radicals and inducing LDL oxidation in culture; this occurs only in iron-containing media, such as Ham’s F10, and not in RPMI-1640, which contains much less iron and copper.1 24 Atherosclerotic lesions sometimes contain detectable iron and copper in a form that promotes lipid peroxidation.7 8 Many researchers, investigating various systems of lipid peroxidation in vitro, have found that the addition of simple iron complexes, which allow iron to act as a redox cycler, is required for lipid peroxidation.25 26 27 Fuhrman et al28 report that iron-induced lipid peroxidation of macrophages increases cell-mediated LDL oxidation and enhances the release of interleukin-1β and apoE without causing cell death. These findings agree with those of the present study.
Iron bound to EDTA is highly effective in promoting production of hydroxyl radicals by the superoxide-driven Fenton reaction,29 and Sakurai et al30 have reported a direct correlation between iron-ADP concentration and LDL peroxidation. We observed that both of these iron compounds had pronounced effects on LDL oxidation and uptake by HMDMs.
About two thirds of the body iron is within hemoglobin, and hemoglobin concentration may influence cardiovascular disease.31 32 33 Hemoglobin and myoglobin may act as catalysts of the Fenton reaction; these metalloproteins have been described as “frustrated oxidases.”34 Rice-Evans and colleagues35 36 37 found that hemoglobin, myoglobin, and ruptured erythrocytes may stimulate oxidation and uptake of LDL by macrophages, and they concluded that iron found in atherosclerotic lesions may be derived from heme proteins released from lysed cells. Our results provide new evidence for this hypothesis.
The presence of lipofuscin (ceroid) pigment in macrophage-derived foam cells is a well-known finding in human atherosclerotic lesions. Extracellular lipoprotein oxidation may be necessary to promote the intracellular appearance of lipofuscin.38 39 We found that LDL and iron both play important roles in macrophage and foam-cell lipofuscinogenesis. To our knowledge, our quantitative data on the relation between lipofuscin accumulation and LDL-HMDM interactions in the presence of iron are the first to have been presented. Our results are also consistent with findings that augmentation of iron in the culture medium of myocytes markedly increases the level of lipofuscin accumulation after iron sequestration in secondary lysosomes.40
In summary, under some conditions HMDMs exocytose iron that may be derived from previously phagocytosed structures rich in metalloproteins, eg, erythrocytes. Such exocytosis may enhance oxidation and uptake of LDL and induce foam cell formation. The present results suggest that iron does play a critical role in atherogenesis. Our experimental model seems to be a suitable system for quantitative and morphological studies of LDL oxidation by human macrophages and the formation of foam cells.
Selected Abbreviations and Acronyms
|FCS||=||Fetal calf serum|
|HBSS||=||Hanks’ balanced salt solution|
|HMDM||=||Human monocyte–derived macrophage|
|TBARS||=||Thiobarbituric acid–reactive substances|
This work was supported by grants from the Swedish Medical Research Council (No. 4481 and 6962). We thank Britt Sigfridsson for help with LDL preparation and agarose gel electrophoresis, Uno Johansson for technical assistance with the fluorescence microscopy, and Dazhong Yin and Dr Bo Ziedén for stimulating discussions.
Leake D, Rankin S. The oxidative modification of low-density lipoproteins by macrophages. Biochem J. 1990;270:741-748.
Heinecke JW, Baker L, Rosen H, Chait A. Superoxide-mediated modification of low density lipoprotein by arterial smooth muscle cells. J Clin Invest. 1986;77:757-761.
Hiramatsu K, Rosen H, Heinecke JW, Wolbauer G, Chait A. Superoxide initiates oxidation of low density lipoprotein by human monocytes. Arteriosclerosis. 1987;7:55-60.
Minotti G, Aust SD. The requirement for iron (III) in the initiation of lipid peroxidation by iron (II) and hydrogen peroxide. J Biol Chem. 1987;262:1098-1101.
Halliwell B, Gutteridge JMC, Cross CE. Free radicals, antioxidants, and human disease: where are we now? J Lab Clin Med. 1992;6:598-620.
Dubick MA, Hunter GC, Casey SM, Keen CL. Aortic ascorbic acid, trace elements, and superoxide dismutase activity in human aneurysmal and occlusive disease. Proc Soc Exp Biol Med. 1987;184:138-143.
Smith C, Mitchinson MJ, Aruoma OI, Halliwell B. Stimulation of lipid peroxidation and hydroxyl-radical generation by the contents of human atherosclerotic lesions. Biochem J. 1992;286:901-905.
Havel RJ, Eder HA, Bragdon JH. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Invest. 1955;34:1345-1353.
Böyum A. Isolation of lymphocytes, granulocytes, and macrophages. Scand J Immunol Suppl. 1968;8:5-15.
Kumagai KK, Hinnuma IS, Tada M. Pretreatment of plastic petri dishes with fetal calf serum: a simple method for macrophage isolation. J Immunol Methods. 1979;67:17-25.
Amador E, Dorfman LE, Wacker WEC. Serum lactic dehydrogenase activity: an analytical assessment of current assays. Clin Chem. 1963;8:391-399.
Allison AC, Harington JS, Birbeck M. An examination of the cytotoxic effect of silica on macrophages. J Exp Med. 1966;124:141-154.
McCord JM, Fridovich I. Superoxide dismutase: an enzymatic function for erythrocuprein (hemocupreine). J Biol Chem. 1969;244:6049-6055.
Koren E, Franzen J, Fugate RD, Alaupovic P. Analysis of cholesterol ester accumulation in macrophages by the use of digital imaging fluorescence microscopy. Atherosclerosis. 1990;38:175-184.
Sohal R, Marzabadi MR, Galaris D, Brunk UT. Effect of ambient oxygen concentration on lipofuscin accumulation in cultured rat heart myocytes: a novel in vitro model of lipofuscinogenesis. J Free Radic Biol Med. 1989;6:23-30.
Nobel RP. Electrophoretic separation of plasma lipoproteins in agarose gel. J Lipid Res. 1968;9:693-700.
Olakanmi O, McGowan SE, Hayek MB, Britigan BE. Iron sequestration by macrophages decreases the potential for extracellular hydroxyl radical formation. J Clin Invest. 1993;91:889-899.
Cathcart MK, Morel DW, Chisolm GM. Monocytes and neutrophils oxidized low density lipoprotein making it cytotoxic. J Leukoc Biol. 1985;38:341-350.
Cathcart MK, McNally AK, Morel W, Chisolm GM III. Superoxide anion participation in human monocyte-mediated oxidation of low density lipoprotein and conversion of low density lipoprotein to a cytotoxin. J Immunol. 1989;142:1963-1969.
Borg DC, Schaich KM. Iron and hydroxyl radicals in lipid oxidation: Fenton reactions in lipid and nucleic acids co-oxidized with lipid. In: Cerutti PA, Fridovich I, McCord JM, eds. Oxy-Radicals in Molecular Biology and Pathology. New York, NY: Alan R. Liss, Inc; 1988:427-441.
Balla G, Jacob HS, Eaton JW, Belcher JD, Vercellotti GM. Hemin: a possible physiological mediator of low-density lipoprotein oxidation and endothelial injury. Arterioscler Thromb. 1991;11:1700-1711.
Paganga G, Rice-Evans C, Andrews B, Leake D. Oxidised low density lipoproteins convert oxyhaemoglobin from ruptured erythrocytes to reactive ferryl forms. Biochem Soc Trans. 1992;20:331S.
Mitchinson MJ, Ball RY, Carpenter KLH, Enright JH. Macrophages and ceroid in human atherosclerosis. Eur Heart J. 1990;11(suppl E):116-121.