Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Acidic functional groups at the surface of mineral oxide particles have the capacity to complex metals (1). The ligands can include -Al-OH, -Mg-OH, -Fe-OH, -Ti-OH, and -Si-OH. With silica, silanol groups (-Si-OH) function to complex metal cations. As a result of its abundance, that metal cation most likely to be complexed in both the environment of the earth's crust and within a living system is iron (2, 3). The silanol group can react with Fe³⁺ to form a coordination complex (three silanol groups complex one ferric ion). This complexation of the ferric ion to the surface of mineral oxides is incomplete, allowing the iron to participate in electron transfer that can lead to the generation of reactive oxygen species (4).

The increased availability of catalytically active iron after silica exposure potentially presents an oxidative injury to a living system (5, 6). Sequestration of reactive iron would, therefore, confer a protective effect. The intracellular storage of iron by macrophages can limit the potential for the generation of free radicals and cellular injury resulting from exposure to a metal chelate (7). Ferritin is considered a safe form of storage for iron because metal sequestered in this protein infrequently participates in electron exchange and oxidant generation (7). The isolation of iron in this chemically less reactive form within intracellular ferritin confers an antioxidant function to ferritin and, in certain cells, provides cytoprotection in vitro against oxidants (8).

Ferritin expression is regulated through a posttranscriptional mechanism (11). A specific sequence at the 5' untranslated end of ferritin messenger RNA (mRNA) called the iron responsive element (IRE) binds the iron regulatory protein (IRP), a cubane iron-sulfur cluster, when the IRP exists in the apoprotein form. Available iron reacts with the IRP to alter the conformation of this protein when it is complexed to the ferritin mRNA. The affinity of the protein for the mRNA is diminished and it is displaced, allowing translation of ferritin to proceed. This posttranscriptional regulation of ferritin allows the cell to respond rapidly to increased concentrations of iron by increasing the ferritin available to sequester the metal.

In vivo exposures to mineral oxide dusts can be associated with increased concentrations of ferritin (18). We tested the hypothesis that the in vitro exposure of human alveolar macrophages to silica increases the expression of ferritin in a posttranscriptional manner. The elevation of this storage protein could function to diminish oxidative injury after inhalation of these particles.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Materials

The silica dust (Min-U-Sil 5) used in all studies was obtained from US Silica (Berkeley Springs, WV). All other reagents were from Sigma Co. (St. Louis, MO) unless otherwise specified.

Preparation of Acid-washed Silica and Silica Particles with Complexed Zinc and Iron

To obtain a particle with diminished concentrations of metal cations complexed to the surface, 150 mg of silica was agitated in 10 ml of 1 N HCl for 1 h. This suspension was centrifuged at 1,200 × g for 10 min and the supernatant was removed. The dust was agitated in 10 ml distilled water and centrifuged at 1,200 × g for 10 minutes. After the supernatant was removed, the silica was washed with distilled water twice more, and the particles were stored at 4°C.

To test for differences in ferritin expression by macrophages exposed to metals with and without a capacity to catalyze oxidant generation, zinc and iron were complexed to the surface of acid-washed silica. Iron was selected because it is the transition metal in highest concentration in a living system that can support electron transport. Zinc was chosen to contrast with iron, as it assumes only one stable valence state and has a dissimilar complexation chemistry relative to iron. Fifty milligrams of acid-washed silica was added to either 10 ml of 100 µM ZnCl₂ or 10 ml of 100 µM FeCl₃ and agitated for 15 min. This suspension was centrifuged and the resultant dusts were washed in distilled water three times to provide silica dusts with surface-complexed zinc and silica with surface-complexed iron.

Concentrations of zinc and iron complexed to the surface of silica, acid-washed particles, silica with complexed zinc cation, and silica with complexed iron cation were measured by agitating 10 mg of dust in 5 ml of 1.0 N HCl for 1 h. The suspension was centrifuged at 1,200 × g for 10 min and zinc and iron concentrations in the supernatant were quantified using inductively coupled plasma emission spectroscopy (ICPES; model P40; Perkin-Elmer, Norwalk, CT). Results are reported in units of micromoles per gram of dust.

Measurement of Oxidative Stress

Oxidant generation by silica, acid-washed particles, silica with complexed zinc cation, and silica with complexed iron cation was measured using thiobarbituric acid (TBA)-reactive products of deoxyribose (4). The reaction mixture containing 1 mM deoxyribose, 1 mM H₂O₂, 1 mM ascorbate, and 100 µg/ml of the dust was incubated in saline at 37°C for 60 min with agitation and then centrifuged at 1,200 × g for 10 min. One milliliter of both 1.0% (wt/vol) TBA and 2.8% (wt/vol) trichloroacetic acid was added to 1 ml of supernatant, heated at 100°C for 10 min, cooled in ice, and the chromophore concentration determined by its absorbance at 532 nm.

Exposure of Alveolar Macrophages to Silica Particles

Healthy, nonsmoking male volunteers (n = 8), 18 to 35 yr of age, underwent fiberoptic bronchoscopy with lavage to procure human alveolar macrophages. The screening procedures for each subject included a Minnesota Multiphasic Personality Inventory, medical history, physical examination, chest X-ray, and routine hematologic and biochemical tests. In addition, none of the subjects had a history of asthma, allergic rhinitis, chronic respiratory disease, or cardiac disease. Subjects were excluded from the study if they had suffered a recent acute respiratory illness and were asked to avoid exposure to air pollutants such as tobacco smoke and paint fumes. Prior to participation in the study, subjects were informed of the procedures and potential risks and each signed a statement of informed consent. The protocol and consent form were approved by the University of North Carolina School of Medicine Committee on the Protection of the Rights of Human Subjects. The fiberoptic bronchoscope was wedged into a segmental bronchus of the lingula. Six 50-ml aliquots of sterile saline were instilled and immediately aspirated. The procedure was repeated on the right middle lobe, again using 300 ml of saline. Samples were put on ice immediately after aspiration and centrifuged at 300 × g for 10 min at 4°C. Cells from all aliquots were pooled, washed twice with RPMI 1640 supplemented with 0.025% gentamycin, and used immediately. Alveolar macrophages were incubated in 12-well plates (Costar, Cambridge, MA). After 2 h, nonadherent cells were removed and fresh medium either with or without particles added.

Measurement of Ferritin Protein Concentrations

Alveolar macrophages (1.0 × 10⁶) in 1.0 ml of RPMI 1640 supplemented with 0.025% gentamycin were exposed for 4 h either to medium, silica, acid-washed particles, silica with complexed zinc cation, or silica with complexed iron cation. The concentration of dust suspension was 100 µg/ ml. The supernatant was removed and assayed for light-subunit ferritin (L-ferritin). Cells were washed with phosphate-buffered saline (PBS) (Life Technologies, Grand Island, NY). Adherent cells were lifted by repeated pipetting of cold PBS, which included EDTA (1:5,000), onto the cells. Cells were then centrifuged at 600 × g for 10 min. This cell pellet was sonicated in 1 ml of PBS for 15 s to disrupt membranes. L-ferritin in these cell lysates was also measured.

L-ferritin was analyzed using a commercially available kit, controls, and standards from Microgenics Corporation (Concord, CA). The enzyme -galactosidase is split into two inactive fragments, which can recombine spontaneously to form a catalytically active enzyme. One fragment is covalently attached to ferritin-specific antibody in a manner that does not affect the reassociation of the enzyme. The binding of L-ferritin in the standard or sample to the antibody inhibits the reassociation of the enzyme. The presence of ferritin in the standard or the sample therefore reduces the -galactosidase formed in the assay reaction. Concentrations of L-ferritin are indirectly proportional to the amount of enzyme formed as monitored by the hydrolysis of the substrate chlorophenol red--D-galactopyranoside, which is measured by changes in absorbance at 550 nm. This assay was modified for use on the Cobas Fara II centrifugal spectrophotometer (Hoffman-La Roche, Branchburg, NJ).

Reverse Transcription-Polymerase Chain Reaction

Cells (1.0 × 10⁶) in 1.0 ml of RPMI 1640 supplemented with 0.025% gentamycin were exposed for 4 h to one of three media: acid-washed particles, silica with complexed zinc cation, or silica with complexed iron cation. The concentration of dust suspension was 100 µg/ml. The supernatant was removed and the cells washed twice with PBS. The cells were lysed with 4 M guanidine thiocyanate (Boehringer Mannheim, Indianapolis, IN), 50 mM sodium citrate, 0.5% Sarkosyl, and 0.01 M dithiothreitol. After dislodging the cells from wells with scrapers (Costar), lysates were sheared with four passes through a 22-gauge syringe. RNA was isolated by ultracentrifugation in a cesium chloride gradient, which included 5.7 M cesium chloride (Boehringer Mannheim) and 0.1 M EDTA. One hundred nanograms of total RNA was reverse transcribed (Moloney murine leukemia virus [Mo-MuLV] reverse transcriptase; Life Technologies). The resultant cDNA was amplified by polymerase chain reaction (PCR) for 24 and 25 cycles for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and L-ferritin, respectively, using gene-specific primers. Oligonucleotide sequences were synthesized using an Applied Biosystems 391 DNA synthesizer (Foster City, CA), on the basis of sequences published in the GenBank human DNA database. The following sense and antisense sequences were used:

GAPDH: 5' CCATGGAGAAGGCTGGGG 3' and 5' CAAATTGTCATGGATGACC 3'

L-Ferritin:  5' TTCCTCTCCGCTTGCAACCT 3' and 5' CACTCATCTTCAGCTGGCTTCT 3'

PCR-amplified DNA was analyzed using gel electrophoresis and ethidium bromide intercalation, visualized under ultraviolet (UV) light, and the resulting polaroid negative (type 55 film; Polaroid Corp., Cambridge, MA) quantitated using a Bio Image analyzer (Bio Image, Ann Arbor, MI). The intensity of the GAPDH DNA band (a housekeeping gene unaffected by the addition of any particle) for each sample was then used to normalize differences between samples. For each sample, the integrated optical densities of L-ferritin DNA bands were divided by that of the GAPDH DNA to rectify any errors in RNA quantitation.

Ferritin Stain

Alveolar macrophages (1.0 × 10⁶) in 1.0 ml RPMI 1640 supplemented with 0.025% gentamycin were exposed for 4 h to one of three media: acid-washed particles, silica with complexed zinc cation, or silica with complexed iron cation. The concentration of dust suspension was 100 µg/ml. The supernatant was removed and the macrophages were lifted by repeated pipetting of cold PBS, which included EDTA (1:5,000), onto the cells. Cells (0.200 ml of cell suspension) were cytocentrifuged onto slides and stained for ferritin using immunohistochemistry. Slides were treated to block endogenous peroxidase with hydrogen peroxide in methanol (1:16). Nonspecific staining of highly charged protein was blocked by incubation in normal goat serum diluted 1:20 in PBS with 1% bovine serum albumin (BSA) for 10 min at 37°C. The serum was removed and the primary antibody (rabbit -ferritin antibody; Dako, Carpinteria, CA) applied at a dilution of 1:100 in PBS with 1% BSA. This is an antibody specific for L-ferritin. After incubation at room temperature for 45 min at 37°C, slides were washed with PBS three times and goat anti-rabbit biotinylated IgG (Vector Laboratories, Burlingame, CA), 1:200 in PBS with 1% BSA, was applied for 30 min at room temperature. The cells were then incubated with peroxidase conjugated streptavidin (Jackson Laboratories, Bar Harbor, ME) (1:800 dilution) in 0.05 M Tris for 30 min at room temperature and rinsed in PBS. Aminoethyl carbazole (Biomeda Corp., Foster City, CA) was applied to the cells for 8 min at room temperature, then rinsed with distilled water. The counterstain employed was hematoxylin (Fisher Scientific, Raleigh, NC). Controls included stains of normal human spleen (positive control) and lung tissue without the polyclonal antibody added (negative control).

Statistics

Measurements were done in replicates of three. Data are expressed as mean values ± standard deviation. Differences between multiple groups were compared using analysis of variance (19). The post hoc test employed was Duncan's multiple range test. Tests of significance were two-tailed. Significance was assumed at P < 0.05. 

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

The Min-U-Sil 5 silica dust had 0.1 ± 0.2 and 8.6 ± 1.1 µmol Zn and Fe/g dust, respectively. This concentration of surface iron is greater than that demonstrated in a silica used previously (4, 6). In vitro exposure of this silica dust to H₂O₂, ascorbate, and deoxyribose resulted in the generation of oxidized products, which was reflected by increases in the absorbance at 532 nm (the A₅₃₂ of reaction mixtures including saline only was 0.02 ± 0.01, while the A₅₃₂ of those with Min-U-Sil 5 was 0.05 ± 0.01).

Incubations of human alveolar macrophages with silica resulted in no increase in cytotoxicity as reflected by a release of lactate dehydrogenase (9 ± 4 and 11 ± 2 units/ml after medium and silica, respectively). Exposure of the macrophages to silica was associated with increases in the concentration of L-ferritin protein in both the cell supernatants and the cell lysates (Figure 1). The amount of L-ferritin in the cell lysate was always slightly greater than that in the supernatant. Inclusion of the iron chelator deferoxamine (final concentration, 1.0 mM) in the reaction mixtures inhibited increases in L-ferritin protein concentrations after silica exposure whereas the identical concentration of radical scavenger dimethylthiourea (DMTU) had no effect (Figure 1). Deferoxamine inhibited elevations of the storage protein in both the supernatant and the cell lysate.

View larger version (15K): [in this window] [in a new window]   Figure 1.   Ferritin protein concentrations after 4 h of human alveolar macrophage incubation with either medium or Min-U-Sil 5 silica. Exposure of Min-U-Sil 5 silica resulted in a significant increase in L-ferritin protein concentrations in both the cell lysate and cell supernatants relative to incubations with medium. Inclusion of deferoxamine in the reaction mixture diminished concentrations of this storage protein while the radical scavenger DMTU had no effect. Solid and open bars designate concentrations in the cell lysate and supernatant, respectively. An asterisk indicates signficant differences when cell lysates and supernatants after a specific exposure are compared to cell lysate and supernatant after incubation with medium.

To test for a posttranscriptional regulation of ferritin protein expression, acid-washed particles, silica with complexed zinc cation, and silica with complexed iron cation were used. Concentrations of zinc cation complexed to the surface of the three particles were 0.0 ± 0.0, 15.3 ± 1.2, and 0.0 ± 0.0 µmol/g dust whereas concentrations of complexed iron cation were 0.0 ± 0.0, 0.0 ± 0.0, and 20.7 ± 0.8 µmol/g dust, respectively. The catalysis of oxygen-based free radicals increased with the complexation of iron at the silica surface. The absorbance values of malondialdehyde-like products of deoxyribose for acid-washed particles, silica with complexed zinc cation, and silica with complexed iron cation were 0.01 ± 0.01, 0.01 ± 0.01, and 0.08 ± 0.01, respectively. Those reaction mixtures with no particles included had A₅₃₂ values of 0.02 ± 0.01. Increases in the TBA-reactive products of deoxyribose after exposure of silica to iron support complexation of the metal rather than a precipitation of oxyhydroxides on the surface of the mineral dust.

Incubations of acid-washed particles, silica with complexed zinc cation, or silica with complexed iron cation with the alveolar macrophages did not result in cytotoxicity as reflected by release of lactic acid dehydrogenase. L-ferritin protein concentrations were increased in both the cell supernatants (Figure 2, open bars) and cell lysates (Figure 2, solid bars) after 4 h of exposure to silica with complexed iron cation only. Incubations of acid-washed silica and silica with complexed zinc cation did not increase L-ferritin protein relative to medium alone. Elevations in L-ferritin production after exposure to silica with complexed iron cation were confirmed by immunohistochemistry (Figures 3A through 3D). Elevations in staining for ferritin protein were evident only in the cytoplasm of those alveolar macrophages incubated with silica with complexed iron. Finally, to demonstrate possible transcriptional differences after exposure, cDNA was isolated from cells exposed for 4 h to medium, acid-washed particles, silica with complexed zinc cation, or silica with complexed iron cation. There were no significant differences between any of the exposures to medium and the three silica dusts, supporting a posttranscriptional control of L-ferritin expression (Figure 4).

View larger version (11K): [in this window] [in a new window]   Figure 2.   L-ferritin protein concentrations after 4 h of human alveolar macrophage incubation with medium, acid-washed particles (silica-H), silica with complexed zinc cation (silica-Zn), and silica with complexed iron cation (silica-Fe). Only exposures of silica with complexed iron resulted in a significant increase in L-ferritin protein concentrations in both the cell lysate and cell supernatant relative to incubations with medium. Solid and open bars designate concentrations in the cell lysate and supernatant, respectively. An asterisk indicates signficant differences when cell lysates and supernatants after a specific exposure are compared to cell lysate and supernatant after incubation with medium.

View larger version (121K): [in this window] [in a new window]   Figure 3.   Immunohistochemical staining of L-ferritin protein after 4 h of human alveolar macrophage incubation with (A) medium, (B) acid-washed particles, (C) silica with complexed zinc cation, and (D) silica with complexed iron cation. Alveolar macrophages exposed to silica with complexed iron demonstrated an increased staining for L-ferritin protein relative to medium, whereas acid-washed particles and silica with complexed zinc did not.

View larger version (10K): [in this window] [in a new window]   Figure 4.   Band intensity of ferritin cDNA after 4 h of human alveolar macrophage incubation with medium, acid-washed particles, silica with complexed zinc cation, and silica with complexed iron cation. There were no significant differences in the intensity of L-ferritin cDNA after 4 h of incubation when normalized to the GAPDH cDNA band.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Exposures of alveolar macrophages to silica resulted in an increased concentration of both intracellular and extracellular L-ferritin. Elevations in ferritin protein concentrations after in vitro incubations with silica are modest relative to increases in the concentration of this protein after exposures to erythrocytes and cigarette smoke (20, 21). This may reflect either dissimilarities in the duration of exposure or disparities in the capacity of metal chelates to affect ferritin expression. Similar to a previous investigation employing exposures of lens epithelial cells to iron (22), deferoxamine significantly diminished ferritin concentrations following incubation of macrophages with silica. The radical scavenger DMTU had no effect on ferritin concentrations. This result contrasts with that of a prior study, in which antioxidants inhibited an elevation in the storage protein after iron exposure (23). The lack of an effect by a radical scavenger suggests that the metal itself, rather than the associated oxidative stress, could be responsible for the increased expression of ferritin after in vitro silica exposure. The dependence of this elevation in ferritin production on iron was confirmed using acid-washed particles, silica with complexed zinc cation, and silica with complexed iron cation. Ferritin mRNA was unaffected by exposures to any silica dust. This supports a posttranscriptional control of ferritin expression in the human alveolar macrophage after exposure to silica. Such posttranscriptional regulation of ferritin expression in monocytes and macrophages after exposures to iron chelates has been previously demonstrated (24).

Iron is complexed to the surface of silica and can catalyze the generation of free radicals. The alveolar macrophage, using superoxide, can mobilize the metal from the surface of a mineral oxide by reducing it to the ferrous state (25). A higher concentration of reactive iron is likely to be the immediate result. Increased concentrations of available iron mobilize ferritin mRNA onto polysomes for translation by reacting with the apoprotein form of the IRP and displacing it from the IRE. An increased production of ferritin results. The metal can then be isolated in this storage protein. This allows a rapid sequestration of the metal in a chemically less reactive state (i.e., ferritin) relative to its initial state (i.e., complexed to the silica surface). The result of ferritin regulation by iron is a decrease in the concentration of reactive iron in the cell, thereby preventing iron toxicity. This capacity to sequester iron in apoferritin is not unique to macrophages but can be a property of many different cell types. If not cleared, the surface of the mineral oxide retains the capacity to complex metal and appears to mobilize iron both from cells in vitro (26) and a living system (6). The source of this metal is not known, but could include the intracellular pool of low molecular weight chelates. The host will respond with continued attempts to sequester the metal, employing ferritin, which will accumulate locally. This process is recognized histologically as a ferruginous body, which is the particle and the accumulated ferritin contiguous to the particle (e.g., the silica body). This formation of ferruginous bodies functions to protect the host by diminishing the oxidative injury mediated by metal complexed by mineral oxides. These results are consistent with previous investigations that have demonstrated diminished in vitro cytotoxicity and in vivo injury by ferruginous bodies relative to uncoated particles (27, 28).

Apoferritin was produced intracellularly in response to exposure to iron complexed to the silica particle. Significant concentrations of this storage protein were measured in the cell supernatant of alveolar macrophages. Some quantity of the ferritin must have been either released or secreted by the macrophages. This was unassociated with any evidence of cytotoxicity to these cells, suggesting this protein was not passively released. The concentrations of ferritin in the supernatant correlated with those in the cell lysates after 4 h of exposure to silica. Previous investigation has demonstrated a release/secretion of ferritin by alveolar macrophages that does not result from passive cell leakage (21). Concentrations of ferritin released/secreted by alveolar macrophages corresponded to those in the lavage fluid (21). Macrophages can increase the release/secretion of intracellular ferritin after iron loading. Following erythrophagocytosis, Kupffer cells release/secrete ferritin, which can then be taken up by hepatocytes (7, 20, 29). Twenty-four hours after exposure to sensitized erythrocytes, peritoneal macrophages release/secrete the majority of acquired iron as ferritin (30). The release/secretion of stored iron can be biphasic with a rapid early phase (approximately 4 h) and a slower prolonged phase (approximately 16 h) (29, 31). This release/secretion of ferritin by macrophages, with subsequent uptake by other cell types through a ferritin receptor, can be a mechanism to diminish iron availability by removing the storage protein from the circulation after cell necrosis (28). Alternatively, ferritin release can be involved in the transport and redistribution of metal in states of iron overload (28, 32). In this manner, metal that presented a potential to injure a living system is detoxified and utilized by that system to meet its own nutritional requirements. Such release/secretion explains the appearance of ferritin in extracellular compartments including serum, semen, and synovial fluid (33). The source of ferritin in serum, and other extracellular fluids, has not been determined. The concentration of this protein can reflect total body iron stores and is used for that purpose. Rather than indicating ferritin derived from senescent cells, this protein is more likely to result from an active secretion. Such transport of iron to meet the metabolic needs of a living system has been demonstrated in plants (34, 35) but the specific metal chelate was not identified.

In conclusion, there is a requirement for a mechanism to regulate the production of ferritin by cells exposed to mineral oxide dusts as a result of catalysis of oxygen-based free radicals by iron complexed to the surface of mineral oxide particles. Cells diminish the availability of reactive metal complexed to the mineral oxide, and subsequently the oxidative stress, by regulating its sequestration into ferritin. Exposures of human alveolar macrophages to silica increased the production of ferritin protein. The expression of this storage protein was controlled by a posttranscriptional mechanism that was dependent on the concentration of available iron.