INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Iron acquisition is necessary for normal cell function; however, unbound intracellular iron can catalyze the generation of highly reactive hydroxyl radicals from less reactive oxygen species, and promote oxidative cell injury (1). Storage of intracellular iron within the iron-storage protein ferritin, which is synthesized in response to iron uptake, provides substantial protection against iron-catalyzed injury (2). However, recent studies indicate that newly acquired iron that is in transit prior to storage within ferritin can enhance the susceptibility of cells to oxidant injury, presumably by increasing the intracellular content of catalytic iron (3).

The presence of extracellular iron in alveolar structures of healthy subjects has been reported by Gutteridge and colleagues as well as by other investigators (4, 5). Alveolar iron concentrations are increased in several chronic and acute respiratory conditions, including cigarette smoking, emphysema, the period following lung transplantation, and acute respiratory distress syndrome (4). Extracellular iron present in the lower respiratory tract of healthy subjects, as well as in patients with acute respiratory distress sydrome, is predominantly bound to transferrin and is not redox-active (4). However, uptake of even transferrin-delivered iron by cells can enhance cellular susceptibility to oxidant injury after intracellular release of iron from transferrin (3).

Alveolar macrophages (AM) can take up iron in various forms, including transferrin-bound iron, low-molecular-weight chelates of iron, and iron present in inhaled particles, and can sequester large amounts of iron in ferritin (8, 9). The sequestration of iron in ferritin within AM may protect other alveolar cells against oxidant injury, since sequestration of extracellular catalytic iron in ferritin inhibits the generation of extracellular hydroxyl radicals (10). Human AM normally contain a substantial amount of iron, and sequestration of iron within AM is markedly enhanced in smokers (11).

Exposure of cells to hyperoxia promotes intracellular accumulation of endogenous, reactive oxygen intermediates, and prior studies have indicated that oxidant stress contributes to hyperoxia-induced injury of AM (12). An increased AM content of unbound iron following iron uptake could enhance, at least transiently, the intracellular generation of hydroxyl radicals in association with exposure to hyperoxia. Prior studies have not assessed the potential of macrophage iron uptake to modulate the susceptibility of AM to hyperoxic injury. Among toxic effects of hyperoxia on AM are inhibition of various cell functions, including the respiratory burst, bactericidal activity, and phagocytosis, and prolonged hyperoxic exposure induces cell death (14). In the studies reported here, we sought to assess the effect of iron uptake by AM on susceptibility to these known manifestations of hyperoxic injury, including effects on phagocytic function and cell death. In addition, we provide new information on the effects iron uptake and exposure to hyperoxia on intracellular concentrations of calcium in AM, since oxidant injury of AM is known to increase cytosolic calcium concentrations (18).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Recovery of AM

AM were recovered from healthy, nonsmoking subjects through bronchoalveolar lavage (BAL), as previously described (6). Total cell recovery was determined with a hemacytometer, and a cell differential profile was determined by counting 200 cells on a Wright-Giemsa-stained cytocentrifuge preparation. AM suspended at a concentration of 1 × 10⁶/ml in RPMI-1640 medium were plated in culture dishes (35 mm) and allowed to adhere. After 2 h, nonadherent cells were washed off and the medium was refreshed.

Exposure to Hyperoxia and Iron

AM were cultured in RPMI-1640 supplemented with fetal calf serum (FCS) (10%). The iron content of media supplemented with 10% FCS was determined to be 3 µM. In studies of the effects of iron supplementation, ferric chloride (FeCl₃; Sigma, St. Louis, MO) was added to media to produce final iron concentrations in the media of 10, 20 or 40 µM. These concentrations were chosen on the basis of preliminary studies showing a lack of signficant toxicity of iron to AM at these concentrations, as well as on the basis of concentrations of iron in lavage fluid recovered from normal subjects, which have ranged from 0.21 µM to 33.6 µM, and concentrations of iron in lavage fluid from patients with acute respiratory distress syndrome and following lung transplantation (4, 5). Since bronchoalveolar lavage fluid (BALF) dilutes alveolar epithelial fluid by as much as 100-fold, the concentrations of extracellular iron that we evaluated (10, 20, and 40 µM) represent a conservative estimate of iron concentrations in the alveolar epithelial fluid of these patients.

Exposure of AM to hyperoxia was done as previously described (19). In brief, AM were plated in 35-mm culture dishes at a density of 10⁶ cells/plate. The cells were incubated at 37° C in air and 5% CO₂ for 1 h. The plates were then washed gently with media to remove nonadherent cells. Subsequently, some culture plates were placed in airtight containers, which were flushed at 7 L/min for 5 min with 95% oxygen and 5% CO₂ or with 60% oxygen, 5% CO₂, and 35% N₂. Controls were exposed to 21% oxygen and 5% CO₂. Before the containers were opened after 6 h or 18 h of incubation, gas samples were drawn from each container and evaluated in a gas analyzer (IL-1306; Instrumentation Laboratory, Lexington, MA). If the PO₂ and PCO₂ in the gas samples were not within 10% of the initial infused gas concentrations, the samples were discarded. At the completion of incubation, media were collected for lactate dehydrogenase (LDH) analysis. Cells were collected from the culture plates with a rubber policeman, were sonicated for 30 s at 90W, and were then stored at 70° C until subsequent analysis.

Measurement of Iron and Isoferritins

The content of iron and isoferritins in AM was determined as previously described (6). The concentration of iron was determined in 25-µl samples in duplicate through a controlled coulometry method (Ferrochem II; Environmental Science Associates, Bedford, MA) as previously described (20). Working standards for iron were prepared from certified ferric chloride suitable for standardization through atomic absorption spectrometry (Fisher Scientific, Fair Lawn, NJ). This method has a sensitivity of 10 ng/ml.

Both L-type ferritin, the predominant ferritin present in AM, and H-type ferritin were measured (6). The L-type ferritin content of AM was measured with a solid-phase, two-sided immunoradiometric assay (IRMA), using antibodies to L-type ferritin (Hybritech, San Diego, CA). This assay has a sensitivity of 0.7 ng/ml of L-ferritin. The assay for H-ferritin was done with an enzyme-linked immunosorbent assay (ELISA) methodology using specific monoclonal antibodies that were developed with human heart as the ferritin source, as previously described (21). This assay has a sensitivity of 0.5 ng/ml. The proportion of H-ferritin reacting in the L-ferritin assay was 5.2%, which represents the minimum L-subunit composition of H-rich ferritin. The cell contents of iron and isoferritins were expressed as µg/mg protein and ng/mg protein, respectively. All studies were done in triplicate. For purposes of data analysis, values for AM from a single volunteer subject were considered as a single experimental value.

Yeast Phagocytosis Assay

The quantitation of phagocytosis by AM was based on previously published assay methods involving ingestion of yeast (Saccharomyces cerevisiae) and counting of intracellular yeast in 100 AM through light microscopy after 4 h of incubation at 37° C (22). Approximately 5 × 10⁷ yeast were added to cultures of AM. The period of 4 h was chosen after initial studies indicating substantial phagocytosis over this period without overloading of cells, which would make accurate counting of intracellular yeast difficult. After incubation with yeast, cells were washed with phosphate-buffered saline (PBS), and basic fuchsin (0.01%) was added to stain extracellular yeast. The number of AM containing yeast, and the numbers of yeast in each macrophage, were counted through light microscopy under high-power, with evaluation of serial random fields. The fuchsin stain allowed visual separation of adherent and ingested yeast. Values were expressed as a phagocytic index, which was determined as a ratio of average number of yeast per AM as compared with values for the same AM population incubated with yeast under normoxia and in control media (3 µM iron). Values determined with AM from a single volunteer subject were considered a single experiment. All studies were performed in duplicate.

LDH and Protein Assays

LDH was measured with a colorimetric method based on the conversion of pyruvic acid to lactic acid (Procedure 500; Sigma). The LDH content of culture supernatant was expressed as a percentage of the total LDH content of the sonicated cell layer after subtracting the LDH content of the medium. The protein content of cell layers was measured with a protein assay reagent consisting of bicinchoninic acid (BCA kit; Pierce, Rockford, IL), with bovine serum albumin (BSA) used as a standard. Values determined with AM from each volunteer subject were considered a single experiment.

Intracellular Calcium Measurement

Intramitochondrial-1 (indo-1), seminaphthorhodafluor (SNARF), bis- (o-aminophenoxy)ethane-N, N, N', N'-tetraacetic acid (BAPTA), and standard Ca²⁺/ethylene glycol-bis-(-aminoethyl ether)-N, N, N', N'-tetraacetic acid (EGTA) solutions were obtained from Molecular Probes (Eugene, OR). The methods used for dye loading and measurement of intracellular calcium concentration were as previously described (23). A solution of the two cell-permeant acetoxymethyl ester dyes indo-1 and SNARF was made at a concentration of 5 µmol/L in each dye, with 12.5% dimethyl sulfoxide and 0.02% pluronic F-127 in Dulbecco's modified Eagle's medium. AM that were allowed to adhere to coverslips were incubated in the dye solution for 30 min. Cells were then returned to culture medium containing 10% fetal bovine serum (FBS) for 1 h. This period provided for cleavage of the esterified dyes to their impermeant forms, and provided optimal levels of image intensity. AM were then placed into the temperature-controlled chamber of a four-channel emission fluorescence microscope. A region of interest (ROI) was defined for each single cell. Epifluorescence excitation was generated from two Nikon 75W xenon lamps, having a 350 nm and a 540 nm bandwidth filter, respectively (Chroma Technology, Brattleboro, VT). Emission light-wave pairs at 405/ 475 nm and 575/640 nm were captured simultaneously. Intracellular calcium concentrations and pH ([Ca²⁺]i and pHi) of individual cells were measured continuously by ratio image analysis of the indo-1 and SNARF fluorescence emissions as previously described (24). The emitted light-wave pairs from each ROI were analyzed with commercial software (MicroMeasure FL-4000; Belvoire Consulting, Long Beach, CA) capable of calculating uncorrected ratios in real time. These real-time ratios were displayed while the raw data was being recorded, thus permitting immediate evaluation of cell viability and changes in [Ca²⁺]i. The [Ca²⁺]i value of each ROI was derived by using calcium standards. Correction values for the pH dependence of the indo-1 dye were obtained from ratios to simultaneously measured SNARF as previously described (25, 26). Time dependent [Ca²⁺]i values for each ROI then were plotted from the corrected data (27). Values for effects of various exposures on [Ca²⁺]i were determined with AM from five different volunteer subjects.

Statistics

Data are expressed as mean ± SEM. Differences between groups were analyzed with the Mann-Whitney rank-sum test. In all tests, statistical significance was identified at the p < 0.05 level. For purposes of data analysis, studies were done with AM recovered from a single volunteer, and mean values were determined from the single values determined for each study subject.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Characterization of BALF Cells

Cell recovery in BALF and the baseline iron and ferritin contents of AM recovered from control subjects are provided in Table 1. AM comprised approximately 97% of recovered cells, and were > 95% viable by trypan blue testing. The iron and ferritin concentrations present in recovered AM were consistent with previously reported values for nonsmoking subjects (6, 11).

LDH Release

Incubation of AM under normoxia and in iron-supplemented media did not significantly increase cytolysis as indicated by release of LDH over 6 h or 18 h as compared with incubation in control media (3 µM iron) (Figure 1). In contrast, exposure to 95% oxygen in control media significantly increased the supernatant content of LDH at both 6 h and 18 h, whereas exposure to 60% oxygen had significant effects only at 18 h. The exposure of AM to either 60% oxygen or 95% oxygen in iron-supplemented media enhanced cytolysis as compared with incubation in control media (Figure 1). At both oxygen concentrations, increasing the iron content of the media from 3 µM to 10 µM enhanced LDH release at 18 h. Coexposure to media with an iron concentration of 40 µM and either 60% oxygen or 95% oxygen also enhanced LDH release at 6 h as compared with identical exposures in control media.

View larger version (48K): [in this window] [in a new window]   Figure 1.   Release of LDH by AM in cultures exposed for 6 h or 18 h to normoxia or hyperoxia (60% and 95% oxygen) in either control media or media supplemented with iron. Values represent results of at least eight experiments done with AM from different subjects. *p < 0.05 compared with normoxia.

Yeast Phagocytosis

Exposure of AM to yeast for 4 h allowed substantial phagocytosis of yeast, with a mean number of phagocytized yeast of 8/cell, and approximately 70% of AM demonstrating phagocytized yeast. Exposure of AM to media with increased concentrations of iron under normoxic conditions decreased phagocytic activity slightly at an iron concentration of 40 µM, but not at lower iron concentrations (Table 2). Exposure of AM to 60% oxygen or 95% oxygen for 18 h decreased the cells' phagocytic activity for yeast. Exposure of AM to these same oxygen concentrations and in iron-supplemented media enhanced the negative effects of hyperoxic exposure on yeast phagocytosis, with iron concentrations as low as 10 µM causing increased toxicity in 95% oxygen.

Effects of Exposures on [Ca²⁺]i of AM

Resting [Ca²⁺]i of the observed AM had a value of 190 ± 6 nM (mean ± SE), with little cell-to-cell variation. Exposure to hyperoxia or iron and hyperoxia caused significant increases in [Ca²⁺]i (Figure 2), with significantly greater increases following exposure to both iron and hyperoxia than to hyperoxia alone. The increases in [Ca²⁺]i started slowly, and the time course of [Ca²⁺]i responses varied from cell to cell during the first hour. Some cells showed a persistent, gradual elevation of [Ca²⁺]i during the first 10 min of exposure, whereas others started to show an increase in [Ca²⁺]i later. This kinetic pattern of [Ca²⁺]i elevation is usually seen during slow influx of Ca²⁺ from the extracellular space. A rapid, sharp, transient peak in [Ca²⁺]i which could be characteristic for [Ca²⁺]i release from inositol-1,4,5-trisphosphate (IP3)-sensitive intracellular stores was not seen under these conditions (data not shown). Exposure of AM for 18 h to hyperoxia, followed by addition of iron to cultures, induced a further slow increase in [Ca²⁺]i in most cells, with some cells showing a more rapid increase denoting more advanced membrane injury (Figure 3). Exposure of AM for 18 h to iron-supplemented media did not increase these cells' [Ca²⁺]i; however, subsequent incubation under hyperoxic conditions was associated with a slow increase in [Ca²⁺]i, with some cells showing a more rapid increase (Figure 4).

View larger version (11K): [in this window] [in a new window]   Figure 2.   Intracellular concentrations of calcium in AM exposed to normoxia, hyperoxia (95% oxygen), iron-supplemented media (40 µM), or iron-supplemented media and hyperoxia for 18 h. Values represent mean ± SE of values determined in studies with AM recovered from at least four different subjects, with determinations made under each condition with at least 50 AM from each subject.

View larger version (23K): [in this window] [in a new window]   Figure 3.   Intracellular calcium concentrations measured in AM exposed to hyperoxia for 18 h and then exposed to iron-supplemented media. There was an increased [Ca2+]i in most of the cells analyzed (> 200 nM) in association with exposure to hyperoxia. Following addition of iron, there was a further rapid increase of [Ca2+]i in one cell, indicative of a more severe cell injury.

View larger version (18K): [in this window] [in a new window]   Figure 4.   Intracellular calcium concentrations measured in AM exposed to iron-supplemented media (40 µM) for 18 h and then subjected to hyperoxia. The [Ca2+]i of cells was not increased following exposure to iron-supplemented media; however, after exposure to hyperoxia there was a slow increase in [Ca2+]i, with a more rapid increase in one cell, indicative of a more severe cell injury.

Iron Accumulation and Isoferritin Synthesis

Normoxia. Incubation of AM in FeCl₃-supplemented media (10 to 40 µM) was associated with a rapid, dose-dependent increase in iron content, whereas incubation in control media (3 µM iron) induced only minimal iron uptake (Figure 5). Iron uptake from iron-supplemented media was relatively rapid, with most of this uptake occurring by 6 h. Increases in AM iron content were accompanied by increases in both L-type ferritin (Figure 6) and H-type ferritin content (Figure 7), with a greater increase in L- than in H-type ferritin. The increase in AM content of both L- and H-type ferritins in response to iron loading occurred more slowly than the increases in iron content, with a substantial amount of ferritin synthesis occurring beyond 6 h of incubation. The ratio of L-ferritin to H-ferritin was approximately 7:1 in AM at baseline, but increased to approximately 11:1 after incubation in 10 µM iron under normoxic conditions, and to 16:1 after incubation in 20 µM iron. The ratio of L-ferritin to H-ferritin did not increase further at higher concentrations, and was approximately 15:1 after incubation in 40 µM iron.

View larger version (41K): [in this window] [in a new window]   Figure 5.   Iron uptake by AM cultured in iron-supplemented media and exposed to normoxia or hyperoxia for 6 h or 18 h. *p < 0.01 compared with normoxia. Values represent mean ± SE. Values determined for AM recovered from a single volunteer were considered as a single experiment; the data represent values determined from experiments with AM from at least four different subjects.

View larger version (27K): [in this window] [in a new window]   Figure 6.   Synthesis of L-type ferritin by AM cultured in iron-supplemented media and exposed to normoxia or hyperoxia for 6 h or 18 h. *p < 0.01 compared with normoxia. All values represent results of experiments with AM from at least four different subjects.

View larger version (27K): [in this window] [in a new window]   Figure 7.   Synthesis of H-type ferritin by AM cultured in iron-supplemented media and exposed to normoxia or hyperoxia for 6 h or 18 h. *p <0.01 compared with normoxia. All values represent results of experiments with AM from at least four different subjects.

Hyperoxia. The influence of hyperoxia on iron uptake by human AM is shown in Figure 5. Exposure to 95% oxygen decreased AM iron uptake as compared with that of controls at both 6 h and 18 h of incubation at all iron concentrations. There was less of an effect on iron uptake in AM exposed to 60% oxygen, with a significant decrease in iron uptake occurring only at 18 h at an iron concentration of 20 µM, although there was a trend toward decreased iron uptake at all iron concentrations. The effects of both 60% oxygen and 95% oxygen on ferritin synthesis were greater than their effects on iron uptake for both L-type ferritin (Figure 6) and H-type ferritin (Figure 7). The greater effects of hyperoxia on ferritin synthesis than on iron uptake resulted in the ratio of iron to ferritin being significantly increased in AM exposed to hyperoxia than in those exposed to normoxia (Table 3). In AM exposed to iron-supplemented media under conditions of both normoxia and hyperoxia, the uptake of iron was more rapid than the increase in ferritin synthesis, so that the cell ratio of iron uptake to ferritin synthesis was substantially greater at 6 h than at 18 h.

Effect of Transferrin on Iron Uptake, Isoferritin Synthesis, and LDH Release

Addition of human apotransferrin (200 µg/ml) to cultures caused a significant increase in AM iron uptake and synthesis of L-ferritins (Figure 8). There was also approximately a doubling of cellular H-ferritin content in the setting of apotransferrin supplementation over the values achieved without apotransferrin, although total amounts of H-ferritin synthesized remained substantially less than those of L-ferritin (data not shown). Iron uptake and L-ferritin synthesis were increased by additional apotransferrin in both normoxia and hyperoxia; although exposure to both 60% oxygen and 95% oxygen continued to decrease iron uptake and ferritin synthesis. Furthermore, the addition of transferrin to media supplemented with 40 µM iron did not significantly reduce the release of LDH by AM with exposure to 95% oxygen for 18 h (data not shown).

View larger version (20K): [in this window] [in a new window]   Figure 8.   Effect of apotransferrin supplementation (200 µg/ml) on iron uptake and L-ferritin synthesis by AM cultured in iron-supplemented media exposed to normoxia or hyperoxia for 18 h. All values represent results of experiments with AM from at least five different subjects.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The principal finding of the present study was that iron uptake enhances the susceptibility of AM to in vitro hyperoxic injury. Exposing AM to hyperoxia in media containing iron at concentrations as low as 10 µM enhanced the cells' susceptibility to hyperoxic injury as compared with exposure in media containing 3 µM iron. Prior studies have shown that hyperoxic injury to AM is mediated, at least in part by enhanced oxidant stress (13, 14). Increased intracellular concentrations of unbound iron could catalyze the intracellular generation of highly reactive hydroxyl radicals from less reactive oxygen species, and hyperoxia promotes the intracellular accumulation of endogenous reactive oxygen species (2, 12). Extracellular iron is present in normal alveolar structures, and its concentrations are increased in patients with some respiratory disorders; consequently, the present study suggests a possible role of iron uptake by AM in modulating susceptibility to oxygen toxicity in vivo both in lungs with normal extracellular concentrations of iron and in conditions marked by an increased alveolar iron content (4, 5).

The mean concentration of iron in BALF recovered from healthy subjects was reported by Gutteridge and colleagues to be 0.21 µM; however, other investigators have reported values as high as 33.6 µM (4, 5). Because BALF dilutes alveolar epithelial fluid by as much as a hundredfold, total iron concentrations in alveolar epithelial lining fluid of healthy subjects may be the same as or greater than the in vitro iron concentrations we utilized in the present study. Because earlier studies have not clearly defined the iron concentrations in alveolar epithelial lining fluid, it is unclear whether iron uptake by AM occurs to a significant extent in normal lungs, although normal AM contain substantial amounts of iron (11).

Iron concentrations within alveolar structures are increased substantially in patients with some respiratory disorders including acute respiratory distress syndrome and in the period after lung transplantation (4, 5). AM probably take up iron under these conditions, although prior studies have not documented the iron content of AM. AM are known to accumulate iron in several chronic conditions associated with increased alveolar iron content, such as cigarette smoking and alveolar deposition of iron-containing dusts (6, 9). The findings of the present study suggest that rapid accumulation of iron by AM acutely enhances these cells' susceptibility to hyperoxia, but it is unclear whether slower accumulation of iron in AM, such as may occur in cigarette smokers, also enhances the cells' susceptibility to hyperoxic injury.

The increase in hyperoxia-induced in vitro injury to AM in the presence of iron-supplemented media is most likely mediated by an increase in the intracellular pool of catalytic iron following iron uptake (3). Ferritin synthesis is induced in cells by iron uptake, and sequestration of iron in ferritin protects against iron-catalyzed injury (2). However, our studies indicate that ferritin synthesis is delayed as compared with iron uptake, so that cells have substantially more iron than ferritin after 6 h of exposure to iron. This finding is similar to the findings of other investigators who have assessed iron uptake and ferritin synthesis in response to exposure of other cell populations to iron-containing compounds (28). During the period of increased cellular iron content relative to ferritin content, ferritin may become saturated with iron, and newly acquired iron may therefore be catalytic.

The findings of the present study are consistent with those of prior studies assessing the toxicity of hyperoxia to AM, including the findings that hyperoxia impairs phagocytic activity and induces cell death (14). As noted in studies by Raffin and colleagues (16), phagocytic activity was substantially impaired by exposure of AM to even moderate concentrations of oxygen (60%). We found that exposure to both 60% and 95% oxygen in media with increased iron content enhanced the negative effects of hyperoxia on yeast phagocytosis and cell death. The potential role of alveolar iron in modulating hyperoxia-induced decreases in phagocytosis is substantial, as our data indicate that exposure of AM to 60% oxygen in iron-supplemented media (40 µM iron) impaired phagocytic activity substantially more than did exposure to 95% oxygen in control media (3 µM iron).

The finding that iron uptake and, to a greater extent, ferritin synthesis by AM are rapidly impaired by hyperoxia has implications in the setting of acute lung injury. AM are capable of storing large amounts of iron, and iron sequestration within newly synthesized ferritin inhibits the extracellular generation of hydroxyl radicals (10). The capacity of AM to sequester iron may be important in limiting the availability of iron in the setting of acute lung injury. Prior animal studies evaluating silica-induced acute lung injury have demonstrated that iron accumulation is rapidly induced within the lungs, and have implicated pulmonary iron accumulation in the pathogenesis of subsequent pulmonary fibrosis (29).

Exposure of AM to 60% or 95% oxygen in iron-supplemented media decreased the synthesis of L-type ferritin as compared with exposure to iron under normoxic conditions. L-type ferritin is the primary isoferritin induced in AM by iron loading and this type of ferritin is generally associated with long-term iron storage (30). In our study, hyperoxia did not shift the type of ferritin synthesized in response to iron loading because synthesis of H-type ferritin, which stores iron in a more metabolically available form, was decreased to an extent similar to that of L-type ferritin. Exposure of cells to reactive oxygen species in vitro, including superoxide and hydrogen peroxide, can promote the cells to synthesize ferritin via effects on the iron regulatory protein (31). In a prior study, increased expression of messenger RNA (mRNA) for L-type ferritin was found in lung tissue after exposure of rats to hyperoxia (32). In the present study, however, we found that exposure of AM to hyperoxia in iron-supplemented media decreased their synthesis of L-ferritin as compared with exposure to iron under normoxic conditions. The decrease in L-ferritin synthesis was probably a result of enhanced toxicity to AM exposed to both iron and hyperoxia.

In the studies reported here, we noted that exposure to hyperoxia was associated with accumulation of intracellular calcium in AM. Prior studies have indicated that oxidative injury to AM alters calcium homeostasis, with a resulting calcium overload (18, 33). We found that exposure of AM to hyperoxia (95%) for 18 h induced a significant increase in intracellular calcium concentrations. Although exposure to iron-supplemented media alone did not increase cytosolic calcium concentrations in AM, combined exposure to both iron and hyperoxia caused greater and more rapid cell accumulation of calcium than did hyperoxia alone. The pattern of increase was suggestive of an influx of calcium, which in the setting of oxidative injury to AM occurs through specific cell membrane Ca²⁺ channels (18, 33). The concentrations of intracellular cytosolic calcium that we noted in human AM at baseline and after injury by hyperoxia and iron are similar to those reported by Rojanasaku and colleagues in rat AM both before and after oxidative injury induced by hydrogen peroxide and iron (33). Increases in intracellular concentrations of calcium induced by exposure to iron and hyperoxia may contribute to cell injury and death by activating cellular proteases or through other degradative processes (34).

Patients with acute respiratory distress syndrome have an increased alveolar content of transferrin-bound iron (4). In our studies, we found that addition of apotransferrin to iron-supplemented media enhanced total iron uptake and ferritin synthesis by AM; however, cytolysis induced by hyperoxia and iron-supplemented media was not significantly decreased. This finding is consistent with findings in prior studies indicating that transferrin-delivered iron can promote oxidative cell injury (3). Since patients with acute respiratory distress syndrome are often treated with high inspired oxygen concentrations, it is possible that iron uptake by AM in these patients could increase susceptibility to hyperoxic injury. However, the findings of the present study may have limited relevance to patients with acute respiratory distress syndrome, since we used AM from healthy subjects, whereas the population of AM in patients with acute respiratory distress syndrome is substantially altered and may respond differently to exposure to iron or hyperoxia (35).

In summary, the study reported here demonstrates that iron uptake by AM substantially increases the susceptibity of these cells to hyperoxic injury. Enhanced hyperoxic injury occurred with exposure to iron concentrations as low as 10 µM, and hyperoxic injury was enhanced at oxygen concentrations of both 60% and 95%. Although prior studies have not yet clearly defined iron concentrations in the alveolar epithelial fluid of normal subjects or patients with acute respiratory disorders, these findings suggest that relatively low concentrations of extracellular total iron are associated with iron uptake by AM and with an increase in the susceptibility of these cells to hyperoxic injury.