Human Alveolar Macrophages and Granulocyte-macrophage Colony-stimulating Factor-induced Monocyte-derived Macrophages Are Resistant to H2O2 via Their High Basal and Inducible Levels of Catalase Activity*

Human alveolar macrophages (A-MΦ) and macrophages (MΦ) generated from human monocytes under the influence of granulocyte-macrophage colony-stimulating factors (GM-MΦ) express high levels of catalase activity and are highly resistant to H2O2. In contrast, MΦ generated from monocytes by macrophage colony-stimulating factors (M-MΦ) express low catalase activity and are about 50-fold more sensitive to H2O2 than GM-MΦ or A-MΦ. Both A-MΦ and GM-MΦ but not M-MΦ can induce catalase expression in both protein and mRNA levels when stimulated with H2O2 or zymosan. M-MΦ but not GM-MΦ produce a large amount of H2O2 in response to zymosan or heat-killed Staphylococcus aureus. These findings indicate that GM-MΦ and A-MΦ but not M-MΦ are strong scavengers of H2O2 via the high basal level of catalase activity and a marked ability of catalase induction and that catalase activity of MΦ is regulated by colony-stimulating factors during differentiation.

Human alveolar macrophages (A-M⌽) 1 can survive for a long duration (1)(2)(3)(4) to exposure to not only chemical pollutants and exogenous oxidants but also inflammatory mediators and endogenously generated reactive oxygen species (ROS) and play important roles in phagocytosis-mediated host defense against microbial infection via the airway (5,6). Superoxide dismutase, catalase, and glutathione are the main cellular ROS-degrading enzyme systems; superoxide dismutase converts superoxide radical (O 2 . ) into H 2 O 2 , which is metabolized by catalase and glutathione peroxidase. Previous studies indicated that these enzymes are abundant in A-M⌽ (7)(8)(9). However, the mechanism to maintain high antioxidant activities in A-M⌽ has not been understood because of its heterogeneity and lack of availability to study.
Colony-stimulating factors (CSFs) such as granulocyte-macrophage CSF (GM-CSF) and macrophage-CSF (M-CSF) play important roles in survival and differentiation of monocytes/ M⌽. Previously, we reported that CSFs such as GM-CSF and M-CSF stimulate M⌽ generation from human monocytes, but GM-CSF-induced M⌽ (GM-M⌽) and M-CSF-induced M⌽ (M-M⌽), however, are distinct in their morphology, cell surface antigen expression (c-fms, CD14, CD71, and 710F), and sensitivity to human immunodeficiency virus, type I (HIV-I) infection (10 -14). Other studies also demonstrated that human monocyte-derived GM-M⌽ and M-M⌽ are distinct in their expression of CD14, integrin, and antibody-dependent cellular cytotoxicity activity (15)(16)(17)(18). Numerous studies show that the phenotype of human A-M⌽ closely resembles that of GM-M⌽ in morphology (fried egg-like shape) (19), the expression of cell surface antigens (c-fms low , CD14 low , CD71 ϩ , and 710F ϩ ) (11, 13, 20 -22), and function (resistance to M⌽-tropic HIV-I infection) (12,23). In contrast, M-M⌽ are elongated and spindleshaped, express c-fms high and CD14 high , which are similar to the phenotype of anaerobic peritoneal M⌽ (11,20,21,24), and are sensitive to M⌽-tropic HIV-I infection (12). These findings suggest that CSF is one of the critical factors in the determination of phenotypical characteristics of tissue M⌽ in the human system, and CSF-induced monocyte-derived M⌽ are available to analyze tissue M⌽.
In the present study, we investigated whether the antioxidant states of GM-M⌽ are at similar levels to those of A-M⌽ by assessment of H 2 O 2 sensitivity and catalase activity. We found that GM-M⌽ express high basal and inducible levels of catalase activity, are highly resistant to H 2 O 2 compared with M-M⌽, and inhibit H 2 O 2 production when stimulated with microbial stimulants. We also observed that catalase activity and sensitivity to H 2 O 2 in GM-M⌽ are at similar levels to those in A-M⌽. These findings suggest that GM-CSF but not M-CSF induces a strong antioxidant system in human tissue M⌽ during the differentiation.
Preparation and Culture of Macrophages-Peripheral blood mononuclear cells were obtained from venous blood drawn from normal healthy volunteers as described previously (12,13). Briefly, peripheral blood mononuclear cells were isolated by centrifugation on a Ficoll-Metrizoate density gradient (Lymphoprep; Nycomed, Oslo, Norway) and then placed into monocyte-isolating plates (MSP plates; Japan Immunoresearch Laboratories, Co., Ltd., Takasaki, Japan) for 2 h at 37°C in a humidified 5% CO 2 atmosphere (CO 2 incubator). More than 97% of the recovered cells were judged to be monocytes based on morphology, nonspecific esterase staining (cells were stained using a kit for ␣-naphthyl butyrate esterase), CD14 positivity, and their ability to phagocytize latex particles. Monocytes (2.5 ϫ 10 5 per ml or 5 ϫ 10 5 per 2 ml in 12or 6-well tissue culture plates, respectively) were then cultured with a optimal concentration of GM-CSF (500 units/ml) or M-CSF (10 4 units/ ml) for 7 days at 37°C in a CO 2 incubator. During the culture, monocytes underwent morphologic changes characteristic of monocytes to M⌽ differentiation such as an increase in their size and adherence.
Human A-M⌽ (2.5 ϫ 10 5 per ml or 5 ϫ 10 5 per 2 ml in 12-or 6-well tissue culture plates, respectively) were obtained from healthy volunteers (non-smokers without pathogenesis) by bronchoalveolar lavage (4,25). All volunteers agreed with a document to permit the use of A-M⌽ in part of this study, as informed consent. A-M⌽ were incubated in plastic dishes for 1 h at 37°C in a CO 2 incubator, and non-adherent cells were removed by repeated washing.
Assessment of Cell Number and Viability-Cell viability was assessed by trypan blue dye exclusion. The number of adherent monocytes and monocyte-derived M⌽s was determined by the method described previously by Nakagawara and Nathan (26). Briefly, cultures were depleted of medium by gentle aspiration and then replenished with 1% (w/v) cetyltrimethyl ammonium bromide (Cetavlon; Wako Pure Chemical Industries, Ltd., Osaka, Japan) in 0.1 M citric acid with 0.05% (w/v) naphthol blue black (Sigma) at room temperature for 3 min. This treatment readily lysed the adherent cells and liberated stained intact nuclei, which were then counted using a TATAI hemocytometer (American Optical).
Measurement of Catalase Activity-Intracellular and extracellular catalase activity was measured according to the method described by Aebi (27). Briefly, M⌽ were cultured in the phenol red-free medium (Life Technologies, Inc.) supplemented with the indicated concentrations of M-CSF or GM-CSF. Culture supernatants were harvested at 48 h for the measurement of extracellular catalase. To measure the intracellular catalase, cell lysates were prepared with a specific lysis buffer (10 mmol/liter EDTA, 2% Triton-X, 0.05% deoxycholic acid in phosphate-buffered saline, pH 7.4) and then diluted with 50 mmol/liter phosphate buffer (pH 7.0). 2 ml of the diluted sample was dispensed into a quartz cube, followed by 1 ml of 30 mmol/liter H 2 O 2 in phosphate buffer (pH 7.0). Catalase activity was measured by the consumption of H 2 O 2 at 240 nm in a spectrophotometer (Graphicord UV-240; Shimadzu Co., Kyoto, Japan) at 20°C. The slope was converted into catalase activity units based on the standard curve of purified human erythrocyte catalase (HEC, 5 ϫ 10 4 units/mg, Lot. number 643793; Calbiochem-Novabiochem). The activity is shown as milliunits/ml per well (2.5 ϫ 10 5 cells) or units per mg of protein using a protein assay kit (Bio-Rad Laboratories, Hercules, CA).
Isolation of RNA and Northern Blot Analysis-Isolation of total RNA and Northern blot analysis were performed as described previously (13). Briefly, cells were lysed with denaturing solution containing 4 mol/liter guanidine thiocyanate, 25 mmol/liter sodium citrate (pH 7.0), 0.5% sarcosyl, and 0.1 mol/liter 2-mercaptoethanol. After transfer to a polypropylene tube, total RNA was extracted by sequential addition of 2 mol/liter sodium acetate (pH 4.0), water-saturated phenol, and chloroform-isoamylalcohol (49:1), followed by centrifugation at 10,000 rpm for 20 min at 4°C and then precipitated with isopropanol and ethanol. Total RNA (10 g/lane) was size-fractionated by electrophoresis after denaturation with 6% (v/v) deionized glyoxal and 50% (v/v) dimethyl sulfoxide and then transferred to a nylon membrane (Pall BioSupport, East Hills, NY). After cross-linking on the membrane under UV irradiation and boiling in 80 mmol/liter Tris-HCl (pH 8.0) for 5 min, the membrane was prehybridized in prehybridization buffer (5 Prime 3 3 Prime, Inc., Boulder, CO) and 50% formamide at 42°C for 3 h. Hybridization for catalase or ␤-actin transcript was performed in hybridization buffer containing a human catalase cDNA probe or ␤-actin cDNA probe as a standard. All probes were labeled using a multiprime DNA labeling system with [␣-32 P]dCTP (PerkinElmer Life Sciences). Blots were washed at 37°C in 2ϫ SSC and 0.1% SDS, 45°C in 2ϫ SSC, 0.2ϫ SSC in 0.1% SDS for 30 min each and then analyzed using a Fuji BAS 2000 bioimage analyzer (Fuji Photo Film Co., Ltd., Tokyo, Japan).
Western Blot Analysis-Cell lysates were prepared with sample buffer containing 4% SDS, 62.5 mmol/liter Tris-HCl (pH 6.8), 10% glycerol, 100 mmol/liter dithiothreitol and 0.005% bromphenol blue. Cell lysates (25 g protein/lane) were separated by 10% SDS-polyacrylamide gel electrophoresis and then transferred to an Immobilon P membrane (Millipore Corp., Bedford, MA) using a semidry electroblotting system (Bio CRAFT BE300, BIO CRAFT Corp., Tokyo, Japan). The membrane was blocked with non-fat milk (BlockAce; Dainippon Medical Corp., Osaka, Japan) at 4°C overnight to avoid nonspecific binding and then incubated at 4°C overnight with 1 g/ml of rabbit anti-HEC antibody (Athens Research and Technology, Inc., Athens, GA) or normal rabbit IgG. After four washes in Tris-buffered saline (10 mmol/liter Tris-HCl, pH 8.0, 150 mmol/liter NaCl) supplemented with 0.1% Tween 20, the membrane was incubated at room temperature for 1 h with horseradish peroxidase-conjugated goat anti-rabbit IgG (Santa Cruz Biotechnology). After four washes with Tris-buffered saline supplemented with 0.1% Tween 20, the specific bands were visualized with Amersham Pharmacia Biotech ECL reagent on Hyper ECL film (Amersham Pharmacia Biotech).
Measurement of H 2 O 2 Production-Cellular release of H 2 O 2 was detected by the semiautomated microassay reported by De la Harpe and Nathan (28). In brief, cultured M⌽ (5 ϫ 10 4 per 100 l in 96-well flat-bottom tissue culture plates) were rinsed with phosphate-buffered saline, 100 l of assay mixture (30 mmol/liter scopoletin (Sigma), 1 mmol/liter NaN 3 , 1 purpurogallin unit/ml horseradish peroxidase (Sigma) in Krebs-Ringer phosphate buffer (145 mmol/liter NaCl, 4.86 mmol/ liter KCl, 0.54 mmol/liter CaCl 2 , 1.22 mmol/liter MgSO 4 , 5.7 mmol/liter sodium phosphate) with 5.5 mmol/liter glucose) was dispensed into the wells. Immediately, after the addition of stimuli (zymosan or heat-killed Staphylococcus aureus), the plate was placed in a fluorometer (Titertek Fluoroskan II; Flow Laboratories Inc., McLean, VA), and fluorescence was recorded for each well (0 -60 min) at 37°C. H 2 O 2 release was calculated from the loss of fluorescence, using the following formula: ϫ S, where E 0 is the initial fluorescence reading for the well, E 60 is the fluorescence reading at 60 min, W is the fluorescence recorded in an empty well, C 0 and C 60 are the mean fluorescence readings in the cell-free control wells at 0 and 60 min, respectively, and S is the amount of scopoletin, 3 nmol, added to each well at the start of the assay.

Distinct Susceptibility of Monocyte-derived M⌽s and A-M⌽
to Exogenously Added H 2 O 2 -M⌽ generated from human monocytes by CSF (M-M⌽ and GM-M⌽) and A-M⌽ were cultured in the medium containing the indicated concentrations of H 2 O 2 for 48 h and then cell viability was determined. There was a marked difference in their susceptibility to exogenously added H 2 O 2 (Fig. 1). When M-M⌽ were treated with 10 and 1 mmol/liter H 2 O 2 , 100 and 75% of the cells died, respectively, whereas almost 100% of the cells were viable in 0.1 mmol/liter H 2 O 2 . In contrast, more than 90% of GM-M⌽ were viable even

Intracellular and Extracellular Catalase Activities and Catalase Gene Expression in Monocyte-derived M⌽s and A-M⌽-
Because GM-M⌽ and A-M⌽ were markedly more resistant to H 2 O 2 than M-M⌽, we examined the levels of cell associatedand extracellular catalase activity that catalyze H 2 O 2 to H 2 O in these M⌽s. Extracellular catalase activity in the culture supernatants obtained from GM-M⌽ incubated for 48 h was about 4-fold higher than that from M-M⌽ (about 160 and 40 milliunits/ml/well in GM-M⌽ and M-M⌽, respectively) ( Fig. 2A). Culture supernatants obtained from A-M⌽ also contained high extracellular catalase activity (about 160 milliunits/ml/well), and the level was similar to that of GM-M⌽ ( Fig. 2A). In accordance with the findings of the enzyme activity, protein levels of extracellular catalase in GM-M⌽ and A-M⌽ cultures were about 4-fold higher than that in M-M⌽ cultures by Western blot analysis using anti-HEC antibody (Fig. 2B). Similarly, catalase activity in M-M⌽ lysates at 24 h was about 1 units/mg protein, whereas those in GM-M⌽-and A-M⌽-lysates were about 5 units/mg protein. (Fig. 3A). In agreement with the enzyme activity, protein levels of catalase among these M⌽ lysates were significantly different; catalase protein levels in GM-M⌽ and A-M⌽ lysates were higher than that in M-M⌽ lysate, and the difference was about 5-fold (Fig. 3B). As the above findings suggest that expression of catalase gene is quite different between M-M⌽ and GM-M⌽ or A-M⌽, we examined the levels of catalase mRNA among these M⌽s at 24 h after their cultivation by Northern blot analysis. Catalase mRNA in GM-M⌽ was about 5-fold higher than that in M-M⌽, which was similar to that in A-M⌽ (Fig. 3C).

Oxidant Stress-or Microbial Stimulant-induced Catalase Gene Expression Is High in Both GM-M⌽ and A-M⌽ but Low in M-M⌽-Although
there is a significant difference in basal levels of catalase activity between M-M⌽ and GM-M⌽ or A-M⌽, the findings cannot fully explain their distinct susceptibility to H 2 O 2 ; the difference in catalase activity was 4 -5-fold, whereas the difference in sensitivity to H 2 O 2 was about 50-fold. We therefore examined whether oxidant stress triggers the augmented expression of catalase gene in monocyte-derived M⌽s. When these M⌽s were treated for 3 h with 0.1 mmol/liter H 2 O 2 , catalase mRNA in GM-M⌽ was augmented up to about 3-fold, whereas that in M-M⌽ did not change significantly (Fig. 4A). Next we examined whether zymosan stimulation induces catalase gene activation in these M⌽s. When M⌽s were stimulated for 3 h with 0.1 mg/ml zymosan, catalase mRNA in GM-M⌽, but not in M-M⌽, also increased about 3-fold (Fig. 4A).
To confirm that catalase protein was synthesized by induction of the catalase gene via oxidant stress or microbial stim- ulant, we conducted a Western blot analysis of catalase protein in M⌽ lysates (Fig. 4B). The levels of catalase protein in lysates of GM-M⌽ stimulated for 24 h with H 2 O 2 or zymosan increased up to about 3-fold, whereas such oxidant-triggered induction of catalase protein was not observed in lysates of M-M⌽ (Fig. 4B). Oxidant stress or microbial stimulant-mediated catalase induction in both gene and protein levels was also observed in A-M⌽ treated with H 2 O 2 or zymosan, and the induction level was similar to that in GM-M⌽ (Fig. 4). These findings indicate that GM-M⌽ and A-M⌽ have a marked ability to induce catalase expression in both gene and protein levels in response to H 2 O 2 or microbial stimulant, but M-M⌽ lacks this ability.  (31,32).

M-M⌽ Releases a Large Amount of H 2 O 2 , but GM-M⌽ Inhibits H 2 O 2 Release by Stimulation with Fungal or Bacterial
In the present study, GM-M⌽ and A-M⌽, but not M-M⌽, have a marked ability to induce catalase gene expression by exposure of low levels of H 2 O 2 and zymosan, which can augment their protection against oxidant-rich environments. Analysis of the 5Ј-flanking region of the catalase gene in human hepatoma cells and bronchoepithelial cells demonstrated that several transcriptional regulation sites in response to oxidant stress exist in the promoter region (33,34). These cells, however, express low levels of catalase activity, and hyperoxia fails to augment catalase transcript but induces transactivation of the heat shock protein 70 gene to support their survival (35)(36)(37). The present study is the first to report that M⌽s such as A-M⌽ and GM-M⌽ have a marked ability to induce catalase gene expression in response to oxidant stress. The precise mechanism, however, remains unknown.
A marked difference in catalase activity as one of the antioxidant systems between GM-M⌽ and M-M⌽ contributes to the generation of M⌽ heterogeneity during the differentiation of monocytes under the influence of CSF. Previous studies demonstrated that morphology, the expression of cell surface antigens, and resistance to HIV-I infection of GM-M⌽ resembled those of human A-M⌽ (11, 12, 14, 19 -23). In the present study, we also show that catalase activity and H 2 O 2 sensitivity of GM-M⌽ also resembled those of human A-M⌽. An important role of GM-CSF in A-M⌽ function was also reported in GM-CSF or GM-CSF receptor gene knockout mice (38 -42). These findings and those of the present study strongly suggest that GM- CSF plays a critical role in the development of ROS scavenging ability via catalase activity in human A-M⌽.
We demonstrated that GM-M⌽ are resistant to H 2 O 2 and a weak producer of H 2 O 2 by bacterial and fungal stimuli via high catalase activity. In contrast, M-M⌽s produce and release a large amount of H 2 O 2 because of their low catalase activity. We previously reported that M-M⌽ has a great capacity to produce HIV-I PAR whereas GM-M⌽ inhibits HIV-I PAR replication (12). Numerous studies have shown that ROS, including H 2 O 2 , trigger HIV-I replication via NF-B transactivation in the HIV-I long terminal repeat promoter region (43,44). Furthermore, a critical role of H 2 O 2 in NF-B-mediated HIV-I replication was confirmed by reduction of HIV-I replication with the scavengers, including catalase in human monocyte/M⌽ lineage cells (45,46). In some studies, exposure to bacterial products rendered M⌽ highly susceptible to T lymphocyte-tropic HIV-I via production of endogenous ROS and proinflammatory cytokines (47). These findings suggest that the difference in catalase activity between M-M⌽ and GM-M⌽ is a critical factor in the determination of their susceptibility to HIV-I replication.
In summary, we present evidence that catalase contributes to protect human tissue M⌽ from oxidant-induced cell death and control their respiratory burst, and the activity is regulated at both the protein and mRNA levels by CSF during their differentiation. GM-CSF but not M-CSF plays a critical role in the induction of a strong antioxidant mechanism. The comparison of GM-M⌽ and A-M⌽ with M-M⌽ in response to ROS helps clarify the self-defense mechanism of M⌽ against oxidant stress in vivo.