Glutathione peroxidase activity and mRNA

Glutathione peroxidase activity and mRNA expression in eosinophils and neutrophils of asthmatic and non-asthmatic subjects Neil L. A. Misso, Darryl J. Peroni, D. Neil Watkins, Geoffrey A. Stewart,* and Philip J. Thompson Asthma and Allergy Research Unit, Departments of Medicine and *Microbiology, University of Western Australia, Queen Elizabeth II Medical Centre, Perth, Australia

Abstract: Asthma has been reported to be associated with a reduction in the activity of glutathione peroxidase (GSH-Px), an important antioxidant enzyme. However, the expression of GSH-Px enzyme activity has not previously been investigated in human eosinophils, which are important inflammatory cells involved in asthma. Reverse transcriptasepolymerase chain reaction and Southern blotting demonstrated that eosinophils express GSH-Px mRNA and the relative expression of GSH-Px was greater in eosinophils than in neutrophils for both asthmatic and non-asthmatic subjects. The presence of GSH-Px protein in eosinophil and neutrophil lysates was confirmed by size exclusion chromatography and by Western blotting. GSH-Px enzyme activity as measured by a spectrophotometric assay was greater in eosinophil (48.4 6 1.6 mmol NADPH oxidized · min21 · g21 protein) than in neutrophil lysates (18.1 6 0.4, n 5 24, P F 0.0001). GSH-Px activities of eosinophils and neutrophils from asthmatic subjects did not differ from those of nonasthmatic subjects. Eosinophil GSH-Px activity was correlated with peripheral blood eosinophil count only in asthmatic subjects (rs 5 0.59, n 5 12, P 5 0.04). Increased GSH-Px expression in eosinophils compared with neutrophils of asthmatic patients may provide antioxidant protection against the greater amounts of reactive oxygen species generated by these cells and may enhance the survival of eosinophils at sites of inflammation in asthma. J. Leukoc. Biol. 63: 124–130; 1998. Key Words: leukocytes · human · antioxidant enzyme

The potentially damaging effects of ROS in the lung are counteracted by a number of antioxidant systems. The glutathione redox cycle in which the tripeptide glutathione (GSH) in conjunction with the enzyme glutathione peroxidase (GSH-Px; EC 1.11.1.9) reduces hydrogen peroxide and lipid peroxides may be the most important of these respiratory antioxidant systems [7]. Indeed, GSH occurs in high concentrations in respiratory epithelial lining fluid [8, 9] and significant GSH-Px activity has been detected in human bronchial epithelial cells [10]. Cellular GSH-Px is a tetrameric protein in which each of the four identical subunits contains one atom of selenium (Se) in the form of selenocysteine at the active site [11]. Previous studies have shown that peripheral blood and platelet GSH-Px activities are reduced in aspirin-sensitive asthmatic patients [12–15] and some, but not all, studies have also demonstrated this in atopic asthmatic patients [15–19]. In our study of an Australian population, we showed that reduced GSH-Px activity was associated with reduced serum Se concentrations in atopic asthmatic patients, although the reasons for the reduced Se levels remain unclear [17]. Neutrophils and eosinophils are generally regarded as important sources of ROS such as superoxide anion and hydrogen peroxide [20, 21] and thereby contribute to inflammatory processes within the lung. However, these cells also possess antioxidant enzymes such as catalase [22] that may protect these cells from the damaging effects of oxidants released at sites of inflammation. Such antioxidant activity could be particularly important in asthma where the antioxidant capacity of the bronchial epithelium may be reduced by epithelial damage, and eosinophils are recruited into the lungs in considerable numbers. Although there have been limited studies investigating GSH-Px activity in human neutrophils [23, 24], studies specifically examining GSH-Px activity in human eosinophils have not been performed. This may partly be due to the difficulty of isolating pure populations of eosinophils in sufficient numbers for bio-

INTRODUCTION Reactive oxygen species (ROS) produced by inflammatory cells have been implicated in the pathogenesis of lung diseases such as asthma, cystic fibrosis, adult respiratory distress syndrome, and idiopathic pulmonary fibrosis [1]. Chronic inflammation and influx of inflammatory leukocytes into the airways in asthma results in increased generation of ROS in asthmatic patients [2, 3] and it is likely that ROS play a significant role in the pathophysiology of asthma [4–6]. 124

Journal of Leukocyte Biology

Volume 63, January 1998

Abbreviations: GSH-Px, glutathione peroxidase; RT-PCR, reverse transcriptasepolymerase chain reaction; ROS, reactive oxygen species; GSH, glutathione; HBSS, Hanks’ balanced salt solution; PIPES, 1,4-piperazinediethanesulfonic acid; FCS, fetal calf serum; BSA, bovine serum albumin; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; NP-40, Nonidet P-40; ECL, enhanced chemiluminescence. Correspondence: P. J. Thompson, Asthma and Allergy Research Unit, University Department of Medicine, Queen Elizabeth II Medical Centre, Nedlands WA 6009, Australia. Received June 11, 1997; revised August 7, 1997; accepted September 8, 1997.

chemical assay of GSH-Px activity, but the recent development of an immunomagnetic technique has greatly increased the ease of isolation of such cells in high yield and purity from the peripheral blood of healthy, non-atopic subjects [25]. Thus, the aim of this study was to determine whether peripheral blood eosinophils express GSH-Px activity and to compare the activity with that of neutrophils. Because previous work suggested a reduction in GSH-Px activity in atopic, asthmatic subjects, eosinophil GSH-Px activity was also compared in atopic, asthmatic and non-atopic, non-asthmatic subjects.

METHODS Subjects Healthy, non-atopic, non-asthmatic subjects (n 5 12, mean age 36.7 years, range 20–59, 7 females) and atopic, asthmatic subjects (n 5 12, mean age 35.7 years, range 29–49, 7 females) who were clinically stable and not taking oral corticosteroids, provided peripheral blood for the isolation of neutrophils and eosinophils. The asthmatic subjects were asymptomatic at the time of blood collection, but all these subjects had a confirmed diagnosis of asthma with severities ranging from mild to moderate. Four subjects used b-agonists alone as their regular medication while the other eight subjects regularly used a combination of b-agonist and inhaled corticosteroid medication. The study was approved by the Committee for Human Rights of the University of Western Australia.

Isolation of neutrophils and eosinophils The negative immunomagnetic selection technique was used with minor modifications [25, 26]. Heparinized blood (60 mL) was diluted (1:1) in Hanks’ balanced salt solution (HBSS) without Ca21 or Mg21 and aliquots (30 mL) were layered onto 15 mL of Percoll (density 1.088). After centrifugation (1,000 g, 30 min, 4°C), plasma and lymphocytes were removed and erythrocytes were lysed by resuspension in 20 mL ice-cold water for 30 s, after which 20 mL of 2 x PIPES buffer (25 mM PIPES, 50 mM NaCl, 5 mM KCl, 40 mM NaOH, 5.4 mM glucose, pH 7.4) were added. The lysis step was repeated and the granulocytes were then washed in HBSS with 2% (v/v) heat-inactivated fetal calf serum (HBSS/FCS). After centrifugation, the pellet of granulocytes was cooled on ice and anti-CD16 immunomagnetic beads (Miltenyi Biotech, Bergisch-Gladbach, Germany; 80 µL per 108 granulocytes) were added. The mixture was incubated for 45 min on ice, then 10 mL of cold HBSS/FCS was added and the suspension transferred onto a column in the magnetic field of a MACS system (Miltenyi Biotech). The granulocyte mixture was allowed to enter the column with the use of a 22-gauge needle to control the flow rate, and unlabeled eosinophils were eluted with a further 35 mL of HBSS/FCS. The column was then removed from the magnet and labeled neutrophils were eluted with 35 mL of HBSS/FCS. The eosinophil and neutrophil suspensions were centrifuged and resuspended in 2 mL of HEPESbuffered HBSS with 0.25% bovine serum albumin (BSA). Cell counts were performed on a Coulter counter and purity was evaluated by differential counts on Diff-Quik-stained cytospin preparations. Viability as assessed by trypan blue exclusion was consistently .95%.

RNA extraction and reverse transcription Cell pellets containing 0.5 to 12 3 106 eosinophils or 20 to 30 3 106 neutrophils were lysed in 1 mL of RNAzolB (Bresatec, Adelaide, Australia) and total cellular RNA was extracted by a modification of the method of Chomczynski and Sacchi [27] as supplied by the manufacturer. RNA was quantitated by spectrophotometry and equal amounts of neutrophil and eosinophil RNA were heated with random hexamer primers (Promega, Madison, WI) at 70°C for 5 min, then placed on ice. First strand cDNA synthesis was performed by the addition of avian myeloblastosis virus reverse transcriptase (Promega, 10 U), 8 mM MgCl2, 1 mM each of dATP, dCTP, dGTP, and dTTP (dNTPs), RNAsin (Promega, 10 U), and buffer to a volume of 20 µL. After incubation at 42°C for 1 h the samples were heated at 75°C for 5 min and the cDNA was stored at 280°C.

Detection of GSH-Px mRNA by polymerase chain reaction (PCR) Intron-spanning primers for cellular GSH-Px [28] were synthesized on a Milligen Cyclone Plus DNA synthesizer (Australian Neuromuscular Research Institute, Perth, Australia). These primers (sense: 58-GGG-GCC-TGG-TGG-TGC-TCGGCT-38 and anti-sense: 58-CAA-TGG-TCT-GGA-AGC-GGC-GGC-38) produce a 354-bp product. As a control, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was co-amplified in the same PCR reaction using sense (58-GGT-CGG-AGT-CAACGG-ATT-TGG-TCG-38) and anti-sense (58-CCT-CCG-ACG-CCT-GCT-TCACCAC-38) primers that produce a 788-bp product. PCR amplifications were performed in a DNA Thermal Cycler (Perkin Elmer, Melbourne, Australia) in 25-µL reactions containing 0.1 µg of neutrophil or eosinophil cDNA, GSH-Px, and GAPDH primers (100 ng of each), 1.2 mM MgCl2, 0.25 mM dNTP mix, 1 U of Amplitaq DNA polymerase (Perkin Elmer), and PCR buffer. After initial denaturation at 94°C for 5 min, amplification was performed for 35 cycles of annealing at 63°C for 1 min, extension at 72°C for 2 min, and denaturation at 94°C for 30 s. cDNA synthesis reaction mixtures from which reverse transcriptase had been omitted were used as negative controls. After amplification, PCR products (5 µL) were electrophoresed on a 2% agarose gel containing ethidium bromide and viewed on a 300-nm UV transilluminator.

Preparation of probes for Southern hybridization Pooled GSH-Px and GAPDH PCR products were electrophoresed on 2% agarose and the appropriate bands (354 and 788 bp, respectively) were excised. The DNA was purified (Geneclean II kit, BIO 101 Inc., Vista, CA) and directly sequenced by the dideoxy terminator technique (Applied Biosystems 373A DNA Sequencing System, Foster City, CA) at the Australian Neuromuscular Research Institute, Perth, Australia. The nucleotide sequences were compared with the GenBank database and had 97.5 and 96.8% identity with the cDNA sequences for human GSH-Px1 and human GAPDH, respectively. The sequenced DNA fragments were amplified for 40 cycles using the appropriate primer pairs and the previously described PCR conditions with the addition of 0.02 mM biotinylated dUTP (dUTP-16-biotin) to the dNTP mixture. Aliquots of the resulting PCR products were electrophoresed on 2% agarose to confirm the presence of the biotinylated probes, which were quantitated by spectrophotometry.

Southern hybridization of GSH-Px and GAPDH PCR products After electrophoresis of PCR products, the gel was denatured in 0.4 M NaOH/0.6 M NaCl and neutralized in 1.5 M NaCl/0.5 M Tris-HCl, pH 7.5. The DNA was transferred to Hybond N1 nylon membrane (Amersham, Sydney, Australia) by capillary blotting in 10 3 SSC (1 3 SSC is 0.15 M NaCl, 0.015 M sodium citrate, pH 7) for 16 h. The membrane was fixed in 0.4 M NaOH for 1 min, washed in 2 3 SSC/1.5 M Tris-HCl, pH 7.5, and pre-hybridized at 42°C for 8 h in hybridization buffer (6 3 SSC, 5 3 Denhardt’s solution, 50% formamide, 20 µg/mL heatdenatured salmon sperm DNA). The membrane was then hybridized at 42°C for 16 h in the same buffer with the addition of biotinylated GSH-Px or GAPDH probe at a concentration of 100 ng/mL. After hybridization, the membrane was washed at 55°C in 2 3 SSC, 1% sodium dodecyl sulfate (SDS; 2 3 30 min) and then in 0.5 3 SSC, 0.1% SDS (2 3 30 min). The membrane was incubated for 30 min with streptavidinperoxidase (4 U/mL, Boehringer Mannheim, Germany) in buffer 1 (50 mM Tris-HCl, pH 7.5; 100 mM NaCl; 5% Triton X-100) followed by two 5-min washes in buffer 1 and two 5-min washes in buffer 2 (50 mM Tris-HCl, pH 7.5; 100 mM NaCl; 5% Triton X-100; 6% urea; 1% dextran). Signals were detected on photographic film by enhanced chemiluminescence (ECL, Amersham).

Detection of GSH-Px protein by size exclusion chromatography and Western blotting Eosinophils and neutrophils were isolated from the peripheral blood of five non-asthmatic subjects. Cell pellets containing 3 to 7 3 106 eosinophils or 6 to 14 3 107 neutrophils were lysed in 25 or 100 µL, respectively, of lysis buffer (2% NP-40, 10 mM ethylenediaminetetraacetic acid, 10 mM sodium azide, 5 mM dithiothreitol, 40 µg/mL aprotinin, 0.1 mM leupeptin, 4 mM Pefabloc SC, and 0.5 mM E-64 in 50 mM N-2-hydroxyethylpiperazine-N8-2-ethanesulfonic acid buffer, pH 7.5). After vortexing and brief sonication, lysates were centrifuged (15,000 g, 10 min, 4°C) and the supernatants were stored at 220°C. Cell lysates

Misso et al.

Glutathione peroxidase in eosinophils

125

from the five subjects were pooled and the eosinophil and neutrophil lysates (,175 µL) and pure human erythrocyte GSH-Px (Sigma) were injected onto a 300 3 7.8 mm Bio-Silect SEC 250 high-performance liquid chromatography column (Bio-Rad, Hercules, CA). The column was eluted at 1 mL/min with 0.05 M phosphate buffer, pH 6.8, containing 0.15 M sodium chloride and 0.01 M sodium azide and 0.5-mL fractions were collected and assayed for GSH-Px activity. The column was calibrated with proteins of known molecular weight (thyroglobulin, 670 K; BSA, 66 K; ovalbumin, 45 K; cytochrome C, 12.2 K) and GSH-Px molecular weights were estimated (6 5 K) by interpolation on a regression plot of log (molecular weight) versus retention time. Fractions containing GSH-Px activity (retention time 8–10 min) were pooled, concentrated (Centricon 10, Amicon, Beverly, MA), mixed with 2 3 sample buffer (1:1), heated at 100°C for 7 min, and proteins were separated by 12.5% SDS-polyacrylamide gel electrophoresis (PAGE) under reducing conditions. Human erythrocyte GSH-Px was also used as a positive control. Proteins were electroblotted onto nitrocellulose membranes (Bio-Rad) and, after incubation in blocking buffer [2% (w/v) skim milk in 0.05 M Tris-buffered saline with 0.05% Tween 20], the membranes were incubated with a mouse anti-human GSH-Px monoclonal antibody (GPX-347, MBL Co. Ltd., Nagoya, Japan) followed by biotinylated rabbit anti-mouse immunoglobulin secondary antibody (DAKO, Sydney, Australia). Membranes were then treated with streptavidin/horseradish peroxidase conjugate (DAKO) followed by chemiluminescence detection (ECL, Amersham).

Assay of GSH-Px activity Suspensions of neutrophils and eosinophils were centrifuged, then resuspended in 0.34 M sucrose, 20 mM Tris-HCl, and 0.1% Triton-X 100 and sonicated for 20 s on ice. The neutrophil and eosinophil lysates were assayed for GSH-Px activity using a coupled spectrophotometric assay [29]. Assays were performed in 1-mL cuvettes containing glutathione reductase (1 U), GSH (3 mM), sodium azide (3.8 mM), NADPH (0.2 mM), and 0.1–0.4 mL of cell lysate. The reaction was initiated by addition of t-butylhydroperoxide (0.3 mM) and the change in absorbance at 340 nm was monitored for 4 min on an Ultrospec III spectrophotometer (Pharmacia LKB, Sweden). Non-enzymatic oxidation of NADPH was measured in an identical assay system except that cell lysate was replaced with water. This blank value was subtracted from the cell lysate reaction rate to obtain the true enzyme activity, which was calculated as micromoles NADPH oxidized · min21 · g21 protein using a molar absorptivity for NADPH of 6.22 3 103 L · mol21 · cm21 and the measured protein concentrations of the cell lysates (Protein Assay, Bio-Rad). The specificity of this assay for GSH-Px activity was confirmed by performing the assay in the absence of glutathione (n 5 3) or the t-butylhydroperoxide substrate (n 5 3). In the absence of either of these substrates, the reaction rates for neutrophil and eosinophil lysates were lower than that for the non-enzymatic oxidation of NADPH, confirming that the enzyme activity measured by this assay was that of a glutathione-dependent peroxidase.

count in non-atopic, non-asthmatic subjects (0.15 3 109 L21, SD 0.08, n 5 12) but the difference was not statistically significant (P 5 0.14). Mean peripheral blood neutrophil counts in the non-asthmatic (3.55 6 1.03 3 109 L21) and asthmatic groups (3.98 6 1.05 3 109 L21) were similar. The mean purities of the eosinophil preparations were 98.5 (SD 1.3) and 97.4% (SD 2.3) in non-asthmatic and asthmatic subjects, respectively, while the neutrophil preparations were 98.9 (SD 1.3) and 98.7% (SD 1.3) pure, respectively.

Expression of GSH-Px mRNA in neutrophils and eosinophils RNA was extracted from neutrophils and eosinophils of four non-atopic, non-asthmatic subjects and four atopic, asthmatic subjects. Using the RT-PCR technique, GSH-Px mRNA was readily detected in all eight neutrophil and eosinophil preparations. Representative ethidium bromide-stained gels of the PCR products from neutrophils and eosinophils of a non-asthmatic and an asthmatic subject are shown in Figure 1. A single PCR product for GSH-Px with the predicted size of 354 bp and a product for GAPDH with the predicted size of 788 bp were detected in both neutrophil and eosinophil preparations. The specificity of the RT-PCR technique for detecting GSH-Px mRNA in these cells was confirmed by Southern blotting using a sequenced biotinylated probe (Fig. 1). Densitometric analysis of the autoradiographs obtained after Southern blotting indicated a GSH-Px/GAPDH PCR signal ratio of 0.62 6 0.1 in neutrophils and 0.92 6 0.14 in eosinophils from non-asthmatic subjects. Similarly, in asthmatic subjects, the GSH-Px/GAPDH signal ratio was 0.58 6 0.11 in neutrophils and 1.67 6 0.69 in eosinophils. The GSH-Px/GAPDH ratio for all eight subjects was significantly greater in eosinophils (1.35 6 0.4) than in neutrophils (0.60 6 0.07, P 5 0.016, Wilcoxon signed rank test, Fig. 2B). The mean density of the neutrophil GAPDH signals was 1093 6 129, whereas the eosinophil GAPDH signals had a mean density of 863 6 135. The respective mean densities for neutrophil and eosinophil GSH-Px signals were 620 6 87 and 894 6 169. Because neutrophil GAPDH had the highest mean density, the

Analysis of data and statistics Autoradiographs of Southern and Western blots were scanned (Umax Data Systems, Taiwan) and Scan Analysis software (Biosoft, Cambridge, UK) was used to digitize the densitometric data. GSH-Px Southern blot densitometric signals were expressed as a ratio of the corresponding GAPDH signal obtained by co-amplification in the same PCR reaction. Although RT-PCR as performed here is only semi-quantitative, it can be used to detect differences in GSH-Px mRNA level. This was demonstrated by amplifying increasing amounts of cDNA (0.05, 0.1, 0.15, 0.25 µg) using the same PCR conditions. There was a corresponding increase in the GSH-Px RT-PCR signal measured by densitometry of Southern blot autoradiographs (see Fig. 2A). The InStat statistics program (GraphPad Software, San Diego, CA) was used to generate mean (6 SEM) values and differences between means were assessed for statistical significance using the non-parametric Wilcoxon rank sum test for paired or unpaired data. Correlations were determined using Spearman’s rank correlation (rs).

RESULTS Isolation of neutrophils and eosinophils The mean peripheral blood eosinophil count in atopic, asthmatic subjects (0.33 3 109 L21, SD 0.28, n 5 12) was higher than the 126



Fig. 1. RT-PCR amplification of GSH-Px and GAPDH in neutrophils (Neu) and eosinophils (Eos) from (A) a non-asthmatic and (B) an asthmatic subject and the corresponding autoradiographs obtained after Southern blotting with biotinylated GSH-Px and GAPDH probes. cDNA synthesis reactions without reverse transcriptase were used as negative controls (2).

min21 · g21 protein) than in neutrophil lysates (18.1 6 0.4, P , 0.0001), and this difference was apparent in both the asthmatic and non-asthmatic groups (P 5 0.0005, Fig. 5). There were no differences between asthmatic and non-asthmatic subjects either in neutrophil or eosinophil enzyme activities. However, there was a significant positive correlation between peripheral blood eosinophil count and eosinophil lysate GSH-Px activity in the asthmatic (rs 5 0.59, n 5 12, P 5 0.04) but not in the non-asthmatic subjects (rs 5 20.22, n 5 12, P 5 0.5, Fig. 6). Neutrophil GSH-Px activity did not correlate with peripheral blood neutrophil count in either asthmatic or non-asthmatic subjects.

DISCUSSION

Fig. 2. (A) Detection of increasing amounts of GSH-Px mRNA by RT-PCR. Increasing amounts of cDNA were amplified by PCR, the products were subjected to Southern blotting, and the resulting signals were quantitated by densitometry. (B) Relative densitometric measurement of GSH-Px/GAPDH PCR products from neutrophils and eosinophils of all subjects (n 5 8), non-asthmatic (n 5 4), and asthmatic subjects (n 5 4). GSH-Px densitometric values from autoradiographs of Southern blots were expressed as ratios of the corresponding GAPDH densitometric values. Data are means 6 SEM. *Eosinophil ratio significantly greater than neutrophil ratio for all subjects, P 5 0.016.

effect of any underestimation of autoradiographic signals from these more dense bands would be that the measured neutrophil GSH-Px/GAPDH ratio was greater than the true ratio. Therefore, the difference in GSH-Px/GAPDH ratio for eosinophils compared with neutrophils may be even greater than estimated in this study.

Detection of GSH-Px protein by size exclusion chromatography and Western blotting When eosinophil and neutrophil lysates were chromatographed on a size exclusion column, the GSH-Px activity in neutrophils eluted with a retention time of 9.25 min, whereas eosinophil GSH-Px activity eluted at 8.75 min (Fig. 3). The estimated apparent molecular weight of the neutrophil protein was 100 K compared with 145 K for the eosinophil protein. On the same column, purified human erythrocyte GSH-Px eluted as two peaks with apparent molecular weights of 100 and 200 K (Fig. 3). SDS-PAGE and Western blotting of eosinophil and neutrophil lysate chromatographic fractions, using a monoclonal antibody specific for erythrocyte GSH-Px showed several immunoreactive bands ranging from 24 to 98 kDa (Fig. 4). The eosinophil lysate showed two major bands at 98 and 45 kDa and a doublet of less intense bands at 26 and 24 kDa that corresponded to the doublet of subunits observed with purified human erythrocyte GSH-Px. The neutrophil lysate showed three bands at 98, 45, and 28 kDa but only a single band at 24 kDa corresponding to the smaller subunit of the erythrocyte doublet.

GSH-Px enzyme activity in neutrophils and eosinophils The mean GSH-Px enzyme activity for all subjects (n 5 24) was significantly greater in eosinophil lysates (48.4 6 1.6 µmol ·

This study is the first to demonstrate that human eosinophils express a significant level of GSH-Px activity. This was confirmed by detection of GSH-Px mRNA by RT-PCR using probes based on the published sequence for cellular GSH-Px [30], by biochemical measurement of the enzyme activity in eosinophil lysates, and by detection of GSH-Px protein in eosinophil lysates. These data suggest that the eosinophil enzyme is similar to the erythrocyte enzyme, but differences in the physicochemical properties of the enzyme from both eosinophils and neutrophils were observed. Thus, during size exclusion chromatography, purified erythrocyte GSH-Px showed two peaks of enzyme activity; one corresponding to the expected tetramer at 100 kDa and a second major peak with an apparent molecular weight of 200 K, suggesting that the latter was a dimer of the basic four-subunit structure. The eosinophil and neutrophil GSH-Px had apparent molecular weights of 145 K and 100 K, respectively, on size exclusion chromatography, indicating that the eosinophil enzyme was present as a tetramer linked to a dimer,

Fig. 3. Detection of GSH-Px protein by size-exclusion chromatography. Eosinophil (W) and neutrophil (j) lysates and pure human erythrocyte GSH-Px (Q) were injected on a Bio-Silect SEC 250 HPLC column and eluted at 1 mL/min with 0.05 M phosphate buffer, pH 6.8, containing 0.15 M NaCl, 0.01 M NaN3. Fractions (0.5 mL) were collected and assayed for GSH-Px activity. The molecular weights and retention times of proteins used to calibrate the column are indicated.

Misso et al.


127

Fig. 4. Detection of GSH-Px protein by Western blotting. Pure human erythrocyte GSH-Px (lane 2) and pooled chromatographic fractions (8–10 min) from eosinophils (lane 3) and neutrophils (lane 4) were separated by 12.5% SDS-PAGE. After transfer to nitrocellulose, blots were probed with mouse antihuman GSH-Px antibody followed by biotinylated anti-mouse immunoglobulin, streptavidin/peroxidase, and ECL detection. Lane 1 shows biotinylated molecular mass markers.

whereas the neutrophil enzyme was eluting as a tetramer. This suggests that the neutrophil, but not the eosinophil enzyme, may have undergone a form of limited NH2-terminal proteolysis during isolation, similar to that described for bovine erythrocyte GSH-Px [31]. This proteolysis of 10–12 amino acids involves loss of a cysteine at position 2 that would otherwise be able to form a disulfide link with another subunit of the enzyme. In the purified erythrocyte enzyme, this proteolysis may have been partial, producing both the tetramer and two tetramers linked by a disulfide bridge. Differential NH2-terminal proteolysis may also partly explain the Western blotting results in this study. Depending on the extent of proteolysis during isolation, either the intact polypeptide, the NH2-terminal cleaved product that is ,2.8 kDa smaller, or varying proportions of both subunit forms may be observed on SDS-PAGE [31]. This explains the observation of a doublet of subunits at 26 and 24 kDa on Western blots of purified human erythrocyte GSH-Px. In eosinophil lysates, the extent of NH2terminal proteolysis was much less, producing only a faint band for the smaller 24-kDa subunit. In neutrophils, however, only this lower-molecular-mass subunit was observed, indicating complete

NH2-terminal proteolysis in neutrophil lysates. The much greater numbers of neutrophils lysed and, therefore, greater concentration of proteases such as elastase in the neutrophil lysates probably account for this. Elastase has been shown to quite specifically cleave the twelve NH2-terminal amino acids [31] and the concentrations of antiproteases used during cell lysis were probably insufficient to inhibit its activity. Both eosinophil and neutrophil lysates also showed highermolecular-weight immunoreactive bands on Western blotting and it seems quite likely that these indicate the presence of dimeric and tetrameric GSH-Px protein. Because SDS-PAGE was performed under reducing conditions that should have ensured complete disruption of disulfide links, an alternative possibility is that tissue transglutaminase catalyzed covalent cross-linking of cellular proteins such as GSH-Px may have been occurring in these cells. Tissue transglutaminase is a widely distributed enzyme that accumulates in cells undergoing apoptosis and forms cross-links between specific glutamyl and lysyl residues that are resistant to disruption by reducing agents [32]. In this study, therefore, the detection of a peak of GSH-Px activity on size exclusion chromatography, the detection of

Fig. 5. GSH-Px enzyme activities in eosinophil and neutrophil lysates from non-asthmatic (n 5 12) and asthmatic subjects (n 5 12). Bars indicate mean 6 SEM. In both groups eosinophil lysate GSH-Px activities were significantly greater than neutrophil lysate GSH-Px activities (P 5 0.0005).

128



Fig. 6. Correlations between eosinophil lysate GSH-Px activity and peripheral blood eosinophil count in (A) non-asthmatic (n 5 12) and (B) asthmatic (n 5 12) subjects. Correlations were determined using Spearman’s rank correlation (rs).

similar immunoreactive bands and NH2-terminal proteolytic cleavage to that obtained with pure erythrocyte GSH-Px on Western blotting, and the spectrophotometric measurement of a glutathione-dependent peroxidase activity in eosinophil lysates provide convincing evidence for the presence of GSH-Px protein in eosinophils. Further work is clearly required to fully characterize this eosinophil protein, but the requirement for large numbers of pure eosinophils and the low level of protein that is likely to be expressed in these cells, make such studies difficult. In asthma, large numbers of eosinophils are recruited into the lungs at localized sites of inflammation where they are the major source of ROS, cationic proteins, and inflammatory mediators [2, 21, 33]. Although ROS generation by eosinophils and neutrophils has been studied in considerable detail [20, 21], antioxidant and antioxidant enzyme metabolism in these cells have received much less attention. However, maintenance of the balance between ROS production and antioxidant capability may be essential to the integrity and function of neutrophils and eosinophils and it is likely that antioxidant enzymes such as GSH-Px within neutrophils and eosinophils are required to protect these cells from the toxic effects of oxidants released at sites of inflammation. In this study, neither eosinophil nor neutrophil GSH-Px activities were significantly different in atopic, asthmatic compared with non-asthmatic subjects contrasting with previous studies showing that platelet and/or whole blood (mainly erythro-

cyte) GSH-Px activities were significantly reduced in asthmatic subjects [15–17]. However, these latter differences were only apparent when large numbers of subjects (.40 in each group) were studied and it is possible similar differences may become apparent in other cells such as eosinophils if large numbers of subjects were studied. Alternatively, GSH-Px may not be reduced in eosinophils from asthmatics because mature leukocytes are capable of increasing the level of GSH-Px activity de novo in response to external stimuli, as suggested by the detection of GSH-Px mRNA in eosinophils and neutrophils. In contrast, mature platelets and erythrocytes that have no nuclei are unlikely to synthesize significant amounts of GSH-Px enzyme de novo. In asthmatic patients, GSH-Px activity may be inherently low in some cells [16, 17], whereas other cells may respond to increased levels of oxidant stress by up-regulating GSH-Px enzyme expression. However, in this study, we did not observe any evidence of increased GSH-Px activity in neutrophils and eosinophils from asthmatics compared with non-asthmatics, possibly because all the asthmatic subjects were asymptomatic and clinically stable. However, in the asthmatic subjects but not in the non-asthmatic subjects, there was a significant positive correlation between eosinophil GSH-Px activity and peripheral blood eosinophil count. Although this phenomenon may be related to eosinophilia in general rather than to asthmatic eosinophilia specifically, it raises the possibility that stimuli, such as increased levels of interleukin-5 or granulocytemacrophage colony-stimulating factor, that induce the formation and survival of greater numbers of eosinophils, may also be up-regulating GSH-Px expression and activity in these cells. It is interesting that, although eosinophil GSH-Px enzyme activities did not differ between asthmatic and non-asthmatic subjects, the GSH-Px/GAPDH PCR signal ratio appeared to be greater in asthmatic compared with non-asthmatic subjects. Expression of GSH-Px activity also inhibits apoptosis in certain cell types by limiting the cellular damage caused by lipid peroxides [34] and this may be contributing to prolonged eosinophil survival in asthmatic patients. GSH-Px activity, expressed on the basis of cellular protein content, was 2.5 times greater in eosinophil lysates than in neutrophil lysates. This was supported by the observation that expression of GSH-Px mRNA appeared to be greater in eosinophils than in neutrophils as the RT-PCR signal for GSH-Px, when expressed relative to that for GAPDH co-amplified in the same reaction, was consistently greater in eosinophils than in neutrophils. The biological significance of this greater antioxidant enzyme mRNA expression and activity in eosinophils is unclear. However, the activity of the NADPH oxidase responsible for the respiratory burst that generates ROS is reportedly three to six times greater in eosinophils than in neutrophils [35], suggesting that ROS production by eosinophils is greater than that of neutrophils [21]. Thus, it may be that eosinophils are exposed to greater concentrations of ROS or are more susceptible to oxidant-induced damage and therefore require greater antioxidant protection. Although a few studies have measured GSH-Px activity and investigated glutathione metabolism in human neutrophils [23, 24, 36, 37], there have been no such studies in human Misso et al.


129

eosinophils. The expression of GSH-Px activity in eosinophils suggests that these cells also contain, and are capable of synthesizing, intracellular GSH, which would not only serve as a substrate for GSH-Px but also function as an intracellular antioxidant and regulator of the production of eosinophil-derived mediators such as leukotriene C4, which are known to be important in asthma [38, 39]. Therefore, further studies are required to investigate the expression and activity of the enzymes involved in GSH synthesis (g-glutamylcysteine synthetase) and metabolism (GSH-Px) in eosinophils. Regulation of these enzyme activities may play an important role in determining the antioxidant capacity of eosinophils and hence their survival and function as producers of ROS and inflammatory mediators in diseases such as asthma.

ACKNOWLEDGMENTS We thank the Medical Research Fund of Western Australia for financial support and Ms. Jacque El-Saleh for technical assistance.

REFERENCES 1. Cross, C. E., Van Der Vliet, A., O’Neill, C. A., Eiserich, J. P. (1994) Reactive oxygen species and the lung. Lancet 344, 930–933. 2. Djukanovic, R., Roche, W. R., Wilson, J. W., Beasley, C. R. W., Twentyman, O., Howarth, P. H., Holgate, S. T. (1990) Mucosal inflammation in asthma. Am. Rev. Respir. Dis. 142, 434–457. 3. Vachier, I., Chanez, P., Le Doucen, C., Damon, M., Descomps, B., Godard, P. (1994) Enhancement of reactive oxygen species formation in stable and unstable asthmatic patients. Eur. Respir. J. 7, 1585–1592. 4. Kanazawa, H., Kurihara, N., Hirata, K., Takeda, T. (1991) The role of free radicals in airway obstruction in asthmatic patients. Chest 100, 1319–1322. 5. Barnes, P. J. (1990) Reactive oxygen species and airway inflammation. Free Rad. Biol. Med. 9, 235–243. 6. Tonnel, A. B., Wallaert, B. (1990) Oxidants and bronchial inflammatory processes. Eur. Respir. J. 3, 987–988. 7. Maier, K. L. (1993) How the lung deals with oxidants. Eur. Respir. J. 6, 334–336. 8. Heffner, J. E., Repine, J. E. (1989) Pulmonary strategies of antioxidant defense. Am. Rev. Respir. Dis. 140, 531–554. 9. Cantin, A. M., North, S. L., Hubbard, R. C., Crystal, R. G. (1987) Normal alveolar epithelial lining fluid contains high levels of glutathione. J. Appl. Physiol. 63, 152–157. 10. Kinnula, V. L., Yankaskas, J. R., Chang, L., Virtanen, I., Linnala, A., Kang, B. H., Crapo, J. D. (1994) Primary and immortalized (BEAS 2B) human bronchial epithelial cells have significant antioxidative capacity in vitro. Am. J. Respir. Cell Mol. Biol. 11, 568–576. 11. Sunde, R. A. (1990) Molecular biology of selenoproteins. Annu. Rev. Nutr. 10, 451–474. 12. Malmgren, R., Unge, G., Zetterstrom, O., Theorell, H., De Wahl, K. (1986) Lowered glutathione peroxidase activity in asthmatic patients with food and aspirin intolerance. Allergy 41, 43–45. 13. Pearson, D. J., Suarez-Mendez, V. J. (1990) Abnormal platelet hydrogen peroxide metabolism in aspirin hypersensitivity. Clin. Exp. Allergy 20, 157–163. 14. Hasselmark, L., Malmgren, R., Unge, G., Zetterstrom, O. (1990) Lowered platelet glutathione peroxidase activity in patients with intrinsic asthma. Allergy 45, 523–527. 15. Pearson, D. J., Suarez-Mendez, V. J., Day, J. P., Miller, P. F. (1991) Selenium status in relation to reduced glutathione peroxidase activity in aspirinsensitive asthma. Clin. Exp. Allergy 21, 203–208. 16. Flatt, A., Pearce, N., Thomson, C. D., Sears, M. R., Robinson, M. F., Beasley, R. (1990) Reduced selenium in asthmatic subjects in New Zealand. Thorax 45, 95–99. 17. Misso, N. L. A., Powers, K. A., Gillon, R. L., Stewart, G. A., Thompson, P. J. (1996) Reduced platelet glutathione peroxidase activity and serum selenium concentration in atopic asthmatic patients. Clin. Exp. Allergy 26, 838–847.

130



18. Stone, J., Hinks, L. J., Beasley, R., Holgate, S. T., Clayton, B. A. (1989) Reduced selenium status of patients with asthma. Clin. Sci. 77, 495–500. 19. Plaza, V., Prat, J., Rosello, J., Ballester, E., Ramis, I., Mullol, J., Gelpi, E., Vives-Corrons, J. L. L., Picado, C. (1995) In vitro release of arachidonic acid metabolites, glutathione peroxidase, and oxygen-free radicals from platelets of asthmatic patients with and without aspirin intolerance. Thorax 50, 490–496. 20. Smith, J. A. (1994) Neutrophils, host defense, and inflammation: a doubleedged sword. J. Leukoc. Biol. 56, 672–686. 21. Kroegel, C., Virchow, J. C., Jr., Luttmann, W., Walker, C., Warner, J. A. (1994) Pulmonary immune cells in health and disease: the eosinophil leucocyte (Part 1). Eur. Respir. J. 7, 519–543. 22. Dvorak, A. M., Ishizaka, T., Weller, P. F., Ackerman, S. J. (1993) Ultrastructural contributions to the understanding of the cell biology of human eosinophils: mechanisms of growth factor-induced development, secretion, and resolution of released constituents from the microenvironment. In Eosinophils: Biological and Clinical Aspects (S. Makino and T. Fukuda, eds.), Boca Raton, FL: CRC Press, 13–32. 23. Cohen, H. J., Brown, M. R., Hamilton, D., Lyons-Patterson, J., Avissar, N., Liegey, P. (1989) Glutathione peroxidase and selenium deficiency in patients receiving home parenteral nutrition: time course for development of deficiency and repletion of enzyme activity in plasma and blood cells. Am. J. Clin. Nutr. 49, 132–139. 24. Tarp, U., Stengaard-Pedersen, K., Hansen, J. C., Thorling, E. B. (1992) Glutathione redox cycle enzymes and selenium in severe rheumatoid arthritis: lack of antioxidative response to selenium supplementation in polymorphonuclear leukocytes. Ann. Rheum. Dis. 51, 1044–1049. 25. Hansel, T. T., De Vries, I. J. M., Iff, T., Rihs, S., Wandzilak, M., Betz, S., Blaser, K., Walker, C. (1991) An improved immunomagnetic procedure for the isolation of highly purified human blood eosinophils. J. Immunol. Meth. 145, 105–110. 26. Neeley, S. P., Hamann, K. J., White, S. R., Baranowski, S. L., Burch, R. A., Leff, A. R. (1993) Selective regulation of expression of surface adhesion molecules Mac-1, L-selectin, and VLA-4 on human eosinophils and neutrophils. Am. J. Respir. Cell Mol. Biol. 8, 633–639. 27. Chomczynski, P., Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156–159. 28. Apostolakos, M. J., Schuermann, W. H. T., Frampton, M. W., Utell, M. J., Willey, J. C. (1993) Measurement of gene expression by multiplex competitive polymerase chain reaction. Anal. Biochem. 213, 277–284. 29. Paglia, D. E., Valentine, W. N. (1967) Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. J. Lab. Clin. Med. 70, 158–169. 30. Sukenaga, Y., Ishida, K., Takeda, T., Takagi, K. (1987) cDNA sequence coding for human glutathione peroxidase. Nucleic Acids Res. 15, 7178. 31. Gettins, P., Dyal, D., Crews, B. (1992) Selenium-dependent glutathione peroxidases from ovine and bovine erythrocytes occur as longer chain forms than previously recognized. Arch. Biochem. Biophys. 294, 511–518. 32. Nagy, L., Thomazy, V. A., Chandraratna, R. A., Heyman, R. A., Davies, P. J. (1996) Retinoid-regulated expression of BCL-2 and tissue transglutaminase during the differentiation and apoptosis of human myeloid leukemia (HL-60) cells. Leukemia Res. 20, 499–505. 33. Sanders, S. P., Zweier, J. L., Harrison, S. J., Trush, M. A., Rembish, S. J., Liu, M. C. (1995) Spontaneous oxygen radical production at sites of antigen challenge in allergic subjects. Am. J. Respir. Crit. Care Med. 151, 1725–1733. 34. Hockenbery, D. M., Oltvai, Z. N., Yin, X. M., Milliman, C. L., Korsmeyer, S. J. (1993) Bcl-2 functions in an antioxidant pathway to prevent apoptosis. Cell 75, 241–251. 35. Dechatelet, L. R., Shirley, M. R., McPhail, L. C., Huntley, C. C., Muss, H. B., Bass, D. A. (1977) Oxidative metabolism of human eosinophils. Blood 50, 525–535. 36. Niwa, Y., Iizawa, O., Ishimoto, K., Akamatsu, H., Kanoh, T. (1993) Age-dependent basal level and induction capacity of copper-zinc and manganese superoxide dismutase and other scavenging enzyme activities in leukocytes from young and elderly adults. Am. J. Pathol. 143, 312–320. 37. Bilzer, M., Lauterburg, B. H. (1991) Glutathione metabolism in activated human neutrophils: stimulation of glutathione synthesis and consumption of glutathione by reactive oxygen species. Eur. J. Clin. Invest. 21, 316–322. 38. Takafuji, S., Bishoff, S. C., De Weck, A., Dahinden, C. A. (1991) IL-3 and IL-5 prime normal human eosinophils to produce leukotriene-C4 in response to soluble agonists. J. Immunol. 147, 3855–3861. 39. Rouzer, C. A., Scott, W. A., Griffith, O. W., Hamill, A. L., Cohn, Z. A. (1981) Depletion of glutathione selectively inhibits synthesis of leukotriene C by macrophages. Proc. Natl. Acad. Sci. USA 78, 2532–2536.