INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Platelet-activating factor (PAF; 1-alkyl-2-acetyl-sn-glycero-3-phosphocholine) is a lipid mediator involved in the pathogenesis of various inflammatory diseases of the lung such as bronchial asthma, chronic obstructive pulmonary disease, lung fibrosis, and adult respiratory distress syndrome (1). In vivo administration of PAF reproduces many signs and symptoms observed in these diseases such as bronchoconstriction, bronchial hyperreactivity, enhanced mucus secretion and vascular permeability, and recruitment of inflammatory cells into the lung (2). Large quantities of PAF can be produced in human lung during inflammatory reactions by resident cells such as macrophages (3) and mast cells (4) and by cells migrating from the blood such as neutrophils, eosinophils, and monocytes (2). In vitro and in vivo studies have shown that nanomolar concentrations of PAF are sufficient to elicit most of its proinflammatory effects (5). Thus, highly efficient systems must be available in vivo to inactivate PAF and to limit the potential damage induced by an excessive accumulation of this inflammatory mediator.

The major enzyme responsible for the catabolism of PAF is acetylhydrolase (AH) (6, 7), a PAF-specific esterase that cleaves the acetate group at the sn-2 position of PAF producing lyso-PAF. AH was initially characterized as an enzyme present in large amounts in serum and plasma associated with low density lipoproteins and, to a lesser extent, with high density lipoproteins (7, 8). AH activity was also detected in various organs and tissues, including the kidney, the lung, the brain, and the liver (2, 6). Intracellular AH is a cytosolic enzyme present in various inflammatory cells such as mast cells, macrophages, and platelets (2). These cells release AH either spontaneously or after cell activation, and it has been speculated that they represent the source of extracellular AH reported in inflammatory fluids. For example, extracellular AH has been detected in the skin chamber fluid of allergic patients after antigen challenge (9), in the nasal lavage fluid after antigen challenge (10, 11), and in the perfusion fluid of hearts undergoing ischemia (12).

There is compelling evidence that AH plays a major role in modulating PAF activity in vivo. An increased plasma level of AH was found in spontaneously hypertensive rats, and it was thought to contribute to hypertension by accelerating the degradation of PAF (13). A congenital deficiency of plasma AH has been reported in children with bronchial asthma (14). Administration of recombinant human AH significantly reduced vascular leakage in PAF-induced experimental pleurisy and paw edema (15). These observations suggest that local inactivation of PAF by extracellular AH may be important in modulating the intensity of inflammation.

Several groups have attempted to quantitate PAF in plasma or bronchoalveolar lavage (BAL) from patients with various respiratory diseases. Although the measurement of PAF in BAL fluid produced conflicting results, high levels of lyso-PAF in the BAL of patients with inflammatory lung diseases were found in most studies (16). This observation led us to hypothesize that PAF could be rapidly converted to lyso-PAF in human BAL fluid by an extracellular AH. The aim of this study was to determine the presence and to characterize the activity of extracellular AH in human BAL fluid obtained from normal donors and from patients with different inflammatory lung diseases.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Materials

Radiolabeled PAF (1-³H-alkyl-2-acetyl-sn-glycero-3-phosphocholine; 60 Ci/mmol) and ³H-arachidonate-labeled Escherichia coli membrane (10.5 µCi/µmol) were purchased from New England Nuclear (Boston, MA). Unlabeled PAF and lyso-PAF (1-alkyl-2-lyso-GPC) were purchased from Biomol (Plymouth, PA). Essentially fatty-acid-free human serum albumin (HSA), phenylmethylsulfonyl fluoride (PMSF), 5,5'-dithiobis-2-nitrobenzoic acid (DTNB), and xanthine, xanthine-oxidase, piperazine-N, N'-bis-2-ethanesulfonic acid (Pipes) were purchased from Sigma Chemicals Co. (St. Louis, MO). Dithiotreitol (DTT) was purchased from Aldrich-Chemie (Steinheim, Germany). Protease (pronase; from Streptomyces griseus) was obtained from Calbiochem (La Jolla, CA). All solvents were HPLC grade from Carlo Erba (Milan, Italy). The Pipes buffer used in these experiments was a mixture of 25 mM Pipes, 110 mM NaCl, and 5 mM KCl at pH 7.4. Pipes-calcium-glucose (PCG) contains, in addition to Pipes, 1 mM CaCl₂ and 6 mM dextrose.

Study Population

The demographic and respiratory function data of the BAL donors are shown in Table 1. Eighteen healthy volunteers (14 male and four female) were recruited as a control group. They were all nonsmokers, free of respiratory symptoms, and had no history of atopic diseases. Fifteen patients (eight male and seven female) met the diagnostic criteria for asthma. These patients were classified as having mild intermittent asthma on the basis of clinical history and pulmonary function studies (19). They were all atopic as documented by a positive skin test for at least one common aeroallergen. At the time of the BAL procedure, all asthmatic patients were asymptomatic and had a FEV₁ greater than 80% of their predicted value (Table 1). They required only irregular ₂-agonist inhalation to control their symptoms and had never been treated with inhaled or systemic corticosteroids. Fifteen patients (12 male and three female) had a diagnosis of idiopathic lung fibrosis according to clinical and radiographic criteria and histology of samples obtained by transbronchial lung biopsy (20). In these patients all identifiable causes of lung fibrosis were excluded. None of the patients had received corticosteroid treatment in the month prior to the study. All subjects included in the study were free of acute respiratory illness during the 6 wk prior to the study. The study protocol was approved by the Ethics Committee of the University of Naples Federico II, and informed consent was obtained from each subject.

Bronchoalveolar Lavage

Bronchoscopy and BAL were performed following a standardized protocol according to current National Heart, Lung, and Blood Institute guidelines (21). Subjects were premedicated intramuscularly with diazepam and atropine 30 min before the procedure. Lidocaine (5 ml or less) was used for topical anesthesia of the upper airways. A flexible fiberoptic bronchoscope (Olympus BF type P20; Olympus Corp. of America, New Hyde Park, NY) was wedged into a segment or subsegment of the right middle lobe, and three fractions 50 ml each of saline preheated at 37 ° C were introduced. Each aliquot of fluid was recovered by gentle suction with a syringe and collected into 50-ml tubes kept in ice. Recovery of fluid ranged from 45 to 70%. All fractions were pooled and filtered through two layers of gauze. The total cell count was obtained by using a Burger chamber, and the results were expressed as cells × 10⁵/ml. The fluid was then centrifuged twice at 800 × g for 10 min at 4° C. An aliquot of cell pellet was resuspended in Pipes buffer and centrifuged with a cytospin. Cytocentrifugates were stained with May-Grunwald-Giemsa, and the differential cell count was obtained by counting 400 nonepithelial cells in random fields. The cell-free supernatant was immediately prepared for AH and phospholipase A₂ (PLA₂) assay. Total protein content of the fluid was determined by the method of Lowry and coworkers (22).

Secretion of Acetylhydrolase from BAL Macrophages

BAL fluids were centrifuged (800 × g for 10 min at 4° C) and the cell pellet was resuspended and washed three times in Pipes buffer containing 0.5 mg/ml HSA. Total and differential cell counts were made and the cells were resuspended (2 × 10  ⁶ cells/ml) in PCG. Cell suspensions consisted mainly of macrophages (> 95% of total cells). Contaminating cells were lymphocytes, erythrocytes, and eosinophils. Cells were incubated in PCG alone or in the presence of TNF- (100 ng/ml) for 15 to 180 min at 37° C or 4° C and then centrifuged (800 × g for 10 min at 4° C). The cell-free supernatant was collected and immediately analyzed for AH activity.

Acetylhydrolase Assay

The AH activity assay is described in detail elsewhere (12). Briefly, cell-free BAL fluid was incubated (15 to 90 min at 37° C) with various amounts of ³H-PAF (0.1 to 10 nmol/ml of fluid) complexed with HSA (final concentration, 0.1 mg/ml). At the end of incubations the reaction was stopped by the addition of 2 vol of methanol, 1 vol of chloroform, and 50 µl of 9% formic acid. Lipids were extracted using the technique of Bligh and Dyer (23) and were separated by thin-layer chromatography (TLC) on layers of silica gel G developed in chloroform/methanol/acetic acid/water (50/ 25/8/4 vol/vol). PAF and lyso-PAF standards were visualized with I₂ vapors and the radioactivity in each product on the TLC plate was located by scanning the plate with a Bioscan System 200 Imaging Scanner (Camberra Packard, Milan, Italy). Radiolabeled products were isolated from the TLC plate, and the radioactivity was determined by liquid scintillation counting. The activity of AH was expressed as pmoles of ³H-PAF converted to ³H-lyso-PAF/min.

Biochemical Characterization of Acetylhydrolase

The biochemical characteristics of BAL-AH and of AH released from macrophages in culture were compared with those of plasma AH. In these experiments, cell-free BAL fluid, supernatant from macrophage cultures, and plasma collected from normal donors by venipuncture in EDTA were incubated (37° C for 30 to 90 min) with the following compounds: DTT (10 mM), PMSF (2 mM), protease (1 mg/ml), and DTNB (2 mM). Because AH is an acid-labile enzyme, BAL fluid was mixed with TRIS-HCl buffer (0.1 M at pH 7.3) to avoid changes in the pH of the reaction. At the end of the incubation, radiolabeled PAF was added and AH activity remaining after each treatment was determined.

Phospholipase A₂ Assay

Eight hundred microliters of cell-free BAL were incubated (1 h at 37°  C) with 10 µl of ³H-arachidonate-labeled E. coli membranes as a source of phospholipids. The final concentration of CaCl₂ in the reaction mixture was 10 mM. The reaction was stopped by adding 2 vol of methanol, 1 vol of chloroform, and 50 µl of 9% formic acid. Immediately before extraction, 10 µg of unlabeled arachidonic acid were added to each sample as a carrier. The lipids were then extracted with the technique of Bligh and Dyer (23) and separated by TLC on layers of silica gel G developed in hexane/diethyl ether/formic acid (90 /60/6 vol/vol). PLA₂ activity was expressed as pmoles of ³H-arachidonate hydrolyzed from the radiolabeled membranes /min.

Statistical Analysis

The data are expressed as the mean ± SEM. Statistical analysis between groups was performed using Student's t test for unpaired samples. Correlations were assessed using Spearman's rank correlation analysis (24).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Characterization of Acetylhydrolase in the BAL Fluid

Initial experiments were performed to determine whether AH activity was present in cell-free BAL fluid. The kinetics of catabolism of exogenous PAF in the BAL fluid of a healthy nonsmoker donor are shown in Figure 1. PAF was rapidly converted to only one metabolite, lyso-PAF. Neither 1-alkyl-2-acyl-GPC nor 1-alkyl-2-acetyl-sn-glycerol was detected during the catabolism of PAF in the BAL. The mean activity of AH in the BAL of 18 normal donors was 2.25 ± 0.26 pmol of PAF converted to lyso-PAF/min. These data indicated that human BAL contains an enzyme capable of removing the acetate group at the sn-2 position of PAF. In addition to AH, hydrolysis of the acyl group at the sn-2 position of PAF could also be accomplished by PLA₂. A Group II secretory PLA₂ has been found in various biologic fluids such as the synovial fluid from patients with rheumatoid arthritis (25), the nasal lavage fluid from patients with allergic rhinitis (26), and the BAL fluid from asthmatic patients (27). The biochemical characteristics of BAL-AH, determined using labeled PAF as a substrate, and of BAL-PLA₂, determined using arachidonate- labeled E. coli membranes as a substrate are shown in Table 2. BAL-AH activity was not influenced by the addition of CaCl₂ (10 mM) or EDTA (5 mM) to the BAL fluid. Furthermore, AH activity was partially inhibited by PMSF (2 mM), but not by DTT (10 mM). In contrast BAL-PLA₂ activity was strictly dependent on the presence of Ca²⁺ in the medium, and it was completely inhibited by DTT (10 mM) but not by PMSF (2 mM).

View larger version (12K): [in this window] [in a new window]   Figure 1.   Kinetics of catabolism of exogenous PAF in BAL fluid. Cell-free BAL from a healthy nonsmoker donor was incubated at 37° C with radiolabeled PAF (0.5 nmol). At each time point the reaction was stopped by the addition of 2 vol of methanol and 1 vol of chloroform. Lipids were extracted (23) and separated by thin-layer chromatography. The areas of silica corresponding to PAF and lyso-PAF were immediately scraped into scintillation vials, and the radioactivity was measured by liquid scintillation counting.

Secretion of Acetylhydrolase from BAL Cells

The next group of experiments was designed to understand if AH in the BAL fluid was secreted by cells recovered in the BAL. Previous studies have shown that human macrophages are a source of AH (3, 15). To support the hypothesis that alveolar macrophages were the source of AH in the BAL, we investigated if there was a correlation between the level of AH and the number of macrophages in the BAL. It can be seen in Figure 2 that there is a significant positive correlation between the levels of extracellular AH and the number of macrophages/milliliter in the lavage of 15 healthy nonsmoker donors (r_s = 0.63; p < 0.02).

View larger version (16K): [in this window] [in a new window]   Figure 2.   Correlation between the levels of extracellular AH and the number of macrophages in BAL fluid of healthy nonsmoker donors. Each point represents an individual donor.

Cells obtained from the BAL of six healthy donors, containing 90 to 98% macrophages, were cultured, and the release of AH in the medium was determined at various time points (Figure 3). BAL cells incubated at 37° C spontaneously released AH in the extracellular medium. Secretion of AH reached a plateau after approximately 60 min of incubation. Release of AH by BAL cells was negligible when the cells were incubated at 4° C. Incubation of BAL cells in the presence of TNF- (100 ng/ml) resulted in a slight increase of AH secretion that was not significantly different from the spontaneous release.

View larger version (16K): [in this window] [in a new window]   Figure 3.   Secretion of AH from BAL cells. Unfractioned cells isolated from BAL of healthy nonsmoker donors (90 to 98% macrophages) were resuspended (2 × 106 cells/ml) in Pipes buffer and incubated at 4° C or 37° C in the absence or in the presence of TNF- (100 ng/ml). At each time point, the cells were centrifuged (800 × g for 8 min at 4° C) and AH activity was measured in the supernatant. The data are expressed as the mean ± SEM of six experiments.

To determine whether AH in the BAL fluid and AH released from macrophages were the same enzyme, we compared their biochemical characteristics. The effect of various compounds on the activity of AH from BAL, AH released from alveolar macrophages in culture, and plasma AH is shown in Table 3. BAL-AH and macrophage-derived AH had the same biochemical properties, i.e., they were both resistant to DTT and DTNB and had a similar sensitivity to PMSF and to protease. In contrast, plasma AH was resistant to protease and had a greater sensitivity to PMSF.

Effect of Oxygen-Free Radicals on BAL-Acetylhydrolase

We have demonstrated that plasma AH is inactivated by exposure to oxygen-free radicals (28). Generation of oxygen radicals is thought to be an important factor in the pathogenesis of several inflammatory diseases of the lung (29). To determine whether BAL-AH activity is influenced by oxygen radicals, we measured the enzyme activity after exposure of the BAL to an oxygen radical-generating system (xanthine/xanthine oxidase). It can be seen in Figure 4 that incubation of BAL with the radical-generating system (X/XO) in vitro results in the inactivation of more than 80% of BAL-AH. This effect of oxygen radicals appears to be specific since inactivation of BAL-AH is completely prevented by preincubation with either superoxide dismutase (SOD) or catalase (CAT).

View larger version (20K): [in this window] [in a new window]   Figure 4.   Effect of oxygen free radicals on BAL-AH. Cell-free BAL was incubated (15 min at 37° C) with an oxygen radical generating system (xanthine, 200 µM, and xanthine oxidase, 100 mU/ml) either in the absence or in the presence of superoxide dismutase (SOD) (300 U/ml), catalase (CAT) (1,000 U/ml), or both. At the end of the incubation, radiolabeled PAF (0.5 nmol) was added and the incubation was continued for 30 min at 37° C. AH activity was expressed as pmoles of PAF converted to lyso-PAF/min. The data are the mean ± SEM of five experiments. *p < 0.001 when compared with control values.

Acetylhydrolase Activity in the BAL Fluid of Patients with Inflammatory Lung Diseases

We compared AH activity in the BAL fluid obtained from 18 healthy donors, from 15 patients with mild bronchial asthma, and from 15 adult patients with lung fibrosis (Figure 5). Levels of BAL-AH were significantly lower in patients with bronchial asthma than in healthy donors of matched ages (1.32 ± 0.18 versus 2.25 ± 0.26 pmol/min; p < 0.001). In contrast, AH levels were significantly higher in patients with lung fibrosis than in normal donors (6.13 ± 0.81 versus 2.25 ± 0.26 pmol/ min; p < 0.01).

View larger version (18K): [in this window] [in a new window]   Figure 5.   Levels of BAL-AH in normal donors and in patients with bronchial asthma or lung fibrosis. Cell-free BAL from normal volunteers (n = 18), patients with bronchial asthma (n = 15), and patients with lung fibrosis (n = 15) were incubated (30 min at 37° C) with radiolabeled PAF (0.5 nmol). At the end of the incubation, radioactive products were extracted (23) and isolated by thin-layer chromatography. The data are expressed as pmoles of PAF converted to lyso-PAF/min. Boxes indicate the mean ± SEM.

To determine whether these differences were due to changes in the number of active molecules or to changes in the affinity of the enzyme, we constructed a dose-response curve of AH activity, with increasing concentrations of PAF (0.1 to 10 nmol/ml), in the BAL from four normal donors, four asthmatics, and four patients with lung fibrosis. The data are reported as a double reciprocal plot of the substrate-velocity curve (Figure 6). The estimated Michaelis constants of the enzyme were 0.92 ± 0.46, 0.96 ± 0.38, and 1.75 ± 0.94 µM for control subjects, asthmatics, and patients with lung fibrosis, respectively. The values between the three groups were not statistically different. These data indicate that the number of enzyme molecules in the BAL is reduced in asthmatic patients and increased in patients with lung fibrosis when compared with healthy donors.

View larger version (9K): [in this window] [in a new window]   Figure 6.   Lineweaver-Burke plot of AH in the BAL fluid of normal donors and of patients with bronchial asthma or lung fibrosis. Cell-free BAL was incubated (30 min at 37° C) with increasing concentrations of radiolabeled PAF (0.05 to 5 nmol). At the end of the incubation, radioactive products were extracted (23) and isolated by thin-layer chromatography. The data are reported as the double reciprocal plot of velocity of the reaction (1/V) at each substrate concentration (1/S) and represent the mean of four patients in each group.

As we show in Figure 2, the level of AH in the BAL of normal donors is significantly correlated with the number of macrophages. Therefore, the reduced or increased AH levels in the BAL of patients with asthma and with fibrosis, respectively, could also be due to variations in the number of alveolar macrophages in the BAL of these patients. To explore this possibility, we correlated the levels of BAL-AH with the number of macrophages in the BAL obtained from patients with asthma and lung fibrosis. There was a significant correlation in patients with fibrosis (r_s = 0.76; p < 0.01; n = 15), but not in patients with bronchial asthma (r_s = 0.14; NS; n = 15). These data indicate that elevated levels of AH in the BAL of patients with fibrosis could be due to an increased number of alveolar macrophages in these patients. In contrast, the reduced levels of AH in asthmatic patients are presumably not related to a lower number of macrophages but rather to inactivation or to a reduced synthesis of the enzyme.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We have identified and characterized an extracellular form of AH in human BAL fluid. BAL-AH converts PAF released in the airways to the inactive metabolite lyso-PAF. The biochemical characterization of BAL-AH indicates that this enzyme is calcium-independent, is not blocked by reducing agents, and is partially sensitive to serine-esterase inhibitors. These characteristics clearly distinguish it from the secretory PLA₂ previously reported in human nasal lavage fluid and in BAL (11, 26, 27). The other characteristics of BAL-AH reported in Tables 2 and 3, i.e., resistance to DTNB and partial sensitivity to PMSF and protease, also suggest that BAL-AH is different from plasma AH (8) and from erythrocyte AH (30). Consequently, it seems that the BAL-AH does not derive from passive diffusion of plasma AH or from extravasation of erythrocytes in the alveoli.

A likely source of AH in BAL is the alveolar macrophage. Previous studies have shown that monocyte-macrophages obtained from peripheral blood (3, 15) and macrophages from human milk (31) spontaneously release AH. In healthy donors we found a positive correlation between the number of macrophages recovered in the BAL and the level of extracellular AH. Furthermore, unfractionated BAL cells, more than 90% of which are macrophages, spontaneously release AH in culture in a temperature- and time-dependent fashion. Finally, AH released from alveolar macrophages in culture has biochemical characteristics identical to those of BAL-AH. Taken together, these observations strongly support the hypothesis that alveolar macrophages are a source of AH in BAL. It should be mentioned, however, that the conditions to which macrophages are exposed in culture do not completely mimic those occurring in vivo. Therefore, although our data implicate the macrophage as a major source of AH in the alveoli, they do not indicate the rate of secretion or the stimuli that can induce AH release in vivo nor do they exclude other cellular sources of BAL-AH such as epithelial cells and glandular cells.

It has recently been suggested that certain mediators released by tissue macrophages may play a role in downregulating inflammation and immune responses (32). In this regard, the secretion of AH, as well as of other macrophage products such as IL-1 receptor antagonist and protease inhibitors, may contribute to turn off inflammation and to initiate the tissue repair process (33).

Levels of AH in plasma may vary significantly in several diseases (5). In particular, previous studies have shown that the activity of plasma AH is reduced in young patients with moderate-to-severe bronchial asthma (14, 18). In the present study we report for the first time that adult patients with mild bronchial asthma have a reduced AH activity in the BAL fluid. The lack of correlation between the BAL-AH levels and the number of macrophages in the BAL of asthmatic patients suggest that the reduced AH activity is not due to a decrease in the number of alveolar macrophages. One possibility is that asthmatic patients may have a genetic deficiency of BAL-AH similar to what has been demonstrated for plasma AH (14). An alternative hypothesis to explain the reduced AH activity in asthmatic patients is the inactivation of the enzyme by oxygen free radicals. A number of experimental and clinical observations suggest that formation of oxygen radicals occurs in BAL in patients with asthma (29, 34). Potential cellular sources of oxygen radicals in such condition include eosinophils, macrophages, and neutrophils (29, 34). We demonstrate that BAL-AH is highly susceptible to oxygen radical-induced damage. Thus, chronic exposure of alveolar fluid to activated eosinophils or macrophages may result in a reduced AH activity. Whatever the mechanism, a reduced AH activity in the BAL of asthmatic patients may be responsible for a slower inactivation of PAF and, ultimately, for potentiation and/or prolongation of the proinflammatory effects of this mediator in bronchial asthma.

Recent studies have shown that antigen challenge of upper (11) and lower respiratory airways (27) in allergic patients results in an increased extracellular AH activity. These results are apparently in conflict with ours. However, it should be remembered that results from bronchoprovocation studies cannot be compared with those obtained under resting conditions. First, bronchoprovocation studies are obviously performed in allergic patients, and no comparable data can be obtained from healthy donors. Second, local challenge induces severe modifications of airway microenvironment. In particular, because of increased vascular permeability, large quantities of albumin are found in BAL after segmental provocation (27). Because plasma AH has a molecular weight (43 Kd) lower than that of albumin, it is likely that a percentage of AH activity found in the BAL after antigen challenge may be of plasma origin. Moreover, albumin concentration in the medium can influence the rate of catabolism of PAF (12). Thus, elevated albumin levels may apparently increase AH activity in the BAL. For these reasons, the finding of an elevated AH activity during antigen challenge does not exclude the possibility that asthmatic patients may have a reduced level of enzyme under resting conditions.

Patients with lung fibrosis have a higher AH activity than do normal donors. In these patients, as in normal donors, a significant correlation can be found between BAL-AH levels and the number of macrophages in the lavage fluid. These data indicate that the increased levels of AH in the BAL of patients with fibrosis are primarily due to a higher number of alveolar macrophages when compared with control subjects. However, it cannot be excluded that, in these patients, the increased number of AH molecules may also be the expression of a higher state of activation of alveolar macrophages. Previous studies have shown that alveolar macrophages from patients with lung fibrosis release larger quantities of inflammatory mediators such as leukotriene B₄ and IL-8 than do macrophages from normal donors (20). In addition, large quantities of AH may be released in the alveolar fluid of patients with lung fibrosis from lysed or damaged epithelial cells (12). The role of PAF in the pathogenesis of lung fibrosis is presently unclear. Therefore, it is difficult to determine whether an increased BAL-AH level may have a pathogenetic role in this disease or that it represents an epiphenomenon. It should be considered, however, that the main products of AH are lysophospholipids such as lyso-PAF and lysophosphatidylcholine, which have a profound detergent activity on cell membranes and may cause an irreversible damage to airway epithelium (35). In addition, lysophospholipids alter the properties of surfactant leading to alveolar collapse and increase collagen synthesis in lung fibroblasts (35). These biologic effects of lysophospholipids generated in the alveolar fluid by BAL-AH, and by other enzymes such as PLA₂, may potentially contribute to the development of lung fibrosis.

In conclusion, the present study demonstrates that an extracellular form of AH is present in human alveolar fluid of normal donors and of patients with different inflammatory lung diseases. This enzyme can effectively regulate the extracellular levels of PAF produced within the airways and, therefore, can modulate the local proinflammatory activity of this mediator. BAL-AH concentration varies in different inflammatory lung diseases and it presumably represents a net balance between secretion and inactivation of the enzyme. Further studies are required to determine whether the changes in the levels of BAL-AH may correlate with the disease activity in chronic lung diseases such as bronchial asthma and lung fibrosis.