INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pulmonary alveolar proteinosis (PAP) is a disease of unknown pathogenesis that is characterized by the accumulation of protein- and lipid-rich insoluble material in alveoli and terminal brochioles of the lung (1). Several studies have shown that alveolar macrophages (AM) obtained from patients with PAP contain large amounts of such material in their cytoplasm, and that their sizes are extremely large compared to those of normal AM (2, 3). Studies have also shown abnormalities in the adherence, chemotaxis, phagocytosis, and phagolysome fusion of these AM, and their poor survival in culture as compared with normal AM (2).

It is suggested that dysfunction of AM in PAP is acquired through exposure to the material that accumulates in the lung, because inhibition of phagocytosis, acquisition of morphologic characteristics similar to those of AM of PAP, and decreased viability can be induced in normal macrophages by incubating them with lavaged material from PAP patients (2, 3, 5). It has also been reported that migration of AM is improved dramatically after therapeutic whole-lung lavage in PAP, and returns to depressed levels when the disease relapses (6). Other studies have shown that AM of PAP are not only functionally abnormal but also decreased in quantity as compared with those of healthy subjects (7). In addition, several reports have observed that the serum lactate dehydrogenase (LDH) level seemed to correlate with disease activity, and that a large amount of LDH was found in lavage fluid from patients with PAP suggesting that pulmonary cell death occurs in the disease. LDH levels decreased to within the normal range when a remission was brought about by treatment with lung lavage (8).

Thus, some substance(s) in the accumulated lung material in PAP is speculated to be harmful to AM, and even seems to induce death of AM. In the present study, we attempted to characterize substances in the lavage material from PAP patients that contribute to the morphologic changes and low viability in vitro of AM from patients with PAP.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Lipids and Reagents

L--phosphatidylcholine--(nitrobenzodiazole-aminohexanoyl)--palmitoyl (NBD-PC) was purchased from Sigma Chemical Co. (St. Louis, MO). Succinimidyl 6-(7-nitrobenz-2-oxa-1,3-diazol-4-yl) aminohexanoate (NBD) and ethidium homodimer were products of Molecular Probes, Inc. (Eugene, OR). Dulbecco's modified Eagle's medium (DMEM) and fetal calf serum (FCS) were purchased from Dainippon Pharmaceutical Co., Ltd. (Castle Hill, NSW, Australia). Xylocaine (2%) was the product of Fujisawa Pharmaceutical Co., Ltd. (Osaka, Japan). N-(2-hydroxyethyl)piperazine-N'-2- ethanesulfonic acid (HEPES), octyl--D-glucopyranoside (OGP), 3,5-diaminobenzoic acid dihydrochloride, thiobarbituric acid, and 1,1,3,3,-tetramethoxypropane were the products of Nacalai Tesque (Kyoto, Japan). The lactate dehydrogenase (LDH) assay kit used in the study, glucose-6-phosphate dehydrogenase, hexokinase, LDH, pyruvate kinase, 5,5'-dithiobis-(2-nitrobenzoic acid), glutathione reductase, reduced nicotinamide-adenine dinucleotide phosphate (NADPH), oxidized glutathione (GSSG), phenyl isothiocyanate, triethylamine, amino acid standard solution, and phenylthiocarbamyl-amino acid elution solutions were purchased from Wako Pure Chemical Industries, Inc. (Osaka, Japan). Adenosine triphosphate (ATP), adenosine diphosphate (ADP), nicotinamide-adenine dinucleotide phosphate (NADP), reduced nicotinamide-adenine dinucleotide (NADH), and phosphoenolpyruvate were purchased from Boehringer Mannheim GmbH (Mannheim, Germany).

Fractionation of Bronchoalveolar Lavage Fluid from Patients with PAP

Bronchoalveolar lavage fluid (BALF) was obtained by therapeutic lung lavage from patients with PAP who were admitted to the Chest Disease Research Institute Hospital at Kyoto University. BALF was stored at 10° C until used. The thawed BALF was fractionated according to the method described by Gonzalez-Rothi and Harris (3). The precipitate obtained by centrifugation of BALF at 250 × g for 10 min at 4° C was referred to as P1. The turbid supernatant was further centrifuged at 20,000 × g for 30 min, and a pellet (P2) and supernatant (S) were obtained. P1 and P2 were resuspended in saline at a protein concentration of 10 mg/ml.

Preparation of Surfactant-associated Protein A

For the preparation of surfactant-associated protein-A (SP-A), BALF of patients was centrifuged at 16,000 × g for 30 min and the precipitate was delipidated by three extractions with ethanol-ether (3:2, vol/ vol) and one extraction with diethylether at 10° C. Dried residue was extracted twice with 20 mM OGP in 10 mM HEPES, and then twice with 5 mM borate buffer (pH 10) by hand homogenization and centrifugation at 100,000 × g for 1 h at 4° C. The borate-buffer extracts were combined and dialyzed against 5 mM sodium borate (pH 9.2), and the dialysate was centrifuged at 100,000 × g for 1 h at 4° C to remove insoluble material. The purity of SP-A was examined through sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS- PAGE) (7.5%, reduced condition) according to the method of Weber and Osborn (9). The SP-A solution was stored at 20° C until used.

Preparation of Lipids and Hydrophobic Proteins

Lipids and hydrophobic apoproteins were extracted from the total precipitate of patient's BALF centrifuged at 16,000 × g for 30 min through the method of Folch and colleagues (10). This fraction was referred to as the crude lipid fraction containing hydrophobic proteins (CLF). A portion of CLF was further fractionated into hydrophobic surfactant apoproteins and a pure lipid fraction (PLF), using a Sephadex LH60 (Pharmacia Biotechnology, Uppsala, Sweden) column according to the method of Curstedt and colleagues (11). Proteins and lipids were eluted with chloroform-methanol-0.1 M HCl (1:1:0.1, vol/ vol), and all fractions containing proteins were examined with SDS-PAGE (12.5%, unreduced condition). Fractions containing surfactant-associated protein B (SP-B) were evaporated to dryness, and an acidic ethanol-soluble protein of the residue was used as SP-B (12). Fractions containing SP-C and a dimeric form of SP-C ([SP-C]₂), which was eluted between SP-B and SP-C, were pooled separately and used without further purification. PLF was dried, reextracted with the method of Folch and colleagues (10), and dissolved in chloroform. Lipids and proteins were stored at 20° C until used.

Pig pulmonary surfactant was isolated through lung lavage and sucrose density-gradient centrifugation (13), and the hydrophobic surfactant-associated apoproteins pig SP-B and pig SP-C were obtained as described earlier.

Purity of isolated proteins was evaluated with SDS-PAGE under reducing and unreducing conditions, with amino-acid-composition analysis. SP-A and SP-B were hydrolyzed at 110° C for 22 h and 72 h, and SP-C and [SP-C]₂ were hydrolyzed at 110° C for 22 h and at 150° C for 72 h (14). Amino acids were converted to their phenylthiocarbamyl derivatives and analyzed with reverse-phase high-pressure liquid chromatography (HPLC) (L-6200; Hitachi Ltd., Tokyo), using a Hibar prepacked column (LiChrosorb RP-18; Ciba-Merck, Darmstadt, Germany).

Preparation of Fluorescence-labeled Hydrophobic Apoproteins

Hydrophobic apoproteins obtained from PAP patients were separately labeled with succinimidyl-NBD according to the method of Horowitz and colleagues (15). After the labeling reaction, aggregated material was removed by centrifugation at 6,000 × g for 10 min, and the labeled proteins were purified with a Sephadex LH60 column. Labeled SP-B, SP-C, and [SP-C]₂ were referred to as NBD-SP-B, NBD-SP-C, and NBD-[SP-C]₂, respectively. Surface activities of liposomes reconstituted with these labeled proteins were not significantly different from those reconstituted with original proteins when examined with a pulsating bubble surfactometer (Electronics Co., Amherst, NY).

Determination of Composition of Proteins and Lipids of Insoluble Material

The content of SP-A and SP-B in P1 was determined with an enzyme-linked immunosorbent assay (ELISA). For determination of SP-A, insoluble material was sonicated in borate buffer (pH 10.0) and applied directly to microtiter-plate wells (Nunc-Immuno Plate I; A/S Nunc, Roskilde, Denmark) after multiple dilutions. After being left overnight at room temperature, wells were washed with phosphate-buffered saline (PBS), and nonspecific adsorbing sites were blocked with 1% bovine serum albumin (BSA) in PBS for 2 h at 37° C. Wells were washed with PBS and reacted with a biotinylated monoclonal antihuman SP-A antibody (35-13) in PBS containing 0.05% Tween-20 (TPBS) for 1 h at room temperature. After washing of the plates with TPBS, streptavidin-horseradish peroxidase (1 µg/ml in TPBS) was added and left for 30 min at room temperature. After vigorous washing with TPBS and PBS, color was developed by incubation for 30 min with 0.04% o-phenylenediamine and 0.01% H₂O₂ in 0.1 M citrate- phosphate buffer (pH 5.0). The chromogenic reaction was stopped by adding 2 M H₂SO₄, and absorbance at 492 nm was determined with a microplate reader (MTP-100; Corona Electric, Ibaragi, Japan). For the determination of SP-B, insoluble material was sonicated in PBS containing 1% Triton X-100, and was further diluted through multiple dilutions with the same solution. A sandwich ELISA was used to quantitate SP-B according to the method described previously (16), with two kinds of monoclonal antihuman SP-B antibodies (unlabeled HS-2 and biotinylated HS-1) (12). The content of SP-C in P1 was determined from the sum of the amount separated by a Sephadex LH60 column after extraction into organic solvent and the amount in proteins that were unextractable with organic solvent. The latter was determined according to the method described previously. (16). The purity of the SP-C thus obtained was assessed by analysis of its amino acid composition after eluting the relevant band from a preparative SDS-PAGE gel.

Total protein was determined by the method of Lowry and colleagues (17). Hydrophobic apoproteins, phospholipids (PL), and PL composition were assayed as described (16). Total and esterified cholesterol were determined with an enzymic method, using a kit (Cholesterol CII Test Wako; Wako Biochemical Co.).

Preparation of Liposomes

CLF and PLF containing 500 µg PL were dried in a 1.5-ml Eppendorf tube under nitrogen and then in vacuo. The residue was dispersed in 1 ml of culture medium (see the subsequent DISCUSSION) by sonication for 2 min on ice with a sonifier having a 2-mm microtip (Astrason; Heatsystems-Ultrasonics, Inc., Farmingdale, NY) at 10 W. To prepare liposomes with hydrophobic proteins, PLF was separately mixed with SP-B, SP-C, and [SP-C]₂ at a weight ratio of protein to PL of 1:40. Pig SP-B and SP-C in combination or pig SP-C alone were mixed with PLF at a ratio of protein to PL of 3:5:100 or 1:40 by weight, respectively.

To assess the accumulation and removal of each component, fluorescence-labeled lipid and proteins were used. NBD-PC was mixed with CLF or PLF at a weight ratio of NBD-PC to PL of 1:80. To examine the effect of apoproteins on the accumulation of NBD-PC in AM, the mixture of PLF and NBD-PC was supplemented separately with unlabeled SP-B, SP-C, and [SP-C]₂ at a weight ratio of protein to PL of 1:40. To test the effect of [SP-C]₂ on the accumulation and removal of NBD-SP-B and NBD-SP-C in the cells, unlabeled [SP-C]₂ was added to the mixture of PLF and labeled proteins at a weight ratio of protein to PL of 1:40. Liposomes were prepared from these various mixtures as described earlier. The liposomes were prepared just before use, and the pH of the final liposome preparations was adjusted to neutral with 0.1 M NaHCO₃, since the eluate from the Sephadex LH60 column contained HCl.

Preparation of Rat AM

Specific pathogen-free adult male Wistar rats (Shimizu Animal Laboratories, Kyoto, Japan) weighing 200 to 250 g were anesthetized with intraperitoneal injections of pentobarbital sodium (100 mg/kg weight) and killed by exsanguination from the abdominal aorta. After the lungs were perfused via the pulmonary artery with saline to eliminate blood, they were removed and lavaged eight times with 5 ml saline containing 1 mM EDTA (pH 7.4, 37° C) under mild massage. The combined lavage fluids were centrifuged at 250 × g for 10 min at 4° C and the cell pellets were washed twice with 20 ml of PBS.

Cells were suspended in DMEM supplemented with 10% FCS and incubated for 1 h at 37° C in 5% CO₂/air in 24-well culture plates (2.5 × 10⁵ cells/ml/well) or in 35-mm culture dishes (5 × 10⁵ cells/2 ml/dish). Nonadherent cells were removed by washing three times with culture medium, and the remaining adherent cells were used for the experiments. Viability of adherent cells, as tested by trypan blue exclusion, was greater than 98%, and the purity of AM tested by acid phosphatase staining (18) was greater than 98%.

Determination of Cell Size and Viability of AM Cultured with Different Preparations

AM were incubated for 3 d with P1, P2, and S (1 mg protein/ml), or with various liposomes with and without SP-A for different periods, as shown in the RESULTS. When culture was continued for more than 3 d, each well or culture dish was supplemented with a half volume of fresh culture medium containing liposomes and SP-A on the 3rd incubation day. After culture, cells were detached from wells or dishes by addition of xylocaine at a concentration of 5 mg/ml followed by 30 min exposure to room temperature (19). The cells were washed twice with cold PBS and were then subjected to determination of cell size and viability.

To avoid changes in cell size as a result of attachment, we used floating cells to determine cell size. Cells suspended in PBS were photographed with a phase-contrast system, and cell volume was calculated from the area of cells in the photographs enlarged to a magnification of ×1,250 by the equation, V = 3/4 (S[S/]^1/2), where V is the cell volume and S is the cell area measured with a JIM-5000 image analyzer (JEOL, Akishima, Japan).

To determine cell viability, 100 µl of ethidium homodimer (2 mM) was added to the cell suspension on a 22-mm square coverslip. After incubation of the treated suspension at room temperature in a moistened dish for 45 min, 10 µl of fresh reagent solution was added and cells were immediately examined with a fluorescence microscope (Axioplan, Carl Zeiss, Jena, Germany). The cells were photographed randomly under UV and under visual light in the same field. Viability of cells was calculated as a percentage of the number of cells without red fluorescent nuclei in the total number of cells counted in the photographs taken with the phase-contrast system under visual light. Usually, three fields were examined for one sample, at a final magnification of ×1,250.

Determination of LDH

After culture of AM with various liposomes for 3 d and 6 d, culture medium was centrifuged at 250 × g for 10 min at 4° C. Cell pellets and cells attached to dishes were lysed by adding 1 ml of distilled water, and were mechanically ruptured with three repeated cycles of freezing and thawing. A 50-µl aliquot of cell-free medium and the cell lysates were subjected to LDH measurement with a commercial kit. The determination of LDH activity with the kit was based on the formation of diformazane by reduction of nitroblue tetrazolium in a reaction catalyzed by diaphorase with NADH. NADH was formed from NAD used as a cofactor in the oxidation of L-lactate to pyruvate, which was catalyzed by LDH. Absorbance at 560 nm was measured with a spectrophotometer (U-2000; Hitachi Ltd.). LDH activity of the medium, from which background LDH activity of the control, cell-free medium incubated for the same period as that from experimental groups was subtracted, was considered to be the liberated LDH activity. This activity was expressed as the percentage of total LDH activity, which was the sum of the LDH activity in the medium and cell lysates.

Measurement of Superoxide Anion Release

Superoxide anion release was assayed through the superoxide dismutase (SOD)-inhibitable reduction of ferricytochrome c. After incubation of AM with liposomes, the nonadherent and adherent cells were collected and washed twice with 2.5 mM EDTA in PBS by centrifugation. The cells were then incubated with 0.5 ml of reaction buffer (2 mM glucose, 1 mM CaCl₂, 1.3 mM MgCl₂, 4 mM KCl, 100 mM NaCl, and 10 mM phosphate buffer, pH 7.4) and ferricytochrome c (final concentration, 70 µM), with and without SOD, for 30 min at 37° C.The rate of superoxide anion formation for 30 min was calculated from the difference in absorbance at 550 nm and 540 nm, using the extinction coefficient ( = 19.1 mM¹ cm¹) (20). Stimulation of the reaction with phorbol myristate acetate (PMA) was also quantitated in cells cultured with SP-A or liposomes for 6 h.

Determination of Intracellular Glutathione and Adenosine Nucleotide Concentration

Cells were harvested as described earlier. They were extracted with 0.5 N perchloric acid and centrifuged at 12,000 × g for 10 min at 4° C. Total glutathione (GSH), ATP, and adenosine diphosphate (ADP) were measured in the supernatant after neutralization with 2 M KOH/ 0.5 M triethanolamine buffer and centrifugation at 30,000 × g for 15 min at 4° C, according to the methods of Tietze (21) and Tornheim and Schultz (22, 23), respectively. The pellet was used for determination of total cell protein by the method of Lowry and associates (17).

Determinaton of Intracellular Lipid Peroxides

Cells harvested as described earlier were suspended in 1 ml distilled water and disrupted by sonication. The intracellular lipid peroxides were measured according to the method of Ohkawa and colleagues (24). Total cell protein was measured as previously described.

Accumulation in and Removal of Fluorescence-labeled Components from Cells

AM were cultured with liposomes labeled with NBD-PC and with NBD-proteins for 2 d in the presence of SP-A, and were further cultured for another 2 d after removing liposomes and SP-A by washing. The amount of fluorescence-labeled substances accumulated in the cells on Day 4 was compared with that for the first 2 d of cell culture.

Cells were harvested as described earlier and washed twice with 2.5 mM EDTA and 1% FCS in PBS to remove the liposomes outside the cells. Complete removal of membrane-associated liposomes was carefully checked with a fluorescence microscope. In the case of NBD-PC, a definite number of cells were pelleted by centrifugation and extracted with 1 ml of methanol for 2 h at room temperature, and the extracted lipids were separated by thin-layer chromatography on silica gel. Spots corresponding to original NBD-PC were eluted and quantified with a fluorescence spectrophotometer (F-2000; Hitachi Ltd.) at an excitation wavelength of 470 nm and an emission wavelength of 530 nm. In the case of labeled apoproteins, cells were sonicated for 1 min in 1 ml chloroform-methanol-0.1 M HCl (1:1:0.1; vol/vol), and extracted NBD-apoproteins were measured as described earlier after centrifugation at 6,000 × g for 10 min. A linear response was obtained for up to 0.1 µg/ml of NBD-PC and 0.4 µg/ml of NBD-apoproteins- the ranges including all of the samples examined.

Statistical Analysis

Differences between means were evaluated with the nonpaired Student's t test, using a commercially available computer software package (StatView; Abacus Concepts, Inc., Berkeley, CA).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Composition of Fractionated Insoluble Material

Protein-to-PL ratios of the three fractions separated by centrifugation, P1, P2, and S, were 2.01, 2.74, and 28.40, respectively. The distribution of proteins was 50.9% in P1, 2.7% in P2, and 46.4% in S. Total proteins in these fractions, as examined with SDS-PAGE, are shown in Figure 1A. The major protein bands in P1 were 67 kDa, 62 kDa, and 35 kDa. The faster migrating bands seemed to correspond to SP-B and SP-C. The profile of P2 was almost the same as that of P1, except for the relative abundance of 43-kDa band. In fraction S, the main protein band corresponded to albumin, at 67 kDa, with bands corresponding to heavy and light chains of immunoglobulins (54 kDa and 28 kDa, respectively). Smaller protein bands were not observed in S. Since the most deleterious effects on macrophages were shown in P1 (see the subsequent DISCUSSION), further analysis of composition was done on P1. Table 1 shows the lipid and protein compositions of P1 and CLF. They are expressed as quantities per milliliter of culture medium used in this experiment. The total quantity of SP-C was greater in P1 than in CLF, but no marked difference was noted in lipid composition.

View larger version (52K): [in this window] [in a new window]   Figure 1.   (A) sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (7.5%) profiles of fractions separated by differential centrifugation of patient's BALF. Lane 1, a fraction obtained by centrifugation at 250 × g for 10 min (P1); Lane 2, a fraction obtained by further centrifugation at 20,000 × g for 30 min (P2); Lane 3, supernatant fraction (S); and Lane 4, purified surfactant-associated protein A (SP-A) from patient's BALF. Gels were stained with Coomassie brilliant blue. (B) SDS-PAGE (12.5%) profiles of fluorescence-labeled surfactant-associated proteins. Lanes 1 and 4, NBD-SP-B; Lanes 2 and 5, NBD-[SP-C]2; Lanes 3 and 6, NBD-SP-C. Lanes 1, 2, and 3, unreduced; Lanes 4, 5, and 6, reduced with 2-mercaptoethanol.

Isolated SP-A consisted of two main bands, one at 62 kDa and the other at 35 kDa (Figure 1A, Lane 4). Hydrophobic proteins in CLF of PAP patients showed three peaks when separated on the Sephadex LH60 column. The first peak corresponded to SP-B and the last peak to SP-C. Between these two peaks there was another large peak, which was not observed in the pig surfactant. On SDS-PAGE under nonreducing conditions, this protein migrated between SP-B and SP-C, but under reducing conditions it migrated at the same position as SP-C. Figure 1B shows the SDS-PAGE profile of the protein (Lane 2, under nonreducing conditions; Lane 5, under reducing conditions) after labeling with succinimidyl-NBD (unlabeled proteins showed a similar profile). The amino-acid composition of the protein was similar to that of SP-C, as summarized in Table 2. The amino-acid compositions of other isolated proteins were similar to those reported previously (14, 16, 25).

Effects of Fractionated BALF and Liposomes on Cell Size and Viability

AM incubated for 3 d with P1 and P2 of BALF showed significantly larger cell volumes than those incubated with S (Table 3). The viabilities of cells treated with P1, P2, and S were 15.3%, 27.1%, and 85.6%, respectively. The results showed that the insoluble fraction of BALF had a significant effect on cell volume and viability, but that albumin and immunoglobulins in S did not affect the cells. We further examined the effect of SP-A and CLF, which were the major constituents of the insoluble fraction. Purified SP-A at a high concentration (500 µg/ml) did not show any significant effect on cell viability, although it showed the same effect on cell-size enlargement as CLF. However, CLF significantly decreased cell viability (85.2%) as compared with SP-A. SP-A at a lower concentration (80 µg/ ml) had no effect on cell volume and viability, but when given at the same time as CLF had greater effects on cell volume and viability than those of CLF alone (p < 0.05). Effects of SP-A were dose-dependent, and heat treatment of SP-A abolished these effects (data not shown).

Effects of Different Hydrophobic Proteins on Cell Size and Viability

The findings in the study suggested that a substance in CLF induced an increase in AM volume and cell death. Figure 2 shows the size distribution of cells and the viability of cells according to their sizes for AM cultured for 6 d with liposomes prepared from PLF and PLF combined with SP-B, SP-C, and [SP-C]₂, which were major constituents of CLF. As controls, we examined cells cultured with SP-A alone or with liposomes prepared from pig SP-B and SP-C in combination. As the size of the cells increased, the cell viability decreased in groups treated with liposomes with hydrophobic proteins. When the cell viability among these four groups was compared with a paired t test, the viability of cells incubated with PLF with [SP-C]₂ was significantly different from that of cells incubated with SP-B and SP-C obtained from PAP patients, and from those incubated with pig SP-B and SP-C (p < 0.05). The finding that [SP-C]₂ induced death in smaller-sized cells suggested that cell death was not only due to overaccumulation of material in the cell, but might also have been due to the toxic nature of [SP-C]₂. The total viabilities of cells were 88.3 ± 3.8% (control), 84.7 ± 2.4% (PLF), 79.5 ± 3.7% (PLF + pig SP-B and SP-C in combination), 81.1 ± 3.8% (PLF + SP-B), 75.9 ± 5.9% (PLF + SP-C), and 55.5 ± 6.7% (PLF + [SP-C]₂) (mean ± SE of three or four independent experiments).

View larger version (28K): [in this window] [in a new window]   Figure 2.   Distribution of cell sizes and viability of rat alveolar macrophages (AM) cultured with different liposomes in the presence of SP-A (80 µg/ml) for 6 d. Control cells (Control) were cultured with SP-A alone. Numbers of cells were counted according to their sizes (open squares) and viability determined with ethidium homodimer (closed circles). PLF = pure lipid fraction; PLF + Pig B, C = PLF combined with pig surfactant-associated proteins B and C added in combination; PLF + SP-B, PLF + SP-C and PLF + [SP-C]2 = PLF combined with surfactant-associated proteins B, C, and [SP-C]2 obtained from PAP patients, respectively.

LDH Release in Culture Medium

Since monitoring cell viability with an ethidium homodimer is limited to cells that do not disintegrate, the release of the cytoplasmic marker enzyme LDH was also compared at 3 d and 6 d of incubation among AM cultured with various liposomes (Figure 3A). The release of LDH from AM incubated with PLF + [SP-C]₂ was 48.7% of the total LDH in the supernatant of the cell medium and in the cell lysate at 3 d and 71.1% at 6 d of incubation, and both of these values were significantly greater than those of both the control and PLF groups on the respective days. The release of LDH from AM incubated with PLF + SP-B and PLF + SP-C was 39.0% and 42.4% at 3 d and 56.5% and 58.7% at 6 d of incubation, respectively. These values were not significantly different from those of the PLF group on the respective days (30.3% and 52.3%, respectively). Changes in cell numbers in several of the experimental groups, calculated from total proteins of harvested cells, are shown in Figure 3B. These changes consisted of a linear decrease after 1 d of culture in all groups to 45% of the initial number of cells for AM treated with PLF + [SP-C]₂, and to 58% of the initial number of cells in the control group at 6 d. There are no significant differences among the various groups.

View larger version (23K): [in this window] [in a new window]   Figure 3.   (A) Lactate dehydrogenase (LDH) release into culture medium and (B) number of cells harvested from culture dish according to time after incubation of rat alveolar macrophages (AM) with SP-A (80 µg/ml) or with different liposomes in the presence of SP-A. No additional material was given to control AM. Pig B, C is SP-B and SP-C obtained from pig surfactant and used in combination. LDH release was expressed as percentage of total LDH in supernatant of medium and in cell lysate. Number of cells was calculated from the total protein of cell lysate, using the value 1 mg protein/3.94 x 106 cells. Each point represents mean ± SE of three or four experiments. *Different from control and PLF groups at p < 0.05; different from control group at p < 0.05.

Effect on Superoxide Anion Release

Superoxide anion release by AM was not significantly different among the control, SP-A-treated, PLF-treated, PLF + pig SP-C treated, and PLF + SP-C-treated cells at 6 h of incubation (Figure 4A). At 12 h, AM treated with PLF, PLF + pig SP-C, and PLF + PAP SP-C released significantly larger amounts of superoxide anion than did control AM and AM treated with SP-A. For AM incubated with PLF + [SP-C]₂, superoxide anion release was significantly greater than for all of the other five groups both at 6 h and 12 h of incubation (3.81 and 4.99 nm/10⁶ cells/30 min, respectively).

View larger version (26K): [in this window] [in a new window]   Figure 4.   (A) Superoxide anion release from rat alveolar macrophages (AM) after incubation with SP-A or with different liposomes in the presence of SP-A for 6 h and 12 h. Control cells (Control) were cultured without liposomes and SP-A. (B) Superoxide anion release in response to phorbol myristate acetate (PMA) in rat AM after 6 h incubation as described previously. Pig C is SP-C obtained from pig surfactant. Each point represents mean ± SE of three or four experiments. * Different from control and SP-A-treated groups at p < 0.05; different from other five groups at p < 0.05.

Figure 4B shows superoxide anion release stimulated by PMA (final concentration, 0.1 µg/ml) in cells after 6 h culture with SP-A or with liposomes. Superoxide anion release increased to 5 to 6 nm/10⁶ cells/30 min, and there were no significant differences among the six groups examined.

Effect on Total Cellular GSH and Lipid Peroxides

As shown in Figure 5A, the cellular GSH contents of all groups increased at 1 d after culture, and the content in AM cultured with PLF + [SP-C]₂ was significantly greater than that of the time = 0 control. At 3 d after culture, the content of GSH in AM cultured with PLF + [SP-C]₂ was significantly smaller than the GSH contents of the other three groups.

View larger version (19K): [in this window] [in a new window]   Figure 5.   Cellular glutathione (GSH) content (A) and lipid peroxide content (B) in control rat alveolar macrophages (AM) and AM cultured with different liposomes with SP-A. Each point represents mean ± SE of three to five experiments. *Different from time = 0 control at p < 0.05. Different from other three groups at p < 0.05; different from control group at p < 0.05.

The lipid peroxide content increased in AM treated with PLF, PLF + SP-C and PLF + [SP-C]₂ at 3 d after culture, and it was significantly greater in AM treated with PLF + [SP-C]₂ than that of the control group (Figure 5B).

Effect of Liposomes on ATP and ADP

As shown in Figure 6, the ATP content decreased in the control, PLF, PLF + SP-C, and PLF + [SP-C]₂ groups as the incubation time increased. At 3 d after culture, the ATP content of the cells cultured with PLF + [SP-C]₂ was significantly smaller (8.6 nmol/mg protein) than those of the other three groups (14.4 nmol/mg protein in the control group and 16 nmol/mg protein in cells cultured with PLF and PLF + SP-C). No change was found in the ADP content of the cells.

View larger version (17K): [in this window] [in a new window]   Figure 6.   Cellular adenosine triphosphate (ATP) and adenosine diphosphate (ADP) content in rat alveolar macrophages (AM) incubated with different liposomes in the presence of SP-A. Open and closed symbols denote ATP and ADP, respectively. Each point represents mean ± SE of three experiments. *Different from other three groups at p < 0.05.

Accumulation and Disappearance of Phagocytosed Lipids, SP-B, SP-C, and [SP-C]₂

We also investigated whether or not the substances accumulated in AM could be reversibly removed from them (Table 4). For this purpose, we cultured AM for 2 d with liposomes prepared from CLF and PLF with NBD-PC with and without unlabeled SP-B, SP-C, and [SP-C]₂. There was no significant difference among the five groups in the amounts of NBD-PC accumulated in the cells after 2 d of culture. After liposomes and SP-A were removed by washing, cells were further cultured for another 2 d. The amount of NBD-PC in the cells after 4 d of culture is shown as the percentage of the amount accumulated after 2 d. The amount of NBD-PC remaining in the cells after 4 d of incubation was 17 to 25% of that at 2 d, and there were no significant differences among the groups.

When AM were cultured for 2 d, with liposomes separately labeled with NBD-SP-B, NBD-SP-C, or NBD-[SP-C]₂ in the presence of SP-A, the absolute amounts of labeled proteins accumulated in the cells were 3.13, 2.30, and 3.19 µg/10⁶ cells. These values were about 60 to 80 times greater than the absolute amount of NBD-PC accumulated in cells cultured for the same period. After a further 2 d of culture without liposomes and SP-A, the amount of NBD-[SP-C]₂ in the cells was unchanged from that after 2 d (102%), whereas the amounts of NBD-SP-B and NBD-SP-C decreased to 67.4% and 72.4%, respectively of the values after 2d. When the latter two liposomes were supplemented with unlabeled [SP-C]₂, these values increased to 75.2% and 79.5%, respectively, but they were not significantly different from those cells cultured with liposomes without [SP-C]₂.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We investigated the prolonged effects of the material obtained from BALF of patients with PAP on the cell size and viability of rat AM, and confirmed the findings reported by Gonzalez-Rothi and Harris (3) that insoluble material from BALF induced the appearance of large cells and cell death. We further determined the substance that had the most marked effects on cellular activities.

The earliest change we observed after incubation of AM with various liposomes was the release of superoxide anion from treated cells. The basal superoxide anion release of control cells was in the same range as those reported by others (26, 27), but significantly increased release was found for AM treated with [SP-C]₂-containing liposomes after 6 h and 12 h of culture. It is unclear why [SP-C]₂-containing liposomes induced stimulation of superoxide anion release, since the quantity of liposomes with SP-C and those with [SP-C]₂ that were phagocytosed by cells did not seem to differ.

The increase in GSH level in cells with liposomes with [SP-C]₂ after 1 d of culture may have been an adaptive response of the cells to oxidative stresses following increased superoxide anion formation induced by [SP-C]₂. A marked increase in GSH was shown in cells at 24 h after treatment with oxidants (28). After 3 d of culture, a significant decrease in GSH and ATP content was found in AM exposed to [SP-C]₂-containing liposomes. These decreased cellular activities probably have a direct relationship to the earlier cell death induced in AM treated with [SP-C]₂. Oxidative stresses are known to induce changes in mitochondrial function (29), and to lead eventually to cell death (30).

In accord with a high death rate among recovered cells, extracellular LDH levels were significantly higher in cells treated with liposomes with [SP-C]₂ than in cells of other groups, suggesting that this protein caused marked destruction of cells with the release of LDH. At 3 d and 6 d of culture of control cells, about 25% and 40%, respectively, of LDH activity was found extracellularly. Although these values for control cell culture were high, the level of LDH in a similar experiment reported by Tangirala and colleagues (31) was 32% at 4 d of culture, which is in the same range as those in our study. SP-A, a soluble protein, also accumulates in large amounts in the lungs of patients with PAP. Since a very high concentration of SP-A (500 µg/ml) did not induce high mortality in AM in our study, the effect of SP-A, if present, is deduced to be mediated by acceleration of phagocytosis, as reported by Wright and colleagues (32). More marked effects on cell size and viability were shown in AM cultured with insoluble material (P1 and P2) than in cells cultured with CLF liposomes and SP-A. It is possible that the difference was due to the form of SP-A; SP-A was combined with lipids in P1 and P2, but in the latter case lipids and SP-A were added separately to cells. Another possibility, which could not be excluded on the basis of the present results, is that macrophages in P1 released some other toxic substance(s). Further investigation is necessary to examine this possibility. Although we did not investigate the effects of globulin and albumin, the supernatant fraction obtained at a higher speed (S), which was composed of these soluble proteins, had no harmful effects on the cells.

The cell volume of AM is considered to reflect the accumulation of phagocytosed material in the cell. Since it was difficult to analyze the components of phagocytosed material in cells because little material was recovered from cells and because of possible interference by cellular lipids and proteins, we measured the accumulation and removal of fluorescence-labeled lipids or proteins. Very little NBD-PC-0.04 µg/10⁶ cells, corresponding to 3.2 µg of the total PL of liposomes-was recovered after 2 d of incubation, and about 82% of accumulated NBD-PC was removed from cells by culturing them for a further 2 d without liposomes and SP-A. On the other hand, about 2 µg to 3 µg of each labeled protein/10⁶ cells was recovered after 2 d of culture. Only 33% of accumulated SP-B and SP-C was removed, whereas no reduction in [SP-C]₂ was observed with a further 2 d of culture. Since each labeled protein was combined with lipids at a concentration of 2.5%, and since cells do not phagocytize only the protein components of reconstituted complexes, more than 80 µg to 120 µg of PL should have been phagocytosed together with proteins. Even if hydrophobic proteins accelerated lipid intake to twice the rate of intake of lipids alone, more than 90% of lipids was estimated to be removed during this period, since only 3.2 µg of PL remained in the cells, as described earlier. These findings suggest that lipids were more rapidly removed from cells than were hydrophobic apoproteins, especially [SP-C]₂, and that relatively large amounts of proteins as compared with lipids accumulated in AM from PAP patients. The present results may explain why fused-membrane structures are rich in hydrophobic surfactant apoproteins rather than in lipids (16), and why a relatively large amount of [SP-C]₂ is found in the BALF of patients with PAP (33).

The presence of [SP-C]₂ in lungs of PAP patients was first reported by Warr and colleagues (34). [SP-C]₂ is present in only small amounts in normal pulmonary surfactant, but is increased in PAP. The present findings are important in understanding the pathophysiology of PAP. Further investigation is necessary to clarify the mechanism by which [SP-C]₂ stimulates superoxide anion release and the mechanism triggering accumulation of this protein.

In conclusion, we have shown that a dimeric form of SP-C was the least digestible of several hydrophobic apoproteins by AM. It was considered to cause early cell death through increased release of superoxide anion and the subsequent derangement of cellular metabolism. The changes in macrophages induced by [SP-C]₂ may contribute to establishing PAP.