Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Pulmonary surfactant protein (SP)-A and SP-D belong to a group of collagenous carbohydrate binding proteins known as collectins that mediate a variety of immune cell functions. For example, a number of studies have shown that SP-A and SP-D stimulate immune cell chemotaxis and the phagocytosis and clearance of a variety of pathogens (1, 2). The role of SP-A and SP-D in regulating production of cytokines and reactive species is less clear. For example, SP-A has been shown to directly stimulate (3) or to have no effect (4) on tumor necrosis factor (TNF)- production by immune cells. In contrast, SP-A has been shown to inhibit lipopolysaccharide (LPS)-induced TNF- production by alveolar macrophages (4, 6), the macrophage-like U937 cells (6), or human buffy coat cells (5). Recently, SP-A has been reported to inhibit production of several cytokines by human alveolar macrophages and monocytes stimulated with Candida albicans (7). To the best of our knowledge, the effects of SP-D on cytokine production have not been reported.

The effects of SP-A on the production of reactive nitrogen species are also controversial. For example, SP-A has been shown to have no effect on (8) or to stimulate (12) the production of nitric oxide (NO) by alveolar macrophages. Miles and coworkers (9) and Blau and colleagues (12) reported that SP-A has no effect on LPS-induced NO production by rat alveolar macrophages. Hickman-Davis and associates (11) showed that SP-A enhanced in vitro killing of Mycoplasma pneumoniae by alveolar macrophages from C57 BL/6 mice through a NO-dependent mechanism. More recently, Hickman-Davis and coworkers (13) reported that SP-A-deficient mice have an increased susceptibility to M. pneumoniae infection compared with wild-type mice. In contrast, Pasula and colleagues (10) reported that SP-A suppresses NO production by interferon (IFN)--treated mice alveolar macrophages in response to Mycobacterium tuberculosis. Both SP-A (14) and SP-D (15) have been reported to stimulate directly the production of oxygen radicals. Although endotoxin contamination can affect SP-A-mediated function (8), the presence of endotoxin in SP-A cannot entirely explain these disparate results because SP-A containing low levels or no detectable endotoxin has been used in many of these investigations.

We hypothesized that the effects of SP-A and SP-D on immune cell function may depend on the mechanism and state of activation of immune cells. To test this hypothesis, we examined the effects of SP-A and SP-D on the production of NO by alveolar macrophages stimulated with either LPS, IFN-, or both agonists. The results indicate that SP-A can exert opposite effects on immune cells activated with different agonists.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Reagents

Water from a Picosystem water purification system (Hydro Water Systems, Research Triangle Park, NC) was used for all studies. The water quality specifications for this system include a pyrogen content less than 0.25 endotoxin U/ml. LPS from the LCD 25 strain of Escherichia coli K12 (Rb-LPS mutant) and LPS of E. coli 0111:B4 (S-LPS) were purchased from List Biological Laboratories (Campbell, CA). Low endotoxin (less than 0.1 ng/mg) bovine serum albumin (BSA) was from Sigma Chemical (St. Louis, MO), nitrate-free basal medium Eagle and Dulbecco's phosphate-buffered saline were from GIBCO BRL (Grand Island, NY), and heat-inactivated fetal calf serum (FCS) was from Atlanta Biologicals (Norcross, GA). Human C1q was from Advanced Research Technologies (San Diego, CA). Rat recombinant IFN- at 2.2 × 10⁵ U/ml was purchased from Calbiochem (La Jolla, CA); the level of endotoxin in a representative preparation was 0.79 pg/U IFN- as measured by the ACL-1000 Limulus Amebocyte Lysate kit (BioWhittaker, Walkersville, MD). Mouse monoclonal anti-macrophage inducible nitric oxide synthase (iNOS) immunoglobulin (Ig) G and a macrophage lysate prepared from RAW 264.7 cells stimulated with IFN- and LPS, which served as the standard, were from Transduction Laboratories (Lexington, KY). Horseradish peroxidase (HRP)-conjugated rabbit anti-mouse IgG was from Pierce Laboratories (Rockford, IL). The BIOXYTECH Nitric Oxide Assay was obtained from OXIS International (Portland, OR). All other chemicals (except as noted) were from Sigma Chemical (St. Louis, MO).

Purification of SP-A

SP-A was isolated from the lung lavage of patients with alveolar proteinosis as previously described in detail (16). Briefly, the whole surfactant pellet from lung lavage was extracted with butanol, butanol-insoluble proteins were resolubilized with octylglucopyranoside (OGP), and SP-A was finally solubilized in 5 mM Tris-buffered water, pH 7.4. Residual OGP was removed by dialysis against 5 mM Tris-buffered water, pH 7.4. To remove endotoxin, SP-A was treated with OGP and polymyxin B agarose beads as previously described in detail (8). SP-A preparations were tested for bacterial endotoxin using a Limulus Amebocyte Lysate assay (BioWhittaker). All SP-A preparations used contained < 0.02 pg endotoxin/µg SP-A. For some experiments, SP-A was heat-inactivated by incubation at 95°C for 10 min.

Purification of SP-D

SP-D was isolated from the lung lavage of silica-treated rats as described (17). Briefly, to induce surfactant accumulation, anesthetized rats received 25 mg silica in 0.5 ml of saline intratracheally, they were killed by exsanguination approximately 4 wk later, and their lungs were lavaged six times with 150 mM NaCl, 5 mM Tris, pH 7.4. Lavage samples were centrifuged at approximately 25,000 × g for 45 min at 4°C (model L8-70; Beckman, Fullerton, CA). The supernatant was mixed with maltose-sepharose (18) in the presence of 5 mM CaCl₂ and incubated for 18 h at 4°C. The slurry was centrifuged at 2,400 × g for 10 min at 10°C. The supernatant was removed and the slurry was washed four times by resuspending the slurry in 20 mM Tris, 150 mM NaCl, 5 mM CaCl₂, pH 7.8, incubating it for 10 min at room temperature, and centrifuging the mixture at 2,400 × g for 10 min. After the final wash, the supernatant was removed and the maltose-Sepharose was resuspended in 50 mM Tris, 150 mM NaCl, 10 mM ethylenediaminetetraacetic acid (EDTA), pH 7.8. The Ca²⁺-dependent maltose-binding proteins were eluted from the beads by washing them six times. The fractions containing the highest SP-D concentrations were pooled and further purified by gel filtration (19). To remove contaminating endotoxin, SP-D was treated with octylglucoside and polymyxin as previously described (8). The two SP-D preparations used contained 0.114 pg endotoxin/µg SP-D and 0.23 pg endotoxin/µg SP-D.

Isolation of Cells for Nitrite Measurement

Alveolar macrophages were isolated by lung lavage of specific pathogen-free, male, Sprague-Dawley rats weighing 250 to 300 g (Charles River Laboratory, Raleigh, NC). In brief, rats were anesthetized with pentobarbital and killed by exsanguination. The lungs were removed and lavaged six times with a macrophage isolation buffer containing 140 mM NaCl, 6 mM glucose, 2.5 mM phosphate buffer, 10 mM N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid (Hepes), and 0.2 mM ethyleneglycol-bis-(-aminoethyl ether)-N,N'-tetraacetic acid (EGTA), followed by two lavages with the same buffer containing 1.3 mM magnesium and 2 mM calcium and no EGTA. Lavage samples were centrifuged at 200 × g for 10 min. The cells were resuspended in nitrate-free basal medium Eagle containing 110 mg/liter of L-glutamine, 50 U/ ml of penicillin, 50 µg/ml of streptomycin, 15 mM Hepes with 10% (vol/vol) heat-inactivated FCS. Cell recovery routinely averaged 3 to 8 × 10⁶ cells/animal. The viability of the cells was determined by erythrosin B exclusion and averaged 95 to 97%. The cell lysate fractions were assayed for protein content by the BCA method (BCA kit; Pierce) with BSA as a standard.

Incubation Conditions

Cells were plated at 2.5 × 10⁵ cells/well in 96-well plates (Costar, Cambridge, MA) and allowed to attach for 2 h. The medium was then removed and, for some conditions, alveolar macrophages were stimulated with rat recombinant IFN- (10 or 100 U/ml) for 24 h at 37°C in a 5% CO₂ atmosphere in the absence or in the presence of SP-A or SP-D. After preincubation, Rb- or S-LPS was added at 100 ng/ml for an additional 24 h. After 48 h, media were removed and briefly centrifuged to remove nonadherent cells before analysis of nitrite content. For time course experiments, cells were preincubated with SP-A (100 µg/ml) in the presence (10 U/ml) or absence of IFN- for 24 h; Rb-LPS (100 ng/ml) was then added for 10 min, 4, 8, 12, 16, or 24 h. In separate experiments, cells were preincubated with SP-A (100 µg/ml) for 24 h and IFN- (10 U/ml) was added for an additional 24 h. For time course experiments with IFN- and SP-A, cells were stimulated with IFN- (10 U/ml) in the absence or presence of SP-A (100 µg/ml) and accumulated nitrite was determined at 10 min, 6, 12, and 24 h. Cell viability was assessed by erythrosin B dye exclusion and averaged 94 to 98%.

Measurement of Nitrite Production

NO production was assessed by measuring nitrite in media fractions by the Greiss reaction as previously described (20). Nitrite concentrations were determined from a standard curve using sodium nitrite at concentrations ranging from 6.7 to 133.3 µM. Results are expressed as nitrite concentration or as a percentage of the levels of nitrite in the control group. For measurement of total nitrite (NO₃/NO₂), the BIOXYTECH Nitric Oxide Assay (OXIS) was used. The kit employs affinity purified Zea Mays nitrate reductase and NADH to reduce nitrate (NO₃) to nitrite NO₃. NO₃ was then measured using the Greiss reaction as described previously. The standard curve ranged from 0.5 to 25 µM.

Microtiter Plate LPS Binding Assay

Immulon-2 96-well microtiter plates were coated with LCD 25 E. coli K 12 LPS or 0111:B4 E. coli LPS (5 µg/well), AP-SP-A (positive control for SP-D binding) and myosin (positive control for SP-A binding [550 ng/well]) or nothing (BSA) in 0.1 M Na₂CO₃, 20 mM EDTA, pH 9.6, for 3 h at 37°C. After incubation, the LPS solution was removed, and the plates were air-dried overnight. Nonspecific binding sites were blocked with 200 µl/well of 10 mg/ ml low endotoxin BSA in 50 mM Hepes, 0.15 M NaCl, pH 7.4 for 30 min. Plates were washed three times and biotinylated SP-A (bSP-A) or bSP-D (1 to 20 µg/ml) was added. Plates were incubated at 4°C overnight. After washing, biotinylated streptavidin-HRP was added at a 1:500 dilution and incubated for 3.5 h at room temperature. After additional washing, color was developed using H₂O₂ and 0-phenylenediamine dihydrochloride in 0.1 M citric acid, pH 5. Color reaction was stopped with 4 N H₂SO₄, and the color intensity was measured in a microtiter plate spectrophotometer at 490 nm.

Detection of iNOS Protein

To determine if SP-A affected LPS or IFN--induced iNOS levels in alveolar macrophages, Western blot analysis was performed. Cell lysate fractions were assayed for protein content by the BCA method (BCA kit; Pierce). Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions was performed on either 30 or 40 µg of cell protein using either 12.5 or 15% polyacrylamide gels. The proteins were transferred to nitrocellulose membranes, which were then blocked with 5% nonfat dry milk in 10 mM Tris, 100 mM NaCl, 0.1% Tween 20 for 1 h at room temperature. The blots were then incubated for an additional hour at room temperature in 0.1% Tween 20 with 5% nonfat dry milk with a mouse monoclonal anti-macrophage iNOS IgG at a dilution of 1:2,500 or 1:1,000. Membranes were then washed three times in Tris buffer, followed by a 1-h incubation at room temperature in blocking buffer containing HRP-conjugated rabbit antimouse IgG at a dilution of 1:5,000 or 1:2,000. After incubation, blots were washed three times and immunoreactive proteins were visualized by enhanced chemiluminescence using the enhanced chemiluminescence Western blotting detection system (Amersham Life Science, Arlington Heights, IL). A macrophage lysate prepared from RAW 264.7 cells stimulated with IFN- and LPS served as the standard. Samples were also assessed for actin content. Membranes were incubated with a mouse monoclonal antiactin IgG from ascites (Chemicon Labs, Temecula, CA) at a 1:400 dilution. Rabbit antimouse IgG-HRP at a 1:5,000 dilution served as the secondary antibody. Autoradiographs were analyzed by densitometric scanning and analyzed using NIH image.

Statistics

Data were either analyzed with the Mann-Whitney U test or Student's t test for unpaired samples or with paired Student's t test when expressed as percentage of the control response. Values were considered significant when P  0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

SP-A Inhibits LPS-Induced NO Production

Consistent with previous reports from Miles and colleagues (9) and Pasula and coworkers (10), we found that SP-A alone did not induce production of detectable nitrite by alveolar macrophages incubated for up to 48 h with SP-A (Figure 1). In contrast, SP-A dramatically inhibited production of nitrite by alveolar macrophages stimulated with Rb-LPS. Preincubation of cells with SP-A for 24 h before the addition of Rb-LPS resulted in an 89, 61, and 44% inhibition of Rb-LPS-induced NO production at 12, 16, and 24 h after addition of Rb-LPS, respectively, suggesting that the inhibitory effect of SP-A is decreasing over time (Figure 1). Of note, SP-A at concentrations as high as 100 µg/ ml did not interfere with the nitrite assay (data not shown).

View larger version (17K): [in this window] [in a new window]   Figure 1.   Time-dependent effects of SP-A on Rb-LPS-induced NO production by alveolar macrophages. Alveolar macrophages (2.5 × 105/well) were preincubated in the absence or presence of SP-A (100 µg/ml) for 24 h at 37°C in a 5% CO2 atmosphere. After 24 h, Rb-LPS (100 ng/ml) was added to some samples and medium was collected at later times points ranging from 10 min to 24 h and analyzed for nitrite by the Greiss reaction. Values shown are means of two representative experiments performed in duplicate.

SP-A Augments IFN--Induced NO Production

In contrast to its inhibitory effects on LPS-induced production of NO, SP-A enhanced the production of NO by cells stimulated with IFN-. As shown in Figure 2, detectable enhancement was evident at 12 h and persisted over the 24-h time course of the experiment. At 24 h, SP-A enhanced IFN--induced NO by approximately 200% over the levels obtained with IFN- in the absence of SP-A. When SP-A was preincubated for 24 h before addition of IFN-, the nitrite level was 6.9 µM. IFN- alone induced nitrite levels of 3.2 µM. Thus, SP-A enhances nitrite induced by IFN- when preincubated before addition of the IFN- or when added simultaneously with the IFN-.

View larger version (16K): [in this window] [in a new window]   Figure 2.   Time-dependent effects of SP-A on IFN--induced NO production by alveolar macrophages. Alveolar macrophages (2.5 × 105/well) were incubated with IFN- (10 U/ml) in the absence or presence of SP-A (100 µg/ml) at 37°C in a 5% CO2 atmosphere. Media fractions were collected at 10 min, 6, 12, and 24 h, and analyzed for nitrite levels. Values shown are means ± standard error (SE) for three to five experiments. *P < 0.05 compared with zero time point. The nitrite levels obtained in the presence of SP-A are significantly greater than values obtained in the absence of SP-A when data were calculated as a percentage of control (no SP-A), P < 0.05.

The effect of SP-A was dependent on the dose of IFN- and was more pronounced at lower doses than at higher doses (Figure 3). The enhancing effects of SP-A on IFN--induced NO, as well as the inhibitory effects of SP-A on LPS-induced NO production, were dependent on the dose of SP-A (Figure 4). SP-A also enhanced production of NO when alveolar macrophages were stimulated with both LPS and IFN- (Figure 4).

View larger version (14K): [in this window] [in a new window]   Figure 3.   Effects of SP-A (solid diamonds) on NO production induced by different concentrations of IFN- (open squares). Alveolar macrophages (2.5 × 105 cells/ml) were incubated with various concentrations of IFN- either in the absence or presence of 100 µg SP-A/ml. Media fractions were collected after 48 h and nitrite levels were determined. Values shown are means ± SE for three to five experiments. *P < 0.05 compared with IFN- alone. +P < 0.01 compared with IFN- alone.

View larger version (15K): [in this window] [in a new window]   Figure 4.   Dose-dependent effects of SP-A on Rb-LPS (open squares), IFN- (open circles), or Rb-LPS plus IFN--induced (solid circles) NO production. Alveolar macrophages (2.5 × 105 cells/ml) were preincubated in the presence or absence of IFN- (100 U/ml) and varying doses of SP-A for 24 h before addition of Rb-LPS (100 ng/ml) to some samples and continuation of the incubation for another 24 h. Media samples were collected and analyzed for nitrite by the Greiss reaction. Data are expressed as percentage of the corresponding (no SP-A) control level of nitrite. Values shown are means ± SE for three to four experiments.  P < 0.01, *P < 0.05 compared with controls (no SP-A).

The effects of SP-A on LPS and IFN--induced NO₃/ NO₂ were analyzed by converting nitrate to nitrite and measuring nitrite with the Greiss reaction. The trends for the effects of SP-A on NO₃/NO₂ were similar to the effects on nitrite. For example, SP-A inhibited LPS-induced NO₂ levels by 40% and inhibited NO₃/NO₂ levels by 26%. In addition, SP-A enhanced NO₂ levels in the presence of IFN- by 325% and NO₃/NO₂ by 141% (means from two different experiments).

Inhibitory Effect of SP-A on LPS-Induced NO Production Does Not Require a High Degree of Binding of SP-A to LPS

To determine if SP-A inhibited LPS action by binding LPS, the effects of SP-A on macrophage responses to a smooth serotype of LPS were investigated. SP-A inhibited production of NO stimulated by S-LPS (Figure 5) as well as Rb-LPS (Figures 1 and 4) in a dose-dependent manner. As shown in Figure 6, biotinylated SP-A bound the Rb-LPS mutant in a dose-dependent manner (Figure 6A) but only minimally bound the smooth LPS serotype from E. coli 0111:B4. Biotinylated SP-D bound in a dose-dependent manner to Rb-LPS and but only slightly to S-LPS (Figure 6B); SP-D did not affect production of NO by alveolar macrophages stimulated with Rb-LPS (see subsequent text).

View larger version (15K): [in this window] [in a new window]   Figure 5.   Dose-dependent effects of SP-A on S-LPS (open squares), IFN- (open circles), or S-LPS plus IFN--induced (solid circles) NO production by alveolar macrophages. Alveolar macrophages were preincubated with varying doses of SP-A either in the presence or absence of IFN- (100 U/ml) for 24 h before addition of S-LPS (100 ng/ml) to some samples for another 24 h. Values shown are means ± SE for three to five experiments. Data are expressed as a percentage of the corresponding controls (no SP-A). *P < 0.05 compared with controls (no SP-A).

View larger version (13K): [in this window] [in a new window]   Figure 6.   Microtiter plate analysis of binding of biotinylated SP-A (bSP-A) or bSP-D binding to Rb- or S-LPS. Rb-LPS and S-LPS were coated onto immulon-2 microtiter plates and incubated with bSP-A (A) or bSP-D (B) followed by biotinylated streptavidin-HRP complex as described in MATERIALS AND METHODS. Data are expressed as the absorbance of the developed reaction at 490 nm. Data are representative of two experiments.

SP-A Affects Levels of iNOS Protein in Alveolar Macrophages

The effects of SP-A on iNOS levels were analyzed by Western blotting of macrophages stimulated with IFN-, LPS, or the combination of IFN- and LPS. As shown in Figure 7A, SP-A alone had no detectable effect on iNOS levels. Smooth LPS dramatically increased iNOS protein, and SP-A inhibited this induction by approximately 88% at a concentration of 20 µg/ml and 99% at 100 µg/ml as assessed by densitometric analysis. SP-A also inhibited Rb-LPS- induced iNOS protein expression (Figure 7A) by approximately 82% at a concentration of 100 µg SP-A /ml. SP-A significantly enhanced the magnitude of IFN--induced iNOS protein expression (Figure 7B). At a concentration of 100 µg/ml, SP-A enhanced iNOS protein levels by approximately 290% after 24 h incubation.

View larger version (31K): [in this window] [in a new window]   View larger version (45K): [in this window] [in a new window]   Figure 7.   Western blot analysis of the effects of SP-A (100 µg/ml) on iNOS protein expression induced by S-LPS and Rb-LPS (upper panels) or IFN--induced iNOS protein expression by rat alveolar macrophages (lower panel). Cells were incubated with either S-LPS, Rb-LPS, or IFN- (10 U/ml) as described in MATERIALS AND METHODS. Cell lysate fractions were assayed for protein content by the BCA method, and SDS-PAGE was used to separate 40-µg aliquots of cell protein. Proteins were transferred to a nitrocellulose membrane and probed with a monoclonal anti-iNOS antibody.

Effects of SP-D and Other Immunomodulatory Proteins on NO Formation Induced by LPS or IFN-

The two pulmonary collectins share some but not all functions (1, 2). Therefore, we assessed the role of SP-D in regulating production of NO by macrophages stimulated with IFN-, Rb-LPS, or Rb-LPS in combination with IFN-. In contrast to the inhibitory effects of SP-A, SP-D (0.1 to 100 µg/ml) had no significant effect on Rb-LPS-induced NO production by alveolar macrophages (Table 1). Also in contrast to SP-A, SP-D did not significantly augment NO production in the presence of IFN- or IFN- plus Rb-LPS. Furthermore, neither IgG, C1q, nor heat-inactivated SP-A affected nitrite production (Table 2) induced by Rb-LPS or IFN-.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The goal of this investigation was to test the hypothesis that SP-A exerts differential effects on alveolar macrophages that are stimulated with different agonists. This hypothesis was generated, at least in part, by contradictory reports on the effects of SP-A on the production of cytokines and reactive nitrogen species by immune cells from a variety of species both in the absence and presence of pathogens (3). Additional support for the hypothesis also comes from our unpublished studies in which we found variable effects of SP-A on LPS-induced NO production. In some experiments SP-A dramatically inhibited NO production and in others it had no effect. When attempting to assess the variables that might account for these differences, we postulated that a contributing factor might be the immune status of the animal, including whether or not the animals were pathogen-free. In the current study, we investigated the effects of SP-A on NO and iNOS protein expression induced by LPS, IFN-, or the combination of both. Our findings show that SP-A does indeed have opposite effects on the production of NO induced by different agonists and suggest that SP-A may exert both cell- and agonist-specific effects that contribute to the inflammatory state of the lung.

Neither SP-A nor SP-D Directly Stimulates NO Production or iNOS Protein Expression by Rat Alveolar Macrophages

In the current investigation, neither SP-A nor SP-D that had been treated to reduce contaminating endotoxin (8) directly stimulated NO production by rat alveolar macrophages. These findings are consistent with those reported for SP-A by Miles and coworkers (9), also using rat alveolar macrophages, and by Pasula and associates (10), using mouse alveolar macrophages. However, these results differ from those of Blau and coworkers (12) who reported that SP-A directly stimulates nitrite production by rat alveolar macrophages. We do not know the reason for these discrepant findings. Blau and coworkers (12) reported that the SP-A used in their studies had low levels of measurable endotoxin, so it seems unlikely that the effects can be attributed to endotoxin. However, they did not use pathogen-free rats for all of their studies, whereas we did use pathogen-free animals. We speculate that, in some cases, rats that are not pathogen-free may be exposed to a stimuli that may alter both their in vivo and ex vivo cell responses.

SP-A, in Contrast to SP-D, Inhibits LPS-Induced NO Production and iNOS Protein Expression

Preincubation of cells with SP-A, but not SP-D, resulted in a significant and dose-dependent inhibition of LPS- induced NO production and iNOS protein expression. This effect of SP-A was independent of the two LPS phenotypes tested. These results are in contrast to those obtained by Miles and coworkers (9) and Blau and colleagues (12), who both reported no inhibitory effect of SP-A on LPS-induced NO production by rat alveolar macrophages. The contradictory results may depend largely on the different incubation conditions and the molar ratios of LPS and SP-A used. For example, Miles and associates (9) coincubated SP-A with LPS at a molar ratio of approximately 1:40 for 22 h; Blau and coworkers (12) coincubated SP-A with LPS at a molar ratio of approximately 1:156 for 48 h. In contrast, we used 100 ng/ml of LPS and SP-A ranging from 20 to 100 µg/ml. Thus, in our studies, the highest ratio of SP-A to LPS tested was 1:0.8. Although the precise concentration of SP-A present in vivo is not known, we have estimated that the hypophase concentration of SP-A ranges from 300 µg/ml to 1.8 mg/ml, although a large proportion of the SP-A is likely bound to lipid and not homogeneously distributed (1). Because the concentration of LPS is an important determinant of CD14-dependent and -independent cell binding and activating pathways in the presence of serum (21), some of the conflicting data may be due to differential inhibitory effects of SP-A on LPS interaction with different receptors. When we coincubated LPS and SP-A with macrophages for 24 h and then measured nitrite levels, we were unable to demonstrate consistent inhibitory effects of SP-A on LPS-induced NO production, even at high SP-A to LPS molar ratios (data not shown). These data are consistent with the possibility that an adequate time period of preincubation of cells with SP-A is necessary for optimal inhibition of LPS-induced NO. We were initially surprised that the effects of SP-A were so long lived in light of the published reports that SP-A is degraded by alveolar macrophages (22, 23). Western blot analysis of media after the 48-h incubation period showed that levels of SP-A were reduced during the incubation, but substantial amounts of SP-A remained. Thus, it appears that although SP-A is degraded, sufficient amounts remain to affect nitrite production under these experimental conditions.

Possible Mechanisms for the Differential Effects of SP-A and SP-D on LPS-Induced iNOS-Derived NO

There are several possible mechanisms by which SP-A may inhibit LPS-induced NO production by alveolar macrophages. To determine if binding of SP-A to LPS was required for inhibition, we tested the effects of SP-A on NO induction by two serotypes of LPS. SP-A significantly bound Rb-LPS but only slightly bound to S-LPS. Despite this differential binding, SP-A inhibited nitrite production induced by both serotypes of LPS. These data are consistent with reports of others showing that SP-A inhibits LPS-mediated function independent of the degree of binding of LPS and SP-A. For example, McIntosh and coworkers (4) and Sano and colleagues (6) found that SP-A suppresses S-LPS-induced TNF- in alveolar macrophages. In addition, Rosseau and associates (7) reported that SP-A inhibits the production of a variety of proinflammatory cytokines by alveolar macrophages and monocytes stimulated with C. albicans and that this inhibition appears to be independent of SP-A binding to Candida. Like SP-A (6, 24), SP-D (26) binds to Rb-LPS. Kuan and associates (26) reported a significant binding of SP-D to Rd and Rc E. coli LPS, but not to the Re and Ra mutants. The functional consequences of this interaction have not yet been investigated. We now show that SP-D also significantly and in a dose-dependent manner binds to the Rb mutant of E. coli LPS but does not affect NO production or iNOS protein levels induced by the same Rb-LPS mutant, suggesting that binding of LPS by SP-D is not sufficient to inhibit LPS-induced NO production.

Sano and coworkers (6) proposed that the effects of SP-A on LPS-induced production of TNF- occur via a direct interaction of SP-A with CD14, which is a component of the LPS-LPS binding protein receptor (27). In their study, SP-A inhibited TNF- production by a smooth serotype of LPS (026:B6), which did not significantly bind to SP-A. In contrast, SP-A did not inhibit and, in fact, increased production of TNF- in response to a rough serotype of LPS from Re595-Salmonella minnesota, which avidly bound to SP-A. SP-A was shown to bind to CD14 and, furthermore, SP-A enhanced binding of rough LPS and inhibited binding of smooth LPS to a soluble recombinant CD14. We have found that SP-A enhances the binding of ³H-labeled Rb-LPS by rat alveolar macrophages in the absence (28) and presence of serum (unpublished data). Unlike the finding of Sano and colleagues (6) that SP-A enhances TNF- production induced by a rough LPS (Re-LPS from S. minnesota), we found that SP-A inhibited TNF- production induced by a different rough LPS (Rb-LPS from E. coli, unpublished data). It is possible that the effects of SP-A may vary with the rough LPS mutants used or other experimental variables such as the presence and/or type of serum used.

SP-A may also affect postreceptor events in the LPS- induced signaling pathways for NO induction. LPS stimulation of NO and iNOS has been shown to be inhibited by a variety of mechanisms, including elevation of intracellular cyclic adenosine monophosphate (29), inhibition of protein kinase C activity (29), inhibition of tyrosine kinase activity (30), and antagonists of Ca²⁺ channel activity (31). Little is known about the regulation of signaling pathways by SP-A, but the available data do not provide a clear explanation. The iNOS promoter has two nuclear factor-B (NF-B) binding sites as well as an IFN-stimulated response element. Koptides and coworkers (32) reported that SP-A enhances NF-B activity in a monocytic cell line, THP-1. If SP-A were acting via NF-B in our system, we would have predicted that SP-A would inhibit NF-B activation. However, we cannot exclude the possibility that SP-A may elicit different responses in different cells.

SP-A Dose-Dependently Enhanced IFN--Induced NO Production and iNOS Protein Levels

SP-A enhanced IFN--induced, iNOS-derived NO both after preincubation of cells with SP-A for 24 h before addition of IFN- and when both were added simultaneously. This effect of SP-A was more pronounced at low doses of IFN- (10 U/ml) than at higher doses of IFN-. In addition, not only is the inhibitory effect of SP-A on LPS-induced NO completely abolished in the presence of low doses of IFN-, but the well-known synergism between IFN- and LPS is further increased.

The effects of SP-A on NO production induced by IFN- in combination with another agonist or a pathogen may be influenced by both the source of the cells and the agonist or pathogen. For example, Pasula and coworkers (10) reported that SP-A inhibits IFN--induced NO production by mouse alveolar macrophages in the presence of M. tuberculosis. In contrast, Hickman-Davis and colleagues (11) reported that SP-A enhances production of NO by mouse macrophages stimulated with IFN- and subsequently infected with Mycoplasma pulmonis. These results suggest that the effects of SP-A may be pathogen specific. Hickman-Davis and colleagues (11) did not report the effects of SP-A on IFN- on NO production in the absence of M. pulmonis. However, Pasula and associates (10) showed that the very low level of NO induced by IFN- (in the absence of M. tuberculosis) was inhibited by SP-A. These studies all suggest that the method of activation of the cell, the pathogen to which the cells are exposed, as well as the source of the cells may have an impact on the response of SP-A.

Possible Mechanisms for SP-A-Enhanced IFN--Induced NO

We have shown that SP-A potentiates IFN--induced iNOS protein expression and has a robust enhancing effect on NO production for up to 48 h. It is unclear at this time how SP-A exerts this enhancing effect on IFN--induced NO. SP-A may enhance functionally active binding of IFN- to its receptor (IFN-R) by altering the expression of IFN-R -chains, which, in contrast to IFN-R -chains, can be upregulated by external stimuli (33). Alternatively, SP-A has been shown to trigger a rapid tyrosine, but not serine or threonine, phosphorylation (34) of alveolar macrophage proteins and could possibly enhance/accelerate initial phosphorylation steps of the JAK/STAT pathway, which participates in the regulation of iNOS expression. Finally, SP-A may enhance IFN--induced NO via an autocrine mechanism, for example, by increasing INF--induced TNF-. However, we were unable to document SP-A- enhanced TNF- production in response to IFN- after 24 h of incubation (unpublished data).

Summary

In a previous study, we investigated the role of SP-A in the regulation of TNF- and nitrite production induced by LPS in wild-type and SP-A-deficient mice (35). We found that intratracheal instillation of a smooth serotype of LPS (026:B6) induced greater lung lavage levels of nitrite and TNF- in SP-A-deficient mice than in wild-type mice, findings consistent with the in vitro studies reported here. Important future studies include investigations of SP-A regulation of immune cell function in SP-A-deficient mice that have been treated with IFN- as well as studies to elucidate the mechanisms by which SP-A elicits differential responses from cells activated with different agonists and different pathogens.