Surfactant protein A (SP-A) has been implicated in the regulation of pulmonary host defense and inflammatory events. We analyzed the impact of SP-A on the Candida albicans-induced cytokine response in human alveolar macrophages (AM) and its precursor cells, the monocytes, which rapidly expand the alveolar mononuclear phagocyte pool under inflammatory conditions. Both recombinant human SP-A and natural canine SP-A were employed. SP-A dose-dependently down-regulated the proinflammatory cytokine response of AM and monocytes to both viable and nonviable Candida, including TNF-α, IL-1β, macrophage inflammatory protein-1α, and monocyte chemoattractant protein-1. In contrast, SP-A did not affect the baseline liberation of these cytokines. The release of the antiinflammatory cytokines IL-1 receptor antagonist and IL-6 was not inhibited by SP-A under baseline conditions and in response to fungal challenge. The inhibitory effect of SP-A on proinflammatory cytokine release was retained upon reassembly of the apoprotein with natural surfactant lipids and in the presence of serum constituents, for mimicry of plasma leakage into the alveolar space. It was not reproduced by the homologous proteins complement component C1q and type IV collagen. It was independent of Candida-SP-A binding and phagocyte C1q receptor occupancy, but apparently demanded SP-A internalization by the mononuclear phagocytes, effecting down-regulation of proinflammatory cytokine synthesis at the transcriptional level. We conclude that SP-A limits excessive proinflammatory cytokine release in AM and monocytes confronted with fungal challenge in the alveolar compartment. These data lend further credit to an important physiological role of SP-A in regulating alveolar host defense and inflammation.

Pulmonary surfactant is a complex mixture of lipids and proteins that lowers the surface tension at the air-liquid interface of the lung. Surfactant protein A (SP-A),3 the major hydrophilic protein component of the surfactant system, is a member of the collectin family (1), sharing sequence homology with serum mannose-binding protein (MBP; Ref. 2), surfactant protein D (3), conglutinin (4), CL-43 (5), and the complement component C1q (6, 7). Collectins have a collagen-like triple helical domain and bind carbohydrates in a calcium-dependent manner. They are important constituents of innate immunity (8, 9), acting as opsonins for a variety of bacteria (10, 11) and viruses (12), and stimulating chemotaxis (13) and oxidative burst (11) of phagocytes.

Leukocytes obtained from the alveolar compartment by lavage techniques have repeatedly been shown to be hyporesponsive to inflammatory stimuli as compared with leukocytes isolated from peripheral blood (14). This relative “dampening” of leukocyte activity within the alveolar space is thought to protect the host from persistent immune cell activation via inhaled Ags, since ongoing inflammatory events in the alveolar compartment might result in progressive destruction and/or fibrotic remodeling of lung tissue and ultimately result in severe impairment of pulmonary gas-exchange function. Surfactant has since long been implicated in this suppression, since surfactant lipid mixtures and individual lipid components were noted to inhibit lymphocyte proliferation and Ig secretion (15), as well as phagocyte oxygen radical production (16). Recently, immunosuppressive activity was also demonstrated for SP-A, since this major surfactant protein inhibited lymphocyte proliferation and IL-2 production (17), and reduced TNF-α generation in endotoxin-stimulated macrophages (18). However, others reported that SP-A per se stimulated proinflammatory cytokine production in mononuclear cells, secretion of Igs by splenocytes, and proliferation of lymphocytes (19, 20).

In this study, we analyzed the impact of SP-A on the Candida albicans-induced cytokine response in alveolar macrophages (AM) and its precursor cells, the monocytes, which rapidly expand the alveolar mononuclear phagocyte pool under inflammatory conditions. Isolated Candida cell wall oligosaccharides have previously been shown to stimulate TNF-α synthesis in AM (21), and TNF-α production in the alveolar compartment contributes significantly to local inflammatory tissue injury and lethal septic shock in disseminated candidemia (22). Further cytokines centrally involved in lung inflammation and leukocyte recruitment include IL-1β, IL-8, macrophage inflammatory protein (MIP)-1α and monocyte chemoattractant protein (MCP)-1 (23). We now show that the Candida-elicited mononuclear phagocyte production of all these proinflammatory cytokines, but not that of the antiinflammatory agents IL-1 receptor antagonist (IL-1Ra) and IL-6 (24, 25), is dramatically inhibited by SP-A. Focussing on TNF-α as central proinflammatory cytokine, we further demonstrate that the SP-A effect is robust, since it is also operative in the presence of surfactant lipids and plasma constituents, and that it is independent of fungal SP-A binding but is rather related to some direct impact of the apoprotein on the phagocyte signaling events with down-regulation of the TNF-α gene expression. Interestingly, the SP-A effect is not reproduced by the homologous proteins complement component C1q and type IV collagen. In summary, these findings strongly support an important role of SP-A in down-regulating AM and monocyte inflammatory response to fungal challenge in the alveolar compartment.

Stock cultures of C. albicans (kindly provided by E. Martin and S. Bakdhi, Institute of Medical Microbiology, Mainz, Germany) were maintained on Columbia agar with 5% sheep blood (Becton Dickinson, Heidelberg, Germany) at 4°C. Agar cultures were restored by culturing an aliquot of frozen yeast stock for 48 h at 37°C. For each experiment, one colony was inoculated into 2 ml of tryptic soy broth (TSB; Sigma, Munich, Germany) and grown overnight at 37°C. After 12 h, 10 μl of yeast suspension was drawn into 50 ml fresh tryptic soy broth and incubated again for 12 h at 27°C on a shaker. At the end of the second culture period, yeast were centrifuged for 10 min at 1500 × g and room temperature, and resuspended in sterile, endotoxin-free saline. Under these conditions C. albicans grew as a >95% pure yeast phase population and remained >96% viable as determined by the exclusion of trypan blue dye. For the use of nonviable yeast, Candida were killed by incubation in 70% ethanol for 60 min at room temperature followed by three washing steps in sterile saline (26). The endotoxin content of saline and Candida supernatants was assayed by the Limulus amebocyte lysate assay (COATEST endotoxin; Chromogenix, Mölndal, Sweden) and always ranged <10 pg/ml. Possible bacterial contamination of the Candida suspension cultures was excluded by reculture experiments on agar plates.

Healthy volunteers underwent single bronchoscopy and bronchoalveolar lavage (BAL). All participants were nonsmokers, had no smoking history and no history of cardiac or pulmonary disease, were free of respiratory symptoms, were not taking any medication, and had normal lung function testing. The study was approved by the local ethics committee, and written, informed consent was obtained from all participants. Bronchoscopy and BAL were performed by means of a fiberoptic bronchoscope, with 10 aliquots of 20 ml warm sterile saline being infused into one segment of the middle lobe or lingula and removed by gentle suction with a total recovered volume of 150–180 ml. The lavage fluid was centrifuged for 10 min at 200 × g and room temperature, and the cell pellet was washed twice in endotoxin-free HBSS without Ca2+ and Mg2+ (HBSS; Life Technologies, Eggenstein, Germany). Cells were counted with a hemocytometer, viability was assessed by trypan blue exclusion, and differential cell counting was performed in Pappenheim-stained cytocentrifuge preparations. The BAL cells were composed of 93–96% macrophages, 3–6% lymphocytes, and 0–1% neutrophil granulocytes. Viability of the cells consistently ranged >95%.

Human monocytes were isolated using a combination of ficoll density gradient centrifugation (Ficoll Hypaque, Pharmacia, Freiburg, Germany) and counterflow centrifugal elutriation (Beckmannn J2–21 M/E centrifuge with JE-B6 Elutriator Rotor, standard 5-ml Elutriation Chamber, Beckman Instruments, Palo Alto, CA). Cell counts and viability were determined by hemocytometer counts of trypan blue-stained aliquots. Cytocentrifuge preparations were examined after Pappenheim staining for differential cell counting. The monocyte fraction consisted of 88–93% monocytes, 7–12% lymphocytes, and 0–1% granulocytes; cell viability always ranged >96%. Buffers and reagents were tested for endotoxin content, which always ranged <10 pg/ml. The elutriation chamber was rinsed with Detoxaclean (Sigma) to eliminate possible endotoxin contamination and was sterilized by the use of 70% ethanol. The endotoxin content of chamber effluents ranged always <10 pg/ml.

Monocytes and AM were cultured at a concentration of 5 × 105/ml in MEM (Life Technologies) with 0.1% human serum albumin (MEM) in 48-well tissue culture plates (Costar, Cambridge, MA) in absence or presence of 5% human AB serum (Life Technologies). After incubation for prescribed periods at 37°C and 5% CO2, and stimulation with viable or nonviable C. albicans, supernatants were collected, centrifuged, and stored at −80°C until cytokine measurements. The endotoxin content of MEM, human serum albumin, and human serum was assayed by the Limulus lysate assay and always ranged below the detection limit of 10 pg/ml.

Human recombinant SP-A (generously provided from K. Schäfer, Byk Gulden Pharmazeutica, Konstanz, Germany) or canine SP-A from silica-treated dogs (a gift from U. Pison, Department of Anesthesiology and Intensive Care Medicine, Humboldt University, Berlin, Germany) were used in concentrations of 0.01, 0.1, 1, 10 and 100 μg/ml during the coincubation of mononuclear phagocytes with C. albicans. In selected experiments, leukocytes or yeast were preincubated with 10 μg/ml SP-A and washed twice in warm MEM to remove nonbound SP-A before starting the coincubation period.

C1q is homologous to SP-A with respect to its macromolecular organization, and both proteins are composed of a collagen-like triple helical domain that binds to the C1qR on monocytes and macrophages and show immunologic homology with type IV collagen (27). Against this background, human recombinant C1q and bovine type IV collagen (Sigma) were employed in concentrations of 0.01, 0.1, 1, and 10 μg/ml to examine whether C1q substituted for SP-A in its impact on cytokine production by mononuclear phagocytes and whether the collagenous domain was involved. To assess the role of C1qR in the modulation of cytokine synthesis, the receptor ligands C1q and SP-A were coated to 48-well tissue culture plates (10 μg/well in 100 μl of 0.1 M carbonate buffer for 3 h at 37°C). Coating of culture plates was verified by ELISA technique using a monoclonal mouse anti-human SP-A (PE10; generously provided by T. Akino, Department of Biochemistry, Sapporo Medical College, Sapporo, Japan) or a polyclonal goat anti-human C1q Ab (Sigma), HRP-coupled goat anti-mouse or rat anti-goat Ig Abs (Dianova, Hamburg, Germany) and the substrate ABTS (Dianova). Monocytes and AM were allowed to adhere to coated wells for 30 min at 37°C to cluster C1qR at the basal surface of the cells (10). C1qR clustering was assumed since a C1q-binding assay, using biotinylated human C1q, streptavidin-coupled HRP (Dianova), and the substrate ABTS demonstrated a decrease of C1q binding to phagocytes that previously adhered to C1q- or SP-A-coated wells in comparison to cells adhered to native tissue culture plates or albumin-coated wells (data not presented). After the adherence step, mononuclear phagocytes were stimulated with C. albicans in absence or presence of 10 μg/ml soluble SP-A for 24 h at 37°C, and supernatants were collected and processed as described above.

The effect of surfactant lipids as well as the impact of SP-A-assembly with these lipids on the profile of mononuclear phagocyte cytokine synthesis was investigated by use of the natural bovine surfactant Alveofact, which is composed of phospholipids, cholesterin, glycerides, the hydrophobic surfactant proteins SP-B and SP-C, and free fatty acids (Dr. Karl Thomae GmbH, Biberach, Germany). Reassembly of SP-A-lipid complexes was generated by incubation of 10 μg SP-A with 200 μg Alveofact for 1 h at 37°C under continuous rotation as described (28). The endotoxin content of buffers and MEM was routinely assayed by the Limulus amebocyte lysate assay and always ranged below the detection limit of 10 pg/ml. SP-A, C1q, and type IV collagen were routinely treated with polymyxin B (Pierce, Rockford, Illinois) to reduce significant endotoxin contamination (>1 μg/mg protein). The endotoxin level in the SP-A, C1q, and type IV collagen preparations after polymyxin B treatment always ranged below 10 pg/mg protein.

TNF-α, IL-1β, IL-6, IL-8, MCP-1, and IL-1Ra in cell culture supernatants were measured by ELISA technique. Maxisorp microtiter plates (Nunc, Wiesbaden, Germany) were coated overnight at 4°C with polyclonal goat Abs to human TNF-α, IL-1β, IL-6, IL-8, MIP-1α, MCP-1, or IL-1Ra (R&D, Abingdon, U.K.) followed by three washing steps with PBS containing 0.05% Tween 20 (Sigma). Fifty-microliter samples of culture supernatant were dispensed into the wells and incubated for 2 h at room temperature. After washing, application of a monoclonal mouse Ab directed against TNF-α, IL-1β, IL-6, IL-8, MIP-1α, MCP-1, or IL-1Ra (R&D) was followed by sequential incubation with a biotinylated donkey anti-mouse Ig Ab, avidin, and biotinylated HRP, and the substrate ABTS (Dianova). Serial dilutions of human recombinant cytokines (R&D) provided a standard curve for each individual ELISA. Plates were read at 405 nm with an ELISA photometer. Quantification of each cytokine was performed in triplicate with detection ranges of 10–1000 pg/ml.

AM (2 × 106) were cocultured with 5 × 107C. albicans in absence or presence of 0.1, 1, and 10 μg/ml SP-A in six-well tissue culture plates in 2 ml MEM at 37°C. After 2 h of coincubation, supernatants were removed, and total cellular RNA was isolated using the acid guanidinium thiocyanate-phenol-chloroform method as previously described (29). The constituent mRNA was reverse transcribed according to the instructions of the manufacturer (StrataScript RT-PCR kit; Stratagene, Heidelberg, Germany) in a final volume of 25 μl. The synthesis of complementary DNA was conducted in a GeneAmp PCR System 2400 (Perkin-Elmer, Norwalk, CA) for 50 min at 37°C, and enzyme inactivation was achieved by heating the reaction to 94°C for 7 min. Subsequently, the reaction mixture was diluted with RNase-free water to 60 μl and stored at −85°C until use. The PCR was performed in 1 × PCR buffer (Perkin-Elmer), 1 mM of each dNTP (dATP, dCTP, dGTP, and dTTP), 1 μM of intron-spanning specific primers (β-actin: 5′-AAAGAACCTGTACGCCAACACAGTGCTGTCT-3′, 5′-CGTCATACTCCTGCTTGCTGATCCACATCTG-3′; TNF-α: 5′-CGGGACGTGGAGCTGGCCGAGGAG-3′, 5′-CACCAGCTGGTTATCTCAGCTC-3′; Stratagene), 0.75 U AmpliTaq DNA Polymerase (Perkin-Elmer), and 2 μl of first strand cDNA in a total volume of 25 μl. PCR profiles consisted of initial denaturation at 94°C (1.5 min), followed by 25 (β-actin) or 35 (TNF-α) cycles of denaturation (94°C, 50 s), primer annealing (60°C, 60 s), and primer extension (72°C, 60 s) in a GeneAmp PCR System 2400. The final extension was performed at 72°C for 7 min. Aliquots of PCR products were electrophoresed through 1.8% (w/v) Nusieve/agarose gels stained with ethidium bromide for ∼2 h at 75 V. Negative controls were routinely performed by running PCR without cDNA-template to exclude false positive amplification products. Positive controls were performed using cDNA preparations obtained from LPS-stimulated AM. To verify the specificity of PCR amplifications obtained from the above mentioned procedure, automated DNA sequencing was conducted on the purified cDNA samples according to the instructions of the manufacturer (model 373 A; Applied Biosystems, Darmstadt, Germany). By comparing the resulting DNA sequences with the corresponding published sequences, we identified PCR products as expected segments of spliced TNF-α or β-actin mRNA species. With PCR conditions optimized for primer and magnesium concentrations and cycle numbers, amplification of cDNA samples was verified to be in the exponential phase of PCR by comparing the amount of input RNA equivalents with the yield of the TNF-α and β-actin PCR products.

C. albicans cell wall oligosaccharides have been previously reported to induce proinflammatory cytokine synthesis in macrophages (21). In the present study we could show that viable Candida hyphae as well as nonviable yeast dose-dependently induced the release of TNF-α from monocytes and AM (Fig. 1).

FIGURE 1.

Dose-dependency of C. albicans-induced TNF-α secretion in AM (upper panel) and monocytes (lower panel). AM or monocytes (5 × 105) were incubated with 1 × 104-1 × 108 nonviable Candida yeast (AM n = 5, monocytes n = 10) or 1 × 104-1 × 106 viable Candida hyphae (AM n = 5, monocytes n = 8) for 24 h at 37°C in 1 ml MEM. TNF-α was measured by ELISA, and values are expressed as multiple of the 24-h baseline TNF-α release from 5 × 105 cells calculated from each individual experiment. Mean values ± SEM are given. The baseline TNF-α release was 278 ± 156 pg/ml for AM and 183 ± 112 pg/ml for monocytes in these experiments.

FIGURE 1.

Dose-dependency of C. albicans-induced TNF-α secretion in AM (upper panel) and monocytes (lower panel). AM or monocytes (5 × 105) were incubated with 1 × 104-1 × 108 nonviable Candida yeast (AM n = 5, monocytes n = 10) or 1 × 104-1 × 106 viable Candida hyphae (AM n = 5, monocytes n = 8) for 24 h at 37°C in 1 ml MEM. TNF-α was measured by ELISA, and values are expressed as multiple of the 24-h baseline TNF-α release from 5 × 105 cells calculated from each individual experiment. Mean values ± SEM are given. The baseline TNF-α release was 278 ± 156 pg/ml for AM and 183 ± 112 pg/ml for monocytes in these experiments.

Close modal

Whereas the baseline TNF-α production in monocytes and AM was not affected by LPS-decontaminated human recombinant (Fig. 2) or natural canine SP-A, the augmented TNF-α secretion in response to C. albicans-challenge was dose-dependently suppressed by human recombinant and natural canine SP-A (Fig. 3). Maximal reduction of Candida-stimulated TNF-α release was achieved by 10 μg/ml human recombinant SP-A, which decreased AM TNF-α secretion by ∼60% and nearly completely inhibited monocyte TNF-α production (Fig. 3). The natural canine SP-A isolated from silica-treated dogs by butanol extraction was less active. A total of 100 μg/ml of the canine SPA preparation reduced the AM and monocyte TNF-α release to approximately the same levels achieved with 10 μg/ml human SP-A (Fig. 3).

FIGURE 2.

Lack of influence of recombinant human SP-A on the baseline secretion of TNF-α by AM (upper panel; n = 8) and monocytes (lower panel; n = 10). AM or monocytes (5 × 105) were incubated with LPS-decontaminated SP-A for 24 h at 37°C in 1 ml MEM. TNF-α release is expressed as multiple of the 24-h baseline TNF-α liberation (mean values ± SEM) calculated from each individual experiment. The baseline TNF-α release was 425 ± 275 pg/ml for AM and 114 ± 63 pg/ml for monocytes in these experiments.

FIGURE 2.

Lack of influence of recombinant human SP-A on the baseline secretion of TNF-α by AM (upper panel; n = 8) and monocytes (lower panel; n = 10). AM or monocytes (5 × 105) were incubated with LPS-decontaminated SP-A for 24 h at 37°C in 1 ml MEM. TNF-α release is expressed as multiple of the 24-h baseline TNF-α liberation (mean values ± SEM) calculated from each individual experiment. The baseline TNF-α release was 425 ± 275 pg/ml for AM and 114 ± 63 pg/ml for monocytes in these experiments.

Close modal
FIGURE 3.

Dose-dependent suppression of C. albicans-induced TNF-α release in AM and monocytes by recombinant human and natural canine SP-A. AM (5 × 105) (upper panel; human SP-A n = 8, canine SP-A n = 3) or monocytes (5 × 105) (lower panel; human SP-A n = 10, canine SP-A n = 3) were incubated with 1 × 107 nonviable C. albicans yeast in absence or presence of SP-A (24 h, 37°C, total volume 1 ml MEM). TNF-α was measured by ELISA, and data are given as percentage of control (absence of SP-A; mean values ± SEM) calculated from each individual experiment. The baseline TNF-α release was 496 ± 222 pg/ml for AM and 144 ± 105 pg/ml for monocytes in these experiments, Candida-stimulated TNF-α secretion was 1871 ± 266 pg/ml for AM and 1723 ± 318 pg/ml for monocytes. ∗, p < 0.01; ∗∗, p < 0.001, compared with control.

FIGURE 3.

Dose-dependent suppression of C. albicans-induced TNF-α release in AM and monocytes by recombinant human and natural canine SP-A. AM (5 × 105) (upper panel; human SP-A n = 8, canine SP-A n = 3) or monocytes (5 × 105) (lower panel; human SP-A n = 10, canine SP-A n = 3) were incubated with 1 × 107 nonviable C. albicans yeast in absence or presence of SP-A (24 h, 37°C, total volume 1 ml MEM). TNF-α was measured by ELISA, and data are given as percentage of control (absence of SP-A; mean values ± SEM) calculated from each individual experiment. The baseline TNF-α release was 496 ± 222 pg/ml for AM and 144 ± 105 pg/ml for monocytes in these experiments, Candida-stimulated TNF-α secretion was 1871 ± 266 pg/ml for AM and 1723 ± 318 pg/ml for monocytes. ∗, p < 0.01; ∗∗, p < 0.001, compared with control.

Close modal

In addition to TNF-α, we analyzed the impact of SP-A on the production of the proinflammatory cytokines IL-1β, IL-8, MIP-1α, and MCP-1, as well as IL-1Ra and IL-6, all known to be produced by AM during the acute inflammatory response in the alveolar space (Ref. 30 and unpublished data). SP-A did not influence the baseline production of these cytokines, but nearly totally inhibited the Candida-elicited secretion of IL-1β and MCP-1, and significantly decreased the release of IL-8 and MIP-1α in response to the fungal agent. In contrast, no significant inhibition of the Candida-induced secretion of the antiinflammatory cytokines IL-6 and IL-1Ra was noted (Fig. 4).

FIGURE 4.

Impact of SP-A on C. albicans-stimulated liberation of IL-1β, IL-6, IL-8, MIP-1α, IL-1Ra, and MCP-1 from AM. AM (5 × 105) were incubated with 1 × 107 C. albicans for 24 h in absence or presence of 10 μg/ml human SP-A. Cytokines were measured by ELISA, and mean values ± SEM are given as percentage of respective controls (cytokine secretion of yeast-stimulated AM in absence of SP-A) calculated from each individual experiment. Baseline secretion of 5 × 105 AM within 24 h was 286 ± 179 pg/ml (n = 10) for IL-1β, 2005 ± 602 pg/ml (n = 9) for IL 8, 89 ± 33 pg/ml (n = 9) for MIP-α, 91 ± 45 pg/ml (n = 10) for MCP-1, 690 ± 301 pg/ml (n = 10) for IL-6, and 438 ± 212 pg/ml (n = 9) for IL-1Ra, respectively. SP-A did not influence baseline secretion of these cytokines. C. albicans significantly stimulated the release of IL-1β (7120 ± 1005; n = 10), IL-8 (17,953 ± 6127 pg/ml; n = 9), MIP-1α (1781 ± 335 pg/ml; n = 9), and MCP-1 (896 ± 264 pg/ml; n = 10), as well as of IL-6 (15,420 ± 4325 pg/ml; n = 10) and IL-1Ra (1676 ± 381 pg/ml; n = 9). ∗, p < 0.01; ∗∗, p < 0.001, compared with respective control.

FIGURE 4.

Impact of SP-A on C. albicans-stimulated liberation of IL-1β, IL-6, IL-8, MIP-1α, IL-1Ra, and MCP-1 from AM. AM (5 × 105) were incubated with 1 × 107 C. albicans for 24 h in absence or presence of 10 μg/ml human SP-A. Cytokines were measured by ELISA, and mean values ± SEM are given as percentage of respective controls (cytokine secretion of yeast-stimulated AM in absence of SP-A) calculated from each individual experiment. Baseline secretion of 5 × 105 AM within 24 h was 286 ± 179 pg/ml (n = 10) for IL-1β, 2005 ± 602 pg/ml (n = 9) for IL 8, 89 ± 33 pg/ml (n = 9) for MIP-α, 91 ± 45 pg/ml (n = 10) for MCP-1, 690 ± 301 pg/ml (n = 10) for IL-6, and 438 ± 212 pg/ml (n = 9) for IL-1Ra, respectively. SP-A did not influence baseline secretion of these cytokines. C. albicans significantly stimulated the release of IL-1β (7120 ± 1005; n = 10), IL-8 (17,953 ± 6127 pg/ml; n = 9), MIP-1α (1781 ± 335 pg/ml; n = 9), and MCP-1 (896 ± 264 pg/ml; n = 10), as well as of IL-6 (15,420 ± 4325 pg/ml; n = 10) and IL-1Ra (1676 ± 381 pg/ml; n = 9). ∗, p < 0.01; ∗∗, p < 0.001, compared with respective control.

Close modal

In one set of experiments, we analyzed whether SP-A directly acts on macrophages or prevented association of fungal mannan and glucan with corresponding receptors on AM and monocytes by shielding these Candida cell wall oligosaccharides with its lectin domain. Although effective calcium-dependent SP-A-binding to Candida blastoconidia is known to occur (verified by flow cytometric analysis of FITC-SP-A binding to C. albicans yeast; Ref. 26), precoating of yeast with SP-A failed to attenuate the Candida-elicited cytokine response of AM (Fig. 5) and monocytes. Preincubation of AM or monocytes with SP-A for 30–60 min and removing unbound SP-A before admixing Candida did not suffice to inhibit TNF-α release, but preincubation for 120–240 min significantly inhibited cytokine secretion (Fig. 5), which demonstrated that SP-A directly acted on the mononuclear phagocytes. Culturing AM or monocytes for another 24 h in MEM after preincubation with SP-A for 240 min and removing unbound protein, however, completely restored the cytokine response to C. albicans (Fig. 5).

FIGURE 5.

Upper panel, Time course of Candida-stimulated AM TNF-α secretion in absence (n = 4; Candida) or presence (Candida + SP-A; n = 4) of 10 μg/ml human recombinant SP-A. AM (5 × 105) were incubated with 1 × 107 nonviable C. albicans yeast for various time periods. TNF-α release is expressed as multiple of the respective 0- to 2-h baseline TNF-α liberation (mean values ± SEM) calculated from each individual experiment. The 24-h baseline TNF-α release of AM was 255 ± 204 pg/ml in these experiments. Lower panel, TNF-α secretion of 5 × 105 AM after preincubation with 10 μg/ml human recombinant SP-A for 1, 2, or 4 h at 37°C (−1 h, −2 h, −4 h; each n = 3). Unbound SP-A was removed before starting the coincubation with 1 × 107 nonviable C. albicans yeast. In one set of experiments, AM were preincubated with 10 μg/ml SP-A, washed, and cultured for 24 h in MEM before commencing yeast admixture (−4 h/+24 h, n = 3). Alternatively, 1 × 107 nonviable C. albicans were preincubated with 10 μg/ml SP-A for 30 min, and nonbound SP-A was removed before adding yeast to 5 × 105 AM for 24 h (SP-A/Candida; n = 5). Mean values ± SEM are given as percentage of control (AM coincubated with yeast in absence of SP-A; 1621 ± 335 pg/ml) calculated from each individual experiment. The baseline TNF-α release of AM was 517 ± 278 pg/ml. ∗, p < 0.01; ∗∗, p < 0.001, compared with control.

FIGURE 5.

Upper panel, Time course of Candida-stimulated AM TNF-α secretion in absence (n = 4; Candida) or presence (Candida + SP-A; n = 4) of 10 μg/ml human recombinant SP-A. AM (5 × 105) were incubated with 1 × 107 nonviable C. albicans yeast for various time periods. TNF-α release is expressed as multiple of the respective 0- to 2-h baseline TNF-α liberation (mean values ± SEM) calculated from each individual experiment. The 24-h baseline TNF-α release of AM was 255 ± 204 pg/ml in these experiments. Lower panel, TNF-α secretion of 5 × 105 AM after preincubation with 10 μg/ml human recombinant SP-A for 1, 2, or 4 h at 37°C (−1 h, −2 h, −4 h; each n = 3). Unbound SP-A was removed before starting the coincubation with 1 × 107 nonviable C. albicans yeast. In one set of experiments, AM were preincubated with 10 μg/ml SP-A, washed, and cultured for 24 h in MEM before commencing yeast admixture (−4 h/+24 h, n = 3). Alternatively, 1 × 107 nonviable C. albicans were preincubated with 10 μg/ml SP-A for 30 min, and nonbound SP-A was removed before adding yeast to 5 × 105 AM for 24 h (SP-A/Candida; n = 5). Mean values ± SEM are given as percentage of control (AM coincubated with yeast in absence of SP-A; 1621 ± 335 pg/ml) calculated from each individual experiment. The baseline TNF-α release of AM was 517 ± 278 pg/ml. ∗, p < 0.01; ∗∗, p < 0.001, compared with control.

Close modal

SP-A admixture 30–120 min, but not 240 min, after starting C. albicans exposure of AM and monocytes also attenuated the TNF-α release (Fig. 6). This time course did suggest a transcriptional modulation of TNF-α synthesis in the presence of SP-A. Indeed, SP-A significantly and dose-dependently down-regulated TNF-α gene expression in Candida-stimulated AM (Fig. 6).

FIGURE 6.

Upper panel, Efficacy of human SP-A added to AM 30–240 min after Candida exposure. AM (5 × 105) were incubated with 1 × 107 nonviable C. albicans yeast for 24 h. SP-A (10 μg/ml) was admixed at the beginning of coincubation (0) or 30, 60, 120, or 240 min after starting AM-Candida coincubation (each n = 4). TNF-α values (mean ± SEM) are given as percentage of control (AM coincubated with yeast for 24 h in absence of SP-A; 1354 ± 298 pg/ml) calculated from each individual experiment. Baseline TNF-α secretion of AM was 309 ± 216 pg/ml in these experiments. ∗, p < 0.01; ∗∗, p < 0.001, compared with control. Lower panel, SP-A down-regulation of TNF-α gene expression in AM. AM (2 × 106) were sham-incubated or incubated with 5 × 107 nonviable C. albicans yeast in absence or presence of 0.1, 1, and 10 μg/ml human SP-A for 2 h, and cellular RNA was forwarded to PCR as described in Materials and Methods (each n = 3). As obvious from this representative experiment, the TNF-α message is dose-dependently down-regulated as compared with the β-actin mRNA.

FIGURE 6.

Upper panel, Efficacy of human SP-A added to AM 30–240 min after Candida exposure. AM (5 × 105) were incubated with 1 × 107 nonviable C. albicans yeast for 24 h. SP-A (10 μg/ml) was admixed at the beginning of coincubation (0) or 30, 60, 120, or 240 min after starting AM-Candida coincubation (each n = 4). TNF-α values (mean ± SEM) are given as percentage of control (AM coincubated with yeast for 24 h in absence of SP-A; 1354 ± 298 pg/ml) calculated from each individual experiment. Baseline TNF-α secretion of AM was 309 ± 216 pg/ml in these experiments. ∗, p < 0.01; ∗∗, p < 0.001, compared with control. Lower panel, SP-A down-regulation of TNF-α gene expression in AM. AM (2 × 106) were sham-incubated or incubated with 5 × 107 nonviable C. albicans yeast in absence or presence of 0.1, 1, and 10 μg/ml human SP-A for 2 h, and cellular RNA was forwarded to PCR as described in Materials and Methods (each n = 3). As obvious from this representative experiment, the TNF-α message is dose-dependently down-regulated as compared with the β-actin mRNA.

Close modal

The size of Candida hyphae prevents phagocytosis by macrophages and monocytes, and hyphae contain much more cell wall oligosaccharides than blastoconidia, which caused the release of large amounts of TNF-α in mononuclear phagocytes (Ref. 21 and Fig. 1). This is important in vivo, since blastoconidia disseminate into the lungs during Candida septicemia and undergo hyphal transition in the alveolar space coincident with an overwhelming, compartmentalized TNF-α production (22). We could show that SP-A not only diminished mononuclear phagocyte TNF-α release induced by nonviable Candida yeast (Figs. 3 and 7), but also inhibited cytokine secretion stimulated by viable hyphae (Fig. 7), which developed from viable blastoconidia during the 24 h coculture period of fungi and leukocytes.

FIGURE 7.

Influence of serum components and SP-A on TNF-α generation in AM (upper panel,n = 5) and monocytes (lower panel,n = 7) in response to viable Candida hyphae and nonviable Candida yeast. Phagocytes (5 × 105) were incubated with 1 × 107 nonviable or 1 × 105 viable C. albicans for 24 h, in absence or presence of 5% human AB serum, with or without additional admixture of 10 μg/ml human SP-A. Viable yeast underwent hyphal transition during the incubation at 37°C in MEM. TNF-α values are expressed as multiple (mean ± SEM) of the 24-h baseline TNF-α release from 5 × 105 cells calculated from each individual experiment. Baseline TNF-α secretion was 445 ± 186 pg/ml in absence and 283 ± 141 pg/ml in presence of serum for AM, and 122 ± 61 pg/ml in absence and 263 ± 206 pg/ml in the presence of serum for monocytes, respectively.

FIGURE 7.

Influence of serum components and SP-A on TNF-α generation in AM (upper panel,n = 5) and monocytes (lower panel,n = 7) in response to viable Candida hyphae and nonviable Candida yeast. Phagocytes (5 × 105) were incubated with 1 × 107 nonviable or 1 × 105 viable C. albicans for 24 h, in absence or presence of 5% human AB serum, with or without additional admixture of 10 μg/ml human SP-A. Viable yeast underwent hyphal transition during the incubation at 37°C in MEM. TNF-α values are expressed as multiple (mean ± SEM) of the 24-h baseline TNF-α release from 5 × 105 cells calculated from each individual experiment. Baseline TNF-α secretion was 445 ± 186 pg/ml in absence and 283 ± 141 pg/ml in presence of serum for AM, and 122 ± 61 pg/ml in absence and 263 ± 206 pg/ml in the presence of serum for monocytes, respectively.

Close modal

During severe pulmonary inflammation with capillary leakage (e.g., acute respiratory distress syndrome), the alveolar space becomes flooded with plasma proteins, and plasma proteins like LPS-binding protein or MBP (31) have been shown to modulate the cytokine response of mononuclear phagocytes. Plasma proteins have also been reported to inhibit biophysical properties of surfactant components, and SP-A counteracts these inhibitory effects of plasma constituents (32). Admixture of serum components (5% v/v) to the phagocyte-yeast coincubation dramatically increased the Candida-induced cytokine secretion of AM and monocytes (Fig. 7); SP-A, however, also abrogated this serum-enhanced Candida-elicited TNF-α response in both cell types (Fig. 7).

The structural homologous proteins C1q and type IV collagen share the N-terminal collagenous region with SP-A but lack the C-terminal carbohydrate recognition domain. Like SP-A, both proteins bind to C1qR with their collagenous domain, but, in contrast to the surfactant protein, they failed to inhibit the Candida-induced cytokine response in monocytes and AM. LPS-decontaminated type IV collagen itself stimulated TNF-α secretion in both cell types and dose-dependently augmented the Candida-induced cytokine release (Fig. 8). Polymyxin B-treated C1q did not influence the baseline secretion of TNF-α, but slightly increased the Candida-stimulated cytokine production at a concentration of 10 μg/ml (Fig. 8).

FIGURE 8.

Influence of C1q and type IV collagen on baseline (upper panel) and C. albicans-induced (lower panel) TNF-α release from AM. AM (5 × 105) were incubated with C1q or type IV collagen in absence or presence of 1 × 107 nonviable C. albicans yeast for 24 h at 37°C in 1 ml MEM (each n = 6). Mean values ± SEM are given as multiple of the 24-h baseline TNF-α release from 5 × 105 AM (upper panel) or as percentage of control (AM coincubated with yeast in absence of SP-A, C1q, or type IV collagen; lower panel) calculated from each individual experiment. Baseline TNF-α release within 24 h was 261 ± 114 pg/ml, and Candida-stimulated TNF-α secretion was 1233 ± 303 pg/ml. Error bars are missing when falling into symbol. ∗, p < 0.01; ∗∗, p < 0.001, compared with baseline release (upper panel) or control (lower panel).

FIGURE 8.

Influence of C1q and type IV collagen on baseline (upper panel) and C. albicans-induced (lower panel) TNF-α release from AM. AM (5 × 105) were incubated with C1q or type IV collagen in absence or presence of 1 × 107 nonviable C. albicans yeast for 24 h at 37°C in 1 ml MEM (each n = 6). Mean values ± SEM are given as multiple of the 24-h baseline TNF-α release from 5 × 105 AM (upper panel) or as percentage of control (AM coincubated with yeast in absence of SP-A, C1q, or type IV collagen; lower panel) calculated from each individual experiment. Baseline TNF-α release within 24 h was 261 ± 114 pg/ml, and Candida-stimulated TNF-α secretion was 1233 ± 303 pg/ml. Error bars are missing when falling into symbol. ∗, p < 0.01; ∗∗, p < 0.001, compared with baseline release (upper panel) or control (lower panel).

Close modal

Adherence of AM to SP-A-coated surfaces did not influence the Candida-elicited TNF-α secretion (Fig. 9). Soluble SP-A, in contrast, maintained its inhibitory capacity even under conditions of AM adherence to SP-A- or C1q-coated tissue culture plates (Fig. 9), which clustered C1qR at the basal surface of adherent cells. Altogether, these data suggested that binding of the collagenous domain of SP-A to C1qR was not involved in reducing TNF-α release from Candida-stimulated monocytes or AM.

FIGURE 9.

Impact of SP-A- and C1q-coated tissue culture plates on C. albicans-induced TNF-α release from AM. Human SP-A (10 μg/ml; n = 7) or C1q (10 μg/ml; n = 7) were coated to tissue culture plates, and 5 × 105 AM were allowed to adhere to coated wells for 30 min at 37°C. Adherence to C1q- or SP-A-coated wells occurs preferentially via C1qR and induces clustering of C1qR at the basal surface of the cells. TNF-α release was induced by the addition of 1 × 107 nonviable C. albicans yeast for 24 h at 37°C in absence (n = 7) or presence of 10 μg/ml soluble SP-A (n = 7). Mean values ± SEM are given as percentage of control (AM adhered to native tissue culture plastic and coincubated with yeast in absence of soluble SP-A; 1173 ± 224 pg/ml, n = 7) calculated from each individual experiment. TNF-α baseline release within 24 h was 372 ± 302 pg/ml (AM adherent to plastic, n = 7). Baseline secretion of AM adhered to SP-A- (401 ± 236 pg/ml, n = 7) or C1q-coated (349 ± 198 pg/ml, n = 7) surfaces did not significantly differ from baseline secretion of cells adhered to native tissue culture plastic.

FIGURE 9.

Impact of SP-A- and C1q-coated tissue culture plates on C. albicans-induced TNF-α release from AM. Human SP-A (10 μg/ml; n = 7) or C1q (10 μg/ml; n = 7) were coated to tissue culture plates, and 5 × 105 AM were allowed to adhere to coated wells for 30 min at 37°C. Adherence to C1q- or SP-A-coated wells occurs preferentially via C1qR and induces clustering of C1qR at the basal surface of the cells. TNF-α release was induced by the addition of 1 × 107 nonviable C. albicans yeast for 24 h at 37°C in absence (n = 7) or presence of 10 μg/ml soluble SP-A (n = 7). Mean values ± SEM are given as percentage of control (AM adhered to native tissue culture plastic and coincubated with yeast in absence of soluble SP-A; 1173 ± 224 pg/ml, n = 7) calculated from each individual experiment. TNF-α baseline release within 24 h was 372 ± 302 pg/ml (AM adherent to plastic, n = 7). Baseline secretion of AM adhered to SP-A- (401 ± 236 pg/ml, n = 7) or C1q-coated (349 ± 198 pg/ml, n = 7) surfaces did not significantly differ from baseline secretion of cells adhered to native tissue culture plastic.

Close modal

Finally, we examined whether SP-A retained its capacity to modulate cytokine synthesis in the presence of surfactant lipids, because the vast majority of SP-A recovered in lung lavage is associated with lipids, and the lipid fraction has been described to counteract certain immune functions of SP-A (19, 20). Alveofact (which contains surfactant lipids and the hydrophobic surfactant apoproteins SP-B and SP-C) exhibited a moderate inhibitory effect on the Candida-stimulated cytokine production by AM (Fig. 10). Despite SP-A assembly with surfactant lipids, the surfactant protein displayed a strong inhibitory capacity, surpassing that of surfactant lipids and even slightly (not significantly) surpassing that of SP-A alone, rather suggesting that surfactant lipids and SP-A act in an additive manner (Fig. 10).

FIGURE 10.

Influence of natural surfactant (Alveofact) and SP-A-surfactant reassembly (Alveofact + SP-A) on C. albicans-induced TNF-α release from AM. AM (5 × 105) were incubated with 1 × 107 nonviable C. albicans yeast for 24 h in absence (n = 8) or presence of 200 μg/ml Alveofact (containing surfactant lipids and hydrophobic apoproteins; n = 7), 10 μg/ml human SP-A (n = 8), or 10 μg SP-A + 200 μg Alveofact/ml (n = 7). For reassembly of SP-A-lipid complexes, preincubation of SP-A with Alveofact was performed for 1 h at 37°C before experimental use. Mean values ± SEM are expressed as percentage of control (AM coincubated with yeast for 24 h in absence of SP-A or Alveofact; 1431 ± 352 pg/ml, n = 8) calculated from each individual experiment. TNF-α baseline release was 421 ± 355 pg/ml in absence of Alveofact and SP-A (n = 8), 454 ± 286 in presence of SP-A (n = 8), 382 ± 172 in presence of Alveofact (n = 7), and 402 ± 313 in the presence of SP-A and Alveofact (n = 7) without significant difference. ∗, p < 0.01; ∗∗, p < 0.001, compared with control.

FIGURE 10.

Influence of natural surfactant (Alveofact) and SP-A-surfactant reassembly (Alveofact + SP-A) on C. albicans-induced TNF-α release from AM. AM (5 × 105) were incubated with 1 × 107 nonviable C. albicans yeast for 24 h in absence (n = 8) or presence of 200 μg/ml Alveofact (containing surfactant lipids and hydrophobic apoproteins; n = 7), 10 μg/ml human SP-A (n = 8), or 10 μg SP-A + 200 μg Alveofact/ml (n = 7). For reassembly of SP-A-lipid complexes, preincubation of SP-A with Alveofact was performed for 1 h at 37°C before experimental use. Mean values ± SEM are expressed as percentage of control (AM coincubated with yeast for 24 h in absence of SP-A or Alveofact; 1431 ± 352 pg/ml, n = 8) calculated from each individual experiment. TNF-α baseline release was 421 ± 355 pg/ml in absence of Alveofact and SP-A (n = 8), 454 ± 286 in presence of SP-A (n = 8), 382 ± 172 in presence of Alveofact (n = 7), and 402 ± 313 in the presence of SP-A and Alveofact (n = 7) without significant difference. ∗, p < 0.01; ∗∗, p < 0.001, compared with control.

Close modal

Several studies provided evidence that surfactant may have major impact on lung immune cell function (reviewed in Ref. 33), including cell proliferation (17, 19), cytokine production (17, 18, 19, 20, 34, 35, 36, 37), and expression of surface molecules on leukocytes (38, 39). Down-regulation of immune cell function has predominantly been ascribed to the lipid moieties in the natural surfactant material (16, 19, 20, 34, 36, 39). In contrast, the hydrophilic surfactant apoproteins were shown to activate alveolar host defense by stimulating phagocytosis, oxidative burst, and chemotaxis of macrophages, monocytes, and granulocytes (7, 11, 12, 13, 40, 41, 42). Accordingly, enhanced mononuclear cell cytokine production, lymphocyte proliferation, and Ig secretion were noted in the presence of SP-A (19, 20). In AM, however, this apoprotein was also demonstrated to inhibit TNF-α synthesis in response to endotoxin challenge (18). Interestingly, the SP-A homologue MBP also decreased the production of TNF-α from oligosaccharide-activated monocytes (31). Various cell types might respond differentially to SP-A to explain these different findings; however, the origin and chemical features of the SP-A batches employed in these studies must also be taken into consideration. Human alveolar proteinosis material, a common source of SP-A, differs from physiological surfactant (43). SP-A might be susceptible to inactivation or denaturation during a commonly employed butanol extraction procedure (44), and significant contamination with endotoxin might have additional impact. To address this issue, we currently employed two different SP-A preparations, one recombinant human protein, the functional integrity of which has been characterized in detail (43, 45), and one extracted from natural material (canine SP-A; Ref. 40). Corresponding results were obtained with these two SP-A preparations, with the only difference of lower spec. act. (related to weight) of the canine SP-A in the human cell types currently investigated. It has to be kept in mind, however, that we presently employed a LPS-decontamination step to minimize the endotoxin load of the SP-A preparations used, since considerable LPS-contamination of these SP-A preparations was noted in pilot experiments, alongside with marked induction of mononuclear phagocyte cytokine production under baseline conditions (data not shown).

The inhibitory effect on proinflammatory cytokine release from mononuclear phagocytes was specific for SP-A and was not reproduced by C1q and type IV collagen, which share the N-terminal collagenous region with SP-A, but lack the C-terminal carbohydrate recognition domain (6, 8, 9). These findings and the previous observation of calcium-dependent SP-A-binding to C. albicans (26) initially favored the hypothesis that the lectin domain of SP-A might “shield” fungal mannan and glucan moieties, thereby blocking their interaction with corresponding receptors on AM and monocytes (2, 31). Fungal cell wall polysaccharides have indeed been reported to provoke TNF-α synthesis in mononuclear phagocytes (21). However, precoating of yeast with SP-A in the absence of soluble SP-A completely failed to inhibit the cytokine release in response to fungal challenge. This finding is in line with the previous observation that SP-A binds only to Lipid A of rough LPS, but interferes with macrophage TNF-α secretion in response to smooth LPS (18, 46, 47). Altogether, these data strongly suggest that SP-A blocked the cytokine response of macrophages and monocytes to Candida challenge not by its binding to the fungal surface, but by some direct impact on the mononuclear phagocytes. This is supported by the finding that preincubation of AM with SP-A and removal of unbound SP-A before starting the AM-Candida coincubation sufficed to suppress the cytokine response to fungal challenge.

Moreover, internalization of SP-A by macrophages seemed apparently necessary for the interference with the cytokine response to Candida exposure, since SP-A immobilized on tissue culture plates was entirely ineffective. Multiple SP-A membrane receptors on monocytes and AM have been suggested, all of which might be operative for achieving SP-A internalization into these cells (48, 49). One of these is C1qR, which recognizes the collagenous region of SP-A and its homologues (49, 50). Employment of immobilized C1q for clustering C1qR at the basal surface of adhering mononuclear phagocytes and blocking C1qR by this technique (10, 48), did, however, not abrogate the inhibitory activity of soluble SP-A. Though further detailed analysis of the mechanisms of SP-A binding and internalization, underlying the suppressive effect on proinflammatory cytokine synthesis, was not undertaken in the current study, this finding supports the notion that the carbohydrate recognition domain of SP-A rather than its collagenous region is centrally involved in this effect.

Analysis of the time-dependency of the SP-A effect on the phagocyte TNF-α response, as well as direct assessment of the TNF-α gene expression in absence and presence of SP-A, indicated that the down-regulation of TNF-α synthesis by SP-A occurred at the transcriptional level. This observation is reminiscent of the previously reported effect of surfactant lipid mixtures, which were noted to reduce mononuclear cytokine mRNA via inhibition of the transcription factor NF-κB (36). Accordingly, the suppressive effect of SP-A was not exclusive for TNF-α, but was also demonstrated for the proinflammatory cytokines IL-1β, IL-8, MCP-1, and MIP-1α. In contrast, the Candida-elicited synthesis of the antiinflammatory cytokines IL-1Ra and IL-6 was not suppressed by SP-A, and SP-A did not affect the baseline levels of cytokine generation in AM and monocytes. These data indicate a distinct immunomodulatory rather than a general inhibitory effect of SP-A on mononuclear phagocyte cytokine responsiveness. The intracellular signaling events underlying this profile of SP-A efficacy in macrophages and monocytes clearly demand further elucidation.

The robustness of the SP-A effect on mononuclear phagocyte cytokine generation is demonstrated by the fact that it was noted for both viable and nonviable Candida as well as AM and monocytes, and that it was also operative in the presence of surfactant lipids and plasma proteins. The vast majority of SP-A recovered from lung lavage is associated with the lipid fraction of surfactant (51), and the lipids have been noted to counteract several effects demonstrated for purified SP-A (11, 19, 20). We could show that the capacity of SP-A to suppress the proinflammatory cytokine response of AM was retained in the presence of the natural surfactant material, including surfactant lipids and the hydrophobic surfactant apoproteins SP-B and SP-C, and this was true after a preceding maneuver to establish reassembly of SP-A-lipid complexes. Most impressively, SP-A also retained its inhibitory capacity under conditions mimicking plasma protein leakage into the alveolar compartment, a prominent feature of acute inflammatory lung disease. In the presence of serum constituents, the TNF-α secretory response to both viable and nonviable Candida was manyfold increased, possibly related to an opsonization effect, and this strong TNF-α response was markedly reduced in the presence of SP-A, to a level approaching that in the absence of serum. This finding is reminiscent of the previous observation that SP-A counteracts the inhibitory effect of plasma constituents on alveolar surfactant function (32).

In conclusion, the major surfactant protein SP-A was found to strongly suppress the proinflammatory cytokine response of AM and monocytes to both viable and nonviable C. albicans. In contrast, the baseline cytokine generation and the release of antiinflammatory cytokines upon fungal challenge were not affected. This SP-A effect was retained upon reassembly of the protein with natural surfactant lipids and in the presence of serum constituents, as mimicry of inflammatory conditions with plasma leakage into the alveolar space. Direct impact of the apoprotein on phagocyte regulatory events rather than its binding to fungal surfaces is suggested as the underlying mechanism. These data lend further credit to a physiological function of SP-A in regulating alveolar host defense and inflammation, by suggesting a fundamental role of this apoprotein in limiting excessive proinflammatory cytokine release in AM and invading monocytes confronted with microbial challenge in the alveolar compartment.

We thank R. Maus and G. Wahler for excellent technical assistance. We are grateful to U. Pison and K. P. Schäfer who supplied the SP-A preparations, and to T. Akino for the anti-SP-A Ab. We are also indebted to all volunteers who underwent bronchoscopy and BAL or donated blood for this study. This work includes parts of the thesis of Peter Hammerl.

1

This work was supported by Deutsche Forschungsgemeinschaft, Grant SFB 547.

3

Abbreviations used in this paper: SP-A, surfactant protein A; MBP, mannose-binding protein; AM, alveolar macrophage; MIP, macrophage inflammatory protein; MCP, monocyte chemoattractant protein; IL-lRa, IL-1R antagonist; BAL, bronchoalveolar lavage.

1
Haagsman, H. P., S. Hawgood, T. Sargeant, D. Buckley, R. T. White, K. Drickamer, B. J. Benson.
1987
. The major lung surfactant protein, SP 28–36, is a calcium-dependent, carbohydrate-binding protein.
J. Biol. Chem.
262
:
13877
2
Sastry, K., G. A. Herman, L. Day, E. G. Bruns, C. C. Morton, R. A. B. Ezekowitz.
1989
. The human mannose-binding protein gene: exon structure reveals its evolutionary relationship to a human pulmonary surfactant gene and localization to chromosome 10.
J. Exp. Med.
170
:
1175
3
Persson, A., D. Chang, K. Rust, M. Moxley, W. Langmore, E. Crouch.
1989
. Purification and biochemical characterization of CP4 (SP-D): a collagenous surfactant-associated protein.
Biochemistry
27
:
6361
4
Strang, C. J., H. S. Slayter, P. Lachman, A. E. Davis.
1986
. Ultrastructure and composition of bovine conglutinin.
Biochem. J.
234
:
381
5
Holmskov, U., B. Teisner, A. C. Willis, K. B. Reid, J. C. Jensenius.
1993
. Purification and characterization of a bovine serum lectin (CL-43) with structural homology to conglutinin and SP-D and carbohydrate specificity similar to mannan-binding protein.
J. Biol. Chem.
268
:
10120
6
Voss, T., H. Eistetter, K. P. Schäfer, J. Engel.
1988
. Macromolecular organization of natural and recombinant lung surfactant protein SP 28–36: structural homology with the complement factor C1q.
J. Mol. Biol.
201
:
219
7
Tenner, A. J., S. L. Robinson, J. Borchelt, J. R. Wright.
1989
. Human pulmonary surfactant protein (SP-A), a protein homologous to C1q, can enhance FcR- and CR1-mediated phagocytosis.
J. Biol. Chem.
264
:
13923
8
Hoppe, H. J., K. B. Reid.
1994
. Collectins—soluble proteins containing collagenous regions and lectin domains—and their roles in innate immunity.
Protein Sci.
3
:
1143
9
Holmskov, U., R. Malhotra, R. Sim, J. C. Jensenius.
1994
. Collectins: collagenous C-type lectins of the innate immune defense system.
Immunol. Today
15
:
67
10
Tino, M. J., J. R. Wright.
1996
. Surfactant protein A stimulates phagocytosis of specific pulmonary pathogens by alveolar macrophages.
Am. J. Physiol.
270
:
L677
11
Iwaarden, F., B. Welmers, J. Verhoef, H. P. Haagsman, L. M. G. van Golde.
1990
. Pulmonary surfactant protein A enhances the host-defense mechanism of rat alveolar macrophages.
Am. J. Respir. Cell. Mol. Biol.
2
:
91
12
Iwaarden, J. F., J. A. G. van Strijp, M. J. M. Ebskamp A. C. Welmers, J. Verhoef, L. M. G. van Golde.
1991
. Surfactant protein A is opsonin in phagocytosis of herpes simplex virus type 1 by rat alveolar macrophages.
Am. J. Physiol.
261
:
L204
13
Wright, J. R., D. C. Youmans.
1993
. Pulmonary surfactant protein A stimulates chemotaxis of alveolar macrophages.
Am. J. Physiol.
264
:
L338
14
Kaltreider, H. B., S. E. Salmon.
1973
. Functional properties of bronchoalveolar lymphocytes obtained from the normal canine lung.
J. Clin. Invest.
52
:
2211
15
Bartman, P., L. Gortner, F. Pohlandt, H. Jaeger.
1992
. In vitro lymphocyte functions in the presence of bovine surfactant and its phospholipid fractions.
J. Perinat. Med.
20
:
189
16
Geertsma, M. F., H. R. Broos, M. T. van den Barselaar, P. H. Nibbering, R. van Furth.
1993
. Lung surfactant suppresses oxygen-dependent bactericidal functions of human blood monocytes by inhibiting the assembly of the NADPH oxidase.
J. Immunol.
150
:
2391
17
Borron, P., R. A. W. Veldhuizen, J. F. Lewis, F. Possmayer, A. Caveney, K. Inchley, R. G. McFadden, L. J. Fraher.
1996
. Surfactant associated protein A inhibits lymphocyte proliferation and IL-2 production.
Am. J. Respir. Cell Mol. Biol.
15
:
115
18
McIntosh, J. C., S. Mervin-Blake, E. Conner, J. R. Wright.
1996
. Surfactant protein A protects growing cells and reduces TNF-α activity from LPS-stimulated macrophages.
Am. J. Physiol.
271
:
L310
19
Kremlev, S. G., T. M. Umstead, D. S. Phelps.
1994
. Effects of surfactant protein A and surfactant lipids on lymphocyte proliferation in vitro.
Am. J. Physiol.
267
:
L357
20
Kremlev, S. G., D. S. Phelps.
1994
. Surfactant protein A stimulation of inflammatory cytokine and immunoglobulin production.
Am. J. Physiol.
267
:
L712
21
Garner, R. E., K. Rubanowice, R. T. Sawyer, J. A. Hudson..
1994
. Secretion of TNF-α by alveolar macrophages in response to Candida albicans mannan.
J. Leukocyte Biol.
55
:
161
22
Lechner, A. J., T. L. Tredway, D. S. Brink, C. A. Klein, G. M. Matuschak.
1992
. Differential systemic and intrapulmonary TNF-α production in Candida sepsis during immunosuppression.
Am. J. Physiol.
263
:
L526
23
Goodman, R. B., R. M. Strieter, D. P. Martin, K. P. Steinberg, J. A. Milberg, R. J. Maunder, S. L. Kunkel, A. Walz, L. D. Hudson, T. R. Martin.
1996
. Inflammatory cytokines in patients with persistence of the Acute Respiratory Distress Syndrome.
Am. J. Respir. Crit. Care Med.
154
:
602
24
Aderka, D., J. Le, J. Vilcek.
1989
. IL-6 inhibits lipopolysaccharide-induced tumor necrosis factor production in cultured human monocytes, U937 cells, and in mice.
J. Immunol.
143
:
3517
25
Xing, Z., J. Gauldie, G. Cox, H. Baumann, M. Jordana, X. F. Lei, M. K. Achong.
1998
. IL-6 is an antiinflammatory cytokine required for controlling local or systemic acute inflammatory responses.
J. Clin. Invest.
101
:
311
26
Rosseau, S., A. Guenther, W. Seeger, J. Lohmeyer.
1997
. Phagocytosis of viable Candida albicans by alveolar macrophages: lack of opsonin function of surfactant protein A.
J. Infect. Dis.
175
:
421
27
Coulombe, P. A., M. Bendayan.
1989
. Lung surfactant-associated proteins and type IV collagen share common epitopes: an immunocytochemical demonstration.
Am. Rev. Respir. Dis.
140
:
1040
28
King, R. J., M. C. Carmichael, P. M. Horowitz.
1983
. Re-assembly of lipid-protein complexes of pulmonary surfactant.
J. Biol. Chem.
258
:
10672
29
Chomczynski, P., N. Sacchi.
1987
. Single-step method of RNA isolation by acid guanidinium thiocyanat-phenol-chloroform extraction.
Anal. Biochem.
162
:
156
30
Maus, U., S. Rosseau, U. Knies, W. Seeger, J. Lohmeyer.
1998
. Expression of pro-inflammatory cytokines by flow sorted alveolar macrophages in severe pneumonia.
Eur. Respir. J.
11
:
534
31
Soell, M., E. Lett, F. Holveck, M. Scholler, D. Wachsmann, J. P. Klein.
1995
. Activation of human monocytes by streptococcal rhamnose glucose polymers is mediated by CD14 antigen, and mannan binding protein inhibits TNF-α release.
J. Immunol.
154
:
851
32
Cockshutt, A. M., J. Weitz, F. Possmeyer.
1990
. Pulmonary surfactant-associated protein A enhances the surface activity of lipid extract surfactant and reverses inhibition by blood proteins in vitro.
Biochemistry
29
:
8424
33
Wright, J. R..
1997
. Immunomodulatory functions of surfactant.
Physiol. Rev.
77
:
931
34
Allen, J. N., S. A. Moore, A. L. Pope-Harman, C. B. Marsh, M. D. Wewers.
1995
. Immunosuppressive properties of surfactant and plasma on alveolar macrophages.
J. Lab. Clin. Med.
125
:
356
35
Blau, H., S. Riklis, V. Kravtsov, M. Kalina.
1994
. Secretion of cytokines by rat alveolar epithelial cells: possible regulatory role for SP-A.
Am. J. Physiol.
266
:
L148
36
Antal, J. M., L. T. Divis, S. C. Erzurum, H. P. Wiedemann, M. J. Thomassen.
1996
. Surfactant suppresses NFκB activation in human monocytic cells.
Am J. Respir. Cell Mol. Biol.
14
:
374
37
Speer, C. P., B. Gotze, T. Cursted, B. Robertson.
1991
. Phagocytic functions and tumor necrosis factor secretion of human monocytes exposed to natural porcine surfactant (Curosurf).
Pediatr. Res.
30
:
69
38
Kremlev, S. G., D. S. Phelps.
1997
. Effect of SP-A and surfactant lipids on expression of cell surface markers in the THP-1 monocytic cell line.
Am. J. Physiol.
272
:
L1070
39
Geertsma, M. F., W. L. Teeuw, P. H. Nibbering, R. van Furth.
1994
. Pulmonary surfactant inhibits activation of human moncytes by recombinant interferon-γ.
Immunology
82
:
450
40
Pison, U., M. Max, A. Neuendank, S. Weissbach, S. Pietschmann.
1994
. Host defense capacities of pulmonary surfactant: evidence for non-surfactant functions of the surfactant system.
Eur. J. Clin. Invest.
24
:
586
41
Hartshorn, K. L., E. Crouch, M. R. White, M. L. Colamussi, A. Kakkanatt, B. Tauber, V. Shepherd, K. N. Sastry.
1998
. Pulmonary surfactant proteins A and D enhance neutrophil uptake of bacteria.
Am. J. Physiol.
274
:
L958
42
Gaynor, C. D., F. X. McCormack, D. R. Voelker, S. E. McGowan, L. S. Schlesinger..
1995
. Pulmonary surfactant protein A enhanced phagocytosis of Mycobacterium tuberculosis by a direct interaction with human macrophages.
J. Immunol.
155
:
5343
43
Voss, T., K. Melchers, G. Scheirle, K. P. Schaefer.
1991
. Structural comparison of recombinant pulmonary surfactant protein SP-A derived from two human coding sequences: implications for the chain composition of natural human SP-A.
Am. J. Respir. Cell Mol. Biol.
4
:
88
44
Van Iwaarden, J. F., F. T. van Berkhout, J. A. Whitsett, R. S. Oosting, L. M. G. van Golde.
1995
. A novel procedure for the rapid isolation of surfactant protein A with retention of its alveolar macrophage-stimulating properties.
Biochem. J.
309
:
551
45
Manz-Keinke, H., H. Plattner, J. Schlepper-Schaefer.
1992
. Lung surfactant protein A enhances serum-independent phagocytosis of bacteria by alveolar macrophages.
Eur. J. Cell Biol.
57
:
95
46
Van Iwaarden, F., J. C. Pikaar, J. Storm, E. Brouwer, J. Verhoef, R. S. Oosting, L. M. G. van Golde, J. A. G. van Strijp.
1994
. Binding of surfactant protein A to the lipid moiety of bacterial lipopolysaccharides.
Biochem. J.
303
:
407
47
Pikaar, J. C., W. F. Voorhout, L. M. G. van Golde, J. Verhoef, J. A. G. van Strijp, F. van Iwaarden.
1995
. Opsonic activities of surfactant proteins A and D in phagocytosis of gramnegative bacteria by alveolar macrophages.
J. Infect. Dis.
172
:
481
48
Chroneos, Z. C., R. Abdolrasulnia, J. A. Whitsett, W. R. Rice, V. L. Shepherd.
1996
. Purification of a cell surface receptor for surfactant protein A.
J. Biol. Chem.
271
:
16375
49
Malhotra, R., S. Thiel, K. B. M. Reid, R. B. Sim.
1990
. Human leukocyte C1q receptor binds other soluble proteins with collagen domains.
J. Exp. Med.
172
:
955
50
Malhotra, R., J. Haurum, S. Thiel, R. B. Sim.
1992
. Interaction of C1q receptor with lung surfactant protein A.
Eur. J. Immunol.
22
:
1437
51
Baritussio, A., A. Alberti, D. Quaglino, A. Pettenazzo, D. Dalzoppo, L. Sartori, I. Pasquali-Ronchetti.
1994
. SP-A, SP-B, and SP-C in surfactant subtypes around birth: reexamination of alveolar life cycle of surfactant.
Am. J. Physiol.
266
:
L436