Surfactant protein D (SP-D) plays important roles in the initial innate defense against influenza A virus (IAV). The collagen domain of SP-D is probably critical for its homeostatic functions in vivo and has been implicated in the modulation of macrophage responses to SP-D-ligand complexes. For the current studies, we used a panel of rat SP-D mutants lacking all or part of the collagen domain to more specifically evaluate the contributions of this domain to viral interactions. SP-D multimers lacking the collagenous sequence efficiently neutralized Phil82 IAV, promoted neutrophil uptake of IAV, and also potentiated the IAV-induced neutrophil respiratory burst response. A dodecameric mutant with shortened collagenous arms showed enhanced viral aggregation and neuraminidase inhibition, and an increased capacity to inhibit a partially collectin-resistant strain of IAV. By contrast, truncated molecules lacking an N-terminal and collagen domain showed no detectable antiviral and opsonizing activity, despite preservation of lectin activity and detectable viral binding. Thus, multimerization, which is mediated by the N-peptide, is more important than the collagen domain for efficient viral neutralization and opsonization. However, the structure of the collagen domain significantly influences the anti-viral activity of multimerized forms of SP-D.

There is considerable evidence that the lung collectins, surfactant protein D (SP-D)3 and surfactant protein A, play important roles in innate defense against acute respiratory viral infections, most notably influenza A virus (IAV; Refs. 1, 2, 3). IAV is an important cause of morbidity and mortality especially in the very young and elderly (4, 5, 6, 7). Although IAV elicits strong adaptive immune responses, it is prone to rapid genomic variation, and a recurrent pandemic of influenza is a growing concern (8, 9). There is a window of 5–7 days before arrival of CD8+ T cells in the lung after exposure to a new IAV strain, and it is during this interval that innate defenses are most critical.

The collectins are collagenous lectins that recognize glycoconjugates on the surface of many pathogenic organisms including IAV (10, 11). They protect against infection by enhancing microbial neutralization or clearance, and by decreasing potentially harmful inflammatory responses. Collectins have four major structural domains: the N-terminal cross-linking domain, the triple helical collagen domain, a short coiled-coil neck domain, and the C-terminal carbohydrate recognition domain. The basic structural unit of all collectins is the trimer, and the minimum functional unit for microbial interactions is the trimeric C-terminal neck carbohydrate recognition domain of rat SP-D (NCRD).

The collagen domain of SP-D has been shown to be required for restoring normal lung homeostasis to SP-D−/− mice in some studies. Although all abnormalities can be corrected by the transgenic expression of full-length rat SP-D, transgenic expression of constructs lacking the collagen domain (rSftpdCDM or SP-Dcdm) or containing the N terminus and collagen domain of surfactant protein A fused to the NCRD of SP-D did not rescue the abnormal structural phenotype or lipid abnormalities of SP-D−/− mice (12, 13). Furthermore, the ability of SP-D to form multimers appears critical for homeostatic lung functions of SP-D. In this regard, overexpression of the RrSP-Dser15,20 single-arm mutant did not restore normal homeostasis and decreased the levels of endogenous SP-D in wild-type mice (14). However, restoration of some critical functions has also been achieved through instillation of the NCRD of human SP-D in SP-D−/− mice (15, 16). Furthermore, the collagen domain and multimerization per se are not absolutely required for antiviral activity against IAV; both SP-Dcdm and RrSP-Dser15,20 overexpressing SP-D−/− mice showed normalization of viral replication and virus-induced inflammation (17).

Lung collectins show lectin-dependent binding to signal inhibitory recognition protein α receptors on macrophages, with inhibition of endotoxin-mediated cellular activation (18). Alternatively, complexes of collectins with particulate ligands such as apoptotic cells can engage receptors for the collagen domain, with associated macrophage activation (18). Although similar receptors are present on neutrophils, it is not yet clear whether these mechanisms can be generally extrapolated to other phagocytic cells or microbial ligands. However, we have shown that binding of free SP-D to neutrophils down-regulates respiratory burst responses elicited by subsequent addition of IAV and that preformed SP-D-IAV complexes strongly up-regulate these responses (19).

The current study was undertaken to further examine the contributions of the N-terminal collagen domain to the anti-IAV and opsonic activities of SP-D. For these studies, we used a panel of rat SP-D mutants that vary in the presence or length of the N-terminal domains.

Dulbecco’s PBS (PBS) containing 0.9 mM calcium and 0.493 mM magnesium and PBS without calcium and magnesium were purchased from Invitrogen. PBS with added calcium and magnesium at pH 7.2 was used unless otherwise indicated. Cytochalasin B, scopoletin, and HRP were purchased from Sigma-Aldrich.

The molecular forms of SP-D used in this study are described in Table I, and their domain structure is depicted in Fig. 1. The methods for preparation of most of these constructs have been previously described. Wild-type recombinant rat SP-D (RrSP-D, referred to here as SP-D), was used as the control (20). Recombinant rat SP-Ddel29–37 (21) and SP-Dcdm (previously designated rSftdpCDM; Ref. 12) were expressed and characterized as previously described. Trimeric rat NCRDs were produced in Escherichia coli, isolated from inclusion bodies, and purified by sequential chelation and gel filtration chromatography (22, 23). One, previously uncharacterized mutant, designated simply MiniSP-D, is described in detail in Expression of MiniSP-D. None of the SP-D preparations showed biologically significant levels of endotoxin contamination, as assessed using a chromogenic assay.

Table I.

Molecular forms of SP-D used in this study

Rat SP-D PreparationPrimary StructureAssembly
Wild-type SP-D Full-length rat SP-D Dodecamer, i.e., a tetramer of trimers 
SP-Ddel29–37 Rat SP-D lacking residues 29–37 at N-terminal end of collagen domain Dodecamer 
MiniSP-D Rat SP-D lacking two internal exons of collagen sequence Dodecamer with short arms 
SPDcdm Rat SP-D deletion mutant lacking the entirety of the collagen domain; it retains the N-terminal peptide cross-linking domain. Multimer of trimers with a high proportion of dodecamers (Fig. 3). 
NCRD Rat SP-D truncation mutant lacking the N-terminal peptide and collagen domains. Trimeric neck + CRD that retains lectin activity 
Rat SP-D PreparationPrimary StructureAssembly
Wild-type SP-D Full-length rat SP-D Dodecamer, i.e., a tetramer of trimers 
SP-Ddel29–37 Rat SP-D lacking residues 29–37 at N-terminal end of collagen domain Dodecamer 
MiniSP-D Rat SP-D lacking two internal exons of collagen sequence Dodecamer with short arms 
SPDcdm Rat SP-D deletion mutant lacking the entirety of the collagen domain; it retains the N-terminal peptide cross-linking domain. Multimer of trimers with a high proportion of dodecamers (Fig. 3). 
NCRD Rat SP-D truncation mutant lacking the N-terminal peptide and collagen domains. Trimeric neck + CRD that retains lectin activity 
FIGURE 1.

Comparison of the primary structures of rat SP-D constructs. Exon derivation and nomenclature are at the top. The collagen domain is encoded on five exons, and the corresponding sequences are designated C1–C5. MiniSP-D lacks two exons of internal collagen sequence (C3–C4), whereas SP-Dcdm lacks the entire collagen domain (C1–5). Both retain the wild-type signal and N-peptide sequences (SN), and the single site of N-linked glycosylation at Asn70 in C2. The NCRD lacks the N-peptide and collagen domains (SNC1–C5). The arrow underneath wild-type recombinant rat SP-D identifies the approximate position of the small deletion in C1 of SP-Ddel29–37. See Table I for a summary of the corresponding oligomeric structures.

FIGURE 1.

Comparison of the primary structures of rat SP-D constructs. Exon derivation and nomenclature are at the top. The collagen domain is encoded on five exons, and the corresponding sequences are designated C1–C5. MiniSP-D lacks two exons of internal collagen sequence (C3–C4), whereas SP-Dcdm lacks the entire collagen domain (C1–5). Both retain the wild-type signal and N-peptide sequences (SN), and the single site of N-linked glycosylation at Asn70 in C2. The NCRD lacks the N-peptide and collagen domains (SNC1–C5). The arrow underneath wild-type recombinant rat SP-D identifies the approximate position of the small deletion in C1 of SP-Ddel29–37. See Table I for a summary of the corresponding oligomeric structures.

Close modal

A rat mutant with deletion of two internal collagenous exons corresponding to exon 4 (C3) and exon 5 (C4) was generated by PCR overlap extension using full-length rat SP-D cDNA as primer (Fig. 1 and Table I). Briefly, forward and reverse primers containing the splice junctions were synthesized. PCR reactions used ∼200 ng of ScaI linearized cDNA/pGEM-3Z template in reaction buffer containing 2 mM MgCl2, 2 mM DTT, and 5 U of Pwo DNA polymerase (Boehringer-Mannheim). Separate reactions containing the required specific and flanking (SP6 or T7) primer pairs were performed for ∼25 cycles at an annealing temperature of 52°C. The products were gel purified, and the final reaction with the flanking SP6 and T7 primers was performed for 30 cycles at an annealing temperature of 55°C. The DNA was purified using QIAquick Gel Extraction Kit (Qiagen), digested with EcoRI, and subcloned into pGEM-3Z. The mutant construct was excised from pGEM-3Z with EcoRI and subcloned into the corresponding site within the multiple cloning site of pEE14 (20). The orientation and final sequence were verified by DNA sequencing. Stable Chinese hamster ovary-K1 transfectants were isolated and cloned as previously described for wild-type rat and human SP-D (20, 24).

The stably transfected cells were incubated overnight in the presence of fresh ascorbate (50 μg/ml). The conditioned medium was collected, inhibitors were added, and any cellular debris was removed by centrifugation before ultrafiltration. The protein was isolated by maltosyl-agarose affinity chromatography (25). Bound proteins that eluted with EDTA were pooled and resolved by gel filtration on a column of 4% agarose (25). The column was precalibrated with purified recombinant rat SP-D dodecamers (RrSP-D) and trimeric subunits (RrSP-Dser15,20; Ref. 24). Peak fractions were pooled, and proteins were resolved by SDS-PAGE in the absence and presence of reduction. The total recovery of protein was significantly lower than wild type, a few hundred micrograms per liter of conditioned medium.

The molecular structures of MiniSP-D and SP-Dcdm were further characterized by electron microscopy essentially as described for natural and recombinant wild-type SP-D (20). Approximately 10 μg of each protein were adsorbed to mica flakes, followed by quick-freezing. The mica pellet was then freeze-fractured, deep-etched, and rotary replicated before transmission electron microscopy and digital photography. Images were obtained at ×70,000 primary magnification, and montages of representative molecules were assembled in Adobe Photoshop. Structures were compared with images of purified recombinant rat SP-D dodecamers.

IAV was grown in the chorioallantoic fluid of 10-day-old chicken eggs and purified on a discontinuous sucrose gradient as previously described (26). The virus was dialyzed against PBS to remove sucrose, aliquoted, and stored at −80°C until needed. Philippines 82/H3N2 (Phil82) strain and the collectin-resistant Phil82/BS strains were kindly provided by Dr. E. Margot Anders (University of Melbourne, Melbourne, Australia). The hemagglutination (HA) titer of each virus preparation was determined by titration of virus samples in PBS with thoroughly washed human type O, Rh− RBC as described (26). Postthawing, the viral stocks contained ∼5 × 108 PFU/ml.

HA inhibition was measured by serially diluting collectin or other host defense protein preparations in round-bottom 96-well plates (Serocluster U-Vinyl plates; Costar) using PBS as a diluent. After addition of 25 μl of IAV, giving a final concentration of 40 or 4 HA U/well, the IAV-protein mixture was incubated for 15 min at room temperature, followed by addition of 50 μl of a type O human erythrocyte suspension. The minimum concentration of protein required to fully inhibit the hemagglutinating activity of the viral suspension was determined by noting the highest dilution of protein that still inhibited hemagglutination. Inhibition of HA activity in a given well is demonstrated by absence of formation of an erythrocyte pellet.

MDCK cell monolayers were prepared in 96-well plates and grown to confluency. These layers were then infected with diluted IAV preparations for 45 min at 37°C in PBS and tested for the presence of IAV-infected cells after 7 h using a mAb directed against the influenza A viral nucleoprotein (provided by Dr. Nancy Cox, Centers for Disease Control and Prevention, Atlanta, GA) as described (27). IAV was preincubated for 30 min at 37°C with SP-D or control buffer, followed by addition of these viral samples to the Madin-Darby canine kidney cells.

NA activity of IAV was measured by an enzyme-linked microplate assay in which Arachis hypogaea peanut lectin was used to detect β-d-galactose-N-acetylglucosamine sequences exposed after the removal of sialic acid from fetuin (28). Wells of microtiter plate were coated with 50 μl of fetuin (Sigma-Aldrich; 20 μl/ml in PBS) overnight at 4°C and washed with PBS. Dilutions of IAV strains with different concentrations of SP-D were preincubated for 30 min at 37°C, and 50 μl of the mixture were added to fetuin-coated wells and incubated at 37°C for 2 h. After the wells were washed, 50 μl of peroxidase-labeled peanut lectin (Sigma-Aldrich; 20 μl/ml in 0.5% BSA) were added to each well for 30 min at room temperature, followed by washing and incubation with 50 μl of tetramethylpyrazine-peroxidase (Bio-Rad Laboratories) for 20 min. Finally, 50 μl of 1 N H2SO4 were added to the wells, and the optical density was measured on an ELISA plate reader at 450 nm.

Neutrophils from healthy volunteers were isolated to >95% purity by using dextran precipitation, followed by Ficoll-Paque gradient separation for the removal of mononuclear cells, and then hypotonic lysis to eliminate any contaminating erythrocytes, as previously described (26). Cell viability was determined to be >98% by trypan blue staining. The isolated neutrophils were resuspended at the appropriate concentrations in control buffer (PBS) and used within 2 h. Neutrophil collection was done with informed consent as approved by the Institutional Review Board of Boston University School of Medicine.

FITC-labeled IAV (Phil82 strain) was prepared as described (29). Uptake of virus by neutrophils was measured as previously described (24). In brief, the labeled IAV was preincubated SP-D for 30 min at 37°C followed by incubation with neutrophils for 30 min at 37°C. Trypan blue (0.2 mg/ml) was added to these samples to quench extracellular fluorescence. After a washing, the neutrophils were fixed with paraformaldehyde, and neutrophil-associated fluorescence was measured using flow cytometry. The mean neutrophil fluorescence (>1000 cells counted per sample) was measured. Rhodamine-labeled E. coli (K12 strain) was purchased from Molecular Probes, and uptake was measured by flow cytometry as described (30).

H2O2 production was measured by assessing reduction in scopoletin fluorescence as previously described (31). Neutrophils were pretreated with cytochalasin B before addition of IAV samples. Measurements were made using a POLARstar OPTIMA fluorescent plate reader (BMG Labtech).

Statistical comparisons were made using Student’s paired, two-tailed t test or ANOVA with a post hoc test (Tukey’s). ANOVA was used for multiple comparisons to a single control.

To examine the role of the collagen domain in the antiviral and opsonizing activities of SP-D, we compared the activities of a panel of wild-type and mutant forms of rat SP-D (Table I and Fig. 1). Rat SP-D was selected as the backbone for all the mutations because it almost exclusively assembles as dodecamers, thereby facilitating purification and analysis.

The length of the SP-D collagen domain is highly conserved. All known SP-Ds have 177 aa (59 Gly-X-Y triplets) in the collagen domain with the exception of bovine SP-D, which has 171. To assess the effect of removing the entire collagen domain, two mutants were examined. The first lacks all SP-D sequence N-terminal to the neck domain and consists of the trimeric NCRD. The functional properties of the mutant have been extensively characterized (32, 33). A second mutant, designated here as SP-Dcdm, lacks only the collagen domain and forms multimers of trimers by virtue of the retained N-terminal peptide (12). This protein binds to carbohydrate and has been shown to correct the defective IAV clearance and associated abnormal inflammatory responses in SP-Dnull mice (12).

To assess the functional impact of a shortened collagen domain, we examined two additional mutants. The first, SP-Ddel29–37, contains a 9-aa deletion at the N terminus of the collagen domain, corresponding to residues 29–37 of the mature protein. The original rationale for the deletion was to remove an embedded DGRDGR sequence hypothesized to be involved in binding to cellular integrins; this mutant eluted near the position of wild-type rat SP-D dodecamers on gel filtration (21). As detailed in Materials and Methods, we also produced a novel construct with an extensive internal deletion, designated MiniSP-D. The deletion consists of 26 consecutive Gly-X-Y triplets corresponding to two consecutive collagenous exons: exon 4 (C3) and exon 5 (C4). Thus, MiniSP-D has a shortened, but continuous and uninterrupted, collagen domain of 33 triplets. It retains the single site of N-linked glycosylation at Asn70 and preserves the normal junctions of the collagen domain with the flanking N- and C-terminal domains. MiniSP-D retained lectin activity, as evidenced by its specific, calcium-dependent binding to maltosyl agarose during the purification. MiniSP-D migrated more rapidly than recombinant rat SP-D, in both the absence and presence of sulfhydryl reduction (Fig. 2, top). Using globular standards, the estimated mass on reduced gels was approximately 34 kDa, consistent with the size of the deletion. Identity of the major species was confirmed by immunoblotting with Ab to rat SP-D (Fig. 2, bottom). The derivation of the lower m.w. immunoreactive species shown on the blot of gel filtration purified MiniSP-D is uncertain. Unreduced SP-D chains migrate faster than the reduced protein because of intrachain bonds within the lectin domain. However, the size is more consistent with proteolytic nicking of MiniSP-D chains. In any case, comparison with the protein stain indicates that the component accounts for a very small fraction of the total protein.

FIGURE 2.

Characterization of MiniSP-D. Aliquots of the MiniSP-D were resolved by SDS-PAGE (10%) in the absence and presence of DTT. Parallel lanes included purified wild-type rat SP-D. Proteins were visualized by protein staining (top) or immunoblotting with polyclonal Ab to rat SP-D (bottom). Left, Mark 12 standards (Invitrogen). The arrows at right identify species corresponding to MiniSP-D.

FIGURE 2.

Characterization of MiniSP-D. Aliquots of the MiniSP-D were resolved by SDS-PAGE (10%) in the absence and presence of DTT. Parallel lanes included purified wild-type rat SP-D. Proteins were visualized by protein staining (top) or immunoblotting with polyclonal Ab to rat SP-D (bottom). Left, Mark 12 standards (Invitrogen). The arrows at right identify species corresponding to MiniSP-D.

Close modal

To better define the molecular structures of MiniSP-D and SP-Dcdm, transmission electron microscopy was performed as described in Materials and Methods. Consistent with previous studies (20), purified rat SP-D dodecamers showed the expected four-armed assembly, with paired arms and terminal globular domains (Fig. 3 A). Only rare incomplete or larger assemblies were identified. Ultrastructural analysis of proteolytic fragments has previously shown that the arms correspond to the pepsin-resistant triple helical domains (43 nm long), whereas the globules (9-nm diameter) correspond to the trimeric lectin domains (20).

FIGURE 3.

Ultrastructural analysis of collagen deletion mutants as compared with wild-type SP-D. A, Recombinant rat SP-D (RrSP-D) dodecamers. Note the central hub, four rigid arms, and terminal globular domains. The arms of SP-D have an overall span of >100 nm. Previous proteolytic mapping demonstrated that the arms correspond to the pepsin-resistant triple helical collagen domains, whereas the globules correspond to the collagenase-resistant C-terminal lectin domains. B, MiniSP-D consists of dodecamers with shortened arms. C, SP-Dcdm consists predominantly of multimers of globular domains, with a large proportion of assemblies containing four globules, consistent with a dodecamer. Some assemblies, particularly in the top row, show a central connecting structure, presumed to include the N-peptide. Selected larger assemblies are shown in the column at far right.

FIGURE 3.

Ultrastructural analysis of collagen deletion mutants as compared with wild-type SP-D. A, Recombinant rat SP-D (RrSP-D) dodecamers. Note the central hub, four rigid arms, and terminal globular domains. The arms of SP-D have an overall span of >100 nm. Previous proteolytic mapping demonstrated that the arms correspond to the pepsin-resistant triple helical collagen domains, whereas the globules correspond to the collagenase-resistant C-terminal lectin domains. B, MiniSP-D consists of dodecamers with shortened arms. C, SP-Dcdm consists predominantly of multimers of globular domains, with a large proportion of assemblies containing four globules, consistent with a dodecamer. Some assemblies, particularly in the top row, show a central connecting structure, presumed to include the N-peptide. Selected larger assemblies are shown in the column at far right.

Close modal

Analysis of purified MiniSP-D confirmed a four-armed structure resembling the native dodecameric protein, but with shortened arms (Fig. 3 B). Unlike the natural protein, the paired arms were sometimes closely apposed, creating a more rod-like structure with close approximation of two terminal globules (e.g., last image in panel). The arms are slightly shorter than predicted and some show additional mass near the hub. Although unexplained, this suggests partial overlap of the collagen domains within the hub region of SP-D dodecamers and/or differences in adsorption of wild-type and mutant molecules to the mica.

Previous gel filtration of SP-Dcdm showed a complex profile, but with major species eluting near the expected position of a 12-chain complex (12). To further assess the extent and pattern of multimeric assembly, non-size-selected preparations were examined as above. SP-Dcdm predominantly consisted of multimers of globular domains (Fig. 3,C). As illustrated in Fig. 3,C, these often consisted of a relatively symmetrical array of four globules, with each globule comparable in size with the trimeric C-terminal globule of native SP-D. In occasional profiles, a small central connecting structure could be identified, presumably representing the associated noncollagenous N-peptide domains. There was a subpopulation of larger assemblies, including apparent dimers of the predominant tetrameric structure (Fig. 3 C, column at far right), and some smaller assemblies consisting of two or three globules were observed (not shown). Although superimposition of globules and differences in orientation precluded a reliable assessment of size distribution, the findings confirm a propensity to form multimers consisting of four trimeric globular domains.

As shown in Table II, and consistent with previous studies of the human NCRD, the rat NCRD had minimal HA-inhibitory activity against the collectin-sensitive Phil82 strain of IAV. Reduced activity is reflected in the higher concentration of SP-D needed to achieve inhibition. In contrast with the rat NCRD, rat SP-Dcdm had substantial HA-inhibitory activity, albeit reduced compared with wild-type SP-D. SP-Ddel29–37 retained HA-inhibitory activity, which was again somewhat reduced compared with wild-type SP-D. MiniSP-D and wild-type SP-D showed similar molar inhibitory activities. We also tested the mutants against the partially collectin-resistant Phil82/BS strain. Although considerably higher concentrations of wild-type SP-D were needed to inhibit this strain as compared with Phil82, MiniSP-D showed substantially increased weight or molar potency. There was no difference in activity of SP-Ddel29–37 against the resistant strain compared with wild-type SP-D.

Table II.

HA-inhibitory activity of SPD constructs against the Phil82 and Phil 82/BS strains of IAVa

SP-D PreparationMean HA-Inhibitory Concentration (ng/ml)
Phil82 IAVPhil82/BS IAV
SP-D wild type 17 ± 5 449 ± 62 
SP-Ddel29–37 52 ± 10c 550 ± 180 
MiniSP-D 14 ± 2 91 ± 6b 
SPDcdm 217 ± 43c >8,400 
NCRD 24,000 ± 1,000c >24,000 
SP-D PreparationMean HA-Inhibitory Concentration (ng/ml)
Phil82 IAVPhil82/BS IAV
SP-D wild type 17 ± 5 449 ± 62 
SP-Ddel29–37 52 ± 10c 550 ± 180 
MiniSP-D 14 ± 2 91 ± 6b 
SPDcdm 217 ± 43c >8,400 
NCRD 24,000 ± 1,000c >24,000 
a

Results are means ± SEM of at least three experiments. NCRD had significantly weaker HA-inhibiting activity than all other mutants for both viral strains.

b

, Instances in which HA-inhibiting activity of mutant was greater than that of wild-type SP-D.

c

, Instances in which HA-inhibiting activity of mutants were significantly reduced compared to wild type.

Overall, similar results were obtained in viral neutralization assays using the Phil82 wild-type viral strain (Fig. 4). The NCRD had no measurable activity, but collagen deletion mutants retained activity comparable with, or slightly less than, that of wild-type SP-D. The SP-Ddel29–37 mutant had relatively greater activity on the neutralization assay as compared with its activity on HA inhibition. We cannot currently account for this difference.

FIGURE 4.

Neutralizing activity of SP-D constructs. NCRD did not cause any reduction in infectious titer of Phil82 strain of IAV; however, SP-Dcdm, SP-Ddel29–37, and MiniSP-D all caused significant reduction in infectious titers (n = 3 or more experiments; p < 0.02). Wild-type SP-D caused significant reductions in viral titers as previously described, but its effects were not significantly greater than those of SP-Dcdm, SP-Ddel29–37, or MiniSP-D as assessed by ANOVA.

FIGURE 4.

Neutralizing activity of SP-D constructs. NCRD did not cause any reduction in infectious titer of Phil82 strain of IAV; however, SP-Dcdm, SP-Ddel29–37, and MiniSP-D all caused significant reduction in infectious titers (n = 3 or more experiments; p < 0.02). Wild-type SP-D caused significant reductions in viral titers as previously described, but its effects were not significantly greater than those of SP-Dcdm, SP-Ddel29–37, or MiniSP-D as assessed by ANOVA.

Close modal

SP-Dcdm caused significant NA inhibition of the Phil82 IAV strain comparable with wild-type SP-D (Fig. 5). By contrast, MiniSP-D caused significantly greater NA inhibition than wild-type SP-D.

FIGURE 5.

Role of collagen domain in neuraminidase inhibition by SP-D. Phil82 IAV was preincubated with wild-type SP-D, SP-Dcdm, or MiniSP-D and NA activity was measured as described. Results represent mean ± SEM of three or more experiments. All three forms of SP-D caused significant inhibition of NA activity. Results obtained with SP-Dcdm were not different from results obtained with wild-type SP-D; however, MiniSP-D caused significantly greater inhibition than wild-type SP-D (p < 0.05 by ANOVA).

FIGURE 5.

Role of collagen domain in neuraminidase inhibition by SP-D. Phil82 IAV was preincubated with wild-type SP-D, SP-Dcdm, or MiniSP-D and NA activity was measured as described. Results represent mean ± SEM of three or more experiments. All three forms of SP-D caused significant inhibition of NA activity. Results obtained with SP-Dcdm were not different from results obtained with wild-type SP-D; however, MiniSP-D caused significantly greater inhibition than wild-type SP-D (p < 0.05 by ANOVA).

Close modal

We have previously shown that wild-type SP-Ds are potent viral agglutinins and that this aggregating activity correlates with the ability of SP-D to promote viral uptake by neutrophils (29). Fig. 6 demonstrates that the rat NCRD lacked viral aggregating activity as assessed by light transmission assays (Fig. 6,C). Higher concentrations of NCRD (e.g., 3.2 or 6.4 μg/ml) also had no activity in this assay (not shown). In contrast, all the collagen deletion mutants readily induced viral aggregation. On a weight basis, SP-Dcdm caused a similar degree of viral aggregation as wild-type SP-D (Fig. 6,A). SP-Ddel29–37 had aggregating activity comparable to wild-type SP-D (Fig. 6A and B); however, MiniSP-D caused substantially greater viral aggregation than wild-type SP-D (Fig. 6, A and B).

FIGURE 6.

Collagen domain deletion mutants retain viral aggregating activity. Viral aggregation was measured by reductions in light transmission through stirred suspensions of Phil82 IAV. Results are mean ± SEM of three or more experiments. NCRD caused no detectable aggregation, whereas the other preparations all had significant aggregating activity (p < 0.01 for all compared with control buffer). The activity of SP-Dcdm and SP-Ddel29–37 were approximately the same as that of wild-type SP-D; however, MiniSP-D caused greater viral aggregation than wild-type SP-D (p < 0.05 as assessed by ANOVA).

FIGURE 6.

Collagen domain deletion mutants retain viral aggregating activity. Viral aggregation was measured by reductions in light transmission through stirred suspensions of Phil82 IAV. Results are mean ± SEM of three or more experiments. NCRD caused no detectable aggregation, whereas the other preparations all had significant aggregating activity (p < 0.01 for all compared with control buffer). The activity of SP-Dcdm and SP-Ddel29–37 were approximately the same as that of wild-type SP-D; however, MiniSP-D caused greater viral aggregation than wild-type SP-D (p < 0.05 as assessed by ANOVA).

Close modal

Although the NCRD did not significantly increase viral uptake by neutrophils, SP-Dcdm, SP-Ddel29–37, and MiniSP-D were comparable to wild-type SP-Ds (Fig. 7). To determine whether the ability of SP-Dcdm and MiniSP-D to act as opsonins was confined to IAV, we also examined effects on uptake of E. coli. SP-Dcdm and MiniSP-D increased the uptake of E. coli, whereas the NCRD did not (Fig. 8).

FIGURE 7.

SP-Dcdm and MiniSP-D increase neutrophil uptake of IAV. Neutrophil uptake of FITC-labeled IAV was assessed by flow cytometry. Results are mean ± SEM of three or more experiments. Preincubation of IAV with NCRD caused no increase in uptake compared with control buffer. In contrast, wild-type SP-D, SP-Dcdm, and MiniSP-D all resulted in dose-related increases in uptake of IAV by neutrophils (p < 0.05 for all compared with control buffer). MiniSP-D caused significantly greater increases in uptake of IAV than wild-type SP-D (p < 0.02).

FIGURE 7.

SP-Dcdm and MiniSP-D increase neutrophil uptake of IAV. Neutrophil uptake of FITC-labeled IAV was assessed by flow cytometry. Results are mean ± SEM of three or more experiments. Preincubation of IAV with NCRD caused no increase in uptake compared with control buffer. In contrast, wild-type SP-D, SP-Dcdm, and MiniSP-D all resulted in dose-related increases in uptake of IAV by neutrophils (p < 0.05 for all compared with control buffer). MiniSP-D caused significantly greater increases in uptake of IAV than wild-type SP-D (p < 0.02).

Close modal
FIGURE 8.

SP-Dcdm increases neutrophil uptake of E. coli. Uptake of Rhodamine-labeled E. coli was assessed by flow cytometry. Results are mean ± SEM of three to five experiments. SP-Dcdm significantly increased neutrophil uptake of E. coli (left; n = 3; p < 0.05 compared with control buffer), whereas NCRD did not. Wild-type SP-D and MiniSP-D also caused dose-related increases in uptake of E. coli and the effect of MiniSP-D was significantly greater than that of wild-type SP-D (right; p < 0.04 at 3.4 μg/ml; n = 5).

FIGURE 8.

SP-Dcdm increases neutrophil uptake of E. coli. Uptake of Rhodamine-labeled E. coli was assessed by flow cytometry. Results are mean ± SEM of three to five experiments. SP-Dcdm significantly increased neutrophil uptake of E. coli (left; n = 3; p < 0.05 compared with control buffer), whereas NCRD did not. Wild-type SP-D and MiniSP-D also caused dose-related increases in uptake of E. coli and the effect of MiniSP-D was significantly greater than that of wild-type SP-D (right; p < 0.04 at 3.4 μg/ml; n = 5).

Close modal

IAV triggers H2O2 response by neutrophils and these are increased by preincubation of IAV with wild-type SP-Ds (3, 19). SP-Dcdm retained the ability to increase neutrophil H2O2, but the NCRD did not (Fig. 9, left). In a separate set of experiments, MiniSP-D caused greater H2O2 response than wild-type SP-D (Fig. 9, right).

FIGURE 9.

SP-Dcdm and MiniSP-D increase neutrophil H2O2 production in response to IAV. Preincubation of IAV with SP-Dcdm caused significant increase in H2O2 production as compared with IAV alone (left; n = 4; p value shown). NCRD did not increase neutrophil H2O2 responses compared with IAV alone. MiniSP-D caused significantly greater increase in H2O2 response than in with wild-type SP-D (right; n = 5; p value shown for comparison; responses also different by ANOVA).

FIGURE 9.

SP-Dcdm and MiniSP-D increase neutrophil H2O2 production in response to IAV. Preincubation of IAV with SP-Dcdm caused significant increase in H2O2 production as compared with IAV alone (left; n = 4; p value shown). NCRD did not increase neutrophil H2O2 responses compared with IAV alone. MiniSP-D caused significantly greater increase in H2O2 response than in with wild-type SP-D (right; n = 5; p value shown for comparison; responses also different by ANOVA).

Close modal

Using a panel of SP-D constructs derived from rat SP-D, we demonstrate that complete or partial deletion of the collagen domain of SP-D does not preclude HA-inhibitory, NA-inhibitory, or IAV-neutralizing activity as long as the molecules form higher order oligomers. Purified trimeric rat NCRDs, which lack the N-terminal peptide and the entirety of the collagen domain, have minimal HA inhibitory activity and no measurable neutralizing activity. Ultrastructural studies of the rat SP-Dcdm and MiniSP-D, confirmed higher order oligomerization with the assembly of dodecamers with missing or shortened arms, respectively. We conclude that the retention of the N terminus permits the formation of multimers, which in turn enhances viral binding via cooperative interactions among lectin domains. The ability to form multimers of trimeric lectin domains is consistent with the substantial viral aggregating activities of SP-Dcdm, SP-Ddel29–37, and MiniSP-D. We have recently reported that cross-linking of trimeric human NCRD of SP-D with mAbs greatly enhances the ability of the NCRD to inhibit infectivity of IAV and to cause viral aggregation (34). Hence, using two different experimental models we have now confirmed the importance of cooperative interactions between NCRD trimers in antiviral activity against IAV and shown that antiviral activity can be obtained without the collagen domain of SP-D. The current study demonstrates this without the need for Ab cross-linking.

For convenience, most of our binding assays compare equivalent weight amounts of purified proteins. However, molar potencies are undoubtedly more relevant to biological phenomena. Because an SP-Dcdm chain has approximately one-half the mass of a wild-type monomer, a given weight of collagenous domain dodecamer will have nearly twice as many dodecameric molecules as the wild-type protein. Likewise, a given weight of NCRD trimers will have at least four times as many molecules as a given weight of SP-Dcdm dodecamer. Given these considerations, it is evident that deletion of the collagen domain generally decreases antiviral activity. Nevertheless, the residual activity is much greater than for trimeric NCRDs, emphasizing the importance of multimerization and cooperative interactions among carbohydrate recognition domains.

Other important new findings of this study are that the MiniSP-D preparation actually has increased antiviral and opsonic activities compared wild-type SP-D and that the DGRDGR sequence in the collagen domain does not contribute to antiviral or opsonic activities. Comparison of the activities of SP-Dcdm, SP-Ddel29–37, and MiniSP-D suggests that there is an optimal collagen domain length for viral aggregation and for aggregation-dependent effects on neutrophils. MiniSP-D had greater aggregating and NA-inhibitory activity than wild-type SP-D, SP-Ddel29–37, or SP-Dcdm. Further testing of preparations of human or rat SP-D with various lengths of collagen domain deletions could be informative.

At present, it is unclear how the structural perturbations contribute to enhanced interactions of MiniSP-D with the SP-D-resistant Phil82BS viral strain. Given that Phil82BS lacks a single glycan on the HA, we infer enhanced interactions with other glycans displayed on the viral envelope. It is intriguing that MiniSP-D has some capacity to form rod-like structures with closely apposed trimeric lectin domains (Fig. 3). This effectively creates a bivalent molecule with hexameric, rather than trimeric, binding surfaces. Close alignment of the paired arms is only rarely observed in preparations of wild-type rat or human SP-D.

It is notable that the SP-Dcdm showed substantial opsonizing activity for IAV and E. coli and was able to stimulate respiratory responses of neutrophils to virus. By contrast, the rat NCRD was inactive in these assays. Consistent with the latter findings, we previously observed that full-length rat SP-D trimers that lack cysteine in the N-terminal cross-linking domain (RrSP-Dser15,20) do not increase viral or bacterial uptake by neutrophils or increase virus-induced respiratory burst responses (30, 35). In contrast, cross-linking of human SP-D NCRD with mAbs allowed the NCRD to promote viral uptake and H2O2 generation by neutrophils (34). The results of the current study are important because they demonstrate opsonizing activity without the collagen domain but also without the need for Ab cross-linking, which may in part involve engagement of FcRs on neutrophils (36). Collectively, the findings are consistent with the concept that multimerization and viral aggregating activity are required for effects of SP-D on viral uptake and H2O2 responses, but that the collagen domain per se is not required. Furthermore, the increased ability of MiniSP-D to promote viral uptake of H2O2 responses by neutrophils is consistent with the increased viral aggregating activity of this preparation.

We cannot exclude the possibility that rat proteins have different interactions with human neutrophil receptors than human SP-D. Although we previously described differences in the saccharide binding preferences of human and rat NCRDs (33), we have not found significant differences between activity of human and rat SP-D dodecamers in assays using human neutrophils. Furthermore, the N-terminal and collagen domains are highly conserved, making it unlikely that there are significant species differences in the recognition of these domains.

As discussed in the Introduction, receptors for the collagen domain of collectins are present on macrophages and neutrophils. In the case of macrophages, there is evidence that these receptors can mediate activating effects of collectins complexed with microbial or apoptotic cells ligands (18). Nevertheless, the current studies demonstrate that interactions with the collagen domain are not required for promotion of neutrophil uptake of IAV or for potentiation of H2O2 responses of neutrophils to IAV. A feature that may distinguish IAV (and E. coli) from other potential ligands is the ability of phagocyte to recognize the ligand in the absence of a classical opsonin. In the case of IAV, sialic acid-rich ligands on the neutrophil surface can serve as receptors for the viral HA; cross-linking of these receptors by SP-D-IAV complexes results in activation of the cell (37, 38). It is possible that collagen domain receptors are more important for responses to ligands that lack natural binding sites on the neutrophil surface.

We thank Pengnian Zhang for generating the MiniSP-D construct.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This research was supported by National Institutes of Health Grants HL-06931 (to K.L.H.), HD28827 (to P.K.), and HL-44015 and HL-29594 (to E.C.C.).

3

Abbreviations used in this paper: SP-D, surfactant protein D (recombinant rat SP-D); NCRD, neck and rat SP-D + carbohydrate recognition domain; IAV, influenza A virus; HA, hemagglutinin; NA, neuraminidase.

1
Pastva, A. M., J. R. Wright, K. L. Williams.
2007
. Immunomodulatory roles of surfactant proteins A and D: implications in lung disease.
Proc. Am. Thorac. Soc.
4
:
252
-257.
2
Hawgood, S., C. Brown, J. Edmondson, A. Stumbaugh, L. Allen, J. Goerke, H. Clark, F. Poulain.
2004
. Pulmonary collectins modulate strain-specific influenza a virus infection and host responses.
J. Virol.
78
:
8565
-8572.
3
Hartshorn, K. L., E. C. Crouch, M. R. White, P. Eggleton, A. I. Tauber, D. Chang, K. Sastry.
1994
. Evidence for a protective role of pulmonary surfactant protein D (SP-D) against influenza A viruses.
J. Clin. Invest.
94
:
311
-319.
4
Morens, D..
2003
. Influenza-related mortality: considerations for practice and public health.
JAMA
289
:
227
-229.
5
Nichol, K., A. Lind, K. Margolis, M. Murdoch, R. McFadden, M. Hauge, S. Magnan, M. Drake.
1995
. The effectiveness of vaccination against influenza in healthy working adults.
N. Engl. J. Med.
333
:
889
-893.
6
Izurieta, H. S., W. W. Thompson, P. Kramarz, D. K. Shay, R. L. Davis, F. DeStefano, S. Black, H. Shinefield, K. Fukuda.
2000
. Influenza and the rates of hospitalization for respiratory disease among infants and young children.
N. Engl. J. Med.
342
:
232
-239.
7
Chiu, S., Y. L. Lau, K. Chan, W. Wong, J. M. Peiris.
2002
. Influenza-related hospitalizations among children in Hong Kong.
N. Engl. J. Med.
347
:
2097
-2103.
8
Trampuz, A., R. M. Prabhu, T. F. Smith, L. M. Baddour.
2004
. Avian influenza: new pandemic threat?.
Mayo Clin. Proc.
79
:
523
-530.
9
Webby, R. J., R. G. Webster.
2003
. Are we ready for pandemic influenza?.
Science
302
:
1519
-1522.
10
Shepherd, V. L..
2002
. Distinct roles for lung collectins in pulmonary host defense.
Am. J. Respir. Cell Mol. Biol.
26
:
257
-260.
11
van de Wetering, J. K., L. M. van Golde, J. J. Batenburg.
2004
. Collectins: players of the innate immune system.
Eur J. Biochem.
271
:
1229
-1249.
12
Kingma, P. S., L. Zhang, M. Ikegami, K. Hartshorn, F. X. McCormack, J. A. Whitsett.
2006
. Correction of pulmonary abnormalities in Sftpd−/− mice requires the collagenous domain of surfactant protein D.
J. Biol. Chem.
281
:
24496
-24505.
13
Zhang, L., M. Ikegami, T. R. Korfhagen, F. X. McCormack, M. Yoshida, R. M. Senior, J. M. Shipley, S. D. Shapiro, J. A. Whitsett.
2006
. Neither SP-A nor NH2-terminal domains of SP-A can substitute for SP-D in regulation of alveolar homeostasis.
Am. J. Physiol.
291
:
L181
-L190.
14
Zhang, L., M. Ikegami, E. Crouch, T. Korfhagen, J. Whitsett.
2001
. Activity of pulmonary surfactant protein-D (SP-D) in vivo is dependent on oligomeric structure.
J. Biol. Chem.
276
:
19214
-19219.
15
Knudsen, L., M. Ochs, R. Mackay, P. Townsend, R. Deb, C. Muhlfeld, J. Richter, F. Gilbert, S. Hawgood, K. Reid, H. Clark.
2007
. Truncated recombinant human SP-D attenuates emphysema and type II cell changes in SP-D deficient mice.
Respir. Res.
8
:
70
16
Clark, H., N. Palaniyar, S. Hawgood, K. B. M. Reid.
2003
. A recombinant fragment of human surfactant protein D reduces alveolar macrophage apoptosis and pro-inflammatory cytokines in mice developing pulmonary emphysema.
Ann. NY Acad. Sci.
1010
:
113
-116.
17
Zhang, L., K. L. Hartshorn, E. C. Crouch, M. Ikegami, J. A. Whitsett.
2002
. Complementation of pulmonary abnormalities in SP-D−/− mice with an SP-D/conglutinin fusion protein.
J. Biol. Chem.
277
:
22453
-22459.
18
Gardai, S., Y. Xiao, M. Dickinson, J. Nick, D. Voelker, K. Green, P. Henson.
2003
. By binding SIRP or calreticulin/CD91, lung collectins act as dual function surveillance molecules to suppress or enhance inflammation.
Cell
115
:
13
-23.
19
White, M. R., E. Crouch, J. Vesona, P. J. Tacken, J. J. Batenburg, R. Leth-Larsen, U. Holmskov, K. L. Hartshorn.
2005
. Respiratory innate immune proteins differentially modulate the neutrophil respiratory burst response to influenza A virus.
Am. J. Physiol.
289
:
L606
-L616.
20
Crouch, E., D. Chang, K. Rust, A. Persson, J. Heuser.
1994
. Recombinant pulmonary surfactant protein D.
J. Biol. Chem.
269
:
15808
-15813.
21
Borron, P., E. Crouch, J. Lewis, J. Wright, F. Possmayer, L. Fraher.
1998
. Recombinant rat surfactant-associated protein D inhibits human T lymphocyte proliferation and IL-2 production.
J. Immunol.
161
:
4599
-4603.
22
Crouch, E., B. McDonald, K. Smith, T. Cafarella, B. Seaton, J. Head.
2006
. Contributions of phenylalanine 335 to ligand recognition by human surfactant protein D: ring interactions with SP-D ligands.
J. Biol. Chem.
281
:
18008
-18014.
23
Crouch, E., Y. Tu, D. Briner, B. McDonald, K. Smith, U. Holmskov, K. Hartshorn.
2005
. Ligand specificity of human surfactant protein D: expression of a mutant trimeric collectin that shows enhanced interactions with influenza A virus.
J. Biol. Chem.
280
:
17046
-17056.
24
Hartshorn, K., D. Chang, K. Rust, M. White, J. Heuser, E. Crouch.
1996
. Interactions of recombinant human pulmonary surfactant protein D and SPD multimers with influenza A.
Am. J. Physiol.
271
:
L753
-L762.
25
Crouch, E., A. Persson, D. Chang, J. Heuser.
1994
. Molecular structure of pulmonary surfactant protein D (SP-D).
J. Biol. Chem.
269
:
17311
-17319.
26
Hartshorn, K. L., M. Collamer, M. Auerbach, J. B. Myers, N. Pavlotsky, A. I. Tauber.
1988
. Effects of influenza A virus on human neutrophil calcium metabolism.
J. Immunol.
141
:
1295
-1301.
27
Hartshorn, K., K. Sastry, D. Chang, M. White, E. Crouch.
2000
. Enhanced anti-influenza activity of a recombinant pulmonary surfactant protein D and serum conglutinin fusion protein.
Am. J. Physiol.
278
:
L90
-L98.
28
Tecle, T., M. R. White, E. C. Crouch, K. L. Hartshorn.
2007
. Inhibition of influenza viral neuraminidase activity by collectins.
Arch. Virol.
152
:
1731
-1742.
29
Hartshorn, K. L., M. R. White, V. Shepherd, K. Reid, J. C. Jensenius, E. C. Crouch.
1997
. Mechanisms of anti-influenza activity of surfactant proteins A and D: comparison with serum collectins.
Am. J. Physiol.
273
:
L1156
-L1166.
30
Hartshorn, K., E. Crouch, M. White, M. Colamussi, A. Kakkanatt, B. Tauber, V. Shepherd, K. Sastry.
1998
. Pulmonary surfactant proteins A and D enhance neutrophil uptake of bacteria.
Am. J. Physiol.
274
:
L958
-L969.
31
Hartshorn, K. L., M. Collamer, M. R. White, J. H. Schwartz, A. I. Tauber.
1990
. Characterization of influenza A virus activation of the human neutrophil.
Blood
75
:
218
-226.
32
Wang, H., J. Head, P. Kosma, H. Brade, S. Muller-Loennies, S. Sheikh, B. McDonald, K. Smith, T. Cafarella, B. Seaton, E. Crouch.
2008
. Recognition of heptoses and the inner core of bacterial lipopolysaccharides by surfactant protein d.
Biochemistry
47
:
710
-720.
33
Crouch, E. C., K. Smith, B. McDonald, D. Briner, B. Linders, J. McDonald, U. Holmskov, J. Head, K. Hartshorn.
2006
. Species differences in the carbohydrate binding preferences of surfactant protein D.
Am. J. Respir. Cell Mol. Biol.
35
:
84
-94.
34
Tecle, T., M. R. White, G. L. Sorensen, D. Gantz, N. Kacak, U. Holmskov, K. Smith, E. C. Crouch, K. L. Hartshorn.
2008
. Critical role for cross-linking of trimeric lectin domains of surfactant protein D in antiviral activity against influenza A virus.
Biochem. J.
412
:
323
-329.
35
Brown-Augsburger, P., K. Hartshorn, D. Chang, K. Rust, C. Fliszar, H. Welgus, E. Crouch.
1996
. Site directed mutagenesis of Cys15 and Cys20 of pulmonary surfactant protein D: expression of a trimeric protein with altered anti-viral properties.
J. Biol. Chem.
271
:
13724
-13730.
36
Tacken, P. J., K. L. Hartshorn, M. R. White, C. van Kooten, J. G. van de Winkel, K. B. Reid, J. J. Batenburg.
2004
. Effective targeting of pathogens to neutrophils via chimeric surfactant protein D/anti-CD89 protein.
J. Immunol.
172
:
4934
-4940.
37
Hartshorn, K. L., L. S. Liou, M. R. White, M. M. Kazhdan, J. L. Tauber.
1995
. Neutrophil deactivation by influenza A virus: role of hemagglutinin binding to specific sialic acid-bearing cellular proteins.
J. Immunol.
154
:
3952
-3960.
38
Daigneault, D. E., K. L. Hartshorn, L. S. Liou, G. M. Abbruzi, M. R. White, S. Oh, A. I. Tauber.
1992
. Influenza A virus binding to human neutrophils and cross-linking requirements for activation.
Blood
80
:
3227
-3234.