Surfactant Protein-D and Surfactant Inhibit Endotoxin-Induced Pulmonary Inflammation^* FREE TO VIEW

Lipopolysaccharide (LPS) and lipoteichoic acid (LTA) are constituents of the outer cell wall of Gram-negative microorganisms (ie, LPS) and Gram-positive microorganisms (ie, LTA) that activate innate immune defenses and recruit inflammatory cells during infection in various organs including the lung. Acute lung injury (ALI) is a common clinical problem that is associated with morbidity and mortality.¹ During bacterial infection, LPS and LTA increase capillary permeability, the expression of cellular adhesion molecules, and the expression of proinflammatory cytokines and chemokines, which are associated with ALI.²³ Surfactant protein (SP)-D is a member of the collectin family of polypeptides that bind LPS and LTA in vitro.⁴⁵⁶ SP-D binds to various bacterial, fungal, and viral surfaces.⁶⁷ The important role of SP-D in the protection of the lung during infection by influenza A, respiratory syncytial viruses, and Pseudomonas aeruginosa was previously demonstrated in Sftpd^−/− mice.⁸⁹¹⁰¹¹ Several clinical conditions associated with chronic lung inflammation and persistent microbial infections, including premature infants, bronchopulmonary dysplasia, and cystic fibrosis¹²¹³¹⁴ are associated with decreased levels of SP-D in BAL fluid (BALF).

Pulmonary surfactant plays an important role in the protection of the lung during LPS-induced injury and infection.¹⁵¹⁶ In addition to the protective effects of pulmonary surfactant against LPS-induced lung injury, exogenous surfactant enhances the distribution of drugs and particles in the lung. For example, surfactant improved the spreading of adenoviral vectors over the alveolar surface in the adult rabbit lung.¹⁷ Exogenous surfactants used for the treatment of respiratory distress syndrome (RDS) in preterm infants contain surfactant lipids and proteins (SP-B and SP-C) but do not contain the pulmonary collectins SP-A or SP-D.¹⁸ The addition of recombinant human SP-D (rhSP-D) to surfactant prevented death from endotoxin shock by reducing the leakage of intratracheally administered Escherichia coli LPS to the systemic circulation in the premature lamb.¹⁹ To assess the potential utility of SP-D in the protection of the adult lung from inflammation, we tested the following assumptions: (1) Sftpd^−/− mice are more susceptible to intratracheal LPS than Sftpd^+/+ mice; (2) rhSP-D treatment suppresses the influx of neutrophils and reduces the lung inflammation caused by intratracheal LPS or LTA; and (3) the addition of exogenous surfactant with rhSP-D influences LPS-induced and LTA-induced pulmonary inflammation.

Sftpd^−/− mice were generated by targeted gene inactivation, as previously described.⁸ National Institutes of Health Swiss black adult (7 weeks old) Sftpd^+/+ and Sftpd^−/− mice were utilized in this study. The mice were housed in barrier containment prior to use. The results of viral serology tests remained negative. The Animal Care & Use Committee at the Cincinnati Children’s Hospital Research Foundation approved all protocols and procedures.

rhSP-D was synthesized, as previously described,¹⁹ by the transfection of Chinese hamster ovary cells with a complementary DNA encoding full-length human SP-D. Purified rhSP-D migrated as a mixture of dodecamers and multimers of > 1 × 10⁶ d on size-exclusion chromatography. The rhSP-D migrated as a trimer under nonreducing conditions and fully converted to a approximately 48-kd monomeric form when reduced on sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels. rhSP-D bound and aggregated E coli 055:B5 LPS in vitro in a calcium-dependent manner. The endotoxin level in the rhSP-D preparations ranged from 0.1 to 0.5 EU/mL (Limulus Lysate Assay; Charles River Laboratories; Wilmington, MA). In a preliminary study, instillation of a treatment dose of rhSP-D into normal adult mice did not induce lung inflammation. Thus, the endotoxin level in rhSP-D either was below levels that induce inflammation or the endotoxin present was bound to rhSP-D and was unable to elicit a response. The addition of rhSP-D did not influence the surface activity of the surfactant (Survanta; Ross Products Division, Abbott Laboratories, Columbus, OH). The mean (± SEM) minimum surface tension analyzed by a captive bubble surfactometer²⁰ was low (surfactant + buffer, 3.4 ± 0.4; and surfactant + rhSP-D, 3.3 ± 0.6 mN/m; n = 3), and rhSP-D containing surfactant (Survanta) improved the lung function of newborn premature lambs with surfactant deficiency.¹⁹ The rhSP-D was diluted to a concentration of 0.5 mg/mL in a buffer containing 20 mmol/L Tris, 200 mmol/L NaCl, and 1 mmol/L ethylenediaminetetraacetic acid at pH 7.4. Control mice were instilled with the same volume of buffer. The number of neutrophils in the BALF after instillation of the buffer into the Sftpd^+/+ and Sftpd^−/− mice lung, without LPS or LTA injection, were similar to that after saline solution instillation (data not shown). In preliminary studies, the addition of 5 mmol/L CaCl₂ to rhSP-D did not provide additional protection against LPS-induced lung inflammation in either Sftpd^−/− or Sftpd^+/+ mice (data not shown). Thus, the physiologic concentration of Ca²⁺ in the airways is likely sufficient for rhSP-D activity.

To assess the neutrophilic infiltrates induced by intratracheal LPS, Sftpd^+/+ and Sftpd^−/− mice were anesthetized using continuous inhalation of isoflurane and orally intubated with a 24-gauge animal feeding needle. A saline solution (80 μL) containing 20 μg of E coli LPS (E coli 055:B5; Sigma; St. Louis, MO) was administered intratracheally to each mouse. In the present study, the volumes of all of the solutions administered by intratracheal intubation were adjusted to 80 μL to facilitate uniform distribution in the lung²¹ and were prepared at room temperature. Mice were extubated immediately after intratracheal injection. Mice recovered from anesthesia instantly, and we did not experience any loss of mice during or after intratracheal injections. Mice were killed for analysis at 45 min, 2 h, 4 h, and 16 h after intratracheal LPS injection. No LPS was given to the 0-h group.

To study the effects of intratracheal rhSP-D on neutrophilic infiltration following intratracheal administration of E coli LPS or Staphylococcus aureus LTA, Sftpd^+/+ and Sftpd^−/− mice were administered either 20 μg of LPS or 15 μg of LTA (Sigma) in saline solution via intratracheal injection. Thirty minutes later, 40 μg of rhSP-D diluted in 80 μL of buffer was administered by intratracheal instillation. Control mice received 80 μL of buffer without rhSP-D. Mice were killed 4 h after the injection of LPS or LTA.

To determine the effects of the surfactant added to rhSP-D, mice were given either 20 μg of LPS or 15 μg of LTA by intratracheal injection. Thirty minutes later, the mice were again anesthetized, and 40 μg of rhSP-D mixed with commercial surfactant (Survanta) was administered at a concentration of 10 μg/μL. For comparison, groups received 20 μg of LPS, 15 μg of LTA, or 10 mg/mL surfactant (Survanta) in buffer without rhSP-D.

The processing of lungs, cell counts in BALF, measurement of protein levels, and proinflammatory cytokine analyses in BALF and lung homogenates were performed as described previously.²⁰ Briefly, the mice were deeply anesthetized with intraperitoneal pentobarbital sodium, and the distal aorta was cut. BAL was repeated five times with a 0.9% NaCl solution through a 20-gauge blunt needle tied into the trachea, and the recovered BALF was pooled. BALF was centrifuged at 284 g for 10 min to isolate cells. The cells in the pellets were counted using trypan blue. Differential cell counts were obtained from a stained cytospin preparation (Diff-Quick; Scientific Products; McGraw Park, IN). Proinflammatory cytokines were quantitated in BALF and in supernatants prepared from lung homogenates after centrifugation at 1,000 g for 10 min with murine sandwich enzyme-linked immunosorbent assay kits (R&D Systems; Minneapolis, MN). The total protein in aliquots of BALF from a mouse was measured by the method of Lowry et al.²² The levels of proinflammatory cytokines in BALF were all low, and in most samples were not detectable. Proinflammatory cytokine values have therefore been presented for lung homogenates only.

The results are given as the mean ± SEM. The groups were compared using two-tailed t tests or one-way analysis of variance (ANOVA) with Tukey-Kramer multiple comparisons test used for post hoc analyses. Significance was accepted at p < 0.05.

E coli LPS was administered intratracheally to Sftpd^+/+ and Sftpd^−/− mice. The total number of cells and neutrophils in BALF were increased by LPS in both Sftpd^+/+ and Sftpd^−/− mice (p < 0.05 vs no LPS 0-h group by ANOVA) and were significantly higher in Sftpd^−/− mice than in Sftpd^+/+ mice 2 h after LPS instillation and thereafter (Figs 1 , top, A, and top center, B). Although alveolar macrophage numbers in Sftpd^−/− mice were increased with chronic inflammation⁸ before LPS treatment (0-h group), intratracheal administration of LPS did not alter alveolar macrophage numbers in either Sftpd^−/− or Sftpd^+/+ mice (Fig 1, bottom center, C). The total protein level in BALF was significantly higher in Sftpd^−/− mice than in Sftpd^+/+ mice 4 and 16 h after LPS instillation, suggesting increased protein permeability in the absence of SP-D (Fig 1, bottom, D). LPS significantly increased the levels of proinflammatory mediators, such as interleukin (IL)-1β, IL-6, and macrophage inflammatory protein (MIP)-2 in the lung homogenate in both Sftpd^+/+ and Sftpd^−/− mice (p < 0.05 [vs 0-h group by ANOVA]) [Fig 2 ]. Lung IL-1β and MIP-2 levels were significantly increased in Sftpd^−/− mice compared to Sftpd^+/+ mice 16 h after intratracheal LPS injection. Since neutrophilic inflammation was consistently observed 4 h after LPS administration, subsequent studies to assess the effects of rhSP-D and surfactant were performed at this time point.

The effects of the intratracheal injection of rhSP-D were evaluated 4 h after LPS injection. The instillation of buffers or normal saline solution can be injurious to the LPS-exposed lung, and the number of neutrophils measured in BALF from Sftpd^+/+ (Fig 3 , top center, B) was significantly increased by buffer injection (p < 0.05 [for LPS/buffer group vs LPS group by t test]). In contrast, the levels of neutrophils in Sftpd^−/− mice were twofold higher than those in Sftpd^+/+ mice after the initial LPS injection (Fig 1), and no further increase was detected by the injection of additional buffer or normal saline solution. The inflammatory responses were activated in Sftpd^−/− mice by the initial inflammation induced by LPS, and further neutrophil infiltration was not observed by buffer instillation. Intratracheal rhSP-D given 30 min after LPS was administered decreased buffer-induced neutrophilic infiltration in BALF by 30% (Figs 3, top center, B, and 4 , top center, B). rhSP-D treatment did not influence macrophage numbers (Figs 3, bottom center, C, and 4, bottom center, C) or total protein levels (Figs 3, bottom, D, and 4, bottom, D) in BALF 3.5 h after the treatment of either Sftpd^+/+ or Sftpd^−/− mice. The observed decrease in cell counts in BALF after rhSP-D treatment was a reflection of the decreased numbers of neutrophils (Figs 3, top, A, and 4, top, A). The levels of IL-1β, IL-6, and MIP-2, induced by LPS and buffer injection, were significantly higher in Sftpd^−/− mice than in Sftpd^+/+ mice (p < 0.001 [by ANOVA]) [Fig 5 ].

rhSP-D was mixed with a diluted commercial surfactant (10 mg/mL), and its effects on LPS-induced lung inflammation were determined. Sftpd^+/+ and Sftpd^−/− mice received surfactant with and without rhSP-D given 30 min after intratracheal LPS injection. Surfactant treatment alone did not cause additional lung inflammation (Fig 3, 4), and the levels of proinflammatory cytokines (Fig 5) were similar to those in Sftpd^−/− and Sftpd^+/+ mice that had not been injected with LPS or LTA (0-h group in Figs 1, 2). Surfactant enhances the antiinflammatory effects of rhSP-D, further decreasing the number of neutrophils in BALF by 50 to 80% compared to levels in the LPS/buffer-injected Sftpd^+/+ and Sftpd^−/− mice (Figs 3, top center, B, and 4, top center, B). LPS did not strongly influence lung cytokine levels in Sftpd^+/+ mice 4 h after LPS injection (Fig 2); rhSP-D did not alter cytokine levels in LPS-treated Sftpd^+/+ mice (Fig 5). The mixture of rhSP-D and surfactant significantly decreased the levels of IL-1β, IL-6, and MIP-2 (Fig 5), and was more effective than rhSP-D alone. Thus, the addition of surfactant to rhSP-D further decreased LPS-induced lung inflammation.

Gram-positive bacterial LTA was also instilled intratracheally in Sftpd^−/− and Sftpd^+/+ mice. As shown in Figure 6 , Sftpd^−/− mice were more susceptible to LTA-induced lung inflammation. The administration of surfactant with rhSP-D was less effective in suppressing LTA-induced inflammation than LPS-induced inflammation.

Systemic and pulmonary infection caused by Gram-negative bacteria are commonly associated with ALI.²³ In the present study, intratracheal LPS caused neutrophilic pulmonary inflammation, and alveolar epithelial injury associated with increased protein permeability and increased levels of proinflammatory mediators, which is consistent with previous findings.²³²⁴ Likewise, LTA, a component of the cell wall of Gram-positive bacteria, triggers lung inflammation and causes neutrophil influx into the lung.⁵²⁵²⁶ Since SP-D binds both LPS and LTA in vitro,⁴⁵⁶ we determined whether rhSP-D inhibited LPS-mediated and LTA-mediated lung inflammation. The administration of intratracheal rhSP-D was effective in decreasing lung inflammation induced by the pulmonary administration of LPS or LTA in both Sftpd^−/− and Sftpd^+/+ mice. The addition of an exogenous surfactant (Survanta) to rhSP-D further inhibited LPS-induced and LTA-induced lung inflammation.

SP-D binds preferentially to monosaccharides.²⁷ Microbial ligands are rich in carbohydrates, and both LTA and LPS polysaccharides consist of repeating monosaccharides that are recognized by the carbohydrate recognition domain of SP-D.⁴²³ The structure of SP-D is critical for its function. Purified rhSP-D migrated as a mixture of dodecamers and multimers, which was demonstrated by atomic force microscopy (data not shown). Similar to native SP-D, rhSP-D protein migrated as a trimer under nonreducing conditions on sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels.¹⁹ This preparation of rhSP-D prevented intratracheal endotoxin-induced septic shock in premature newborn lambs.¹⁹ Thus, both in vivo and in vitro studies support the concept that rhSP-D is biologically active, in a manner similar to that of native SP-D.

Although advances in the ventilatory management of patients with ALI improved outcomes, ALI is still a common cause of morbidity and mortality following pulmonary or systemic infections. Various pharmacologic therapies, including surfactant, inhaled nitric oxide, corticosteroids, antifungal drugs, and phosphodiesterase inhibitors, have not been clinically effective.²⁸ Exogenous surfactant administration in patients with ALI was tested 20 years ago.²⁹ Among many clinical trials, there are only two randomized controlled trials³⁰³¹ that have succeeded in improving both oxygenation and mortality in patients with ALI. The commercial replacement surfactants used for the treatment of RDS and ALI do not contain SP-D or SP-A. Pulmonary surfactant has unique biophysical properties, including its rapid surface adsorption, spreading, and film formation, which are critical for lung function. With its high surface activity, surfactant enhances the spreading of solutions in airways and on alveolar surfaces. Surfactant (Survanta) is routinely used for the treatment of premature infants with RDS. Since surfactant levels are increased fourfold to fivefold in Sftpd^−/− mice,⁸ we utilized a dose that was previously shown to improve the spreading of intratracheally instilled adenovirus,¹⁷ using a concentration (10 mg/mL) that is considerably less than the usual concentration used for the treatment of RDS (25 mg/mL). In the present study, the addition of surfactant (Survanta) to rhSP-D improved the efficacy of rhSP-D in the suppression of lung inflammation induced by LPS and LTA. The surfactant also reduced the lung injury caused by the instillation of buffer given to the LPS-exposed or LTA-exposed lung. There are at least two possible mechanisms by which surfactant enhanced the antiinflammatory effects of rhSP-D. One is that the surfactant enhanced the distribution of the rhSP-D in the lung.¹⁷ The second is that the surfactant directly inhibits lung inflammation induced by LPS.¹⁵ Previous studies¹⁶³² have demonstrated that surfactant (Survanta) blocked LPS signaling and inhibited proinflammatory cytokine secretion in human alveolar macrophages in vitro. The mixture of rhSP-D and surfactant has additive effects on the treatment of pulmonary inflammation, suggesting that surfactant and rhSP-D have independent biological effects.

The severity and time course of lung inflammation are dependent on the dose of intratracheal LPS or LTA. The doses of LPS or LTA used in the present study were based on preliminary studies with 0.5 to 2 μg per gram of body weight LPS and 0.1 to 4 μg per gram of LTA (data not shown), and are consistent with the doses used in previous studies.²⁴³³³⁴ The doses of LPS or LTA used in the present study induced neutrophilic inflammation within 4 h after administration. The dose of rhSP-D used in the present study represents a dose of 1.6 mg/kg. The normal pool size of SP-D in the mouse lung has been estimated to be 0.25 mg/kg. In a previous study,¹⁹ the administration of 2 mg/kg rhSP-D was effective in rescuing the preterm newborn lamb from LPS-induced shock. This dose is similar to the content of SP-D that is present in the term newborn lamb lung. In previous studies,³⁵ examining the clearance of exogenous radiolabeled SP-D from the alveoli showed that 30% of exogenous SP-D was recovered in BALF 4 h after instillation. Thus, the levels of rhSP-D are increased approximately twofold compared to normal adult SP-D levels 4 h after intratracheal injection, indicating that the dose of SP-D used in the present study is physiologically relevant.

Pulmonary diseases, including cystic fibrosis, bronchopulmonary dysplasia, and COPD, are associated with decreased SP-D concentrations in the lung as well as an increased risk of pulmonary infection.^7121314 We have shown that the intratracheal injection of rhSP-D, particularly when mixed with surfactant, reduced the severity of LPS-induced or LTA-induced inflammation, both in Sftpd^+/+ and Sftpd^−/− mice, indicating that rhSP-D may be protective against lung inflammation caused by Gram-negative and Gram-positive pathogens in vivo.

The authors thank Karen Carter, Kimberly Bishop, Annuradha Yadav, and Elizabeth Masterjohn for their expert technical support.