Gary F. Ross

Flagellin, a Novel Mediator of Salmonella-Induced Epithelial Activation and Systemic Inflammation: IκBα Degradation, Induction of Nitric Oxide Synthase, Induction of Proinflammatory Mediators, and Cardiovascular Dysfunction

Gram-negative sepsis is mediated by the actions of proinflammatory genes induced in response to microbes and their products. We report that flagellin, the monomeric subunit of flagella, is a potent proinflammatory species released by Salmonella. Flagellin (1 μg/ml) induces IκBα degradation, NF-κB nuclear translocation, and inducible NO synthase expression in cultured intestinal epithelial cells (IEC). Aflagellic Salmonella mutants do not induce NF-κB activation or NO production by cultured IEC. Antiserum to flagellin blocks NO production in IEC induced by medium conditioned by a variety of motile Gram-negative enteric pathogens (Escherichia coli, Salmonella muenchen, Serratia marcescens, Proteus mirabilis, and Proteus vulgaris). Flagellin, when injected systemically (∼10 μg/mouse), induces systemic inflammation characterized by the systemic expression of a range of proinflammatory cytokines and chemokines and of inducible NO synthase. At higher doses (∼300 μg/mouse), flagellin induces shock, characterized by hypotension, reduced vascular contractility in mice, and death. The effects of flagellin do not diminish in C3H/HeJ LPS-resistant mice, indicating that the Toll-like receptor-4 receptor is not involved in flagellin’s actions. In LPS-resistant mice, i.p. injection of S. dublin flagellin or medium conditioned by wild-type S. dublin induces serum IFN-γ and TNF-α, whereas medium conditioned by aflagellic mutants has no effect. Flagellin can be detected in the blood of rats with septic shock induced by live bacteria at approximately 1 μg/ml. We propose that flagellin released by Gram-negative pathogens may contribute to the inflammatory response by an LPS- and Toll-like receptor-4-independent pathway.

Altered stability of pulmonary surfactant in SP-C-deficient mice

The surfactant protein C (SP-C) gene encodes an extremely hydrophobic, 4-kDa peptide produced by alveolar epithelial cells in the lung. To discern the role of SP-C in lung function, SP-C-deficient (−/−) mice were produced. The SP-C (−/−) mice were viable at birth and grew normally to adulthood without apparent pulmonary abnormalities. SP-C mRNA was not detected in the lungs of SP-C (−/−) mice, nor was mature SP-C protein detected by Western blot of alveolar lavage from SP-C (−/−) mice. The levels of the other surfactant proteins (A, B, D) in alveolar lavage were comparable to those in wild-type mice. Surfactant pool sizes, surfactant synthesis, and lung morphology were similar in SP-C (−/−) and SP-C (+/+) mice. Lamellar bodies were present in SP-C (−/−) type II cells, and tubular myelin was present in the alveolar lumen. Lung mechanics studies demonstrated abnormalities in lung hysteresivity (a term used to reflect the mechanical coupling between energy dissipative forces and tissue-elastic properties) at low, positive-end, expiratory pressures. The stability of captive bubbles with surfactant from the SP-C (−/−) mice was decreased significantly, indicating that SP-C plays a role in the stabilization of surfactant at low lung volumes, a condition that may accompany respiratory distress syndrome in infants and adults.

Hydrophobic Surfactant-Associated Protein in Whole Lung Surfactant and Its Importance for Biophysical Activity in Lung Surfactant Extracts Used for Replacement Therapy

ABSTRACT. Hydrophobic protein of 6,000 and 14,000 daltons was isolated from mammalian pulmonary surfactant obtained from canine, human, and bovine alveolar lavage material. Low molecular weight, hydrophobic, surfactant- associated protein (SAP), herein referred to as SAP 6-14, was distinguished from SAP-35, the major glycoprotein in mammalian surfactants (the 35,000 dalton glycoprotein A or apolipoprotein A) by amino acid composition, peptide mapping, and by resistance of SAP 6-14 to digestion by endoglycosidase F, collagenase, trypsin, and other proteases. The amino acid composition of SAP 6-14 was found to be highly enriched in leucine and other hydrophobic amino acids. The characteristics of protein isolated from bovine replacement surfactant extracts utilized for the treatment of hyaline membrane disease in humans were also studied. SAP 6-14 isolated from calf lung surfactant replacement extracts (CLSE) and surfactant- TA were found to be identical to SAP 6-14 isolated from ether/ethanol extracts of various mammalian surfactants. By contrast, SAP-35, the major surfactant-associated glycoprotein of molecular weight=35,000, and other higher molecular weight proteins were not detected in significant quantities in the CLSE or surfactant-TA replacement surfactants, either by highly sensitive silver stain analysis or by immunoblot using monospecific antisera generated against bovine SAP-35. Biophysical studies of the CLSE replacement surfactant containing only SAP 6-14 and native phospholipids demonstrated full surface activity compared to natural lung surfactant. Dynamic surface tension lowering and adsorption properties of CLSE were essentially identical to those of freshly isolated bovine whole surfactant. Thus, hydrophobic SAP 6-14 is the only protein detected in bovine lung extract surfactants with full biophysical activity. The major surfactant associated protein, SAP-35, was not a significant component of either the CLSE or surfactant-TA replacement preparations.

Granulocyte-macrophage colony-stimulating factor in the innate immune response to Pneumocystis carinii pneumonia in mice.

Innate immunity plays an important role in pulmonary host defense against Pneumocystis carinii, an important pathogen in individuals with impaired cell-mediated immunity. We investigated the role of GM-CSF in host defense in a model of P. carinii pneumonia induced by intratracheal inoculation of CD4-depleted mice. Lung GM-CSF levels increased progressively during the infection and were significantly greater than those in uninfected controls 3, 4, and 5 wk after inoculation. When GM-CSF gene-targeted mice (GM−/−) depleted of CD4+ cells were inoculated with P. carinii, the intensities of infection and inflammation were increased significantly compared with those in CD4-depleted wild-type mice. In contrast, transgenic expression of GM-CSF directed solely in the lungs of GM−/− mice (using the surfactant protein C promoter) dramatically decreased the intensity of infection and inflammation 4 wk after inoculation. The concentrations of surfactant proteins A and D were greater in both uninfected and infected GM−/− mice compared with those in wild-type controls, suggesting that this component of the innate response was preserved in the GM−/− mice. However, alveolar macrophages (AM) from GM−/− mice demonstrated impaired phagocytosis of purified murine P. carinii organisms in vitro compared with AM from wild-type mice. Similarly, AM production of TNF-α in response to P. carinii in vitro was totally absent in AM from GM−/− mice, while GM-CSF-replete mice produced abundant TNF in this setting. Thus, GM-CSF plays a critical role in the inflammatory response to P. carinii in the setting of impaired cell-mediated immunity through effects on AM activation.

Immunologic identification of a pulmonary surfactant-associated protein of molecular weight = 6000 daltons.

ABSTRACT. Hydrophobic, small molecular weight, surfactant- associated protein of Mr = 6000 (SAP-6) was isolated from bovine, canine, and human alveolar lavage and identified by silver staining after sodium dodecyl sulfate polyacrylamide gel electrophoresis gels. Lesser amounts of protein of Mr = 14,000, 20,000, and 26,000 daltons also copurified with SAP-6, likely representing oligomers of the Mr = 6,000 dalton protein. In the absence of sulfhydryl-reducing agents, increased amounts of the larger forms of the protein were observed. Antisera generated against bovine SAP-6 were used to further characterize the protein and distinguish it from the more abundant surfactant-associated glycoprotein of Mr = 35,000 (SAP- 35) present in mammalian surfactants. Rabbit antisera generated against the bovine hydrophobic protein recognized SAP-6 and lesser amounts of the proteins of Mr = 14,000, 20,000, and 26,000 daltons. The SAP-6 antisera were reactive against the hydrophobic proteins from human, bovine, and canine surfactants as assessed by immunoblot analysis after sodium dodecyl sulfate polyacrylamide gel electrophoresis. SAP-6 antisera did not detect bovine SAP-35 the abundant surfactant-associated glycoprotein, by immunoblot analysis; however, some reactivity of the anti-SAP-6 was detected against purified bovine SAP-35 by a sensitive enzyme-linked immune-adsorbant assay. Anti-SAP-6-did not react with bovine serum components either by immunoblot or by enzyme-linked immune- adsorbant assay. Monospecific antisera generated against bovine SAP-35 did not detect SAP-6 by immunoblot analysis. Immunoblot analysis of the protein in exogenous replacement surfactant preparations, Surfactant-TA and calf lung surfactant extract, utilizing anti-SAP-6 and anti-SAP-35 demonstrated that both preparations were enriched in SAP-6 and lacking in SAP-35. Surfactant-associated proteins, present in bovine lung-based replacement surfactants, are immunologically identified as SAP-6 and its oligomers.

Characteristics of Human Surfactant-Associated Glyoproteins A

ABSTRACT: Surfactant-associated glycoprotein A [molecular weight (Mr) = 34,000, isoelectric point (pi) 4.6-5.0] and its sulfhydryl dependent oligomers were purified and partially characterized from surfactant obtained from human alveolar lavage. Two major forms of the protein were identified by silver stain and immunoblot analysis of surfactant using human surfactant-associated glycoprotein A antisera: glycoprotein A2, Mr = 34,000 and glycoprotein Ai, Mr = 28,000. The larger form was reduced to Mr = 28,000 by treatment with endogylcosidase F, indicating the presence of complex N-linked oligosaccharide on the molecule. Charge heterogeneity was decreased and the isoelectric point increased by treatment with neuroaminidase, supporting the presence of sialic acid. Homology between the proteins Mr = 34,000 and 28,000 was confirmed by analysis of two-dimensional tryptic and chymotryptic peptides of 125I-iodo-glycoproteins Ai and A2 which were identical. The protein was very rich in glycine and its amino acid composition was similar to that of glycoprotein A previously reported for the dog and rat. Treatment of glycoproteins A with bacterial collagenase resulted in the generation of highly glycosylated peptides Mr = 20,000- 22,000, pi 4.6-5.0, which no longer formed sulfhydryldependent oligomers, supporting the presence of significant collagen-like region in the molecule. In the absence of reducing agents, glycoprotein A from surfactant was present as sulfhydryl-dependent dimers and larger oligomers. Higher molecular weight aggregates of glycoproteins A were also present in lavage material even after sulfhydryl reduction. Glycoproteins A were identified in surfactant from amniotic fluid, normal adult lung lavage, human cadaver lung lavage, and material obtained from lung lavage from a patient with alveolar proteinosis. Alveolar proteinosis proteins contained larger amounts of the higher molecular weight aggregates and smaller molecular weight proteolytic fragments of glycoproteins A than material obtained from other sources. Peptide mapping of the I25Iiodinated aggregates, approximately Mr = 50,000, 70,000, 100,000, and greater were identical to glycoproteins A (Mr = 34,000 and 28,000) from normal human lung lavage. A smaller immunoreactive form (Mr = 20,000) shared several peptides but lacked others, supporting its origin as a proteolytic fragment of glycoproteins Ai or A2. Human glycoprotein A2 is a complex N-linked glycoprotein likely representing the glycosylated form of a polypeptide precursor Mr = 28,000.

Macrophage Dysfunction and Susceptibility to Pulmonary Pseudomonas aeruginosa Infection in Surfactant Protein C-Deficient Mice

To determine the role of surfactant protein C (SP-C) in host defense, SP-C-deficient (Sftpc−/−) mice were infected with the pulmonary pathogen Pseudomonas aeruginosa by intratracheal injection. Survival of young, postnatal day 14 Sftpc−/− mice was decreased in comparison to Sftpc+/+ mice. The sensitivity to Pseudomonas bacteria was specific to the 129S6 strain of Sftpc−/− mice, a strain that spontaneously develops interstitial lung disease-like lung pathology with age. Pulmonary bacterial load and leukocyte infiltration were increased in the lungs of Sftpc−/− mice 24 h after infection. Early influx of polymorphonuclear leukocytes in the lungs of uninfected newborn Sftpc−/− mice relative to Sftpc+/+ mice indicate that the lack of SP-C promotes proinflammatory responses in the lung. Mucin expression, as indicated by Alcian blue staining, was increased in the airways of Sftpc−/− mice following infection. Phagocytic activity of alveolar macrophages from Sftpc−/− mice was reduced. The uptake of fluorescent beads in vitro and the number of bacteria phagocytosed by alveolar macrophages in vivo was decreased in the Sftpc−/− mice. Alveolar macrophages from Sftpc−/− mice expressed markers of alternative activation that are associated with diminished pathogen response and advancing pulmonary fibrosis. These findings implicate SP-C as a modifier of alveolar homeostasis. SP-C plays an important role in innate host defense of the lung, enhancing macrophage-mediated Pseudomonas phagocytosis, clearance and limiting pulmonary inflammatory responses.

Surfactant protein C in fetal and ventilated preterm rabbit lungs

The developing lung contains surfactant protein (SP) C mRNA levels comparable to term values before mature type II cells and alveolar surfactant lipids are detectable. Estimates of the amount of mature SP-C in the alveolar lavages of preterm lungs are not available. We used an antibody to a recombinant human SP-C to measure the amount of SP-C in alveolar lavages of preterm fetal rabbits, ventilated preterm rabbits, and term rabbits. The amounts of SP-C were compared with the amounts of saturated phosphatidylcholine (Sat PC). Median Sat PC amounts increased about 680-fold, and median SP-C values increased by over 5,000-fold in alveolar washes from 27 days gestation to term. There was no increase in Sat PC or SP-C with ventilation at 27 and 28 days gestation, but ventilation increased both Sat PC and SP-C at 29 days gestation. The molar percent of SP-C relative to Sat PC also increased with gestational age and with ventilation at 29 days gestation. proSP-C was abundant in a membrane fraction from lung tissue at 27 and 28 days gestation when minimal mature SP-C was detected in alveolar washes. At 29 days and at term, proSP-C decreased in membrane fractions. The preterm lung that is surfactant lipid deficient is also severely deficient in mature SP-C.The developing lung contains surfactant protein (SP) C mRNA levels comparable to term values before mature type II cells and alveolar surfactant lipids are detectable. Estimates of the amount of mature SP-C in the alveolar lavages of preterm lungs are not available. We used an antibody to a recombinant human SP-C to measure the amount of SP-C in alveolar lavages of preterm fetal rabbits, ventilated preterm rabbits, and term rabbits. The amounts of SP-C were compared with the amounts of saturated phosphatidylcholine (Sat PC). Median Sat PC amounts increased about 680-fold, and median SP-C values increased by over 5,000-fold in alveolar washes from 27 days gestation to term. There was no increase in Sat PC or SP-C with ventilation at 27 and 28 days gestation, but ventilation increased both Sat PC and SP-C at 29 days gestation. The molar percent of SP-C relative to Sat PC also increased with gestational age and with ventilation at 29 days gestation. proSP-C was abundant in a membrane fraction from lung tissue at 27 and 28 days gestation when minimal mature SP-C was detected in alveolar washes. At 29 days and at term, proSP-C decreased in membrane fractions. The preterm lung that is surfactant lipid deficient is also severely deficient in mature SP-C.

1780 HUMAN SURFACTANT ASSOCIATED APOLIPOPROTEIN A: MOLECULAR COMPOSITION AND PRIMARY TRANSLATION PRODUCTS

Surfactant apolipoprotein A (apo A) is a major lipid binding protein and an active component of pulmonary surfactant. Apo A was purified from human alveolar lavage, amniotic fluid and alveolar proteinosis fluid. Two major forms were identified by silver stain and immunoblot: apo A2 (Mr=34,000), apo A1 (Mr=28,000). Larger forms were reduced to Mr=28;000 by treatment with endoglycosidase F demonstrating complex N-linked oligosaccharide. Isoelectric point was increased by neuraminidase demonstrating presence of sialic acid. Poly A mRNA isolated from adult human lung was translated in vitro; primary translation products were identified at Mr=28,000. Homology between Mr=34,000 and 28,000 was confirmed by analysis of 2D tryptic peptide maps of apo A1 and A2 which were identical, thus providing evidence that apo A results from processing of the Mr=28,000 precursor by addition of carbohydrate. Amino acid composition of purified apo A was rich in glycine and contained a large portion of collagen-like sequence. Apoproteins A were identified in surfactant from amniotic fluid, normal adult lung lavage, human cadaver lavage and material obtained from a patient with alveolar proteinosis. Apo A was identified in all of these samples by silver stain and immunoblot analysis. Alveolar proteinosis fluid contained acidic aggregates whose peptide maps were identical to apo A from normal human lung lavage. These studies clarify the identity of human apo A, a complex N-linked glycoprotein derived from a polypeptide of Mr=28,000.