A comparative study of secretory immunoglobulin A and immunoglobulin G in host defense in an in vitro pneumonia model.

BACKGROUND: An exaggerated inflammatory response to infections including nosocomial pneumonia may be detrimental to the host and contribute to morbidity and mortality. Both secretory immunoglobulin A (SIgA) and IgG contribute to the immune defense of respiratory surfaces. However, their relative ability to protect against invasive infections and the resultant host inflammatory response are unknown and were the basis for this study. METHODS: Calu-3 cell (a respiratory epithelial cell line) monolayers were established in a Transwell system (Costar Corp., Cambridge, MA). Escherichia coli and either polyclonal SIgA or IgG were inoculated into the apical chamber and neutrophils (polymorphonuclear neutrophils) were added to the basal chamber. PMN cytotoxic potential was indexed by %CD11b expression, superoxide anion (O(2)-) production, and % elastase release. Bacterial translocation into the basal chamber was quantitated after log transformation. Calu-3 monolayer integrity was indexed by permeability to dextran fluorescein isothiocyanate. RESULTS: The addition of E coli to Calu-3/polymorphonuclear cocultures led to increases in O(2)- generation, elastase release, and CD11b expression. These effects were diminished by the addition of SIgA but not IgG. A similar effect was noted with Calu-3 barrier function. CONCLUSIONS: SIgA may function to protect against microbial invasion of respiratory surfaces and protect against tissue injury from an exaggerated inflammatory response.

Immunoglobulin a modulates inflammatory responses in an in vitro model of pneumonia.

BACKGROUND: Preservation of mucosal immunity has been shown to affect the risk and outcome of pneumonia in severely injured patients. Secretory immunoglobulin A (SIgA) is the principle humoral defense of mucosal surfaces in the body and has several antiinflammatory properties. Polymorphonuclear neutrophils (PMN) function to kill invading microorganisms, but their exaggerated inflammatory responses may cause tissue injury to the host. The purpose of this study was to compare the ability of different immunoglobulin (Ig) isotypes to modulate PMN cytotoxic potential cocultured with respiratory epithelial cells challenged with bacteria. METHODS: Calu-3 cell monolayers were established on membranes (0.1-microm pore) in a two-chamber culture system. Escherichia coli (EC) incubated with either polyclonal SIgA or IgG was inoculated into the apical chamber and PMNs (10/mL) added to the basal chamber. PMN cytotoxic potential was indexed by % CD11b expression, superoxide anion (O2) production, and elastase release. Dextran flux was used to index Calu-3 monolayer permeability. RESULTS: Addition of EC to PMN-Calu-3 cell coculture increased % CD11b expression, O2 production, and elastase release. IgG had no effect on PMN activation after EC challenge. SIgA abrogated PMN activation and the increase in Calu-3 cell monolayer permeability noted with EC or EC + IgG treatment groups. CONCLUSION: PMN cytotoxic potential was decreased by the presence of SIgA but not IgG in an in vitro model to simulate pneumonia in vivo. SIgA may not only function to protect against microbial invasion of mucosal surfaces, but may also protect against tissue injury from an exaggerated inflammatory response.

Intestinal epithelial cells modulate PMN activation and apoptosis following bacterial and hypoxic challenges.

BACKGROUND: The post-ischemic gut may serve to prime and activate neutrophils which may lead to the subsequent development of the systemic inflammatory response syndrome (SIRS) and multiple organ failure. However, the initiating event which may trigger this immunoinflammatory cascade from the gut is unknown. Recent studies have indicated that intestinal epithelial cells (IEC) play an integral role in generating and transmitting signals between luminal bacteria and the host cells in the underlying gut tissues. The purpose of this study was to investigate the ability of IEC to modulate PMN responses to bacteria and/or hypoxia/reoxygenation (H/R) challenges in vitro. METHODS: Caco2 cell monolayers were established in a two-chamber cell culture system. Neutrophils from normal human volunteers were placed in the basal chamber and the cell co-culture exposed to either apical bacteria (E. coli) and/or H/R challenge. PMN apoptosis, and percentage of CD11b expression, superoxide anion production, and elastase release were subsequently quantitated. RESULTS: Coculture of PMNs with Caco2 cells led to a significant reduction in neutrophil apoptosis in both normoxic and H/R conditions. CD11b expression was increased in PMNs exposed to bacteria but the greatest expression was noted with PMN cocultured with Caco2 cells and H/R. Superoxide anion production was increased in all groups following either H/R or bacterial challenge and H/R. Elastase release was highest in neutrophils following H/R and exposure to E. coli. CONCLUSION: IEC modulate PMN response to bacteria and H/R insults. This results in the production of activated neutrophils with an exaggerated lifespan which may promote remote organ failure. Attempts to modulate this response may be useful in preventing multiple organ failure following severe traumatic shock.

The pivotal role of tumor necrosis factor-alpha in signaling apoptosis in intestinal epithelial cells under shock conditions.

BACKGROUND: Apoptosis is essential for the regulation of cell number and function of intestinal epithelial cells but may contribute to intestinal barrier failure after shock and other low-flow conditions to the gut. METHODS: Caco2 intestinal cell monolayers were challenged with recombinant tumor necrosis factor (TNF). In a second group of experiments, Caco2 cells were exposed to bacteria and/or hypoxia followed by reoxygenation. Apoptosis was detected using annexin-V propidium-iodide staining. Cell culture supernatants were also obtained in the second group of experiments and TNF levels quantitated. Monolayer integrity was assessed by measurement of paracellular permeability and transepithelial electrical resistance. RESULTS: Apical but not basal recombinant TNF increased Caco2 apoptosis. Exposure to either bacteria alone or hypoxia/reoxygenation alone did not increase apoptosis; however, the combined insults significantly increased apoptosis. The increased apoptosis occurred in a delayed fashion in both groups. TNF was released in a polar fashion, and the greatest levels were noted after exposure to both bacteria and hypoxia-reoxygenation. There was also an increase in paracellular permeability in this group; however, no change in transepithelial electrical resistance was noted. The effects on apoptosis and permeability were abrogated by anti-TNF antibodies. CONCLUSION: Intestinal epithelial cell apoptosis contributes to barrier failure after shock conditions and is related to augmented TNF release.

Enterocyte apoptosis and barrier function are modulated by SIgA after exposure to bacteria and hypoxia/reoxygenation.

BACKGROUND: Secretory immunoglobulin A (SIgA) is the principal immune defense against luminal pathogens at gut mucosal surfaces. It also has anti-inflammatory activities that may be important for the maintenance of mucosal surface integrity. Enterocyte apoptosis (Apo) is increased after challenge with invasive bacteria and ischemia-reperfusion insults. Increased Apo also has been associated with impaired intestinal barrier function. However, the impact of SIgA on enterocyte apoptosis and mucosal barrier integrity after challenge with commensal bacteria and ischemia-reperfusion is unknown. METHODS: Caco2 intestinal epithelial cell monolayers were subjected to 21% O(2) (control) or 95% N(2)/15% CO(2) (hypoxic) conditions for 90 minutes, followed by 21% O(2). Escherichia coli and SIgA were added in subsets. Caco2 cell Apo was identified by flow cytometry and barrier function indexed by permeability to dextran-fluorescein isothiocyanate. RESULTS: There were no differences in the percentage of Apo Caco2 cells after exposure to either bacteria or hypoxic-reoxygenation versus control. There was a significant increase in Apo after the combined bacteria/hypoxia-reoxygenation challenge. SIgA abrogated the Apo response and preserved barrier function after this combined challenge. CONCLUSION: Modulation of enterocyte Apo by SIgA may serve to maintain intestinal barrier function and thereby decrease the systemic inflammatory response after clinical conditions associated with gut ischemia-reperfusion insults.

The synergistic effects of hypoxia/reoxygenation or tissue acidosis and bacteria on intestinal epithelial cell apoptosis.

BACKGROUND: Clinical data indicate that gut perfusion deficits must be rectified within 24 hours after traumatic injury to decrease organ failure and death. Ischemia/reperfusion injury to the gut causes enterocyte apoptosis (Apo), which may contribute to intestinal barrier failure. The temporal response of enterocyte Apo to acidosis and hypoxia/reoxygenation (H/R) in vitro is unknown. The purpose of this study was to examine the effect of various time points of acidosis or H/R on enterocyte apoptosis and monolayer integrity in an in vitro model. METHODS: Caco-2 cell monolayers were made acidic (Dulbecco's modified Eagle's medium, pH 6.9) by hydrochloric acid or exposed to 95% nitrogen/5% carbon dioxide (hypoxia) and then 21% oxygen (reoxygenation). Escherichia coli C-25 were added to the apical media in subsets. Apo and necrosis were quantified by flow cytometry. Permeability was determined by fluorescein isothiocyanate-dextran. Transepithelial electrical resistance (TEER) indexed monolayer. RESULTS: Extracellular acidosis and C-25 significantly increased apoptosis of Caco-2 cells at 18 hours (extracellular acidosis [EC] + C-25, 14.5 +/- 3.0; control, 3.8 +/- 0.8; p < 0.001 by analysis of variance). Similarly, the H/R + C-25 group showed a significant increase in apoptosis at 12 hours (H/R + C-25 vs. control, 22.86 +/- 2.12 vs. 3.74 +/- 0.7; p < 0.001 by analysis of variance). The permeability difference was not significant for EC + C-25 versus control at 18 hours (0.68 +/- 0.25 vs. 0.43 +/- 0.0.0.36, respectively; p > 0.05). The H/R + C-25 group had a profound increase in permeability over control at 12 hours (10.8 +/- 0.5 vs. 2.1 +/- 0.3, respectively; p < 0.001). The TEER was significantly lowered for EC versus control at 18 hours (458 +/- 1.5 vs. 468 +/- 8.2) and at 0, 6, and 18 hours for EC + C-25 (409 +/- 28.1, 443 +/- 16.8, and 438 +/- 8.9 vs. 455 +/- 6.5, 467 +/- 6.5, and 469 +/- 8.2, respectively). There was no significant change in the H/R and H/R + C-25 groups. CONCLUSION: Synergism of H/R or tissue acidosis and bacteria caused increased Apo, TEER, and permeability in vitro.

The effects of varying oxygen conditions and immunoglobulin A on barrier defense to bacterial invasion.

Tissue oxygenation is a critical factor in host defense against bacteria. Gut mucosal tissue oxygenation (partial pressure of O2) is normally low putting the gut at risk of invasion by luminal microbes. Secretory immunoglobulin (Ig) A (sIgA) is the principal immune defense at mucosal surfaces. The protective effect of IgA under low oxygen conditions is unknown. We studied the interaction of varying O2 environments and sIgA on protection against bacterial invasion in our in vitro model. Cell monolayers of Madin-Darby canine kidney (MDCK) cells transfected with the cDNA for polymeric immunoglobulin receptor were established in a two-chamber cell culture system. A commensal strain of Escherichia coli (10(8) colony-forming units) was added to the apical medium and cell cultures were placed in either a 5, 21, or 95 per cent O2 environment at 37 degrees C. Polyclonal sIgA (100 microg/mL) was added to the apical chamber in subsets. Basal medium was sampled at intervals and bacterial translocation quantitated. The cell monolayers of MDCK transfected cells then had 100 microg/mL IgA added to the basal compartment at 4 degrees C for 2 hours followed by various oxygen environments for 90 minutes. Afterwards apical medium was removed at one, 3, and 12 (overnight) hours. The bacterial translocation data showed a significance increase in translocation with hypoxia. Both increased oxygen and IgA abrogated these effects significantly. The transcytosis of IgA was increased during hypoxic conditions. Normal and hyperoxic conditions did not produce any significant difference in IgA transcytosis. We conclude that O2 and sIgA are protective against bacterial invasion at epithelial surfaces. Effects to either boost O2 delivery to the gut or enhance mucosal IgA production and delivery may be protective in the critically ill surgical patient.

Alternatively spliced FGFR-1 isoform signaling differentially modulates endothelial cell responses to peroxynitrite.

Mounting experimental evidence has suggested that the trophic environment of cells in culture is an important determinant of their vulnerability to the cytotoxic effects of reactive oxidants such as peroxynitrite (ONOO(-)). However, acidic fibroblast growth factor (FGF-1)-induced signaling renders some cells more sensitive and others resistant to the cytotoxic effects of ONOO(-). To determine whether alternatively spliced fibroblast growth factor receptor (FGFR-1) isoforms are responsible for this differential response, we have stably transfected FGFR-negative rat brain-derived resistant vessel endothelial cells (RVEC) with human cDNA sequences encoding either FGFR-1 alpha or FGFR-1 beta. FGF-1 treatment of RVEC(R-1 alpha) transfectants enhanced ONOO(-)-mediated cell death in a manner dependent upon FGFR-1 tyrosine kinase, MEK/Erk 1/2 kinase, and p38 MAP kinase activities and independent of Src-family kinase (SFK) activity. FGF-1 treatment of RVEC(R-1 beta) transfectants inhibited the cytotoxic effects of ONOO(-) in a manner dependent upon FGFR-1 tyrosine kinase, MEK/Erk 1/2 kinase, and SFK activities and independent of p38 MAP kinase activity. FGF-1-induced preactivation of both FGFR-1 tyrosine and Erk 1/2 kinases was detected in both RVEC(R-1 alpha) and RVEC(R-1 beta) transfectants. FGF-1-induced preactivation of p38 MAPK was restricted to RVEC(R-1 alpha) transfectants, whereas, ligand-induced preactivation of SFK was limited to RVEC(R-1 beta) transfectants. Collectively, these results both reemphasize the role of extracellular trophic factors and their receptor-mediated signaling pathways during cellular responses to oxidant stress and provide a first indication that the alternatively spliced FGFR-1 isoforms induce differential signal transduction pathways.

Ethanol impairs intestinal barrier defense by modulation of immunoglobulin A transport.

BACKGROUND: Increased epithelial permeability may contribute to septic complications after alcohol intoxication. Prior ethanol (EtOH) exposure leads to structural and functional effects on cytoskeletal components important in maintaining barrier integrity. The cytoskeleton is also important in the transcellular transport of proteins including the polymeric immunoglobulin receptor (pIgR) mediated transport of immunoglobulin A (IgA) to the luminal surface of mucosal tissues. We (1) investigated the effects of acute EtOH exposure on cell monolayer barrier integrity and IgA transport and (2) assessed the importance of microtubule function on these functions in vitro. METHODS: Confluent Madin-Darby canine kidney (MDCK) epithelial cells transfected with pIgR cDNA were exposed to 1% or 5% EtOH, and transcytosis of dimeric IgA was measured. Paclitaxel was used to pretreat a subset of MDCK cells. Results. EtOH exposure resulted in a concentration and time dependent decrease in IgA transcytosis. This was associated with physical derangements in the integrity of the cell monolayer. Pretreatment with paclitaxel completely abrogated these effects. CONCLUSIONS: EtOH causes both physical and immune derangement of epithelial cell barrier function. Decreased secretory IgA at mucosal surfaces may initiate septic complications after EtOH intoxication.

Actin mediates secretory immunoglobulin A transport: effect of ethanol.

Secretory immunoglobulin A (IgA) is the principle antibody protecting against pathogens at mucosal sites. Ethanol (EtOH) exposure is related to adverse effects on the enterocyte cytoskeleton. The aim of this study was to assess the role of normal cytoskeletal function on IgA transcytosis and its modulation by EtOH by studying Madin-Darby canine kidney (MDCK) cells transfected with the polyimmunoglobulin receptor. MDCK cells were grown as confluent monolayers and treated with 5 per cent EtOH, cytochalasin D (Cyto-D, a cytoskeletal destabilizer), or pretreatment with prostaglandin E2 (a cytoskeletal stabilizer) followed by EtOH. Media alone served as control. IgA was then added to the basolateral side of the chambers, and apical samples were taken for enzyme-linked immunosorbent assay analysis at 0, 3, and 12 hours. Dimeric IgA transcytosis increased in all groups and was significantly depressed by 5 per cent EtOH and Cyto-D. Morphological slides revealed aggregation of actin after Cyto-D treatment. Prostaglandin E2 prevented the decrease in IgA transcytosis seen otherwise with 5 per cent EtOH treatment. We conclude that IgA transcytosis is dependent on actin microfilaments of the cytoskeleton. Decreased IgA transport may lead to mucosal immunodeficiency and infectious complications after EtOH exposure.

Phosphatidylinositol 3-kinase may mediate isoproterenol-induced vascular relaxation in part through nitric oxide production.

Phosphatidylinositol 3-kinase (PI3-K) has been shown to mediate insulin and insulin-like growth factor-1 (IGF-1)-induced nitric oxide (NO) generation and, thus, vascular tone. A role for PI3-K in G-protein-coupled receptor signal transduction has also been reported. As beta2 -adrenergic vascular actions are partly dependent on NO, this study the role of PI3-K on in vitro isoproterenol (Iso)-induced endothelial cell (EC) nitric oxide synthase (NOS) activation and rat aortic vascular relaxation. Cell lysates of rat aortic EC (RAEC), exposed to Iso (10 micromol/L) for 5 minutes, were immunoprecipitated with an antiphosphotyrosine antibody prior to assay for Western blot for the p85-kd regulatory subunit of PI3-K. Endothelial NOS activity was determined by measuring nitrite production. Endothelium-intact aortic rings from male Wistar rats were preincubated with the PI3-K inhibitors, wortmannin (WT), or LY294002 (LY), precontracted with phenylepinephrine (PE), and relaxation to graded doses of Iso was measured. NO contribution to vascular relaxation was assessed by L-N(G)-nitroarginine methyl ester (L-NAME), a NOS inhibitor. Both Iso and IGF-1 induced an increase in p85 subunit phosphorylation as demonstrated by Western analysis, effects inhibited by preincubation with WT. Iso also enhanced association of p85 with the Triton X-100-insoluble fraction of RAEC, reflecting translocation of this enzyme to a cytoskeletal fraction. In addition, Iso as well as IGF-1 significantly increased eNOS activity measured by nitrite production. Both WT and LY markedly inhibited relaxation to Iso, while L-NAME nearly abolished this beta-adrenergic-mediated vasorelaxation. These data indicate that both Iso and IGF-1 activate the EC PI3-K pathway which mediates, in part, the release of NO and subsequent vasorelaxation in response to this beta-agonist Iso as well as to IGF-1.

Tyrosine Phosphorylation of a ~30 kD Protein Precedes avb3 Integrin-signaled Endothelial Cell Spreading and Motility on Matrix Proteins.

A microvascular endothelial cell line (CD clone 4) isolated from murine lung adheres to and spreads well on fibronectin, vitronectin, and fibrinogen, but poorly on collagen type IV and laminin. Ligating cell surface av, b3, a4, a5, or b1 integrin receptors with monospecific antibodies promoted a dramatic cell spreading and motility on vitronectin or collagen IV. Antibodies directed to other adhesion molecules, including aIIb, PECAM-1, and P-selectin were ineffective. Ligation with monoclonal anti-av or -b3, but not -a4, -a5, or -b1 antibodies, induced a rapid, and dose-dependent tyrosine phosphorylation of a ~30 kD protein, which preceded CD clone 4 endothelial cell spreading and motility and was partially inhibited by genistein and completely inhibited by BAPTA. All other antibodies tested did not induce the tyrosine phosphorylation of the 30 kD protein as well as cell spreading and motility. The present results suggest that b1 and b3 integrins employ different biochemical mechanisms in signaling endothelial cell spreading and motility and that the tyrosine phosphorylation of the 30 kD protein (and probably other proteins) may play an important role in signaling b3 integrin-mediated endothelial cell interaction with other cells (e.g., tumor cells) and extracellular matrix.