| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Channing Laboratory, Departments of
*
Medicine and
Pathology, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA 02115
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production, which when transferred to monolayers of MMCs elicited
significant ICAM-1 expression by these cells. This response resulted in
enhanced polymorphonuclear leukocyte attachment to MMCs that could be
inhibited by Abs specific for TNF- or ICAM-1. Mice treated with
TNF-- or ICAM-1-specific Abs failed to develop intraabdominal
abscesses following challenge with purified CPC. These results
illustrated the role of the CPC in promoting adhesion of B.
fragilis to the peritoneal wall and coordinating the cellular
events leading to the development of abscesses associated with
experimental intraabdominal sepsis. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Abscess formation is a complex host response that involves the recruitment and accumulation of neutrophils, fibrin deposition, and other incompletely defined processes. In experimental models, abscesses develop following i.p. challenge with B. fragilis or purified CPC, PSA, or PSB (15, 17, 18, 19). Early studies (20) showed that encapsulated B. fragilis bound to the peritoneal walls of rats better than unencapsulated Bacteroides species, enabling B. fragilis to resist clearance from the peritoneal cavity by the diaphragmatic lymph system (20). Several groups have demonstrated that peritoneal mesothelium, a layer of cells that constitutes a line of structural and immunologic defense in the abdominal cavity, potentiates the deposition of fibrin (21), and the production of an array of cytokines and cell adhesion molecules (21, 22, 23, 24, 25, 26, 27, 28), plays an important role in abdominal sepsis. Despite the lack of information describing a role for peritoneal mesothelium during the formation of intraabdominal abscesses, it is likely that inflammatory cells interact with this physical barrier during the migration from host tissues to the peritoneal lumen. The processes governing accumulation of these cells in the peritoneal cavity remain unclear (29); however, these mechanisms most likely parallel those elucidated for migration of immune cells from the vasculature to a focus of infection: a complex process regulated by cytokines, cell adhesion molecules, and cell activation (23, 30, 31).
Several studies have shown that the host immune response is
critical to abscess formation and that several cell types are important
in the development of intraabdominal abscesses (9, 19, 32, 33).
Intraperitoneal challenge of animals with B. fragilis is
followed by immune cell infiltration, with an initial influx of
lymphocytes into the peritoneal cavity and the appearance of
neutrophils and macrophages approximately 4 days postchallenge (9).
Recent studies have shown that purified phagocytic cells from mice or
humans cultured in vitro with CPC produce TNF-, IL-1, IL-8, and
IL-10 (33). It has been suggested that cytokines may be responsible for
triggering the migration of immune cells into the peritoneal cavity
following contamination with B. fragilis (33); however, the
source and role of these cytokines remain undefined.
The prevalence of isolation of encapsulated B. fragilis from
clinical cases of abscess formation led us to hypothesize that
associated surface polysaccharides allow this organism to persist
preferentially within the peritoneal cavity and initiate cellular
events that lead to the formation of this pathobiologic host response.
In this work, we present data that demonstrate the preferential binding
of B. fragilis, as well as purified surface polysaccharides
from this organism, to MMCs in vitro. These polysaccharides stimulate
TNF- production by peritoneal macrophages that in turn elicited the
production of ICAM-1 by MMCs. ICAM-1 expression on MMCs served as a
functional ligand that supports increased PMNL binding to these cells.
mAb-blocking experiments in mice demonstrated that both TNF- and
ICAM-1 expression are necessary for the development of intraabdominal
abscesses in vivo. These studies define cellular events critical for
intraabdominal abscess formation by B. fragilis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B. fragilis NCTC 9343 was obtained from National Culture Type Collection, Rockville, MD. Bacteroides thetaiotaomicron strain 491909 and Bacteroides distasonis strain 8503 are clinical strains obtained from the Channing Laboratory’s (Brigham and Women’s Hospital, Boston, MA) anaerobe stock culture collection, and were identified to the species level by long chain fatty acid analysis and conventional biochemical reactions. Each strain was passaged once on Brucella agar supplemented with 5% defibrinated sheep’s blood and stored frozen at -80°C in peptone-yeast extract broth. When needed, frozen aliquots were thawed, grown overnight at 37°C in an anaerobic chamber. Bacterial growth was collected and resuspended in DMEM without serum to the desired multiplicity of infection before bacterial binding experiments.
Isolation of CPC; purification of PSA and group B Streptococcus type III capsular polysaccharide
The CPC used in these studies was isolated from B. fragilis grown in proteose-peptone yeast extract broth supplemented with hemin and menadione in a 20-L pH-controlled (pH 7.2) batch culture overnight at 37°C, as described previously (34).
PSA was generated from pure CPC by isoelectric focusing with a Rotofor chamber (Bio-Rad, Hercules, CA) in 2% ampholytes (range 3–10) for 4 to 5 h at 12 watts constant power. Focused fractions were collected, and a sample of each fraction was subjected to immunoelectrophoresis and subsequent immunoprecipitation with high titer rabbit antiserum to B. fragilis NCTC 9343 (34). Samples containing PSA were pooled and dialyzed against 1 M NaCl overnight, and then against distilled water for 2 days.
The purity of CPC and PSA was assessed by nuclear magnetic resonance spectroscopy, gas chromatography-mass spectrometry, immunoelectrophoresis (pH 7.3), UV spectroscopy (260 and 280 nm), and reducing PAGE on gradient gels with subsequent silver staining, as described (34). The CPC and PSA used for these experiments were isolated from a single extraction, tested for purity by the above methods, and stored dry at 4°C. Before use, each Ag was diluted to 1 mg/ml in pyrogen-free water and tested for endotoxin by the Limulus amebocyte lysate assay (Cape Cod Associates, Woods Hole, MA). All Ags used in these studies tested free of endotoxin.
The native and tritiated GBSTIII polysaccharides used in these experiments were a kind gift from Dr. Lawrence Paoletti (Channing Laboratory).
Radiolabeling of PSA
3H-radiolabeled PSA was generated by oxidation and subsequent reduction. In brief, PSA was treated with sodium metaperiodate (0.01 M) to oxidize approximately 25% of the vicinal hydroxyl groups on the galactofuranose of the PSA side chain to carbonyl groups. Ethylene glycol was added to stop oxidation, and the sample was dialyzed overnight against water. The newly generated carbonyl groups underwent reduction with tritiated sodium borohydride (DuPont NEN, Boston, MA) to form isotope-labeled hydroxymethyl groups on PSA. Excess unlabeled sodium borohydride was added to completely modify any remaining carbonyl groups, and the modified Ag was dialyzed overnight against water lyophilized, and stored dry at 4°C. We have demonstrated that this procedure does not alter the biologic activity of the polymer.
Isolation and culture of MMCs, murine peritoneal macrophage, and PMNLs
MMCs. MMCs were isolated from the peritoneum of C57BL/6 mice by enzymatic digestion and were cultivated in wells of collagen-coated culture vessels (35). Briefly, omentum was harvested and digested with collagenase-dispase for 30 min at room temperature. Liberated cells were collected by centrifugation and washed extensively to remove enzyme. Cells were grown in DMEM with 12% FCS supplemented with 2-hydrocortisone and epidermal growth factor until confluent (4–6 days), then subcultured into 24-well or 96-well collagen-coated plates, grown for 24 to 48 h, and used upon reaching confluency (approximately 2.5 x 105 and 3.3 x 104 cells/well, respectively, and >98% pure by morphology and immunofluorescent assay). Bacteroides sp. were added at various multiplicities of infection, polysaccharide Ags were added at various concentrations, or supernatant fluids from Ag-stimulated macrophage were added to these cells. The number of MMCs per monolayer was determined for each assay.
Murine peritoneal macrophage. pMo were elicited in C57BL/6 mice by peritoneal injection of thioglycolate broth. After 3 days, cells were harvested by lavage, washed, added to 24-well tissue culture plates at 1 x 106 cells/ml, and allowed to adhere to plastic for 2 h at 37°C. The monolayers were washed to remove nonadherent cells and cultured (>90% macrophage) with 1 ml culture medium, or medium containing 10 µg/ml CPC, CPC treated with 20 µg/ml polymyxin B, PGG-glucan (Alpha-Beta Technologies, Worcester, MA), or GBSTIII. After 24 h, supernatants were harvested, centrifuged at 1000 x g for 20 min to remove cells, and stored frozen at -80°C.
PMNLs. To isolate PMNLs, peripheral blood was taken from healthy human donors and layered over a bed of Mono-Poly resolving medium (ICN Biomedicals, Palo Alto, CA). Following separation by centrifugation, the PMNL fraction was collected (>95% pure), washed with ice-cold DMEM to remove separation medium, resuspended in DMEM to 2 to 3 x 107 cells/ml, and maintained on ice (<30 min). A 1-ml sample of these cells was warmed to 37°C and added to monolayers of MMCs in 24-well plates.
Bacteroides- and CPC-binding assays and blocking experiments
Monolayers of MMCs were cocultured with B. fragilis, B. thetaiotaomicron, or B. distasonis for 1 h at 37°C with 5% CO2. Monolayers were washed extensively to remove unbound bacteria, and an equivalent volume of sterile water was added to the monolayers. Following lysis, vigorous aspiration-expulsion cycles were performed with a pipet to evenly distribute bacteria. The lysate was serially diluted in 1% peptone, plated on Brucella agar, and grown for 48 h for viable count (CFU/ml) determination. Additional experiments involved the addition of B. fragilis CPC, PSA, [3H]PSA, and GBSTIII polysaccharide to MMCs. These Ags were weighed, diluted to a concentration of 1 mg/ml in DMEM without serum, and vortexed until completely dissolved; dilutions were then made with DMEM, and the Ags were added to MMCs. The amount of Ag bound to cells was evaluated by ELISA or liquid scintillation.
In experiments designed to block the binding of B. fragilis to MMCs, bacteria were left untreated or treated with B. fragilis strain 9343-specific capsular polysaccharide antiserum or irrelevant Ab for 1 h at 37°C before addition of bacteria to MMC monolayers. Blocking of PSA binding to MMCs was accomplished by adding various dilutions of a PSA-specific mouse mAb (clone CE3) or nonimmune mouse control ascites (Sigma, St. Louis, MO) to PSA (10 µg/ml) for 1 h at 37°C. Untreated or Ab-treated B. fragilis or PSA was added to MMCs for 1 h, and binding was evaluated by CFU/ml determinations or ELISA.
Competition experiments were performed to demonstrate specific binding of PSA to MMCs or pMo. To tritiated PSA (10 µg/ml), we added a 50-fold excess of native unmodified PSA (500 µg/ml). This polysaccharide mixture was added to monolayers of cells in 24-well plates and cocultured for 1 h at 37°C. The cells were washed three times with DMEM and lysed with 1 ml of sterile distilled water, and the lysates were collected and processed for liquid scintillation enumeration of 3H-radiolabeled PSA binding.
Quantitation of bacteria and Ag binding
Colony counts. After incubation with bacteria, MMCs were washed with DMEM to remove unbound bacteria (with the efficiency of washing determined by plating of the final wash), and 100 µl of sterile water was added to the monolayers for 30 min to lyse MMCs. These lysates were subjected to cycles of vigorous aspiration and expulsion to disrupt cells and evenly disperse bacteria. The lysates were subjected to serial 10-fold dilution in 1% peptone, plated onto Brucella blood agar plates, and incubated at 37°C in an anaerobic chamber. It was noted that the treatment of these bacteria in this manner did not affect organism viability. After 2 days, colonies were enumerated.
ELISA. MMCs cocultured with CPC were gently washed to remove excess unbound Ag and were fixed with 2% formaldehyde in PBS (pH 7.2) for 1 h. After fixation, monolayers were washed with PBS + 0.05% Tween-20 (pH 7.2). High titer rabbit serum specific for B. fragilis was added at a 1/2000 dilution in PBS (100 µl/well), and the monolayers were incubated for 1 h at 37°C. Incubation was followed by three washes, after which 100 µl/well of a 1/4000 dilution of goat antiserum to rabbit IgG/alkaline phosphatase conjugate (Biosource, Camarillo, CA) was added to monolayers and incubated for 1 h at 37°C. The wells were washed, and 100 µl of p-nitrophenyl phosphate solution (Kirkegaard & Perry Laboratories, Gaithersburg, MD) was added for 15 min. The reaction was stopped, and plates were read with a microtiter plate reader at 405 nm.
Liquid scintillation. After incubation for 1 h with tritiated polysaccharides, cells in either 24- or 96-well plates were washed with DMEM to remove unbound Ag, lysed with water in situ, and harvested onto glass fiber filters with a PHD cell harvester (Cambridge Technologies, Watertown, MA). The glass fiber disks were placed into glass vials with 2-ml liquid scintillation mixture, and cpm were measured with a liquid scintillation reader (Packard International, Downers Grove, IL). The sp. act. of tritiated PSA and GBSTIII were determined using a 10-µg sample of each polysaccharide during each binding experiment, and cpm/ng dry weight of each polymer was calculated.
Cytokine and ICAM-1 detection by ELISA
MMCs were grown to confluency in 96-well collagen-coated cell
culture plates. Proinflammatory cytokines were detected from MMCs
cultured with CPC or PGG-glucan (10 µg/ml). Supernatants were
collected 1, 4, 8, and 24 h after stimulation, and levels of
TNF- and IL-1 were determined with cytokine-specific ELISA kits
(Endogen, Cambridge, MA).
To detect ICAM-1 on the surface of MMCs, an in situ ELISA assay was
developed. MMCs in 96-well plates were incubated for 18 h with 100
µl culture medium, 500 U/ml murine rTNF- (R&D Systems, Cambridge,
MA), CPC, or PGG-glucan (10 µg/ml), and supernatants from
Ag-stimulated peritoneal macrophage. After stimulation, MMCs were
washed to remove Ags and fixed with 2% buffered formaldehyde. After
fixation, cell monolayers were incubated with 100 µl/well of a 1/500
dilution of rat anti-murine ICAM-1 mAb (clone YN/1.7.4; American
Type Culture Collection, Rockville, MD) for 1 h. The monolayers
were washed and then incubated with 1/2000 dilution of goat antiserum
specific for rat IgG/alkaline phosphatase (Sigma). Alkaline phosphatase
substrate reagent (Kirkegaard & Perry Laboratories) was added to each
well and developed for 1 h, and absorbance was read at 405
nm.
Treatment of peritoneal macrophage supernatants with
TNF--neutralizing Ab
Supernatant fluids from CPC-stimulated peritoneal macrophage
were thawed and either added directly to monolayers of MMCs in 96-well
plates or treated with 10, 1, or 0.1 µg of goat neutralizing Ab to
murine TNF- (R&D Systems) or nonimmune goat IgG (Sigma) for 1 h
at 37°C before addition to MMCs. After the addition of these
supernatants, MMCs were incubated for 18 h and then assayed for
surface-expressed ICAM-1 by ELISA or in PMNL-binding experiments.
PMNL/MMC cell adherence assays
We adapted the method for studying the binding of human PMNLs to murine endothelium (36) and modified this for MMCs. Supernatant fluids from Ag-stimulated peritoneal macrophage were added to MMCs in 24-well plates for 18 h, as previously described (this study). PMNLs were added to MMC and allowed to adhere for 30 min. After binding, the cocultures were washed to remove unbound PMNLs while taking care to maintain intact MMC monolayers. Additional experiments were performed to characterize the mechanism of PMNL attachment to MMCs stimulated with CPC-treated macrophage supernatant fluids. Following stimulation, MMCs were treated with ICAM-1-specific mAb (100 µg/ml) or irrelevant Ab matched to isotype (clone IXB2; a kind gift from Dr. Gene Muller, Channing Laboratory, Brigham and Women’s Hospital) for 1 h before PMNL attachment. An observer blinded as to the treatments counted the number of PMNLs bound to MMCs per x200 magnification field with an Olympus CK2 phase contrast microscope. Five random fields were counted per sample.
In vivo Ab blocking of ICAM-1 and TNF-
A murine model of peritoneal abscess formation was adapted to
assess the role of TNF- and ICAM-1 during abscess formation (19). In
brief, C57BL/6 mice received 100 µl i.p. injections of rat Ab to
murine ICAM-1, goat Ab to murine TNF-, or sham mAb (clone IXB2) in
PBS (1 mg/ml) 24 and 4 h before implantation of an
abscess-inducing inoculum of 100 µg CPC in the adjuvant sterile cecal
contents. Mice received additional i.p. injections of mAb 4, 24, 48,
72, and 96 h after challenge to down-regulate TNF- or ICAM-1 in
vivo (37). Six days after B. fragilis CPC challenge, an
observer blinded as to the treatment, then graded for presence of i.p.
abscesses in these animals.
Statistical analyses
All statistical analyses were performed with InStat statistical analysis software (Graphpad Software, San Diego, CA) on an IBM Personal Computer AT. Results of in vitro data were calculated from three experiments, recorded as the mean ± SD, and analyzed with the Kruskal-Wallis nonparametric test. In vivo data were analyzed by the Fisher’s exact test. A p value of less than 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Initial experiments defined the binding kinetics of B.
fragilis to MMCs. B. fragilis was added to MMCs at
various multiplicities of infection ranging from 1 to 10,000. Bacterial
viable counts indicated that a multiplicity of infection of 1000
saturated binding sites on MMCs (data not shown). Additional
experiments compared the binding of B. fragilis with other
Bacteroides species (Fig. 1
).
B. fragilis (1.39 x 106 CFU/ml) bound more
avidly than B. thetaiotaomicron (2.35 x
105 CFU/ml; p = 0.0004 vs B.
fragilis) or B. distasonis (8.88 x
104 CFU/ml; p = 0.0001 vs B.
fragilis). In similar experiments to characterize the
attachment of B. fragilis to MMCs, bacteria were treated
with either CPC-specific rabbit polyclonal Ab or irrelevant Ab before
the addition to MMC monolayers. Irrelevant Ab-treated B.
fragilis (1.28 x 106 CFU/ml) bound to similar
levels as untreated B. fragilis (1.15 x
106 CFU/ml), while CPC-specific Ab treatment significantly
reduced B. fragilis attachment (1.21 x 105
CFU/ml; p < 0.002 vs irrelevant Ab treatment).
|
MMCs.
The binding of CPC to MMCs was measured by ELISA. With the addition of
increasing doses of CPC (ranging from 10 ng/ml to 200 µg/ml) in DMEM,
saturation was achieved at a dose of 10 µg/ml (Fig. 2A). Maximal binding of
this dose of CPC occurred within 15 min.
|
B). PSA binding was inhibited with mAb CE3 (specific
for PSA), as measured by ELISA (48% inhibition; p <
0.0001). Competitive binding experiments demonstrated specificity of
PSA binding to MMCs. The addition of 50-fold unlabeled PSA with
tritiated PSA (10 µg/ml) to MMC monolayers significantly reduced the
binding of tritiated PSA (Fig. 2B). Addition of
3H-radiolabeled GBSTIII (sp. act. 42.78 cpm/ng) to MMCs
failed to bind appreciably irrespective of dose administered.
Peritoneal macrophage.
CPC and PSA of B. fragilis bound readily to pMo. The
time-dependent binding of tritiated PSA to these cells is shown in
Figure 2C. Saturation of binding sites on pMo occurred with
a dose of 50 µg/ml (3% of the input Ag or 0.289 pg/cell; Fig. 2D) and occurred at 1 h following the addition
of PSA (Fig. 2C). PSA bound specifically to pMo as
50-fold excess unlabeled PSA significantly inhibited PSA attachment
(Fig. 2D); furthermore, the addition of 50-fold
excess GBSTIII polysaccharide failed to inhibit PSA binding. GBSTIII
polysaccharide did not bind appreciably to these cells.
CPC stimulation of TNF- and ICAM-1
Direct in vitro stimulation of MMCs with CPC failed to elicit
detectable levels of the proinflammatory cytokines TNF- or IL-1
from these cells, but resulted in a modest increase in
surface-expressed ICAM-1 compared with untreated or PGG-glucan-treated
cells (p < 0.02 and p < 0.04,
respectively; data not shown). Additional experiments, in which culture
supernatants from CPC-stimulated murine peritoneal macrophages were
added to MMCs for 18 h, resulted in a potent ICAM-1 response
(p < 0.0001 vs medium supernatant transfer;
Fig. 3) by these cells. This effect was
dependent on the dose and time of CPC administration to the macrophages
and was not elicited by PGG-glucan or GBSTIII. In addition, incubation
of CPC with polymyxin B did not affect ICAM-1 expression. Based on our
previous data in which CPC was shown to elicit a potent TNF-
response from murine peritoneal macrophages (33), we hypothesized that
this cytokine was responsible for ICAM-1 expression by MMCs. Therefore,
CPC-stimulated macrophage supernatants were treated with neutralizing
Ab specific for murine TNF- (1 µg/ml) before addition of the
supernatants to MMCs. This treatment significantly reduced the level of
ICAM-1 expressed by MMCs (p = 0.0022 vs
nonimmune goat IgG; Fig. 3).
|
To assess the biologic function of ICAM-1 expression by MMCs, a
PMNL-binding assay was performed. In this assay, human PMNLs were added
to MMC monolayers following culture with supernatants from pMo
stimulated with medium, PGG-glucan, GBSTIII, or CPC. In these
experiments, direct stimulation of MMCs with TNF- for 18 h
resulted in enhanced PMNL binding (p < 0.002
vs medium control; Fig. 4). MMCs treated
with supernatant fluids from CPC-stimulated macrophages supported
increased PMNL binding compared with supernatants from PGG-glucan- or
GBSTIII-treated macrophages (p < 0.002 and
p < 0.002, respectively, vs CPC stimulation; Fig. 4).
Treatment of CPC-stimulated supernatant fluids with murine
TNF--neutralizing Ab significantly reduced PMNL binding to MMCs
(p < 0.002 vs irrelevant Ab treatment; Fig. 4). Furthermore, PMNL binding to MMCs was inhibited by treatment of
monolayers with ICAM-1-specific mAb (p < 0.002
vs irrelevant Ab treatment; Fig. 4).
|
and ICAM-1 in abscess formation
The role of TNF- and ICAM-1 in the development of
intraabdominal abscesses was studied in a murine model of peritoneal
sepsis. Mice received i.p. injections of TNF--neutralizing Ab,
ICAM-1-specific mAb, or sham Ab (100 µg/injection) 24 and 4 h
before challenge, and 2, 24, 48, 72, and 96 h after B.
fragilis CPC challenge. Treatment with these mAbs significantly
reduced the development of abdominal abscesses following CPC challenge,
while treatment with a sham Ab did not affect abscess formation
(p < 0.0005 for TNF-, and p
< 0.0005 for ICAM-1 vs irrelevant Ab treatment; Table I).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B. fragilis adhered more avidly to MMCs than either B. distasonis or B. thetaiotaomicron. This result suggested that the CPC functions as an attachment factor. Previous studies have shown that B. thetaiotaomicron has only a thin capsule layer (41), while B. distasonis lacks a capsule. This difference most likely explains why B. thetaiotaomicron binds less avidly than B. fragilis but more avidly than the unencapsulated B. distasonis. Although little is known about the capsular polysaccharide of B. thetaiotaomicron, its binding capacity is interesting since B. thetaiotaomicron is the second most frequently isolated Bacteroides species in human disease.
The finding that CPC adhered to different cell types (MMCs and pMo) was
not surprising, as surface-expressed polysaccharides from
Actinobacillus actinomycetem comitans and
Staphylococcus aureus type 5 and 8 bind to a variety of host
cells (42, 43, 44). Furthermore, recent studies have shown that binding of
microbial polysaccharides to host cells is important for eliciting
proinflammatory cytokines (43, 44, 45). Previous work by our group has
demonstrated that the CPC of B. fragilis elicits potent
TNF-, IL-1, IL-8, and IL-10 response from phagocytic cells of
human or murine origin (33). In the present study, we were unable to
detect the proinflammatory cytokines TNF- or IL-1 from MMCs
cocultured with CPC. Although other cytokines may be produced from
CPC-stimulated MMCs, we limited our current studies to these cytokines
since they are major inflammatory stimuli following microbial
contamination in the peritoneal cavity (27, 46) and are prominent in
the induction of cell adhesion molecules on cell surfaces (36, 47).
Based on our previous observations that B. fragilis and CPC promote rapid infiltration of lymphocytes, neutrophils, and macrophages into the peritoneal cavity of animals following i.p. challenge, we hypothesized that cell adhesion molecules such as ICAM-1 might play a role in the extravasation and localization of these cells to the peritoneum (47, 48). Direct stimulation of MMCs with CPC produced higher levels of surface-expressed ICAM-1 than cells in medium alone, or PGG-glucan-treated cells, although this increase was modest. Additional experiments showed that transfer of culture supernatants from CPC-stimulated peritoneal macrophages elicited a maximal ICAM-1 response from MMCs.
Since we have shown previously that murine peritoneal macrophages
cultured with the CPC produced TNF-, we believed that this
macrophage-derived cytokine played a major role in up-regulating the
expression of ICAM-1 on MMCs and is critical to the development of
intraabdominal abscesses. Treatment of pMo supernatants from
CPC-stimulated macrophages with TNF--neutralizing Ab significantly
reduced the ICAM-1 response, indicating that TNF- is a major factor
eliciting expression of this cell adhesion molecule on MMCs. Taken
together with our demonstration of CPC and PSA binding to pMo, these
data suggest that following challenge, peritoneal macrophages recognize
B. fragilis capsular polysaccharide, either bound to
mesothelium or in the peritoneal cavity, and secrete TNF-, which in
turn activates a potent inflammatory response leading to ICAM-1
expression on MMCs. The binding of human PMNLs to MMCs cultured with
supernatants from CPC-stimulated macrophages confirmed the importance
of TNF- and ICAM-1 in the localization of these cells to mesothelial
tissue.
The ability of TNF-- and ICAM-1-specific Abs to significantly reduce
abscess formation in the mouse model confirmed the biologic importance
of these immune mediators in the formation of this host response. We
propose that the binding of B. fragilis to MMCs serves two
roles: 1) localization of the organism on the mesothelial surface to
form a nidus of infection in the peritoneal cavity; and 2) stimulation
of ICAM-1 expression to provide a ligand for infiltrating PMNLs. These
two factors most likely form the first stages of intraabdominal abscess
formation in the infected host. The binding of CPC to MMCs is probably
insufficient to induce cell infiltration into the peritoneal cavity on
its own since proinflammatory cytokines were not detected from MMCs
after CPC attachment and elicited only modest ICAM-1 expression.
However, it appears that TNF- produced by resident or infiltrating
phagocytes in response to B. fragilis CPC plays the major
role in up-regulating ICAM-1 expression. This latter response leads to
the accumulation of PMNLs within the abdominal cavity, the hallmark of
abscess formation.
In summary, this work demonstrates that the CPC of B. fragilis interacts with the host immune system in a number of ways to coordinate a cellular response leading to abscess formation. We are currently investigating the chemotactic properties of CPC and PSA that may be responsible for the recruitment of PMNLs to the peritoneal cavity and the possible contribution of these factors to abscess formation associated with intraabdominal sepsis.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Frank C. Gibson III, Channing Laboratory, Departments of Medicine and Pathology, Brigham and Women’s Hospital and Harvard Medical School, 181 Longwood Avenue, Boston, MA 02115.
3 Abbreviations used in this paper: CPC, capsular polysaccharide complex; GBSTIII, group B streptococcal type III capsular polysaccharide; MMC, murine mesothelial cell; PGG, poly(1–6)-
-glucotriosyl-(1–3)--glucopyranose; PMNL, polymorphonuclear leukocyte; pMo, murine peritoneal macrophage; PSA, polysaccharide A; PSB, polysaccharide B.
Received for publication September 30, 1997. Accepted for publication January 13, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-inducible intercellular adhesion molecule-1. J. Immunol. 152:5883.[Abstract]
and tumor necrosis factor-. Am. J. Pathol. 142:1876.[Abstract]
3)--D-glucan biologic response modifier in two monocyte/macrophage cell lines. J. Immunol. 156:3418.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  G. d'Empaire, M. T. Baer, and F. C. Gibson III The K1 Serotype Capsular Polysaccharide of Porphyromonas gingivalis Elicits Chemokine Production from Murine Macrophages That Facilitates Cell Migration Infect. Immun., November 1, 2006; 74(11): 6236 - 6243. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. P. Robertson, C. J. Smith, A. M. Gough, and E. R. Rocha Characterization of Bacteroides fragilis Hemolysins and Regulation and Synergistic Interactions of HlyA and HlyB Infect. Immun., April 1, 2006; 74(4): 2304 - 2316. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. K. Mullarky, F. M. Szaba, K. N. Berggren, L. W. Kummer, L. B. Wilhelm, M. A. Parent, L. L. Johnson, and S. T. Smiley Tumor Necrosis Factor Alpha and Gamma Interferon, but Not Hemorrhage or Pathogen Burden, Dictate Levels of Protective Fibrin Deposition during Infection Infect. Immun., February 1, 2006; 74(2): 1181 - 1188. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. K. Gupta, K. M. Hasan, A. M. Mishra, D. Jha, M. Husain, K. N. Prasad, and P. A. Narayana High Fractional Anisotropy in Brain Abscesses versus Other Cystic Intracranial Lesions AJNR Am. J. Neuroradiol., May 1, 2005; 26(5): 1107 - 1114. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Stingele, B. Corthesy, N. Kusy, S. A. Porcelli, D. L. Kasper, and A. O. Tzianabos Zwitterionic Polysaccharides Stimulate T Cells with No Preferential V{beta} Usage and Promote Anergy, Resulting in Protection against Experimental Abscess Formation J. Immunol., February 1, 2004; 172(3): 1483 - 1490. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. L. Montenegro, F. G. C. Abath, A. L. C. Domingues, W. G. Melo, C. N. L. Morais, E. M. Coutinho, S. Mahanty, and T. A. Wynn Enhanced Interleukin-12 and CD40 Ligand Activities but Reduced Staphylococcus aureus Cowan 1-Induced Responses Suggest a Generalized and Progressively Impaired Type 1 Cytokine Pattern for Human Schistosomiasis Infect. Immun., November 1, 2002; 70(11): 5903 - 5912. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. R. Schopf, K. F. Hoffmann, A. W. Cheever, J. F. Urban Jr., and T. A. Wynn IL-10 Is Critical for Host Resistance and Survival During Gastrointestinal Helminth Infection J. Immunol., March 1, 2002; 168(5): 2383 - 2392. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Echtenacher, K. Weigl, N. Lehn, and D. N. Mannel Tumor Necrosis Factor-Dependent Adhesions as a Major Protective Mechanism Early in Septic Peritonitis in Mice Infect. Immun., June 1, 2001; 69(6): 3550 - 3555. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M. P. J. Reijnen, H. van Goor, P. Falk, M. Hedgren, and L. Holmdahl Sodium Hyaluronate Increases the Fibrinolytic Response of Human Peritoneal Mesothelial Cells Exposed to Tumor Necrosis Factor {alpha} Arch Surg, March 1, 2001; 136(3): 291 - 296. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. O. Tzianabos, A. Chandraker, W. Kalka-Moll, F. Stingele, V. M. Dong, R. W. Finberg, R. Peach, and M. H. Sayegh Bacterial Pathogens Induce Abscess Formation by CD4+ T-Cell Activation via the CD28-B7-2 Costimulatory Pathway Infect. Immun., December 1, 2000; 68(12): 6650 - 6655. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. O. Tzianabos Polysaccharide Immunomodulators as Therapeutic Agents: Structural Aspects and Biologic Function Clin. Microbiol. Rev., October 1, 2000; 13(4): 523 - 533. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. E. Comstock, M. J. Coyne, A. O. Tzianabos, A. Pantosti, A. B. Onderdonk, and D. L. Kasper Analysis of a Capsular Polysaccharide Biosynthesis Locus of Bacteroides fragilis Infect. Immun., July 1, 1999; 67(7): 3525 - 3532. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
