Sustained Lipopolysaccharide-Induced Lung Inflammation in Mice Is Attenuated by Functional Deficiency of the Fas/Fas Ligand System

To determine whether the Fas/Fas ligand (FasL) (CD95/CD178)system contributes to the development of an inflammatory responsein vivo, 2.5 µg of bacterial lipopolysaccharide (LPS;endotoxin) per g was administered intranasally to healthy mice(C57BL/6) and mutant mice deficient in either Fas (lpr mice)or FasL (gld mice). Sustained LPS-induced neutrophilic inflammationin the lungs was attenuated in both lpr and gld mice. Theseobservations provide further evidence of a proinflammatory rolefor the Fas/FasL system in the lungs.

INTRODUCTION

Top

Introduction

The Fas/Fas ligand (FasL) (CD95/CD178) system is recognizedfor its importance in the regulation of apoptosis in a widevariety of tissues. As a "death receptor," Fas is generallyconsidered to exclusively transmit a death signal, resultingin a biologically silent, noninflammatory mechanism for theelimination of cells, especially during the regulation and resolutionof immune and inflammatory responses (6, 13). However, accumulatingevidence indicates that the Fas/FasL system may also preferentiallysignal cellular proinflammatory responses associated with tissueinjury under certain circumstances (1, 4, 14). Data from bothanimals and humans implicating the Fas/FasL system as a potentialmediator of acute inflammatory lung injury have previously beenreported (7-10).

A growing body of literature suggests that a complex interactionexists between bacterial lipopolysaccharide (LPS; endotoxin)-inducedinflammatory responses and the Fas/FasL system. Experimentalevidence suggests that LPS may induce apoptosis via Fas-associateddeath domain (FADD)-dependent mechanisms (3). LPS has been reportedto initiate caspase activation and apoptosis in endothelialcell lines via a CD14- and FADD-dependent mechanism (3). Inanother study, mice developed lung endothelial injury, as wellas evidence of apoptosis in neutrophils (polymorphonuclear leukocytes[PMN]), macrophages, and cells of the alveolar walls, within24 h of intratracheal instillation of LPS (5). These changeswere attenuated by the administration of a Fas inhibitory antibody,suggesting that the Fas/FasL system participates in apoptoticlung injury induced by intratracheal LPS in mice. In a thirdstudy, it was shown that FADD modulates proinflammatory responsesin endothelial cells. Specifically, the overexpression of FADDblocked the activation of NF- ${kappa}$ B following LPS exposure in anendothelial cell line (2). Furthermore, LPS-induced expressionof the cytokines interleukin-6 and KC was enhanced in FADD-deficientmurine fibroblasts (2).

The present study was conducted to determine whether the Fas/FasLsystem plays a role in the development of an inflammatory responseto LPS in vivo. The response to intranasal instillation of LPS(2.5 µg/g) was examined in healthy mice (C57BL/6) andmutant mice deficient in either Fas (lpr mice) or FasL (gldmice).

MATERIALS AND METHODS

Top

Materials and Methods

The animal protocol was approved by the Animal Care Committeeof the University of Washington, Seattle. Briefly, male miceweighing 20 to 30 g were anesthetized with inhaled halothane.The mice were either C57BL/6 mice (B&K Universal, Seattle,Wash.) or naturally occurring mutant mice lacking the Fas receptor(lpr mice) or FasL (gld mice) (Jackson Laboratories) on theC57BL/6 background. Each mouse received Escherichia coli O111:B4LPS (Sigma, St. Louis, Mo.) at a dose of 2.5 µg/g by intranasalinstillation as previously described (11). The animals wereallowed to recover from anesthesia, returned to their cages,and given free access to water and food. After 6, 24, or 48h, the animals were euthanized with ketamine and xylazine, theirthoraxes were rapidly opened, and the animals were exsanguinatedby direct cardiac puncture. The lungs were dissected free, andthe tracheas were cannulated in order to perform bronchoalveolarlavage (BAL) or to fix the lungs. BAL was performed by instilling0.9% NaCl containing 0.6 mM EDTA in two separate 0.5-ml aliquots.The BAL fluid (BALF) was recovered by gentle suction and placedon ice for immediate processing. The lungs were fixed by inflationwith 10% neutral buffered formalin at a transpulmonary pressureof 15 cm H₂O and embedded in paraffin. An aliquot of BALF wasprocessed immediately for total and differential cell counts.Total cells were counted with a hemacytometer, and differentialcell counts were obtained from cytospin preparations stainedwith modified Wright-Giemsa stain (Diff-Quik; American ScientificProducts, McGaw Park, Ill.). The remainder of the lavage fluidwas spun at 200 x g for 30 min, and the supernatant was removedaseptically and stored in individual aliquots at -70°C.The total protein concentration in BALF was measured by usingthe bicinchoninic acid method (Pierce Co., Rockford, Ill.).The cytokines KC and tumor necrosis factor alpha (TNF- ${alpha}$ ) weremeasured in BALF by using commercially available immunoassays(R&D Systems, Minneapolis, Minn.). Comparisons among groupswere made with factorial analysis of variance and the Tukeypost hoc test with Graph-Pad InStat software (San Diego, Calif.).A P value of <0.05 was considered significant.

RESULTS

Top

Results

All of the animals survived for the duration of the experiments.Six hours after intranasal instillation of LPS, the numbersof leukocytes in BALF specimens were similar for the three groupsof mice (Fig. 1). In contrast, at 24 and 48 h, BALF from thelpr and gld animals contained significantly fewer white bloodcells (WBC) than BALF from wild-type C57BL/6 animals (P <0.05) (Fig. 1A). The difference in numbers of WBC was due tothe presence of lower numbers of PMN in the BALF from lpr animals[PMN at 24 h, (104 ± 25) x 10³; PMN at 48 h, (171 ±75) x 10³] and gld animals [PMN at 24 h, (96 ± 37) x10³; PMN at 48 h, (184 ± 52) x 10³] than in wild-typeC57BL/6 animals [PMN at 24 h, (315 ± 53) x 10³; at 48h, (439 ± 90) x 10³] (P < 0.05) (Fig. 1B). The numbersof mononuclear cells remained similar in all three groups ofanimals at all times examined (data not shown).

View larger version (23K):
[in this window]
[in a new window]
FIG. 1. Total numbers of WBC (A) and PMN (B) in BALF from wild-type C57BL/6, Fas-deficient lpr, and FasL-deficient gld mice 6, 24, and 48 h after intranasal administration of E. coli LPS (2.5 µg/g). Asterisk, P < 0.05.

Concentrations of the proinflammatory cytokine TNF- ${alpha}$ and themurine chemokine KC were measured in the BALF from animals euthanized24 and 48 h after the administration of LPS. The concentrationsof KC in BALF were not significantly different among the groupsof animals at the times examined. In contrast, the concentrationof TNF- ${alpha}$ was significantly decreased in BALF from lpr mice (53± 28 ng/ml [mean ± standard error of the mean])compared to that in wild-type C57BL/6 mice (186 ± 98ng/ml) at the 48-h time point following LPS administration (P< 0.05) (Fig. 2).

View larger version (17K):
[in this window]
[in a new window]
FIG. 2. Concentrations of TNF- ${alpha}$ in the BALF of wild-type C57BL/6, Fas-deficient lpr, and FasL-deficient gld mice 24 and 48 h after intranasal administration of E. coli LPS (2.5 µg/g). Asterisk, P < 0.05.

Representative histopathological samples from animals euthanized24 h after receiving intranasal LPS are shown in Fig. 3. Denseneutrophilic infiltrates and poor expansion were present inlungs from wild-type C57BL/6 mice (Fig. 3A and B). In contrast,only moderate neutrophilic infiltration was present in lungsfrom lpr (Fig. 3C and D) and gld (Fig. 3E and F) mice.

View larger version (125K):
[in this window]
[in a new window]
FIG. 3. Lung tissue sections stained with hematoxylin-eosin from wild-type C57BL/6 (A and B), Fas-deficient lpr (C and D), and FasL-deficient gld (E and F) mice 24 h after intranasal administration of E. coli LPS (2.5 µg/g). Lungs from C57BL/6 mice show dense neutrophilic alveolar infiltrates and poor inflation (magnifications, x100 [A] and x400 [B]). Lungs from lpr (magnifications, x100 [C] and x400 [D]) and gld (magnifications, x100 [E] and x400 [F]) mice show healthy alveolar walls and moderate neutrophilic infiltrates without intra-alveolar protein deposition.

DISCUSSION

Top

Discussion

These results demonstrate that the Fas/FasL system mediates,at least in part, sustained pulmonary neutrophilic inflammationin response to LPS administration in mice. The lung neutrophilicinflammatory response was impaired in both Fas-deficient lprmice and FasL-deficient gld mice at 24 and 48 h following theadministration of LPS. Decreased recruitment of PMN to the airspacesresulting from functional impairment of the Fas/FasL systemmight account for the differences in responses observed in lprand gld mice compared to that in wild-type C57BL/6 mice. TheFas/FasL system might play a direct role in PMN recruitmentto the lung airspaces, as suggested by previous reports of solubleFasL serving as a PMN chemoattractant (11, 15). However, wehave been unable to confirm that FasL functions directly asa chemotactic factor for neutrophils (12). Another possibilityis that the Fas/FasL system may induce the production of chemotacticcytokines from resident tissue cells, such as macrophages (12).Interestingly, the decreased neutrophilic inflammatory responsesobserved in lpr and gld mice were not associated with reducedconcentrations of the potent murine PMN chemokine, KC, in theBALF from lpr and gld mice compared to that from wild-type C57BL/6mice, although there were lower concentrations of TNF- ${alpha}$ in thelpr mice at the 48-h time point. A third possibility, givenprevious findings suggesting that the Fas/FasL system may playan important role in the pathogenesis of acute lung injury inboth animals and humans (7-10), is that Fas/FasL-mediated disruptionof the alveolar-endothelial barrier is required to allow chemotacticcytokines to reach the intraluminal surface for presentationto and recruitment of circulating PMN (5).

In summary, our results demonstrate that the acute inflammatoryresponse to LPS is blunted in the lungs of mice functionallydeficient in either Fas or FasL. To our knowledge, these findingsrepresent the first reported evidence indicating that the Fas/FasLsystem plays a role in LPS-induced inflammation in vivo. Furthermore,these observations provide further evidence of a proinflammatoryrole for the Fas/FasL system in the lungs.

ACKNOWLEDGMENTS

This work was supported in part by the Medical Research Serviceof the Seattle VAMC and by Public Health Service grants HL70840-01(G.M.-B.), HL65892 (T.R.M), and HL62995 (W.C.L.) from the NationalInstitutes of Health.

FOOTNOTES

* Corresponding author. Mailing address: Pulmonary Research Group, Seattle VAMC, 1660 S. Columbian Way, 151L, Seattle, WA 98108. Phone: (206) 277-4434. Fax: (206) 768-5289. E-mail: matuteb{at}u.washington.edu.

REFERENCES

Top