Increased Resistance of LFA-1-Deficient Mice toLipopolysaccharide-Induced Shock/Liver Injury in thePresence of TNF-� and IL-12 Is Mediated by IL-10: A NovelRole for LFA-1 in the Regulation of the Proinflammatory andAnti-Inflammatory Cytokine Balance1
Masashi Emoto,2* Yoshiko Emoto,* Volker Brinkmann,† Mamiko Miyamoto,*Izumi Yoshizawa,* Manuela Staber,* Nico van Rooijen,‡ Alf Hamann,§ andStefan H. E. Kaufmann*
Challenge with low doses of LPS together with D-galactosamine causes severe liver injury, resulting in lethal shock (low doseLPS-induced shock). We examined the role of LFA-1 in low dose LPS-induced shock. LFA-1�/� mice were more resistant to lowdose LPS-induced shock/liver injury than their heterozygous littermates, although serum levels of TNF-� and IL-12 were higherin these mice. C57BL/6 mice were not rescued from lethal effects of LPS by depletion of NK1� cells, granulocytes, or macrophages,and susceptibility of NKT cell-deficient mice was comparable to that of controls. High numbers of platelets were detected in theliver of LFA-1�/� mice after low dose LPS challenge, whereas liver accumulation of platelets was only marginal in LFA-1�/� mice.Following low dose LPS challenge, serum levels of IL-10 were higher in LFA-1�/� mice than in LFA-1�/� mice, and susceptibilityto low dose LPS-induced shock as well as platelet accumulation in the liver of LFA-1�/� mice were markedly increased by IL-10neutralization. Serum levels of IL-10 in LFA-1�/� mice were only marginally affected by macrophage depletion. However, inLFA-1�/� mice macrophage depletion markedly reduced serum levels of IL-10, and as a corollary, susceptibility of LFA-1�/� miceto low dose LPS-induced shock was markedly elevated despite the fact that TNF-� levels were also diminished. We conclude thatLFA-1 participates in LPS-induced lethal shock/liver injury by regulating IL-10 secretion from macrophages and that IL-10 playsa decisive role in resistance to shock/liver injury. Our data point to a novel role of LFA-1 in control of the proinflammatory/anti-inflammatory cytokine network. The Journal of Immunology, 2003, 171: 584–593.
S eptic shock is mainly attributed to exaggerated proinflam-matory cytokine production in response to Gram-negativebacteria and their unique cell wall component, LPS (1).
Proinflammatory cytokines, notably TNF-�, are pivotal mediatorsof septic shock (2–4). Although mice are relatively resistant toLPS-induced shock, high dose LPS challenge induces pathophys-iological reactions, including fever, hypotension, leukocyte infil-tration, and inflammation in various organs, resulting in a syn-drome resembling septic shock with a high mortality (2).D-galactosamine (D-GalN)3 increases the susceptibility of mice toLPS-induced shock by impairing liver metabolism (5, 6). In con-
trast to high dose LPS-induced shock which induces a systemicdisorder including multiple organ failures (2), liver is a major tar-get organ after challenge with low doses of LPS in conjunctionwith D-GalN (5, 6). Similarly to high dose LPS-induced shock,TNF-� plays a central role in low dose LPS-induced shock/liverinjury (6–9). In addition to TNF-�, other cytokines, includingIFN-�, participate in low dose LPS-induced shock/liver injury(10).
LFA-1 (CD11a/CD18) belongs to the �2 integrin family and isexpressed on the surface of virtually all leukocytes, albeit at dif-ferent levels (11). In mice, ICAM-1 (CD54) and ICAM-2(CD102), expressed on leukocytes, epithelial cells, endothelialcells, and fibroblasts, are the ligands for LFA-1 (11). In addition toICAM(s), LPS is considered a ligand for LFA-1 (12). Interactionsof LFA-1/ICAMs promote firm adhesion of leukocytes to vascularendothelium as the initiating event for transmigration of leuko-cytes into sites of inflammation (11). Infiltration of granulocytesinto the liver has been suggested as crucial event in low doseLPS-induced liver damage (13–15). Although the �2 integrin fam-ily member, Mac-1 (CD11b/CD18), has been suggested to partic-ipate in low dose LPS-induced shock/liver injury (16, 17), the roleof LFA-1 in low dose LPS-induced shock/liver injury remainselusive.
In the present study we examined the role of LFA-1 in low doseLPS-induced shock/liver injury. Our data reveal that LFA-1 ex-pression is a critical prerequisite for low dose LPS-induced shock/liver injury and that IL-10 produced by tissue macrophages plays
*Department of Immunology and †Central Core Facility Microscopy, Max-Planck-Institute for Infection Biology, Berlin, Germany; ‡Department of Cell Biology andImmunology, Faculty of Medicine, Vrije Universiteit, Amsterdam, The Netherlands;and §Experimental Rheumatology, Medical Clinic, Charite, Humboldt University,Berlin, Germany
Received for publication August 29, 2002. Accepted for publication May 5, 2003.
The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by a grant from the German Science Foundation(SFB421).2 Address correspondence and reprint requests to Dr. Masashi Emoto, Department ofImmunology, Max-Planck-Institute for Infection Biology, Schumannstrasse 21/22,10117 Berlin, Germany. E-mail address: [email protected] Abbreviations used in this paper: D-GalN, D-galactosamine; �2m, �2-microglobulin;Cl2 MBP-L, liposome-encapsulated dichloromethylene bisphosphonate; PBS-L, li-posome-encapsulated PBS.
a decisive role in resistance to low dose LPS-induced damage inthe absence of LFA-1. Our data define a novel role for LFA-1 inregulation of the proinflammatory/anti-inflammatory cytokine net-work, which determines the outcome of harmful inflammatory re-sponses, such as low dose LPS-induced shock.
Materials and MethodsMice
Breeding pairs of LFA-1�/� mice (18) and CD1d�/� mice were providedby Drs. R. Schmits (University of Saarland, Homburg, Germany) and A.Bendelac (University of Chicago, Chicago, IL), respectively. Breedingpairs of ICAM-1�/�, �2-microglobulin (�2m)�/�, and TCR��/� micewere purchased from The Jackson Laboratory (Bar Harbor, ME). Thesemutants backcrossed onto C57BL/6 (LFA-1�/� and LFA-1�/� mice, 4thgeneration; CD1d�/� mice, �8th generation; ICAM-1�/�, �2m�/� andTCR��/� mice, �15th generation) and C57BL/6 mice were maintainedunder specific pathogen-free conditions at our animal facilities, and weight-and generation-matched female mice were used at 8–10 wk of age.
Antibodies
mAbs against IFN-� (R4-6A2 and XMG1.2), IL-12 (p40/p70) (C17.8),IL-12 (p40) (C15.6.7), IL-10 (JES5-2A5), TNF-� (XT22), NK1.1(PK136), and Ly6G (RB6-8C5) were purified from hybridoma culture su-pernatants. Anti-IFN-� mAb (XMG1.2) and anti-IL-12 mAb (C15.6.7)were biotinylated by standard methods. F4/80 mAb (CI:A3-1) was ob-tained from Serotec (Oxford, U.K.). Anti-CD41 mAb (MWReg30), bio-tinylated anti-mouse IgG2a mAb (R19-15), FITC-conjugated-anti-mouseIgG2a mAb (R19-15), and FITC-conjugated anti-rat IgG2b mAb (G15-337) were purchased from BD PharMingen (Hamburg, Germany). Cy2-conjugated goat anti-rat IgG and Cy2-conjugated Fab of goat anti-rat IgGwere obtained from Jackson ImmunoResearch Laboratories (WestGrove, PA).
LPS and septic shock models
Salmonella typhimurium-derived LPS and D-GalN were purchased fromSigma-Aldrich (Deisenhofen, Germany). Highly purified S. abortus equi-derived LPS was provided by Dr. M. A. Freudenberg (Max-Planck-Insti-tute for Immunobiology, Freiburg, Germany). Mice received (i.v.) variousdoses of LPS and/or D-GalN (8 mg) dissolved in sterile PBS in a totalvolume of 200 �l.
Histopathology and TUNEL assay
For histology, specimens were embedded in Tissue-Tek (Sakura FinetekEurope, Zoeterwoude, The Netherlands), frozen, and cut on a Cryotome(Leica Microsystems, Bensheim, Germany). Sections (3–5 �m) were air-dried, fixed with acetone, rehydrated, and stained with H&E (Merck, Haar,Germany). For TUNEL assay, formalin-fixed specimens were infiltratedwith 20% sucrose (Merck) in PBS, placed in Tissue-Tek, frozen, and cut ona Cryotome. Sections (3–5 �m) were partially digested with 20 �g/mlproteinase K (Sigma-Aldrich) at room temperature for 15 min and sub-jected to the TUNEL reaction (kit from Roche, Mannheim, Germany) fol-lowing the manufacturer’s instructions. TUNEL-positive cells are brightgreen due to the incorporation of FITC-labeled nucleotides.
ELISA for cytokines
Serum levels of IFN-� and IL-12 (p40) were determined by ELISA asdescribed previously (19). In brief, serum samples were incubated in im-munoassay plates (Nunc, Copenhagen, Denmark) precoated with anti-IFN-� mAb (R4-6A2) or anti-IL-12 (p40/p70) mAb (C17.8), respectively.After washing, plates were incubated with biotinylated anti-IFN-� mAb(XMG1.2) or biotinylated anti-IL-12 (p40) mAb (C15.6.7), respectively,followed by streptavidin-conjugated alkaline phosphatase (Dianova, Ham-burg, Germany) and the chromogen p-nitrophenyl phosphate (Sigma-Aldrich). The cytokine concentration in each sample was determined usingserially diluted mouse rIFN-� (R&D Systems, Wiesbaden, Germany) ormouse rIL-12 (Genzyme, Alzenau, Germany). Serum levels of TNF-�,IL-12 (p70), and IL-10 were assayed using the Quantikine M kit (R&Dsystems) following the manufacturer’s instruction.
Cell preparation
Hepatic leukocytes were prepared as described previously (20). Spleno-cytes were prepared by standard methods. Blood samples were obtainedfrom the axillary vein, and leukocytes were collected after hypotonichemolysis.
ELISPOT
The frequencies of IL-12 (p40)-producing cells were determined as de-scribed previously (21) with slight modifications. Briefly, appropriate di-lutions of cells were cultured overnight in ELISPOT plates (Millipore,Eschborn, Germany) precoated with anti-IL-12 (p40/p70) mAb (C17.8).Plates were then washed and incubated with biotinylated anti-IL-12 (p40)mAb (C15.6.7) at 37°C for 2 h. For developing spots, streptavidin-conju-gated alkaline phosphatase (Dianova) and 5-bromo-4-chloro-3-indolylphosphate/nitro blue tetrazolium tablets (Sigma-Aldrich) were used. Thefrequencies of IL-10- or TNF-�-producing cells were determined byELISPOT assay using the mouse IL-10 or TNF-� measurement kit (R&DSystems), respectively, following the manufacturer’s instruction.
Cell depletion and blocking
Multilamellar liposome-encapsulated dichloromethylene bisphosphonate(Cl2MBP-L) was prepared as described previously (22). Cl2MBP was agift from Roche. To deplete tissue macrophages, mice were injected i.v.with 200 �l of Cl2MBP-L (containing 1 mg of Cl2MBP) suspended in PBS2 days before LPS challenge as described previously (22). As a control,mice were treated i.v. with 200 �l of liposome-encapsulated PBS (PBS-L).For depletion of granulocytes, mice were treated i.p. with 150 �g of anti-Ly6G mAb 1 day before LPS challenge as described previously (23). Todeplete NK1� cells, mice were treated i.p. with 500 �g of anti-NK1.1 mAb2 days before LPS challenge. Depletion of Kupffer cells (�99%), splenicmacrophages (�95%), granulocytes (�95%), and NK1� cells (�98%) inthe liver was verified by immunohistochemistry and/or flow cytometry. Forneutralization of endogenous IL-10 or TNF-�, mice were treated i.p. with500 �g of anti-IL-10 mAb or anti-TNF-� mAb, respectively, 2 h beforeLPS challenge. Isotype-matched mAbs purified by the same procedure asthat described for specific mAbs or PBS used for mAb purification wereused as a control, and it was verified that the reaction produced by mAbtreatment is not caused by LPS contamination of mAbs or PBS.
Immunohistochemistry
For immunohistochemistry, specimens were embedded in Tissue-Tek, fro-zen, and cut on a Cryotome. Sections (3–5 �m) were air-dried, fixed withacetone, rehydrated, and treated with blocking buffer (PBS containing 1%BSA and 0.05% Tween 20) for 15 min. Sections were then incubated withmAbs against Ly6G, CD41, or F4/80 diluted in blocking buffer at a con-centration of 10 �g/ml at 37°C for 30 min. After washing with PBS, thesections were incubated with Cy2-conjugated goat anti-rat IgG. For im-munodetection in tissues of animals pretreated with anti-IL-10 mAb, pri-mary Abs were labeled with Cy2-conjugated Fab of anti-rat IgG and thenallowed to react with tissue sections.
Statistical analysis
Statistical significance was determined by log-rank test (survival time) orpost hoc multiple range test (serum cytokine levels and frequencies ofcytokine-producing cells). A value of p � 0.05 was regarded as significant.
ResultsLFA-1�/� mice are resistant to low dose LPS-induced lethalshock
We compared LFA-1�/�, LFA-1�/�, ICAM-1�/�, and C57BL/6mice for susceptibility to low dose LPS-induced shock. Mice werechallenged with different doses of LPS together with D-GalN, andsurvival times were monitored thereafter. All C57BL/6 mice suc-cumbed to challenge with 0.01 �g of LPS within 8 h (Table I). AllLFA-1�/� and ICAM-1�/� mice also succumbed to 0.01-�g LPSchallenge, although the survival times were slightly prolonged(mean survival time, 10 h). In contrast, LFA-1�/� mice resisted upto 1000 times higher LPS doses compared with LFA-1�/�, ICAM-1�/�, and C57BL/6 mice (Table I). Similar results were obtainedwith highly purified S. abortus equi-derived LPS (data not shown),verifying that the reaction was induced by LPS, but not other con-taminating components such as lipoprotein, which can be presentin commercially obtained LPS (24). Thus, LFA-1�/� mice werehighly resistant to low dose LPS-induced lethal shock.
585The Journal of Immunology
LFA-1�/� mice are resistant to low dose LPS-induced liverinjury
Mice were challenged with LPS (1 �g) together with D-GalN (8mg), liver tissues were prepared at different time points thereafter,and histopathology was analyzed by H&E and TUNEL staining.No measurable alterations were found in liver sections fromC57BL/6 mice by 4.5 h after LPS challenge (data not shown). Yet,liver sections from C57BL/6 mice at 6 h after LPS challenge dis-played characteristic features of hepatocyte destruction, includingpyknosis and karyorrhexis of hepatocyte nuclei, as well as exten-sive parenchymal hemorrhage, whereas areas of necrosis werescarcely detected (Fig. 1A). In LFA-1�/� mice, hepatocyte de-struction was marginal at 6 h after challenge, but severe hepatocyte
destruction was found at 8 h, comparable to that in C57BL/6 miceat 6 h after challenge (Fig. 1A). Results in ICAM-1�/� mice weresimilar to those in LFA-1�/� mice (data not shown). In contrast,only marginal signs of liver injury were detected in LFA-1�/�
mice at 8 h after LPS challenge (Fig. 1A), and no exacerbation wasfound thereafter (data not shown). High numbers of parenchymalcell nuclei were positively stained in C57BL/6 mice at 6 h afterLPS challenge by the TUNEL method, suggesting apoptotic deathof these cells (Fig. 1B). Marginal signs of apoptosis were detectedin parenchymal cells from LFA-1�/� mice at 6 h after LPS chal-lenge, but high numbers of parenchymal cell nuclei were positivelystained at 8 h (Fig. 1B), and this was also true in ICAM-1�/� mice(data not shown). In contrast, only marginal signs of apoptosis
Table I. Mortality rates of various mouse strains following challenge with D-GalN plus low dose LPS
a Groups of five mice were challenged i.v. with various doses of LPS and/or D-GalN (8 mg) and mortality rates were scored on day 3. Determinations were performed twicewith comparable results.
b p � 0.01, �2m�/� vs C57BL/6.c p � 0.001, LFA-1�/� vs LFA-1�/�.d p � 0.05, ICAM-1�/� vs C57BL/6; LFA-1�/� vs C57BL/6.
FIGURE 1. Liver injury in LFA-1�/�, LFA-1�/�, and C57BL/6 mice following challenge with D-GalN plus low dose LPS. Mice were injected i.v. withLPS (1 �g) together with D-GalN (8 mg), and livers were collected at 6 and/or 8 h. Livers were cryosectioned and stained with H&E (A), TUNEL (B),anti-Ly6G mAb (C), or F4/80 mAb (D). Bar � 50 �m.
586 ROLE OF LFA-1 IN LIVER INJURY
were found in the liver of LFA-1�/� mice at 8 h after LPS chal-lenge (Fig. 1B), and no exacerbation occurred at later time points(data not shown). LPS (1 �g) or D-GalN (8 mg) alone did not causeany alterations in livers, and treatment with both LPS (1 �g) andD-GalN (8 mg) did not affect other organs, including spleen (datanot shown). Thus, LFA-1�/� mice were highly resistant to lowdose LPS-induced liver injury/hepatocyte apoptosis.
Elevated serum levels of TNF-� and IL-12 in LFA-1�/� micefollowing low dose LPS challenge
TNF-� plays a central role in low dose LPS-induced shock (6–9).We therefore compared serum levels of TNF-� in LFA-1�/�,LFA-1�/�, and C57BL/6 mice following challenge with LPS to-gether with D-GalN. Since IFN-� is involved in low dose LPS-induced shock (10), and this cytokine is induced by IL-12 (25),serum levels of these cytokines were analyzed as well. Serum lev-els of TNF-�, IL-12 (p40), IL-12 (p70), and IFN-� in C57BL/6mice peaked at 1, 4, 4, and 6 h, respectively, after LPS challenge(data not shown). We therefore compared cytokine levels in thesera of LFA-1�/�, LFA-1�/�, and C57BL/6 mice at the respectivetime points. Serum levels of TNF-� and IL-12 (p40) were signif-icantly higher in LFA-1�/� mice than in LFA-1�/� and C57BL/6mice after challenge (Fig. 2A). Serum levels of IL-12 (p70) werealso significantly higher in LFA-1�/� mice than in LFA-1�/� andC57BL/6 mice, although the levels were markedly lower thanIL-12 (p40) levels (Fig. 2A). In contrast, serum levels of IFN-�slightly increased in these mouse strains following challenge, andthe levels were comparable among these mouse strains (data notshown). Thus, serum levels of TNF-� and IL-12 following lowdose LPS challenge were elevated in LFA-1�/� mice, althoughthese mice resisted low dose LPS-induced shock.
Elevated levels of TNF-� and IL-12 in LFA-1�/� mice correlatewith numerical increase of cytokine-producing cells
We compared numbers of TNF-�- and IL-12-producing cells inblood, liver, and spleen of LFA-1�/� and LFA-1�/� mice follow-ing low dose LPS challenge. Before challenge, TNF-� and IL-12(p40) producers were low in both mouse strains, and no significantdifference was found in these mice (data not shown). The frequen-cies of TNF-� and IL-12 (p40) producers among hepatic leuko-cytes and splenocytes were markedly increased in both mousestrains following low dose LPS challenge, and they were signifi-cantly higher in LFA-1�/� mice than in LFA-1�/� mice (Fig. 2B).Whereas high frequencies of TNF-�-producing cells were detectedamong peripheral blood leukocytes in both LFA-1�/� and LFA-1�/� mice following LPS challenge, those of IL-12 (p40) produc-ers were low in both mouse strains (Fig. 2B). The frequencies ofTNF-�- and IL-12 (p40)-producing cells were virtually compara-ble among peripheral blood leukocytes in LFA-1�/� and LFA-1�/� mice. Thus, higher levels of TNF-� and IL-12 in LFA-1�/�
mice following low dose LPS challenge correlate with higher num-bers of the respective cytokine-producing cells in liver and spleen.
V�14�NKT cells, granulocytes, and tissue macrophages are notessential for low dose LPS-induced shock/liver injury in thepresence of LFA-1
V�14�NKT cells have been shown to play a crucial role in theinduction of liver injury using various experimental systems (26–30). We and others have previously shown that the numbers ofV�14�NKT cells are markedly reduced in the liver of LFA-1�/�
mice (31, 32). We therefore wondered whether increased resis-tance of LFA-1�/� mice to low dose LPS-induced shock was dueto the reduction of V�14�NKT cells in the liver. To clarify thisissue, the susceptibilities of TCR��/�, �2m�/�, and CD1d�/�
mice, all of which are devoid of V�14�NKT cells (20, 33–36), tolow dose LPS-induced shock were compared. The susceptibilitiesof TCR��/� and CD1d�/� mice to low dose LPS-induced shockwere comparable to that of C57BL/6 mice, and susceptibility wasslightly increased in �2m�/� mice (Table I). Consistent with this,C57BL/6 mice were not rescued from the lethal effects of LPS byin vivo depletion of NK1� cells (Table II). These results not onlyexclude the possibility that increased resistance of LFA-1�/� miceto low dose LPS-induced shock is caused by reduced numbers ofV�14�NKT cells in the liver, but also suggest that neither con-ventional T cells nor NK cells are required in low dose LPS-in-duced shock.
Since granulocytes have been suggested to participate in lowdose LPS-induced shock (13–15), we compared numbers of gran-ulocytes in the liver following challenge with LPS and D-GalNbetween LFA-1�/� and LFA-1�/� mice. Comparable numbers ofLy6G� cells were detected in the liver of LFA-1�/� and LFA-
FIGURE 2. Levels of TNF-�, IL-12, and IL-10 in sera, and frequenciesof these cytokine-producing cells in peripheral blood, liver, and spleen ofLFA-1�/�, LFA-1�/�, and/or C57BL/6 mice following challenge with D-GalN plus low dose LPS. Mice were injected i.v. with LPS (1 �g) togetherwith D-GalN (8 mg). A, Sera were collected at 1 h (for TNF-�), 4 h (forIL-12 (p40, p70)), and 2 h (for IL-10) after challenge, and serum levels ofthese cytokines were determined by ELISA. Each marker represents serumlevels of cytokine in an individual animal. The horizontal lines indicatemean serum levels. Before LPS challenge, the above cytokines were vir-tually undetectable in sera. B, Peripheral blood, liver, and spleen werecollected at 20 min (for TNF-�), 2.5 h (for IL-12 (p40, p70)), and 1 h (forIL-10) after challenge, and PBL, hepatic leukocytes, and splenocytes wereprepared thereafter. The frequencies of these cytokine-producing cells weredetermined by ELISPOT assay. Data represent the mean of five mice pergroup. SFC, spot-forming cells. �, p � 0.01, LFA-1�/� vs LFA-1�/�.
587The Journal of Immunology
1�/� mice before LPS challenge (Fig. 1C). Although numbers ofLy6G� cells were markedly increased in the liver of both mousestrains following LPS challenge, they were slightly higher in LFA-1�/� mice than in heterozygous littermates (Fig. 1C). Susceptibil-ity of C57BL/6 mice to low dose LPS-induced shock was virtuallyunchanged by in vivo depletion of granulocytes (Table II). Theseresults exclude a role of granulocytes in low dose LPS-inducedshock and indicate a dispensable role of LFA-1 in the infiltrationof granulocytes in the liver.
Because macrophages have been considered to play a major rolein low dose LPS-induced shock (37–39), we compared numbers ofKupffer cells following challenge with LPS and D-GalN betweenLFA-1�/� and LFA-1�/� mice. Comparable numbers of F4/80�
cells were detected in the liver of homozygous and heterozygousmouse mutants before LPS challenge, which were equally dimin-ished in both mouse strains following LPS challenge (Fig. 1D).The susceptibility of C57BL/6 mice to low dose LPS-inducedshock was slightly, although significantly, reduced by in vivo de-pletion of tissue macrophages (Table II). These results argueagainst a critical role of Kupffer cells in low dose LPS-inducedshock, at least in the presence of LFA-1.
Markedly reduced numbers of platelets in the liver of LFA-1�/�
mice following low dose LPS challenge
Blood coagulation is a major event at the terminal stage of endotox-emia (40, 41). Because LFA-1 is expressed on platelets (42), we com-pared numbers of platelets in the liver following challenge with LPSand D-GalN between LFA-1�/� and LFA-1�/� mice. Immunohisto-chemical analysis of liver sections revealed an early prominent accu-mulation of platelets in the liver of LFA-1�/� mice following LPSchallenge (Fig. 3). In contrast, the numbers of platelets were onlymarginally increased in LFA-1�/� mice following LPS challenge(Fig. 3). Thus, in the absence of LFA-1, the numbers of platelets weremarkedly reduced in livers following low dose LPS challenge.
Elevated serum levels of IL-10 in LFA-1�/� mice following lowdose LPS challenge
IL-10 plays a protective role against both high dose (43–45) andlow dose LPS-induced shock (46, 47). To determine whether IL-10participates in the increased resistance of LFA-1�/� mice to lowdose LPS-induced shock, serum levels of IL-10 were compared inLFA-1�/�, LFA-1�/�, and C57BL/6 mice following challengewith LPS and D-GalN. Since serum levels of IL-10 peaked at 2 hafter LPS challenge in C57BL/6 mice (data not shown), serumlevels of IL-10 were compared at this time point. In these mousestrains, IL-10 was undetectable in sera before LPS challenge. After
LPS challenge, serum levels of IL-10 were significantly higher inthe absence of LFA-1 (Fig. 2A).
Elevated levels of IL-10 in LFA-1�/� mice correlate with thenumerical increase in IL-10-producing cells
We compared numbers of IL-10-producing cells in blood, liver,and spleen of LFA-1�/� and LFA-1�/� mice following low doseLPS challenge. IL-10 producers were virtually undetectable inboth mouse strains before challenge (data not shown). In contrast,the frequencies of IL-10 producers among peripheral blood leuko-cytes, hepatic leukocytes, and splenocytes were markedly in-creased in both mouse strains following low dose LPS challenge,and they were significantly higher in LFA-1�/� mice than in LFA-1�/� mice (Fig. 2B). Thus, higher levels of IL-10 in LFA-1�/�
mice following low dose LPS challenge correlate with higher num-bers of IL-10 producers.
Endogenous IL-10 neutralization increases the susceptibility ofLFA-1�/� mice to low dose LPS-induced shock
The susceptibility of LFA-1�/� and C57BL/6 mice to low doseLPS-induced shock was virtually unchanged by endogenous IL-10neutralization (Table III). In contrast, the susceptibility of LFA-1�/� mice to low dose LPS-induced shock was increased up to1000-fold (Table III), and the numbers of platelets in the liver wereelevated (Fig. 3). These results indicate a direct relationship be-tween resistance and elevated IL-10 levels in low dose LPS-in-duced shock in the absence of LFA-1.
FIGURE 3. Accumulation of platelets in the liver of LFA-1�/� andLFA-1�/� mice following challenge with D-GalN plus low dose LPS andthe influence of endogenous IL-10 neutralization. Mice were left untreatedor were treated i.p. with anti-IL-10 mAb (500 �g) and 2 h later injected i.v.with LPS (1 �g) together with D-GalN (8 mg). Livers were collected at 8 hafter LPS challenge, fixed in 4% paraformaldehyde in PBS, sectioned, andstained with anti-CD41 mAb. Bar � 50 �m.
Table II. Influence of in vivo depletion of NK11 cells, granulocytes, or Kupffer cells on mortality rates of C57BL/6 and/or LFA-1�/� mice followingchallenge with D-GalN plus low dose LPS
D-GalN (mg) LPS (�g)
Mortality Rate (%)a
C57BL/6 LFA-1�/�
Anti-NK1.1 mAb Anti-Ly6G mAb PBS C12MBP-L PBS-L PBS C12MBP-L PBS-L
a Mice were treated i.p. with anti-NK1.1 mAb (500 �g), anti-Ly6G mAb (150 �g), C12MBP-L (200 �l), or PBS-L (200 �l) 2, 1, 2, or 2 days, respectively, before challengewith various doses of LPS and D-GalN (8 mg), and mortality rates were scored on day 3 after challenge. Numbers in parentheses represent the numbers of mice examined. Sinceno significant difference was found among mouse IgG2a-treated (isotype-matched mAb for anti-NK1.1 mAb), rat IgG2a-treated (isotype-matched mAb for anti-Ly6G mAb), andPBS-treated groups in another experiment, PBS was used as a control.
b p � 0.05, C12MBP-L-treated group vs PBS-L-treated group.c p � 0.001, C12MBP-L-treated group vs PBS-L-treated group.
588 ROLE OF LFA-1 IN LIVER INJURY
Tissue macrophages as a source of elevated IL-10 levels inLFA-1�/� mice following low dose LPS challenge
TNF-�-induced release of IL-10 during endotoxemia has been re-ported (48). Because not only IL-10, but also TNF-�, serum con-centrations were elevated in LFA-1�/� mice following low doseLPS challenge, we assessed the influence of TNF-� neutralizationon IL-10. Serum levels of IL-10 were virtually unaffected by neu-tralization of endogenous TNF-� (Fig. 4). The in vivo efficacy ofanti-TNF-� mAb was verified in a parallel group of C57BL/6 miceby assessing susceptibility to LPS-induced shock. Neutralizationof TNF-� resulted in a �1000-fold increase in the resistance ofthese mice to low dose LPS-induced shock.
Recently, LPS was found to induce IL-10 secretion by mono-cytes in the presence of apoptotic granulocytes (49). Because thenumbers of granulocytes infiltrating the liver were higher in LFA-1�/� mice than in LFA-1�/� mice, we examined the influence ofgranulocyte depletion on serum levels of IL-10 following chal-lenge with LPS and D-GalN. In vivo depletion of granulocytes didnot induce IL-10 production by itself (data not shown), and serumlevels of IL-10 in LFA-1�/� mice following LPS challenge werevirtually unaffected by in vivo depletion of granulocytes (Fig. 4).These results argue against participation of TNF-� and granulo-cytes in the control of IL-10 secretion in LFA-1�/� mice.
Because macrophages secrete IL-10 in response to LPS (50, 51),we assessed the influence of tissue macrophage depletion on serumlevels of IL-10 following low dose LPS challenge. Tissue macro-
phage depletion did not result in measurable alterations in serumlevels of IL-10 in LFA-1�/� mice (Fig. 5). In contrast, in LFA-1�/� mice serum levels of IL-10 were significantly diminishedafter tissue macrophage depletion (Fig. 5). Note that serum levelsof IL-10 in tissue macrophage-depleted LFA-1�/� mice werecomparable to those in LFA-1�/� and C57BL/6 mice (see Fig.2A). These results suggest that tissue macrophages are responsiblefor higher levels of IL-10 in LFA-1�/� mice.
Tissue macrophage depletion markedly increases thesusceptibility of LFA-1�/� mice to low dose LPS-induced shockdespite reduced serum levels of TNF-�
We examined the influence of tissue macrophage depletion on thesusceptibility of LFA-1�/� mice to low dose LPS-induced shock.In contrast to C57BL/6 mice, the susceptibility of LFA-1�/� micewas �100-fold increased by tissue macrophage depletion (TableII). These results suggest that tissue macrophages have a protectiverole in LFA-1�/� mice to low dose LPS-induced shock. Tissue
FIGURE 4. Influence of endogenous TNF-� neutralization or granulo-cyte depletion on serum levels of IL-10 in LFA-1�/� mice following chal-lenge with D-GalN plus low dose LPS. Mice were treated i.p. with anti-TNF-� mAb (500 �g) or anti-Ly6G mAb (150 �g) 2 h or 1 day,respectively, before challenge with LPS (1 �g) together with D-GalN (8mg), and sera were collected at 2 h after challenge. Serum levels of IL-10were determined by ELISA. Each symbol represents serum levels of IL-10in an individual animal. The horizontal lines indicate mean serum levels.Since no significant difference was found in the rat IgG1-treated group(isotype-matched mAb for anti-TNF-� mAb), the rat IgG2b-treated group(isotype-matched mAb for anti-Ly6G mAb), and the PBS-treated group inanother experiment, PBS was used as a control. Before LPS challenge, theabove cytokines were virtually undetectable in sera.
FIGURE 5. Influence of tissue macrophage depletion on serum levels ofIL-10, TNF-�, and IL-12 in LFA-1�/� and/or LFA-1�/� mice followingchallenge with D-GalN plus low dose LPS. Mice were treated i.p. with 200�l of CL2 MBP-L or PBS 2 days before challenge with LPS (1 �g) to-gether with D-GalN (8 mg), and sera were collected at 2 h (for IL-10), 1 h(for TNF-�), or 4 h (for IL-12 (p70)) after challenge. Serum levels of thesecytokines were determined by ELISA. Each symbol represents serum lev-els of each cytokine in an individual animal. The horizontal lines indicatemean serum levels. Before LPS challenge, the above cytokines were vir-tually undetectable in sera.
Table III. Influence of endogenous IL-10 neutralization on mortality rates of LFA-1�/�, LFA-1�/�, and C57BL/6 mice following challenge with D-GalN plus low dose LPS
D-GalN (mg) LPS (�g)
Mortality Rate (%)a
C57BL/6 LFA-1�/� LFA-1�/�
PBS Anti-IL-10 mAb PBS Anti-IL-10 mAb PBS Anti-IL-10 mAb
a Groups of four mice were treated i.p. with anti-IL-10 mAb (500 �g) or PBS 2 h before challenge with various doses of LPS and D-GalN (8 mg), and mortality rates werescored on day 3 after challenge. Since no significant difference was found between rat IgG1-treated (isotype-matched mAb for anti-IL-10 mAb) and PBS-treated groups in anotherexperiment, PBS was used as a control. ND, not determined.
b p � 0.05, anti-IL-10 mAb-treated group vs PBS-treated group.
589The Journal of Immunology
macrophage depletion diminished serum levels of TNF-� follow-ing LPS challenge (Fig. 5). In contrast, serum levels of IL-12 (p70)were increased by tissue macrophage depletion. Note that the lev-els of TNF-� in tissue macrophage-depleted LFA-1�/� mice werecomparable to those in LFA-1�/� and C57BL/6 mice (see Fig.2A). Thus, the susceptibility of LFA-1�/� mice to low dose LPS-induced shock was markedly increased by tissue macrophage de-pletion despite reduced TNF-� levels.
DiscussionThis paper describes participation of LFA-1 in low dose LPS-induced shock/liver injury and points to IL-10 as critical mediatorof resistance to shock/liver injury in its absence. Although TNF-�is critical for low dose LPS-induced shock (6–9), and LPS is con-sidered a ligand for LFA-1 (12), serum levels of TNF-� weresignificantly higher in the resistant LFA-1�/� mice compared withcontrols. These results imply that a factor(s) downstream of TNF-�signaling participates in the increased resistance of LFA-1�/�
mice to low dose LPS and suggest that this increased resistance isnot caused by the absent interactions between LPS and LFA-1.Moreover, our data suggest dominant effects of the anti-inflammatory cytokine IL-10 over the proinflammatory cytokinesTNF-� and IL-12.
It could be speculated that increased resistance of LFA-1�/�
mice to low dose LPS was due to reduced numbers of V�14�NKTcells in the liver. However, mice deficient in V�14�NKT cellswere susceptible to low dose LPS, and the susceptibility ofC57BL/6 mice was only slightly affected by NK1� cell depletion.The susceptibility of �2m�/� mice to low dose LPS was slightlyhigher than that of other mouse strains. Following low dose LPSchallenge, serum levels of IFN-� in �2m�/� mice were signifi-cantly higher than those in other mouse strains (M. Emoto and Y.Emoto, unpublished observations). Hence, we consider it likelythat slightly increased susceptibility of �2m�/� mice to low doseLPS-induced shock is due to higher levels of IFN-�. In any case,our data suggest that a numerical reduction of V�14�NKT cells inthe liver is not responsible for the increased resistance of LFA-1�/� mice to low dose LPS-induced shock/liver injury.
Granulocytes infiltrate the liver in response to low dose LPSchallenge and play a critical role in low dose LPS-induced shock/liver injury (13–17). Although by 8 h after low dose LPS chal-lenge, granulocytes infiltrated the liver, the number of liver gran-ulocytes was slightly higher in LFA-1�/� mice than inheterozygous littermates. Moreover, susceptibility of C57BL/6mice to low dose LPS was virtually unchanged by in vivo deple-tion of granulocytes. Hence, we assume that granulocytes are notresponsible for increased resistance to low dose LPS in the absenceof LFA-1.
Our study does not formally exclude that increased resistance ofLFA-1�/� mice to low dose LPS was caused by a lack of inter-actions between LFA-1 and its physiological ligands. It has beenreported that anti-LFA-1 mAb treatment does not rescue micefrom low dose LPS-induced lethal shock (17). This finding raisesthe possibility that increased resistance of LFA-1�/� mice to lowdose LPS occurred independently from interactions betweenLFA-1 and its physiological ligands. We cannot exclude that themAb treatment failed to block LFA-1 interactions with a ligand(s)hidden within tissue or involving additional, unknown ligands andbinding epitopes. Consecutive administration of anti-LFA-1 mAbincreases the resistance of Propionibacterium acnes-primed miceto low dose LPS challenge (52). However, LFA-1 is expressed onvirtually all leukocytes, which could be depleted by anti-LFA-1
mAb treatment, resulting in increased resistance to LPS-inducedshock. Although in the mouse, ICAM-1 is one of the physiologicalligands for LFA-1 (11), we found that the susceptibility of ICAM-1�/� mice was comparable to that of controls. We assume thatincreased resistance of LFA-1�/� mice to low dose LPS is not adirect consequence of cognate interactions between LFA-1 andICAM-1 in situ. In addition to ICAM-1, ICAM-2 is a ligand forLFA-1 (11). Hence, we cannot exclude that interactions betweenLFA-1 and ICAM-2 are involved in this mechanism.
Mac-1 can compensate for the lack of LFA-1 (53). Moreover,Mac-1 expression is up-regulated on leukocytes following LPSchallenge, and this molecule participates in low dose LPS-inducedshock (16, 17). It is also possible that Mac-1 participates in in-creased resistance of LFA-1�/� mice to low dose LPS. However,we found that Mac-1 expression on various cell populations wascomparable in LFA-1�/� and LFA-1�/� mice even after LPSchallenge (M. Emoto and Y. Emoto, unpublished observations).We therefore consider it unlikely that increased resistance of LFA-1�/� mice to low dose LPS occurred independently of Mac-1expression.
IL-10 plays a protective role in low dose LPS-induced shock/liver injury (46, 47). After low dose LPS challenge, serum levelsof IL-10 were significantly, although modestly, higher in LFA-1�/� mice than in LFA-1�/� and C57BL/6 mice. Whereas endog-enous IL-10 neutralization only marginally increased the suscep-tibility of LFA-1�/� mice to low dose LPS, IL-10 neutralizationdiminished the resistance of LFA-1�/� mice by several orders ofmagnitude. Differential efficacy of IL-10 neutralization in thesemouse strains was probably due to different serum levels ofTNF-�. IL-10 has been found to limit TNF-� secretion after lowdose LPS challenge (47). Yet serum levels of both TNF-� andIL-10 were higher in LFA-1�/� mice than in heterozygous litter-mates. We therefore consider it unlikely that increased resistanceof LFA-1�/� mice to low dose LPS is a direct consequence ofimpaired TNF-� secretion by IL-10. The production of variousproinflammatory cytokines and chemokines is regulated by IL-10(54, 55), and IL-10 increases the secretion of soluble TNF-� re-ceptor p55 (56). It is thus possible that LFA-1 participates in thesemechanisms by controlling IL-10 secretion.
IFN-� participates in low dose LPS-induced shock/liver injury(10). Because serum levels of IFN-� were comparable in LFA-1�/�, LFA-1�/�, and C57BL/6 mice, we assume that IFN-� is notresponsible for the increased resistance of LFA-1�/� mice to lowdose LPS-induced shock/liver injury. At present, we cannot pro-vide a conclusive answer for why serum levels of IFN-� werecomparable in LFA-1�/�, LFA-1�/�, and C57BL/6 mice after lowdose LPS challenge, although IL-12 levels were higher in LFA-1�/� mice than in other mouse strains. However, IL-10 preventsIFN-� production (57, 58), and it is possible that IFN-� productionin LFA-1�/� mice was prevented by higher levels of IL-10.
Blood coagulation is a major event at the terminal stage of en-dotoxemia (40, 41). 1) Platelets play a pivotal role in the bloodcoagulation cascade; 2) a critical role of platelets in low dose LPS-induced shock has been suggested (59); 3) LFA-1 is expressed onplatelets (42); and 4) high numbers of platelets were detected in theliver of LFA-1�/� mice following low dose LPS challenge,whereas platelet accumulation in LFA-1�/� mice was marginal.Hence, it is possible that LFA-1 expressed on platelets directlyparticipates in blood coagulation during endotoxemia. However,endogenous IL-10 neutralization markedly increased the numbersof platelets in the liver of LFA-1�/� mice following low dose LPSchallenge, and an inhibitory role of IL-10 in fibrin formation hasbeen reported (60–64). Therefore, we consider it more likely that
590 ROLE OF LFA-1 IN LIVER INJURY
higher levels of IL-10 increased the resistance of LFA-1�/� miceto low dose LPS by impairing the blood coagulation cascade.
Serum levels of IL-12 in LFA-1�/� mice following LPS chal-lenge were further increased by tissue macrophage depletion. Thisraises the question of whether increased levels of IL-12 are re-sponsible for the increased susceptibility of tissue macrophage-depleted LFA-1�/� mice to low dose LPS-induced shock. Yet thesusceptibility of tissue macrophage-depleted LFA-1�/� mice tolow dose LPS-induced shock/liver injury was unchanged by en-dogenous IL-12 neutralization (M. Emoto and Y. Emoto, unpub-lished observation). Hence, the role and cellular source of IL-12 inlow dose LPS-induced shock/liver injury of LFA-1�/� mice re-main elusive.
Depletion of tissue macrophages did not significantly reduce theserum levels of IL-10 in LFA-1�/� mice, arguing against tissuemacrophages as a major source of IL-10 during endotoxemia in thepresence of LFA-1. In contrast, in LFA-1�/� mice serum levels ofIL-10 were markedly reduced by tissue macrophage depletion. Se-rum levels of IL-10 were comparable in tissue macrophage-de-pleted LFA-1�/� mice and nondepleted LFA-1�/� and C57BL/6mice following low dose LPS challenge, and higher frequencies ofIL-10 producers were detected in LFA-1�/� mice compared withLFA-1�/� mice. We conclude that tissue macrophages are respon-sible for higher levels of IL-10 in the absence of LFA-1.
The susceptibility of C57BL/6 mice to low dose LPS was onlyslightly reduced by in vivo depletion of tissue macrophages. More-over, considerable levels of TNF-� were detected in the sera oftissue macrophage-depleted C57BL/6 mice following low doseLPS challenge (M. Emoto and Y. Emoto, unpublished observa-tions). Hence, we assume that in wild-type mice cells other thantissue macrophages play a central role in low dose LPS-inducedshock/liver injury by producing TNF-�. In contrast, in LFA-1�/�
mice tissue macrophage depletion reduced serum levels of TNF-�.Serum levels of TNF-� were comparable in tissue macrophage-depleted LFA-1�/� mice and nondepleted LFA-1�/� andC57BL/6 mice following low dose LPS challenge, and high fre-quencies of TNF-� producers were detected in LFA-1�/� micecompared with LFA-1�/� mice. We conclude that tissue macro-phages are a major source not only of the anti-inflammatory cy-tokine IL-10 but also of the proinflammatory cytokine TNF-� inthe absence of LFA-1.
Because the susceptibility of LFA-1�/� mice to low dose LPS-induced shock was markedly increased by tissue macrophage de-pletion despite the fact that not only IL-10, but also TNF-�, werediminished, we consider it likely that the host is rescued fromLPS-induced lethal shock/liver injury when IL-10 levels exceed acertain threshold level even in the presence of elevated levels ofTNF-�. Consistent with this idea, our additional experiments re-vealed that serum levels of IL-10 in the susceptible mouse strainswere lower than those in LFA-1�/� mice (M. Emoto and Y.Emoto, unpublished observations; see also Fig. 2A).
The present study does not conclusively answer the question ofwhy IL-10 is elevated in the absence of LFA-1 following low doseLPS challenge. TNF-� is a major mediator of the cytokine cascadethat leads to endotoxic shock, and this cytokine has been suggestedto participate in the release of IL-10 during endotoxemia (48).However, in our hands serum levels of IL-10 were virtually unaf-fected by endogenous TNF-� neutralization, which is consistentwith previous findings by others (65). It is therefore possible thata factor(s) independent from TNF-� signaling participates in IL-10production in LFA-1�/� mice. LPS has recently been shown toinduce the prompt release of IL-10 from monocytes in the presence
of apoptotic granulocytes (49). However, granulocyte depletiondid not affect serum levels of IL-10 in LFA-1�/� mice. We there-fore consider it unlikely that elevated levels of IL-10 in LFA-1�/�
mice were caused by increased numbers of granulocytes.Mice deficient in LFA-1, CD18, P-selectin, L-selectin, and/or
E-selectin display leukocytosis (66–72), and serum levels of G-CSF and IL-17 are elevated in mice deficient in CD18, P-selectin,and/or E-selectin mice (72). We have recently shown that numbersof leukocytes in peripheral blood and serum levels of G-CSF andIL-17 are markedly increased in LFA-1�/� mice (73). It is there-fore possible that the higher levels of pro- and anti-inflammatorycytokines in LFA-1�/� mice are a consequence of leukocytosisand altered regulatory interactions as observed in other mousestrains deficient in cell adhesion molecules. It is tempting to as-sume that leukocytosis at least in part is responsible for the ele-vated levels of IL-10 in LFA-1�/� mice following low dose LPSchallenge.
Our results show that LFA-1 deficiency confers resistance tolow dose LPS-induced shock/liver injury, and that LFA-1 defi-ciency results in higher IL-10 production in response to low doseLPS. Elevated serum levels of the anti-inflammatory cytokineIL-10 and the proinflammatory cytokines TNF-� and IL-12 inLFA-1�/� mice suggest a dominant role for the inhibitory cyto-kine IL-10 over the proinflammatory cytokines TNF-� and IL-12.Hence, IL-10 is the critical mediator of resistance to low doseLPS-induced shock/liver injury as a corollary of LFA-1 deficiency.In summary, therefore, our findings define a novel role of the celladhesion molecule LFA-1 in the regulation of the proinflamma-tory/anti-inflammatory cytokine balance.
AcknowledgmentsWe thank Drs. Marina A. Freudenberg, Rudolf Schmits, and Albert Ben-delac for S. abortus equi-derived LPS and breeding pairs of LFA-1�/� andCD1d�/� mice, respectively. We are grateful to Daniela Groine-Triebkornfor screening of mice, and to Beatrix Fauler and Ulrike Reichard for helpwith histological procedures.
References1. Glauser, M. P., G. Zanetti, J. D. Baumgartner, and J. Cohen. 1991. Septic shock:
pathogenesis. Lancet 338:732.2. Beutler, B., and A. Cerami. 1989. The biology of cachectin/TNF-� primary me-
diator of the host response. Annu. Rev. Immunol. 7:625.3. Gutierrez-Ramos, J. C., and H. Bluethmann. 1997. Molecules and mechanisms
operating in septic shock: lessons from knockout mice. Immunol. Today 18:329.4. Hack, C. E., L. A. Aarden, and L. G. Thijs. 1997. Role of cytokines in sepsis. Adv.
Immunol. 66:101.5. Galanos, C., M. A. Freudenberg, and W. Reutter. 1979. Galactosamine-induced
sensitization to the lethal effects of endotoxin. Proc. Natl. Acad. Sci. USA 76:5939.
6. Lehmann, V., M. A. Freudenburg, and C. Galanos. 1987. Lethal toxicity of li-popolysaccharide and tumor necrosis factor in normal and D-galactosaminetreated mice. J. Exp. Med. 165:657.
7. Tiegs, G., M. Wolter, and A. Wendel. 1989. Tumor necrosis factor is a terminalmediator in D-galactosamine/endotoxin-induced hepatitis in mice. Biochem.Pharmacol. 38:627.
8. Pfeffer, K., T. Matsuyama, T. M. Kundig, A. Wakeham, K. Kishihara,A. Shahinian, K. Wiegmann, P. S. Ohashi, M. Kronke, and T. W. Mak. 1993.Mice deficient for the 55 kd tumor necrosis factor receptor are resistant to en-dotoxin shock, yet succumb to L. monocytogenes infection. Cell 73:457.
9. Rothe, J., W. Lesslauer, H. Lotscher, Y. Lang, P. Koebel, F. Kotgen, A. Althage,R. Zinkernagel, M. Steinmetz, and H. Bluethmann. 1993. Mice lacking the tumornecrosis factor receptor 1 are resistant to TNF-mediated toxicity but highly sus-ceptible to infection by Listeria monocytogenes. Nature 364:798.
10. Car, B. D., V. M. Eng, B. Schnyder, L. Ozmen, S. Huang, P. Gallay,D. Heumann, M. Aguet, and B. Ryffel. 1994. Interferon � receptor deficient miceare resistant to endotoxic shock. J. Exp. Med. 179:1437.
11. Springer, T. A. 1995. Trafic signals on endothelium for lymphocyte recirculationand leukocyte emigration. Annu. Rev. Physiol. 57:827.
12. Wright, S. D., and M. T. C. Jong. 1986. Adhesion-promoting receptors on humanmacrophages recognize Escherichia coli by binding to lipopolysaccharide.J. Exp. Med. 164:1876.
591The Journal of Immunology
13. Hewett, J. A., P. A. Jean, S. L. Kunkel, and R. A. Roth. 1993. Relationshipbetween tumor necrosis factor and neutrophils in endotoxin-induced liver injury.Am. J. Physiol. 265:G1011.
14. Komatsu, Y., Y. Shiratori, T. Kawase, N. Hashimoto, K. Han, S. Shiina,M. Matsumura, Y. Niwa, M. Tada, Y. Ikeda, et al. 1994. Role of polymorpho-nuclear leukocytes in galactosamine hepatitis: mechanism of adherence to hepaticendothelial cells. Hepatology 20:1548.
15. Jaeschke, H., and C. W. Smith. 1997. Mechanisms of neutrophil-induced paren-chymal cell injury. J. Leukocyte Biol. 61:647.
16. Jaeschke, H., A. Farhood, and C. W. Smith. 1991. Neutrophil-induced liver cellinjury in endotoxin shock is a CD11b/CD18-dependent mechanism. Am. J.Physiol. 261:G1051.
17. Chang, H. R., C. Vesin, G. E. Grau, P. Pointaire, D. Arsenijevic, M. Strath, J.-C.Pechere, and P.-F. Piguet. 1993. Respective role of polymorphonuclear leuko-cytes and their integrins (CD11/18) in the local or systemic toxicity of lipopoly-saccharide. J. Leukocyte Biol. 53:636.
18. Schmits, R., T. M. Kundig, D. M. Baker, G. Shumaker, J. J. L. Simard,G. Duncan, A. Wakeham, A. Shahinian, A. van der Heiden, M. F. Bachmann, etal. 1996. LFA-1-deficient mice show normal CTL responses to virus but fail toreject immunogenic tumor. J. Exp. Med. 183:1415.
19. Emoto, M., M. Miyamoto, I. Yoshizawa, Y. Emoto, U. E. Schaible, E. Kita, andS. H. E. Kaufmann. 2002. Critical role of NK cells rather than V�14�NKT cellsin lipopolysaccharide-induced lethal shock in mice. J. Immunol. 169:1426.
20. Emoto, M., Y. Emoto, and S. H. E. Kaufmann. 1995. IL-4 producing CD4� TCR��int liver lymphocytes: influence of thymus, �2-microglobulin and NK1.1 ex-pression. Int. Immunol. 7:1729.
21. Emoto, Y., M. Emoto, and S. H. E. Kaufmann. 1997. Transient control of inter-leukin-4-producing natural killer T cells in the livers of Listeria monocytogenes-infected mice by interleukin-12. Infect. Immun. 65:5003.
22. van Rooijen, N., and A. Sanders. 1994. Liposome mediated depletion of macro-phages: mechanism of action, preparation of liposomes and applications. J. Im-munol. Methods 174:83.
23. Seiler, P., P. Aichele, B. Raupach, B. Odermatt, U. Steinhoff, andS. H. E. Kaufmann. 2000. Rapid neutrophil response controls fast-replicatingintracellular bacteria but not slow-replicating Mycobacterium tuberculosis. J. In-fect. Dis. 181:671.
24. Hirschfeld, M., Y. Ma, J. H. Weis, S. N. Vogel, and J. J. Weis. 2000. Repurifi-cation of lipopolysaccharide eliminates signaling through both human and murineToll-like receptor 2. J. Immunol. 165:618.
25. Trinchieri, G. 1995. Interleukin-12: a proinflammatory cytokine with immuno-regulatory functions that bridge innate resistance and antigen-specific adaptiveimmunity. Annu. Rev. Immunol. 13:251.
26. Toyabe, S., S. Seki, T. Iiai, K. Takeda, K. Shirai, H. Watanabe, H. Hiraide,M. Uchiyama, and T. Abo. 1997. Requirement of IL-4 and liver NK1� T cells forconcanavalin A-induced hepatic injury in mice. J. Immunol. 159:1537.
27. Takeda, K., Y. Hayakawa, L. van Kaer, H. Matsuda, H. Yagita, and K. Okumura.2000. Critical contribution of liver natural killer T cells to a murine model ofhepatitis. Proc. Natl. Acad. Sci. USA 97:5498.
28. Kaneko, Y., M. Harada, T. Kawano, M. Yamashita, Y. Shibata, F. Gejyo,T. Nakayama, and M. Taniguchi. 2000. Augmentation of V�14 NKT cell-me-diated cytotoxicity by interleukin 4 in an autocrine mechanism resulting in thedevelopment of concanavalin A-induced hepatitis. J. Exp. Med. 191:105.
29. Ogasawara, K., K. Takeda, W. Hashimoto, M. Satoh, R. Okuyama, N. Yanai,M. Obinata, K. Kumagai, H. Takada, H. Hiraide, et al. 1998. Involvement ofNK1� T cells and their IFN-� production in the generalized Shwartzman reac-tion. J. Immunol. 160:3522.
30. Dieli, F., G. Sireci, D. Russo, M. Taniguchi, J. Ivanyi, C. Fernandez,M. Troye-Blomberg, G. de Leo, and A. Salerno. 2000. Resistance of natural killerT cell-deficient mice to systemic Shwartzman reaction. J. Exp. Med. 192:1645.
31. Ohteki, T., C. Maki, S. Koyasu, T. W. Mak, and P. S. Ohashi. 1999. LFA-1 isrequired for liver NK1.1� TCR�/�� cell development: evidence that liverNK1.1� TCR�/�� cells originate from multiple pathways. J. Immunol. 162:3753.
32. Emoto, M., H.-W. Mittrucker, R. Schmits, T. W. Mak, and S. H. E. Kaufmann.1999. Critical role of leukocyte function-associated antigen-1 in liver accumula-tion of CD4�NKT cells. J. Immunol. 162:5094.
33. Bendelac, A., M. N. Rivera, S. Park, and J. H. Roark. 1997. Mouse CD1-specificNK1 T cells: development, specificity, and function. Annu. Rev. Immunol. 15:535.
34. Mendiratta, S. K., W. D. Martin, S. Hong, A. Boesteanu, S. Joyce, andL. van Kaer. 1997. CD1d1 mutant mice are deficient in natural T cells thatpromptly produce IL-4. Immunity 6:469.
35. Smiley, S. T., M. H. Kaplan, and M. J. Grusby. 1997. Immunoglobulin E pro-duction in the absence of interleukin-4-secreting CD1-dependent cells. Science275:977.
36. Chen, Y., N. M. Chiu, M. Mandal, N. Wang, and C. Wang. 1997. Impaired NK1�
T cell development and early IL-4 production in CD1-deficient mice. Immunity6:459.
37. Chojkier, M., and J. Fierer. 1985. D-Galactosamine hepatotoxicity is associatedwith endotoxin sensitivity and mediated by lymphoreticular cells in mice. Gas-troenterology 88:115.
38. Freudenberg, M. A., D. Keppler, and C. Galanos. 1986. Requirement for lipopo-lysaccharide-responsive macrophages in galactosamine-induced sensitization toendotoxin. Infect. Immun. 51:891.
39. Shiratori, Y., T. Kawase, S. Shiina, K. Okano, T. Sugimoto, H. Teraoka,S. Matano, K. Matsumoto, and K. Kamii. 1988. Modulation of hepatotoxicity bymacrophages in the liver. Hepatology 8:815.
40. McKay, D. G., W. Margaretten, and I. Csavossy. 1966. An electron microscopestudy of the effects of bacterial endotoxin on the blood-vascular system. Lab.Invest. 15:1815.
41. Dhainaut, J. F., N. Marin, A. Mignon, and C. Vinsonneau. 2001. Hepatic re-sponse to sepsis: interaction between coagulation and inflammatory processes.Crit. Care Med. 29:S42.
42. McCaffery, P. J., and M. V. Berridge. 1986. Expression of the leukocyte func-tional molecule (LFA-1) on mouse platelets. Blood 67:1757.
43. Gerard, C., C. Bruyns, A. Marchant, D. Abramowicz, P. Vandenabeele,A. Delvaux, W. Fiers, M. Goldman, and T. Velu. 1993. Interleukin-10 reducesthe release of tumor necrosis factor and prevents lethality in experimental endo-toxemia. J. Exp. Med. 177:547.
44. Howard, M., T. Muchamuel, S. Andrade, and S. Menon. 1993. Interleukin-10protects mice from lethal endotoxemia. J. Exp. Med. 177:1205.
45. Marchant, A., C. Bruyns, P. Vandenabeele, M. Ducarme, C. Gerard, D. Delvaux,D. de Groote, D. Abramowicz, T. Velu, and M. Goldman. 1994. Interleukin-10controls interferon-gamma and tumor necrosis factor production during experi-mental endotoxemia. Eur. J. Immunol. 24:1167.
46. Santucci, L., S. Fiorucci, M. Chiorean, P. M. Brunori, F. M. Di Matteo, A. Sidoni,G. Migliorati, and A. Morelli. 1996. Interleukin 10 reduces lethality and hepaticinjury induced by lipopolysaccharide in galactosamine-sensitized mice. Gastro-enterology 111:736.
47. Louis, H., O. Le Moine, M.-O. Peny, B. Gulbis, F. Nisol, M. Goldman, andJ. Deviere. 1997. Hepatoprotective role of interleukin 10 in galactosamine/lipo-polysaccharide mouse liver injury. Gastroenterology 112:935.
48. van der Poll, T., J. Jansen, M. Levi, H. ten Cate, J. W. ten Cate, andS. J. van Deventer. 1994. Regulation of interleukin 10 release by tumor necrosisfactor in humans and chimpanzees. J. Exp. Med. 180:1985.
49. Byrne, A., and D. J. Reen. 2002. Lipopolysaccharide induces rapid production ofIL-10 by monocytes in the presence of apoptotic neutrophils. J. Immunol. 168:1968.
50. Moore, K. W., A. O’Garra, R. de Waal Malefyt, P. Vieira, and T. R. Mosmann.1993. Interleukin-10. Annu. Rev. Immunol. 11:165.
51. Knolle, P., J. Schlaak, A. Uhrig, P. Kempf, K. H. Meyer zum Buschenfelde, andG. Gerken. 1995. Human Kupffer cells secrete IL-10 in response to lipopolysac-charide (LPS) challenge. J. Hepatol. 22:226.
52. Tanaka, Y., K. Kobayashi, A. Takahashi, I. Arai, S. Higuchi, S. Otomo, S. Habu,and T. Nishimura. 1993. Inhibition of inflammatory liver injury by a monoclonalantibody against lymphocyte function-associated antigen-1. J. Immunol.151:5088.
53. Andrew, D. P., J. P. Spellberg, H. Takimoto, R. Schmits, T. W. Mak, andM. M. Zukowski. 1998. Transendothelial migration and trafficking of leukocytesin LFA-1-deficient mice. Eur. J. Immunol. 28:1959.
54. de Waal Malefyt, R., J. Abrams, B. Bennett, C. G. Figdor, and J. E. de Vries.1991. Interleukin 10 (IL-10) inhibits cytokine synthesis by human monocytes: anautoregulatory role of IL-10 produced by monocytes. J. Exp. Med. 174:1209.
55. de Waal Malefyt, R., C. G. Figdor, R. Huijbens, S. Mohan-Peterson, B. Bennett,J. Culpepper, W. Dang, G. Zurawski, and J. E. de Vries. 1993. Effects of IL-13on phenotype, cytokine production, and cytotoxic function of human monocytes:comparison with IL-4 and modulation by IFN-� or IL-10. J. Immunol. 151:6370.
56. Leeuwenberg, J. F., T. M. Jeunhomme, and W. A. Buurman. 1994. Slow releaseof soluble TNF receptors by monocytes in vitro. J. Immunol. 152:4036.
57. Nagaki, M., M. Tanaka, A. Sugiyama, H. Ohnishi, and H. Moriwaki. 1999. In-terleukin-10 inhibits hepatic injury and tumor necrosis factor-� and interferon-�mRNA expression induced by staphylococcal enterotoxin B or lipopolysaccha-ride in galactosamine-sensitized mice. J. Hepatol. 31:815.
58. Moore, K. W., R. de Waal Malefyt, R. L. Coffman, and A. O’Garra. 2001. In-terleukin-10 and the interleukin-10 receptor. Annu. Rev. Immunol. 19:683.
59. Piguet, P. F., C. Vesin, J. E. Ryser, G. Senaldi, G. E. Grau, andF. Tacchini-Cottier. 1993. An effector role for platelets in systemic and locallipopolysaccharide-induced toxicity in mice, mediated by a CD11a- and CD54-dependent interaction with endothelium. Infect. Immun. 61:4182.
60. Ramani, M., V. Ollivier, F. Khechai, T. Vu, C. Ternisien, F. Bridey, andD. de Prost. 1993. Interleukin-10 inhibits endotoxin-induced tissue factor mRNAproduction by human monocytes. FEBS Lett. 334:114.
61. Praidier, O., C. Gerard, A. Delvaux, M. Lybin, D. Abramowicz, P. Capel,T. Velu, and M. Goldman. 1993. Interleukin-10 inhibits the induction of mono-cyte procoagulant activity by bacterial lipopolysaccharide. Eur. J. Immunol. 23:2700.
62. Ramani, M., F. Khechai, V. Ollivier, C. Ternisien, F. Bridey, J. Hakim, andD. de Prost. 1994. Interleukin-10 and pentoxifylline inhibit C-reactive protein-induced tissue factor gene expression in peripheral human blood monocytes.FEBS Lett. 356:86.
63. Vasse, M., J. Paysant, J. Soria, J. P. Collet, J. P. Vannier, and C. Soria. 1996.Regulation of fibrinogen biosynthesis by cytokines: consequences on the vascularrisk. Haemostasis 26(Suppl. 4):331.
64. Okada, K., T. Fujita, K. Minamoto, H. Liao, Y. Naka, and D. J. Pinsky. 2000.Potentiation of endogenous fibrinolysis and rescue from lung ischemia/reperfu-sion injury in interleukin (IL)-10-reconstituted IL-10 null mice. J. Biol. Chem.275:21468.
592 ROLE OF LFA-1 IN LIVER INJURY
65. Junger, W. G., D. B. Hoyt, H. Redl, F. C. Liu, W. H. Loomis, J. Davies, andG. Schlag. 1995. Tumor necrosis factor antibody treatment of septic baboonsreduces the production of sustained T-cell suppressive factors. Shock 3:173.
66. Scharffetter-Kochanek, K., H. Lu, K. Norman, N. van Nood, F. Munoz,S. Grabbe, M. McArthur, I. Lorenzo, S. Kaplan, K. Ley, et al. 1998. Spontaneousskin ulceration and defective T cell function in CD18 null mice. J. Exp. Med.188:119.
67. Ding, Z. M., J. E. Babensee, S. I. Simon, H. Lu, J. L. Perrard, D. C. Bullard,X. Y. Dai, S. K. Bromley, M. L. Dustin, M. L. Entman, et al. 1999. Relativecontribution of LFA-1 and Mac-1 to neutrophil adhesion and migration. J. Im-munol. 163:5029.
68. Mayadas, T. N., R. C. Johnson, H. Rayburn, R. O. Hynes, and D. D. Wagner.1993. Leukocyte rolling and extravasation are severely compromised in P selec-tin-deficient mice. Cell 74:541.
69. Forlow, S. B., E. J. White, S. C. Barlow, S. H. Feldman, H. Lu, G. J. Bagby,A. L. Beaudet, D. C. Bullard, and K. Ley. 2000. Severe inflammatory defect and
reduced viability in CD18 and E-selectin double mutant mice. J. Clin. Invest.106:1457.
70. Wilson, R. W., C. M. Ballantyne, C. W. Smith, C. Montogomery, A. Bradley,W. E. O’Brien, and A. L. Beaudet. 1993. Gene targeting yields a CD18-mutantmouse for study of inflammation. J. Immunol. 151:1571.
71. Frenette, P. S., T. N. Mayadas, H. Rayburn, R. O. Hynes, and D. D. Wagner.1996. Susceptibility to infection and altered hematopoiesis in mice deficient inboth P- and E-selectins. Cell 84:563.
72. Forlow, S. B., J. R. Schurr, J. K. Kolls, G. J. Bagby, P. O. Schwarzenberger, andK. Ley. 2001. Increased granulopoiesis through interleukin-17 and granulocytecolony-stimulating factor in leukocyte adhesion molecule-deficient mice. Blood98:3309.
73. Miyamoto, M., M. Emoto, Y. Emoto, V. Brinkmann, I. Yoshizawa, P. Seiler, P.Aichele, E. Kita, and S. H. E. Kaufmann. 2003. Neutrophilia in LFA-1-deficientmice confers resistance to listeriosis: possible contribution of granulocyte-colo-ny-stimulating factor and IL-17. J. Immunol. 170:5228.
