Plasminogen Activator Inhibitor Type-1-Deficient Mice Have an Enhanced IFN-γ Response to Lipopolysaccharide and Staphylococcal Enterotoxin B This information is current as of September 26, 2021. Rosemarijn Renckens, Jennie M. Pater and Tom van der Poll J Immunol 2006; 177:8171-8176; ; doi: 10.4049/jimmunol.177.11.8171 http://www.jimmunol.org/content/177/11/8171 Downloaded from

References This article cites 26 articles, 14 of which you can access for free at: http://www.jimmunol.org/content/177/11/8171.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 26, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2006 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Plasminogen Activator Inhibitor Type-1-Deficient Mice Have an Enhanced IFN-␥ Response to Lipopolysaccharide and Staphylococcal Enterotoxin B1

Rosemarijn Renckens,2*† Jennie M. Pater,*† and Tom van der Poll*†

Plasminogen activator inhibitor type-1 (PAI-1) is a major inhibitor of fibrinolysis by virtue of its capacity to inhibit urokinase-type plasminogen activator (uPA) and tissue-type plasminogen activator (tPA). Systemic inflammation is invariably associated with elevated circulating levels of PAI-1, and during human sepsis plasma PAI-1 concentrations predict an unfavorable outcome. Knowledge about the functional role of PAI-1 in a systemic inflammatory response syndrome is highly limited. In this study, we determined the role of endogenous PAI-1 in cytokine release induced by administration of LPS or staphylococcal enterotoxin B (SEB). Both LPS and SEB elicited secretion of PAI-1 into the circulation of normal wild-type (Wt) mice. Relative to Wt mice, Downloaded from PAI-1 gene-deficient (PAI-1؊/؊) mice demonstrated strongly elevated plasma IFN-␥ concentrations after injection of either LPS or SEB. In addition, PAI-1؊/؊ splenocytes released more IFN-␥ after incubation with LPS or SEB than Wt splenocytes. Both PAI-1؊/؊ CD4؉ and CD8؉ T cells produced more IFN-␥ upon stimulation with SEB. LPS-induced IFN-␥ release in mice deficient for uPA, the uPA receptor, or tPA was not different from IFN-␥ release in LPS-treated Wt mice. These results identify a novel function of PAI-1 during systemic inflammation, where endogenous PAI-1 serves to inhibit IFN-␥ release by a mechanism that

does not depend on its interaction with uPA/uPA receptor or tPA. The Journal of Immunology, 2006, 177: 8171–8176. http://www.jimmunol.org/

ediators of the fibrinolytic system have been associ- meningococcal sepsis (5, 6). Although these observations suggest ated with functions besides their classical fibrin-dis- that PAI-1 plays a functional role in the inflammatory response solving properties (1, 2). Plasminogen activator inhib- during severe infection, such a role has not been established thus M 3 itor type-1 (PAI-1) is the main inhibitor of the fibrinolytic system. far. One can hypothesize that PAI-1 influences the innate immune By inactivating both urokinase-type plasminogen activator (uPA) response by inhibition of mediators of the fibrinolytic system. For and tissue-type plasminogen activator (tPA), PAI-1 inhibits acti- instance, plasmin can activate the p38 MAPK signaling pathway in vation of the key-enzyme plasmin and subsequent fibrin degrada- monocytes (7, 8), and activation of this pathway was recently tion. However, PAI-1 has also been implicated in several other shown to be of key importance for the inflammatory response to by guest on September 26, 2021 processes and diseases by actions that are not or only partially endotoxin in humans (9, 10). Furthermore, in vitro, plasmin was related to its capacity to inhibit plasminogen activation, including demonstrated to stimulate the release of cytokines and other in- wound healing, tumor angiogenesis, rheumatoid arthritis, fibrosis, flammatory mediators by different cell types (7, 11–13). Moreover, metabolic disturbances, glomerulonephritis, cell migration, bleo- PAI-1 inhibits uPA, which can enhance LPS-induced cytokine ex- mycin-induced lung injury, asthma, and nasal allergy. pression in vitro and in vivo (14), and studies using genetically During severe infection or sepsis, the systemic inflammatory modified mice have implicated uPA as an important regulator of response is almost invariably associated with a strong rise in cir- inflammatory responses to bacterial and other stimuli (15, 16). culating PAI-1 levels. In sepsis patients, PAI-1 concentrations pre- LPS, present in the outer membrane of Gram-negative bacteria, dict lethality in a very sensitive manner (3, 4). Moreover, a (4G/ plays a pivotal role in triggering inflammatory responses during 5G) promoter deletion/insertion polymorphism in the PAI-1 gene Gram-negative sepsis. Staphylococcal enterotoxin B (SEB) is a has been found to influence the risk of the development of septic product of Staphylococcus aureus, which stimulates both APCs shock, and to be associated with a poor outcome in patients with and T cells in vivo (17, 18). We investigated the role of PAI-1 in cytokine responses to LPS and SEB-induced inflammation in vivo, using PAI-1 gene-deficient (PAI-1Ϫ/Ϫ) mice. We show that PAI-1 *Center for Infection and Immunity Amsterdam, and †Center for Experimental and Molecular Medicine, Academic Medical Center, University of Amsterdam, Amster- deficiency is associated with a strongly enhanced LPS and SEB- dam, The Netherlands induced IFN-␥ release in vivo and in vitro by a mechanism Received for publication May 8, 2006. Accepted for publication September 11, 2006. that does not depend on its interaction with tPA, uPA, or its The costs of publication of this article were defrayed in part by the payment of page receptor uPAR. charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by a grant from the Netherlands Heart Foundation (2001B114; to R.R.). Materials and Methods 2 Address correspondence and reprint requests to Dr. Rosemarijn Renckens, Center Animals for Experimental and Molecular Medicine, Academic Medical Center, Room G2-132, Ϫ/Ϫ Ϫ/Ϫ Ϫ/Ϫ Ϫ/Ϫ Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. E-mail address: r.renckens@ PAI-1 , uPA , uPAR , and tPA mice, all backcrossed to a amc.uva.nl C57BL/6 genetic background, and C57BL/6 wild-type (Wt) mice were 3 Abbreviations used in this paper: PAI-1, plasminogen activator inhibitor type-1; purchased from The Jackson Laboratory. Eight-week-old female mice were uPA, urokinase-type plasminogen activator; tPA, tissue-type plasminogen activator; used for all experiments. All experiments were approved by the institu- SEB, staphylococcal enterotoxin B; Wt, wild type; CBA, cytometric bead array; IC, tional animal care and use committee of the Academic Medical Center intracellular. (Amsterdam, The Netherlands).

Copyright © 2006 by The American Association of Immunologists, Inc. 0022-1767/06/$02.00 8172 ENHANCED IFN-␥ RELEASE IN PAI-1-DEFICIENT MICE

FIGURE 1. Effect of PAI-1 deficiency on LPS-induced cytokine release. TNF-␣, IL-6, MCP-1, IL-10, IL-12p70, and IFN-␥ were measured in plasma at several time points after i.p. injection of LPS (200 ␮g). Results are ex-

pressed as means Ϯ SE of eight mice per strain Downloaded from p Ͻ ,ءء p Ͻ 0.05, and ,ء .at each time point 0.001 vs Wt mice at the same time point. http://www.jimmunol.org/ by guest on September 26, 2021

LPS and SEB-induced inflammation in vivo Ex vivo peritoneal macrophage stimulation A total of 200 ␮gofLPS(Escherichia coli 055:B5; Sigma-Aldrich) or 100 ␮g Peritoneal macrophages were harvested from Wt and PAI-1Ϫ/Ϫ mice by of SEB (Sigma-Aldrich) was injected i.p. in 200 ␮l of sterile NaCl 0.9%. washing the peritoneal cavity with 5 ml of sterile saline. Collected cells were allowed to adhere to 96-well tissue-culture plates (105 cells/well) for Sample harvesting 2 h at 37°C, after which nonadherent cells were removed by rinsing with medium. More than 95% of the cells were peritoneal macrophages, as At the time of sacrifice, mice were first anesthetized by i.p. injection of identified by cytospin preparations stained with modified Giemsa stain. 0.07 ml/g FFM mixture (Fentanyl (0.315 mg/ml)-Fluanisone (10 mg/ml) Macrophage monolayers were stimulated with medium, 10 ng/ml LPS or (Janssen), Midazolam (5 mg/ml) (Roche)). Next, blood was drawn via 10 ␮g/ml SEB in an end-volume of 200 ␮l for 48 h at 37°C; then super- direct heart puncture with a sterile syringe, and transferred to tubes con- natants were aspirated, and cytokine levels were analyzed by CBA. taining heparin; plasma was prepared by centrifugation at 1400 ϫ g for 10 min at 4°C, after which aliquots were stored at Ϫ20°C. Assays FACS analysis ␣ For FACS analysis, splenocytes suspensions obtained from Wt and PAI- Murine PAI-1 was measured by ELISA (Korida). TNF- , IL-6, MCP-1, Ϫ/Ϫ IL-10, IL-12p70, IFN-␥, IL-5, IL-4, and IL-2 were measured by cytometric 1 mice were washed with FACS buffer (PBS supplemented with 0.5% BSA, 0.01% NaN , and 100 mM EDTA) and resuspended in 150 ␮lof bead array (CBA) multiplex assay (BD Pharmingen) in accordance with 3 FACS buffer. Immunostaining for cell surface molecules was performed the manufacturer’s recommendations. The detection limit of all cytokines for 30 min at 4°C, using directly labeled Abs against CD3, CD4, CD8, was 2.5–5.0 pg/ml. NK1.1, CD11b, CD11c, F4/80, and GR1. All Abs were used in concen- Ex vivo splenocyte stimulation trations recommended by the manufacturer (BD Pharmingen). To correct for nonspecific staining, an appropriate control Ab (rat IgG2; BD Pharm- Single-cell suspensions were obtained from Wt and PAI-1Ϫ/Ϫ mice by ingen) was used. For the ex vivo stimulation experiments, splenocytes were crushing spleens through a 40-␮m cell strainer (BD Pharmingen). Eryth- seeded in 24-well tissue-culture plates at a cell density of 1 ϫ 106 cells/well ␮ rocytes were lysed with ice-cold isotonic NH4Cl solution (155 mM NH4Cl, in duplicate, and stimulated with medium or 10 g/ml SEB in an end- 10 mM KHCO3, 100 mM EDTA (pH 7.4)), and the remaining cells were volume of 2 ml. After1hofincubation at 37°C in 5% CO2, the protein washed twice with RPMI 1640 (BioWhittaker Europe). Splenocytes were transport inhibitor brefeldin A (2 ␮g/ml; Sigma-Aldrich) or medium was

suspended in medium (RPMI 1640 with L-glutamine, 5% autologous se- added to the wells. Cell cultures were incubated at 37°C in 5% CO2 for rum, 5% antibiotic-antimycotic (Invitrogen Life Technologies), seeded in 48 h, after which cells were washed with FACS buffer and resuspended in 96-well round-bottom culture plates at a cell density of 1 ϫ 106 cells/ FACS buffer. Cells were fixed with 4% formaldehyde and permeabilized well in triplicate, and stimulated with medium, 10 ng/ml LPS or 10 using 100 ␮l of Cytofix/Cytoperm (BD Pharmingen) for 20 min at 4°C. ␮g/ml SEB in an end-volume of 200 ␮l. Supernatants were harvested Next, cells were stained for intracellular (IC) IFN-␥ (BD Pharmingen) for

after a 48-h incubation at 37°C in 5% CO2, and cytokine levels were 30 min at 4°C, after which they were washed and resuspended in FACS analyzed by CBA. buffer for FACS analysis (using FACSCalibur; BD Biosciences). The Journal of Immunology 8173 Downloaded from

FIGURE 2. Effect of PAI-1 deficiency on SEB-induced cytokine release. TNF-␣, IL-6, MCP-1, IL-10, IL-12p70, IFN-␥, IL-5, and IL-2 were measured in plasma at 4 and 8 h after i.p. injection of SEB (100 ␮g). Results are expressed as means Ϯ SE of eight mice per strain at each http://www.jimmunol.org/ p Ͻ 0.001 vs Wt ,ءء p Ͻ 0.05 and ,ء .time point mice at the same time point. by guest on September 26, 2021

Statistical analysis injection also resulted in an elevation of plasma PAI-1 concentrations Ϯ Ϯ Data are expressed as means Ϯ SEM, unless indicated otherwise. Com- at 4 and 8 h postinjection (18 2.3 and 31 2.3 ng/ml, respectively, parison between time curves were conducted using two-way ANOVA, with both p Ͻ 0.05 vs baseline). Bonferroni post hoc tests. Comparisons between groups were conducted using the Mann-Whitney U test. A p value of Ͻ0.05 was considered to The LPS-induced cytokine response represent a statistically significant difference. To investigate whether PAI-1 deficiency influences LPS-induced cytokine production, we measured plasma levels of TNF-␣, IL-6, Results MCP-1, IL-10, IL-12p70, and IFN-␥ at various time points up to PAI-1 is up-regulated by LPS and SEB 48 h after i.p. injection of LPS. At 48 h, all cytokine levels were To confirm PAI-1 production in Wt mice in our models, we measured back to baseline in both groups of mice, therefore only results PAI-1 levels in plasma before and at 4 and 8 h after i.p. injection of obtained during the first 24 h are shown. The plasma concentra- LPS (200 ␮g) or SEB (100 ␮g). Baseline plasma PAI-1 concentra- tions of all cytokines measured showed a profound rise after LPS tions were 5 Ϯ 0.3 ng/ml. At 4 and 8 h after LPS administration, injection (Fig. 1). Plasma TNF-␣, MCP-1, and IL-10 levels were PAI-1 levels were significantly increased compared with baseline val- not different between Wt and PAI-1Ϫ/Ϫ mice. In Wt mice, plasma ues (512 Ϯ 34 and 669 Ϯ 54 ng/ml, respectively, both p Ͻ 0.05). SEB IL-6 levels showed a peak at 2 h postinjection; in contrast, in 8174 ENHANCED IFN-␥ RELEASE IN PAI-1-DEFICIENT MICE

Table I. Cellular composition of spleens from Wt and PAI-1Ϫ/Ϫ micea

Cell Type (%) Wt PAI-1Ϫ/Ϫ

CD4ϩ T cells 11.5 Ϯ 0.7 12 Ϯ 0.6 CD8ϩ T cells 8.6 Ϯ 0.5 8.1 Ϯ 0.7 NK cells 3.6 Ϯ 0.1 3.8 Ϯ 0.2 Dendritic cells 0.7 Ϯ 0.1 0.8 Ϯ 0.1 Macrophages 4.4 Ϯ 0.3 3.5 Ϯ 0.4 Granulocytes 2.7 Ϯ 0.3 2.1 Ϯ 0.3

a Percentages of total splenocyte cell counts. Data are means Ϯ SEM. n ϭ 6 mice per genotype.

PAI-1Ϫ/Ϫ mice IL-6 levels peaked after 4 h and were still signif- icantly higher than in Wt mice at 8 h after injection. IL-12p70 levels peaked at 4 h after LPS in both groups of mice, but levels were much higher in PAI-1Ϫ/Ϫ mice compared with Wt mice. The plasma concentrations of IFN-␥ peaked at 8 h after LPS injection in both groups. However, IFN-␥ levels were strongly increased in PAI-1Ϫ/Ϫ mice compared with Wt mice. Downloaded from

The SEB-induced cytokine response Having found that PAI-1 deficiency is associated with strongly en- hanced LPS-induced IFN-␥, IL-6, and IL12p70 release, we decided to investigate the cytokine response to a polyclonal T cell activator. http://www.jimmunol.org/ by guest on September 26, 2021 FIGURE 4. PAI-1Ϫ/Ϫ CD4ϩ and CD8ϩ T cells produce more IFN-␥ in vitro. Splenocytes were stimulated with SEB (10 ␮g/ml) for 48 h. The percentage of cells that positively stained for IC IFN-␥ was measured by p Ͻ ,ء .FACS analysis. Results are expressed as means Ϯ SE of six mice 0.01 vs Wt mice.

Therefore, we measured plasma levels of TNF-␣, IL-6, MCP-1, IL- 10, IL-12p70, IFN-␥, IL-5, IL-4, and IL-2 at 4 and 8 h after i.p. injection of SEB in Wt and PAI-1Ϫ/Ϫ mice (Fig. 2). We chose these two time points because they were found to be representative of su- perantigen-induced cytokine release in vivo; in particular, IFN-␥ reaches peak levels at 8 h after SEB administration (17, 19). The SEB-induced cytokine response was less strong than the LPS-induced cytokine response, although all cytokines with the exception of IL-4 displayed elevated plasma concentrations after SEB administration. TNF-␣ levels were similar in the PAI-1Ϫ/Ϫ mice and Wt mice. In line with the LPS-induced IL-6 response, SEB-induced IL-6 levels were higher in PAI-1Ϫ/Ϫ mice than in Wt mice at both time points. MCP-1 concentrations were higher in PAI-1Ϫ/Ϫ mice at 4 h after SEB ad- ministration. IL-10, IL-12p70, IL-5, and IL-2 concentrations were similar between both genotypes. In line with the results obtained after LPS injection, IFN-␥ levels were markedly elevated in PAI-1Ϫ/Ϫ mice when compared with Wt mice at both 4 and 8 h after SEB injection (Fig. 2).

PAI-1Ϫ/Ϫ splenocytes release more IFN-␥ during ex vivo Ϫ/Ϫ FIGURE 3. PAI-1 splenocytes produce more IFN-␥ in vitro. stimulation TNF-␣, IL-6, and IFN-␥ were measured in the supernatants of splenocyte Ϫ/Ϫ cultures after 48 h of incubation with either LPS (10 ng/ml) or SEB (10 In an attempt to determine which cell populations in PAI-1 p Ͻ 0.05 and mice show a changed cytokine response and to determine which ,ء .␮g/ml). Results are expressed as means Ϯ SE of six mice p Ͻ 0.01 vs Wt mice. cytokine changes are influenced directly by PAI-1 deficiency, we ,ءء The Journal of Immunology 8175

FIGURE 5. uPA, uPAR, or tPA deficiency does not influence LPS-induced IFN-␥ release. Plasma IFN-␥ levels were measured 8 h after i.p. injection of LPS (200 ␮g) in uPAϪ/Ϫ, uPARϪ/Ϫ, tPAϪ/Ϫ, or Wt mice. Results are expressed as means Ϯ SE of eight mice per group. Differences between groups were not significant.

stimulated splenocytes and peritoneal macrophages harvested from Discussion Ϫ Ϫ Wt and PAI-1 / mice that had not received LPS or SEB in vivo Patients with sepsis consistently display elevated plasma concen- with LPS, SEB, or culture medium for 48 h ex vivo, and measured trations of PAI-1, and such elevated circulating PAI-1 levels are TNF-␣, IL-6, IL-12p70, IFN-␥, IL-5, IL-4, and IL-2 levels in the highly predictive for an unfavorable outcome (3, 4). However, supernatants. The cellular distribution of spleens obtained from these clinical studies failed to provide insight into a possible func- Ϫ Ϫ PAI-1 / and Wt mice did not differ (Table I). None of these tional role of PAI-1 in sepsis, i.e., the elevated PAI-1 levels could cytokines was detectable in the supernatants of cell cultures incu- merely be indicative of a strong inflammatory response of the host, bated with medium only. Stimulation with either LPS or SEB re- rather than bearing any pathophysiological significance. In this sulted in TNF-␣, IL-6, and IFN-␥ release by splenocytes (Fig. 3), study, we attempted to provide a first insight into the role of en- Downloaded from whereas IL-12p70, IL-5, IL-4, and IL-2 remained undetectable. dogenous PAI-1 in the systemic inflammatory response to two dif- LPS and SEB-induced TNF-␣ and IL-6 levels were similar in the ferent challenges: LPS, the most important proinflammatory com- Ϫ Ϫ supernatants of Wt and PAI-1 / splenocytes. However, in line ponent of Gram-negative bacteria that mainly activates monocytes Ϫ Ϫ with the in vivo findings, IFN-␥ release by PAI-1 / splenocytes and macrophages (20); and SEB, a superantigen derived from the stimulated with either LPS or SEB was significantly increased. Gram-positive pathogen S. aureus, which triggers polyclonal T ␣ Peritoneal macrophages only released detectable levels of TNF- cell activation (17, 18). The main finding of this study was that http://www.jimmunol.org/ and IL-6 upon stimulation with LPS or SEB; these concentrations PAI-1Ϫ/Ϫ mice demonstrated markedly elevated plasma IFN-␥ Ϫ Ϫ did not differ between Wt or PAI-1 / macrophages (data not levels upon administration of either LPS or SEB. Further investi- shown). gations identified CD4ϩ and CD8ϩ T cells, and to a lesser extent ␥ ϩ ϩ NK cells, as likely cellular sources for the enhanced IFN- release CD4 and CD8 T cells are involved in the stronger IFN-␥ Ϫ Ϫ in PAI-1 / mice, and, in addition, showed that the effect of PAI-1 response in PAI-1Ϫ/Ϫ mice on IFN-␥ secretion occurred independent of tPA, uPA, or uPAR. To determine which cell types are involved in the production of LPS has been used extensively to obtain insight in the mecha- IFN-␥ in response to SEB and to examine whether they were in- nisms contributing to systemic inflammation during sepsis. Bolus fluenced by PAI-1 deficiency, we measured the percentage of cells injection of LPS into humans results in a strong rise in plasma by guest on September 26, 2021 positive for IC IFN-␥ in splenocyte cultures after 48 h of SEB PAI-1 concentrations peaking after 4 h (10, 21). In this study, we ϩ ϩ ϩ ϩ stimulation and differentiated between CD3 CD4 , CD3 CD8 , used the LPS challenge model to obtain a first insight in the role of ϩ and NK1.1 cells by FACS analysis (Fig. 4). In these studies, we PAI-1 in systemic release of cytokines. We first confirmed PAI-1 focused on SEB-induced IFN-␥ production because this stimulus release after LPS administration to Wt mice. When we subse- had proved to be a far more potent IFN-␥ inducer than LPS (see quently found elevated IFN-␥ concentrations in LPS-challenged Fig. 3). The percentage of IFN-␥ positive T cells was low, and PAI-1Ϫ/Ϫ mice, we decided to investigate whether this observation there was no increase in IC IFN-␥ after SEB stimulation in Wt could be reproduced after administration of a polyclonal T cell ϩ Ϫ Ϫ CD4 T cells; in contrast, in PAI-1 / cultures, we did find an activator. For this model we used SEB, a superantigen produced by ϩ increase in the percentage of IFN-␥-positive CD4 T cells that was S. aureus implicated in nonmenstrual toxic shock syndrome (18). ϩ ϩ significantly different from Wt CD4 T cells. CD8 T cells Indeed, SEB injection elicited PAI-1 secretion into the circulation, Ϫ Ϫ showed a rise in IC IFN-␥ in both Wt and PAI-1 / cells; how- and higher IFN-␥ levels were detected in PAI-1Ϫ/Ϫ mice. Of note, Ϫ Ϫ ever, the percentage was significantly higher in PAI-1 / cultures. not all cytokine responses were similarly affected by PAI-1 defi- Finally, NK-cells also showed a rise in the percentage of IC IFN- ciency in the LPS and SEB models. These differences might be Ϫ Ϫ ␥-positive cells after SEB stimulation; although PAI-1 / NK cells explained by the fact that LPS and SEB activate different cell types showed a higher percentage with positive IC IFN-␥ staining, the dif- (20). LPS primarily activates mononuclear cells (20). SEB can ference with Wt cells did not reach statistical significance. These data bind directly to regions of the class II MHC molecule that are ϩ ϩ Ϫ Ϫ show that both CD4 and CD8 T cells from PAI-1 / mice pro- outside the physiological MHC haplotype-restricted Ag-binding duced more IFN-␥ upon SEB stimulation in vitro. groove, which eventually results in activation of both APCs and SEB-reactive V␤8ϩ T cells (17). Such differences in cell activation uPA, uPAR, or tPA deficiency does not influence LPS-induced likely contributed to the differential effects of PAI-1 deficiency on ␥ IFN- release IL-12p70 and MCP-1 release. In particular, the different impact of A major function of PAI-1 in the fibrinolytic system is inhibition PAI-1 deficiency on IL-12p70 release is important. IL-12p70 is a of the plasminogen activators uPA and tPA. Therefore, we were strong inducer of IFN-␥ production (22); the fact that IFN-␥ re- interested to determine the role of uPA and the uPA receptor and lease was enhanced in PAI-1Ϫ/Ϫ mice after both LPS and SEB tPA in IFN-␥ release induced by LPS. To address this issue, injection, whereas IL-12p70 was increased only after LPS injec- uPAϪ/Ϫ, uPARϪ/Ϫ, and tPAϪ/Ϫ mice were i.p. injected with LPS tion strongly suggests that the effect of endogenous PAI-1 on and after 8-h plasma IFN-␥ concentrations were measured, i.e., the IFN-␥ release does not rely on IL-12p70. This notion is further time point at which plasma IFN-␥ reached peak levels in this supported by the fact that PAI-1Ϫ/Ϫ splenocytes released more model. None of the mice displayed an altered IFN-␥ response IFN-␥ upon stimulation with either LPS or SEB under condi- when compared with Wt mice (Fig. 5). tions in where IL-12p70 release remained undetectable. 8176 ENHANCED IFN-␥ RELEASE IN PAI-1-DEFICIENT MICE

Mice with a targeted deletion of the gene-encoding uPA have Disclosures been found to mount a reduced type 1 response upon pulmonary The authors have no financial conflict of interest. infection with the opportunistic yeast Cryptococcus neoformans, as reflected by lower IFN-␥ levels in bronchoalveolar lavage fluid References (16). In addition, lung mononuclear cells and regional lymph node 1. Mondino, A., and F. Blasi. 2004. uPA and uPAR in fibrinolysis, immunity and Ϫ/Ϫ pathology. Trends Immunol. 25: 450–455. cells obtained from infected uPA mice released less IFN-␥ 2. Lijnen, H. R. 2005. Pleiotropic functions of plasminogen activator inhibitor-1. upon Ag-specific stimulation in this model (16). These findings led J. Thromb. Haemost. 3: 35–45. 3. Pralong, G., T. Calandra, M. P. Glauser, J. Schellekens, J. Verhoef, F. Bachmann, us to hypothesize that PAI-1 may facilitate a type 2 response and E. K. Kruithof. 1989. Plasminogen activator inhibitor 1: a new prognostic through inhibition of uPA. However, this appeared not to be the marker in septic shock. Thromb. Haemost. 61: 459–462. case: not only uPAϪ/Ϫ mice, but also uPARϪ/Ϫ and tPAϪ/Ϫ mice 4. Lorente, J. A., L. J. Garcia-Frade, L. Landin, R. de Pablo, C. Torrado, E. Renes, and A. Garcia-Avello. 1993. Time course of hemostatic abnormalities in sepsis demonstrated an unremarkable IFN-␥ response upon injection of and its relation to outcome. Chest 103: 1536–1542. LPS. These data suggest that the effect of endogenous PAI-1 is not 5. Westendorp, R. G., J. J. Hottenga, and P. E. Slagboom. 1999. Variation in plas- minogen-activator-inhibitor-1 gene and risk of meningococcal septic shock. Lan- mediated by an effect on uPA/uPAR or tPA. cet 354: 561–563. Although our study was in progress, Sejima et al. (23) reported 6. Hermans, P. W., M. L. Hibberd, R. Booy, O. Daramola, J. A. Hazelzet, on the consequences of PAI-1 deficiency in a model of nasal al- R. de Groot, and M. Levin. 1999. 4G/5G promoter polymorphism in the plas- minogen-activator-inhibitor-1 gene and outcome of meningococcal disease: Me- lergy. In this latter investigation, OVA-sensitized mice demon- ningococcal Research Group. Lancet 354: 556–560. strated increased PAI-1 levels in nasal washings upon intranasal 7. Burysek, L., T. Syrovets, and T. Simmet. 2002. The serine protease plasmin triggers expression of MCP-1 and CD40 in human primary monocytes via acti- OVA challenge; this locally produced PAI-1 contributed to a Th2 vation of p38 MAPK and Janus kinase (JAK)/STAT signaling pathways. J. Biol. Downloaded from response as reflected by Th1-biased responses in sensitized PAI- Chem. 277: 33509–33517. 1Ϫ/Ϫ mice characterized by reduced OVA-specific circulating IgE 8. Syrovets, T., A. Schule, M. Jendrach, B. Buchele, and T. Simmet. 2002. Cigli- tazone inhibits plasmin-induced proinflammatory monocyte activation via mod- and elevated plasma IgG2a levels, as well as lower IL-4 and IL-5 ulation of p38 MAP kinase activity. Thromb. Haemost. 88: 274–281. and higher IFN-␥ concentrations in nasal lavage fluid (23). Our 9. Branger, J., B. van den Blink, S. Weijer, J. Madwed, C. L. Bos, A. Gupta, C. L. Yong, S. H. Polmar, D. P. Olszyna, C. E. Hack, et al. 2002. Anti-inflam- present findings extend and are partly in line with these results: matory effects of a p38 mitogen-activated protein kinase inhibitor during human Ϫ Ϫ naive PAI-1 / mice displayed enhanced systemic IFN-␥ release endotoxemia. J. Immunol. 168: 4070–4077. after i.p. injection of either LPS or SEB, and these in vivo obser- 10. Branger, J., B. van den Blink, S. Weijer, A. Gupta, S. J. van Deventer, C. E. Hack, http://www.jimmunol.org/ M. P. Peppelenbosch, and T. van der Poll. 2003. Inhibition of coagulation, fibri- Ϫ/Ϫ vations could be reproduced using PAI-1 splenocytes in vitro. nolysis and endothelial cell activation by a p38 mitogen-activated protein kinase However, in contrast to the results obtained by Sejima et al. (23), inhibitor during human endotoxemia. Blood 101: 4446–4448. 11. Syrovets, T., M. Jendrach, A. Rohwedder, A. Schule, and T. Simmet. 2001. we did not find a diminished release of the type 2 cytokine IL-5 in Plasmin-induced expression of cytokines and tissue factor in human monocytes the SEB model; whereas IL-4 could not be studied because even in involves AP-1 and IKK␤-mediated NF-␬B activation. Blood 97: 3941–3950. Wt mice, plasma IL-4 remained undetectable after SEB adminis- 12. Weide, I., J. Romisch, and T. Simmet. 1994. Contact activation triggers stimu- lation of the monocyte 5-lipoxygenase pathway via plasmin. Blood 83: tration. Nonetheless, these data together suggest that PAI-1 inhib- 1941–1951. its IFN-␥ release under markedly different conditions, thereby 13. Chang, W. C., G. Y. Shi, Y. H. Chow, L. C. Chang, J. S. Hau, M. T. Lin, C. J. Jen, L. Y. Wing, and H. L. Wu. 1993. Human plasmin induces a receptor-mediated identifying a novel biological activity of this protein. Further stud- arachidonate release coupled with G proteins in endothelial cells. Am. J. Physiol. by guest on September 26, 2021 ies are warranted to unravel the mechanisms by which PAI-1 in- 264: C271–C281. ␥ 14. Abraham, E., M. R. Gyetko, K. Kuhn, J. Arcaroli, D. Strassheim, J. Park, fluences IFN- production. S. Shetty, and S. Idell. 2003. Urokinase-type plasminogen activator potentiates Inhibition or elimination of IFN-␥ has been reported to reduce lipopolysaccharide-induced neutrophil activation. J. Immunol. 170: 5644–5651. LPS-induced lethality in mice (24–26). In contrast, pretreatment 15. Gyetko, M. R., G. H. Chen, R. A. McDonald, R. Goodman, G. B. Huffnagle, ␥ C. C. Wilkinson, J. A. Fuller, and G. B. Toews. 1996. Urokinase is required for with recombinant IFN- increased LPS-induced lethality (24). the pulmonary inflammatory response to Cryptococcus neoformans: a murine Thus, it is possible that PAI-1 deficiency influences survival during transgenic model. J. Clin. Invest. 97: 1818–1826. 16. Gyetko, M. R., S. Sud, T. Kendall, J. A. Fuller, M. W. Newstead, and endotoxemia. We considered survival studies beyond the scope of T. J. Standiford. 2000. Urokinase receptor-deficient mice have impaired neutro- the current investigations, because our study focused on the reg- phil recruitment in response to pulmonary Pseudomonas aeruginosa infection. ulation of IFN-␥ production by PAI-1 and because many other J. Immunol. 165: 1513–1519. ␥ 17. Miethke, T., C. Wahl, D. Regele, H. Gaus, K. Heeg, and H. Wagner. 1993. factors besides IFN- have been implicated as mediators of LPS Superantigen mediated shock: a cytokine release syndrome. Immunobiology 189: toxicity and lethality. In addition, in this respect, it should be noted 270–284. ␥ 18. Dinges, M., P. Orwin, and P. Schlievert. 2000. Exotoxins of Staphylococcus that one study reported that IFN- receptor-deficient mice sensi- aureus. Clin. Microbiol. Rev. 13: 16–34. tized with D-galactosamine were protected against LPS-induced 19. Lauw, F. N., S. Florquin, P. Speelman, S. J. van Deventer, and T. van der Poll. lethality, whereas the toxicity evoked by high-dose LPS in the 2001. Role of endogenous interleukin-12 in immune response to staphylococcal enterotoxin B in mice. Infect. Immun. 69: 5949–5952. absence of D-galactosamine sensitization was similar in IFN-␥ re- 20. Beutler, B., and E. Rietschel. 2003. Innate immune sensing and its roots: the story ceptor-deficient and Wt mice (26). Of note, the influence of en- of endotoxin. Nat. Rev. Immunol. 3: 169–176. ␥ 21. Renckens, R., S. Weijer, A. F. de Vos, J. M. Pater, J. C. Meijers, C. E. Hack, dogenous PAI-1 on IFN- production could be of relevance to M. Levi, and T. van der Poll. 2004. Inhibition of plasmin activity by tranexamic different conditions and diseases besides sepsis in which IFN-␥ has acid does not influence inflammatory pathways during human endotoxemia. Ar- been found to play a role, including allergy (see above), viral, and terioscler. Thromb. Vasc. Biol. 24: 483–488. 22. Trinchieri, G. 2003. Interleukin-12 and the regulation of innate resistance and parasitic infections, and IC infections such as tuberculosis. adaptive immunity. Nat. Rev. Immunol. 3: 133–146. Enhanced PAI-1 release is a consistent part of the systemic in- 23. Sejima, T., S. Madoiwa, J. Mimuro, T. Sugo, K. Okada, S. Ueshima, O. Matsuo, T. Ishida, K. Ichimura, and Y. Sakata. 2005. Protection of plasminogen activator flammatory response syndrome induced by sepsis or administra- inhibitor-1-deficient mice from nasal allergy. J. Immunol. 174: 8135–8143. tion of bacterial products. In this study, we demonstrate for the first 24. Heinzel, F. P. 1990. The role of IFN-␥ in the pathology of experimental endo- time that elevated PAI-1 concentrations are functionally important toxemia. J. Immunol. 145: 2920–2924. 25. Doherty, G. M., J. R. Lange, H. N. Langstein, H. R. Alexander, C. M. Buresh, and for attenuating IFN-␥ release after injection of LPS or SEB. These J. A. Norton. 1992. Evidence for IFN-␥ as a mediator of the lethality of endotoxin results further exemplify the complex role of PAI-1 in the host and tumor necrosis factor-␣. J. Immunol. 149: 1666–1670. 26. Car, B. D., V. M. Eng, B. Schnyder, L. Ozmen, S. Huang, P. Gallay, response to severe infection, which reaches far beyond its classical D. Heumann, M. Aguet, and B. Ryffel. 1994. Interferon ␥ receptor deficient mice role as an inhibitor of fibrinolysis. are resistant to endotoxic shock. J. Exp. Med. 179: 1437–1444.