Immune Responses in Mesothelial Cells Chemokine and ...

The Journal of Immunology

Nod1/RICK and TLR Signaling Regulate Chemokine and Antimicrobial Innate Immune Responses in Mesothelial Cells1 Jong-Hwan Park,* Yun-Gi Kim,* Michael Shaw,* Thirumala-Devi Kanneganti,* Yukari Fujimoto,† Koichi Fukase,† Naohiro Inohara,* and Gabriel Nuń˜ez2* Mesothelial cells that line the serous cavities and outer surface of internal organs are involved in inflammatory responses induced by microbial stimuli and bacterial infection. Upon exposure to bacterial products, mesothelial cells secrete chemokines, but the signaling pathways by which these cells recognize bacteria to mediate innate immune responses remain largely unknown. We report that stimulation of primary peritoneal mesothelial cells via nucleotide-binding oligomerization domain (Nod)1, a member of the intracytoplasmic Nod-like receptor family, induced potent secretion of the chemokines CXCL1 and CCL2 as well as expression of inducible NO synthase and such responses required the kinase RICK. Mesothelial cells also produced chemokines in response to TLR2, TLR3, TLR4, and TLR5 agonists, but unlike that induced by Nod1 stimulation, the TLR-mediated responses were independent of RICK. Yet, Nod1 stimulation of mesothelial cells via RICK enhanced chemokine secretion induced by LPS or IFN-␥ and cooperated with IFN-␥ in the production of NO. The i.p. administration of KF1B, a synthetic Nod1 agonist, elicited chemokine production in the serum and peritoneal fluid as well as the recruitment of neutrophils into the peritoneal cavity of wild-type mice, but not RICK-deficient mice. Finally, infection of mesothelial cells with Listeria monocytogenes induced production of CXCL1 and this response was significantly reduced in Nod1- or RICK-deficient cells. These results define mesothelial cells as microbial sensors through TLRs and Nod-like receptors and identify Nod1 and RICK as important mediators of chemokine and antimicrobial responses in mesothelial cells. The Journal of Immunology, 2007, 179: 514 –521.

M

esothelial cells form a monolayer of specialized cells of mesodermal origin that line the pleural, peritoneal, and pericardial cavities as well as internal organs (1). The mesothelium not only provides a surface barrier, but also plays an important role in diverse cellular processes including fluid transport, tissue repair, tumor cell adhesion, inflammation, and host defense (1, 2). Mesothelial cells are also involved in the pathogenesis of several diseases. A common pathology is the inflammation of the peritoneal and pleural cavities by toxic material or microbial infection (1, 2). In the latter conditions, the serous cavities are infiltrated by neutrophils and macrophages, which is accompanied by the production of chemokines and cytokines (2). The function of resident inflammatory cells particularly macrophages in peritonitis induced by microbial stimuli has been extensively studied (3). However, little is known about the role of mesothelial cells in inflammation triggered by bacterial products or infection. Mesothelial cells secrete several CC and CXC chemokines including CXCL8/IL-8, CCL2/MCP-1, and CXCL2/MIP-2 upon exposure to bacterial products or stimulation with cytokines such as TNF-␣ or IL-1␤ in both humans and rodents (4 –9). In addition, chemokines produced by mesothelial cells have been

*Department of Pathology and Comprehensive Cancer Center, University of Michigan Medical School, Ann Arbor, MI 48109; †Department of Chemistry, Graduate School of Science, Osaka University, Osaka, Japan Received for publication February 21, 2007. Accepted for publication April 27, 2007. The costs of publication of this article were defrayed in part by the payment of page 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 Grants DK61707 (to G.N.) and GM-60421 (to N.I.) from the National Institutes of Health. Y.-G.K. was supported by a fellowship from the University of Michigan Comprehensive Cancer Center. 2 Address correspondence and reprint requests to Dr. Gabriel Nuńez, Department of Pathology and Comprehensive Cancer Center, University of Michigan Medical School, 1500 East Medical Center Drive, Ann Arbor, MI 48109. E-mail address: [email protected]

www.jimmunol.org

suggested to mediate influx of acute inflammatory cells into the peritoneal cavity (6 – 8, 10). However, the mechanism by which mesothelial cells recognize bacterial products to induce chemokine secretion remains poorly understood. Sensing of bacterial pathogens by the host is mediated by specific pattern-recognition molecules, such as the TLRs and nucleotide-binding oligomerization domain (Nod)3-like receptors (NLRs) (11, 12). Both TLRs and NLRs sense microbial structures that are common to a large number of microorganisms and induce the activation of the NF-␬B transcription factor and the MAPK (11, 12). However, although TLRs mediate bacterial recognition at the cell surface or endosomes, NLRs induce innate immune responses through cytosolic recognition of bacterial molecules (12, 13). Two NLR family members, Nod1 and Nod2, sense bacterial molecules produced during the synthesis and/or degradation of peptidoglycan (14 –17). Nod1 recognizes peptidoglycan-related molecules containing the amino acid meso-diaminopimelic acid (meso-DAP) that are produced by most Gram-negative and specific Gram-positive bacteria such as Listeria monocytogenes (15, 17). Once activated, both Nod1 and Nod2 induce gene transcription through NF-␬B and MAPK signaling pathways by interacting with the kinase RICK/RIP2 (14 –18). Initial studies suggested that RICK also contributed to cytokine responses induced by TLR2, TLR3, and TLR4 but not to TLR9 (19). However, more recent experiments have challenged the notion that RICK contributes to TLR signaling and concluded that this kinase only mediates Nod1and Nod2-induced immune responses (20). In the present report, we have studied a role for Nod1 in the recognition of bacterial

3 Abbreviations used in this paper: Nod, nucleotide-binding oligomerization domain; NLR, Nod-like receptor; iNOS, inducible NO synthase; BMDM, bone marrow-derived macrophage; poly(I:C), polyinosinic-polycitidylic acid; meso-DAP, meso-diaminopimelic acid.

Copyright © 2007 by The American Association of Immunologists, Inc. 0022-1767/07/$2.00


515

FIGURE 1. Nod1 stimulation induces chemokine secretion and enhances TLR-mediated responses in mesothelial cells. A, Gene expression of Nod1, Nod2, TLR2, and TLR4 was compared between BMDM and mesothelial cells by real-time RT-PCR. mRNA expression of each gene was normalized to that of GAPDH. B, Endogenous RICK expression was examined in BMDM and mesothelial cells from wild-type and RICK-deficient mice by immunoblotting. C and D, Mesothelial cells from wild-type and RICK-deficient mice were stimulated with muramyl dipeptide (10 ␮g/ml), KF1B (1 ␮g/ml), lipoteichoic acid (10 ␮g/ml), lipid A (10 ␮g/ml), LPS (100 ng/ml), poly(I:C) (100 ␮g/ml), and flagellin (100 ng/ml). E, Mesothelial cells from wild-type and Nod1-deficient mice were also stimulated with KF1B at different concentrations. Supernatant was collected at 24 h after stimulation, and CXCL1 or CCL2 was measured by ELISA. Results are from one representative experiment of three independent experiments, and data are mean ⫾ SD.

products and bacterial infection in mesothelial cells. Our results revealed that Nod1 stimulation of mesothelial cells induces chemokine responses and neutrophil recruitment in the peritoneal cavity via RICK. Furthermore, we show that Nod1 regulates TLRmediated chemokine production and cooperates with IFN-␥ to induce proinflammatory and antimicrobial responses in mesothelial cells. We also provide evidence that infection of mesothelial cells with L. monocytogenes induces chemokines and this response is mediated in part through RICK.

Materials and Methods Mice RICK- and Nod1-deficient mice in a C57BL/6J background have been described (20). C57BL/6J mice were purchased from The Jackson Laboratory. Mice were housed in specific pathogen-free facility. Animal studies were approved by the University of Michigan Committee on Use and Care of Animals (Ann Arbor, MI).

Reagents and bacterial culture Ultrapure LPS from Escherichia coli O111:B4, lipoteichoic acid, polyinosinic-polycitidylic acid (poly(I:C)), and purified flagellin were purchased from InvivoGen. Synthetic lipid A has been described (21). Muramyl dipeptide (Ac-(6-O-Stearoyl)-muramyl-Ala-D-Glu-NH2) was purchased from Bachem. N-myristoyl (C-14) iE-DAP (referred to as KF1B), a Nod1 activator, has been described (22). L. monocytogenes strain 10403S was a gift from Dr. M. O’Riordan (University of Michigan, Ann Arbor, MI). Single colonies were inoculated into 5 ml of brain-heart infusion medium and grown overnight at 37°C with shaking. A 1/10 dilution of the overnight culture was prepared and allowed to grow at 37°C with shaking to A600 ⫽ 0.5, which corresponds to ⬃109 CFU/ml. Bacteria were diluted to the desired concentration and used in subsequent experiments.

Preparation of bone marrow-derived macrophages (BMDM) and primary mesothelial cells and stimulation with bacterial products and bacteria BMDMs were prepared as previously described (23). Mesothelial cells were prepared from the peritoneum and external surface of liver, spleen, and kidneys of adult mice as described (24). Briefly, pieces of the peritoneum and intact organs were obtained from sacrificed mice and digested with 0.25%-trypsin-EDTA solution for 50 min at 37°C. Intact tissues and tissue debris were discarded and the cell suspension was centrifuged at 1000 rpm for 5 min. The pellet was resuspended in DMEM supplemented with 15% heat-inactivated FBS (Invitrogen Life Technologies) and antibiotics and cultured overnight. Next day, nonadherent cells were removed and the mesothelial cells grown in the presence of 5 ng/ml epithelial growth factor. Mesothelial cells were used between passages 2 and 4. Mesothelial cells exhibited the classical cobblestone morphology and expressed cytokeratins. Less than 1% of the mesothelial cells expressed the macrophage surface F4/80 marker by flow cytometric analysis. The cells were stimulated with various ligands singly or in different combinations at the concentrations described in each experiment. Culture supernatants were collected at the indicated times and assayed for cytokine production.

Real-time PCR RNA was extracted using RNeasy Mini kit (Qiagen). PCR was conducted with cDNA as a template in an ABI PRISM R7000 sequence detection system using SYBR green buffer according to the manufacturer’s instructions (Applied Biosystems). The DNA template was subjected to 10 min of AmpliTaq Gold Activation at 95°C, followed by 45 cycles of amplification (95°C for 15 s, 60°C for 1 min). PCR amplification of the housekeeping gene GAPDH was performed for each sample as the control for sample loading and to allow normalization between samples. The primer set for inducible NO synthase (iNOS) was (forward) 5⬘-CAGCCCAACAATAC AAGATGACCC-3⬘ and (reverse) 5⬘-CAGTTCCGAGCGTCAAAGACC TGC-3⬘. The other primer sets were previously described (25).

516

ROLE OF Nod1/RICK AND TLR SIGNALING IN MESOTHELIAL CELLS

FIGURE 2. Nod1 signaling via RICK enhances LPS-induced CXCL1 secretion in mesothelial cells. Mesothelial cells from WT and RICKdeficient mice were stimulated with KF1B or LPS alone and their combinations. Supernatant was collected at 3 (A and C) and 6 h (B and D) after stimulation, and CXCL1 levels were measured. Results are from one representative experiment of three independent experiments conducted, and data are the mean ⫾ SD.

Measurement of NO and cytokines

In vivo induction of cytokines and neutrophil recruitment

NO synthase activity in the supernatant of cultured cells was assayed for nitrite accumulation by the Griess reaction (26). Mouse cytokines were measured in culture supernatants by ELISA kits from R&D Systems, respectively.

Mice were injected i.p. with 50 ␮g of KF1B and blood samples were collected at 3, 6, and 24 h after injection. Sera were isolated by centrifugation at 3000 rpm for 10 min and submitted to ELISA for analysis of chemokines. Peritoneal cells were collected from mice 24 h after injection of KF1B. A total of 2 ⫻ 104 cells were centrifuged on slide glass by cytospin (Shandon) at 500 rpm for 5 min and stained with Diff-Quik staining kit (Dade Behring). The percentage of neutrophils was determined based on 400 cells counted.

Western blot For analysis of phosphorylation of I␬-B␣, p38, ERK, and JNK cells were stimulated with KF1B, harvested, and lysed in buffer containing 1% Nonidet P-40 supplemented with complete protease inhibitor cocktail (Roche) and 2 mM DTT. Lysates were resolved by SDS-PAGE, transferred to polyvinylidene difluoride membranes, and immunoblotted with primary Abs. Abs against mouse I␬-B␣, p38, ERK1/2, and JNK (phosphorylated and unphosphorylated forms) were purchased from Cell Signaling Technology. Monoclonal Ab to mouse RICK was purchased from Alexis. Proteins were detected by ECL.

FIGURE 3. Nod1 and RICK are required for NF-␬B and MAPK activation induced by KF1B in mesothelial cells. Mesothelial cells from wild-type (WT), Nod1-deficient (A), and RICK-deficient (B) mice were stimulated with 10 ␮g/ml KF1B. Cell lysates were prepared and blotted with indicated Abs.

Statistical analysis Statistical significance between groups was determined by two tailed Student’s t test (Excel; Microsoft). Differences were considered significant at p ⬍ 0.05.


FIGURE 4. Nod1 stimulation induces iNOS expression and NO production via RICK in mesothelial cells. Mesothelial cells from wild-type (WT) and RICK-deficient mice were stimulated with KF1B (10 ␮g/ml), and mRNA was extracted at different time points. A, Expression of iNOS was evaluated by real-time PCR and fold increase was obtained by comparison to the level at 0 h. Mesothelial cells from wild-type and RICK-deficient mice were also stimulated with KF1B (10 ␮g/ml) or IFN-␥ (2 ␮g/ml) alone and in combination. Supernatant was collected at 48 (B) and 72 h (C) after stimulation, and NO production in the supernatant was determined using Griess assay. Results are from one representative experiment of three independent experiment conducted, and data are the mean ⫾ SD.

Results Nod1 stimulation induces chemokine secretion in mesothelial cells via RICK We first assessed the expression of Nod1, Nod2, and several TLRs in mesothelial cells isolated from the mouse peritoneal cavity and BMDMs by quantitative real-time PCR analysis. The analyses revealed that primary mesothelial cells expressed Nod1 and TLR2 at levels that were almost comparable to those in BMDMs (Fig. 1A). In contrast, Nod2 was expressed at higher levels in BMDMs than mesothelial cells and TLR4 expression was higher in mesothelial cells than in BMDMs (Fig. 1A). Furthermore, the RICK protein was expressed at comparable levels in BMDMs and mesothelial cells and the expression, as expected, was abolished in cells isolated form RICK-deficient mice (Fig. 1B). To assess a functional role for Nod1 signaling in chemokine secretion, we stimulated primary mesothelial cells from wild-type and RICK-deficient mice with KF1B, a synthetic Nod1 agonist, and several TLR ligands as a control. We found that CXCL1 and CCL2 were induced by stimulation of mesothelial cells with KF1B and several TLR agonists including lipoteichoic acid (TLR2), synthetic lipid A (TLR4), LPS

517

FIGURE 5. IFN-␥ enhances KF1B-mediated CXCL1 secretion in mesothelial cells from wild-type (WT) but not RICK-deficient mice. A, Mesothelial cells from wild-type mice were stimulated with various concentrations of KF1B with and without IFN-␥ (1 ␮g/ml). B, In another experiment, wild-type mesothelial cells untreated or pretreated with IFN-␥ for 24 h were stimulated with various concentrations of KF1B. Supernatant was collected at 24 h after stimulation, and CXCL1 was measured by ELISA. C, Mesothelial cells from wild-type and RICK-deficient mice were untreated or pretreated with IFN-␥ for 24 h were stimulated with various concentrations of KF1B. Results are from one representative experiment of two independent experiments, and data are mean ⫾ SD.

(TLR4), poly(I:C) (TLR3), and flagellin (TLR5) (Fig. 1C). The induction of chemokine secretion by TLR ligands is consistent with the observation that the corresponding TLRs are expressed in mouse mesothelial cells (27). Notably, secretion of both CXCL1 and CCL2 induced by KF1B, but not TLR agonists, was abrogated in mesothelial cells deficient in RICK (Fig. 1, C and D). Similarly, the induction of CXCL1 induced by KF1B was abolished in mesothelial cells derived from Nod1-deficient mice that was used as control (Fig. 1E). These results indicate that Nod1 stimulation induces secretion of chemokines in mesothelial cells through RICK, whereas that induced by TLRs is independent of RICK. Nod1 signaling enhances TLR-induced CXCL1 secretion in mesothelial cells Given that mesothelial cells respond to both Nod1 and TLR agonists via different pathways, we tested whether the secretion of CXCL1 triggered by TLR ligands might be regulated by Nod1 stimulation. Dose response experiments revealed that 10 ng/ml, but not with 1 or 0.1 ng/ml, of LPS-induced detectable CXCL1 secretion over background in the absence of Nod1 stimulation

518


FIGURE 6. RICK is required for chemokine production and neutrophil recruitment induced by Nod1 stimulation. Wild-type (n ⫽ 4) and RICK-deficient (n ⫽ 4) mice were i.p. injected with 50 ␮g of KF1B. A, Blood was collected at different time points and serum CXCL1, CCL2, and CXCL2 was measured by ELISA. Another group of wild-type and RICK-deficient mice (n ⫽ 4 each) i.p. administered KF1B were sacrificed at 3 h after injection, and peritoneal fluid was collected. B, CXCL1 and CCL2 were measured by ELISA. The third group of mice (n ⫽ 9 each) was i.p. administered KF1B and sacrificed 24 h after injection. Peritoneal fluid was then collected. Control mice were i.p. administered PBS. Neutrophil recruitment in the peritoneal fluid was examined by cytospin and Diff-Quik staining (C) and the percentage of neutrophils among total peritoneal cells was obtained (D). Data are presented as mean ⫾ SE.

(Fig. 2, A and B). Furthermore, dose response experiments showed that 1 ␮g/ml KF1B was required to induce significant levels of CXCL1 in mesothelial cells when the response was assessed 3 or 6 h poststimulation (Fig. 2, C and D). Notably, mesothelial cell production of CXCL1 induced by LPS was enhanced in an additive manner by coincubation with KF1B, when compared with that induced by LPS or KF1B alone (Fig. 2). Consistent with results showed in Fig. 1, the enhancement of LPS-mediated CXCL1 secretion was abrogated in mesothelial cells deficient in RICK (Fig. 2). These results indicate that Nod1 signaling enhances LPS-induced CXCL1 production in mesothelial cells. Nod1 and RICK are required for NF-␬B and MAPK activation induced by KF1B in mesothelial cells Secretion of chemokines by microbial stimuli is induced transcriptionally via NF-␬B and MAPK activation (11). To determine whether KF1B induces NF-␬B and MAPK activation via Nod1 and RICK in mesothelial cells, extracts from wild-type and mutant cells were prepared at different times after KF1B stimulation and immunoblotted with Abs that recognize activated forms of NF-␬B, ERK, JNK, and p38. By 15–30 min of stimulation, KF1B induced phosphorylation and degradation of I␬-B␣, events that are associated with NF-␬B activation, as well as phosphorylation of p38 and ERK, which was abolished or greatly inhibited in Nod1- and RICK-deficient mesothelial cells (Fig. 3). Similarly, KF1B stimulation induced JNK phosphorylation and this event was abrogated in both Nod1- and RICK-deficient mesothelial cells (Fig. 3). These results indicate that Nod1 and RICK are required for signaling induced by KF1B stimulation in mesothelial cells.

Nod1 stimulation induces iNOS expression and NO production through RICK in mesothelial cells Induction NO production in macrophages by bacterial products is important for host defense against intracellular bacteria (28). To determine whether Nod1 stimulation regulates the expression of iNOS, wild-type, and RICK-deficient mesothelial cells were stimulated with KF1B and iNOS mRNA levels were measured over time. The expression of iNOS was induced in mesothelial cells by stimulation with KF1B with a peak response at 3 h postchallenge in wild-type but not in RICK-deficient cells (Fig. 4A). IFN-␥ is a cytokine that is produced by peritoneal T cells in response to infection and modulates chemokine secretion by mesothelial cells (29 –31). Therefore, we tested whether IFN-␥ regulates Nod1-mediated immune responses in mesothelial cells. We found that costimulation with IFN-␥ and KF1B, but not with each agent alone, induced significant NO levels in mesothelial cells (Fig. 4, B and C). The production of NO induced by costimulation with IFN-␥ and KF1B required RICK (Fig. 4, B and C). These results indicate that Nod1 stimulation via RICK induces iNOS and cooperates with IFN-␥ for NO production in mesothelial cells. IFN-␥ enhances CXCL1 secretion induced by Nod1 stimulation in mesothelial cells We next tested whether IFN-␥ regulates the ability of Nod1 to induce chemokine secretion. We found that costimulation of mesothelial cells with Nod1 agonist and IFN-␥ enhanced the production of CXCL1 when compared with stimulation with KF1B or IFN-␥ alone (Fig. 5A). Similarly, KF1B stimulation of mesothelial


519 RICK regulates chemokine production and neutrophil recruitment in the peritoneal cavity. RICK contributes to CXCL1 responses in response to i.p. L. monocytogenes infection The results shown indicate that Nod1 stimulation through RICK induces the production of chemokines in mesothelial cells. To determine whether RICK plays a role in chemokine production in response to bacterial infection, we infected wild-type and RICK-deficient mesothelial cells with L. monocytogenes, an intracellular bacterium that expresses the Nod1-stimulating meso-DAP structure and can cause peritonitis in patients with certain underlying conditions such as cirrhosis or diabetes mellitus (32, 33). The results revealed reduced levels of CXCL1 in Nod1- or RICK-deficient mesothelial cells when compared with wild-type cells at 6 h postinfection (Fig. 7). Similar results were obtained when production of CXCL1 was assayed 18 h after infection (data not shown). These results indicate that Nod1 and RICK signaling contribute to chemokine production by mesothelial cells in response to L. monocytogenes infection.

Discussion

FIGURE 7. Nod1 and RICK contribute to CXCL1 production induced by L. monocytogenes infection in mesothelial cells. Mesothelial cells from wild-type (A), Nod1-deficient (A), or RICK-deficient mice (B) were stimulated with live L. monocytogenes at the indicated MOI. Supernatant was collected 6 h after infection, and CXCL1 was measured by ELISA. Results are from one representative experiment of two independent experiments conducted, and data are mean ⫾ SD.

cells previously treated with IFN-␥ increased CXCL1 secretion over that observed by incubation with each stimulus alone (Fig. 5B). The enhancement of Nod1-mediated CXCL1 secretion by IFN-␥ was abolished in mesothelial cells deficient in RICK (Fig. 5C). These results indicate that the production of CXCL1 induced by Nod1 stimulation via RICK is regulated by IFN-␥. RICK mediates chemokine production and neutrophil recruitment induced by i.p. Nod1 stimulation To examine the role of Nod1 signaling in chemokine production in vivo, KF1B was i.p. administered to wild-type and RICK-deficient mice and the serum and peritoneal levels of chemokines were determined at different times postchallenge. At i.p. administration of the Nod1 agonist, but not control PBS, induced serum production of CCL2, CXCL1, and CXCL2, which was abrogated in RICK-deficient mice (Fig. 6A). Thus, serum production of chemokines, which is likely to reflect secretion by multiple cell types including mesothelial cells, requires RICK. The production of CXCL1 and CCL2 was also elicited in the peritoneal fluid of mice after i.p. administration of KF1B and this was reduced significantly in RICK-null mice (Fig. 6B). To assess whether RICK regulates neutrophil recruitment, KF1B was i.p. administrated to wild-type and RICK-deficient mice and the number of neutrophils was measured 24 h after challenge. After KF1B administration, ⬃20% of the cells in the peritoneal fluid were neutrophils when compared with ⬍2% after PBS administration (Fig. 6, C and D). Notably, the number of neutrophils elicited by KF1B administration was greatly reduced in the peritoneal fluid of RICK-deficient mice (Fig. 6D). These results indicate that Nod1 stimulation via

Inflammatory processes of the mesothelium such as peritonitis induced by bacterial products or infection are a common cause of pathology and mortality in humans, but the mechanism by which mesothelial cells sense bacteria to trigger innate immune responses is poorly understood. In the current studies, we provide evidence that stimulation of mesothelial cells with the dipeptide iE-DAP (KF1B), a Nod1 agonist, induces secretion of the chemokines CXCL1 and CCL2. Secretion of chemokines induced by Nod1 stimulation required RICK, a downstream effector that mediates the activation of NF-␬B and MAPK signaling pathways. These results indicate Gram-negative and Gram-positive bacteria that express the critical meso-DAP amino acid residue can be sensed by mesothelial cells through Nod1 to trigger inflammatory responses. Previous work revealed that mesothelial cells express TLRs and respond to LPS stimulation through TLR4 (27). We confirmed and extended these earlier observations by showing that mesothelial cells produce CXCL1 and CCL2 upon stimulation not only with LPS and its proinflammatory moiety lipid A, but also with TLR2, TLR3, and TLR5 agonists in a RICK-independent manner. Furthermore, Nod1 stimulation via RICK enhanced the production of chemokines induced by LPS. However, we found that the response induced by Nod1 and TLR4 agonists was additive and not synergistic as it was previously observed for Nod1 or Nod2 and TLRs in macrophages (15, 20, 22, 34). Epithelial cells lining the intestinal tract express Nod1 and also respond to Nod1 stimulation by secreting chemokines (22, 35, 36). However, unlike mesothelial cells, intestinal epithelial cells are hyporesponsive to several TLR agonists including LPS presumably to avoid inappropriate induction of inflammatory responses by commensal bacteria and their products (37). Together these results indicate that mesothelial cells function as bona fide innate immune cells to sense bacteria and initiate host defense immune responses in mesothelial surfaces. Secretion of chemokines by mesothelial cells has been shown to promote the influx of neutrophils and monocytes from the vascular compartment into the serosal cavity via transmesothelial migration (10). Previous studies revealed that i.p. administration of the Nod1 agonist into mice induced CXCL1 mRNA levels in peritoneal tissue, which was accompanied of peritoneal neutrophil influx in wild-type but not Nod1-null mice (22). We extend these observations by showing that mesothelial cells respond to Nod1 stimulation by producing CXCL1 and CCL2 and demonstrate that RICK is required for both chemokine production and neutrophil recruitment into the peritoneal cavity upon Nod1 stimulation. IFN-␥ is produced locally in response

520


to peritoneal infection and could contribute to innate immunity through several mechanisms including modulation of CXC and CC chemokine secretion (29 –31). We show in this study that Nod1 stimulation via RICK and IFN-␥ synergizes to induce secretion of CXCL1 and NO production by mesothelial cells. This synergism could be mechanistically explained, at least in part, by the finding that IFN-␥ induces Nod1 expression (38). Although the relevance of the synergistic induction of CXCL1 remains to be elucidated, it is possible that cooperation between Nod1 and IFN-␥ might play a role in the production of optimal chemokine or NO responses when the amount of Nod1 ligand produced during bacterial infection is limiting. Because there is wide variation in the levels of Nod1-stimulatory activity produced by individual bacteria, the latter scenario might be particularly relevant during infection with bacteria such as L. monocytogenes that produce low levels of Nod1 ligand (34). We found that 100 ng/ml KF1B was capable of inducing responses by mesothelial cells. Overnight culture of ⬃109 L. monocytogenes can produce ⬃3 kU/ml of Nod1-stimulatory activity, which corresponds to ⬃60 ng/ml KF1B (34). Certain bacteria such as Bacillus species produce 20- to 40-fold higher concentration of Nod1-stimulatory activity than L. monocytogenes (34). Thus, the amount of KF1B appears to be close or within the physiological range. However, the concentration of Nod1-stimulatory products produced by L. monocytogenes and other bacteria at the site of infection or present inside host cells is unknown. Thus, further work is needed to determine the physiological relevance of our observations. Several studies have assessed the response of mesothelial cells to bacterial products, chemokines and cytokines (27, 29). However, few reports have examined chemokine responses induced by infection with pathogenic bacteria and none, to our knowledge, have determined the innate immune signaling pathways involved (27, 29). We showed in this study that mesothelial cells produce CXCL1 after L. monocytogenes infection and that in the absence of RICK, the chemokine response is significantly reduced when compared with wildtype cells. L. monocytogenes is known to stimulate cytokine and chemokine production via TLR2 and TLR5 in macrophages (39, 40). Consistently, we provide evidence that that mesothelial cells secrete CXCL1 in response to TLR2 and TLR5 agonists. Because the absence of RICK resulted in decrease, but not abrogation, of the CXCL1 response, these results suggest the existence of functional redundancy between RICK and TLR signaling pathways in the response of mesothelial cells to L. monocytogenes. The peritoneal cavity contains a mixture of hemopoetic cells including macrophages and lymphocytes in addition to mesothelial cells. Further studies are needed to understand the contribution of these distinct mesothelial cell populations and their innate immune receptors to host defense and inflammatory processes in vivo.

Acknowledgments We are grateful to Richard Flavell and Tak Mak for mutant mice and Joel Whitfield from the Cellular Immunology Core Facility of the University of Michigan Cancer Center for ELISA.

Disclosures The authors have no financial conflict of interest.

References 1. Mutsaers, S. E. 2004. The mesothelial cell. Int. J. Biochem. Cell Biol. 36: 9 –16. 2. Topley, N., T. Liberek, A. Davenport, F. K. Li, H. Fear, and J. D. Williams. 1996. Activation of inflammation and leukocyte recruitment into the peritoneal cavity. Kidney Int. Suppl. 56: S17–S21. 3. Matsukawa, A., C. M. Hogaboam, N. W. Lukacs, and S. L. Kunkel. 2000. Chemokines and innate immunity. Rev. Immunogenet. 2: 339 –358. 4. Witowski, J., K. Pawlaczyk, A. Breborowicz, A. Scheuren, M. Kuzlan-Pawlaczyk, J. Wisniewska, A. Polubinska, H. Friess, G. M. Gahl, U. Frei, and A. Jo¨rres. 2000. IL-17 stimulates intraperitoneal neutrophil infiltration through the release of GRO␣ chemokine from mesothelial cells. J. Immunol. 165: 5814 –5821.

5. Visser, C. E., J. Tekstra, J. J. Brouwer-Steenbergen, C. W. Tuk, D. M. Boorsma, S. C. Sampat-Sardjoepersad, S. Meijer, R. T. Krediet, and R. H. Beelen. 1998. Chemokines produced by mesothelial cells: huGRO-␣, IP-10, MCP-1 and RANTES. Clin. Exp. Immunol. 112: 270 –275. 6. Tekstra, J., C. E. Visser, C. W. Tuk, J. J. Brouwer-Steenbergen, C. W. Burger, R. T. Krediet, and R. H. Beelen. 1996. Identification of the major chemokines that regulate cell influxes in peritoneal dialysis patients. J. Am. Soc. Nephrol. 7: 2379 –2384. 7. Jose, P. J., P. D. Collins, J. A. Perkins, B. C. Beaubien, N. F. Totty, M. D. Waterfield, J. Hsuan, and T. J. Williams. 1991. Identification of a second neutrophil-chemoattractant cytokine generated during an inflammatory reaction in the rabbit peritoneal cavity in vivo: purification, partial amino acid sequence and structural relationship to melanoma-growth-stimulatory activity. Biochem. J. 278(Pt. 2): 493– 497. 8. Beaubien, B. C., P. D. Collins, P. J. Jose, N. F. Totty, J. Hsuan, M. D. Waterfield, and T. J. Williams. 1990. A novel neutrophil chemoattractant generated during an inflammatory reaction in the rabbit peritoneal cavity in vivo: purification, partial amino acid sequence and structural relationship to interleukin 8. Biochem. J. 271: 797– 801. 9. Betjes, M. G., C. E. Visser, D. Zemel, C. W. Tuk, D. G. Struijk, R. T. Krediet, L. Arisz, and R. H. Beelen. 1996. Intraperitoneal interleukin-8 and neutrophil influx in the initial phase of a CAPD peritonitis. Perit. Dial. Int. 16: 385–392. 10. Li, F. K., A. Davenport, R. L. Robson, P. Loetscher, R. Rothlein, J. D. Williams, and N. Topley. 1998. Leukocyte migration across human peritoneal mesothelial cells is dependent on directed chemokine secretion and ICAM-1 expression. Kidney Int. 54: 2170 –2183. 11. Akira, S., S. Uematsu, and O. Takeuchi. 2006. Pathogen recognition and innate immunity. Cell 124: 783– 801. 12. Inohara, C., C. McDonald, and G. Nuńez. 2005. NOD-LRR proteins: role in hostmicrobial interactions and inflammatory disease. Annu. Rev. Biochem. 74: 355–383. 13. Ting, J. P., and B. K. Davis. 2005. CATERPILLER: a novel gene family important in immunity, cell death, and diseases. Annu. Rev. Immunol. 23: 387– 414. 14. Inohara, N., Y. Ogura, A. Fontalba, O. Gutierrez, F. Pons, J. Crespo, K. Fukase, S. Inamura, S. Kusumoto, M. Hashimoto, et al. 2003. Host recognition of bacterial muramyl dipeptide mediated through NOD2: implications for Crohn’s disease. J. Biol. Chem. 278: 5509 –5512. 15. Chamaillard, M., M. Hashimoto, Y. Horie, J. Masumoto, S. Qiu, L. Saab, Y. Ogura, A. Kawasaki, K. Fukase, S. Kusumoto, et al. 2003. An essential role for NOD1 in host recognition of bacterial peptidoglycan containing diaminopimelic acid. Nat. Immunol. 4: 702–707. 16. Girardin, S. E., I. G. Boneca, J. Viala, M. Chamaillard, A. Labigne, G. Thomas, D. J. Philpott, and P. J. Sansonetti. 2003. Nod2 is a general sensor of peptidoglycan through muramyl dipeptide (MDP) detection. J. Biol. Chem. 278: 8869 – 8872. 17. Girardin, S. E., I. G. Boneca, L. A. Carneiro, A. Antignac, M. Jehanno, J. Viala, K. Tedin, M. K. Taha, A. Labigne, U. Zahringer, et al. 2003. Nod1 detects a unique muropeptide from gram-negative bacterial peptidoglycan. Science 300: 1584 –1587. 18. Kobayashi, K. S., M. Chamaillard, Y. Ogura, O. Henegariu, N. Inohara, G. Nuńez, and R. A. Flavell. 2005. Nod2-dependent regulation of innate and adaptive immunity in the intestinal tract. Science 307: 731–734. 19. Kobayashi, K., N. Inohara, L. D. Hernandez, J. E. Galan, G. Nuńez, C. A. Janeway, R. Medzhitov, and R. A. Flavell. 2002. RICK/Rip2/CARDIAK mediates signalling for receptors of the innate and adaptive immune systems. Nature 416: 194 –199. 20. Park, J. H., Y.-G. Kim, C. McDonald, T.-D. Kanneganti, M. Hasegawa, M. Body-Malapel, N. Inohara, and G. Nuńez. 2007. RICK/RIP2 mediates innate immune responses induced through Nod1 and Nod2 but not TLRs. J. Immunol. 178: 2380 –2386. 21. Liu, W.-C., M. Oikawa, K. Fukase, Y. Suda, and S. Kusumoto. 1999. A divergent synthesis of lipid A and its chemically stable unnatural analogues. Bull. Chem. Soc. Jpn. 72: 1377–1385. 22. Masumoto, J., K. Yang, S. Varambally, M. Hasegawa, S. A. Tomlins, S. Qiu, Y. Fujimoto, A. Kawasaki, S. J. Foster, Y. Horie, et al. 2006. Nod1 acts as an intracellular receptor to stimulate chemokine production and neutrophil recruitment in vivo. J. Exp. Med. 203: 203–213. 23. Celada, A., P. W. Gray, E. Rinderknecht, and R. D. Schreiber. 1984. Evidence for a ␥-interferon receptor that regulates macrophage tumoricidal activity. J. Exp. Med. 160: 55–74. 24. Gary Lee, Y. C., D. Melkerneker, P. J. Thompson, R. W. Light, and K. B. Lane. 2002. Transforming growth factor ␤ induces vascular endothelial growth factor elaboration from pleural mesothelial cells in vivo and in vitro. Am. J. Respir. Crit. Care Med. 165: 88 –94. 25. Kim, Y.-G., T. Ohta, T. Takahashi, A. Kushiro, K. Nomoto, T. Yokokura, N. Okada, and H. Danbara. 2006. Probiotic Lactobacillus casei activates innate immunity via NF-␬B and p38 MAP kinase signaling pathways. Microbes Infect. 8: 994 –1005. 26. Green, L. C., D. A. Wagner, J. Glogowski, P. L. Skipper, J. S. Wishnok, and S. R. Tannenbaum. 1982. Analysis of nitrate, nitrite, and [15N]nitrate in biological fluids. Anal. Biochem. 126: 131–138. 27. Kato, S., Y. Yuzawa, N. Tsuboi, S. Maruyama, Y. Morita, T. Matsuguchi, and S. Matsuo. 2004. Endotoxin-induced chemokine expression in murine peritoneal mesothelial cells: the role of Toll-like receptor 4. J. Am. Soc. Nephrol. 15: 1289 –1299. 28. Bogdan, C., M. Rollinghoff, and A. Diefenbach. 2000. Reactive oxygen and reactive nitrogen intermediates in innate and specific immunity. Curr. Opin. Immunol. 12: 64 –76. 29. Robson, R. L., R. M. McLoughlin, J. Witowski, P. Loetscher, T. S. Wilkinson, S. A. Jones, and N. Topley. 2001. Differential regulation of chemokine production in human peritoneal mesothelial cells: IFN-␥ controls neutrophil migration across the mesothelium in vitro and in vivo. J. Immunol. 167: 1028 –1038. 30. Dasgupta, M. K., M. Larabie, and P. F. Halloran. 1994. Interferon-␥ levels in peritoneal dialysis effluents: relation to peritonitis. Kidney Int. 46: 475– 481.

The Journal of Immunology 31. Lu, Y., B. Hylander, and A. Brauner. 1996. Interleukin-10, interferon ␥, interleukin-2, and soluble interleukin-2 receptor alpha detected during peritonitis in the dialysate and serum of patients on continuous ambulatory peritoneal dialysis. Perit. Dial. Int. 16: 607– 612. 32. Adeonigbagbe, O., A. Khademi, M. Karowe, N. Gualtieri, and J. Robilotti. 2000. Listeria monocytogenes peritonitis: an unusual presentation and review of the literature. J. Clin. Gastroenterol. 30: 436 – 437. 33. Nolla-Salas, J., M. Almela, I. Gasser, C. Latorre, M. Salvado, and P. Coll. 2002. Spontaneous Listeria monocytogenes peritonitis: a population-based study of 13 cases collected in Spain. Am. J. Gastroenterol. 97: 1507–1511. 34. Hasegawa, M., K. Yang, M. Hashimoto, J. H. Park, Y.-G. Kim, Y. Fujimoto, G. Nuńez, K. Fukase, and N. Inohara. 2006. Differential release and distribution of Nod1 and Nod2 immunostimulatory molecules among bacterial species and environments. J. Biol. Chem. 281: 29054 –29063. 35. Kim, J. G., S. J. Lee, and M. F. Kagnoff. 2004. Nod1 is an essential signal transducer in intestinal epithelial cells infected with bacteria that avoid recognition by Toll-like receptors. Infect. Immun. 72: 1487–1495.

521 36. Uehara, A., Y. Fujimoto, A. Kawasaki, S. Kusumoto, K. Fukase, and H. Takada. 2006. Meso-diaminopimelic acid and meso-lanthionine, amino acids specific to bacterial peptidoglycans, activate human epithelial cells through NOD1. J. Immunol. 177: 1796 –1804. 37. Chamaillard, M., N. Inohara, and G. Nuńez. 2004. Battling enteroinvasive bacteria: Nod1 comes to the rescue. Trends Microbiol. 12: 529 –532. 38. Hisamatsu, T., M. Suzuki, and D. K. Podolsky. 2003. Interferon-␥ augments CARD4/NOD1 gene and protein expression through interferon regulatory factor-1 in intestinal epithelial cells. J. Biol. Chem. 278: 32962–32968. 39. Ozoren, N., J. Masumoto, L. Franchi, T.-D. Kanneganti, M. Body-Malapel, I. Erturk, R. Jagirdar, L. Zhu, N. Inohara, J. Bertin, et al. 2006. Distinct roles of TLR2 and the adaptor ASC in IL-1␤/IL-18 secretion in response to Listeria monocytogenes. J. Immunol. 176: 4337– 4342. 40. Hayashi, F., K. D. Smith, A. Ozinsky, T. R. Hawn, E. C. Yi, D. R. Goodlett, J. K. Eng, S. Akira, D. M. Underhill, and A. Aderem. 2001. The innate immune response to bacterial flagellin is mediated by Toll-like receptor 5. Nature 410: 1099 –1103.