A Soluble Form of LMIR5/CD300b Amplifies Lipopolysaccharide-Induced Lethal Inflammation in Sepsis

Yoshinori Yamanishi, Mariko Takahashi, Kumi Izawa, Masamichi Isobe, Shinichi Ito, Akiho Tsuchiya, Akie Maehara, Ayako Kaitani, Tomoyuki Uchida, Katsuhiro Togami, Yutaka Enomoto, Fumio Nakahara, Toshihiko Oki, Masunori Kajikawa, Hiroki Kurihara, Toshio Kitamura and Jiro Kitaura
J Immunol August 15, 2012, 189 (4) 1773-1779; DOI: https://doi.org/10.4049/jimmunol.1201139

Abstract

Leukocyte mono-Ig–like receptor 5 (LMIR5, also called CD300b) is an activating receptor expressed in myeloid cells. We have previously demonstrated that T cell Ig mucin 1 works as a ligand for LMIR5 in mouse ischemia/reperfusion injury of the kidneys. In this article, we show that LMIR5 is implicated in LPS-induced sepsis in mice. Notably, neutrophils constitutively released a soluble form of LMIR5 (sLMIR5) through proteolytic cleavage of surface LMIR5. Stimulation with TLR agonists augmented the release of sLMIR5. LPS administration or peritonitis induction increased serum levels of sLMIR5 in mice, which was substantially inhibited by neutrophil depletion. Thus, neutrophils were the main source of LPS-induced sLMIR5 in vivo. On the other hand, i.p. administration of LMIR5-Fc, a surrogate of sLMIR5, bound to resident macrophages (Mϕ) and stimulated transient inflammation in mice. Consistently, LMIR5-Fc induced in vitro cytokine production of peritoneal Mϕ via its unknown ligand. Interestingly, LMIR5 deficiency profoundly reduced systemic cytokine production and septic mortality in LPS-administered mice, although it did not affect in vitro cytokine production of LPS-stimulated peritoneal Mϕ. Importantly, the resistance of LMIR5-deficient mice to LPS- or peritonitis-induced septic death was decreased by LMIR5-Fc administration, implicating sLMIR5 in LPS responses in vivo. Collectively, neutrophil-derived sLMIR5 amplifies LPS-induced lethal inflammation.

Introduction

Sepsis is a complex clinical syndrome that results from a harmful host response to infection. The inflammatory response is initiated after detection of microbial infection by pattern-recognition receptors (PRRs) in immune cells; PRRs recognize pathogen-associated molecular patterns that are expressed by microorganisms (1–6). Among PRRs, TLRs play a crucial role in the initiation of inflammatory responses: TLR4 recognizes LPS, a component of the outer membrane of Gram-negative bacteria (7, 8). During sepsis, systemic activation of immune cells by high levels of pathogen-associated molecular patterns causes an imbalanced cytokine response (1–8). In addition, innate immune responses are regulated by receptors that signal through ITAM, ITIM, or both (9, 10). Complete understanding of the immunopathology of sepsis will culminate in the establishment of sepsis-stage–specific therapy (3–5).

Leukocyte mono-Ig–like receptor (LMIR, also called CD300) (11–15) belongs to a paired activating and inhibitory receptor family (9, 10, 16, 17). Mouse LMIR is also known to be a CMRF-35–like molecule (18) or myeloid-associated Ig-like receptor (19). LMIR5 is an activating receptor that interacts with DNAX-activating protein of 12 kDa (DAP12), an ITAM-containing adaptor molecule (13). LMIR5 is predominantly expressed in myeloid cells including neutrophils, macrophages (Mϕ), and mast cells (13). We have recently identified T cell Ig mucin 1 (TIM1) and TIM4 as endogenous ligands for LMIR5; the intact Ig-like domain of LMIR5 is required for the binding to TIM1/4 (15, 20, 21). LMIR5 deficiency alleviates mouse ischemia/reperfusion injury in the kidneys in which TIM1 expression is upregulated (15, 21). However, the pathophysiological role of LMIR5 in other types of inflammation models remains elusive.

In this study, we demonstrate that a soluble form of LMIR5 (sLMIR5) is released by neutrophils, and that the release is augmented by TLR agonists. Remarkably, LMIR5−/− mice are protected against LPS- or peritonitis-induced death, implicating LMIR5 in LPS-induced sepsis. According to recent studies, triggering receptors expressed on myeloid cells-1 (TREM-1) amplifies inflammatory response in the context of microbial infection (22–24). Interestingly, a soluble form of TREM-1 (sTREM-1) is released in infection and counterregulates TREM-1–mediated inflammation (22–26). In contrast, we demonstrate that sLMIR5 is an amplifier of LPS-induced inflammation, highlighting a novel mechanism of LPS-induced lethality in sepsis.

Materials and Methods

Cells and mice

Murine cell lines used in this study were as follows: RAW264.7, Ba/F3, and BW5147. Bone marrow (BM)-derived mast cell (BMMC) and peritoneal Mϕ were generated as described previously (12, 15, 27). Neutrophils were isolated from BM by using three-layer gradient as described previously (28). We used LMIR5−/− female mice and their wild type (WT) littermates. LMIR5−/− mice had been backcrossed at least eight generations with C57BL/6 mice (Charles River Laboratories Japan) (15). All the procedures were approved by an Institutional Review Committee of the University of Tokyo (approval no. 20-8).

Abs and other reagents

Anti-LMIR5 polyclonal Ab was obtained from R&D Systems. Anti-LMIR5 mAb (2C12) was from ACTGen. Both anti-LMIR5 polyclonal Ab and anti-LMIR5 mAb (2C12) specifically recognized mouse LMIR5. The recombinant extracellular domain of LMIR5 (aa 18–157) was obtained from R&D Systems. Anti-Flag mAb (M2), goat IgG, human IgG1, and LPS (Escherichia coli 0111:B4) were from Sigma-Aldrich. PE-conjugated donkey F(ab′)2 anti-goat IgG and anti-human IgG Abs were from Jackson Immunoresearch Laboratories. Anti–Gr-1 Ab, rat IgG2b, FITC-conjugated anti-mouse F4/80, Gr-1, and CD11b Abs, and PE-conjugated anti–Gr-1 Ab were from eBioscience. Anti-TIM4 mAb (RMT4-54) was a gift from Drs. H. Akiba and K. Okumura (Juntendo University, Tokyo, Japan). Oligodeoxynucleotide (ODN) control, ODN1826, Loxoribine, Flagellin, polyriboinosinic-polyribocytidylic acid, and peptidoglycan were purchased from InvivoGen. GM6001 (catalog no. 364205) and GM control (catalog no. 364210) were from Calbiochem.

Gene expression analysis

Relative gene expression of LMIR5 or IL-6 among samples was measured by real-time RT-PCR using gene-specific primers as described previously (13, 15).

Generation of Fc fusion proteins

The Fc fusion proteins were purified as described previously (15). To prepare the Fc segment alone, we used pME18S-hIgG1 Fc vector containing a signaling lymphocyte-activation molecule (SLAM)/CD150 signal sequence (a gift from Dr. H. Arase, Osaka University, Osaka, Japan) (29). Endotoxin levels of Fc fusion proteins used in this study were <0.01 ng/μg protein, when measured by Limulus Amebocyte Lysate (Lonza).

Flow cytometry

Cells were stained as described previously (15). The mice were i.p. injected with 100 μg LMIR5-Fc to identify the target cells of LMIR5-Fc in vivo. Five minutes after injection, the peritoneal cells were collected and stained with PE-conjugated F(ab′)2 donkey anti-human IgG and FITC-conjugated anti-mouse F4/80. Flow cytometric analysis was performed with FACSCalibur (BD Biosciences) equipped with CellQuest software and FlowJo software.

Transfection and infection

Retroviral transfection and infection were performed as described previously (13, 15, 30). Retroviruses were generated by transient transfection of PLAT-E packaging cells (31).

Biochemistry

Western blotting was performed as described previously (12, 13). In brief, the culture supernatants of RAW264.7 cells or neutrophils were immunoprecipitated with anti-LMIR5 polyclonal Ab (R&D Systems) or goat IgG as control to detect sLMIR5. The immunoprecipitates were immunoblotted with biotinylated anti-LMIR5 polyclonal Ab (R&D Systems) followed by HRP-conjugated avidin (eBioscience).

Measurement of sLMIR5

Neutrophils were incubated with or without LPS in the presence of 100 ng/ml G-CSF (R&D Systems). The concentrations of sLMIR5 in the culture supernatants were measured by ELISA using anti-LMIR5 mAb (2C12; ACTGen) and biotinylated anti-LMIR5 polyclonal Ab (R&D Systems) as capture and detection Abs, respectively.

LPS-induced endotoxinemia in mice

Mice (20–25 g) were i.p. injected with LPS dissolved in 500 μl PBS. In some experiments, 100 μg LMIR5-Fc or control Fc was coinjected with LPS. Mice were sacrificed at different time intervals. Peritoneal lavage fluid (PLF; 1 ml PBS) and serum were collected and assayed for sLMIR5, IL-6, and TNF-α by ELISA as described previously (15, 22). Survival was assessed every day for at least 7 d. We used 30 and 4 mg/kg LPS in the experiments for assessing survival and measuring serum cytokines/sLMIR5, respectively.

The colon ascendens stent peritonitis model

The surgical procedure of colon ascendens stent peritonitis (CASP) was performed as described previously (32). In brief, under complete anesthesia, the abdominal wall was opened. An 18-gauge venous catheter was inserted through into the lumen of the ascending colon and fixed with two sutures. Fluid resuscitation was carried out using 0.3 ml sterile sodium chloride. After surgery, the mice were either observed over a period of 7 d for survival studies or were sacrificed at indicated times after CASP to obtain serum.

Statistical analysis

Data are shown as means ± SD, and statistical significance was determined by the Student t test.

Results

An sLMIR5 is released from the RAW264.7 cell line

First, we found that LPS stimulation reduced the surface expression levels of LMIR5 in RAW264.7 cells in a time-dependent manner (Fig. 1A). Real-time PCR analysis showed that LPS stimulation did not significantly affect LMIR5 transcript levels in RAW264.7 cells, although it quickly increased mRNA for proinflammatory cytokine IL-6 at transcript levels (Fig. 1B). Therefore, we thought it possible that surface LMIR5 in RAW264.7 cells might be shed by LPS stimulation. To test this, we examined whether sLMIR5 was detected by anti-LMIR5 Ab in the culture supernatants of RAW264.7 cells. As reported previously (13), Western blot analysis showed that an N-glycoprotein LMIR5 in RAW264.7 cell lysates was recognized as a broad band (22–38 kDa), including a major band (25 kDa; Fig. 1C). Notably, anti-LMIR5 Ab detected a shorter form of LMIR5 as a broad band (17–24 kDa), including two major bands (17 and 21 kDa) in the culture supernatants of RAW264.7 cells, but not in a control medium. These results indicated that sLMIR5 was constitutively shed from an extracellular LMIR5 in RAW264.7 cells. In fact, sLMIR5 accumulated in the culture supernatants of RAW264.7 cells in a time-dependent manner (Fig. 1D). Interestingly, the release of sLMIR5 in RAW264.7 cells was augmented in the presence of LPS (Fig. 1D). Next, to evaluate the concentrations of sLMIR5 in the culture supernatants, we established an ELISA system using anti-LMIR5 Abs; LMIR5-Fc containing an extracellular domain of LMIR5 was efficiently detected, whereas neither an extracellular domain of LMIR3 highly homologous with that of LMIR5 or human IgG1 as control was recognized (Supplemental Fig. 1A). With this assay, we were able to determine the concentrations of sLMIR5 in the culture supernatants of RAW264.7 cells expressing high levels of surface LMIR5 (Supplemental Fig. 1B). In contrast, sLMIR5 was not detected in the culture supernatants of Ba/F3 or BW5147 cells expressing no detectable levels of surface LMIR5 (Supplemental Fig. 1B). Moreover, when LMIR5 was further transduced into RAW264.7 cells, higher concentrations of sLMIR5 were detected in the culture supernatants of LMIR5-transduced cells as compared with mock-transduced cells (Supplemental Fig. 1C). Thus, both the sensitivity and the specificity of this assay system were verified. We then determined the concentrations of sLMIR5 or IL-6 in the culture supernatants of RAW264.7 cells stimulated by various concentrations of LPS. Interestingly, even treatment with 1 ng/ml LPS caused the rapid increase of sLMIR5 as early as 2 h after stimulation (Fig. 1E). Of note, stimulation with 10–1000 ng/ml LPS induced a time- and dose-dependent increase of IL-6 production in RAW264.7 cells, whereas the same stimulation led to a nearly identical time-dependent increase of sLMIR5 (Fig. 1E). To test whether stimulation with other TLR ligands induced the release of sLMIR5, we incubated RAW264.7 cells with different concentrations of CpG ODN, Loxoribine, Flagellin, polyriboinosinic-polyribocytidylic acid, or peptidoglycan. ELISA and flow cytometric analysis demonstrated that dose-dependent increase of sLMIR5 as well as the reciprocal reduction of cell-surface LMIR5 levels were caused by stimulation with all the TLR ligands tested (Fig. 1F). Altogether, TLR stimulation led to the increase of sLMIR5 in RAW264.7 cells.

FIGURE 1.
FIGURE 1.

The release of sLMIR5 from RAW264.7 cells. (A) RAW264.7 cells were incubated with or without 100 ng/ml LPS for 12 h (left panel) or for indicated hours (right panel). Staining with anti-LMIR5 Ab (bold line histogram) or with control Ab (shaded histogram) (left panel). The mean fluorescent intensity (MFI) of LMIR5 staining (right panel). (B) Relative gene expression levels of LMIR5 or IL-6 in RAW264.7 cells at different time intervals after stimulation with 100 ng/ml LPS. (C) The cell lysates or culture supernatants of RAW264.7 cells were immunoprecipitated with anti-LMIR5 Ab or control Ab and then immunoblotted with anti-LMIR5 Ab. (D) RAW264.7 cells were incubated with or without 100 ng/ml LPS for indicated hours. The bands correspond to sLMIR5 in the culture supernatants. (E) RAW264.7 cells were incubated with different concentrations of LPS for indicated hours. The concentrations of sLMIR5 or IL-6 in the culture supernatants were measured by ELISA. (F) RAW264.7 cells were stimulated with different concentrations of indicated stimuli for 12 h. The concentrations of sLMIR5 in the culture supernatants were measured by ELISA (upper panel). The MFI of LMIR5 staining (lower panel). (A–F) Data are presented as mean ± SD (*p < 0.05) and represent three independent experiments.

Neutrophils are the main source of LPS-induced sLMIR5 in mice

We next determined whether sLMIR5 was released in the culture supernatants of primary myeloid cells. Notably, sLMIR5 was constitutively released by neutrophils of BM (Fig. 2A). In addition, as in RAW264.7 cells, LPS stimulation increased sLMIR5 and downregulated surface LMIR5 expression in neutrophils (Fig. 2A, 2B). However, detectable levels of sLMIR5 were not released by either peritoneal Mϕ or BMMC, despite the sufficient expression of surface LMIR5 in these cells (Fig. 2A). These results led us to postulate the involvement of neutrophil-specific proteases in the generation of sLMIR5. To test this, we incubated neutrophils with GM6001, a broad-spectrum inhibitor of matrix metalloproteinases (MMPs), in the presence or absence of LPS because several MMPs were known to be expressed and activated in neutrophils (33, 34). Notably, GM6001 significantly impaired both constitutive and LPS-induced release of sLMIR5 from neutrophils (Fig. 2C). These results indicated that sLMIR5 was released from neutrophils by proteolytic cleavage of surface LMIR5 partly through MMPs. To next examine whether sLMIR5 was indeed released in vivo, we measured the concentrations of sLMIR5 in serum or PLF of C57BL/6 mice that had received i.p. injection of LPS. Remarkably, both serum and PLF levels of sLMIR5, as well as IL-6 and TNF-α, rapidly elevated at 1.5 h after injection and diminished through 6–24 h (Fig. 3A and data not shown). Concurrently, we found a transient decrease followed by a gradual recovery of surface expression levels of LMIR5 in BM neutrophils of mice given an injection of LPS (Fig. 3B). In addition, pretreatment of mice with anti–Gr-1 Ab, which was able to deplete circulating neutrophils (Fig. 3C), substantially inhibited the LPS-induced release of sLMIR5 in serum (Fig. 3D). Taken together, these results showed that neutrophils are the main source of LPS-induced sLMIR5 in vivo.

FIGURE 2.
FIGURE 2.

Soluble LMIR5 is released by neutrophils in vitro. (A) Surface expression of LMIR5 in BM neutrophils, peritoneal Mϕ, or BMMC. Staining with anti-LMIR5 Ab (bold line histograms) or with control Ab (shaded histogram, upper panel). BM neutrophils, peritoneal Mϕ, or BMMC were incubated in the presence or absence of 100 ng/ml LPS for indicated hours. The concentrations of sLMIR5 in the culture supernatants were measured by ELISA (lower panel). (B) BM neutrophils were incubated with or without 100 ng/ml LPS for 24 h. Staining with anti-LMIR5 Ab (bold line histograms) or with control Ab (shaded histogram). (C) BM neutrophils were preincubated with 10 μM of indicated MMP inhibitors for 30 min before incubation with or without 100 ng/ml LPS for 24 h. The concentrations of sLMIR5 in the culture supernatants were measured by ELISA. Data are presented as mean ± SD (*p < 0.05, **p < 0.01) and represent three independent experiments.

FIGURE 3.
FIGURE 3.

Neutrophils are the main source of sLMIR5 in vivo. (A and B) WT mice were i.p. injected with 4 mg/kg LPS. Serum and BM cells were collected from the mice at 1.5, 6, or 24 h after injection (n = 6 per group). (A) The concentrations of sLMIR5, IL-6, or TNF-α in serum were measured by ELISA. (B) Surface expression levels of LMIR5 and Gr-1 in BM cells derived from the mice before or at 6 h after injection (left panel). The MFI of LMIR5 staining in Gr-1high BM neutrophils derived from the mice at different time intervals after injection (right panel). Data are presented as mean ± SD and represent three independent experiments. (C and D) WT mice were i.p. injected with 500 μg of either anti–Gr-1 Ab (n = 4) or control Ab (n = 6) 24 h before i.p. injection of 4 mg/kg LPS. Serum was collected from the mice at 1.5 h after LPS injection. (C) Percentages of CD11bhighGr-1high neutrophils in peripheral blood were examined by flow cytometry. (D) The concentrations of sLMIR5 in serum were measured by ELISA. All bars correspond to the means (*p < 0.05).

LMIR5-Fc binds to peritoneal Mϕ, resulting in the production of proinflammatory cytokines

To investigate the in vivo role of sLMIR5, we gave the mice an i.p. injection of LMIR5-Fc, a surrogate for sLMIR5. Notably, administration of LMIR5-Fc, but not control Fc, led to transient but significant levels of IL-6 and TNF-α production in both serum and PLF, although it did not cause lethal inflammation (Fig. 4A and data not shown). To determine the in vivo target of sLMIR5, we examined what types of the cells were bound by LMIR5-Fc after i.p. administration of LMIR5-Fc. Flow cytometric analysis showed that LMIR5-Fc strongly bound to F4/80+ Mϕ among the peritoneal cells (Fig. 4B). In contrast, we did not find the strong binding of LMIR5-Fc to any other cells in BM and blood after i.p. or i.v. administration of sLMIR5-Fc (data not shown). Consistently, peritoneal Mϕ, but not BMMC, were also strongly stained by LMIR5-Fc in vitro (Fig. 4C). In addition, stimulation with LMIR5-Fc, but not with human IgG1 as a control, significantly increased IL-6 or TNF-α production in peritoneal Mϕ, but not in BMMC (Fig. 4D). Moreover, stimulation with LMIR5-Fc led to a dose-dependent production of IL-6 in peritoneal Mϕ (Fig. 4E). Similarly, recombinant LMIR5 harboring only an extracellular domain of LMIR5 stimulated IL-6 production of peritoneal Mϕ in a dose-dependent manner (Supplemental Fig. 2A). It should be noted that pretreatment with anti-LMIR5 Ab, but not control Ab, dose-dependently suppressed IL-6 production of peritoneal Mϕ stimulated by LMIR5-Fc (Fig. 4F). Collectively, these results indicated that LMIR5-Fc activated peritoneal Mϕ through the interaction of an extracellular domain of LMIR5 with its ligand expressed by peritoneal Mϕ. Because LMIR5 bound to TIM4 that was highly expressed in peritoneal Mϕ (15), we asked whether TIM4 was involved in LMIR5-Fc–induced cytokine production of peritoneal Mϕ. To this end, we used LMIR5 del4-Fc lacking both the C-terminal portion of an Ig-like domain and an extracellular stalk region; unlike LMIR5-Fc, LMIR5 del4-Fc fails to bind TIM1 or TIM4 (15). Indeed, LMIR5-Fc, but not LMIR5 del4-Fc, stained TIM1- or TIM4-expressing Ba/F3 cells (Supplemental Fig. 2B). Notably, LMIR5 del4-Fc significantly stained peritoneal Mϕ expressing TIM4, albeit at weaker levels than LMIR5-Fc (Supplemental Fig. 2B). Because peritoneal Mϕ did not express significant levels of TIM1 (15), it is possible that in peritoneal Mϕ, LMIR5-Fc bound to TIM4 as well as to an unknown ligand other than TIM1/4 via different regions of LMIR5. Importantly, stimulation with LMIR5 del4-Fc caused cytokine production of peritoneal Mϕ at levels comparable with stimulation with LMIR5-Fc (Supplemental Fig. 2C). Taken together, these results suggested that the interaction of LMIR5-Fc with an unidentified ligand other than TIM1/4 was involved in LMIR5-Fc–induced cytokine production of peritoneal Mϕ.

FIGURE 4.
FIGURE 4.

LMIR5-Fc bound to and stimulated the cytokine production of peritoneal Mϕ. (A) WT mice were i.p. injected with 100 μg LMIR5-Fc or control Fc. The concentrations of IL-6 or TNF-α in serum of the mice before or at 1.5, 6, or 24 h after injection were measured by ELISA (n = 6 per group). Data are presented as mean ± SD. (B) Five minutes after i.p. injection with 100 μg LMIR5-Fc, the mice were sacrificed. The peritoneal cells were collected and stained with PE-conjugated donkey F(ab′)2 anti-human IgG Ab and FITC-conjugated anti-mouse F4/80 Ab. (C) Staining of peritoneal Mϕ or BMMC with LMIR5-Fc (bold line histograms) or with control Fc (shaded histogram). Data represent four independent experiments. (D) Peritoneal Mϕ or BMMC were incubated with 20 μg/ml LMIR5-Fc or human IgG1 for 24 h. The concentrations of IL-6 or TNF-α in the culture supernatants were measured by ELISA. (E and F) The concentrations of IL-6 in the culture supernatants were measured by ELISA. (E) Peritoneal Mϕ were incubated with various concentrations of LMIR5-Fc, human IgG1, or control Fc for 24 h. (F) After pretreatment of LMIR5-Fc (10 μg/ml) with indicated concentrations of anti-LMIR5 Ab or goat IgG as control, peritoneal Mϕ were stimulated with LMIR5-Fc for 24 h. (C–F) Data are presented as mean ± SD (*p < 0.01) and represent four independent experiments.

LMIR5−/− mice are protected against LPS- or peritonitis-induced lethal inflammation

To delineate the in vivo significance of sLMIR5, we compared LPS responses between WT and LMIR5−/− mice. First, we confirmed that irrespective of LPS administration, sLMIR5 was absent both in the culture supernatants of LMIR5−/− neutrophils and in serum and PLF of LMIR5−/− mice (Supplemental Fig. 3). We then measured levels of proinflammatory cytokines (IL-6 and TNF-α) in the serum or PLF of WT and LMIR5−/− mice before and after i.p. injection of LPS. Intriguingly, when compared with WT mice, LMIR5−/− mice exhibited lower levels of LPS-induced cytokine production in both serum and PLF (Fig. 5A). Importantly, LMIR5 deficiency decreased the mortality induced by LPS administration: all the WT mice died within 4 d after a lethal dose of LPS injection, whereas 6 of 12 LMIR5−/− mice survived even after 7 d (Fig. 5B). These results suggested that LMIR5 deficiency improved LPS-induced death by suppressing proinflammatory cytokine production in mice. In contrast, we found no significant difference in LPS-induced cytokine production in vitro between WT and LMIR5-deficient myeloid cells, including peritoneal Mϕ, indicating that LMIR5 was not intrinsically involved in LPS-TLR4 signaling in these cells (Supplemental Fig. 4). Based on these observations, we speculated that the reduced mortality and cytokine production in LPS-injected LMIR5−/− mice might be explained by the absence of sLMIR5. To test this possibility, we administered LMIR5-Fc or control Fc together with a sublethal dose of LPS to LMIR5−/− mice, and found that the addition of LMIR5-Fc, but not control Fc, accelerated LPS-induced death in LMIR5−/− mice (Fig. 5C). These results suggested that sLMIR5 played a critical role in LPS-induced lethality. Next, we induced CASP as a murine abdominal sepsis model in WT and LMIR5−/− mice. The increase of sLMIR5 in serum of WT mice was confirmed in this model (Fig. 6A). Remarkably, LMIR5−/− mice were resistant to death in a CASP model (Fig. 6B). Serum levels of TNF-α at 4 h after the induction of a CASP model were lower in LMIR5−/− mice in comparison with WT mice (Fig. 6C). In addition, the peritoneal administration of LMIR5-Fc, but not control Fc, accelerated septic death in LMIR5−/− mice that had been subjected to a CASP model (Fig. 6D). Taken together, sLMIR5 was implicated in lethal inflammation induced by a pathophysiological sepsis model.

FIGURE 5.
FIGURE 5.

The resistance of LMIR5-deficient mice to LPS-induced mortality was decreased by LMIR5-Fc administration. (A) WT or LMIR5−/− mice were i.p. injected with 4 mg/kg LPS or PBS (WT/PBS: n = 4; knockout [KO]/PBS: n = 3; WT/LPS: n = 10; KO/LPS: n = 9). The concentrations of IL-6 (upper panel) or TNF-α (lower panel) in serum (left panel) or PLF (right panel) of the mice at 1.5 h after injection were measured by ELISA. All the bars correspond to the means. (B) WT or LMIR5−/− mice were i.p. injected with 30 mg/kg LPS and monitored for survival (n = 12 per group). (C) LMIR5−/− mice were i.p. injected with 30 mg/kg LPS together with 100 μg LMIR5-Fc or control Fc, and monitored for survival (n = 8 per group).

FIGURE 6.
FIGURE 6.

LMIR5−/− mice are resistant to death induced by CASP. (A) WT mice were subjected to CASP model and sacrificed at 1.5, 6, or 24 h after the surgery. The concentrations of sLMIR5 in serum of the mice (n = 4 per group) were measured by ELISA. Data are presented as mean ± SD. (B) WT or LMIR5−/− mice were subjected to CASP model and monitored for survival (n = 12 per genotype). (C) The concentrations of IL-6 (left panel) or TNF-α (right panel) in serum of WT mice (n = 5) or LMIR5−/− mice (n = 4) at 4 h after the induction of CASP were measured by ELISA. All bars correspond to the means (*p < 0.05). (D) LMIR5−/− mice that had been subjected to CASP model were further i.p. injected with 100 μg of either LMIR5-Fc or control Fc and monitored for survival (n = 5 per group).

Discussion

The ideal therapy for sepsis is to suppress the excessive inflammatory responses in its early phase. Therefore, understanding the precise mechanism of the sepsis progression is indispensable to improving therapy. In this study, we delineated a novel mechanism as to how inflammatory responses were accelerated in LPS-induced sepsis: sLMIR5 was an inflammatory mediator linking the initial activation of LPS-stimulated neutrophils with lethal inflammation in sepsis. The precise analysis revealed that neutrophils were the main source of LPS-induced sLMIR5 in mice (Figs. 2, 3). The reason why sLMIR5 is released by RAW264.7 cells (Mϕ cell lines), but not by peritoneal Mϕ, is uncertain. Similar to the release of sLMIR5 by neutrophils, that by RAW264.7 cells was significantly, but not completely, suppressed by treatment with GM6001 (data not shown). Real-time PCR analysis of expression of MMP1, MMP2, MMP3, MMP8, and MMP9 in RAW264.7 cells or peritoneal Mϕ showed that MMP3 and MMP9 were expressed in RAW264.7 cells, whereas several MMPs, including MMP3 and MMP9, were highly expressed in peritoneal Mϕ (data not shown). Therefore, MMPs other than these MMPs or GM6001-sensitive ADAM and/or ADAM-TS family members might be involved in the release of sLMIR5 by RAW264.7 cells. Alternatively, it is possible that MMP3 and/or MMP9 may be aberrantly activated in RAW264.7 cells, but not in peritoneal Mϕ. Although the involvement of neutrophil-specific proteinases in the release of sLMIR5 was suggested, further examination is required to completely understand the relevant mechanism.

Importantly, the pathophysiological role of sLMIR5 in LPS-induced sepsis was clarified by a number experiments: 1) LMIR5−/− mice were more resistant to LPS- or peritonitis-induced septic death; 2) LMIR5−/− mice exhibited lower levels of proinflammatory cytokine production in response to LPS; 3) systemic and local sLMIR5 quickly increased only in WT mice, but not LMIR5−/− mice, in response to LPS administration or peritonitis induction; 4) in vitro stimulation with LMIR5-Fc, a mimic form of sLMIR5, or recombinant LMIR5 harboring only an extracellular domain of LMIR5 caused cytokine production of peritoneal Mϕ; 5) in vivo administration of LMIR5-Fc bound to peritoneal Mϕ and induced significant, but not lethal, inflammation; and 6) the resistance to LPS- or peritonitis-induced septic death of LMIR5−/− mice was reduced by in vivo administration of LMIR5-Fc. Although neither LMIR5-Fc nor recombinant soluble LMIR5 is completely equivalent to sLMIR5 (15), these results indicated that sLMIR5 is a positive regulator of LPS-induced lethal inflammation. In contrast, because the mechanism of pathophysiological sepsis is much more complex than that of LPS-induced systemic inflammation, it is possible that sLMIR5 plays a limited role in pathophysiological sepsis. A possible mechanism of sLMIR5-mediated, LPS-induced lethal inflammation may be that LPS-activated neutrophils released sLMIR5 that stimulated cytokine production of resident Mϕ, resulting in the lethal inflammation. Given that the release of sLMIR5 was enhanced by not only LPS but also other TLR agonists, it is important to examine whether sLMIR5 plays a role in viral or fungal infection models.

Stronger binding of LMIR5-Fc to peritoneal Mϕ led us to postulate that the specific ligand for sLMIR5 was highly expressed in peritoneal Mϕ. Although we have recently identified TIM1 and TIM4 as endogenous ligands for LMIR5 (15), it was unlikely that these surface molecules were involved in LMIR5-Fc–induced cytokine production in peritoneal Mϕ. First, peritoneal Mϕ did not express TIM1 at high levels (15, 35). Second, peritoneal Mϕ expressed TIM4 but were bound at weak but significant levels by LMIR5 del4-Fc that did not bind to TIM4 (15). Third, LMIR5 del4-Fc as compared with LMIR5-Fc stimulated comparable levels of cytokine production in peritoneal Mϕ. It is therefore possible that sLMIR5 binds to and activates an unknown surface molecule other than TIM1/4, resulting in cytokine production of peritoneal Mϕ. To identify the unknown ligand, several attempts, including retrovirus-mediated expression cloning, are now under way.

The comparison between sLMIR5 and sTREM-1 is interesting in light of the Ig-like receptor family’s involvement in the pathogenesis of sepsis. Stimulation with TLR agonists augmented the release of both sLMIR5 and sTREM-1, although it increased transcript levels of TREM-1, but not LMIR5 (22–26). The mechanism of sTREM-1 generation in neutrophils and monocytes was controversial (22–26), but a recent report has shown that MMPs shed their TREM-1 ectodomain from LPS-stimulated human monocytes (36). Whereas sTREM-1 attenuates excessive inflammatory response by counterregulating TREM-1, an inflammatory amplifier in sepsis (22–26), sLMIR5 amplifies LPS-induced lethal inflammation. Thus, sLMIR5 or sTREM-1 is a positive or negative feedback regulator of LPS-induced inflammation, respectively. Considering that LMIR5 is a DAP12-coupled activating receptor and that DAP12 deficiency downregulates surface expression of LMIR5 (13), the resistance of DAP12−/− mice to LPS-induced lethality might be, in part, because of the lack of sLMIR5 (37). In any case, it is conceivable that cross talk between TLRs and paired immune receptors modulates innate immune responses in sepsis.

In conclusion, sLMIR5 is released by neutrophils in response to LPS, and amplifies lethal inflammation in LPS-induced sepsis. Our results provide a mechanistic insight into septic death and will help develop the novel therapies to treat sepsis.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank Dr. Hisashi Arase for providing plasmids. We also thank Drs. Hisaya Akiba and Ko Okumura for providing Abs. We are grateful to Dr. Dovie Wylie for excellent language assistance.

Footnotes

  • This work was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology, Japan; grants from the Global Center of Excellence program “Center of Education and Research for the Advanced Genome-Based Medicine–For personalized medicine and the control of worldwide infectious diseases,” Ministry of Education, Culture, Sports, Science and Technology, Japan; and a postdoctoral fellowship from the Japan Society for the Promotion of Science.

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    BM
    bone marrow
    BMMC
    BM-derived mast cell
    CASP
    colon ascendens stent peritonitis
    DAP12
    DNAX-activating protein of 12 kDa
    LMIR
    leukocyte mono-Ig–like receptor
    Mϕ
    macrophage
    MFI
    mean fluorescent intensity
    MMP
    matrix metalloproteinase
    ODN
    oligodeoxynucleotide
    PLF
    peritoneal lavage fluid
    PRR
    pattern-recognition receptor
    sLMIR5
    soluble form of LMIR5
    sTREM-1
    soluble form of triggering receptors expressed on myeloid cells-1
    TIM
    T cell Ig mucin
    TREM-1
    triggering receptors expressed on myeloid cells-1
    WT
    wild type.

  • Received April 18, 2012.
  • Accepted June 11, 2012.

References

  1. ↵
    1. Cohen J.
    2002. The immunopathogenesis of sepsis. Nature 420: 885–891.
    OpenUrlCrossRefPubMed
    1. Riedemann N. C.,
    2. R. F. Guo,
    3. P. A. Ward
    . 2003. Novel strategies for the treatment of sepsis. Nat. Med. 9: 517–524.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Buras J. A.,
    2. B. Holzmann,
    3. M. Sitkovsky
    . 2005. Animal models of sepsis: setting the stage. Nat. Rev. Drug Discov. 4: 854–865.
    OpenUrlCrossRefPubMed
    1. Hotchkiss R. S.,
    2. D. W. Nicholson
    . 2006. Apoptosis and caspases regulate death and inflammation in sepsis. Nat. Rev. Immunol. 6: 813–822.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Rittirsch D.,
    2. M. A. Flierl,
    3. P. A. Ward
    . 2008. Harmful molecular mechanisms in sepsis. Nat. Rev. Immunol. 8: 776–787.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Palm N. W.,
    2. R. Medzhitov
    . 2009. Pattern recognition receptors and control of adaptive immunity. Immunol. Rev. 227: 221–233.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Aderem A.,
    2. R. J. Ulevitch
    . 2000. Toll-like receptors in the induction of the innate immune response. Nature 406: 782–787.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Beutler B.
    2004. Inferences, questions and possibilities in Toll-like receptor signalling. Nature 430: 257–263.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Ivashkiv L. B.
    2009. Cross-regulation of signaling by ITAM-associated receptors. Nat. Immunol. 10: 340–347.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Blank U.,
    2. P. Launay,
    3. M. Benhamou,
    4. R. C. Monteiro
    . 2009. Inhibitory ITAMs as novel regulators of immunity. Immunol. Rev. 232: 59–71.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Kumagai H.,
    2. T. Oki,
    3. K. Tamitsu,
    4. S. Z. Feng,
    5. M. Ono,
    6. H. Nakajima,
    7. Y. C. Bao,
    8. Y. Kawakami,
    9. K. Nagayoshi,
    10. N. G. Copeland,
    11. et al
    . 2003. Identification and characterization of a new pair of immunoglobulin-like receptors LMIR1 and 2 derived from murine bone marrow-derived mast cells. Biochem. Biophys. Res. Commun. 307: 719–729.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Izawa K.,
    2. J. Kitaura,
    3. Y. Yamanishi,
    4. T. Matsuoka,
    5. T. Oki,
    6. F. Shibata,
    7. H. Kumagai,
    8. H. Nakajima,
    9. M. Maeda-Yamamoto,
    10. J. P. Hauchins,
    11. et al
    . 2007. Functional analysis of activating receptor LMIR4 as a counterpart of inhibitory receptor LMIR3. J. Biol. Chem. 282: 17997–18008.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Yamanishi Y.,
    2. J. Kitaura,
    3. K. Izawa,
    4. T. Matsuoka,
    5. T. Oki,
    6. Y. Lu,
    7. F. Shibata,
    8. S. Yamazaki,
    9. H. Kumagai,
    10. H. Nakajima,
    11. et al
    . 2008. Analysis of mouse LMIR5/CLM-7 as an activating receptor: differential regulation of LMIR5/CLM-7 in mouse versus human cells. Blood 111: 688–698.
    OpenUrlAbstract/FREE Full Text
    1. Izawa K.,
    2. J. Kitaura,
    3. Y. Yamanishi,
    4. T. Matsuoka,
    5. A. Kaitani,
    6. M. Sugiuchi,
    7. M. Takahashi,
    8. A. Maehara,
    9. Y. Enomoto,
    10. T. Oki,
    11. et al
    . 2009. An activating and inhibitory signal from an inhibitory receptor LMIR3/CLM-1: LMIR3 augments lipopolysaccharide response through association with FcRgamma in mast cells. J. Immunol. 183: 925–936.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. Yamanishi Y.,
    2. J. Kitaura,
    3. K. Izawa,
    4. A. Kaitani,
    5. Y. Komeno,
    6. M. Nakamura,
    7. S. Yamazaki,
    8. Y. Enomoto,
    9. T. Oki,
    10. H. Akiba,
    11. et al
    . 2010. TIM1 is an endogenous ligand for LMIR5/CD300b: LMIR5 deficiency ameliorates mouse kidney ischemia/reperfusion injury. J. Exp. Med. 207: 1501–1511.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Ravetch J. V.,
    2. L. L. Lanier
    . 2000. Immune inhibitory receptors. Science 290: 84–89.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Lanier L. L.
    2009. DAP10- and DAP12-associated receptors in innate immunity. Immunol. Rev. 227: 150–160.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Chung D. H.,
    2. M. B. Humphrey,
    3. M. C. Nakamura,
    4. D. G. Ginzinger,
    5. W. E. Seaman,
    6. M. R. Daws
    . 2003. CMRF-35-like molecule-1, a novel mouse myeloid receptor, can inhibit osteoclast formation. J. Immunol. 171: 6541–6548.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Yotsumoto K.,
    2. Y. Okoshi,
    3. K. Shibuya,
    4. S. Yamazaki,
    5. S. Tahara-Hanaoka,
    6. S. Honda,
    7. M. Osawa,
    8. A. Kuroiwa,
    9. Y. Matsuda,
    10. D. G. Tenen,
    11. et al
    . 2003. Paired activating and inhibitory immunoglobulin-like receptors, MAIR-I and MAIR-II, regulate mast cell and macrophage activation. J. Exp. Med. 198: 223–233.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Kuchroo V. K.,
    2. D. T. Umetsu,
    3. R. H. DeKruyff,
    4. G. J. Freeman
    . 2003. The TIM gene family: emerging roles in immunity and disease. Nat. Rev. Immunol. 3: 454–462.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Ichimura T.,
    2. J. V. Bonventre,
    3. V. Bailly,
    4. H. Wei,
    5. C. A. Hession,
    6. R. L. Cate,
    7. M. Sanicola
    . 1998. Kidney injury molecule-1 (KIM-1), a putative epithelial cell adhesion molecule containing a novel immunoglobulin domain, is up-regulated in renal cells after injury. J. Biol. Chem. 273: 4135–4142.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Bouchon A.,
    2. F. Facchetti,
    3. M. A. Weigand,
    4. M. Colonna
    . 2001. TREM-1 amplifies inflammation and is a crucial mediator of septic shock. Nature 410: 1103–1107.
    OpenUrlCrossRefPubMed
    1. Klesney-Tait J.,
    2. I. R. Turnbull,
    3. M. Colonna
    . 2006. The TREM receptor family and signal integration. Nat. Immunol. 7: 1266–1273.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Ford J. W.,
    2. D. W. McVicar
    . 2009. TREM and TREM-like receptors in inflammation and disease. Curr. Opin. Immunol. 21: 38–46.
    OpenUrlCrossRefPubMed
    1. Gibot S.,
    2. M. N. Kolopp-Sarda,
    3. M. C. Béné,
    4. P. E. Bollaert,
    5. A. Lozniewski,
    6. F. Mory,
    7. B. Levy,
    8. G. C. Faure
    . 2004. A soluble form of the triggering receptor expressed on myeloid cells-1 modulates the inflammatory response in murine sepsis. J. Exp. Med. 200: 1419–1426.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Knapp S.,
    2. S. Gibot,
    3. A. de Vos,
    4. H. H. Versteeg,
    5. M. Colonna,
    6. T. van der Poll
    . 2004. Cutting edge: expression patterns of surface and soluble triggering receptor expressed on myeloid cells-1 in human endotoxemia. J. Immunol. 173: 7131–7134.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Kitaura J.,
    2. J. Song,
    3. M. Tsai,
    4. K. Asai,
    5. M. Maeda-Yamamoto,
    6. A. Mocsai,
    7. Y. Kawakami,
    8. F. T. Liu,
    9. C. A. Lowell,
    10. B. G. Barisas,
    11. et al
    . 2003. Evidence that IgE molecules mediate a spectrum of effects on mast cell survival and activation via aggregation of the FcepsilonRI. Proc. Natl. Acad. Sci. USA 100: 12911–12916.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Mócsai A.,
    2. E. Ligeti,
    3. C. A. Lowell,
    4. G. Berton
    . 1999. Adhesion-dependent degranulation of neutrophils requires the Src family kinases Fgr and Hck. J. Immunol. 162: 1120–1126.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Shiratori I.,
    2. K. Ogasawara,
    3. T. Saito,
    4. L. L. Lanier,
    5. H. Arase
    . 2004. Activation of natural killer cells and dendritic cells upon recognition of a novel CD99-like ligand by paired immunoglobulin-like type 2 receptor. J. Exp. Med. 199: 525–533.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Kitamura T.,
    2. Y. Koshino,
    3. F. Shibata,
    4. T. Oki,
    5. H. Nakajima,
    6. T. Nosaka,
    7. H. Kumagai
    . 2003. Retrovirus-mediated gene transfer and expression cloning: powerful tools in functional genomics. Exp. Hematol. 31: 1007–1014.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Morita S.,
    2. T. Kojima,
    3. T. Kitamura
    . 2000. Plat-E: an efficient and stable system for transient packaging of retroviruses. Gene Ther. 7: 1063–1066.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Traeger T.,
    2. W. Kessler,
    3. V. Assfalg,
    4. K. Cziupka,
    5. P. Koerner,
    6. C. Dassow,
    7. K. Breitbach,
    8. M. Mikulcak,
    9. I. Steinmetz,
    10. K. Pfeffer,
    11. et al
    . 2008. Detrimental role of CC chemokine receptor 4 in murine polymicrobial sepsis. Infect. Immun. 76: 5285–5293.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Renckens R.,
    2. J. J. Roelofs,
    3. S. Florquin,
    4. A. F. de Vos,
    5. H. R. Lijnen,
    6. C. van’t Veer,
    7. T. van der Poll
    . 2006. Matrix metalloproteinase-9 deficiency impairs host defense against abdominal sepsis. J. Immunol. 176: 3735–3741.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Tester A. M.,
    2. J. H. Cox,
    3. A. R. Connor,
    4. A. E. Starr,
    5. R. A. Dean,
    6. X. S. Puente,
    7. C. López-Otín,
    8. C. M. Overall
    . 2007. LPS responsiveness and neutrophil chemotaxis in vivo require PMN MMP-8 activity. PLoS ONE 2: e312.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Rodriguez-Manzanet R.,
    2. R. DeKruyff,
    3. V. K. Kuchroo,
    4. D. T. Umetsu
    . 2009. The costimulatory role of TIM molecules. Immunol. Rev. 229: 259–270.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Gómez-Piña V.,
    2. A. Soares-Schanoski,
    3. A. Rodríguez-Rojas,
    4. C. Del Fresno,
    5. F. García,
    6. M. T. Vallejo-Cremades,
    7. I. Fernández-Ruiz,
    8. F. Arnalich,
    9. P. Fuentes-Prior,
    10. E. López-Collazo
    . 2007. Metalloproteinases shed TREM-1 ectodomain from lipopolysaccharide-stimulated human monocytes. J. Immunol. 179: 4065–4073.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Turnbull I. R.,
    2. J. E. McDunn,
    3. T. Takai,
    4. R. R. Townsend,
    5. J. P. Cobb,
    6. M. Colonna
    . 2005. DAP12 (KARAP) amplifies inflammation and increases mortality from endotoxemia and septic peritonitis. J. Exp. Med. 202: 363–369.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section