Immunoreceptor CD300a on mast cells and dendritic cells regulates neutrophil recruitment in a murine model of sepsis

Abstract

Sepsis is a life-threatening syndrome caused by abnormal host immune responses against bacterial infection. Although innate immune cells are known to be important in the pathogenesis of sepsis, how their activation is regulated during sepsis remains incompletely understood. Here, we examined the role of the inhibitory immunoreceptor CD300a, which is expressed on various types of myeloid cells, in the pathogenesis of sepsis induced by cecal ligation and puncture (CLP). To this end, we used mice in which CD300a was specifically deleted on mast cells (MCs; Cd300a^fl/flMcpt5-Cre), dendritic cells (DCs; Cd300a^fl/flItgax-Cre), or macrophages and neutrophils (Cd300a^fl/flLyz2-Cre). We show that mice with CD300a-deleted MCs or DCs but not macrophages survived significantly longer than did control Cd300a^fl/fl mice. In addition, whereas neutrophil recruitment into the peritoneal cavity was increased within 1 h after CLP in mice with CD300a-deleted MCs, peritoneal neutrophils did not increase in number until the 12 h time point in mice with CD300a-deficient DCs. These results indicate that CD300a on MCs and DCs regulates neutrophil recruitment into the peritoneal cavity after CLP.

CD300a, cecal ligation and puncture, dendritic cells, mast cells, sepsis

Introduction

Sepsis, a life-threatening syndrome caused by abnormal host immune responses against infection, is the most frequent cause of death in hospitalized patients (1). Cecal ligation and puncture (CLP), a well-known murine model of sepsis (2), prompts an overwhelming inflammatory reaction driven by innate mast cells (MCs), dendritic cells (DCs), macrophages and neutrophils with subsequent septic peritonitis. However, these innate immune cells also protect against sepsis during early stages of infection. MCs trigger the innate immune response by producing chemoattractants for neutrophils, such as TNF-α, which are crucial to bacterial clearance in the peritoneal cavity after CLP (3–6). However, how neutrophil recruitment is regulated in the CLP model remains incompletely understood.

The immunoreceptor CD300a, containing a consensus immunoreceptor tyrosine-based inhibitory motif (ITIM) in the cytoplasmic region, suppresses FcεRI-mediated signalling and the degranulation of human and mouse MCs in vitro (7–9). In addition, CD300a inhibits signalling through TLR4 and suppresses cytokine and chemokine production from MCs and macrophages in vitro (3). Compared with wild-type (WT) mice, Cd300a^−/− mice showed enhanced bacterial clearance from the peritoneal cavity and prolonged survival after CLP (3). Similarly, bacterial clearance and survival after CLP were improved in MC-deficient Kit^W-sh/W-sh mice intraperitoneally transferred with Cd300a^−/− bone marrow-derived cultured MCs (BMMCs) compared with WT BMMCs (3), suggesting that CD300a on MCs is involved in bacterial clearance and survival. However, given that Kit^W-sh/W-sh mice show additional non-MC hematopoietic abnormalities, including neutrophil accumulation in several tissues, as well as various non-hematopoietic abnormalities (10–12), whether CD300a on MCs directly suppresses neutrophil recruitment and influences survival after CLP is unclear.

To address this issue, we generated mice that were specifically deficient in CD300a on MCs, DCs, or macrophages and neutrophils. We show that CD300a deficiency on both MCs and DCs enhanced neutrophil recruitment and increased bacterial clearance in the peritoneal cavity and improved survival after CLP. We also found that the effects due to CD300a on DCs occur after those due to the same immunoreceptor on MCs. Together, our findings imply that CD300a on MCs and DCs suppresses neutrophil recruitment in the peritoneal cavity after CLP.

Methods

Mice

C57BL/6J mice were purchased from Clea Japan (Tokyo, Japan). Cd300a^−/− mice were generated in our laboratory, as previously described (3). Mcpt5-Cre mice were obtained from Dr A. Roers (University of Technology, Dresden, Germany). Lyz2-Cre and Itgax-Cre mice were purchased from Jackson Laboratory (Bar Harbor, ME, USA). To delete Cd300a specifically on MCs, macrophage and neutrophils, or DCs, we mated Cd300a^fl/fl mice with Mcpt5-Cre, Lyz2-Cre or Itgax-Cre mice, which express the Cre enzyme under the control of the MC protease promoter (11), the endogenous Lyz2 promoter (13) or integrin α-X (CD11c) (14), respectively, thus generating Cd300a^fl/flMcpt5-Cre, Cd300a^fl/flLyz2-Cre and Cd300a^fl/flItgax-Cre mice. Genomic DNA isolated from tail tissues was used as the template for PCR genotyping of the loxP-flanked CD300a allele (primers: forward, 5′-GTACACAGTCACTGTCTTCAGACACA-3′; reverse, 5′-CCTTCAAGAAGGCCACACCTCCTGAT-3′); the PCR products from WT and Cd300a^fl/fl mice were 287 and 355 bp in length, respectively. All animal experiments were performed in accordance with the regulations and the guidelines of the animal ethics committee of the University of Tsukuba Animal Research Center.

CLP procedure

CLP was performed as previously described (3, 4). Briefly, the cecum was exposed through a 1-cm midline incision on the mouse’s ventral abdomen; the distal 1.5 cm of the cecum was ligated and then punctured by using a 21-gauge needle. The cecum was replaced into the abdomen, 0.5 ml of sterile saline was injected into abdominal cavity, and the abdominal incision was sutured closed.

Antibodies and flow cytometry

Anti-mouse CD45.2 (clone 104), CD11b (M1/70), CD11c (HL3), c-kit (2B8), Ly6C (AL-21), Ly6G (1A8), I-Ad (M5/114.15.2), TNF-α and IL-6 mAbs were purchased from BD Pharmingen (San Diego, CA, USA). Anti-F4/80 (MCA497) mAb was purchased from SeroTec (Raleigh, NC, USA), and anti-FcεR1α (MAR-1) mAb was obtained from eBioscience (San Diego, CA, USA). Anti-CD300a mAb (TX41) was developed in our laboratory as described previously (8). All samples were evaluated by using flow cytometry (Fortessa, Becton Dickinson, Franklin Lakes, NJ, USA), and data were analysed by using FlowJo software (Ashland, OR, USA).

Analysis of intracellular cytokines

Cells in the peritoneal lavage fluid were harvested 1 and 12 h after CLP and cultured at 2 × 10⁵ cells per 1 ml of 10% fetal bovine serum containing DMEM with brefeldin A (10 μg ml⁻¹; Sigma, St Louis, MO, USA) for 3 h. Intracellular cytokines were stained by using Fix and Perm kits (Invitrogen, Carlsbad, CA, USA) and analysed by flow cytometry.

Measurement of colony-forming units

Bacterial colony-forming units (CFUs) were quantified by plating serial dilutions of peritoneal lavage fluid harvested after CLP on brain–heart infusion agar plates and culturing them at 37°C for 24–48 h; data are reported as the number of CFU ml⁻¹, as described previously (15).

Statistical analysis

The Kaplan–Meier log-rank test was used to analyse survival. All other statistical analyses were performed by using unpaired Student’s t-tests. Differences were considered statistically significant at P < 0.05 by using Graph Pad Prism software (San Diego, CA, USA).

Results and discussion

To examine the role of CD300a on MCs in the pathogenesis of sepsis after CLP, we created conditional CD300a-deficient mice by using the Cre-loxP system. Transgenic mice harbouring a floxed Cd300a gene, in which the exon-1-containing region was flanked by two loxP sites (Fig. 1A), were mated with mice carrying Cre recombinase in their MCs (Mcpt5-Cre). We similarly crossed Cd300a^fl/fl mice with mice harbouring Cre recombinase in DCs (Itgax-Cre) or in macrophages and neutrophils (Lyz2-Cre). Both heterozygous and homozygous progeny mice were viable and fertile, without phenotypic differences between littermates on a mixed genetic background, and correct targeting, homologous recombination and germline transmission were confirmed (Fig. 1B).

Fig. 1.

Generation of Cd300a^fl/fl mice and targeted Cre-loxP-mediated conditional CD300a-deficient mice. (A) This diagram depicts the strategy for generating the Cd300a^fl/fl mouse line. (B) PCR analysis of genomic DNA extracted from the tail tissue of Cd300a^fl/fl mice. (C) CD300a expression on the indicated cells was analysed by means of flow cytometry. Data are representative of at least three independent experiments.

We then used flow cytometry to analyse the CD300a expression on several cell types. As expected, Cd300a^fl/flMcpt5-Cre mice, Cd300a^fl/flItgax-Cre and Cd300a^fl/flLyz2-Cre mice lacked CD300a expression specifically on MCs from the peritoneal cavity, DCs from the spleen, and macrophages and neutrophils from the spleen, respectively (Fig. 1C). These results confirm that conditional CD300a-deficient mice were generated. In addition, we confirmed that the cellular populations in the spleen and peritoneal cavity were comparable among all genotypes before CLP (data not shown).

To confirm that CD300a on MCs is primarily responsible for survival during sepsis, we performed CLP in Cd300a^fl/flMcpt5-Cre mice. We observed that these mice survived significantly longer than did Cd300a^fl/fl (control) mice (Fig. 2A), whereas survival rates were comparable between Cd300a^fl/flLyz2-Cre and control mice (Fig. 2B). In addition, survival after CLP was unexpectedly longer in Cd300a^fl/flItgax-Cre mice than in control animals (Fig. 2C). These results indicate that CD300a on DCs as well as MCs is involved in the pathogenesis of CLP.

Fig. 2.

CD300a on MCs and DCs influences survival after CLP. (A–C) Mice of the indicated lines underwent CLP, after which their survival was observed. In all experiments, Cd300a^−/− mice served as the positive control group (n = 10 in each group). Survival rate was compared between (A) Cd300a^fl/fl and Cd300a^fl/flMcpt5-Cre mice (n = 10 in each group), (B) Cd300a^fl/fl and Cd300a^fl/flLyz2-Cre mice (n = 10 in each group) and (C) Cd300a^fl/fl and Cd300a^fl/flItgax-Cre mice (n = 10 in each group). All the data were pooled from two independent experiments. *P < 0.05 (Kaplan–Meier log-rank test).

To elucidate the mechanism underlying the enhanced survival of Cd300a^fl/flMcpt5-Cre and Cd300a^fl/flItgax-Cre mice, we analysed the peritoneal bacterial content after CLP. At 4 h after CLP, Cd300a^fl/flMcpt5-Cre mice had significantly fewer CFUs than did control Cd300a^fl/fl mice, whereas bacterial counts were comparable between Cd300a^fl/flItgax-Cre and control Cd300a^fl/fl mice at this time point (Fig. 3A). However, at 24 h after CLP, both Cd300a^fl/flMcpt5-Cre and Cd300a^fl/flItgax-Cre mice had lower CFU counts than did control mice (Fig. 3B). These results indicate that bacterial clearance in Cd300a^fl/flMcpt5-Cre and Cd300a^fl/flItgax-Cre mice was more efficient than that in control mice at early and later time points after CLP, respectively. We therefore examined the number of neutrophils recruited into the peritoneal cavity at 1 and 12 h after CLP. As we expected, the number of peritoneal neutrophils at 1 h after CLP was significantly greater in Cd300a^fl/flMcpt5-Cre mice than in control mice (Fig. 3C). In contrast, neutrophil recruitment into the peritoneal cavity at 12 h after CLP was significantly higher in Cd300a^fl/flItgax-Cre mice compared with control mice (Fig. 3D).

Fig. 3.

CD300a on MCs and DCs regulates bacterial clearance from and the recruitment of neutrophils into the peritoneal cavity. (A–D) The indicated conditional CD300a-deficient mice underwent CLP. Peritoneal lavage fluid was harvested, and the numbers of CFUs at 4 h (A) and 24 h (B) after CLP and of neutrophils at 1 h (C) and 12 h (D) were determined (n = 4–6 in each group). (E–G) Flow cytometric analysis of CD300a expression on CD11b⁺Ly6C⁺CD11c⁺ DCs (E), the numbers of CD11b⁺Ly6C⁺CD11c⁺ DCs and CD11b⁺F4/80^high macrophages (F) and intracellular cytokine production of CD11b⁺Ly6C⁺CD11c⁺ DCs at 12 h after CLP (G) in the peritoneal cavity (n = 4 in each group). Data are representative of three (A–D and G) or two (E and F) experiments. *P < 0.05 and **P < 0.01 (Student’s t-test).

In the peritoneal cavity, CD300a was expressed on CD11b⁺Ly6C⁺CD11c⁺ DCs (Fig. 3E), which correspond to inflammatory DCs derived from inflammatory monocytes (16). The migration of inflammatory DCs from the peripheral blood to the peritoneal cavity started at 4 h after CLP. In contrast, we observed that CD11b⁺ F4/80^high resident macrophages rapidly disappeared from the peritoneal cavity after CLP (Fig. 3F), which is called the macrophage disappearance response (MDR) from the peritoneal cavity (17, 18). We previously demonstrated that LPS-stimulated Cd300a^−/− MCs produced greater amounts of neutrophil chemoattractants, such as MCP-1, IL-13 and TNF-α, than did WT MCs (3). Similarly, we observed that the amounts of TNF-α and IL-6 produced by peritoneal CD11b⁺Ly6C⁺CD11c⁺ DCs at 12 h after CLP was significantly greater in Cd300a^fl/flItgax-Cre mice than in control mice (Fig. 3G). These results suggest that CD300a regulates the production of neutrophil chemoattractants from DCs as well as MCs in the peritoneal cavity after CLP.

We showed here that CD300a on MCs and DCs regulate the neutrophil recruitment into the peritoneal cavity and survival of mice after CLP. Although residential MCs in the peritoneal cavity trigger the immunity against microbial invasion by CLP (4, 5), inflammatory DCs derived from the circulating CD11b⁺Ly6C⁺ monocytes are also required for the reduction of bacteria after CLP (19–21). The gradual recruitment of inflammatory DCs after CLP may result in the delayed neutrophil accumulation. In contrast, we showed an MDR after CLP; this may be the reason why CD300a on the macrophages was not involved in the neutrophil recruitment into the peritoneal cavity after CLP.

We previously showed that apoptotic cells are abundantly generated in the peritoneal cavity after CLP (3). Although the origin of the apoptotic cells after CLP remains unclear, it is likely that phosphatidylserine (PS) on the apoptotic cells is the ligand for CD300a on MCs and inflammatory DCs (3, 22). Indeed, we also showed that CD300a on inflammatory DCs directly bound to PS on apoptotic cells that had been generated by injection of aluminium salt adjuvant (16). Our results suggest that the blockade of the interaction between CD300a on MCs and inflammatory DCs and apoptotic cells in the peritoneal cavity is useful for controlling MCs and inflammatory DCs to recruit the neutrophils during polymicrobial infections.

Funding

Japan Society for the Promotion of Science (KAKENHI) (15H01365, 16H06387; to A.S. and 16H05350; to C.N-O.).

Acknowledgement

We thank S. Tochihara and Y. Nomura for secretarial assistance, and F. Abe and R. Hirochika for technical assistance.

Conflict of interest statement: The authors have no competing financial interests.

Author notes

Correspondence to: C. Nakahashi-Oda; E-mail:

*These authors contributed equally to this work.