Extracellular CIRP decreases Siglec-G expression on B-1a cells skewing them towards a pro-inflammatory phenotype in sepsis

Abstract

Background: Sepsis is a life-threatening disease syndrome caused by a dysregulated host response to infection and injury. Extracellular cold-inducible RNA-binding protein (eCIRP) acts as a damage-associated molecular pattern. Peritoneal cavity (PerC) B-1a cells attenuate inflammation and tissue injury by spontaneous releasing natural IgM and IL-10. Sialic acid-binding immunoglobulin-type lectin-G (Siglec-G) is a CD33-related receptor highly expressed in B-1a cells to serve critical immunoregulatory functions. In sepsis, B-1a cell numbers in PerC are decreased. We hypothesized that eCIRP causes the reduction of PerC B-1a cells and alters their function during sepsis.

Methods: Sepsis was induced in WT and CIRP^-/- mice by cecal ligation and puncture (CLP). PerC washout cells were collected and B-1a cells and Siglec-G were assessed by flow cytometry. Mice were i.p. injected with recombinant murine (rm) CIRP and after 20 h, Siglec-G expression in PerC B-1a cells were assessed. PerC B-1a cells were treated with rmCIRP for 4 h and Siglec-G expression was assessed. PerC B-1a cells were pre-treated with anti-Siglec-G Ab and then after stimulated with rmCIRP for 24 h, IL-6 levels in the culture supernatants were assessed.

Results: eCIRP levels in the PerC were elevated in septic mice. In WT mice, the frequencies and numbers of total and Siglec-G⁺ B-1a cells in the PerC were significantly decreased in the CLP group compared to sham group, whereas in CIRP^-/- mice, their frequencies and numbers in sepsis were significantly rescued compared to WT septic mice. Mice injected with rmCIRP showed decreased frequencies and numbers of total and Siglec-G⁺ PerC B-1a cells compared to PBS-injected mice. In vitro treatment of PerC B-1a cells with rmCIRP demonstrated significant reduction in Siglec-G mRNA and protein compared to PBS group. PerC B-1a cells treated with anti-Siglec-G Ab had significantly higher production of IL-6 in response to rmCIRP compared to IgG control. Anti-Siglec-G Ab treated B-1a cells co-cultured with macrophages produced significantly higher levels of IL-6, and TNF-α, and lower levels of IL-10 compared to IgG-treated B-1a cells and macrophage co-cultures stimulated with rmCIRP.

Conclusion: eCIRP reduces PerC B-1a cell pool and skews them to a pro-inflammatory phenotype by downregulating Siglec-G expression. Targeting eCIRP will retain Siglec-G expressing B-1a cells in the PerC and preserve their anti-inflammatory function in sepsis.

Keywords: B-1a cells; Sepsis; Siglec-G; TLR4; eCIRP.

Fig. 1

Intraabdominal sepsis results in a loss of PerC B-1a cells, while CIRP^−/− mice are protected from the loss of B-1a cells after sepsis. A After 20 h of CLP, peritoneal cavity lavage samples were collected. PerC lavage eCIRP concentration was assessed with ELISA. Data are expressed as means ± SE (n = 5–7 mice/group). The groups were compared by Student’s t-test (*p < 0.05 vs. sham mice). WT and CIRP^−/− mice were randomly assigned to sham or CLP groups. 20 h after surgery, PerC lavage samples were collected for analysis. B Representative dot blots of the gating strategy of B-1a cells of sham and CLP mice are shown. C, D Status of B-1a cell’s frequencies (%) and numbers after surgery are shown. Data are expressed as means ± SE (n = 10–14 mice/group). The groups were compared by one-way ANOVA and Student–Newman–Keuls (SNK) method (*p < 0.05 vs. WT sham and ^#p < 0.05 vs. WT CLP mice)

Fig. 2

Expression of Siglec-G by B-1a cells is dramatically reduced after sepsis, but its expression is protected in CIRP^−/− mice after sepsis. WT and CIRP^−/− mice were randomly assigned to sham or CLP groups. 20 h after CLP, peritoneal lavage samples were collected for analysis. A Representative dot blots of the gating strategy of Siglec-G expression on B-1a cells of sham and CLP mice are shown. B, C Status of Siglec-G expression by B-1a cells in terms of frequency (%) and number after surgery are shown. Data are expressed as means ± SE (n = 10–14 mice/group). The groups were compared by one-way ANOVA and Student–Newman–Keuls (SNK) method (*p < 0.05 vs. WT sham and ^#p < 0.05 vs. WT CLP mice)

Fig. 3

Intraperitoneal injection of rmCIRP results in a loss of B-1a cells from the PerC. A After 20 h of i.p. injection of rmCIRP in mice, PerC lavage samples were collected. Peritoneal cells were stained with CD23, B220, and CD5 Abs and subjected to flow cytometric detection of B-1a cells frequencies. Representative dot blots of the gating strategy of B-1a cells of vehicle and rmCIRP treated mice are shown. B, C Status of B-1a cell’s frequencies and numbers after i.p. injection of PBS vehicle or rmCIRP are shown. Data are expressed as means ± SE (n = 5–7 mice/group). The groups were compared by Student’s t-test (*p < 0.05 vs. vehicle-treated mice)

Fig. 4

rmCIRP induces a downregulation of Siglec-G expression by B-1a cells in vivo and in vitro. 20 h after i.p. injection of rmCIRP, PerC lavage samples were collected. Peritoneal cells were stained with CD23, B220, CD5, and Siglec-G Abs and subjected to flow cytometric detection of B-1a cells and analysis of Siglec-G expression. Status of Siglec-G expression by B-1a cells in terms of A frequency and B number after i.p. injection of rmCIRP are shown. PerC B-1a cells were stimulated with rmCIRP in vitro for 24 h. Cells were stained with Siglec-G Ab and subjected to flow cytometric analysis. C Frequency of Siglec-G expression on B-1a cells after stimulation with rmCIRP are shown. Data are expressed as means ± SE (n = 6–7 mice/group). The groups were compared by Student’s t-test (*p < 0.05 vs. vehicle treatment)

Fig. 5

An absence of Siglec-G results in greater production of IL-6 by B-1a cells in response to rmCIRP than WT cells. A PerC B-1a cells were collected from WT and Siglec-G^−/− mice and stimulated in vitro with rmCIRP or PBS vehicle for 24 h. Supernatant was assessed by ELISA for IL-6 production. Data are expressed as means ± SE (n = 4–8 mice/group) and compared by one-way ANOVA and Student–Newman–Keuls (SNK) method (*p < 0.05 vs. WT PBS treated mice and #p < 0.05 vs. WT rmCIRP treated mice). B Blocking Siglec-G by its neutralizing Abs increases IL-6 production following rmCIRP treatment. PerC B-1a cells were collected from WT mice and pretreated with Siglec-G blocking Ab or IgG control. The cells were stimulated in vitro with rmCIRP or PBS vehicle for 24 h. Supernatant was assessed by ELISA for IL-6 production. Data are expressed as means ± SE (n = 8 mice/group) and compared by one-way ANOVA and Student–Newman–Keuls (SNK) method (*p < 0.05 vs. PBS IgG treated mice and ^#p < 0.05 vs. rmCIRP IgG treated mice)

Fig. 6

Neutralizing Siglec-G in B-1a cells causes more TNF-α, and IL-6 and less IL-10 production in rmCIRP-treated macrophage co-cultures. PerC macrophages were co-cultured with B-1a cells that had been pretreated with Siglec-G blocking Ab or IgG control and then stimulated with rmCIRP or PBS vehicle for 24 h. Supernatant A TNF-α, B IL-6, and C IL-10 concentration were assessed with ELISA. Data are expressed as means ± SE (n = 8 mice/group) and compared by one-way ANOVA and Student–Newman–Keuls (SNK) method (*p < 0.05 vs. PBS stimulated macrophages, ^#p < 0.05 vs. macrophages cultured with IgG treated B-1a cells stimulated with rmCIRP, ^†p < 0.05 vs. untreated B-1a cells stimulated with rmCIRP)