Soluble P-selectin rescues mice from anthrax lethal toxin-induced mortality through PSGL-1 pathway-mediated correction of hemostasis

Abstract

As one of the virulence factors of Bacillus anthracis, lethal toxin (LT) induces various pathogenic responses including the suppression of the coagulation system. In this study, we observed that LT markedly increased the circulating soluble P-selectin (sP-sel) levels and microparticle (MP) count in wild-type but not P-selectin (P-sel, Selp^-/-) or P-sel ligand-1 (PSGL-1, Selplg^-/-) knockout mice. Because sP-sel induces a hypercoagulable state through PSGL-1 pathway to generate tissue factor-positive MPs, we hypothesized that the increase in plasma sP-sel levels can be a self-rescue response in hosts against the LT-mediated suppression of the coagulation system. In agreement with our hypothesis, our results indicated that compared with wild-type mice, Selp^-/- and Selplg^-/- mice were more sensitive to LT. In addition, the recombinant sP-sel treatment markedly ameliorated LT-mediated pathogenesis and reduced mortality. As a result, elicitation of circulating sP-sel is potentially a self-rescue response, which is beneficial to host recovery from an LT-induced hypocoagulation state. These results suggest that the administration of sP-sel is likely to be useful in the development of a new strategy to treat anthrax.

Keywords: P-selectin ligand 1; Soluble P-selectin; anthrax lethal toxin; hemostasis; microparticle.

Figure 1.

Anthrax lethal toxin (LT)-induced lung damage. H & E staining of lung sections from protective antigen (PA)- (A, B) and LT-treated (lethal dose) (C, D) mice (72 h after PA and LT treatments). The lung tissue retaining Evans blue levels indicated the plasma leakage (E) (72 h after PA and LT treatments). Plasma soluble P-selectin (sP-sel) levels in mice treated with PA and LT for 72 h are shown (F). Normal saline was used for the vehicle treatments (E, F). All quantified results in this report are presented as mean and standard deviation (SD). n = 3 (A–D). *P < 0.05, **P < 0.01, compared with respective indicated groups. n = 6 (E–F; 3 experiments with 2 replicates).

Figure 2.

Circulating sP-sel and microparticle (MP) levels. Circulating sP-sel (ELISA) (A) and MPs (flow cytometry) (B) levels of mice treated with LT (4.5 mg/kg) for 0, 24, 48, and 72 h were analyzed. Because 72-h treatments of LT showed marked increases in circulating sP-sel and MPs (A, B), the following experiments (C–E) were conducted using a 72-h time course. The relative circulating sP-sel levels in MP and MP-poor plasma (MPPP) fractions were analyzed in wild type C57Bl/6J mice 72 h after the LT treatments (C). The MP levels were analyzed in wild type, P-sel and PSGL-1 KO mice 72 h after the LT treatments (D); the plasma sP-sel levels were analyzed in wild-type and PSGL-1 KO mice 72 h after the LT treatments (E). There were no detectable sP-sel levels in P-sel KO mice. *P < 0.05, compared with respective 0-h untreated groups (A-B, D-E). *P < 0.05, compared with respective PA groups, ^† P < 0.05, compared with respective total groups (C). n = 8 (A-E). Treatments of normal saline were used the vehicle controls (C-E).

Figure 3.

Plasma clotting time and tissue factor-positive MPs. The recalcification plasma clotting time of plasma samples obtained from the vehicle-, control Ig- (Ctrl Ig), recombinant P-selectin IgG-Fc fusion protein- (P-sel-Fc), PA-, and LT-treated mice (A, 30 min, treated in vitro; B, 72 h, treated in vivo). Flow cytometry of TF-positive MPs (C-D). The green and purple labels show the representative data of the PA- and LT-treated groups, respectively (C). Percentage of TF-positive MPs was quantified in wild-type and PSGL-1 KO mice treated with vehicle, Ctrl Ig, P-sel-Fc, PA, and LT for 72 h. *P < 0.05, †P < 0.05, ^#P < 0.05, compared with PA (panel 4 vs. panel 5), Ctrl Ig (panel 2 vs. panel 3) and Ctrl Ig+LT groups (panel 5 vs. panel 6), respectively (B). ^†P < 0.05, ^#P < 0.05, Ctrl Ig (panel 2 vs. panel 3) and Ctrl Ig+LT groups (panel 6 vs. panel 7), respectively (D). n = 6 (A, B, D).

Figure 4.

Mortality in LT-treated mice. PA (3.9 mg/kg) and LT (3.9 mg/kg) were used in wild-type (WT; C57BL/6J; Selp^+/+, Selplg^+/+) (A, B), P-sel KO (C57BL/6J; Selp^−/−) (A), and PSGL-1 KO (C57BL/6J; Selplg^−/−) (B) mice. LT-induced mortality in wild mice was compared with LT-induced mortality of P-sel KO and PSGL-1 KO mice (B and C) (WT, LT vs. KO [both P-sel and PSGL-1 KO], LT, *P < 0.05). The mortality was plotted as Kaplan–Meier curves (A, B) (n = 8) (3 experiments with 2–3 mice/group).

Figure 5.

Plasma leakage and blood parameters revealed P-sel-mediated rescue. Experimental outline (A). To evaluate the amelioration of PA or LT (4.5 mg/kg) treatment-induced pathogenesis, the effects of P-sel-Fc pretreatment were compared with those of control Ig treatment. Lung retaining Evans blue (B), platelet counts (C), hematocrit count (D), and serum aspartate aminotransferase (AST) levels (E) of mice were recorded before (0 h) and after (48 and 72 h) treatment with PA and LT. *P < 0.05, **P < 0.01, vs. respective PA + control Ig groups; ^#P < 0.05, significant amelioration vs. respective LT + control Ig groups (B-E). n = 8. The mouse drawing used in this report was originally published in Blood by Huang, HS et al. © the American Society of Hematology. Reproduced by permission of Hsin-Hou Chang. Permission to reuse must be obtained from the rightsholder.

Figure 6.

Recombinant P-sel-mediated rescue of LT-induced mortality in mice. PA (4.5 mg/kg) and LT (4.5 mg/kg; intravenous injection; a lethal dose) (A, B) and the spores of B. anthracis (1 × 10⁶ CFU/mouse; intraperitoneal injection; a lethal dose) (C) were used in wild-type (WT) C57BL/6J mice. The rescue of LT and spore induced mortality 4 h after pretreatment with recombinant P-selectin IgG-Fc fusion protein (P-sel-Fc; 1.2 mg/kg) (A, C) and anti-P-selectin Ig (anti-P-sel Ig; 1.2 mg/kg) (B) was compared with respective control Ig (A-C) and bovine serum albumin (C). The mortality was plotted as Kaplan–Meier curves (A-C; control Ig vs. P-sel-Fc, **P < 0.01 in A; albumin and control Ig vs. P-sel-Fc, both *P < 0.05 in C). n = 12 (A-B, and spore + control Ig groups in C; 3 experiments with 4 mice/group); n = 17 (PA groups and spore + P-sel-Fc groups in C; 3 experiments with 4 mice/group, and 1 experiment with 5 mice/group); n = 5 (albumin groups in C; 2 experiments with 2 and 3 mice/group).