Expression of factor V by resident macrophages boosts host defense in the peritoneal cavity

Abstract

Macrophages resident in different organs express distinct genes, but understanding how this diversity fits into tissue-specific features is limited. Here, we show that selective expression of coagulation factor V (FV) by resident peritoneal macrophages in mice promotes bacterial clearance in the peritoneal cavity and serves to facilitate the well-known but poorly understood "macrophage disappearance reaction." Intravital imaging revealed that resident macrophages were nonadherent in peritoneal fluid during homeostasis. Bacterial entry into the peritoneum acutely induced macrophage adherence and associated bacterial phagocytosis. However, optimal control of bacterial expansion in the peritoneum also required expression of FV by the macrophages to form local clots that effectively brought macrophages and bacteria in proximity and out of the fluid phase. Thus, acute cellular adhesion and resident macrophage-induced coagulation operate independently and cooperatively to meet the challenges of a unique, open tissue environment. These events collectively account for the macrophage disappearance reaction in the peritoneal cavity.

Figure 1.

Coagulation and adhesion additively cooperate to account for the MDR in response to inflammation. (A) Gene array analysis (from ImmGen; n = 3 separate pools) of classical coagulation factors in major tissue resident macrophages, including those from the spleen, central nervous system (CNS), lung, and peritoneum. (B) Quantification of LPMs in peritoneal lavage 3 h after zymosan injection i.p. when clotting and/or adhesion was inhibited. (C) Aggregates retrieved from the peritoneum 5 h after zymosan injection. (D–G) Immunofluorescence staining of the aggregates for fibrin(ogen) and macrophage markers. D and G are stained frozen sections of the clots; E and F are whole-mount preparations. Scale bars represent 100 µm (D and G), 50 µm (E), and 10 µm (F). (H) Flow cytometry on peritoneal exudate cells from untreated mice (left), 3 h after zymosan i.p. (middle), and clots 3 h after zymosan i.p. (right). (I) Quantification of LPMs 3 h after zymosan injection in clots and omenta in WT and Lyz2^Cre;Tln1^fl/fl mice. One-way ANOVA was used to test statistical significance. Symbols represent individual mice studied. Error bars represent ± SEM. All experiments were repeated at least two or three times. **, P < 0.01; ***, P < 0.001.

Figure 2.

FV and TF from resident peritoneal macrophages are responsible for peritoneal fluid clotting. (A) Quantification of FV activity in plasma or peritoneal fluid of various genotypes of mice or after bone marrow transplant of indicated donor genotypes into irradiated WT recipients (last bars on the right). Zym, zymosan. (B) Quantification of LPMs 3 h after zymosan injection. (C) Quantification of LPMs 3 h after zymosan injection. One-way ANOVA was used to test statistical significance, except for the bone marrow chimera result in B and the hF^+/+ groups and F3^fl/fl groups in C, which used a two-tailed t test. Symbols represent individual mice studied. Error bars represent ± SEM. All experiments were repeated at least two or three times. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 3.

Macrophage adhesion and macrophage-driven clotting cooperatively promote peritoneal bacterial clearance. (A) Quantification of TAT complex in peritoneal fluid and plasma in the steady state or after E. coli infection. (B) Quantification of TAT complex in peritoneal fluid 4 h after E. coli infection. Ctrl, control. (C) CFUs per microliter peritoneal fluid from untreated, hirudin-treated WT, Lyz2^Cre;Talin^fl/fl mice, hirudin-treated Lyz2^Cre;Talin^fl/fl mice, and clodronate liposome-treated mice 4 h after E. coli infection. (D) CFUs per microliter peritoneal fluid from Gata6^fl/fl and Lyz2^Cre;Gata6^fl/fl mice 4 h after E. coli infection. (E) CFUs per microliter peritoneal fluid from C57BL/6J treated with or without hirudin 4 h after E. coli infection. (F) CFUs per microliter peritoneal fluid from control and F5^−/−;AlbF5Tg mice 4 h after E. coli infection. One-way ANOVA was used to test statistical significance, except for C, D, and F and the F5^−/−;AlbF5Tg groups in B, which were examined by a two-tailed t test. Symbols represent individual mice studied. Error bars represent ± SEM. All experiments were repeated at least two or three times. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 4.

MDR restricts bacterial dissemination out of the peritoneal cavity to the spleen. (A) CFUs per spleen from Tln1^fl/fl mice or Lyz2^Cre;Tln1^fl/fl mice at 16 h after i.v. E. coli infection. (B) CFUs per spleen from mice untreated or treated with heparin, anti-integrin β1 and β2 blocking antibodies, or hirudin at 16 h after i.v. E. coli infection. Differences between groups are not statistically significant. (C) CFU per spleen from F5 littermates and F5^−/−;AlbF5Tg mice at 16 h after i.v. E. coli infection. Differences are not statistically significant. Ctrl, control. (D) CFUs per spleen from untreated and heparin-treated mice at 16 h after i.p. E. coli infection. (E) CFUs per spleen from F5 littermates and F5^−/−;AlbF5Tg mice treated with anti-integrin β1 and β2 blocking antibodies at 16 h after i.p. E. coli infection. (F) CFUs per spleen from mice treated with anti-integrin β1 and β2 blocking antibodies and mice treated with hirudin and anti-integrin β1 and β2 blocking antibodies at 16 h after i.p. E. coli infection. (G) CFUs per spleen from Tln1^fl/fl mice or Lyz2^Cre;Tln1^fl/fl mice treated with hirudin at 16 h after i.p. E. coli infection. Differences are not statistically significant. Two-tailed t tests were used to test statistical significance for A and C–G. One-way ANOVA was used to test statistical significance for B. Symbols represent individual mice studied. Error bars represent ± SEM. All experiments were repeated at least one to three times. *, P < 0.05; **, P < 0.01; ***, P < 0.001.