A role for connexin43 in macrophage phagocytosis and host survival after bacterial peritoneal infection

Abstract

The pathways that lead to the internalization of pathogens via phagocytosis remain incompletely understood. We now demonstrate a previously unrecognized role for the gap junction protein connexin43 (Cx43) in the regulation of phagocytosis by macrophages and in the host response to bacterial infection of the peritoneal cavity. Primary and cultured macrophages were found to express Cx43, which localized to the phagosome upon the internalization of IgG-opsonized particles. The inhibition of Cx43 using small interfering RNA or by obtaining macrophages from Cx43 heterozygous or knockout mice resulted in significantly impaired phagocytosis, while transfection of Cx43 into Fc-receptor expressing HeLa cells, which do not express endogenous Cx43, conferred the ability of these cells to undergo phagocytosis. Infection of macrophages with adenoviruses expressing wild-type Cx43 restored phagocytic ability in macrophages from Cx43 heterozygous or deficient mice, while infection with viruses that expressed mutant Cx43 had no effect. In understanding the mechanisms involved, Cx43 was required for RhoA-dependent actin cup formation under adherent particles, and transfection with constitutively active RhoA restored a phagocytic phenotype after Cx43 inactivation. Remarkably, mortality was significantly increased in a mouse model of bacterial peritonitis after Cx43 inhibition and in Cx43 heterozygous mice compared with untreated and wild-type counterparts. These findings reveal a novel role for Cx43 in the regulation of phagocytosis and rearrangement of the F-actin cytoskeleton, and they implicate Cx43 in the regulation of the host response to microbial infection.

FIGURE 1

Cx43 is expressed in macrophages and localizes to phagosomes. A, Macrophage cell lines, including J774 macrophages (J774), RAW 264.7 (RAW), and primary murine peritoneal macrophages (Mθ) from Cx43 wild-type (+/+) and heterozygous (+/−) mice, were subjected to SDS-PAGE and were assessed for the expression of Cx43 by immunoblotting. Blots were stripped and reprobed with Abs against F-actin. B, RAW264.7 macrophages were allowed to internalize Dynabeads for the indicated periods of time (5, 15, 30, 60, and 90 min) and then washed with copious amounts of PBS, fixed, permeabilized, and immunostained with Abs to Cx43 and imaged by confocal microscopy. Shown are the corresponding images for Cx43 (green) and DIC at each time point indicated. Arrows show the presence of bound (at 5 min) or internalized particles (at all other time points). Representative of three separate experiments. Size bar = 3 μm. C, Phagosomes (Phag) were isolated from J774 macrophages that had been fed a meal of latex beads and immunoblotted alongside J774 cells lysates (Mθ) using Abs against Cx43, LAMP-2, a phagosomal marker, and p38-MAPK (p38), a cytoplasmic protein expectedly absent from phagosomal membranes. Representative of three separate experiments. D, Rate of phagocytosis after 1 h of either Dynabeads (open bars) or latex beads (filled bars) in RAW264.7 macrophages (open bars) or J774 cells (filled bars) that had been either untreated, treated with anti-FcR, or treated with nonspecific IgG.

FIGURE 2

Cx43 inhibition decreases phagocytosis by macrophages. A, J774 macrophages were either untreated (Ctl) or were treated with oleamide (OLM) or vehicle (Veh) and assessed for the ability to undergo phagocytosis of opsonized sheep erythrocytes. Representative of three separate experiments. *, p < 0.05. B, RAW264.7 macrophages were either untreated (control) or were treated with siRNA against either Cx43 (Cx43 siRNA) or against no known targets (Ctl siRNA). The expression of Cx43 and F-actin were assessed by Western blotting (upper panel). Shown is the quantification of phagocytosis. *, p < 0.05; representative of at least three separate experiments (lower panel). C, The adhesion of Dynabeads to RAW264.7 macrophages that were either treated with control siRNA (open bars) or with siRNA to Cx43 (filled bars). Adhesion occurred as shown either at 4°C or after treatment with the phagocytosis inhibitor cytochalasin D. D, The extent of phagocytosis of opsonized erythrocytes by peritoneal macrophages harvested from either wild-type (+/+) or Cx43 heterozygous mice (+/−). *, p < 0.05. Representative DIC micrographs of macrophages from wild-type (ii) and Cx43^+/− mice (iii) that had been allowed to internalize SRBCs. Size bar = 5 μm.

FIGURE 3

Heterologous transfection of HeLa cells with Fc-Cx43 confers phagocytic ability. HeLa cells deficient in Cx43 were transfected with FcIIA and/or GFP and/or GFP-Cx43, then allowed to phagocytose SRBCs A, DIC micrograph of nontransfected HeLa cells in association with SRBCs (arrow). B and C, Confocal micrograph showing the GFP emission (B) and DIC image (C) revealing the distribution of SRBCs adherent to the surface of Fc-transfected HeLa cell. D and E, Confocal micrograph showing the GFP emission (D) and DIC image (E) of HeLa cells that had been cotransfected with FcRIIA and GFP-Cx43 and allowed to engulf SRBCs. Arrows show the presence of internalized SRBCs. Size bar = 10 μm. F, Phagocytosis capacity among the different groups. *, p < 0.05 vs control of at least seven separate experiments. G, SDS-PAGE with Abs to Cx43 in lysates obtained from RAW264.7 macrophages (Mθ, positive control), nontransfected HeLa cells (−ve), GFP-transfected HeLa cells (GFP), and GFP-Cx43-transfected HeLa cells (GFP-Cx43). Arrows show the location of Cx43 at 43 kDa in macrophages (the positive control) and GFP-Cx43 at 70 kDa in GFP-Cx43-transfected HeLa cells. Representative of three separate experiments.

FIGURE 4

Expression of adenovirus GFP-Cx43 reverses the impairment in phagocytosis observed in Cx43-deficient macrophages. A, Macrophages were harvested from the livers of embryonic Cx43^+/+, Cx43^+/−, and Cx43^−/− mice at gestational age embryonic days 16–18 as described in Materials and Methods, immunostained with Abs against the macrophage marker CD45, and examined by confocal microscopy. Representative CD45 staining (i-iii) and the corresponding DIC images (iv-vi) are shown. B, Embryonic macrophages were harvested from livers of Cx43^+/+, Cx43^+/−, and Cx43^−/− mice and allowed to internalize opsonized RBCs as described in Materials and Methods. The phagocytosis index (no. of cells with at least one internalized RBC per 100 cells relative to wild-type macrophages) is shown. Representative of five separate experiments. *, p < 0.05 vs wild-type macrophages. C, Macrophages were harvested from Cx43^+/+, Cx43^+/−, and Cx43^−/− mice and infected with adenoviruses expressing GFP (yellow bars), GFP-wtCx43 (green bars), or GFP-dnCx43 (blue bars). Cells were then allowed to undergo phagocytosis of opsonized RBCs as described in Materials and Methods. The phagocytosis index (number of macrophages with at least one internalized particle per 100 cells relative to GFP-infected macrophages from each strain) is shown. †, p < 0.05 vs GFP-infected cells from wild-type mice. Note that in each case the rate of phagocytosis was decreased in GFP-infected cells from +/− and −/− mice vs +/+ mice. *, p < 0.05 vs GFP-infected macrophages for each strain. Note that in each case infection with GFP-wtCx43 leads to a significant increase in phagocytosis; **, p < 0.05 vs GFP-wtCx43. Note that for each mouse strain, infection with GFP-dnCx43 leads to a significant decrease in the rate of phagocytosis compared with infection with wild-type Cx43. Representative of at least five separate experiments with more than five mice and 100 cells per group.

FIGURE 5

Cx43 plays a role in the FcR-induced activation of RhoA and actin cup formation. A, J774 macrophages were treated with LPA (10 μM, a Rho activator), Y27632 (10 μM, a Rho inhibitor), or Cx43 siRNA, and assessed for the ability to internalize opsonized SRBCs, as compared with vehicle-treated control cells. *, p < 0.05; representative of at least five separate experiments. B, Fold activation of RhoA in RAW264.7 cells as assessed by ELISA after RAW264.7 cells had been transfected with either control siRNA or Cx43 siRNA, under the following conditions: untreated control (Ctl), IgG-opsonized SRBCs (RBC), or LPA (10 μM); *, p < 0.05 vs control; representative of at least five separate experiments; **, p < 0.01 vs control-siRNA RBC-treated cells; †, p < 0.01 vs control-siRNA LPA-treated cells. C, Phagocytosis capacity of RAW264.7 macrophages that had been treated with Cx43 siRNA or control siRNA then subsequently treated with media alone (Ctl) or transfected with constitutively active RhoA-GTP (V14-RhoA). *, p < 0.05 vs control for both control-siRNA or Cx43-siRNA; representative of at least three separate experiments. D, Percentage of macrophages shown to exhibit the formation of actin cups at the surface of RAW264.7 cells that had internalized IgG-opsonized RBCs, under the following conditions: untreated (Ctl), or transfected with control siRNA (Ctrl-siRNA) or siRNA to Cx43 (Cx43-siRNA). E–J, Representative confocal micrographs showing actin cup formation around nascent SRBC phagosomes in RAW264.7 cells. Cells were either nontransfected (Control) or were transfected with the indicated siRNA. Size bar = 2 μm. Corresponding DIC images (E, G, and I) and F-actin fluorescence (F, H, and J) are displayed. Arrows indicate the formation of actin cups.

FIGURE 6

Cx43 participates in phagocytosis in vivo and plays a role in host survival in response to bacterial peritonitis. A, Phagocytosis rate of Oregon Green-labeled E. coli by peritoneal macrophages of mice that had been injected i.p. either with saline (Ctl), vehicle (Veh), or oleamide (Olm, 25 mg/kg) daily for the preceding 3 days. B and C, Representative merged confocal images showing DIC (gray) and Oregon Green fluorescence (green staining) of macrophages obtained from mice that had been injected with either vehicle control (B) or oleamide (C). Arrows point to macrophages that had internalized bacteria. Representative of three separate experiments; p < 0.05. D, Rate of phagocytosis of IgG-opsonized SRBCs by peritoneal macrophages that had been harvested from mice that were injected i.p. with different concentrations of oleamide (Olm-1, 12.5 mg/kg; Olm-2, 25 mg/kg) or vehicle (Veh). *, p < 0.05, n = 3 separate experiments with three mice per experiment involving >100 macrophages per mouse. E, Percentage mortality 48 h after injection of mice with 8 × 10⁸ CFU live E. coli in addition to either oleamide alone (Olm), vehicle alone (Veh), E. coli alone, E. coli plus vehicle (E. coli + Veh), or E. coli plus oleamide (50 mg/kg, E. coli + Olm). *, p < 0.05, representative of five separate experiments. F–H, Representative merged confocal images showing DIC (gray) and Oregon Green fluorescence (green staining) of macrophages obtained from wild-type mice (F) or Cx43^+/− mice (G). Arrows point to macrophages that had internalized bacteria. Quantification is shown in H. Representative of three separate experiments; *, p < 0.05.

FIGURE 7

Cx43-deficient mice demonstrate reduced survival after bacterial peritoneal sepsis. Age- and weight-matched Cx43^+/+ (solid lines) and Cx43^+/− mice (dashed lines) (n = 10/group) were injected with E. coli daily for 3 days as described in Materials and Methods. Shown are Kaplan-Meier survival curves revealing the proportion of mice that are alive in each group as a function of time from the first injection of bacteria.