Brucella coopts the small GTPase Sar1 for intracellular replication

Abstract

The pathogen Brucella abortus resides inside macrophages within a unique, replication-permissive organelle that is derived from the endoplasmic reticulum (ER). Although dependent on the Brucella type IV secretion system VirB, the mechanisms governing the biogenesis of this compartment remain elusive. Here, we investigated a putative role of the early secretory pathway in ER membrane accretion by the Brucella-containing vacuoles (BCVs). We show that BCVs interact with ER exit sites (ERES), and blockade of Sar1 activity, which disrupts ERES, prevents intracellular replication of Brucella. In cells expressing the dominant interfering form Sar1[T39N], BCVs do not acquire ER membranes, suggesting that they are unable to mature into replicative organelles. By contrast, treatments that block subsequent secretory events do not affect bacterial replication. We propose that Sar1-dependent ERES functions, but not subsequent secretory events, are essential for the biogenesis of the Brucella replicative compartment and, thus, bacterial replication. These results assign an essential role for Sar1 in pathogenesis of an intracellular bacterium.

Fig. 1.

BCVs interact with COPII-positive compartments in macrophages. (A) Confocal microscopy micrographs of BMDMs infected for 4, 12, or 24 h with either WT or ΔvirB9 B. abortus strains expressing GFP. When required, BFA was applied to infected BMDMs for 2 h before analysis. Fixed cells were stained with anti-sec31 antibodies (red) to label COPII-positive compartments. (Insets) GFP-Brucella in close apposition to COPII-positive compartments. Arrows indicate examples of BCVs in close contact with COPII-labeled compartments. (B) Quantitation of BCV apposition to COPII-positive compartments over time. Single confocal microscopy 0.2-μm sections of random fields were acquired, and the appositions of BCVs to COPII-positive compartments were counted as positive when a coincident yellow fringe was observed (see A Insets). Data are means ± SD of three independent experiments. (C) Confocal microscopy micrographs of BMDMs infected with either WT or ΔvirB9 B. abortus strains for 4, 12, or 24 h. Fixed cells were stained with anti-β-COP antibodies to label COPI-positive compartments (green) and propidium iodide (red) to detect bacterial and host cell DNA. Arrows indicate BCVs positioned away from COPI-positive compartments. (D) Quantitation of BCV apposition to COPI-positive compartments over a 24-h infection period. Single confocal microscopy sections of random fields were acquired, and the appositions of BCVs to COPI-positive compartments were counted as positive when a coincident yellow fringe was observed. Data are means ± SD of three independent experiments. All micrographs shown are projections of three consecutive confocal sections. (Scale bars: 10 μm; Insets, 0.5 μm.)

Fig. 2.

BCVs interact with functional ERES. (A) Confocal microscopy micrographs of GFP-p58 expressing BMDMs infected with WT B. abortus for either 4 or 12 h. Fixed cells were stained with anti-sec31 antibodies (red) to label COPII-positive compartments and propidium iodide (pseudocolored in blue) to detect bacterial and host cell DNA. GFP-p58 is shown in green. Arrows indicate BCVs apposed to p58- and sec31-positive ERES. (Insets) Magnified views of typical BCV–ERES appositions. The micrographs shown are projections of three consecutive confocal sections. (Scale bars: 10 μm; Insets, 1 μm.) (B) Quantitation of BCVs apposed to p58-positive compartments in untreated or BFA-treated BMDMs expressing GFP-p58. Data are means ± SD of three independent experiments. Where indicated, BFA was applied for 2 h before processing to ensure that GFP-p58 was relocated to ERES after disruption of the secretory pathway.

Fig. 3.

Disruption of Sar1 activity and ERES prevents Brucella replication. (A) Confocal microscopy micrographs of B. abortus-infected HeLa cells transfected with various Sar1 or ARF1 alleles after 24 h of infection. HeLa cells were infected with WT GFP-expressing Brucella then transfected with plasmids encoding either HA-tagged Sar1, Sar1[T39N], Sar1[H79G], ARF1, or ARF1[T31N]. After 24 h of infection, cells were fixed and stained with rabbit anti-sec31 (red) and mouse anti-HA (Upper) antibodies to label ERES and detect transfected cells, respectively. The micrographs shown are projections of three consecutive confocal sections and depict typical bacterial replication patterns with the various GTPase alleles. Asterisks indicate clustered ERES either around the Golgi apparatus (Sar1) or in a juxtanuclear area (Sar1[H79G]). Only the expression of Sar1[T39N] prevents bacterial replication (arrow). (B) Quantitation of the effect of Sar1 and ARF1 alleles expression upon Brucella replication, expressed as the percentage of infected cells displaying bacterial replication. Scoring of bacterial replication was performed after 24 h of infection by epifluorescence microscopy analysis of untransfected (control) or transfected, infected cells. Data are means ± SD of five independent experiments.

Fig. 4.

Inhibition of Sar1 activity prevents ER membrane acquisition by BCVs. (A) Quantitation of the presence of the ER marker calnexin or the late endosomal/lysosomal marker LAMP-1 on BCVs in HeLa cells expressing various Sar1 alleles. HeLa cells were infected with WT GFP-expressing B. abortus, then transfected with plasmids encoding either HA-tagged Sar1, Sar1[T39N], or Sar1[H79G]. After 24 h of infection, cells were fixed and stained with either rabbit anticalnexin or rabbit anti-LAMP-1 (red) and mouse anti-HA (blue) antibodies to detect transfected cells. Calnexin- or LAMP-1-positive BCVs were scored by confocal microscopy on single sections of untransfected cells (control) or cells transfected with either of the Sar1 alleles. Data are means ± SD of three independent experiments. (B) Confocal microscopy micrographs illustrating calnexin (Left) and LAMP-1 (Right) staining of BCVs in HeLa cells expressing either Sar1[T39N] (Upper) or Sar1[H79G] (Lower). GFP-Brucella are shown in green, calnexin or LAMP-1 in red, and HA-tagged Sar1 alleles in blue. The micrographs shown are projections of three consecutive confocal sections. (Insets) Magnifications of areas indicated by arrows, in which the HA staining (blue) has been removed for clarity. (Scale bars: 10 μm; Insets, 1 μm.)