The protein SdhA maintains the integrity of the Legionella-containing vacuole

Abstract

Legionella pneumophila directs the formation of a specialized vacuole within host cells, dependent on protein substrates of the Icm/Dot translocation system. Survival of the host cell is essential for intracellular replication of L. pneumophila. Strains lacking the translocated substrate SdhA are defective for intracellular replication and activate host cell death pathways in primary macrophages. To understand how SdhA promotes evasion of death pathways, we performed a mutant hunt to identify bacterial suppressors of the ΔsdhA growth defect. We identified the secreted phospholipase PlaA as key to activation of death pathways by the ΔsdhA strain. Based on homology between PlaA and SseJ, a Salmonella protein associated with vacuole degradation, we determined the roles of SdhA and PlaA in controlling vacuole integrity. In the absence of sdhA, the Legionella-containing vacuole was unstable, resulting in access to the host cytosol. Both vacuole disruption and host cell death were largely dependent on PlaA. Consistent with these observations, the ΔsdhA strain colocalized with galectin-3, a marker of vacuole rupture, in a PlaA-dependent process. Access of ΔsdhA strains to the macrophage cytosol triggered multiple responses in the host cell, including degradation of bacteria, induction of the type I IFN response, and activation of inflammasomes. Therefore, we have demonstrated that the Legionella-containing vacuole is actively stabilized by the SdhA protein during intracellular replication. This vacuolar niche affords the bacterium protection from cytosolic host factors that degrade bacteria and initiate immune responses.

Fig. 1.

Growth defect of a ΔsdhA strain is suppressed by the absence of PlaA phospholipase activity. (A) Growth of L. pneumophila strains in bone marrow-derived macrophages from A/J mice, determined by plating for CFU. (B and D) Quantification of the number of bacteria per vacuole 16 h after uptake in macrophages, as described previously (2). (C) Quantification of aberrant nuclei at 14 h after uptake. Infected macrophages were fixed and stained with Hoechst stain (25).

Fig. 2.

Absence of SdhA results in loss of vacuole integrity. (A–C) Fixation of bone marrow-derived macrophages from the A/J mouse is sufficient to allow immunodetection of cytosolic markers such as tubulin (green). Infected macrophages were stained for bacteria before (red) and after (blue) permeabilization of the cells, and the percentage of internalized bacteria detected before permeabilization was quantified (Materials and Methods). (D) Vacuoles were isolated from infected U937 cells (10), fixed on glass coverslips, and stained for bacteria before and after permeabilization of the vacuole membrane. (E and F) Infected macrophages were stained for bacteria (red) and galectin-3 (green) (30), and the frequency of colocalization was quantified. All images were obtained at 8 h after uptake. (Scale bars: 5 μm.)

Fig. 3.

Deletion of plaA in a ΔsdhA strain is sufficient to prevent vacuole disruption. (A) Infected bone marrow-derived macrophages from A/J mice were fixed and stained for bacteria before and after permeabilization, as in Fig. 2. (B) Macrophages were challenged with L. pneumophila strains for 6 h, fixed, and stained for bacteria and galectin-3, as in Fig. 2. The frequency of colocalization was quantified.

Fig. 4.

Inflammasome activation contributes to ΔsdhA-induced cell death and is dependent on PlaA but independent of FlaA. (A) Bone marrow-derived macrophages from A/J mice were challenged with L. pneumophila strains for 6 h and labeled with FITC-VAD before fixation. The frequency of FITC-VAD retention was quantified. (B) RNA was harvested from infected macrophages at 4 h after uptake. The relative levels of IFN-β transcript were determined by quantitative RT-PCR. (C and D) Intracellular growth of L. pneumophila was determined by plating for CFUs. Caspase activity was inhibited by the addition of Z-VAD, a pan-caspase inhibitor, or YVAD, a caspase-1 inhibitor.

Fig. 5.

Cytosolic L. pneumophila is targeted for degradation. (A, B, and D) Macrophages from A/J mice were challenged with L. pneumophila strains, then fixed at the indicated times, and intracellular bacteria were stained with Legionella-specific antisera. Degradation of bacteria was determined and quantified based on aberrant bacterial morphology. Example images are of ΔsdhA bacteria at 8 h after uptake. (Scale bars: 5 μm.) Caspases were inhibited by the addition of Z-VAD. (C) Infected macrophages were fixed and stained for intracellular bacteria and LC3, and the frequency of colocalization was quantified.

Fig. 6.

Activation of host cell death and degradation of bacteria occur downstream of vacuole disruption. (A) Macrophages from A/J mice were challenged with the ΔsdhA strain for the indicated times. FITC-VAD was added to the cells before fixation to stain activate caspases. Fixed cells were stained for bacteria before and after permeabilization to allow quantification of vacuole disruption. Replicative vacuoles are those that contain more than one bacterium. (B) Macrophages were challenged with L. pneumophila strains for 8 h. Cell-associated bacteria were stained before and after cell permeabilization to allow quantification of vacuole disruption. Caspases were inhibited by the addition of Z-VAD.