Specificity of Legionella pneumophila and Coxiella burnetii vacuoles and versatility of Legionella pneumophila revealed by coinfection

Abstract

Legionella pneumophila and Coxiella burnetii are phylogenetically related intracellular bacteria that cause aerosol-transmitted lung infections. In host cells both pathogens proliferate in vacuoles whose biogenesis displays some common features. To test the functional similarity of their respective intracellular niches, African green monkey kidney epithelial (Vero) cells, A/J mouse bone marrow-derived macrophages, human macrophages, and human dendritic cells (DC) containing mature C. burnetii replication vacuoles were superinfected with L. pneumophila, and then the acidity, lysosome-associated membrane protein (LAMP) content, and cohabitation of mature replication vacuoles was assessed. In all cell types, wild-type L. pneumophila occupied distinct vacuoles in close association with acidic, LAMP-positive C. burnetii replication vacuoles. In murine macrophages, but not primate macrophages, DC, or epithelial cells, L. pneumophila replication vacuoles were acidic and LAMP positive. Unlike wild-type L. pneumophila, type IV secretion-deficient dotA mutants trafficked to lysosome-like C. burnetii vacuoles in Vero cells where they survived but failed to replicate. In primate macrophages, DC, or epithelial cells, growth of L. pneumophila was as robust in superinfected cell cultures as in those singly infected. Thus, despite their noted similarities, L. pneumophila and C. burnetii are exquisitely adapted for replication in unique replication vacuoles, and factors that maintain the C. burnetii replication vacuole do not alter biogenesis of an adjacent L. pneumophila replication vacuole. Moreover, L. pneumophila can replicate efficiently in either lysosomal vacuoles of A/J mouse cells or in nonlysosomal vacuoles of primate cells.

FIG. 1.

Replication vacuole formation in Vero cells infected with C. burnetii and/or L. pneumophila. Vero cells infected with C. burnetii phase II (Cb; arrows) for 48 h were superinfected with wild-type L. pneumophila (Lp; arrowheads) or the isogenic dotA strain (Lp dotA; arrowheads) for 24 h. Vero cells singly infected with C. burnetii for 72 h or L. pneumophila for 24 h were used as controls. Live cells were stained with the acidotropic base acridine orange, and the same field was viewed by phase-contrast and fluorescence microscopy. Fixed cells were examined for localization of the lysosomal glycoprotein LAMP-1 by indirect immunofluorescence. Representative electron micrographs of infected cells are also depicted. In both singly and superinfected cells, acridine orange was sequestered by replication vacuoles harboring C. burnetii but not L. pneumophila. In singly infected cells, LAMP-1 (green) localized to the membrane of distinct replication vacuoles harboring C. burnetii (red) but not L. pneumophila (red). In superinfected cells, LAMP-1 (green) also localized to the C. burnetii (blue) replication vacuole membrane but not to the L. pneumophila (red) replication vacuole membrane. C. burnetii and L. pneumophila were never observed in the same replication vacuole. In C. burnetii-infected cells superinfected with L. pneumphila dotA type IV secretion mutants, nonreplicating L. pneumophila (red) localized to the LAMP-1-positive (green) vacuole containing replicating C. burnetii (blue). Vacuoles harboring both L. pneumophila dotA and C. burnetii were observed in 60.7% ± 5.0% of cells infected with both pathogens. In electron micrographs, L. pneumophila are distinguished from C. burnetii by their larger size (>1 μm in length), uniform rod morphology, electron-translucent hydroxybutyrate granules, and lack of filamentous chromatin. Percent colocalization of dotA mutants is expressed as the mean ± the standard deviation of three independent experiments in which at least 50 superinfected cells were evaluated. Images are representative of three independent experiments in which at least 100 singly infected or superinfected cells were examined.

FIG. 2.

Growth of L. pneumophila in C. burnetii-infected Vero cells. Vero cells infected with C. burnetii (phase II) for 48 h were superinfected with wild-type L. pneumophila or the isogenic dotA mutant. Vero cells were singly infected with L. pneumophila as a control. Intracellular L. pneumophila were released from host cells at 2, 24, and 48 h p.i., and CFU assays performed. The 0 h time point represents the titer of the starting inoculum prior to washing extracellular organisms from the monolayer at 2 h p.i. Wild-type L. pneumophila CFU increased 53- and 86-fold between 2 and 48 h p.i. in singly infected (solid line with squares) and superinfected cells (dotted line with squares), respectively. No replication of the dotA mutant was observed in singly infected (dotted line with circles) or superinfected (solid line with circles) cells. The results are expressed as the mean from three independent experiments with error bars representing the standard deviation.

FIG. 3.

Replication vacuole formation in murine BMDM infected with C. burnetii and/or L. pneumophila. BMDM from A/J mice were infected with phase I C. burnetii (Cb I; arrows) for 48 h and then superinfected with L. pneumophila (Lp; arrowheads) for 18 h. BMDM singly infected with phase I or phase II C. burnetii (Cb II; arrows) for 66 h or L. pneumophila for 18 h were also evaluated. Live cells were stained with the acidotropic base acridine orange, and the same field was viewed by phase-contrast and fluorescence microscopy. Fixed cells were examined for localization of the lysosomal glycoprotein LAMP-2 by indirect immunofluorescence. In singly infected cells, acridine orange was sequestered by replication vacuoles harboring both phase I C. burnetii and L. pneumophila. In singly infected cells, LAMP-2 (green) localized to the vacuole membrane surrounding both phase II (red) and phase I C. burnetii (red), as well as L. pneumophila (red). In phase II C. burnetii-infected cells, single organisms were observed scattered throughout the cytoplasm tightly bounded by a LAMP-2-positive membrane. In C. burnetii phase I- or L. pneumophila-infected cells, replicating organisms were observed in multiple large replication vacuoles. In superinfected cells, LAMP-2 (green) also localized to vacuoles harboring phase I C. burnetii (blue) and L. pneumophila (red). In no instance were phase I C. burnetii and L. pneumophila observed in the same vacuole. Images are representative of three independent experiments where at least 100 singly infected or superinfected cells were examined.

FIG. 4.

Replication vacuole formation in human primary macrophages infected with C. burnetii and/or L. pneumophila. Cells were infected with phase II C. burnetii (Cb; arrows) for 36 h and then superinfected with L. pneumophila (Lp; arrowheads) for 12 h. Cells singly infected with C. burnetii for 48 h or L. pneumophila for 12 h were used as controls. Fixed cells were examined for localization of the lysosomal glycoprotein LAMP-1 by indirect immunofluorescence. In singly infected cells, LAMP-1 (green) localized to the membrane of distinct replication vacuoles harboring C. burnetii (red) but not L. pneumophila (red). In superinfected cells, LAMP-1 (green) also localized to the C. burnetii (blue) replication vacuole membrane but not to the L. pneumophila (red) replication vacuole membrane. In no instance were C. burnetii and L. pneumophila observed in the same replication vacuole. Images are representative of three independent experiments in which at least 100 singly infected or superinfected cells were examined.

FIG.5.

Replication vacuole formation in human primary DC infected with C. burnetii and/or L. pneumophila. Cells were infected with phase II C. burnetii (Cb; arrows) for 36 h and then superinfected with L. pneumophila (Lp; arrowheads) for 12 h. Cells singly infected with C. burnetii for 48 h or L. pneumophila for 12 h were used as controls. Fixed cells were examined for localization of the lysosomal glycoprotein LAMP-1 by indirect immunofluorescence. Representative electron micrographs of infected cells are also depicted. In singly infected cells, LAMP-1 (green) localized to the membrane of distinct replication vacuoles harboring C. burnetii (red) but not L. pneumophila (red). In superinfected cells, LAMP-1 (green) also localized to distinct vacuoles harboring replicating C. burnetii (blue) but not replicating L. pneumophila. On rare occasions one or two L. pneumophila organisms were observed in C. burnetii replication vacuoles in superinfected cells (lower dual-infection panels). L. pneumophila are distinguished from C. burnetii in electron micrographs by their larger size (>1 μm in length), uniform rod morphology, electron-translucent poly-3-hydroxybutyrate granules, and lack of filamentous chromatin. Images are representative of three independent experiments where at least 100 singly infected or superinfected cells were examined.

FIG. 6.

Growth of L. pneumophila in C. burnetii-infected primary monocyte-derived macrophages and DC. Intracellular L. pneumophila were released from host cells and CFU were quantified to assay replication in singly infected macrophages and DC and in cells infected with C. burnetii (phase II) for 36 h. The 0-h time point represents the titer of the starting inoculum prior to washing extracellular organisms from the monolayer at 2 h p.i. (This wash step was not conducted with nonadherent DC cultures.) (A) L. pneumophila CFU increased 188 and 165-fold between 2 and 48 h p.i. in singly infected (solid line with squares) and superinfected macrophages (dotted line with squares), respectively. (B) L. pneumophila CFU increased 27- and 16-fold between 2 and 72 h p.i. in singly infected (solid line with triangles) and superinfected DC (dotted line with triangles), respectively. The results are expressed as the means from three independent experiments, with error bars representing the standard deviations.