Identification of mechanisms for attenuation of the FSC043 mutant of Francisella tularensis SCHU S4

Abstract

Previously, we identified a spontaneous, essentially avirulent mutant, FSC043, of the highly virulent strain SCHU S4 of Francisella tularensis subsp. tularensis. We have now characterized the phenotype of the mutant and the mechanisms of its attenuation in more detail. Genetic and proteomic analyses revealed that the pdpE gene and most of the pdpC gene were very markedly downregulated and, as previously demonstrated, that the strain expressed partially deleted and fused fupA and fupB genes. FSC043 showed minimal intracellular replication and induced no cell cytotoxicity. The mutant showed delayed phagosomal escape; at 18 h, colocalization with LAMP-1 was 80%, indicating phagosomal localization, whereas the corresponding percentages for SCHU S4 and the ΔfupA mutant were <10%. However, a small subset of the FSC043-infected cells contained up to 100 bacteria with LAMP-1 colocalization of around 30%. The unusual intracellular phenotype was similar to that of the ΔpdpC and ΔpdpC ΔpdpE mutants. Complementation of FSC043 with the intact fupA and fupB genes did not affect the phenotype, whereas complementation with the pdpC and pdpE genes restored intracellular replication and led to marked virulence. Even higher virulence was observed after complementation with both double-gene constructs. After immunization with the FSC043 strain, moderate protection against respiratory challenge with the SCHU S4 strain was observed. In summary, FSC043 showed a highly unusual intracellular phenotype, and based on our findings, we hypothesize that the mutation in the pdpC gene makes an essential contribution to the phenotype.

FIG 1

Western blot analysis of the Francisella pathogenicity island proteins. Lysates from strains SCHU S4 and FSC043 were separated by SDS-PAGE, blotted onto nitrocellulose, and probed with an antibody to the indicated F. tularensis proteins. (A) Western blot analysis of selected FPI proteins revealed no differences between strains FSC043 and SCHUS4. (B) The truncated form of PdpC (26 kDa) is present in FSC043 in contrast to the full-length form (156 kDa) (arrows), which is visible only in the SCHU S4 lysate. There are also a number of nonspecific bands due to the reactivity of the polyclonal antibody used for detection.

FIG 2

Map of the FPI region of F. tularensis. The pdpC and pdpE genes (black arrows) and the two adjacent genes are magnified. The location of the spontaneous mutation in the pdpC gene in FSC043 is indicated by a gray line. The locations of the primers used for RT-qPCR analysis are indicated by horizontal bars. The arrows indicate the direction of transcription.

FIG 3

F. tularensis infection of J774 cells. (A) Cells were infected for 1 h with the indicated F. tularensis strains at an MOI of 30 and then incubated for 18 h. Bacterial replication was determined and expressed as mean log₁₀ CFU of triplicate wells. Experiments were repeated at least twice with similar results. The horizontal lines indicate the bacterial numbers after uptake. The asterisks indicate that the bacterial numbers were significantly different from the replication of the SCHU S4 strain at the indicated time point (*, P ≤ 0.05;**, P ≤ 0.01). (B) Culture supernatants of the infected J774 cells were assayed for LDH activity at the indicated time points, and the activity was expressed as a percentage of the level of noninfected lysed cells. Means and standard deviations (SD) of triplicate wells from one representative experiment of at least two are shown. The asterisks indicate that the cytotoxicity levels were significantly different from those of SCHU S4-infected cells for a given time point, as determined by a t test (*, P ≤ 0.05;***, P ≤ 0.001).

FIG 4

Infection of J774 cells with the ΔpdpE mutant. (A) Cells were infected for 1 h with the indicated F. tularensis strains at an MOI of 30 and then incubated for 20 h. Bacterial replication was determined and expressed as mean log₁₀ CFU of triplicate wells. Experiments were repeated twice with similar results. The asterisks indicate that the bacterial numbers were significantly different from the replication of the SCHU S4 strain at the 20-h time point. (*,P ≤ 0.05). (B) Culture supernatants of the infected J774 cells were assayed for LDH activity at the indicated time points, and the activity was expressed as a percentage of the level of noninfected lysed cells. Means and standard deviations of triplicate wells from one representative experiment of two are shown. The asterisks indicate that the cytotoxicity levels were significantly different from those of SCHU S4-infected cells for the 20-h time point, as determined by a t test (***, P ≤ 0.001).

FIG 5

Quantification of LAMP-1 colocalization with F. tularensis strain SCHU S4. J774 cells were infected for 1 h with the indicated F. tularensis strains at an MOI of 30 or latex beads at an MOI of 10, and after washing, they were further incubated for up to 18 h. Fixed samples were labeled for the late endosomal/lysosomal marker LAMP-1. (A) Percentages representing the fractions of F. tularensis- or latex bead-containing phagosomes stained for the late endosomal/lysosomal marker LAMP-1. (B) A separate analysis was performed for strain FSC043 at 18 h. LAMP-1 colocalization was determined for two populations of bacteria observed in infected host cells: (i) a majority (95%) of J774 cells with individual bacteria (designated PI) and (ii) a minority (5%) of J774 cells with clusters of replicating bacteria (designated PII). The results are expressed as mean values and SD from one representative experiment in which 100 bacteria each on triplicate coverslips were counted. The asterisks indicate colocalization levels significantly different from those of SCHU S4 for each time point. *, P < 0.05; ***, P < 0.001, according to a t test. Experiments were repeated 2 to 4 times for the 0- and 3-h time points and twice for the 18-h time point.

FIG 6

Colocalization of GFP-expressing F. tularensis strains and the late endosomal marker LAMP-1. J774 cells were infected for 1 h with the indicated F. tularensis SCHU S4 strain at an MOI of 30 or latex beads at an MOI of 10 and further incubated for 3 h (A) or 18 h (B). In the representative confocal images, the green channel shows bacteria or latex particles and the red channel shows LAMP-1 staining for the indicated strain or latex beads. Confocal images were acquired with the Nikon C1 confocal microscope and assembled using Adobe Photoshop CS4 (Adobe Systems, San Jose, CA).

FIG 7

Electron micrographs of J774 cells infected for 1 h with F. tularensis FSC043 or SCHU S4, the ΔiglC mutant, or the ΔiglC mutant and then further incubated for 18 h. (A) The ΔiglC strain. (B and C) A host cell containing an individual FSC043 bacterium enclosed by a phagosomal membrane (B) and a host cell containing a cluster of FSC043 bacteria without discernible phagosomal membranes (C). (D and E) A host cell containing an individual ΔpdpC mutant bacterium enclosed by a phagosomal membrane (D) and a host cell containing a cluster of ΔpdpC mutant bacteria without discernible phagosomal membranes (E). The electron micrographs were acquired with a JEOL 1200 EX-II electron microscope (JEOL Ltd., Tokyo, Japan) and assembled using Adobe Photoshop CS4.

FIG 8

Intracellular replication and cytotoxicity in BMMs of F. tularensis strains. (A) BMMs were infected by the indicated strains of F. tularensis at an MOI of 200 for 2 h. Upon gentamicin treatment, the cells were allowed to recover for 30 min, after which they were lysed immediately (0 h) or after 24 h and plated to determine the number of viable bacteria (log₁₀). All infections were repeated twice, with triplicate data sets, and a representative experiment is shown. Each bar represents the mean value, and the error bar indicates the standard deviation. The asterisks indicate that the log₁₀ number of CFU was significantly different from that of the FSC043 strain as determined by a 2-sided t test with equal variance (**, P ≤ 0.01; ***, P ≤ 0.001). (B) The cytotoxicity of the infected BMMs was determined using the LDH assay (*, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001). No bar is shown for the 48-h time point for SCHU S4, since all cells had lysed before that.