Binding and activation of host plasminogen on the surface of Francisella tularensis

Abstract

Background: Francisella tularensis (FT) is a gram-negative facultative intracellular coccobacillus and is the causal agent of a life-threatening zoonotic disease known as tularemia. Although FT preferentially infects phagocytic cells of the host, recent evidence suggests that a significant number of bacteria can be found extracellularly in the plasma fraction of the blood during active infection. This observation suggests that the interaction between FT and host plasma components may play an important role in survival and dissemination of the bacterium during the course of infection. Plasminogen (PLG) is a protein zymogen that is found in abundance in the blood of mammalian hosts. A number of both gram-positive and gram-negative bacterial pathogens have the ability to bind to PLG, giving them a survival advantage by increasing their ability to penetrate extracellular matrices and cross tissue barriers.

Results: We show that PLG binds to the surface of FT and that surface-bound PLG can be activated to plasmin in the presence of tissue PLG activator in vitro. In addition, using Far-Western blotting assays coupled with proteomic analyses of FT outer membrane preparations, we have identified several putative PLG-binding proteins of FT.

Conclusions: The ability of FT to acquire surface bound PLG that can be activated on its surface may be an important virulence mechanism that results in an increase in initial infectivity, survival, and/or dissemination of this bacterium in vivo.

Figure 1

FT binds to PLG from human plasma. FTLVS cultured to mid-log phase in BHI broth were bound to wells of microtiter plates and then incubated for 1 hour with fresh frozen human plasma (FFP) in the presence or absence of 100 mM ε-amino caproic acid (εACA), a PLG-binding inhibitor. A modified ELISA was performed to measure FTLVS-bound PLG. The results shown are representative of 3 experiments of similar design. Bars indicate +/- SEM in triplicate. Statistical analysis was performed via one-way ANOVA using a Dunnett's Multiple Comparison post-test (*** P < .001).

Figure 2

Purified huPLG binds to FTLVS and FTSchuS4. FTLVS (Panel A) and FTSchuS4 (Panel B) were bound to microtiter wells and incubated for 2 hours with purified huPLG (3 μg/ml) in the presence or absence of 10 mM εACA). A modified ELISA was performed to measure FTLVS-bound huPLG. The results shown are representative of four (Panel A) and one (Panel B) experiments, respectively, of similar design. Bars indicate +/- SEM in triplicate. Statistical analysis was performed via one-way ANOVA using a Dunnett's Multiple Comparison post-test (*** P < .001).

Figure 3

εACA inhibits huPLG binding to FT in a dose-dependent fashion. FTLVS was coated onto microtiter plate wells and incubated for 2 hours with purified huPLG (3 μg/mL) in the presence or absence of titrated concentrations of εACA. The results shown are representative of 3 experiments of similar design. Bars indicate +/- SEM in triplicate. Statistical analysis performed via one-way ANOVA using a Kruskal-Wallis test determined a p-value of < 0.0001.

Figure 4

PLG binds to the outer envelope of FT. Laser scanning confocal microscopy of PLG-associated FTLVS was performed as described in "Materials and Methods". Bound huPLG ligand was detected using sheep anti-human PLG antibody followed by incubation with Dylight-488 conjugated donkey, anti-sheep/goat IgG secondary antibody. Samples were visualized using a Zeiss LSM 510 confocal microscope.

Figure 5

FT surface-bound huPLG can be converted to plasmin. FTLVS was incubated with huPLG at a concentration of 96 μg/mL. After removal of unbound huPLG, a chromogenic plasmin substrate (D-VLK-pNA), tissue PLG activator (tPA), or both were then added to test the proteolytic ability of each sample preparation. Conversion of the chromogenic substrate was measured by comparison of Δ405 nm. The results shown are representative of 3 experiments of similar design. Bars indicate +/- SEM in triplicate. Statistical analysis was performed via one-way ANOVA using a Dunnett's Multiple Comparison post-test (*** P < .001).

Figure 6

Fibronectin is a substrate for plasmin bound to FT. FTLVS (10⁹ CFU) were incubated with 100 μg of huPLG and 0.5 μg tissue tPA for 1 hour at 37°C. After removal of unbound huPLG and tPA, 3 μg fibronectin was added and allowed to incubate for 24 hours at 37°C. Supernatant from each preparation were separated by SDS-PAGE and transferred to PVDF membrane. Degradation of fibronectin was detected by Western blot analysis as described in "Materials and Methods".

Figure 7

Identification of putative PLG-binding proteins of FT. Sarkosyl-soluble and insoluble protein fractions of FTLVS were separated by SDS-PAGE and transferred to PVDF membrane. Membranes were then blotted with huPLG (3 ug/mL) followed by anti-PLG antibody and HRP-conjugated secondary antibody to detect PLG-binding proteins (Panel A). Protein bands on an identical Coomassie Blue-stained SDS-PAGE gel corresponding to those identified via blotting (Panel B) were excised and identified using proteomic methodologies (Panel C).