Reduced thrombin generation increases host susceptibility to group A streptococcal infection

Abstract

Bacterial plasminogen activators are commonplace among microbial pathogens, implying a central role of host plasmin in supporting bacterial virulence. Group A streptococci (GAS) secrete streptokinase, a specific activator of human plasminogen (PLG). The critical contribution of the streptokinase-PLG interaction to GAS pathogenicity was recently demonstrated using mice expressing human PLG. To examine the importance of thrombin generation in antimicrobial host defense, we challenged mice with deficiency of factor V (FV) in either the plasma or platelet compartment. Reduction of FV in either pool resulted in markedly increased mortality after GAS infection, with comparison to heterozygous F5-deficient mice suggesting a previously unappreciated role for the platelet FV pool in host defense. Mice with complete deficiency of fibrinogen also demonstrated markedly increased mortality to GAS infection relative to controls. Although FV Leiden may be protective in the setting of severe sepsis in humans, no significant survival advantage was observed in GAS-infected mice carrying the FV Leiden mutation. Taken together, our data support the hypothesis that local thrombosis/fibrin deposition limits the survival and dissemination of at least a subset of microbial pathogens and suggest that common variation in hemostatic factors among humans could affect host susceptibility to a variety of infectious diseases.

Figure 1

Control mice of the FV-transgenic lines inoculated with GAS strain UMAA2616. F5^+/−, Tg⁺F5^+/−, Tg⁺F5^+/+, and F5^+/+ mice for all 3 FV-transgenic lines were infected with UMAA2616 at doses of 3.6 × 10⁶ to 2.6 × 10⁸ CFU. Data were pooled from 7 independent experiments using littermates, constituting a total of 18 F5^+/−, 27 Tg⁺F5^+/−, 3 Tg⁺F5^+/+, and 7 F5^+/+ mice.

Figure 2

AlbfvB mice inoculated with GAS strain UMAA2616. P values for differences between the survival curves are indicated in all panels. The solid line represents Tg⁺F5^−/−; dashed lines, F5⁺ littermate controls. (A) Inoculum of 3.6 × 10⁶ CFU. Data were pooled from 4 independent experiments (totaling 17 AlbfvBTg⁺F5^−/− mice and 32 littermate F5⁺ controls), at doses ranging from 1.2 × 10⁶ to 4.9 × 10⁶ CFU. (B) Inoculum of 2.5 × 10⁷ CFU. Data were pooled from 2 independent experiments (totaling 6 AlbfvBTg⁺F5^−/− mice and 18 littermate F5⁺ controls) at doses of 1.4 × 10⁷ and 3.6 × 10⁷ CFU. (C) Inoculum of 2.6 × 10⁸ CFU. Data were from a single experiment with 5 AlbfvBTg⁺F5^−/− mice and 5 F5⁺ littermate controls. (D) Pooled data from all the experiments shown in panels A to C (a total of 28 AlbfvBTg⁺F5^−/− mice and 55 F5⁺ littermate controls).

Figure 3

AlbfvC and Pf4fvA mice inoculated with GAS strain UMAA2616. P values for survival differences are indicated in each panel. — represents Tg⁺F5^−/−; ----, F5⁺ littermate controls. (A) AlbfvC mice at an inoculum of approximately 4 × 10⁶ CFU. Data were pooled from 2 independent experiments (totaling 17 AlbfvCTg⁺F5^−/− mice and 32 littermate F5⁺ controls), at doses of 3.7 × 10⁶ and 4.4 × 10⁶ CFU, respectively. (B) AlbfvC mice at an inoculum of 3.6 × 10⁷ CFU. Data were derived from 1 experiment (4 AlbfvCTg⁺F5^−/− mice and 22 littermate F5⁺ controls). (C) AlbfvC mice at an inoculum of 2.6 × 10⁸ CFU. Data were derived from 1 experiment (4 AlbfvCTg⁺F5^−/− mice and 9 littermate F5⁺controls). (D) Pooled data from all experiments (25 AlbfvCTg⁺F5^−/− mice and 63 littermate F5⁺ controls). (E) Comparison of the AlbfvC and AlbfvB lines (combined from Figures 3D and 2D). P value of the difference between Tg⁺F5^−/− mice from AlbfvC and AlbfvB lines is indicated. (F) Pf4fvA mice at an inoculum of 1.4 × 10⁷ CFU. Data were derived from 1 experiment (6 Pf4fvATg⁺F5^−/− mice and 25 littermate F5⁺ controls).

Figure 4

Fga^−/− and Fga^+/− mice infected with UMAA2616. P values of survival difference are indicated in all panels. — represents Fga^−/−; ----, Fga^+/− littermate controls. (A) Five Fga^−/− mice and 5 sibling control Fga^+/− mice were infected with 7 × 10⁷ CFU UMAA2616. (B) Five Fga^−/− mice and 5 sibling control Fga^+/− mice were infected with 10⁶ CFU UMAA2616. (C) Ten Fibγ^390-396A and wild-type sibling controls, 9 Fga^−/−, and 10 Fga^+/− sibling control were infected with 2.8 × 10⁶ CFU UMAA2616.

Figure 5

GAS systemic spread in Fga^−/− and Fga^+/− mice infected with UMAA2616. Bacterial counts from spleens of 8 Fga^−/− and 9 Fga^+/− mice inoculated with 2 × 10⁶ CFU UMAA2616 were calculated by a plating method. Data were pooled from 2 independent experiments (2.8 × 10⁶ and 2.0 × 10⁶ CFU). The difference of CFU between Fga^−/− and Fga^+/− mice was tested by the Mann-Whitney test. P value is indicated.

Figure 6

Fga^+/− and Fga^+/+ mice infected with UMAA2616. P values of survival difference are indicated in all panels. — represents Fga^+/−; ----, Fga^+/+ littermate controls. (A) Fga^+/− mice and sibling control Fga^+/+ mice were infected with 2 × 10⁷ to 10⁸ CFU UMAA2616. Data were pooled from 4 independent experiments totaling 38 Fga^+/− mice and 33 Fga^+/+ mice. (B) Fga^+/− mice and sibling control Fga^+/+ mice were inoculated with 1.4 × 10⁹ CFU UMAA2616, totaling 17 Fga^+/− mice and 22 Fga^+/+ mice in 1 experiment.

Figure 7

FV Leiden mice inoculated with GAS strain UMAA2616. Data were pooled from 11 independent experiments with UMAA2616 doses ranging from 10⁶ to 10⁸ CFU. The total number of mice in each group are 68 hPLG⁻F5^+/+, 74 hPLG⁺F5^+/+, 61 hPLG⁻F5^Q/+, and 57 hPLG⁺F5^Q/+. P values for survival difference are indicated. Among the P values indicated are difference between hPLG⁻F5^+/+ and hPLG⁻F5^Q/+ (P > .26), difference between combined hPLG⁻ (hPLG⁻F5^+/+ and hPLG⁻F5^Q/+) and combined hPLG⁺ (hPLG⁺F5^+/+ and hPLG⁺F5^Q/+) (P < 5.5 × 10⁻⁹), and difference between hPLG⁺F5^Q/+ and hPLG⁺F5^+/+ (P > .52).