Bacteria differentially induce degradation of Bcl-xL, a survival protein, by human platelets

Abstract

Bacteria can enter the bloodstream in response to infectious insults. Bacteremia elicits several immune and clinical complications, including thrombocytopenia. A primary cause of thrombocytopenia is shortened survival of platelets. We demonstrate that pathogenic bacteria induce apoptotic events in platelets that include calpain-mediated degradation of Bcl-x(L), an essential regulator of platelet survival. Specifically, bloodstream bacterial isolates from patients with sepsis induce lateral condensation of actin, impair mitochondrial membrane potential, and degrade Bcl-x(L) protein in platelets. Bcl-x(L) protein degradation is enhanced when platelets are exposed to pathogenic Escherichia coli that produce the pore-forming toxin α-hemolysin, a response that is markedly attenuated when the gene is deleted from E coli. We also found that nonpathogenic E coli gain degrading activity when they are forced to express α-hemolysin. Like α-hemolysin, purified α-toxin readily degrades Bcl-x(L) protein in platelets, as do clinical Staphylococcus aureus isolates that produce α-toxin. Inhibition of calpain activity, but not the proteasome, rescues Bcl-x(L) protein degradation in platelets coincubated with pathogenic E coli including α-hemolysin producing strains. This is the first evidence that pathogenic bacteria can trigger activation of the platelet intrinsic apoptosis program and our results suggest a new mechanism by which bacterial pathogens might cause thrombocytopenia in patients with bloodstream infections.

Figure 1

E coli induces apoptotic-like features in platelets. (A) Platelets were incubated for 8 hours alone (left) or with E coli (EC-SP), an isolate from the bloodstream of a patient diagnosed with sepsis (middle and right). Left and middle: Confocal microscopy of polymerized actin (phalloidin, green stain) and DNA (topro 3, magenta stain) in platelets and EC-SP (E), respectively. Far right: Transmission electron micrograph of the platelet and EC-SP (E) suspension cultures. The white arrows point to lateral condensation of polymerized actin, whereas the magenta arrows (right) identify granules that have moved to one side of the cell. Scale bars = 2 μm in the left and middle, 1 μm in the right. These micrographs are representative of 3 independent experiments. (B) Mitochondrial membrane potential (ΔΨ_m) in platelets or EC-SP that were cultured (8 hours) alone (left and middle) or together (right). Bacteria population in middle and right panels is indicated by green box. This flow cytometric experiment is representative of 3 independent studies. The green box identifies the major population of EC-SP.

Figure 2

EC-SP degrades Bcl-x_L protein in platelets. (A) Western blot analysis of Bcl-x_L protein expression in platelets incubated with EC-SP (4 or 8 hours). (B) Western blot analysis of platelets stimulated with vehicle (control), thrombin (Thr), PAF, or A23187 (1μM) for 8 hours. The Western blots in panels A and B are representative of 2-4 independent experiments for each experimental group.

Figure 3

E coli that express HlyA invoke Bcl-x_L protein degradation in platelets. (A) Platelets were incubated with wild-type UTI89 or UTI89 ΔhlyA bacteria for 4 or 8 hours and Bcl-x_L and actin protein were assessed by Western blot analysis. (B) Platelets were incubated with wild-type UTI89 or UTI89 ΔhlyA bacteria for 4 or 8 hours and intracellular Bcl-x_L protein was quantified by ELISA. The bars in the graph are the mean ± SEM (n = 3) and the single asterisk indicates a significant decrease (P < .01) in Bcl-x_L protein levels compared with all other conditions. (C) Platelets were incubated with WAM582 or WAM783 for 2, 4, or 8 hours, and Bcl-x_L protein was assessed by ELISA. The bars in the graph are the mean ± SEM (n = 3), and the asterisk indicates a significant decrease (P < .01) in Bcl-x_L protein levels in WAM582 treated platelets compared with all other conditions. (D) Platelets were incubated with wild-type UTI89, UTI89 ΔhlyA, WAM783, WAM582, and S aureus isolated from the bloodstream of a patient diagnosed with sepsis (SA-SP), α-toxin (500 ng/mL), or thrombin (1 U/mL). After 8 hours, protein for Bcl-x_L and actin were assessed by Western blot analysis. The Western blots in this figure are representative of 2 independent experiments for each group.

Figure 4

Inhibition of calpain activity rescues Bcl-x_L degradation in platelets. (A-B) Platelets were stimulated with increasing concentrations of A23187 in the presence or absence of MG132 (10μM). After 1 hour, Bcl-x_L and actin protein were assessed by Western blot analysis (A) or intracellular Bcl-x_L protein was measured by ELISA (B). The bars in panel B represent the mean ± SEM of 4 independent experiments, and the asterisk indicates a significant decrease (P < .01) in Bcl-x_L protein levels compared with control and MG132-treated platelets. (C) Platelets were pretreated with or without MG132 and then incubated alone (control) or with UTI89 for 2 hours. Bcl-x_L and actin protein in this blot and panel A are representative of 3 independent experiments.

Figure 5

Inhibition of calpain activity rescues E coli–induced Bcl-x_L degradation in platelets. (A-B) Platelets were left alone or pretreated with specific inhibitors (10μM epoxomicin [EPO]; 25μM lactacystin [Lact]; 10μM MG132; 50μM calpeptin [Calp]; 50μM E64d; and 50μM MDL28170 [MDL]) before being stimulated with A23187 (1μM) for 1 hour and intracellular Bcl-x_L protein was assessed by Western analysis (A) or quantified by ELISA (B). The bars for panel B show the mean ± SEM of 4 independent experiments, and the asterisk indicates a significant decrease (P < .01) in Bcl-x_L protein levels compared with control. (C-D) Platelets were pretreated with inhibitors (A and B) and then left alone (control) or incubated with wild-type UTI89 for 4 hours. Intracellular Bcl-x_L protein was subsequently assessed by Western analysis (C) or quantified by ELISA (D). The bars for panel D depict the mean ± SEM of 4 independent experiments, and the single asterisk indicates a significant decrease (P < .01) in Bcl-x_L protein levels compared with control. (E) Platelets were incubated in the presence or absence of EGTA (5mM) for 15 minutes and then left alone or stimulated with 1μM A23187, UTI89, or UTI89 ΔhlyA for 2 hours and Bcl-x_L protein was assessed by Western analysis. This blot is representative of 3 independent experiments. (F) Whole blood was incubated with UTI89 in the presence or absence of calpeptin. After 4 hours, platelets were isolated, and intracellular Bcl-x_L protein was measured by ELISA. The bars represent the mean ± SEM of 3 independent experiments, and the single asterisk indicates a significant decrease (P < .05) in Bcl-x_L protein levels compared with control.