Fig 1.
Unencapsulated, but not encapsulated, S. pneumoniae induces human platelet aggregation.
Platelet aggregation was measured by light transmission as a percentage to the transmission through PPP in a stirring cuvette. TRAP was used as a positive control; PBS as a negative control. Aggregation curves are depicted for stimulation with S. pneumoniae D39, ΔcpsD39, TIGR4 and 6303 (A) and rCPS2 (B). PRP was pre-incubated with α-TLR2, α-FcyRII, PGE1, abciximab or PBS prior to ΔcpsD39 stimulation in (C). Toll like receptor 2 and 4 expression on human platelets are shown in (D). Aggregation curves as a result of TLR agonist stimulation with LTA, Pam3CSK4 or LPS are shown in (E). All aggregation curves are representative of 3 independent experiments using different donors. HEK cells stably transfected with TLR2 and CD14 were pre-incubated with α-TLR2 and stimulated for 16 hours with LTA, Pam3-CSK4 and ΔcpsD39, IL-8 was measured in the supernatant (n = 4) (F). * P<0.05.
Fig 2.
Prestimulation with S. pneumoniae fails to modulate human platelet aggregation in response to subthreshold concentrations of TRAP.
Aggregation curves testing minimal TRAP concentration to induce aggregation are depicted in (A). Prior to stimulation with this subthreshold TRAP concentration, PRP was incubated with S. pneumoniae D39, ΔcpsD39, TIGR4 and 6303 for 10 minutes in a stirring cuvette (B). Aggregation curves are representative of 3 independent experiments using different donors.
Fig 3.
S. pneumoniae D39, ΔcpsD39, TIGR4 and 6303 all induce human platelet degranulation.
Whole blood was stimulated with S. pneumoniae D39, ΔcpsD39, TIGR4 or 6303. Following 30 minutes of incubation platelets were stained and analysed by flow cytometry for surface expression of CD62p (A) and CD63 (B). Percentages were determined using isotype control antibodies to set the gate. TRAP was used as a positive control and induced CD62p—and CD63 expression on 87% and 56% of platelets respectively; PBS induced CD62p—and CD63 expression on 10% and 2% of platelets. Histograms are representative of 2 independent experiments using different donors.
Fig 4.
TLR2, 4, FcyRII and GPIIb/IIIa are not involved in S. pneumoniae induced human platelet degranulation.
Following pre-incubation with blocking antibodies to TLR2, TLR4, or FcyRII, or GPIIb/IIIa (Abciximab (ABC)) or with PGE1, whole blood was stimulated with S. pneumoniae ΔcpsD39. Platelets were stained and analysed by flow cytometry for surface expression of CD62p (A). As an opposite approach, whole blood was incubated with TLR2 and 4 agonists LTA, Pam3CSK4 and LPS and analysed by flow cytometry for surface expression of CD62p (B) or CD63 (C). Histograms are representative of 2 independent experiments using different donors.
Fig 5.
Human whole blood S. pneumoniae incubation results in platelet-leukocyte complex formation.
Whole blood was stimulated with S. pneumoniae D39, ΔcpsD39, TIGR4 or 6303. Following 30 minutes of incubation leukocytes subsets were stained and analysed for surface expression of CD61. Percentages were determined using isotype control antibodies to set the gate. Neutrophil-CD61 is depicted in (A), monocyte-CD61 in (B) and lymphocyte-CD61 in (C). TRAP was used as a positive control and PBS as a negative control. Histograms are representative of 2 independent experiments using different donors. In a similar fashion, neutrophil-platelet (D) and monocyte-platelet (E) and lymphocyte-platelet (F) complex formation was analysed following stimulation with TLR agonists LTA, Pam3CSK4 and LPS.
Fig 6.
Wild-type mouse platelets respond to S. pneumoniae D39 and ΔcpsD39 in a similar manner as platelets from Tlr2-/-, Tlr4-/-, Tlr2/4-/-, Tlr9-/- and Myd88-/- mice.
Mouse wild-type, Tlr2-/-, Tlr4-/-, Tlr2/4-/-, Tlr9-/- and Myd88-/- whole blood was stimulated with S. pneumoniae D39 (A) or ΔcpsD39 (B). Following 30 minutes of incubation platelets were stained and analysed by flow cytometry for surface expression of CD62p. N = 2–3 mice per group; histograms are representatives for the mice genotypes.
Fig 7.
Platelet MyD88 is not involved in host defence to ΔcpsD39 in vivo.
Control (closed dots, grey bars) and Plt-Myd88-/- mice (open dots, white bars) were infected with S. pneumoniae ΔcpsD39 via the intranasal route and euthanized 16 hours thereafter. Bacterial counts were determined in lungs, blood, spleen and liver (A). Platelet counts (B) and platelet activation (CD62p; C) were determined by FACS analysis for CD61 and CD62p. PF4 (D), sP-selectin (E), TATc (F) and E-selectin (G) were measured in plasma using ELISA. Data are expressed as scatter dot plots or box- and whisker plots depicting the smallest observation, lower quartile, median, upper quartile and largest observation. N = 8 mice per group.
Fig 8.
Platelet MyD88 is not involved in the inflammatory response to ΔcpsD39 in vivo.
WT (gray bars) and Plt-Myd88-/- mice (open bars) were infected with S. pneumoniae ΔcpsD39 via the intranasal route and euthanized 16 hours thereafter. Lung cytokine levels of TNF-α (A), IL-1β (B), IL-6 (C), and KC (D) were measured by ELISA. Lung histopathology was scored by an independent pathologist; representative microphotographs are shown in (E; 10x magnification) and pathology scores in (F). Data are depicted as are expressed as box- and whisker plots depicting the smallest observation, lower quartile, median, upper quartile and largest observation. N = 8 mice per group.