Platelet polyphosphates are proinflammatory and procoagulant mediators in vivo

Abstract

Platelets play a central role in thrombosis, hemostasis, and inflammation. We show that activated platelets release inorganic polyphosphate (polyP), a polymer of 60-100 phosphate residues that directly bound to and activated the plasma protease factor XII. PolyP-driven factor XII activation triggered release of the inflammatory mediator bradykinin by plasma kallikrein-mediated kininogen processing. PolyP increased vascular permeability and induced fluid extravasation in skin microvessels of mice. Mice deficient in factor XII or bradykinin receptors were resistant to polyP-induced leakage. PolyP initiated clotting of plasma via the contact pathway. Ablation of intrinsic coagulation pathway proteases factor XII and factor XI protected mice from polyP-triggered lethal pulmonary embolism. Targeting polyP with phosphatases interfered with procoagulant activity of activated platelets and blocked platelet-induced thrombosis in mice. Addition of polyP restored defective plasma clotting of Hermansky-Pudlak Syndrome patients, who lack platelet polyP. The data identify polyP as a new class of mediator having fundamental roles in platelet-driven proinflammatory and procoagulant disorders.

Figure 1. Activated platelets secrete long-chain polyP

Washed platelets (3.8×10¹³ platelets each) were stimulated with ADP (0.1 mM), Trap6 (0.2 mM), collagen (0.2 mg/ml), thrombin (2 U/ml), or buffer (“w/o”) and polyP was isolated from the supernatants. PolyP were separated by agarose gel electrophoresis and stained with toluidine blue (A) or DAPI (B). Synthetic polyP with mean chain lengths of 20 (polyP₂₀), 70 (polyP₇₀) or 125 (polyP₁₂₅) phosphate units were loaded in lanes 1-3 as size standards. Purified material from thrombin-stimulated platelets was incubated with Psp (0.05 U/μg polyP; “thrombin + Psp”) prior to electrophoresis. Negatively charged chondroitin sulfate (CS) confirmed specificity of staining. A DNA ladder as size standard is indicated on the left. (C) ³¹P NMR spectrum of platelet polyP at 202.4 MHz. The terminal and internal phosphates of the polyP chain give signals at 10.2 (doublet, blue arrow) and 22.1 (red arrow) ppm, respectively; ratios of their intensities yielded a mean chain length of 80 for platelet polyP. Other phosphate species present are monophosphate (0.0 ppm, 5 % of total platelet released phosphate), diphosphate (-10.4 ppm, 13 %), cyclotriphosphate (-21.4 ppm, 10 %) and ADP (-11.2 ppm, 3 %).

Figure 2. PolyP triggers contact activation of factor XII

(A) Human plasma was incubated with 100 μg/ml polyP (▲) or kaolin (○). Platelet polyP was either untreated (0) or pretreated with Psp (0.05 U/μg polyP) for 10 (), 30 (), or 60 min (). Data are means ± SD, n=5. (B and C) Plasma-free activation of purified FXII (200 nM) was analyzed in the presence (B) or absence (C) of PK (0.5 nM). The mixture was stimulated with dextran sulfate (■; 1 μg/ml), polyP (▲; 5 μg/ml), or buffer (●). FXIIa concentrations were determined using S-2302 hydrolysis. Data are means ± SD, n=4. (D) Gel-mobility shift analyses of polyP binding to proteins. Purified FXI, FXII, PK, HK, antithrombin (AT) or albumin (ALB) were incubated with buffer (“w/o”) or with increasing concentrations of polyP₁₂5 (0 - 100 μg) for 10 min, after which mixtures were resolved on a 10 % Tris-Glycine gel under native conditions and stained for protein using Gel Code Blue. (E) Protein binding to immobilized polyP. PolyP (left) or albumin (ALB, right) was coupled to zirconia (zirconium dioxide) beads. FXII, FXI, PK, or HK (5 μg each) were incubated with the beads, after which the flow-through (“FT”) fraction was collected by centrifugation using mini spin columns. The beads were washed with a low salt buffer (50 mM NaCl; “LS”) followed by a high salt buffer (1 M NaCl; “HS”). Fractions were probed by Western blotting with appropriate antibodies. Starting material (“SM”) refers to the initial protein load added to the beads.

Figure 3. PolyP activates the kallikrein-kinin system

(A) PolyP-induced cleavage of FXII, PK and HK in plasma. Human plasma was incubated with increasing concentrations of polyP (0.5 - 1000 μg/ml) or buffer (“w/o”), and then subjected to SDS-PAGE and Western blotting. Samples were probed with antibodies against FXII, HK, and PK. (B) PolyP-induced bradykinin release in plasma. BK in polyP-incubated samples was quantified by ELISA. Data are means ± SD, n=4.

Figure 4. PolyP triggers contact system-mediated bradykinin-induced edema in mice

(A-F) Evans blue was intravenously infused as a tracer into the following: (A) WT mice, (B) B2R^-/-mice, (C) FXII^-/-mice, (D) C1INH^-/- mice, (E) PCK-infused WT mice (8 mg/kg body weight), and (F) Icat-administered WT mice (175 μg/kg body weight). Dorsal skin edema formation was induced by intradermal injection of 50 μl normal saline (NaCl), BK (100 μM), platelet polyP (10 μg), or synthetic polyP₁₂₅ (10 μg), respectively, and visualized by tracer extravasation after 30 min. (G) Evans blue tracer from skin welts was extracted and quantified. Extravasated tracer is plotted as fold-increase of saline-induced signal in WT mice, to control for inter-animal variability. Data are means ± SD, n=10. (H) Psp- (0.05 U/μg polyP, 30 min; “Psp”) or buffer-treated E. coli polyP (“w/o”, 2 μg/lane each) were separated by agarose gel electrophoresis and stained with toluidine blue. A DNA ladder as size standard is indicated on the left. (I, J) Murine plasma was incubated with Psp (“Psp”) or buffer-treated (“w/o”) E. coli polyP (1 μg/ml, 30 min) and was probed for FXII, FXI, PK zymogen forms, single-chain HK, and LK by Western blotting. ELISA determined BK in samples supplemented with polyP. Means ± SD, n=5. (K) Mortality associated with intraperitoneal injection of E. coli polyP (750 μg/g body weight) in WT, FXII^-/-, or B2R^-/- mice. For control, polyP was Psp digested prior to infusion into WT animals (“+Psp”). Animals alive in each group (n=10) 30 min after challenge, were considered as survivors.

Figure 5. PolyP activates the intrinsic pathway of coagulation in plasma

(A) PolyP-induced cleavage of FXII and FXI in plasma. Plasma was incubated for 30 min at 37°C with increasing concentrations of polyP (0.5 - 1000 μg/ml) and analyzed for zymogen forms of FXII and FXI by Western blotting. Untreated plasma (“w/o”) shows initial FXII and FXI levels. (B) Thrombin generation in plasma incubated with buffer alone (□), 100 μg/ml kaolin (○), or platelet polyP (; “polyP”), synthetic polyP₁₂₅ (▲), platelet polyP treated with Psp prior to addition of plasma (; “polyP+Psp”). Data are means ± SD, n=5. (C) Plasma clotting times. PolyP was added to normal plasma (●) or FXII-immunodepleted plasma (▲) at a range of concentrations (0 - 5 μg/ml), with clot formation quantified as the half-maximal change in turbidity at 405 nm. Data are means ± SD, n=3. (D) Recalcification clotting times in normal (white columns) and FXII-deficient human plasma (black columns) incubated with TF (1 pM), polyP (3 μg/ml), TF and polyP (1 pM and 3 μg/ml), or buffer (w/o). Data are means ± SD, n=5. (E) Recalcification clotting times triggered by polyP (0 - 2 μg/ml) in FXII-deficient plasma (FXII-def.) and normal human plasma (NP) supplemented with FVa (4 nM, ▼), αTFPI antibody (10 μg/ml, □), or PCI (30 μg/ml, ○). Means ± SD, n=3. (F-H) Real time thrombin generation: (F) PolyP (0; 1.2; 4.8 μg/ml) triggered thrombin formation in plasma preincubated with an extrinsic pathway-inhibitor FVIIai. PolyP (4.8 μg/ml) failed to generate thrombin in FVIIai treated FXII-deficient plasma (FXII-def.). (G) Thrombin generation triggered by TF in FXII-deficient (red), FXI-deficient (blue), and normal plasma (green) in the absence (dashed lines) or presence of polyP (solid lines, 4.8 μg/ml). (H) Maximum clot firmness assessed by thromboelastography: Clot firmness in polyP (5 μg/ml), CTI (100 μg/ml), or polyP plus CTI-treated whole blood relative to untreated samples (100 %). Data are means ± SD, n=5.

Figure 6. PolyP triggers thrombosis in vivo

(A and B) Survival times following polyP challenge. (A) Pulmonary thromboembolism was induced by i. v. infusion of platelet polyP (300 μg/g body weight) in WT mice, FXII^-/- mice, FXII^-/- mice reconstituted with human FXII (“hFXII”, 2 μg/g body weight), WT mice infused with FXIIa inhibitor CSL829 (15 μg/g body weight), FXI^-/- mice, FXII- and FXI-gene double-deficient (“FXII^-/-/FXI^-/-”) mice, or B2R^-/- mice. Mortality was assessed in each group (n=15); animals still alive 30 min after challenge, were considered survivors. (B) PolyP was Psp digested (0.05 U/μg polyP) prior to infusion into WT animals and survival was analyzed as in panel A. (C) Plasma supplemented with CSL829 (150 μg/ml) or buffer (“w/o”) was incubated with polyP (500 μg/ml) for 30 min and analyzed for zymogen FXII, FXI, PK and HK by Western blotting. (D) Hematoxylin and eosin stained sections of lungs of WT, FXII^-/-, CSL829 infused WT, and hFXII reconstituted FXII^-/- mice 30 min after polyP administration (bar = 100 μm). (E) Thrombi per visual field were counted at 10× magnification from sections such as those in panel D. Data are mean ± SD for 100 fields. (F) Accumulation of fibrin in lung tissue of polyP treated WT, FXII^-/-, CSL829 pretreated WT, and reconstituted FXII^-/- mice. Fibrin formation 30 min after polyP challenge was analyzed by Western blotting.

Figure 7. Targeting polyP blocks platelet procoagulant activity in vitro and thrombosis in vivo

(A and B) Recalcification clotting times were determined in platelet-rich (A) human or (B) WT mouse plasma stimulated with A23187 (5 μM) in the presence (+) or absence (-) of Psp (10 U/ml). Reductions in clotting times are given relative to untreated plasma. Data are means ± SD, n=6. (C) Recalcification clotting times were determined in Trap6 (30 μM) stimulated human platelet-rich plasma in the presence (+) or absence (-) of PGE1 (5 μM), dBcAMP (0.5 mM), and Psp (10 U/ml). Reductions in clotting times are given relative to untreated PRP. Data are means ± SD, n=6. (D) Mortality associated with i.v. injection of 0.7 μg/g body weight Trap6 in WT mice; WT mice i.v. infused with PGE1 (0.85 μg/g body weight “WT+PGE1”) or dBcAMP (100 μg/g body weight “WT+dBcAMP”); FXII^-/- mice; and WT mice injected i.v. with Psp (15 U/g body weight; “WT+Psp”) before Trap6 challenge. 13 of 15 WT mice died within 10 min of challenge. In contrast WT animals injected with PGE1 or dBcAMP, FXII^-/-, and Psp-treated WT mice were significantly protected from lethal pulmonary embolism induced by Trap6; 10/15, 10/15, 13/15, and 13/15, respectively, survived for >30 min each (WT+PGE1, WT+dBcAPMP, FXII^-/- and WT+Psp vs. WT p<0.01, n=15 per genotype). (E) PolyP “rescues” defective fibrin formation in Hermansky-Pudlak Syndrome patients. Human platelets were isolated from healthy blood donors (control) and from patients with Hermansky-Pudlak Syndrome (HPS) and stimulated for 10 min at 37°C with Trap6 (50 μM) and collagen (10 μg/ml). Recalcification clotting times in normal plasma on addition of either normal (control) or HPS platelets (HPS, 1.5 ×10⁷) in the absence (open bars) or presence (closed bars) of additional synthetic polyP₁₂₅ (10 μg/ml). The graph shown is representative of 3 different experiments performed on different HPS and normal individuals.