Polyphosphate, Zn2+ and high molecular weight kininogen modulate individual reactions of the contact pathway of blood clotting

Abstract

Background: Inorganic polyphosphate modulates the contact pathway of blood clotting, which is implicated in thrombosis and inflammation. Polyphosphate polymer lengths are highly variable, with shorter polymers (approximately 60-100 phosphates) secreted from human platelets, and longer polymers (up to thousands of phosphates) in microbes. We previously reported that optimal triggering of clotting via the contact pathway requires very long polyphosphates, although the impact of shorter polyphosphate polymers on individual proteolytic reactions of the contact pathway was not interrogated.

Objectives and methods: We conducted in vitro measurements of enzyme kinetics to investigate the ability of varying polyphosphate sizes, together with high molecular weight kininogen and Zn²⁺ , to mediate four individual proteolytic reactions of the contact pathway: factor XII autoactivation, factor XII activation by kallikrein, prekallikrein activation by factor XIIa, and prekallikrein autoactivation.

Results: The individual contact pathway reactions were differentially dependent on polyphosphate length. Very long-chain polyphosphate was required to support factor XII autoactivation, whereas platelet-size polyphosphate significantly accelerated the activation of factor XII by kallikrein, and the activation of prekallikrein by factor XIIa. Intriguingly, polyphosphate did not support prekallikrein autoactivation. We also report that high molecular weight kininogen was required only when kallikrein was the enzyme (ie, FXII activation by kallikrein), whereas Zn²⁺ was required only when FXII was the substrate (ie, FXII activation by either kallikrein or FXIIa). Activation of prekallikrein by FXIIa required neither Zn²⁺ nor high molecular weight kininogen.

Conclusions: Platelet polyphosphate and Zn²⁺ can promote subsets of the reactions of the contact pathway, with implications for a variety of disease states.

Keywords: blood coagulation factors; polyphosphates; prekallikrein; thrombosis; zinc.

FIGURE 1

The following numbered reactions that contribute to initiation of the contact pathway of blood clotting were examined in this study: 1, activation of FXII by FXIIa (FXII autoactivation); 2, activation of FXII by PKa; 3, activation of PK by PKa (PK autoactivation); and 4, activation of PK by FXIIa. Propagation of the clotting cascade happens when FXIIa activates FXI, leading ultimately to thrombin generation

FIGURE 2

The ability of polyP to accelerate the individual reactions of the contact pathway depends on its polymer length. In each panel, reactions were conducted without (◆) or with 10 μmol/L polyP (●,▼). Data points for size-fractionated polyP preparations (●) are plotted on the x-axes according to their polymer lengths, whereas data points for heterogeneous polyP₁₂₀₀ (▼) are plotted on the x-axes at the modal length of this preparation (1200 phosphate units). In each panel, a pink rectangle represents the approximate length range of platelet polyP (60–100 phosphates), and a yellow rectangle indicates the length range of microbial polyP. A, PolyP length requirement for FXII autoactivation. In an endpoint assay (stopped at 20 min), 100 nmol/L FXII was incubated with 5 μmol/L ZnCl₂ without polyP (◆) or with 10 μmol/L polyP (●,▼). HK was not included in this experiment. B, PolyP length requirement for FXII activation by PKa. In an endpoint assay (stopped at 4 min), 100 nmol/L FXII was incubated with 100 pmol/L PKa, 100 nmol/L HK and 10 μmol/L ZnCl₂, without polyP (◆) or with 10 μmol/L polyP (●,▼) C, PolyP length requirement for PK activation by FXIIa. In an endpoint assay (stopped at 3 min), 100 nmol/L PK was incubated with 100 pmol/L FXIIa and 10 μmol/L ZnCl₂, without polyP (◆) or with 10 μmol/L polyP (●,▼). HK was not included in this experiment. Data in all panels are mean ± SE (n ≥ 3)

FIGURE 3

Zn²⁺ differentially influences the ability of polyP to accelerate individual reactions of the contact pathway. A, Zn²⁺ requirement for polyP-mediated FXII autoactivation. In an endpoint assay (stopped at 20 min), 100 nmol/L FXII without HK was incubated with 10 μmol/L polyP₁₂₀₀ in the presence of either 2 mmol/L EDTA (●) or varying ZnCl₂ concentrations (◇). B, Zn²⁺ requirement for polyP-mediated FXII activation by PKa. In an endpoint assay (stopped at 4 min), 100 nmol/L FXII and 100 pmol/L PKa were incubated with 100 nmol/L HK and 10 μmol/L polyP₁₂₀₀ in the presence of either 2 mmol/L EDTA (●) or varying ZnCl₂ concentrations (◇). C, Zn²⁺ requirement for polyP-mediated PK activation by FXIIa. In an endpoint assay (stopped at 3 min), 100 nmol/L PK and 100 pmol/L FXIIa without HK were incubated with 10 μmol/L polyP₁₂₀₀ in divalent metal ion-depleted buffer to which varying ZnCl₂ concentrations were added. D, EDTA inhibits polyP-mediated PK activation by FXIIa in a concentration-dependent manner. In an endpoint assay (stopped at 3 min), 100 nmol/L PK and 100 pmol/L FXIIa without HK were incubated with 10 μmol/L polyP₁₂₀₀ in the presence of varying EDTA concentrations. Data in all panels are mean ± SE (n ≥ 3)

FIGURE 4

HK differentially influences the ability of polyP to accelerate individual reactions of the contact pathway. A, Influence of HK on polyP-mediated FXII autoactivation. FXIIa levels were measured as a function of time after incubating 100 nmol/L FXII with 10 μmol/L polyP₁₂₀₀ in the presence of 5 μmol/L ZnCl₂ and 0 (●), 50 (Δ), 100 (■), 200 (▼), 300 (○), 500 (□), or 900 (▲) nmol/L HK. B-C, Calculation of second-order rate constants of polyP-mediated FXII autoactivation at various HK concentrations. Data from panel A (time course of FXII autoactivation) were replotted in panel B as ln(FXII/FXIIa) values vs time, to which lines were fitted by linear regression. Second-order rate constants, k₂, were derived from the slopes of these lines as described in Experimental Procedures, and plotted in panel C as a function of HK concentration. D, Influence of HK on polyP-mediated FXII activation by PKa. FXIIa levels were measured as a function of time after incubating 100 nmol/L FXII and 100 pmol/L PKa with 10 μmol/L polyP₁₂₀₀ in the presence of 10 μmol/L ZnCl₂ and varying HK concentrations. Initial FXII activation rates are plotted as nmol/L FXII activated per min divided by nmol/L PKa used. E, Influence of HK on polyP-mediated PK activation by FXIIa. PKa levels were measured as a function of time after incubating 100 nmol/L PK and 100 pmol/L FXIIa with 10 μmol/L polyP₁₂₀₀ in the presence of 10 μmol/L ZnCl₂ and varying HK concentrations. Initial PK activation rates were plotted as nmol/L PK activated per min divided by nmol/L FXIIa used. In panel C-E, blue rectangles represent the approximate concentration range of total HK in plasma. Data in all panels are mean ± SE (n ≥ 3)

FIGURE 5

Activation of PK in the presence of polyP and HK appears to be due to contamination of HK with traces of FXII(a), rather than PK autoactivation. A, Apparent PK autoactivation was inhibited by CTI. PKa levels were measured as a function of time after incubating 100 nmol/L PK with 100 nmol/L HK, 10 μmol/L polyP₁₂₀₀ and 10 μmol/L ZnCl₂ in presence of 0 (●), 0.5 (◆) or 2.5 (▲) μmol/L CTI. Data are mean ± SE (n ≥ 2). B, In the presence of an inhibitory anti-FXII antibody, activation of PK-S559A was undetectable by western blot. Left panel: 200 nmol/L PK-S559A without HK was incubated with or without 70 μmol/L polyP₇₀₀ in the presence of 10 μmol/L ZnCl₂ and 50 nmol/L anti-FXII IgG (1B2). At indicated times, samples were removed into reducing SDS sample buffer. Proteins were resolved by SDS-PAGE, followed by western blot analysis with a polyclonal antibody to human PK. Right panel: Parallel time courses of the same reaction conditions as in the left panel, but performed in the presence of 200 nmol/L HK. In both panels, t = 0 represents the reactants without polyP or HK. C, In the presence of an inhibitory anti-FXII antibody, cleavage of PK-S559A by added PKa was undetectable by western blot, in the presence or absence of polyP. Left panel: 200 nmol/L PK-S559A and 25 nmol/L purified PKa were incubated without polyP or HK, in the presence of 10 μmol/L ZnCl₂ and 50 nmol/L anti-FXII IgG (1B2). Samples were analyzed by western blot with the same polyclonal antibody to human PK as in panel B. Right panel: Parallel time courses of the reaction performed in the presence of 70 μmol/L long-chain polyP₇₀₀, with or without 200 nmol/L HK. In both panels, t = 0 represents the reaction without HK, polyP or PKa. PKa alone is resolved on the gels in the both panels for comparison purposes. Blots are representative of two independent experiments