Histidine-Rich Glycoprotein Modulates the Toxic Effects of High-Dose Polyphosphate in Mice.

BACKGROUND: Polyphosphate (polyP), a procoagulant released from platelets, activates coagulation via the contact system and modulates cardiomyocyte viability. High-dose intravenous polyP is lethal in mice, presumably because of thrombosis. Previously, we showed that HRG (histidine-rich glycoprotein) binds polyP and attenuates its procoagulant effects. In this study, we investigated the mechanisms responsible for the lethality of intravenous polyP in mice and the impact of HRG on this process. METHODS: The survival of wild-type or HRG-deficient mice given intravenous synthetic or platelet-derived polyP in doses up to 50 mg/kg or saline was compared. To determine the contribution of thrombosis, the effect of FXII (factor XII) knockdown or enoxaparin on polyP-induced fibrin deposition in the lungs was examined. To assess cardiotoxicity, the ECG was continuously monitored, the levels of troponin I and the myocardial band of creatine kinase were quantified, and the viability of a cultured murine cardiomyocyte cell line exposed to polyP in the absence or presence of HRG was determined. RESULTS: In HRG-deficient mice, polyP was lethal at 30 mg/kg, whereas it was lethal in wild-type mice at 50 mg/kg. Although FXII knockdown or enoxaparin administration attenuated polyP-induced fibrin deposition in the lungs, neither affected mortality. PolyP induced dose-dependent ECG abnormalities, including heart block and ST-segment changes, and increased the levels of troponin and myocardial band of creatine kinase, effects that were more pronounced in HRG-deficient mice than in wild-type mice and were attenuated when HRG-deficient mice were given supplemental HRG. Consistent with its cardiotoxicity, polyP reduced the viability of cultured cardiomyocytes in a dose-dependent manner, an effect attenuated with supplemental HRG. CONCLUSIONS: High-dose intravenous polyP is cardiotoxic in mice, and HRG modulates this effect.

PAR4 Inhibition Reduces Coronary Artery Atherosclerosis and Myocardial Fibrosis in SR-B1/LDLR Double Knockout Mice.

BACKGROUND: SR-B1 (scavenger receptor class B type 1)/LDLR (low-density lipoprotein receptor) double knockout mice fed a high-fat, high-cholesterol diet containing cholate exhibit coronary artery disease characterized by occlusive coronary artery atherosclerosis, platelet accumulation in coronary arteries, and myocardial fibrosis. Platelets are involved in atherosclerosis development, and PAR (protease-activated receptor) 4 has a prominent role in platelet function in mice. However, the role of PAR4 on coronary artery disease in mice has not been tested. METHODS: We tested the effects of a PAR4 inhibitory pepducin (RAG8) on diet-induced aortic sinus and coronary artery atherosclerosis, platelet accumulation in atherosclerotic coronary arteries, and myocardial fibrosis in SR-B1/LDLR double knockout mice. SR-B1/LDLR double knockout mice were fed a high-fat, high-cholesterol diet containing cholate and injected daily with 20 mg/kg of either the RAG8 pepducin or a control reverse-sequence pepducin (SRQ8) for 20 days. RESULTS: Platelets from the RAG8-treated mice exhibited reduced thrombin and PAR4 agonist peptide-mediated activation compared with those from control SRQ8-treated mice when tested ex vivo. Although aortic sinus atherosclerosis levels did not differ, RAG8-treated mice exhibited reduced coronary artery atherosclerosis, reduced platelet accumulation in atherosclerotic coronary arteries, and reduced myocardial fibrosis. These protective effects were not accompanied by changes in circulating lipids, inflammatory cytokines, or immune cells. However, RAG8-treated mice exhibited reduced VCAM-1 (vascular cell adhesion molecule 1) protein levels in nonatherosclerotic coronary artery cross sections and reduced leukocyte accumulation in atherosclerotic coronary artery cross sections compared with those from SRQ8-treated mice. CONCLUSIONS: The PAR4 inhibitory RAG8 pepducin reduced coronary artery atherosclerosis and myocardial fibrosis in SR-B1/LDLR double knockout mice fed a high-fat, high-cholesterol diet containing cholate. Furthermore, RAG8 reduced VCAM-1 in nonatherosclerotic coronary arteries and reduced leukocyte and platelet accumulation in atherosclerotic coronary arteries. These findings identify PAR4 as an attractive target in reducing coronary artery disease development, and the use of RAG8 may potentially be beneficial in cardiovascular disease.

Histidine-rich glycoprotein attenuates catheter thrombosis.

Factor XII (FXII) knockdown attenuates catheter thrombosis in rabbits. Because histidine-rich glycoprotein (HRG) modulates FXIIa activity, we hypothesized that HRG depletion would promote catheter thrombosis. To test this, rabbits were given either antisense oligonucleotides (ASOs) against HRG or FXII, a control ASO, or saline. The activated partial thromboplastin time (aPTT), prothrombin time (PT), and catheter-induced thrombin generation were determined in blood collected before and after treatment. Compared with the controls, the HRG- and FXII-directed ASOs reduced hepatic messenger RNA and plasma levels of HRG and FXII, respectively, by >90%. Although HRG knockdown shortened the aPTT by 2.5 fold, FXII knockdown prolonged it by fourfold; neither of the ASOs affected the PT. Catheter segments shortened the lag time and increased peak thrombin in the plasma from control rabbits; effects were significantly enhanced and attenuated in the plasma from rabbits given the HRG- and FXII-directed ASOs, respectively. Catheters were then inserted into the right external jugular vein of the rabbits, and the time for catheter occlusion was determined. The catheter occlusion times with the control ASO or saline were 62 ± 8 minutes and 60 ± 11 minutes, respectively. The occlusion time was significantly reduced to 34 ± 9 minutes, with HRG knockdown and significantly prolonged to 128 ± 19 minutes with FXII knockdown. HRG levels are decreased with sepsis or cancer, and such patients are prone to catheter thrombosis. Because HRG modulates catheter thrombosis, our findings suggest that HRG supplementation may prevent this problem.

Allosteric modulation of exosite 1 attenuates polyphosphate-catalyzed activation of factor XI by thrombin.

BACKGROUND: Polyphosphate (polyP) promotes feedback activation of factor (F) XI by thrombin by serving as a template. The contribution of thrombin's exosites to these interactions is unclear. OBJECTIVES: To determine the contribution of thrombin exosites 1 and 2 to polyP-induced potentiation of FXI activation by thrombin. METHODS: The affinities of α-thrombin; K109E/110E-thrombin, an exosite 1 variant, or R93E-thrombin, an exosite 2 variant; FXI; and FXIa for polyP-70 were quantified using surface plasmon resonance in the absence or presence of exosite ligands. FXI was activated with α-thrombin or thrombin variants in the absence or presence of polyP-70 and exosite ligands. RESULTS: α-Thrombin, K109/110E-thrombin, FXI, and FXIa bound polyP-70, whereas R93E-thrombin exhibited minimal binding. Exosite 1 and exosite 2 ligands attenuated thrombin binding to polyP-70. PolyP-70 accelerated the rate of FXI activation by α-thrombin and K109E/110E-thrombin but not R93E-thrombin up to 1500-fold in a bell-shaped, concentration-responsive manner. Exosite 1 and exosite 2 ligands had no impact on FXI activation by thrombin in the absence of polyP-70; however, in its presence, they attenuated activation by 40% to 65%. CONCLUSION: PolyP-70 binds FXI and thrombin and promotes their interaction. Exosite 2 ligands attenuate activation because thrombin binds polyP-70 via exosite 2. Attenuation of FXI activation by exosite 1 ligands likely reflects allosteric modulation of exosite 2 and/or the active site of thrombin because exosite 1 is not directly involved in FXI activation. Therefore, allosteric modulation of thrombin's exosites may represent a novel strategy for downregulating FXI activation.

Identification of the histidine-rich glycoprotein domains responsible for contact pathway inhibition.

BACKGROUND: Previously, we showed that histidine-rich glycoprotein (HRG) binds factor (F) XIIa with high affinity, inhibits FXII autoactivation and FXIIa-mediated activation of FXI, and attenuates ferric chloride-induced arterial thrombosis in mice. Therefore, HRG downregulates the contact pathway in vitro and in vivo. OBJECTIVE: To identify the domains on HRG responsible for contact pathway inhibition. METHODS: Recombinant HRG domain constructs (N-terminal [N1, N2, and N1N2], proline-rich regions, histidine-rich region [HRR], and C-terminal) were expressed and purified. The affinities of plasma-derived HRG, HRG domain constructs, and synthetic HRR peptides for FXII, FXIIa, ß-FXIIa, and polyphosphate (polyP) were determined using surface plasmon resonance, and their effects on polyP-induced FXII autoactivation, FXIIa-mediated activation of FXI and prekallikrein, the activated partial thromboplastin time (APTT), and thrombin generation were examined. RESULTS: HRG and HRG domain constructs bind FXIIa, but not FXII or ß-FXII. HRR, N1, and N1N2 bind FXIIa with affinities comparable with that of HRG, whereas the remaining domains bind with lower affinity. Synthetic HRR peptides bind FXIIa and polyP with high affinity. HRG and HRR significantly inhibit FXII autoactivation and prolong the APTT. Like HRG, synthetic HRR peptides inhibit FXII autoactivation, attenuate FXIIa-mediated activation of prekallikrein and FXI, prolong the APTT, and attenuate thrombin generation. CONCLUSION: The interaction of HRG with FXIIa and polyP is predominantly mediated by the HRR domain. Like intact HRG, HRR downregulates the contact pathway and contributes to HRG-mediated down regulation of coagulation.

Polyphosphate-induced thrombosis in mice is factor XII dependent and is attenuated by histidine-rich glycoprotein.

Histidine-rich glycoprotein (HRG) is an abundant plasma protein that binds factor XIIa (FXIIa) and inhibits factor XII (FXII) autoactivation and FXIIa-mediated activation of FXI. Polyphosphate (polyP), a potent procoagulant released from activated platelets, may serve as a physiological activator of the contact system. Previously, we showed that HRG binds DNA and neutralizes its procoagulant activity. Consequently, our goal was to determine whether the capacity of HRG to bind polyanions enables it to regulate polyP-induced thrombosis. In a plate-based assay, immobilized polyP bound HRG, FXII, and FXIIa in a zinc-dependent manner. Basal and polyP-induced thrombin generation was greater in plasma from HRG-deficient mice than in plasma from wild-type mice. Intraperitoneal injection of polyP shortened the activated partial thromboplastin time, enhanced thrombin generation, increased thrombin-antithrombin levels, reduced lung perfusion, and promoted pulmonary fibrin deposition to a greater extent in HRG-deficient mice than in wild-type mice, effects that were abrogated with FXII knockdown. HRG thus attenuates the procoagulant and prothrombotic effects of polyP in an FXII-dependent manner by modulating the contact system.