Selective depletion of plasma prekallikrein or coagulation factor XII inhibits thrombosis in mice without increased risk of bleeding

Abstract

Recent studies indicate that the plasma contact system plays an important role in thrombosis, despite being dispensable for hemostasis. For example, mice deficient in coagulation factor XII (fXII) are protected from arterial thrombosis and cerebral ischemia-reperfusion injury. We demonstrate that selective reduction of prekallikrein (PKK), another member of the contact system, using antisense oligonucleotide (ASO) technology results in an antithrombotic phenotype in mice. The effects of PKK deficiency were compared with those of fXII deficiency produced by specific ASO-mediated reduction of fXII. Mice with reduced PKK had ∼ 3-fold higher plasma levels of fXII, and reduced levels of fXIIa-serpin complexes, consistent with fXII being a substrate for activated PKK in vivo. PKK or fXII deficiency reduced thrombus formation in both arterial and venous thrombosis models, without an apparent effect on hemostasis. The amount of reduction of PKK and fXII required to produce an antithrombotic effect differed between venous and arterial models, suggesting that these factors may regulate thrombus formation by distinct mechanisms. Our results support the concept that fXII and PKK play important and perhaps nonredundant roles in pathogenic thrombus propagation, and highlight a novel, specific and safe pharmaceutical approach to target these contact system proteases.

Figure 1

Effect of PKK and fXII ASO treatment on PKK and fXII hepatic mRNA and plasma protein levels. Male BALB/c mice were treated subcutaneously with PKK or fXII ASO for 3 weeks at indicated doses (n = 8 per treatment group). Two days after final dosing mice were collected for quantification of PKK and fXII hepatic mRNA levels (A) and PKK and fXII immunoblot analysis on pooled (B) and individual (C-D) samples. PKK and fXII relative plasma protein levels were quantified by densitometry. *P ≤ .05, **P ≤ .01 and ***P ≤ .001 compared with untreated control.

Figure 2

Effect of PKK and fXII ASOs on PKK and fXII activation in vivo. Pooled plasma samples from PKK, fXII or Control (Ctrl) ASO treated male C57BL/6 mice (3 weeks at 80 mg/kg/wk dose, n = 8 per treatment group) were subjected to 2-color immunoblot analysis of PKK (A, bottom row, shown in red) and fXII (A, top row, shown in red) complexes with α2-antiplasmin (A2AP, shown in green). PKK (A, bottom row) and fXII (A, top row) proteins are marked by asterisks. Complexes of PKK (A, bottom row) or fXII (A, top row) with A2AP are stained by both antibodies and are marked by arrows. Pooled plasma samples from PKK or fXII ASO treated male C57BL/6 mice (3 weeks at indicated dose, n = 8 per treatment group) were subjected to immunoblot analysis of kallikrein heavy chain levels (D). fXIIa complexes with antithrombin (AT) were analyzed in plasma samples from PKK and fXII ASO treated male C57BL/6 mice (3 weeks at indicated dose, n = 8 per treatment group) by ELISA (B). Mean values of fXIIa-AT complex levels and fXII plasma protein levels were plotted against PKK protein inhibition for the direct comparison (C). *P ≤ .05, **P ≤ .01 and ***P ≤ .001 compared with untreated control.

Figure 3

Effect of PKK and fXII ASOs on FeCl₃-induced inferior vena cava thrombosis. Male BALB/c mice were treated with either PKK (A) or fXII (B) ASO at indicated doses administered subcutaneously for 3 weeks (n = 8 per treatment group). FXI ASO treatment was used as a reference antithrombotic drug (B). Thrombosis was induced by 3 minute exposure of the inferior vena cava to a 10% FeCl₃ solution and assessed by RT-PCR measurement of platelet factor 4 (PF4) mRNA levels at the site of injury. Comparison of effects of PKK and fXII protein inhibition in FeCl₃-induced vena cava (IVC) thrombosis (C). *P ≤ .05, **P ≤ .01 and ***P ≤ .001 compared with untreated control.

Figure 4

Effect of PKK and fXII ASOs on stenosis-induced inferior vena cava thrombosis. Male BALB/c mice were treated subcutaneously with PKK or fXII ASO for 3 weeks at indicated doses (n = 6-8 per treatment group). Vena cava thrombosis was induced by combination of reduced blood flow with minor endothelial damage. Twenty-four hours after thrombosis induction thrombi were collected, fixed in formalin, weighed (A-B) and photographed (C-D). Thrombus weight relative to PBS control mice was calculated (A-B). FXI ASO was used as a reference antithrombotic drug (B,D). Comparison of effects of PKK and fXII protein inhibition in stenosis-induced vena cava (IVC) thrombosis (E). *P ≤ .05, **P ≤ .01 and ***P ≤ .001 compared with untreated control.

Figure 5

Antithrombotic effect of PKK and fXII ASO treatments in mouse model of FeCl₃-induced mesenteric artery thrombosis. Male Swiss Webster mice were treated with either PKK (A) or fXII (B) ASO at indicated doses administered subcutaneously for 3 weeks (n = 6-8 per treatment group). Mesenteric artery thrombosis was induced by a 3 minute exposure of the mesenteric vein to a 6% FeCl₃ solution. Platelets were labeled through subcutaneous injection of Rhodamine-6G and fibrin was labeled through intravenous injection of Alexa 647–conjugated anti-fibrinogen antibody. Continuous recording of venous thrombus formation by confocal intravital microscopy (confocal-IVM) was done for a total period of 70 minutes. Normalized integral intensity for each of 2 channels (fibrin and platelet detection) was plotted against time. Intensities for control group are shown in black, 10 mg/kg/wk dose of PKK and fXII ASOs are shown in blue, 20 mg/kg/wk dose in green, 40 mg/kg/wk dose in yellow, and 80 mg/kg/wk dose in red. Platelet and fibrin content of formed thrombi in the vena cava in different treatment groups at different time points is shown in panel C. Comparison of effects of PKK and fXII protein inhibition in FeCl₃-induced platelet aggregation (D) and fibrin accumulation (E) in mesenteric artery. *P ≤ .05, **P ≤ .01 and ***P ≤ .001 compared with untreated control.

Figure 6

Effect of PKK and fXII ASO treatments on bleeding. Effect of PKK (A) and fXII (B) antisense treatment at indicated doses on tail bleeding was evaluated in male BALB/c mice (n = 10 per treatment group). FII ASO treatment at indicated dose was shown to cause increased bleeding and was used as reference drug (A-B). *P ≤ .05, **P ≤ .01 and ***P ≤ .001 compared with untreated control.