Conservation of Hemostatic System Component Function Between Zebrafish and Mammals.

Abstract 3165

Poster Board III-102

Extensive DNA sequence similarity has previously been demonstrated for most known components of the mammalian hemostatic system and their orthologous zebrafish genes, and vascular occlusion has been observed in zebrafish following laser-induced vascular injury (Jagadeeswaran P, et al. Methods Mol Med 2006).To test potential conservation of function across species, activated bovine thrombin (Sigma-Aldrich), over a range of concentrations was injected into the orbital vascular plexus of zebrafish embryos at 72 hours post fertilization (hpf). The severity of thrombus formation was relative to the thrombin concentration, with thrombosis consistently observed in 72 hpf wild type (WT) embryos at 30-90 ng of thrombin (1-3 nl of a 30mg/ml solution diluted in 0.9% NaCl, with 0.25% phenol red as a tracer). With injection of higher doses (90 ng), complete occlusion of the circulation was generally observed within 30 seconds, with a large thrombus visible in the heart. Results were scored as zero when no thrombosis was observed, grade 1 when thrombosis was seen in the vessels but blood flow was preserved, and grade 2 for total occlusion of the circulation. Thirty of 30 WT embryos with grade 1 thrombosis showed resolution of the thrombus by 5 minutes post injection, consistent with the activity of an endogenous fibrinolytic system in zebrafish, compared to only 22 of 30 among WT embryos scored as grade 2. No thrombosis was observed with injection of BSA (N=30) or thrombin inactivated by incubation at room temperature for 24 hours (N=30). One cell stage zebrafish embryos were injected with antisense morpholino oligonucleotides (MOs) designed to target the mRNAs transcribed from the zebrafish genes for the fibrinogen alpha-chain, von Willebrand Factor (VWF), or ADAMTS13. Each resulting embryo (only embryos without morphologic defects were used), along with a matched non-MO-treated control embryo, was then injected at 72 hpf with 30-60 ng bovine thrombin, and thrombus formation scored blindly by 2 independent observers at 30 seconds and 2 minutes post injection. An ∼70% reduction in mean thrombosis score was observed for embryos treated with the fibrinogen alpha-chain MO (N=29) compared to controls (p<0.01). A similar ∼60% reduction in mean thrombosis score was observed for VWF MO-treated embryos (N=27) compared to controls (p< 0.01). In contrast, injection of an ADAMTS13 MO (N=31) resulted in an ∼2.5 fold increase in mean thrombosis score (p< 0.01). Co-injection of the VWF and ADAMTS13 MOs (N=15) resulted in a mean thrombosis score that no longer differed significantly from control (p =0.36). In conclusion, bovine thrombin injection induces prompt thrombosis in zebrafish, which appears to be dependent on zebrafish fibrinogen and VWF function, with subsequent spontaneous thrombus dissolution suggesting endogenous zebrafish fibrinolytic activity. Inhibition of ADAMTS13 appears to enhance thrombosis in this system, in a VWF-dependent manner. These results suggest that the structure and function of key hemostatic components is tightly conserved between zebrafish and mammals and that the injection of bovine thrombin into the zebrafish could serve as the basis for future high throughput mutagenic and chemical screens.