Loss of Protein C in Zebrafish Results in Impaired Neutrophil Migration after Tissue Injury

Hemostasis is a natural protective process that developed to retain a circulating blood system, conferred by a complicated yet sophisticated balance of factors. Disturbances of this network result in thrombosis or hemorrhage. Among many well-characterized coagulation factors, protein C (PC) exhibits multifunctional roles including anticoagulant, cytoprotective, and anti-inflammatory activities. The importance of PC has been demonstrated not only by the increased risk of venous thrombosis in individuals with heterozygous deficiency, but also the observed neonatal lethality in patients. Knockout mice exhibit similar neonatal lethality, which has made it difficult to further study complete deficiency. The zebrafish is a vertebrate organism that is characterized by a powerful genetic system, prolific breeding, rapid and transparent development, and a well described and highly conserved coagulation cascade. Here we utilize genome editing to generate a null allele of the PC gene (proc) in zebrafish and discover that its loss not only impairs hemostatic balance, but also affects neutrophil recruitment to sites of tissue injury. Through examination of publicly available zebrafish genome sequence, we determined that the proc locus is duplicated in tandem, resulting in two closely adjacent copies with nearly identical sequences. We used CRISPR/Cas9 with two single guide RNAs flanking the entire locus to produce a 17.3 kilobase deletion that knocks out both copies of proc to produce a complete null mutation, verified by sequencing and quantitative PCR. proc^-/- mutants survived well into adulthood, with ~50% lethality by seven months of age. The embryonic survival and accessibility enabled us to perform intravital microscopy to evaluate the hemostatic effects of PC deficiency. We used laser-induced endothelial injury on the posterior cardinal vein (PCV) at 3 days post fertilization (dpf), which typically results in rapid formation of an occlusive fibrin-rich thrombus. proc^-/- mutants had an average time to occlusion of 60 seconds versus 13 seconds in controls (p < 0.0001), consistent with a consumptive coagulopathy, as previously seen in antithrombin III (at3) mutants. A transgenic background with fluorescently labeled fibrinogen showed that more than 95% of proc^-/- mutants had spontaneous thrombi in the PCV, which was not present in controls. To assess the role of PC in inflammation, we used two different injury strategies, non-vascular tail transection and chemical treatment (copper sulfate), on 3 dpf zebrafish larvae. Staining for neutrophil granules revealed homing to the site of injury within 60-75 minutes. In proc^-/- mutants we found an average 50% reduction in the number of neutrophils recruited to the site of injury yet counts in the caudal hematopoietic tissue (the site of larval hematopoiesis) were unchanged. Since protein S (PS) is a cofactor for PC anticoagulant function, we hypothesized that the consumptive coagulopathy, but not the neutrophil recruitment, would be PS-dependent. We used genome editing to disrupt the PS gene (pros1) and found that loss of PS also results in a mild consumptive coagulopathy, but spontaneous thrombus formation was less common in the PCV (25%) and was often in the heart instead (80%). Neutrophil recruitment was unaffected in pros1 mutants, and evaluation of double proc/pros1 mutants revealed no synergy in any of the phenotypes. In conclusion, PC and PS deficiency in zebrafish show some similarity to our previously reported model of AT3 deficiency, but the effects are less potent, allowing robust survival that enables in vivo analyses. Our data suggest that the thrombotic phenotypes of PC and PS deficiency are not identical, and display tissue-specific phenotypes. We also found evidence for PS-independent functions of PC in neutrophil migration. We speculate this is due to the role that PC plays in inflammation and signaling but cannot exclude a role in neutrophil extracellular trap (NET) formation. This model of complete proc^-/- deficiency in an accessible organism will facilitate further in vivo study of PS-dependent and independent functions of PC, as well as interplay between the two factors.