SEARCH

SEARCH BY CITATION

Keywords:

  • coagulation;
  • hemostasis;
  • knockdown;
  • laser injury;
  • thrombocytes;
  • thrombosis;
  • zebrafish

Hemostasis and thrombosis

  1. Top of page
  2. Abstract
  3. Hemostasis and thrombosis
  4. Realization of the potential utility of zebrafish as a model for hemostasis
  5. Zebrafish genetics
  6. Zebrafish hemostasis
  7. Characterization of plasma factors
  8. Hemostatic screens and isolation of hemostasis mutants
  9. Focused screens
  10. Global screens
  11. Knockdowns of hemostatic gene functions
  12. Transgenesis
  13. Future directions
  14. Acknowledgements
  15. Supplementary material
  16. References

Hemostasis is a defense mechanism that is a well-orchestrated and controlled process initiated in response to injury [1]. When a blood vessel is injured, the subendothelial matrix, which contains cells, collagen, and von Willebrand factor (VWF), is exposed and the flowing platelets roll and tether to the site of injury [2]. Platelets then adhere to this matrix via a collagen receptor and GPIb, an adhesive receptor that binds to VWF [3,4]. The attachment of platelets to the matrix by these receptors initiates signaling events in the platelets so that their contents, both from α granules [factor (F)V, etc.] and dense granules (ADP, etc.) are released. The granular contents, such as ADP, act as agonists to further activate more platelets and recruit them onto the matrix-attached platelets by using the GPIIb/IIIa receptor and fibrinogen [5,6]. This process is the primary response that is followed after the initiation of coagulation by tissue factor (TF) binding to factor (F)VIIa, which cleaves factor (F)X to FXa. FXa further cleaves FV released from platelets as well as FV present in plasma to generate FVa. FXa and FVa together generate thrombin from prothrombin in picomolar quantities [7,8]. This thrombin also acts as an agonist by cleaving thrombin receptors and enhancing the signal to activate more platelets, while also cleaving FV, factor (F)VIII and factor (F)XI. FXIa cleaves factor (F)IX and along with FVIIIa cleaves more FX, which now along with excessive FVa generates copious amounts of thrombin, thus generating the fibrin needed to seal the wound.

The activated coagulation pathways are inhibited by anticoagulant pathways involving the TF pathway inhibitor, which acts upon the TF–FVIIa–FXa complex, thereby stopping initiation. In addition, antithrombin inhibits thrombin activity, and activated protein C degrades cofactors VIIIa and Va, thus limiting the production of thrombin [9,10]. The fibrinolytic pathways remove the fibrin formed [11] and platelets in thrombi are retracted and further cleared [12]. The vessel wall also acts as an inhibitor of hemostasis by providing a layer of endothelial cells that secrete anticoagulant and platelet inhibitory factors, such as nitric oxide [9,13–15].

Human disease conditions in which these processes are not properly maintained result in either excessive bleeding or clotting inside the vessel [16–18]. The formation of clots inside the vessel is referred to as thrombosis and is a pathological process caused by disturbance of the vessel wall, perturbation of blood-borne factors, or stasis. These three components are classically referred to as Virchow's triad [19], which essentially encompasses all of the hemostatic components. Despite extensive investigations into thrombotic processes, only a limited number of genetic risk factors have been identified to account for venous thrombophilia. For example, the two most prevalent genetic variations associated with venous thrombosis are FV Leiden [20], which confers resistance of FV to degradation by activated protein C or prothrombin G20210A [21], a polymorphism in the 3′ untranslated region of the prothrombin gene, that is associated with higher levels of prothrombin. However, these two variations account for at most only 70% of venous thrombophilia cases [22]. In contrast to these insights into the genetic causes of venous thrombosis, no major genetic factor has yet been identified as a definitive risk for arterial thrombosis [23–25].

Overall, the genetic factors that influence thrombosis are poorly understood. For example, the mechanism of FVIIa generation to initiate coagulation is not understood, although suggestions were made that FIXa is involved [26,27]. Despite this suggestion, the question still continues as to what generates the initial FIXa. While the genetics of disease-causing mutations has complemented biochemical investigations, the identification of novel genetic risk factors for thrombosis is difficult, due to the complexity of the problem. Thus, alternative approaches to identifying these genetic factors are needed to fill in the gap in our knowledge of this important life-threatening clinical problem.

Realization of the potential utility of zebrafish as a model for hemostasis

  1. Top of page
  2. Abstract
  3. Hemostasis and thrombosis
  4. Realization of the potential utility of zebrafish as a model for hemostasis
  5. Zebrafish genetics
  6. Zebrafish hemostasis
  7. Characterization of plasma factors
  8. Hemostatic screens and isolation of hemostasis mutants
  9. Focused screens
  10. Global screens
  11. Knockdowns of hemostatic gene functions
  12. Transgenesis
  13. Future directions
  14. Acknowledgements
  15. Supplementary material
  16. References

One of our earlier studies involved the human FIX Leiden mutation, which results in childhood hemophilia that is spontaneously corrected as the patient advances in age past puberty [28]. The proposed mechanism is that the FIX gene promoter is activated by an age-dependent transcription factor to restore functional levels of FIX [29]. In an attempt to understand this mechanism, we constructed a transgenic mouse model with the FIX Leiden promoter driving a reporter gene [30]. Although the transgenic mouse showed an age-dependent increase in gene expression from this promoter, it did not advance our understanding of this mechanism beyond what was already understood from human patients. Thus far, only age-dependent cis-elements in the FIX gene promoter have been noted [31], but the trans-acting factors still remain unidentified. At that time, we realized that to identify these factors was difficult using the existing biochemical and transgenic approaches. We needed classical genetics approaches, such as those used by Thomas Hunt Morgan in Drosophila, where mutations are randomly created and screened for by phenotypic alterations [32,33]. It was also obvious that such an approach could not only address the above problem, but may resolve even more complex problems such as thrombosis. However, we realized that using classical genetics in mice to study hemostasis and thrombosis would be an arduous task [34,35].

At that time, the zebrafish, a tiny fish from the River Ganges, was gaining recognition as a vertebrate genetic model for developmental studies [36,37]. Streisinger first introduced zebrafish because of their many advantages [38]. For one, the embryos were transparent, allowing easy observation of the developing embryo and its morphology. Development occurs in 3 days and it is easy to breed under laboratory conditions. Other advantages included short generation times, amenability for large-scale mutagenesis, high fecundity (female fish can lay up to 200 eggs), and ease of in-vitro fertilization. Since its introduction, saturation mutagenesis by ethylnitrosourea (ENU), which creates as many as 500–1000 mutations per genome, has become feasible and is very efficient [39,40]. Linkage methods are now available making identification of genes possible [41]. Thus, a large number of developmental mutants have been isolated because the mutants are easy to score by morphological methods [42]. The following is a brief description of how mutants are isolated and mapped to isolate the genes that are affected.

Zebrafish genetics

  1. Top of page
  2. Abstract
  3. Hemostasis and thrombosis
  4. Realization of the potential utility of zebrafish as a model for hemostasis
  5. Zebrafish genetics
  6. Zebrafish hemostasis
  7. Characterization of plasma factors
  8. Hemostatic screens and isolation of hemostasis mutants
  9. Focused screens
  10. Global screens
  11. Knockdowns of hemostatic gene functions
  12. Transgenesis
  13. Future directions
  14. Acknowledgements
  15. Supplementary material
  16. References

Mutagenesis is performed by dipping zebrafish males in ENU to cause multiple, random point mutations throughout the genome of the germ cells (spermatogonia) [40]. Mating of an ENU-treated male with a wild-type female zebrafish will yield progeny (F1 generation) that will harbor as many as 1000 random mutations. Theoretically, in 100 fish, 105 mutations can be generated, thus saturating the genome. The F1 generation is next bred to a wild-type fish to generate heterozygous progeny (the F2 generation). Performing brother–sister matings of the F2 generation, which generates homozygous fish and completes a classical three-generation screen, reveals mutant phenotypes of interest. Using this strategy, more than 1000 developmental mutants have been isolated [37,43,44], and more recently, several imaginative screens have been performed to isolate mutants related to lipid absorption, bone defects, etc. [45–47]. Embryonic and larval screens are easier when compared with adult screens, because rearing the F3 generation fish to adulthood requires more time and space. Even the embryonic screens are time consuming, as well as labor and cost intensive, since they require raising and maintaining multiple lines of zebrafish.

Many laboratories have also used an alternative method that makes the screen possible in two generations [48]. In this approach, haploid or homozygous mutant progeny can be generated directly from F1 females. Haploid embryos are generated by in-vitro fertilization of eggs taken from a heterozygous F1 female with male sperm, whose DNA was destroyed by treatment with ultraviolet light. Although these sperm do not contribute DNA, they can still initiate fertilization and allow the development of haploid embryos, which are viable for up to 3 days postfertilization. If these eggs are subjected to early high pressure treatment (EP) immediately after fertilization, these haploids will become diploid embryos, which are fully viable. In the EP treatment, hydrostatic pressure causes disruption of spindle formation, which is required for the separation of sister chromatids and polar body extrusion following fertilization. The resulting gynogenetic diploid larvae possess only maternally derived genes that are homozygous for most loci (heterozygosity may arise from recombination events in meiosis I). Thus, this EP treatment allows for a two-generation mutagenesis screen that will identify mutant-carrier females one generation ahead of classical approaches.

Once a mutant of interest is identified, the mutant loci can be identified by a positional cloning strategy [49]. For mapping purposes, an F1 female mutant carrier is crossed with a wild-type male zebrafish that is polymorphic for many loci when compared with the inbred strain used to generate mutations. From this progeny, homozygous mutants are generated by brother–sister mating. Genomic DNA is then isolated from mutant and normal progeny and the affected genetic locus is identified by linkage analysis using polymerase chain reaction-based microsatellite markers [50]. To rapidly identify linkage of a mutant phenotype to a marker, the method of bulk segregant analysis has been employed [41,51,52]. In this analysis, two pools of genomic DNA are created from normal and mutant zebrafish (20 zebrafish per genomic DNA preparation) and multiple sets of primers are used to amplify known microsatellite markers that span the zebrafish genome at an average resolution of 10 cM. The idea here is that at meiosis, recombination will shuffle both genes and markers. Thus, markers not linked to the mutant loci, or distal to the loci, will be present in both pools, while linked markers will be present in only one of the two pools. Once a marker is identified showing linkage to a mutant gene, analysis of additional flanking markers using a larger number of genomes will establish close linkage. Sequencing isolated bacterial artificial chromosomes (BACs) that contain the linked markers will identify the mutant gene. Confirmation that the gene causes the mutant phenotype is performed by rescuing the mutant through introducing the functional cDNA or by inhibiting the gene function by antisense approaches. However, with the availability of genome information, the painstaking efforts of positional cloning will be replaced by the candidate gene approach once a closer linkage is identified.

Zebrafish hemostasis

  1. Top of page
  2. Abstract
  3. Hemostasis and thrombosis
  4. Realization of the potential utility of zebrafish as a model for hemostasis
  5. Zebrafish genetics
  6. Zebrafish hemostasis
  7. Characterization of plasma factors
  8. Hemostatic screens and isolation of hemostasis mutants
  9. Focused screens
  10. Global screens
  11. Knockdowns of hemostatic gene functions
  12. Transgenesis
  13. Future directions
  14. Acknowledgements
  15. Supplementary material
  16. References

In spite of advances in developmental genetics, whether or not the zebrafish could be used as a genetic model to study mammalian hemostasis was questionable. Even though earlier studies suggested similarities to mammalian coagulation, the idea that fish coagulation pathways were primitive still lingered [53–56]. Furthermore, studying biochemistry in such a miniature model was intimidating, and convincing biochemists used to bovine models was daunting [57]. The first prerequisite then for the use of the zebrafish model was to show that most of the hemostatic pathways are similar to those found in humans. Therefore, we undertook the characterization of zebrafish hemostasis.

Plasma biochemistry  To characterize zebrafish coagulation, a technique was developed to collect reliably a few microliters of blood from the adult zebrafish by clipping the dorsal aorta without activating coagulation [58]. To test zebrafish coagulation using existing assays, which required relatively large blood samples, we used pooled plasma from multiple fish (200 fish). We also used concentrated human fibrinogen which could serve as a substrate for zebrafish thrombin. Using these modifications, we found that zebrafish plasma responded to Dade ACTIN (rabbit brain partial thromboplastin reagent containing ellagic acid), zebrafish thromboplastin, and Russell viper venom factor X activator (RVV-Xa), respectively. These biochemical data established the presence of both contact-activated (propagation) and TF- activated (initiation) pathways similar to those found in mammals [59,60].

To test whether oral anticoagulants would work, it was important to treat the zebrafish with warfarin. However, it was difficult to administer warfarin orally to zebrafish, as had been done previously in Atlantic salmon. Therefore, we hypothesized that dissolving the warfarin tablet in water might provide a sufficient dose to obtain the drug effect. To prove that zebrafish consume water, using radio-opaque dyes and X-ray imaging we provided evidence for water consumption [59]. Subsequently, we treated zebrafish with dissolved warfarin and showed that coagulation times were prolonged after treatment, suggesting that the vitamin K-carboxylase and reductase system may be present in zebrafish. In our later studies, we confirmed this vitamin K-dependent carboxylase activity by a biochemical assay [61]. In addition, using Protac, a protein C activator, we found evidence for the presence of anticoagulant protein C [59]. The ability to inhibit coagulation reactions by the addition of heparin also suggested the presence of antithrombin in zebrafish [59].

To circumvent the limitation of existing analytical techniques, which required 200 µL of blood, we established microassays, which require < 2 µL of blood [62]. Based on the observation that human fibrinogen was specifically cleaved by zebrafish thrombin, we developed kinetic coagulation assays using dilute zebrafish plasma and exogenously added human fibrinogen to measure the prothrombin time (kPT), the common pathway activation (kRVVT), and TF-mediated pathways or partial thromboplastin time (kPTT). These assays measure the rate of thrombin generation by recording the increase of light absorbance at 405 nm due to fibrin formation and are sensitive enough to detect clotting activity in a single fish.

Molecular characterization To complement the biochemical similarities of zebrafish coagulation with mammalian pathways, we began our studies to characterize zebrafish coagulation factors by cloning their cDNAs and genes. By aligning cDNAs encoding various coagulation factors from multiple species, we identified highly conserved regions that were then used to design primers to isolate zebrafish cDNAs [63,64]. The accumulation of expressed sequence tags during this investigation was helpful in characterization of cDNA sequences, not only of coagulation factors, but also of anticoagulant proteins and fibrinolytic factors [65]. In some cases, the high degree of homology in domain organization among various plasma factors made it difficult to identify FVII, FIX and FX, when determined by sequence data alone. Therefore, other methods were used to discern cDNA identities. For example, to confirm the identity of the cDNA corresponding to FVII, a recombinant zebrafish FVII was expressed, and its function was confirmed to be similar to human FVII in biochemical assays. Since the FVII gene is linked to the FX gene in humans, we reasoned that the same might be true of zebrafish by reason of gene organization or synteny. Probing a BAC library with the FVII cDNA isolated a BAC clone containing the zebrafish FVII gene. By sequencing this clone, we identified the FX gene that was closely linked to the FVII gene. By elimination, these data allowed us to distinguish between FIX and FX cDNAs.

Identification of novel factors Based upon molecular and biochemical data, a schematic of the procoagulant and anticoagulant pathways was generated which revealed a great deal of homology between zebrafish and mammalian systems. However, a novel cDNA was isolated that showed high similarity to FVII, but lacked the critical amino acids necessary for FVII activity [65]. This gene may have arisen as a result of the genome-wide duplication that is believed to have occurred in teleosts [66]. Subsequent to our finding, similar duplicated FVII genes were found in the Fugu genome [67]. Expression of the above novel cDNA allowed us to demonstrate that it functions as an inhibitor of blood coagulation in kPT and kPTT assays and we termed this FVIIi. Interestingly, zebrafish FVIIi showed prolongation of kPTT, but not kPT, using human plasma [65]. The mechanism of FVIIi inhibition, and its possible presence in the mammalian genome, still requires further investigation. However, the presence of this factor demonstrates that much remains to be understood about the regulation of coagulation, and that the zebrafish will continue to reveal novel hemostatic factors.

Characterization of thrombocytes  Previous reports showed the presence of nucleated equivalents of platelets in trout and salmon [56], known as thrombocytes, but these had not yet been shown in zebrafish blood. To substantiate further the relevance of the zebrafish to human hemostasis, we sought to identify and characterize thrombocytes in the zebrafish using morphological, immunological, and functional studies. Electron microscopy of zebrafish blood showed a cell population with structures resembling the open canaliculi and filopodia of platelets [58]. It was shown that these cell populations were thrombocytes by specifically labeling them with antibodies directed against the human platelet receptors GPIb and GPIIb/IIIa [58].

To measure thrombocyte aggregation activity, a novel whole-blood plate-tilt assay was developed [58]. In this assay, citrated zebrafish blood was added into the agonist solution within the well of a microtiter plate. After 2 min, the plate was tilted 45°. A positive or normal thrombocyte response resulted in a firm button formation and delayed migration of blood down the slope of the well. Using this assay, it was shown that whole-blood aggregates formed in response to collagen and ADP, thus suggesting that receptors for these agonists were present on thrombocytes. Additionally, ristocetin was shown to cause thrombocyte agglutination, a process that in mammals is mediated by the binding of the GpIb receptor to VWF.

Another hallmark of platelet function is the secretion of various platelet-activating agonists [68]. Using a luminometry assay, we demonstrated the secretion of ATP by thrombocytes in response to treatment of zebrafish blood with collagen [58]. Another important secreted component is thromboxane A2, which is generated from arachidonic acid through the action of cyclooxygenase [69]. We demonstrated the presence of cyclooxygenase-1 by aspirin inhibition of agonist activation. Subsequently, the presence of two isoforms of cyclooxygenase in the zebrafish, COX-1 and COX-2, was shown [70].

Although these findings suggested the functional equivalence between mammalian platelets and zebrafish thrombocytes, assays that directly and quantitatively measured thrombocyte function were important. However, isolation of thrombocytes from small quantities of blood was not trivial. Therefore, we turned to fluorescent labeling methods. We observed that zebrafish thrombocytes were selectively labeled by the lipophilic dye DiI-C18 when blood was labeled in vitro by incubation, or in vivo by direct injection into the circulation. This ability to label the thrombocyte membrane fluorescently allowed the visualization of filopodial extensions, which had been shown in human platelets to play a role in adhesion, aggregation, and possibly in clot retraction [71]. Additionally, we had also observed that with DiI-C18 labeling, bright fluorescent foci on the cell membrane accumulated after agonist treatment. Since DiI partitions favorably in ordered lipids, and lipid rafts are believed to be an ordered lipid structure associated with cell signaling complexes [72], the presence of these fluorescent foci may represent the formation of lipid raft-like structures in zebrafish thrombocytes.

With the identification and ability to label thrombocytes selectively, a fluorescent assay was developed to quantify thrombocyte aggregation. This assay was based upon the platelet activation properties of phosphatidylserine (PS) translocation to the outer plasma membrane and the binding of annexin V, a calcium-dependent phospholipid binding protein with high affinity for PS [73]. By labeling zebrafish thrombocytes with DiI-C18 and using FITC-conjugated annexin V, the rate of increased FITC fluorescence as a result of annexin V binding to thrombocyte aggregates was used to quantify the rate of aggregate formation. With this method, thrombocyte agglutination and aggregation curves were established using ristocetin, collagen, and ADP agonists. Platelet response to these agonists is mediated through a variety of specific receptors. The P2 family of receptors mediates the platelet response to ADP, and mammalian platelets express three ADP receptors, namely, P2Y1, P2Y12, and P2X1[74,75]. We were able to inhibit the aggregation of thrombocytes in response to ADP by using the ADP analog adenosine 3′-5′ diphosphate which selectively inhibits the P2Y1 receptor [76], suggesting that this receptor is conserved in zebrafish thrombocytes [71]. Recently, we have shown that thrombin also activates washed zebrafish thrombocytes (unpublished). In conclusion, all the components of the thrombocyte response appear to be functionally similar to that found in mammalian platelets. Searching the zebrafish genome database has yielded information on sequences of factors, such as VWF and the receptors present on the surface of platelets, which has provided evidence that most of the components of the platelet response are present in zebrafish.

Hemostatic screens and isolation of hemostasis mutants

  1. Top of page
  2. Abstract
  3. Hemostasis and thrombosis
  4. Realization of the potential utility of zebrafish as a model for hemostasis
  5. Zebrafish genetics
  6. Zebrafish hemostasis
  7. Characterization of plasma factors
  8. Hemostatic screens and isolation of hemostasis mutants
  9. Focused screens
  10. Global screens
  11. Knockdowns of hemostatic gene functions
  12. Transgenesis
  13. Future directions
  14. Acknowledgements
  15. Supplementary material
  16. References

Having shown the homology between zebrafish and mammalian hemostasis, as well as between thrombocytes and platelets, the next challenge was to develop genetic screens for hemostatic function. To perform a genetic screen, it must be easy, reliable, and should have high throughput. Screening for the components of Virchow's triad would be the ideal and global method for identifying mutants related to hemostasis and thrombosis. Initially, in absence of such a screen, we developed several focused screens to test whether hemostatic screening was possible. We summarize below these screens for defects in coagulation and thrombocyte functions.

Focused screens

  1. Top of page
  2. Abstract
  3. Hemostasis and thrombosis
  4. Realization of the potential utility of zebrafish as a model for hemostasis
  5. Zebrafish genetics
  6. Zebrafish hemostasis
  7. Characterization of plasma factors
  8. Hemostatic screens and isolation of hemostasis mutants
  9. Focused screens
  10. Global screens
  11. Knockdowns of hemostatic gene functions
  12. Transgenesis
  13. Future directions
  14. Acknowledgements
  15. Supplementary material
  16. References

We have developed focused screens such as kPT, kPTT and RVVT assays to identify coagulation pathway defects [62]. This provided the first proof of principle that hemostatic mutants could be isolated in zebrafish. By using these assays, it is possible to screen for defects such as warfarin resistance. Such a screen may yield mutants related to the coagulation pathway and its regulators. However, the possibility of obtaining non-hemostatic mutations such as mutants of warfarin absorption complicates sorting out such putative hemostatic mutants.

It is also now possible to screen for thrombocyte functions by the whole-blood aggregation plate-tilt assay [58]. This assay is sensitive enough to detect thrombocyte signaling and aggregation responses qualitatively and may be used to measure the contribution of other blood cells, such as leukocytes, to thrombocyte aggregation. Using this assay it should be possible to screen for mutations affecting ADP-induced or collagen-induced platelet aggregation and ristocetin-mediated agglutination. It should also be possible to use this assay to screen for aspirin resistance and for resistance to other antiplatelet drugs. However, a limitation of this assay was its inability to quantify the extent of thrombocyte aggregation.

Thus, currently we have adult screens for delineating the extrinsic pathway, intrinsic pathway, and common pathway. We also have adult screens to study collagen activation, ADP activation, ristocetin activation, thrombocytopenia or thrombocytosis, warfarin resistance, and aspirin resistance. In addition to these screens, specific expression patterns of hemostatic genes could be studied in zebrafish embryos by focused in situ hybridizations [77].

Global screens

  1. Top of page
  2. Abstract
  3. Hemostasis and thrombosis
  4. Realization of the potential utility of zebrafish as a model for hemostasis
  5. Zebrafish genetics
  6. Zebrafish hemostasis
  7. Characterization of plasma factors
  8. Hemostatic screens and isolation of hemostasis mutants
  9. Focused screens
  10. Global screens
  11. Knockdowns of hemostatic gene functions
  12. Transgenesis
  13. Future directions
  14. Acknowledgements
  15. Supplementary material
  16. References

An important global screen for detecting hemostatic function is to induce vascular damage in vivo by subjecting fish to chemical or laser injuries [78]. We developed assays using the chemical ferric chloride (FeCl3), which was found to cause thrombus formation in the caudal vessels of young larvae. The mechanism of FeCl3 is to cause endothelial injury by free radical production. Additionally, using a nitrogen pulsed dye laser we demonstrated that we could cause a targeted vascular lesion leading to thrombus formation (see Figs S1 and S2). Since in all these methods an in-vivo thrombus is formed, we deduced it was possible to conduct larval screens for thrombocyte and coagulation deficiencies by measuring the time to occlusion (TTO) of the vessel. In fact, we used the FeCl3 method and isolated a mutant victoria which was later confirmed by the laser method. In this mutant, we found a marker linked to prothrombin, and recent sequencing of the prothrombin gene demonstrated that the mutation is in the prothrombin locus (unpublished). Another mutant leopold, which was identified earlier by Mary Mullins of the University of Pennsylvania as having red spots that suggested a probable bleeding condition, was identified as a hemostatic mutant by the above TTO assay.

Currently, another large-scale screen is underway for venous thrombosis. We prefer laser thrombosis for global screens because measuring TTO is more reliable and reproducible than with FeCl3 injury. Further, it is possible to cause injury to either an artery or vein at a defined location by the laser method. In addition to the above TTO measurement, it has also been possible to measure the time taken to adherence of the first cell (TTA) and time to dissolution of the thrombus (TTD).

Knockdowns of hemostatic gene functions

  1. Top of page
  2. Abstract
  3. Hemostasis and thrombosis
  4. Realization of the potential utility of zebrafish as a model for hemostasis
  5. Zebrafish genetics
  6. Zebrafish hemostasis
  7. Characterization of plasma factors
  8. Hemostatic screens and isolation of hemostasis mutants
  9. Focused screens
  10. Global screens
  11. Knockdowns of hemostatic gene functions
  12. Transgenesis
  13. Future directions
  14. Acknowledgements
  15. Supplementary material
  16. References

The function of genes can be studied not only by mutagenesis, but also by inhibiting the gene functions with antisense approaches. An antisense approach that has gained rapid application is the knockdown of gene functions using modified oligonucleotides called morpholinos [79]. These antisense morpholinos are injected into single- to four-cell embryos to inhibit translation or normal processing of the pre-mRNA. Morpholinos could be selectively used against Kozak sequences around the translation initiator codon or around splice junctions [80,81]. Morpholinos have been shown to be stable as long as 8 days in the growing embryos and have been shown to inhibit between 50% and 95% of translation. Thus, if one requires a null mutant phenotype it is difficult to obtain by this method. However, the advantage of inhibition of translation at varying degrees is that embryonic lethal phenotypes would not arise as in knockout methods [82].

With the advent of the vascular occlusion assay, it has been possible to study knockdown phenotypes for genes relevant to hemostasis and thrombosis. So far, the hemostasis-related genes that have been subjected to this technology are FVII, FVIIi, and prothrombin [78,83]. The FVII and FVIIi knockdowns resulted in prolongation and shortening of TTO as predicted [78]. In the prothrombin gene knockdown, two types of mutant phenotypes were observed [83]. One of these, the early phenotype, had severe early abnormalities while the other, the weaker phenotype, advanced normally but had prolonged time to occlusion. These results recapitulate the prothrombin knockout in mice [83].

Developmental functions are also studied by over-expressing a specific mRNA by simple microinjections and subsequently altering the phenotype [84]. The altered phenotypes could be studied by using various probes to identify the downstream affected genes.

Transgenesis

  1. Top of page
  2. Abstract
  3. Hemostasis and thrombosis
  4. Realization of the potential utility of zebrafish as a model for hemostasis
  5. Zebrafish genetics
  6. Zebrafish hemostasis
  7. Characterization of plasma factors
  8. Hemostatic screens and isolation of hemostasis mutants
  9. Focused screens
  10. Global screens
  11. Knockdowns of hemostatic gene functions
  12. Transgenesis
  13. Future directions
  14. Acknowledgements
  15. Supplementary material
  16. References

Recently, many transgenic lines have been created in zebrafish, by the technology of Lin and colleagues [85], using green fluorescent reporters driven by the expression of cell/tissue-specific promoters. Although transgenic technology has been available in the zebrafish for more than a decade [86], only a few lines with relevance to hemostasis and thrombosis have been generated. For the study of thrombopoiesis, a transgenic line using the GpIIb promoter [87] has been developed which is useful to address the developmental role of thrombocytes and the subsequent downstream genes. Additionally, these fluorescently labeled thrombocytes will prove useful in studying the adherence, spreading, and aggregation of thrombocytes during thrombus formation. Another transgenic line that is relevant to hemostasis is the Fli-1 promoter-driven green fluorescent protein (GFP) line that shows specific GFP expression in endothelial cells and some blood cells [88]. This line could be used to study the responses of endothelial cells to the thrombus. Other GFP transgenic lines utilizing promoters such as GATA-1, GATA-2, and Pu.1 could be used to study the contribution of other blood cells [89] in the thrombus formation [85,90,91].

Future directions

  1. Top of page
  2. Abstract
  3. Hemostasis and thrombosis
  4. Realization of the potential utility of zebrafish as a model for hemostasis
  5. Zebrafish genetics
  6. Zebrafish hemostasis
  7. Characterization of plasma factors
  8. Hemostatic screens and isolation of hemostasis mutants
  9. Focused screens
  10. Global screens
  11. Knockdowns of hemostatic gene functions
  12. Transgenesis
  13. Future directions
  14. Acknowledgements
  15. Supplementary material
  16. References

Even though characterization of zebrafish hemostasis revealed that the zebrafish is a good model for studying mammalian hemostasis, continued characterization is important in defining the physiology of hemostasis to draw stronger parallels to human hemostasis. With the advent of laser-induced thrombosis, characterization of interactions of the coagulation proteins and blood cells in vivo is approachable. Due to the large size of thrombocytes and other blood cells, the kinetics of thrombus formation could be easily studied [78]. Characterization of thrombocytes and their development may further elucidate the mechanisms of platelet function and their production. Thrombocyte-specific genes will be identified which may complement efforts to identify platelet-specific gene expression. Microarray analysis could be employed to identify thrombocyte-specific genes and, coupled with knockdown technology, the function of these genes could be studied [83].

Developmental functions of hemostatic proteins could be better addressed using the zebrafish model. For example, using microarray analysis with knockdown of these factors, genes affected by signaling events initiated by these factors in development could be identified. In addition, null mutants of hemostatic genes could be generated by using methods employed in plants, termed genome TILLING (targeting induced local lesions in genomes) [92,93]. In this method, a library of ENU-mutagenized F1 animals is screened for mutations in a target gene by CEL-I-mediated heteroduplex cleavage or sequencing to identify a number of target-selected mutations, including null mutations. Null mutations identified in hemostasis genes are virtually equivalent to knockouts and could be used to complement knockdown approaches for understanding their developmental functions.

In the future, the global saturated mutagenesis screens utilizing laser thrombosis will be exploited to better understand all the components of Virchow's triad [19]. Currently, only venous thrombosis screens have been performed, but screens involving arterial thrombosis and fibrinolytic pathways are being explored. Furthermore, using the laser thrombosis assay for screening zebrafish with a library of chemical compounds provides a potential novel antithrombotic drug discovery system [58].

We hope that the debate regarding the utility of the zebrafish model for studying human hemostasis, which still lingers, will soon disappear, so that the power of the zebrafish model can be better utilized to advance the field of hemostasis and thrombosis. Could the zebrafish model address the puzzle of FIX Leiden, the problem which originally took us down this road?

References

  1. Top of page
  2. Abstract
  3. Hemostasis and thrombosis
  4. Realization of the potential utility of zebrafish as a model for hemostasis
  5. Zebrafish genetics
  6. Zebrafish hemostasis
  7. Characterization of plasma factors
  8. Hemostatic screens and isolation of hemostasis mutants
  9. Focused screens
  10. Global screens
  11. Knockdowns of hemostatic gene functions
  12. Transgenesis
  13. Future directions
  14. Acknowledgements
  15. Supplementary material
  16. References