Research Article: Zebrafish Thrombocytes: Functions and Origins

Date Published: June 24, 2012

Publisher: Hindawi Publishing Corporation

Author(s): Gauri Khandekar, Seongcheol Kim, Pudur Jagadeeswaran.


Platelets play an important role in mammalian hemostasis. Thrombocytes of early vertebrates are functionally equivalent to mammalian platelets. A substantial amount of research has been done to study platelet function in humans as well as in animal models. However, to date only limited functional genomic studies of platelets have been performed but are low throughput and are not cost-effective. Keeping this in mind we introduced zebrafish, a vertebrate genetic model to study platelet function. We characterized zebrafish thrombocytes and established functional assays study not only their hemostatic function but to also their production. We identified a few genes which play a role in their function and production. Since we introduced the zebrafish model for the study of hemostasis and thrombosis, other groups have adapted this model to study genes that are associated with thrombocyte function and a few novel genes have also been identified. Furthermore, transgenic zebrafish with GFP-tagged thrombocytes have been developed which helped to study the production of thrombocytes and their precursors as well as their functional roles not only in hemostasis but also hematopoiesis. This paper integrates the information available on zebrafish thrombocyte function and its formation.

Partial Text

Hemostasis is a defense mechanism to prevent loss of blood in the event of an injury in an organism that has a vasculature [1]. It consists of the platelet response to injury which results in platelet aggregation and plugging the wound, termed primary hemostasis, followed by the interplay of a complex cascade of coagulation factors on the platelet surface ultimately resulting in a fibrin clot, termed secondary hemostasis. After their primary hemostatic function platelets, also repair the damaged endothelium [2]. In primary hemostasis platelets adhere to collagen in the subendothelial matrix in response to injury and are subsequently activated by a complex signaling cascade resulting in secretion of their granular contents. These contents also result in the amplification of platelet aggregation at the site of injury and formation of a platelet plug which is stabilized further with help of fibrin [1]. This hemostatic plug prevents loss of blood from the site of injury. Thus, platelets that play a role in hemostasis and defects in platelet function have been shown to be involved in bleeding disorders as well as many pathophysiological conditions like thrombosis, inflammation, and even cancer [3]. Platelets have a number of receptors on their membrane surface that help regulate signaling pathways in platelets. A substantial amount of research has been done in studying platelet development and function mostly using human platelets [2–4] murine models [4], and identification of a number of factors and their roles in platelet function [2–4]. Recently, to identify novel factors involved in platelet function, N-ethyl-N-nitrosourea (ENU) mutagenesis and genomic screens of genes affecting platelet development and function have been attempted in mice [5]. However, they are expensive, less efficient, and have lower throughput. In humans, several novel quantitative trait loci associated with platelet-signaling pathways have been identified: however, these studies require additional functional evaluation using either animal models or human subjects [6]. Thus, study of platelet function requires a model system that is efficient, less costly, and amenable to higher-throughput screen, with hemostatic pathways similar to those found in humans [7]. The hemostatic system of invertebrates differs from that of vertebrates and therefore cannot be used as a model organism to study hemostasis [8]. In this regard, we wondered whether Danio rerio (Zebrafish) previously used as a genetic model to study developmental biology could be used as a genetic model to study hemostasis especially platelet biology [1]. Its high fecundity, external fertilization, transparency at early stages of development, and availability of large-scale mutagenesis methods are some of the features that make it a useful model system, thus attracting our attention [9, 10]. However, the challenge was to prove whether zebrafish thrombocytes and their functional pathways are similar to those found in platelets. For this, characterization of thrombocytes and their functional pathways was required as well as technology suitable for large-scale screens. Therefore, we developed the required technologies ourselves and found them sufficient enough to warrant their utility for the study of hemostatic function. Recently, several groups utilized our zebrafish model to study hemostasis and discovered several factors regulating hemostasis [11]. This paper provides an overview on the zebrafish thrombocyte characterization and development as well as other advances made not only in our laboratory but also from other laboratories which have applied the knowledge and technology that we developed in studying thrombocyte biology.

Unlike mammalian platelets which are anucleated, zebrafish thrombocytes have a nucleus. Our work has shown morphological and functional similarities between the zebrafish thrombocytes and human platelets [12]. Zebrafish thrombocytes have a sparse cytoplasm with large nuclei. The ultrastructure analysis of thrombocytes demonstrated that the cytoplasm contains many vesicles that open to the cell surface, similar to the open canalicular system in mammalian platelets (Figure 1). To demonstrate thrombocyte function, we developed blood collection and thrombocyte aggregation assays using less than one microliter of blood and established that zebrafish thrombocytes are stimulated by agonists including collagen, ADP, ristocetin, and arachidonic acid consistent with the human platelet aggregation methods. The results from such analyses revealed that the receptors for collagen, ADP, vWF, and thromboxane are conserved [12]. By using immunological methods, we have shown that αIIb integrin receptor and GpIb are present on thrombocyte membrane. Cox1 and Cox2 enzymes involved in arachidonic acid metabolism have also been identified in zebrafish [13]. Recently, we have shown that the thrombin receptor PAR-1 and its paralogue PAR-2 are also present on thrombocytes [14]. Using antibody staining and RT-PCR, we have also shown the presence of vWF in thrombocytes [15]. In a recent review, Lang et al. provide a detailed result of BLAST searches between human adhesion proteins and zebrafish proteins confirming our evidence for their similarities [16]. Thus, receptors for both thrombocyte adhesion and aggregation have been shown to be conserved in zebrafish. Subsequently, we developed a laser-induced thrombosis assay to study thrombocyte function and established that thrombosis assays are physiologically relevant in this model [17]. This study resulted in three assays, time to occlusion of artery from the time of laser injury (TTO), time to attachment of first cell from the time of laser injury (TTA) and also time taken to dissolution of the aggregate (TTD). Several reviews regarding the development of the zebrafish model for the study of thrombocyte function using laser-induced thrombosis assays from our laboratory are available [18–21].

To visualize thrombus formation, we wanted to perform intravital staining of the blood cells in zebrafish larvae by intravenous injection of lipophilic dye DiI-C18 (DiI) [22]. Surprisingly, we found only a few cells in the circulating blood were labeled in contrast to the entire blood cells. Subsequently, we identified that only a small proportion of thrombocytes in zebrafish blood was labeled by DiI alone, whereas all thrombocytes were labeled by mepacrine and, thus, giving two populations of thrombocytes (DiI+ and DiI−) (Figure 2). We found that DiI+ thrombocytes have higher levels of rough endoplasmic reticulum and thus higher protein synthesis than the DiI− thrombocytes. Furthermore, labeling the thrombocytes with BrdU for 24 hours resulted in BrdU-labeled circulating thrombocytes which were DiI+, but there were no BrdU-labeled thrombocytes that were DiI−. These results suggested that DiI+ thrombocytes were the first ones to appear in the circulation and, therefore, we called them young thrombocytes which are generated by their precursor cells by thrombopoiesis; by contrast, DiI− thrombocytes were called mature thrombocytes since in the circulation young thrombocytes presumably progress through the maturation process. By performing annexin V binding assays and estimating P-selectin levels on these two types of thrombocytes, we found that young thrombocytes are functionally more active than the mature thrombocytes [23]. In addition, we also found that young thrombocytes first appear at the site of injury and form their own clusters followed by the subsequent appearance of a mature thrombocyte cluster [23].

Platelet microparticles are the microvesicles released by platelets upon activation and have been shown to be involved in thrombin generation [25]. These are 0.1–1.0 μm in diameter and posses most receptors found on platelets such as P-selectin, GPIb, and αIIbβ3 [26]. Microparticle formation from platelets is believed to occur when the asymmetry of the membrane phospholipid is lost and phophotidylserine is externalized [27, 28]. Platelet derived microparticles are thought to promote platelet interaction with subendothelial matrix in an αIIbβ3-dependent manner [29]. Elevated levels of microparticles are observed in many pathological conditions including meningococcal sepsis [30], disseminated intravascular coagulation [31], and myocardial infarction [32].

ENU mutagenesis has been used extensively in forward genetic screens in an unbiased manner [1]. With the laser-induced thrombosis method a relatively high throughput screen is possible to select zebrafish mutants which have hemostatic defects. We proposed that such mutagenesis methods, combined with the laser-induced thrombosis method may lead to the discovery of novel thrombocyte-specific genes and so we pursued this approach. We performed a large-scale screen and found several mutants which have hemostatic defects; however, one mutant which we characterized has a defect in a novel orphan GPCR suggesting it plays a role in thrombocyte function (manuscript in preparation). Thus, we have established it is possible to conduct forward genetic screens for hemostatic function. Another mutant which has relevance to thrombocyte function is the fade  
out mutant which reiterates several aspects of Hermansky-Pudlak syndrome [34]. Furthermore, in the large-scale genome TILLING project spearheaded by Sanger Institute, several mutations in genes related to thrombocyte function were found. However, these will have to be sorted out and their functional evaluation performed in the near future.

Despite the advances in genetic studies of thrombocyte function and development in zebrafish, many novel genes involved in thrombocyte origins and functions remain to be identified. For example, even though embryonic GFPLow thrombocytes have been identified as HSCs and their role in repopulating the kidney for initiating the subsequent generation of thrombocytes from HSCs has not yet been investigated. Thus, we have no information regarding genes involved in the production of thrombocyte precursor cells in adult zebrafish. Likewise, studies of genes involved in maturation from young to mature thrombocytes, as well as genes controlling the production of thrombocyte microparticles are in the beginning stages. Since our laser-induced thrombosis assays for studying hemostasis have already found applications, we anticipate more such studies of this kind will be performed to assess the role of novel human genes relevant to hemostasis and thrombocyte development and function [35]. Recently developed technologies such as Genome TILLING [63, 64], zinc finger nuclease, or other nuclease/s (TALEN) based knockout methods [65–68] are also anticipated to complement the already available methods for studying functions of genes involved in thrombocyte function and production. However, large-scale silencing of genes to study thrombocyte development and production are still prohibitively expensive. Thus, future development of cost-effective gene silencing methodologies is required to attempt a functional genomics approach to analyze thrombocytes using the zebrafish model. Once the genes are identified, utilizing Vivo morpholino technology, we predict that characterization of the phenotypes by thrombocyte aggregation/adhesion functional assays, and determination of their mechanism of action will all be within reach in the next decade.