SEARCH

SEARCH BY CITATION

Keywords:

  • Danio rerio;
  • in vivo thrombosis model;
  • laser-injury;
  • morpholino;
  • protein kinase C;
  • zebrafish

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental procedures
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References
  10. Supporting Information

Summary. Background: The zebrafish (Danio rerio) is becoming an attractive model organism for the assessment of gene function in thrombosis in vivo. Zebrafish, as a thrombosis model, have several advantages, with the capacity to follow thrombus formation at high resolution in real time using intravital microscopy, without the need for complex surgical techniques, and the capability to rapidly knockdown gene expression using morpholino antisense approaches. Objectives: We have recently shown, in mouse models, that protein kinase C alpha (PKCα) plays a critical role in regulating thrombus formation in vivo. PKC beta (β) plays a non-redundant role also in platelet function in vitro, but the function of this gene had not yet been assessed in vivo. Methods: In the present study, we analyzed the function of both PKCα and PKCβ in the zebrafish model in vivo, by live imaging using a laser-induced injury of the main caudal artery in 3-day-old larvae. Results: We showed that D. rerio express orthologs of both the PKCα and PKCβ genes, with high sequence identity. Translation blocking and splice-blocking morpholinos effectively and specifically knockdown expression of these genes and knockdown with either morpholino leads to attenuated thrombus formation, as assessed by several quantitative parameters including time to initial adhesion and peak thrombus surface area. Conclusions: Our data indicate that these two highly related genes play non-redundant roles in regulating thrombosis, an observation that supports our previous in vitro murine data, and suggests unique roles, and possibly unique regulation, for PKCα and PKCβ in controlling platelet function in vivo.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental procedures
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References
  10. Supporting Information

Platelets play a critical role in hemostasis, detecting and adhering to damaged blood vessel walls, resulting in the formation of a platelet plug (thrombus) that seals the damaged area. In hemostasis, this process is critical for maintaining vascular integrity, with defects resulting in spontaneous bleeding [1,2]. In cardiovascular and cerebrovascular disease, uncontrolled thrombus formation, as a result of atherosclerotic plaque rupture, can lead to the formation of occlusive thrombi, which can result in a myocardial infarction or stroke.

To understand better the molecular mechanisms of thrombosis, various in vivo models are increasingly being utilized, with gene knockdown and transgenic animals enabling the analysis of gene function. Recently, zebrafish have emerged as a model organism with which to investigate thrombosis owing to the translucency of the developing larvae, which allows live visualization of thrombus formation, and resolution, after ferric chloride damage or laser-induced injury [3–5]. Recently, thrombosis in zebrafish has been well characterized. The coagulation pathways, hematopoiesis and vasculogenesis appear to all be highly conserved, indicating that zebrafish are suitable for modeling human disease [6–8]. Likewise, in spite of being nucleated, thrombocytes (the zebrafish equivalent of platelets) share the same functional and regulatory systems as human platelets, including the αIIbβ3 integrin, GPIb, protease-activated receptors (PAR)1–4 and P2Y1/P2Y12 receptors, which are all present, as well as similar functional responses (such as aggregation, Ca2+ release, dense-granule secretion, annexin V-binding and P-selectin exposure) [3,9–12]. Finally, the use of morpholino antisense oligonucleotides has enabled the rapid functional assessment of genes through whole animal knockdown, in larvae up to the age of approximately 5 days [13].

The protein kinase C (PKC) family of serine/threonine kinases has a well-established, critical role in platelet function and thrombosis [14]. While the PKCs are generally viewed as being positive regulators of platelet function and thrombus formation, through broad spectrum inhibitor studies [15], isoform-specific gene knockout mice have shown that the PKC isoforms expressed in human platelets (nominally PKCα, β, δ and θ [16]) can play distinct and often opposing roles [14,17–22]. Of the two conventional PKC (cPKC) isoforms expressed in platelets, PKCα has been shown to be an important positive regulator of platelet function and a critical regulator of thrombosis in vitro and in vivo [17,18]. The role of PKCβ in platelet function and thrombus formation has been much less studied; however, PKCβ has been implicated in the regulation of outside-in signaling through the αIIbβ3 integrin [23] as well as being a positive regulator of in vitro thrombosis [18]. In spite of this, there is no information regarding the role of PKCβ in thrombus formation in vivo. As PKCα and PKCβ are the two conventional PKC isoforms expressed in platelets, an important question concerns how these two isoforms interact in the regulation of platelet function.

In the present study, we used the zebrafish model of laser-induced thrombosis to perform a comparative analysis of the in vivo roles of PKCα and PKCβ in thrombosis. Using anti-sense morpholinos in this model, we reproduced the in vivo thrombus formation phenotype seen in PKCα−/− mice and similarly showed that PKCβ also plays a positive role in regulating thrombus formation in vivo– the first such description.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental procedures
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References
  10. Supporting Information

Reagents

Anti-PKCα and anti-PKCβ antibodies were purchased from Enzo Life Sciences (Exeter, UK) and BD Biosciences (Oxford, UK), respectively. Anti-mouse and anti-rabbit horseradish peroxidase (HRP)-conjugated antibodies and ECL reagent were purchased from GE Healthcare (Amersham, UK). Anti-α-tubulin antibody, Type IX-A (low melting-point) agarose and ethyl 3-aminobenzoate methanesulfonate salt (tricaine) were purchased from Sigma-Aldrich Ltd (Gillingham, UK). Morpholinos targeting zebrafish PKCα (translation-blocking: 5′-TCGTTGCTTTGTGTATCAGCCA TTG-3′ and splice-blocking: 5′-ACCCCCTGATGAAGAGAAGAGAGAA-3′) and PKCβ (translation-blocking: 5′-CCGGCTCTGCCATTCTAAAAGCGGG-3′ and splice-blocking: 5′-ACAA AGCAGCAGACTGAAAGAAGGA-3′) and standard control morpholino (5′-CCTCTTACCTCAGTTACAATTTATA-3′) were obtained from GeneTools LLC (Philomath, OR, USA).

Pairing of adult zebrafish for mating and embryo collection

Adult LonAB (wild-type), Tg(-6.0itga2b:EGFP (CD41:GFP) [24] or pU1-Gal4-UAS-GFP [25] zebrafish were paired in breeding tanks at room temperature (25 °C) the evening before embryo collection. Embryos were collected into 0.3× Danieau’s solution 30 min after the expected laying time and kept at room temperature (20–25 °C) until the completion of morpholino injections. Larvae were raised in 0.3× Danieau’s solution at 28 °C.

Morpholino injection of zebrafish larvae

The translation-blocking and splice-blocking morpholinos targeting PKCα and PKCβ were designed by Gene Tools LLC, USA, after the provision of suitable sequences as outlined by Gene Tools. Translation- and splice-blocking morpholinos were successfully designed for both PKCα and PKCβ, with the splice-blocking morpholinos predicted to result in the cessation of protein synthesis mid-way through the C1A domains (Fig. S1) through the deletion of exon 2 of PKCα and exon 3 of PKCβ (Fig. 1A,B). Morpholinos were re-suspended in sterile dH2O and approximately 1.5 nL was injected into zebrafish larvae in the one- to two-cell stage. The amount of injected morpholino was titrated to find the optimal concentration for the experimental morpholinos. These were established as being 5 ng PKCα translation-blocking morpholino, 2.5 ng PKCα splice-blocking morpholino, 6 ng PKCβ translation-blocking morpholino and 4 ng PKCβ splice-blocking morpholino. Post-injection, larvae were maintained in 0.3× Danieau’s solution containing 0.0002% methylene blue for 72–96 h at 28 °C.

image

Figure 1.  Morpholinos to protein kinase C alpha (PKCα) and PKC beta (β) knock down their respective protein expression. A (i) Illustration of the design of the PKCα and PKCβ translation-blocking morpholinos to the transcription start sequences of their respective mRNAs. (ii) Illustration of the design of the PKCα and PKCβ splice-blocking morpholinos to the splice-acceptor sites of the second (PKCα) and third (PKCβ) exons in the pre-mRNA transcripts, resulting in the excision of the respective exons. B (i) Lysates were generated from 3 days postfertilization (dpf) wild-type zebrafish larvae injected with standard control morpholino (MO), or PKCα or PKCβ translation-blocking morpholinos (tb MO) before immunoblot analysis for protein expression. A human platelet whole cell lysate (HP WCL) was used as a positive control with α-tubulin used as a loading control. (ii) Densitometric analysis of western blots for translation-blocking morpholino experiments was performed using ImageJ 1.43. Protein levels were adjusted for protein loading and values were standardized to the control protein expression. PKCα tb MO, n = 4. PKCβ tb MO, n = 5. C (i) Lysates were generated from 3 dpf wild-type zebrafish larvae injected with standard control morpholino (MO), or PKCα or PKCβ splice-blocking morpholinos (sb MO) before immunoblot analysis for protein expression. A human platelet whole cell lysate (HP WCL) was used as a positive control, with α-tubulin as a loading control. (ii) Densitometric analysis of western blots for translation-blocking morpholino experiments was performed as described above. PKCα sb MO, n = 3. PKCβ sb MO, n = 4. Densitometry data were compared using a Kruskal–Wallis non-parametric test with Dunn’s post-test, with * indicating < 0.05. Data are shown as mean ± standard error of the mean (SEM).

Download figure to PowerPoint

Laser-injury model of thrombosis in zebrafish larvae

For laser-injury studies, 3 day postfertilization (dpf) morpholino-injected larvae, at a developmental stage of 28 somites of more, were sedated in 0.3× Danieau’s solution containing 0.008% Tricaine [26]. Larvae were mounted on their sides in a petri dish using 1.5% low melting-point agarose made up in 0.3× Danieau’s solution containing 0.008% Tricaine. The mounted larvae were kept submerged in 0.3× Danieau’s solution containing 0.008% Tricaine. For all wounding experiments, the target site was taken to be the ventral wall of a section of the caudal artery above the cloaca (anal pore). Wounds were generated using a MicroPoint laser system (Photonic Instruments Inc., St. Charles, IL, USA) equipped with a coumarin 488 dye-cell. The laser was set to fire two bursts at a rate of 10 Hz with the laser attenuator plate on the microscope unit set to 10 notches in from the maximum filter setting (i.e. lowest power). The laser was administrated through a 40× water-dipping objective. Laser injury and resulting thrombus formation were followed for 10 min by time-lapse microscopy using StreamPix 4.0 digital video recording software (Norpix, Montreal, Canada), with the administration of the laser injury at t = 10 s. Control larvae were processed at both the beginning and end of the experimental session to confirm that the time taken to complete the experiment did not affect the data.

Assessment of morpholino treatment on thrombocyte count using CD41-GFP zebrafish

Morpholino-injected 4 dpf CD41-GFP larvae were euthanized to stop the circulation using 0.3× Danieau’s solution containing 0.8% Tricaine. The larvae were mounted in 1.5% low melting-point agarose and viewed by epifluorescence microscopy. The total number of bright, CD41-positive cells was counted in individual larvae under a 40× water-dipping objective.

Assessment of morpholino treatment on blood cell velocity using pU1-Gal4-UAS-GFP zebrafish

Morpholino-treated 3 dpf pU1-Gal4-UAS-GFP zebrafish larvae were sedated, mounted and imaged as per the laser-injury model, with frames captured every 50 ms for 5 s.

Immunoblotting

Zebrafish larvae (3–4 dpf) were euthanized using 0.3× Danieau’s solution containing 0.8% Tricaine. Larvae were lysed by adding 10 μL of 5× SDS sample buffer per larva, before homogenization and passage through a 25-gauge needle/syringe to complete homogenization and shear genomic DNA. Lysates were heated to 97 °C for 5 min to complete protein denaturation and run on Tris-glycine polyacrylamide gels before transfer to a poly(vinylidene difluoride) membrane. Membranes were blocked with 10% (w/v) bovine serum albumin (BSA) in Tris-buffered saline containing 0.1% (v/v) Tween-20 (TBS-T). Primary antibodies were diluted in 10% (w/v) BSA/TBS-T and incubated with the membranes for 1 h at room temperature, before extensive washing with TBS-T. The application of secondary antibodies followed the same procedure, before ECL detection of bound antibodies.

Data analysis and interpretation

Gene sequence alignment was performed using the EMBL-EBI ClustalW2 web tool (EU). Image analysis was done using ImageJ 1.43 (National Institutes of Health, Bethesda, MD, USA). Data manipulations were performed using Microsoft Excel and GraphPad Prism software (La Jolla, CA, USA). Statistical analyzes were performed using GraphPad Prism software. Results were regarded as significant when < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental procedures
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References
  10. Supporting Information

Zebrafish PKCα and PKCβ share high sequence homology with their human orthologs

The Ensembl database was queried for zebrafish orthologs of human PRKCA (PKCα) and PRKCB (PKCβ), with one ortholog each found for PKCα and PKCβ (Table 1A). A complementary search was made for zebrafish Prkca and Prkcb genes, to confirm these findings. To support the assertion that these zebrafish genes are genuine orthologs of human PKCα and PKCβ, synteny analysis of the appropriate human and zebrafish chromosomes was performed, showing that numerous genes local to both PKCs on the human chromosomes are conserved on their zebrafish counterparts (Fig. S2). The protein sequences for zebrafish PKCα and PKCβ were compared with human PKCα and PKCβ using the EMBL-EBI ClustalW2 sequence alignment tool to assess their homology. Both zebrafish PKCα and PKCβ showed a high percentage sequence identity when compared with their respective human sequences, with 83% and 84% sequence identity, respectively (Table 1A, Fig. S1). Comparing the main regulatory and functional domains of zebrafish and human PKCα and PKCβ (Table 1B) identified higher levels of sequence homology. The PKCα C1A and C1B diacylglycerol (DAG)-binding domains showed 94% and 96% sequence identity to their human orthologs, respectively, whereas the C1A and C1B domains of PKCβ showed 98% and 90% identity to their orthologs, respectively. The catalytic kinase-domains also showed high sequence homology with percentage sequence identities of 90% and 93% for PKCα and PKCβ, respectively. The C2 Ca2+-binding domains are the least homologous, yet still have high percentage sequence identities of 86% and 89% for PKCα and PKCβ, respectively.

Table 1.   Assessment of human and zebrafish protein kinase C alpha (PKCα) and PKC beta (β) orthologs. (A) The Ensembl genome database was queried for the presence of zebrafish orthologs of human PKCα and PKCβ Zebrafish PKCα and PKCβ were assessed by EMBL-EBI ClustalW2 sequence alignment tool to identify the percentage sequence identity to their human orthologues. (B) The functional domains of PKCα and PKCβ were identified from the UniProt protein database and the percentage sequence identity between the human and zebrafish domains was assessed using EMBL-EMI ClustalW2
(A)
Zebrafish HGNC IDZebrafish Ensembl gene IDHuman orthologHuman Ensembl gene ID% Sequence identity
PrkcaENSDARG00000055102PRKCAENSG0000015422983
PrkcbbENSDARG00000022254PPKCBENSG0000016650184
(B)
PKC isoformDomainHuman UniProt ID% Sequence identity
αC1AP17252 [36–86]94
C1BP17252 [101–151]96
C2P17252 [172–260]86
CatalyticP17252 [339–597]90
βC1AP05771 [36–86]98
C1BP05771 [101–151]90
C2P05771 [173–260]89
CatalyticP05771 [342–600]93

Anti-sense morpholinos targeted against zebrafish PKCα and PKCβ efficiently knockdown expression of their respective proteins

To take advantage of the speed and simplicity of using zebrafish to study gene function in thrombosis in vivo, anti-sense morpholinos were designed for zebrafish PKCα and PKCβ. The standard approach of using two, sequence-independent morpholinos for each gene was utilized, as previously described [27]. Morpholinos were designed to target the mRNA translation start sites (Fig. 1Ai) and pre-mRNA splice junctions (Fig. 1Aii). Translation-blocking morpholinos block protein synthesis, allowing the detection of morpholino activity by western blot. To allow detection of splice-blocking morpholino activity by western blot, the design of the splice-blocking morpholinos was intended to introduce premature stop codons via the deletion of an early exon. Both the PKCα and PKCβ splice-blocking morpholinos lead to truncation within the C1A diacylglycerol-binding domain.

Microinjection of PKCα and PKCβ morpholinos into zebrafish larvae at the one to two cell stage led to a marked and specific knockdown of PKCα and PKCβ expression in larvae 3 dpf by both the translation and splice-blocking morpholinos (Fig. 1B,C respectively). The injection of these morpholinos gave rise to limited toxicity at the amounts that gave knockdown of protein expression, with no spontaneous bleeding (data not shown); however, the dual microinjection of PKCα and PKCβ translation-blocking or splice-blocking morpholinos resulted in significant toxicity to the larvae, which manifested as embryonic lethality (data not shown), preventing any analysis of dual knock-down on thrombus formation.

Effect of morpholino treatment on thrombocyte count

When studying thrombosis in transgenic and gene knockout mice, performing a platelet count is necessary to determine whether the thrombosis model is standardized. The morpholino treatment of zebrafish may also affect thrombocyte count, which may impact on the ability to form a thrombus, and we have therefore determined thrombocyte count in larvae for this reason. As such, morpholino-treated larvae derived from CD41-GFP zebrafish were used to assess the total thrombocyte count, with the bright, CD41-positive thrombocytes identified by epifluorescence microscopy. The control morpholino-injected larvae in the translation-blocking and splice-blocking data sets had mean thrombocyte counts of 71.6 and 67.2 thrombocytes per larva, respectively (Table 2A). The PKCβ translation-blocking and splice-blocking morpholinos had no significant effects on thrombocyte count, with mean counts of 80.7 and 66.5 thrombocytes per larva. Likewise, the PKCα splice-blocking morpholino had no significant effect on thrombocyte count with 61.3 thrombocytes per larva; however, the PKCα translation-blocking morpholino caused a significant reduction in mean thrombocyte count to 49.9 thrombocytes per larva. Analysis of blood cell velocity, using pU1-Gal4-UAS-GFP larvae, which express GFP in monocytes allowing visualization of blood cell motion through vessels, demonstrated that neither PKCα nor PKCβ translation-blocking or splice-blocking morpholinos significantly altered the blood flow rate (Table 2A). Because the PKCα translation-blocking morpholino caused a reduction in thrombocyte count, it was not possible reliably to determine the effect of PKCα knockdown on thrombus formation, and so this morpholino was not used further in the present study.

Table 2.   Morpholinos designed against protein kinase C alpha (PKCα) and PKC beta (β) attenuate thrombus formation and thrombus initiation in thrombosis in zebrafish larvae. (A) Thrombocyte counts were determined for whole larvae by manually counting GFP-positive cells in morpholino-injected CD41-GFP larvae (4 dpf). Blood cell velocity was determined by time-lapse microscopy of morpholino-treated pU1-Gal4-UAS-GFP larvae (3 dpf), with images captured every 50 ms. Using ImageJ 1.43, GFP-positive cells were manually tracked across the field of view. Cell velocity was calculated from the measured cell displacement and the time taken to cross the field of view. Data were compared with control by one-way anova (with Dunnett’s post-test). (B) At t = 120 s, thrombus surface area (TSA) was measured as a readout of approximate peak thrombus size and expressed as a percentage of the daily control average. The time to attachment was measured as the time post laser-injury at which the first cell adhered to the site of injury. PKCβ translation-blocking (TB) data were compared with control using Students’t-test. Splice-blocking (SB) data were compared with control by one-way anova (with Dunnett’s post-test)
ParameterDataControl MOPKCα MosPKCβ MOs
Mean ± SEMnMean ± SEMnStatistical significanceMean ± SEMnStatistical significance
  1. SEM, standard error of the mean.

(A)
 Mean thrombocyte  count per embryoTB71.6 ± 6.01949.9 ± 4.819P < 0.0146.3 ± 12.519ns
SB67.2 ± 3.23461.3 ± 4.622Ns66.5 ± 4.026ns
 Mean cell velocity (μm s−1)TB512.6 ± 43.816410.6 ± 34.112Ns488.5 ± 46.618ns
SB375.8 ± 24.529496.7 ± 70.08Ns357.5 ± 30.412ns
(B)
 TSA at 2 mins (% of control)TB100.0 ± 12.81446.3 ± 12.518< 0.01
SB100.0 ± 9.93528.8 ± 9.912P < 0.00145.3 ± 11.920P < 0.001
 Time to initial attachment(s)TB28.0 ± 4.51450.2 ± 7.118< 0.05
SB31.8 ± 3.53543.4 ± 5.712Ns69.4 ± 6.620P < 0.01

Knockdown of PKCβ, but not PKCα, consistently delays the initial adhesion of thrombocytes to the site of injury

To assess the effect of the knockdown of PKCα and PKCβ on thrombus formation, a laser-injury model of thrombosis was used, which was based upon descriptions in previous studies in mice [28] and zebrafish [3–5]. The transparency of zebrafish larvae lend them self to this technique owing to the easy visualization of the larval circulation via microscopy, without the need for complex surgery as in the murine model. Using intravital light microscopy, the attachment of the initial cell to the site of laser damage can be identified and subsequent thrombus growth followed over time, allowing for precise measurement of thrombus dynamics post laser-injury. In the model applied here, an ablating dye laser was used to wound the ventral wall of the caudal artery at a site above the cloaca (anal pore) in morpholino-treated 3 dpf larvae. Thrombus formation was followed over 10 min.

As thrombus formation is dependent on thrombocyte adhesion to the site of damage, followed by full activation and additional thrombocyte recruitment, the ability to observe clearly the adhesion of the initial cell in zebrafish provides useful information with regards to thrombus initiation and thrombocyte–endothelial interactions [5]. The time to adhesion of the initial cell for the control data in the translation-blocking and splice-blocking data sets was 28.0 and 31.8 s, respectively. Both the PKCβ translation-blocking and splice-blocking morpholinos significantly prolonged the time to adhesion, to 50.2 and 69.4 s respectively; however, the PKCα splice-blocking morpholino did not have any significant effect (Table 2B).

Knockdown of PKCα and PKCβ expression in zebrafish larvae attenuates thrombus formation in a laser-injury model

To assess thrombus formation in morpholino-treated larvae, an endpoint of thrombus surface area (TSA) was used (as described by O’Connor et al. [5]). The use of additional endpoints based around time to occlusion/thrombus dissolution (as described by Jagadeeswaran et al. [29]) were explored, but considered too insensitive as a result of relatively few thrombi becoming occlusive (data not shown). In the control groups, TSA peaked around 2–3 min post laser-injury at 1279 and 949 μm2 for the controls for translation-blocking and splice-blocking data sets, respectively (Fig. 2A(i/ii),B(i/ii)). The PKCβ translation-blocking morpholino caused a significant reduction in thrombus formation over time compared with the control (< 0.01, Fig. 2A). To control for daily experimental variability, the extent of thrombus formation at 2 min was also compared (as described by O’Connor et al. [5]). The PKCβ translation-blocking morpholino caused a significant reduction in TSA at 2 min by 53.7% (Table 2B). This latter observation was reproduced in larvae injection with the PKCβ splice-blocking morpholino, with a significant attenuation of thrombus formation by 54.7% (Table 2B); however, a significant reduction in thrombus formation over time was not seen compared with the control (Fig. 2B). The PKCα splice-blocking morpholino caused a significant attenuation of thrombus formation over time compared with the control (< 0.05, Fig. 2B) and significantly reduced thrombus formation by 71.2% at 2 min (Table 2B).

image

Figure 2.  Morpholinos designed against protein kinase C alpha (PKCα) and PKC beta (β) attenuate thrombus formation in zebrafish larvae. Morpholino-injected wild-type larvae 3 days postfertilization (dpf) were wounded by laser on the ventral wall of the caudal artery, with the resulting thrombus formation followed over 10 min by differential interference contrast (DIC) time-lapse microscopy. A (i) Thrombus formation in zebrafish larvae injected with 6 ng PKCβ (open squares, n = 18) translation-blocking morpholino (tb MO) was compared with thrombus formation in standard control morpholino-injected siblings (black squares, n = 45). Data were compared using two-way anova with < 0.01 for control vs. PKCβ. (ii–iii) Representative images of thrombus formation at 120 s post-injury in larvae injected with (ii) control and (iii) PKCβ translation-blocking morpholinos. B (i) Thrombus formation in zebrafish larvae injected with 2.5 ng PKCα (gray circles, n = 12) or 4 ng PKCβ (open circles, n = 14) splice-blocking morpholinos (sb MO) were compared with thrombus formation standard control morpholino-injected siblings (black circles, n = 33). Data were compared by two-way anova with < 0.05 for control vs. PKCα. All data are shown as mean ± standard error of the mean (SEM). (ii–iv) Representative images of thrombus formation at 120 s post-injury in larvae injected with (ii) control, (iii) PKCα and (iv) PKCβ splice-blocking morpholinos.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental procedures
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References
  10. Supporting Information

The zebrafish has been shown to be a highly relevant model with which to model thrombosis [4,5,30]. The zebrafish model is attractive for quickly assessing gene function in thrombosis as a result of the close homologies between zebrafish and human thrombosis and hemostasis, the conservation of the signaling pathways involved, the transparency of the developing larvae, the ease of knocking-down gene function using morpholinos and the rapid development of the larvae to a stage when thrombosis can be studied [3,24].

In the present study, we aimed to use a zebrafish laser-injury model of thrombosis to perform a comparative analysis of the roles of PKCα and PKCβ in thrombosis. The laser-injury model in zebrafish has several advantages over the equivalent model in mice, as it is able to provide real-time information on thrombus formation with high optical and temporal resolution, yet without needing the surgical expertise and licensing requirements of the murine model. We have previously shown, using PKCα−/− mice, that PKCα has a positive regulatory role in platelet function and thrombus formation, in vitro and in vivo [17]. In the present study, we were able to reproduce these results using the zebrafish model and anti-sense morpholinos. This was to both validate the murine study and to provide confidence when comparing the roles of PKCα and PKCβ. Previous studies into the role of PKCβ in platelet function have been limited to integrin regulation [23], although a positive regulatory function for PKCβ in thrombosis in vitro [18] has recently been identified. Zebrafish PKCα and PKCβ possess high sequence identity to their human orthologs (> 83%), suggesting that gene function is likely to be conserved from zebrafish through to human thrombosis. The high percentage sequence identity of the catalytic kinase domains (> 90%) also suggests that zebrafish will be a suitable model with which to perform preliminary in vivo screens of PKC inhibitors as novel anti-thrombotic drugs.

To control for any non-specific effects that may affect the interpretation of the results, whole larva thrombocyte counts were performed using morpholino-treated CD41-GFP zebrafish larvae, and blood cell velocity measurements were calculated using morpholino-treated pU1-Gal4-UAS-GFP zebrafish larvae. Clearly, alterations in either of these parameters, which may result from off-target effects of the morpholinos or from specific effects of them on tissues other than thrombocytes, would make interpretation of in vivo thrombus formation difficult. However, none of the morpholinos used significantly affected the blood flow. With regard to thrombocyte counts, neither of the PKCβ morpholinos nor the PKCα splice-blocking morpholino affected the counts; however, the PKCα translation-blocking morpholino caused a significant reduction. This is in contrast to PKCα−/− mice where there is no change in the resting platelet count (data not shown). This would suggest that the reduced thrombocyte count in the PKCα translation-blocking morpholino-injected larvae may be developmental [31,32], and required the exclusion of the thrombus formation data using this morpholino. For the rest of this study, therefore, only the splice-blocking morpholino for PKCα was used for analysis of thrombus formation, whereas both translation-blocking and splice-blocking morpholinos of PKCβ were used.

Both of the PKCβ morpholinos attenuated thrombus growth in vivo and peak thrombus size was consistently attenuated with both of these morpholinos. Interestingly, the time to attachment of the first cell is also consistently delayed with the PKCβ morpholinos when compared with the control. This could be indicative of reduced thrombocyte activation in response to the damaged endothelium as PKCβ−/− platelet activation is attenuated upon adhesion to collagen under in vitro flow conditions [18]; however, as injection of morpholinos in the one to two cell stage causes widespread knockdown of gene function, it is possible that a dysfunctional endothelium could be contributing to the effect. Considering that no effect on time to attachment is seen with the PKCα splice-blocking morpholino, yet PKCα−/− platelets in the in vitro thrombosis model [17,18] have a similar phenotype to PKCβ−/− platelets, this could be the case. However, platelets from PKCβ null mice have been shown to have defective αIIbβ3 integrin outside-in signaling [23], which is not the case for PKCα null platelets [17]. There is also recent evidence for a lack of functional redundancy in vitro for these two kinases, from inhibitor studies in human platelets and from PKCα−/− and PKCβ−/− mouse platelets [18], in regulating calcium signaling, α-granule secretion and thrombus formation on collagen in vitro. Here, the difference between the PKCα splice-blocking morpholino and PKCβ morpholinos in time to attachment may therefore reflect these observations in mammalian systems, and demonstrate in vivo for the first time, the absence of redundancy between the two isoforms in regulating thrombus formation.

In conclusion, we have demonstrated positive non-redundant regulatory roles for PKCα and PKCβ in thrombus formation in vivo using a zebrafish laser-injury model of thrombosis. While PKCα was shown to primarily regulate thrombus growth, this first in vivo description of the role of PKCβ in thrombosis demonstrated that it may principally play a role in regulating thrombus initiation, whilst also contributing to thrombus growth.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental procedures
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References
  10. Supporting Information

The authors would like to thank Elizabeth Aitken for invaluable technical support. The work was supported by a programme grant from the British Heart Foundation to A.W.P. (grant no. RG/05/015). C.M.W. was a supported by British Heart Foundation studentship (FS/06/044). A.W.P. was supported by a BBSRC Research Development Fellowship (BB/E024637/1).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental procedures
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental procedures
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interest
  9. References
  10. Supporting Information

Figure S1. Expected truncation of PKCα and PKCβ protein sequences after the action of the relevant splice-blocking morpholinos.

Figure S2. Conserved synteny between human and zebrafish PKCa and PKCb. The syntenic analysis was performed on the human and zebrafish chromosomes containing PKCa (PRKCA) and PKCb (PRKCB) using SyMAP v3.4 (Soderlund et al., Nucleic Acids Res, 2011; 39: e68) with syntenic hits between chromosomes represented by intersecting lines. The Ensembl zebrafish and human genome databases were manually queried to determine the local genomic neighbours (genes within 2 MB) that are conserved between human and zebrafish PRKCA and PRKCB, with the conserved genes displayed.

FilenameFormatSizeDescription
JTH_4520_sm_FigS1.tif7478KSupporting info item
JTH_4520_sm_FigS2.tif3139KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.