Short Report | Open | Published:
Ryanodine receptors, a family of intracellular calcium ion channels, are expressed throughout early vertebrate development
BMC Research Notesvolume 4, Article number: 541 (2011)
Calcium signals ([Ca2+]i) direct many aspects of embryo development but their regulation is not well characterised. Ryanodine receptors (RyRs) are a family of intracellular Ca2+ release channels that control the flux of Ca2+ from internal stores into the cytosol. RyRs are primarily known for their role in excitation-contraction coupling in adult striated muscle and ryr gene mutations are implicated in several human diseases. Current evidence suggests that RyRs do not have a major role to play prior to organogenesis but regulate tissue differentiation.
The sequences of the five zebrafish ryr genes were confirmed, their evolutionary relationship established and the primary sequences compared to other vertebrates, including humans. RyRs are differentially expressed in slow (ryr1a), fast (ryr3) and both types (ryr1b) of developing skeletal muscle. There are two ryr2 genes (ryr2a and ryr2b) which are expressed exclusively in developing CNS and cardiac tissue, respectively. In addition, ryr3 and ryr2a mRNA is detectable in the initial stages of development, prior to embryonic axis formation.
Our work reveals that zebrafish ryr genes are differentially expressed throughout the developing embryo from cleavage onwards. The data suggests that RyR-regulated Ca2+ signals are associated with several aspects of embryonic development, from organogenesis through to the differentiation of the musculoskeletal, cardiovascular and nervous system. These studies will facilitate further work to explore the developmental function of RyRs in each of these tissue types.
Transient changes in the concentration of intracellular calcium ions ([Ca2+]i) act as a powerful signal that is crucial for the establishment of form and function in the embryo. Detailed imaging studies have revealed that the spatial and temporal organisation of Ca2+ signals during embryogenesis are associated with many of the major phases of development, from early cell division to the differentiation of tissues. Despite their importance little is known about the generation and regulation of embryonic Ca2+ signals. A comprehensive understanding of the pathways that regulate [Ca2+]i during development is essential to understand the functional relevance of these signals in the embryo.
Ryanodine receptors (RyR) are a family of intracellular Ca2+ release channels that regulate the entry of Ca2+ into the cytosol from the intracellular organelles (the endoplasmic and sarcoplasmic reticulum). The RyR is a large homotetrameric protein (approximately 2,200 kDa), each subunit is comprised of a large N-terminal cytoplasmic domain which modulates the gating of the channel, as well as luminal and transmembrane spanning (TM) domains. In mammals, there are three distinct ryr genes (ryr1, ryr2 and ryr3) that encode three differentially expressed RyR proteins. RyR1 and RyR2 are expressed predominantly in skeletal and cardiac muscle respectively, whilst RyR3 is found in many tissues at relatively low levels . The primary role of the RyR is to increase [Ca2+]i during excitation-contraction coupling (E-C coupling) in both skeletal and cardiac muscle. In humans, mutations in the ryr1 and ryr2 genes cause skeletal myopathies and cardiac disease respectively . The generation of mouse knockout lines has provided some insight into the role of the receptors in the developing tissues of intact animals. Homozygous mice from ryr1(−/−) (skrrm1) and ryr2(−/−) knockout lines display gross morphological defects in either the skeletal muscle (ryr1) or heart tube (ryr2) and die at the perinatal or embryonic day 10 (E10) stages respectively [3, 4]. In contrast, ryr3(−/−) knockout mice appear to have no gross developmental defects and evidence suggests that RyR3 act to augment the [Ca2+]i response of the other RyR isoforms in striated muscle [5, 6]. The observation that RyR expression does not occur until relatively late in mammalian development , coupled with the fact that knock out lines are not lethal at very early stages has been interpreted as indicating that RyRs do not function during initial development.
The zebrafish has been used extensively as a model for vertebrate development. The rapid development ex utero and embryonic transparency has proved advantageous for imaging the spatial and temporal organisation of Ca2+ signals. These signals are involved in many of the early embryonic events; the initiation of fertilisation (0 hours post fertilisation, hpf), the early cell divisions associated with the cleavage period (up to 2 hpf) and the more extensive cellular rearrangements that occur in the blastula period (up to 5 hpf) (as reviewed in ). Evidence suggests that the release of Ca2+ from intracellular stores via the phosphatidylinositol (PI) signalling pathway is largely responsible for these early transient changes in [Ca2+]i . RyRs have not been implicated in very early developmental events and their expression during these stages has not been documented. Fluxes in the levels of embryonic [Ca2+]i continue to occur throughout gastrulation (up to 10 hpf) (as reviewed in ). Initially changes in [Ca2+]i occur as localised events but, as gastrulation progresses, co-ordinated waves of Ca2+ signalling appear across the embryo. These later signals are proposed to coordinate a wide range of cellular movements (epiboly, involution, convergence and extension) that give rise to the embryonic body plan. Release of Ca2+ from intracellular stores via the PI signalling pathway is again implicated at this stage; however, the contribution to gastrulation from other Ca2+ signalling pathways remains undefined. Finally, the segmentation period (from 10 up to 24 hpf) is characterised by organogenesis and the emergence of the body systems. The Ca2+ signals that occur during segmentation are again more localised and typically associated with developing tissues. Transient changes in [Ca2+]i have been recorded within the nervous system, somites and cardiac tissue [10–12]. Several studies in zebrafish have shown that inhibition of ryanodine receptor function, using both pharmacological and genetic inhibitors, leads to impaired excitation-contraction coupling and gross morphological defects in the skeletal muscle, suggestive of a role in the development of this tissue [11, 13, 14].
This study set out to acquire a more comprehensive understanding of ryanodine receptor expression in early vertebrate development, using the zebrafish as an in vivo model. Our initial work confirmed the sequence of the five zebrafish ryr genes, established their evolutionary relationship to those in other vertebrate species and provided a direct comparison between the structural features of the primary protein sequences found in the zebrafish and mammals. An overview of ryr gene expression during zebrafish embryogenesis will inform work aimed at establishing the developmental significance of this family of Ca2+-release channels. Therefore we conducted a comprehensive temporal and spatial analysis of ryr mRNA expression in the embryo using a combination of semi-quantitative PCR and wholemount in situ hybridisation. We observed strong maternal expression of ryr mRNA (ryr3 and ryr2a) during the cleavage and blastula periods suggestive of a novel role in early development. At 24 hours post fertilisation (hpf) ryr1a, ryr1b and ryr3 are expressed in skeletal muscle, whereas ryr2a is localized to the central nervous system (CNS) and ryr2b is found exclusively in the cardiac muscle. Our study suggests that RyR channels have a role in early development prior to organogenesis as well as in the differentiation of different cell types.
Wildtype (WT) zebrafish strains (Tubingen and Tupfel long fin) were bred and raised in-house at the zebrafish facility of Queen Mary College, University of London, UK, as described previously . Smoothened (smo) mutants were received as a gift from Prof. Simon Hughes (King's College London, UK). Embryos were collected by natural spawning and staged according to Kimmel and colleagues , given in the text as standard developmental time at 28.5°C (hours post fertilisation, hpf). Work on zebrafish embryos (prior to independent feeding) is exempt under the U.K. Animals (Scientific Procedures) Act 1986 and does not require ethical approval.
Genomic analysis and gene prediction of ryr genes
To identify zebrafish ryr genes encoding RyRs as described in Hirata and colleagues , protein sequences corresponding to the human RyR family [http://www.ensembl.org/Homosapiens/familyview?family=ENSF00000000736] were used as template for searching in GenBank and the zebrafish genome resources at Ensembl http://www.ensembl.org/Danio_rerio/;Zv7Ensemblassembly using a protein tBLASTn approach. Several hits were identified, which were then compared to the 3'-UTR sequence of the EST to identify the corresponding genomic region. The gene predictions program, Genwise (http://www.ebi.ac.uk/Wise2/) was applied to reveal genomic contigs and full sequence relating to the ryr1a, ryr2a, ryr2b and ryr3 in zebrafish (Table 1). For ryr1a and ryr3, based on the tBLASTn result, the Accession numbers (Table 2) and annotated gene sequences for these isoforms were selected. For ryr2a, owing to the limitations to the resources available on the database, 5 contigs (NA_1034, NA_1216, NA_3083, NA_1397 and NA_1713) were identified with nucleotide similarity to the human ryr2, based on the approach described above. Bioinformatics database search approaches using human ryr2 did not reveal any fully annotated ryr2b gene sequences in zebrafish. However, database blast results located a gene (EntrezGene Name LOC568506 on chromosome 17:18.7 m) encoding 805 residues. The partial ryr2b gene located in a 19.98 kb region in the Ensembl Zv7 assembly shared very limited conserved symmetry with the human ryr2. An alignment approach revealed that the zebrafish ryr2b gene was reverse transcribed and the region of 19.98 kb encoding 805 residues at the C-terminus end shows high peptide identity to the human RyR2 C-terminus. To help to find the missing large N-terminal cytoplasmic domain of the RyR2b peptide, Genscan features on Ensembl Zv7 assembly were employed to reveal an extra 1014 residues, giving a total of 1819 residues from the C-terminal end of the peptide. In order to search further for the missing portion of the RyR2 peptide towards the N-terminus, a genomic region of 107,783 bp spanning the C-terminal end of ryr2b gene to the start of neighbouring actn2 gene [Ensembl: ENSDARG00000071090], was reverse complemented using REQSEV program http://bioweb.pasteur.fr/seqanal/interfaces/revseq.html before performing a pairwise BLAST alignment with Gasterosteus aculeatus (stickleback) RyR2 peptide sequence on NCBI (http://www.ncbi.nlm.nih.gov/blast/bl2seq/wblast2.cgi) to confirm its sequence identity. Based on the alignment result, the full length zebrafish RyR2b peptide was first constructed with reference to Takifugu rubripes (pufferfish) and stickleback RyR2 peptide sequences using the Genewise program (http://www.ebi.ac.uk/Wise2/). The complete RyR2b peptide was finalised by the alignment of the translated genomic region using SHOWORF program http://bioweb.pasteur.fr/seqanal/interfaces/showorf.html. The sequence for the ryr2b gene was deposited in EMBL Nucleotide Sequence Database (Accession no. FR822741). For ryr1b, sequence was obtained via the NCBI CoreNucleotide search website [NCBI - AB247454] and isoform-specific PCR primers (set 3) were used, as described previously . Isoform-specific PCR primers for ryr1a, ryr2a, ryr2b and ryr3 were designed by retrieving the nucleotide sequences based on divergent regions of the RyRs identified from peptide sequence (Figure 1). These regions were re-confirmed by both BLAST against different databases (NCBI nucleotide blast, Ensembl and ZFIN) and DNA sequencing.
Protein sequences equivalent to the RIH_assoc (pfam08454) and RR_TM4-6 (pfam06459) domains in the five zebrafish RyR protein sequences (RyR1a, RyR1b, partial RyR2a, RyR2b and RyR3) and other well characterised vertebrates RyR homologue sequences were extracted from Ensembl and GenBank databases (Table 2). The RyR sequences were pre-aligned using the ClustalW alignment program to remove any gaps generated within the sequences, followed by a multiple RyR sequence alignment using the T-coffee program available from the European Bioinformatics Institute webpage (http://www.ebi.ac.uk/Tools/t-coffee/index.html) with the default parameters. The output from the multiple sequence alignment result obtained from T-coffee was used subsequently as the template for the generation of a Guide Tree using the equal angle method available on the SplitsTree4 program with its default parameters (http://www.splitstree.org/forfreedownload).
Semi-quantitative end-point PCR
Total RNA was isolated from nine stages of zebrafish development: 4- to 128-cell stage (1-2.25 hpf), 50%-epiboly stage (5.3 hpf), 100%-epiboly stage (10 hpf), 3- to 6-somite stage (11-12 hpf), 18-somite stage (18 hpf), 24 hpf, 72 hpf, 144 hpf stages and a 6-month old adult, using TRIZOL® Reagent (Invitrogen). Total RNA (5 μg) was subjected to 30 minutes treatment with DNase I (Promega) at 37°C prior to cDNA synthesis performed using a Cloned AMV First-Strand cDNA Synthesis kit (Invitrogen) with random hexamers. For semi-quantitative end-point PCR, 1.5 μg of cDNA templates made from the developmental series were used alongside with the five sets of isoform-specific oligonucleotide primer pairs for amplification and normalised against the expression of β-actin (Table 3).
Whole mount in situ hybridisation
In situ hybridisation of whole-mounted zebrafish embryos was performed as described previously . Briefly, ryr1a, ryr1b, ryr2a, ryr2b and ryr3 sense and anti-sense digoxygenin (DIG) labelled RNA probes covering 1,405 bp, 1,361 bp, 1,313 bp, 1,266 bp and 1,549 bp, were generated (Table 3). Briefly recombinant vector templates (pGEM-T Easy-ryrs) were linearised with an appropriate restriction enzyme and then subjected to phenol/chloroform purification and alcohol precipitation. Purified, linear DNA (0.5 μg) was used for the in vitro transcription reaction (a 20 μl reaction contained 1X Transcription Buffer (Invitrogen), 2 μl DTT (0.1 M), 2 μl nucleotide mix (1 mM GTP, 1 mM ATP, 1 mM CTP, 0.65 mM UTP and 0.35 mM DIG-11-UTP), 50U placental ribonuclease inhibitor and 10U of RNA polymerase). The mixture was incubated at 37°C for 4 hours before the removal of the original DNA template by incubating with 2U of DNase at 37°C for 1 hour. DIG-labelled cardiac myosin light chain 2 (cmlc2;), myogenic differentiation (myoD;), nkx2.5 () and fluorescein-11-UTP labelled myosin heavy chain 1 (myhz1;) anti-sense probes were also prepared under the same approach and used as positive markers.
Zebrafish embryos were fixed in 4% paraformaldehyde (PFA)/phosphate buffered Triton X100 (PBT) at 4°C overnight or at room temperature (RT) for 2 hours, then washed in 100% methanol and incubated at −20°C for at least 30 minutes. Embryos were rehydrated in a methanol gradient and treated with 10 μg/ml of Proteinase K/PBT for 1 to 20 minutes at RT depending on their stages prior to fixation in 4% PFA/PBT for 20 minutes at RT (embryos before or at 50% epiboly stage were not treated with Protease K or re-fixed). Embryos were washed with PBT and pre-hybridised in hybridisation mix (50% formamide, 5X SSC (0.75 M NaCl and 75 mM triNa citrate) at pH 5.0, 500 μg/ml yeast RNA, 50 μg/ml heparin, 0.1% Tween20) for 2 hours at 65°C. The mixture was replaced by fresh hybridisation mix containing DIG-labelled or Fluorescein-labelled RNA probe at the appropriate dilution and incubated at 65°C overnight. The post hybridisation washes were carried out at 65°C. The excess probe was removed by sequential washing in 25% formamide in 2X SSC, 2X SSC, 0.2X SSC at 65°C and finally in PBT at RT. Embryos were incubated in maleic acid buffer (0.1 M Maleic acid pH 7.5 and 0.15 M NaCl) containing 2% Blocking Reagent (Roche; MAB) at RT for a minimum of 1 hour. Staining was carried out by replacing the MAB solution with the 1:5000 MAB diluted alkaline phosphatise (AP) conjugated anti-DIG antibody (Roche) or 1:4000 MAB diluted AP conjugated anti-Fluorescein antibody (Roche) at 4°C overnight. The embryos were washed with PBT and then detection buffer (0.1 M Tris-HCl pH 9.5, 0.1 M NaCl and 50 mM MgCl2 and 0.1% Tween20) at RT. The AP substrates used for colour development were BM purple (Roche), Fast Red tablets (Roche) and NBT/BCIP (Roche). The colour reaction was terminated by washing in PBT and embryos were re-fixed in 4% PFA/PBT for 2 hours at RT.
Wholemount immunocytochemistry (WICC) was conducted as described previously . Briefly, embryos were fixed overnight at 4°C and all subsequent steps were performed at RT. Embryos were incubated in blocking buffer prior to incubation in primary antibody in 1% goat serum in phosphate buffered saline supplemented with 0.8% triton X100 (PBST) at room temperature. The primary antibodies used were 34 C, F59 and MF20 at dilutions of 1:250, 1:10 and 1:100, respectively. Embryos were rinsed in phosphate buffer with 0.8% triton (PBT) and incubated in a 1:1000 dilution of goat anti-mouse IgG Cy™-5 linked secondary antibody (Amersham) made up in 1% goat serum in PBT overnight. Embryos were rinsed in PBT and stored in 50% glycerol/50% PBS.
For the double immuno-labelling (WISH/WICC), the WISH labelled embryo was initially developed using reagents as described above, followed by antibody labelling using a 1:10 dilution of F59 and a 1:5000 dilution of goat anti-mouse horseradish peroxidase (HRP) conjugated secondary antibody (Merck) carried out as described in the Zebrafish Book . Colour development by HRP was activated by the addition of 0.01% v/v DAB and 6% v/v H2O2. Reaction was terminated by the addition of PBS and fixed overnight in 4% PFA at 4°C. For double-fluorescent labelling, the WISH labelled embryo was initially developed using Fast Red, followed by antibody labelling as described above.
Embryos were mounted in 100% glycerol and for flat mounting the yolk of the stained embryo was initially removed. Brightfield and fluorescent single or Z-stack images were collected using a X10, X20, X63 oil immersion or X100 water immersion objectives on a Zeiss LSM 510 microscope and LSM Image Examiner software. LSM Image Browser (Version 4,2,0,121) software was used for image post-processing such as preparing three dimensional (3D) projected cross-sections from the acquired Z-stacks of images. For the observation of the Cy™5, Fast Red and Alexa Fluor® (Alexa Fluor® 488 nm) staining, argon lasers 633 nm, 543 nm and 488 nm lines and LP 650, LP 560 and BP 505-550 emission filter(s) were used, respectively.
The relationship between ryanodine receptor genes in zebrafish and those of other vertebrates
We conducted a tBLASTn search of the zebrafish genomic database (Zv7 Ensembl), using the protein sequences for human RyR1, RyR2 and RyR3 as queries, and detected four zebrafish RyRs homologues: ryr1a, ryr1b, ryr2b and ryr3 (as summarised in Table 1). ryr2a was identified from scaffold sequences by blasting the human RyR2 sequence in the updated NCBI database. Despite the fact that the sequence obtained from ryr2a is not a completed version based on the available information in current databases, the five genomic contigs (NA_1034, NA_1216, NA_3083, NA_1397 and NA_1713) identified to date show high similarity to other vertebrate RyR orthologues. Zebrafish ryr gene sequences display significant homology to their respective human isoforms (Table 1).
The evolutionary relationship of vertebrate RyR genes, including the five zebrafish sequences was examined (Figure 2). Three major monophyletic clades were identified that correspond to RyR1, RyR2 and RyR3 subfamilies. We found that the zebrafish RyR1a, RyR1b, RyR2a, RyR2b and RyR3 all clustered with their respective gene families into the three monophyletic clades . The zebrafish RyR1a and RyR1b cluster to their respective subgroups with those of the other teleost species, Gasterosteus aculeatus (stickleback), Oryzias latipes (medaka) and Takifugu rubripes (pufferfish). We identified two ryr2 genes in zebrafish, which currently appears to be the only species in which this gene duplication event can be observed. The zebrafish ryr2a gene appears to be more closely related to other teleosts (i.e. stickleback and pufferfish) whereas ryr2b gene is more diverged. Although most teleosts appear to have two ryr3 genes, we confirmed previous studies which report a single zebrafish ryr3 gene [11, 13, 24].
Zebrafish RyRs contain many of the conserved structural domains with similarities to other vertebrates
The molecular structure of the RyR protein has been explored extensively, partly in the drive to understand its regulation. Each monomeric RyR protein is approximately 5,000 amino acids in length with a molecular mass of 565 kDa. The receptor has a large N-terminal cytoplasmic domain containing many regulatory binding sites (as reviewed in [25, 26]) that modulate the gating of the channel pore located in the C-terminus. The N-terminus cytoplamic domain of RyR interacts with a host of regulatory proteins, such as calstabin and calmodulin. Physiological modulators of RyR function include ATP, Ca2+, Mg2+, cyclic ADP ribose, posttranslational modifications (e.g. phosphorylation, oxidation) and pharmacological substances (e.g. ryanodine, caffeine) .
We conducted an analysis to explore whether the zebrafish RyR protein sequences contain characterised conserved domains that may contribute to receptor regulation and their comparison to their human counterparts was analysed (Figure 1a). MIR (Mannosyltransferase, Inositol 1,4,5-trisphophate receptor (IP3R) and RyR [pfam02815]) and RIH (RyR and IP3R Homology [pfam01365]) domains were identified within the N-terminal of all five zebrafish RyRs. The MIR and RIH domains are common to all the members of the intracellular Ca2+-release channel super family [28, 29]. The MIR domain has been suggested to have a ligand transferase function and the RIH domain may form a binding site for IP3; however, very little is known regarding their role in receptor regulation to date . All of the zebrafish RyRs contain three SPRY domains (SPla and the RyR [pfam00622]), which have been proposed to interact with voltage gated channels [30, 31]. Each zebrafish RyR also contains four copies of the RyR domains (RyR repeated domain [pfam02026]), a sequence unique to these channels . Furthermore, zebrafish RyRs contain the eukaryotic RIH associated (RyR and IP3R Homology associated [pfam08454]) domain, which currently has no known function. EF-hand motifs may have a functional significance in activation of the channel by Ca2+ themselves. Putative Ca2+ binding sequences, EF1 and EF2 , have been identified in the receptor and are thought to be the major Ca2+ regulatory sites. We identified similar putative EF-hand motifs towards the C-terminal of the receptor and these appear to be conserved in all zebrafish and human isoforms. The RyR pore determines the conductance and ion selectivity of the channel; however, the structure of the TM and pore forming region is still unresolved. There are eight proposed TM sequences, of which the last six are suggested to form the Ca2+ release channel . According to the topological model by Du and colleagues six to eight TM sequences (i.e. M4a/M4b, M5, M6, M7a/M7b, M8 and M10) were identified, with the M9 (pore segment) inserted between the M8 and M10 TM segments. This M9 region is proposed to act as the selectivity filter allowing Ca2+ to transverse the membrane . All five zebrafish RyRs contain RR TM 4-6 domains (RyR TM 4-6 [pfam06459]). Comparison of sequence in the core pore-forming region demonstrate that all five zebrafish RyR sequences are highly conserved with other vertebrate species, although it should be noted that there are subtle changes at the single amino acid level (Figure 1b).
Zebrafish ryr mRNA expression can be detected from the earliest stages of development through to adulthood
We examined the temporal expression of the ryr genes during nine key stages of zebrafish development using semi-quantitative end-point PCR (Figure 3a). Expression of ryr1a was detected from 11-12 hpf (3- to 6-somite stage) through to adulthood, whereas ryr1b expression was evident from 18 hpf (18-somite stage) onwards. Results also showed that levels of mRNA expression for both ryr1a and ryr1b were at the highest in 144 hpf and adult stages. The ryr2a gene showed strong maternal expression at 1-2 hpf (4- to 128-cell stage) and zygotic expression was observed from 18 hpf through to adulthood. Expression of ryr2b mRNA was detected from 18 hpf through to adulthood with relatively weak expression observed at 24 hpf and 144 hpf stages. Finally, ryr3 maternal expression was observed at 1-2 hpf, followed by weak zygotic mRNA expression from 5.3 hpf (50% epiboly stage) through to 18 hpf (18-somite stage), after which the levels became much stronger ryr1a , ryr1b and ryr3 are expressed exclusively in skeletal muscle during segmentation period
We examined the spatial and temporal distribution pattern of the zebrafish ryr mRNA in the developing embryo from 12 to 24 hpf of development using WISH. The expression of ryr1a, ryr1b and ryr3 in the developing somites was compared to the myogenic determinant (myoD), one of the earliest markers of myogenic commitment with a key role in regulating muscle differentiation (Figure 4a) .
We observed low levels of ryr1a expression in the adaxial cells located on either side of the notochord beginning at 11 hpf (Figure 4a). Our findings demonstrate that in slow muscle ryr1a appears at 11 hpf, 1 to 2 hpf prior to fast muscle ryr1b mRNA expression. After the lateral migration of the slow muscle precursors is completed, the expression of ryr1a mRNA is clearly visible in the superficial layer of slow muscle (Figure 4b). There is conflicting evidence that by 24 hpf ryr1a mRNA is confined either to slow muscle fibres , or can be detected at low levels throughout the somite. In this study we found that ryr1a mRNA co-localises with a known marker of slow muscle fibres at 24 hpf (Figure 4b). Furthermore, ryr1a mRNA staining was not detected in the somites of smoothened (smo−/−) embryos, a mutant line which lacks slow muscle and the muscle pioneers (Figure 4c). Thus, we conclude that ryr1a mRNA is expressed exclusively in embryonic slow muscle.
We observed low levels of ryr1b expression in adaxial cells adjacent to the notochord beginning at 12-13 hpf (Figure 4a). At 18 and 24 hpf, strong ryr1b expression was detected throughout the somites in a pattern analogous to that of myoD, suggestive of its presence in both fast and slow muscle. Currently there is conflicting reports suggesting that ryr1b expression is confined either to the fast muscle  or appears in both fast and slow muscle . In the later study the observation that ryr1b mRNA expression is found in both muscle types was attributed to an artefact arising as a consequence of contaminated tissue . In order to clarify our findings we performed double-fluorescent labeling experiments, to observe ryr1b expression in the presence of a slow muscle marker. In contrast to previous work which used a ryr1b in situ probe that targeted the relatively conserved pore forming region at the C-terminus we generated a probe that recognised a more variable region at the N-terminus of the receptor . Our data revealed that ryr1b mRNA is localised to both fast and slow developing skeletal muscle (Figure 4b, Additional file 1: Figure S1).
Maternal expression of ryr3 in the dividing cells of 2 to 2.25 hpf (64- and 128-cell) embryos was confirmed by wholemount in situ hybridisation (WISH) (Figure 3b); however, the low level expression of ryr3 from 5.3 to 18 hpf detected in the PCR analysis was not confirmed by WISH (data not shown). In this study ryr3 mRNA expression was detected in the skeletal muscle from 24 hpf using WISH (Figure 4a,b), with stronger expression in the anterior compared to developing posterior somites (Figure 4a). There is conflicting evidence that ryr3 mRNA in zebrafish is expressed exclusively in slow and fast skeletal muscle from 14 hpf  or in many tissues including the CNS during the somitogenesis [14, 36]. The discrepancies in the expression patterns of ryr3 mRNA may be explained in part by the target sequence used for probe synthesis in previous studies. The ryr3 clone (019-D04-2) used previously targeted the conserved pore-forming region at the C-terminus of the receptor which shares high similarity to both the ryr1a and ryr1b sequence, this raises the possibility that cross hybridisation with ryr1 isoforms occurred . In the current work the ryr3 probe was designed to a more divergent region located in the N-terminus of the receptor. In this study ryr3 mRNA expression was not detected in the Kuppfer's vesicles, adaxial cells at 12 hpf or in any region of the CNS at 24 hpf (data not shown). Double staining experiments revealed that ryr3 mRNA expression was confined to the fast skeletal muscle at 24 hpf (Figure 4b, Additional file 1: Figures S1 and Additional file 2: Figure S2). Thus we conclude that ryr3 mRNA is expressed exclusively in the fast skeletal muscle of embryos at 24 hpf.
ryr2a and ryr2b are expressed in the developing zebrafish CNS and heart, respectively
Expression of ryr2a mRNA was detected initially in the CNS of zebrafish embryos starting at 24 hpf (Figure 5a). At 24 and 48 hpf, ryr2a mRNA expression can be observed at several sites in the developing forebrain (telencephalon and diencephalon), midbrain (mesencephalon) and hindbrain (rhombencephalon) of the zebrafish embryo (Figure 5a). A lateral view shows ryr2a mRNA expression in all seven hindbrain rhombomeres, r1-r7 by 48 hpf (Figure 5b). The dorsal view clearly revealed two bilateral patches of ryr2a expression lying dorsal to the eyes which may correspond to the developing tegmental region (Figure 5c). Low level staining was also observed in the anterior portion of the spinal cord at 48 hpf (Figure 5a). Analysis of ryr2a expression by PCR showed that 3 month old zebrafish display a strong ryr2b mRNA expression in the brain compared to the body (Figure 3b). However, the significant maternal expression of ryr2a mRNA detected by PCR (Figure 3a) was not detected by WISH (data not shown). We conclude that zygotic ryr2a expression is expressed exclusively in the developing CNS of zebrafish embryos.
The expression of ryr2b was compared to two well-characterised cardiac markers, nkx2.5 and cmlc2 [18, 20, 37] (Figure 6). Expression of nkx2.5, a homeodomain transcription factor, is the earliest known marker of zebrafish precardiac mesoderm [20, 37]. Cardiac myosin light chain 2 (cmlc2), another marker of myocardial precursor cells, is expressed throughout heart fusion and early heart tube assembly . We show that nkx2.5 but not cmlc2 and ryr2b expression can be observed in the heart progenitor cells at 11 hpf; however, by 14 hpf the bilateral expression of cmlc2 and ryr2b was apparent in the precardiac mesoderm. As reported previously , nkx2.5 expression extends beyond the anterior tip of the notochord and the most posterior nkx2.5 cells (nkx2.5+) do not express cmlc2 (cmlc2-). ryr2b expression appears to be expressed in the region corresponding to cmlc2+ consistent with its appearance in cells that will contribute to the myocardium. By 18 hpf the bilateral cardiac primordial cells expressing cmlc2 and ryr2b bend towards each other, make contact and begin to fuse (Figure 7a). The posterior portions fuse initially followed by anterior portions to create a central lumen and cardiac cone by 20 hpf. At 24 hpf, the heart has formed a linear tube with atrial and ventricular precursor cells and ryr2b expression appears to mirror that of cmlc2+cells, suggestive of its expression throughout the heart at this stage. Using WICC, we observed RyR protein in the atrium and ventricle of the zebrafish heart at 48 hpf (Figure 7b).
Structure of the zebrafish ryr genes and their products: comparison to other species
Previous work identified a total of 14 genomic contigs for ryr from the zebrafish genome assembly . Classification of RyR sequences by radiation hybrid mapping suggested that there are at least five different zebrafish genes: ryr1a, ryr1b, ryr2a, ryr2b and ryr3 . Our study has confirmed the sequences of the five zebrafish ryr genes, although currently the ryr2a gene annotation is still incomplete. Zebrafish, like other teleosts, have undergone a gene duplication event and appear to have retained two distinct copies of ryr1 and ryr2. The teleost ryr1 genes are differentially expressed in the skeletal muscle tissue and the receptors have distinctive Ca2+ binding sensitivities, suggesting that the genes have evolved to perform different physiological functions [13, 38]. Here we report that the zebrafish genome contains two copies of the ryr2 gene, currently the only teleost in which this gene duplication can be observed. Our study has revealed that expression of ryr2a and ryr2b is confined to the embryonic nervous and cardiac tissue respectively. The differential expression of the ryr2 supports the idea that the gene products have different roles within these tissues. The reported sub-functionalisation of the ryr2 in the zebrafish embryo in this study will facilitate the study of this receptor in the development of nervous and cardiac tissue.
Our work has revealed that the primary sequence of the zebrafish RyRs contains many of the conserved domains associated with the regulation and function of this intracellular ion channel in other species, mostly notably humans. The primary sequence of the Ca2+-conducting pore domain was found to be extensively conserved between zebrafish and other species examined. Our study of RyR primary structure supports previous work to show that the biophysical properties and pharmacological regulation of the zebrafish RyR1 is similar to its mammalian homolog . Taken together this data reveals that the zebrafish RyR functions in a similar manner to those found in mammals, this information is of significant relevance to work using the zebrafish as a model for human disease. However, differences in the predicted primary sequence of the zebrafish ryanodine receptor proteins and those of other species were also recorded. There are reports species-specific differences in the Ca2+ regulation and single channel conductance of the RyR1 channel . Further comparative analysis of the RyR family will provide a better insight into the physiological functions of the receptor at a tissue or whole organism level.
The expression of ryr genes during early development, up to and including axis formation
Calcium signalling is required throughout development; however the signal pathways have not been well defined. The prevailing view is that IP3R-driven Ca2+ signals have a major role in axis formation prior to organogenesis whereas RyR-induced Ca2+ signals are necessary for the later aspects of tissue specific differentiation (e.g. muscle formation) [3–5, 7, 9, 40]. Our current study revealed strong maternal expression of ryr2b and ryr3 genes during cleavage, with low levels of ryr3 expression detectable throughout the blastula and gastrula periods. The significance of early ryr mRNA expression remains to be determined; however, it raises the possibility that RyR-generated Ca2+ signals act in development prior to 10 hpf. The characterisation of expression in zebrafish establishes a basis for future experimental work aimed at determining the action of RyR-induced Ca2+ signalling events in early embryonic patterning.
The expression of ryr genes in developing skeletal muscle
Several studies have implicated RyR function in organogenesis, particularly in striated muscle development (as reviewed in ). Our study has used both mutant lines and double staining to establish that RyRs are differentially expressed in slow (ryr1a), fast (ryr3) and both types (ryr1b) of developing skeletal muscle during the segmentation period. In E-C coupling within mammalian skeletal muscle, RyR1 is directly coupled to a voltage gated-Ca2+ channel (VOC) on the sarcolemma. Activation of VOCs via membrane depolarisation then triggers the opening of the ryanodine receptor (RyR1) and release of Ca2+ from the sarcoplasmic reticulum (SR) stores. Zebrafish have non-Ca2+-conducting voltage-gated Ca channels that have evolved solely as voltage sensors to trigger opening of the RyR . The central subunit of the VOC (Cav1.1α1s) acts as the pore, selectivity filter and voltage sensor. In zebrafish two Cav1.1α1s genes (zf-α1s- a and zf-α1s-b) have been identified. The Cav1.1α1s gene products are proposed to interact in a tissue specific manner with the ryr1 genes; that is zf-α1s-a and ryr1a are expressed in slow muscle whilst zf-α1s-b and ryr1b are confined to fast muscle. However, our data suggest that the situation is not quite as clear cut as first proposed because ryr1b is not expressed exclusively in fast muscle but is also located in slow muscle. In addition, Ca2+ release can also be regulated by the ryr3 gene product which is also located in the fast skeletal muscle. RyR3 is proposed to act as an uncoupled calcium-induced calcium release (CICR) channel to propagate the Ca2+ signal . Therefore we propose that in the developing fast muscle zf-α1s-b and ryr1b act together to generate an increase in [Ca2+]i, with the ryr3 gene product acting to amplify the signal. Our data in the embryonic fast muscle showed that ryr1b expression occurs prior to ryr3 and suggests that the RyR-generated [Ca2+]i increase occurs initially via ryr1b with the proposed amplification step via ryr3 developing subsequently. In zebrafish the role of the RyR in E-C coupling within the developing slow muscle appears more complex. Our data reveals that both ryr1a and ryr1b are expressed in the developing slow muscle, but ryr3 is not. This presents the possibility that zf-α1s-a could couple to both ryr1a and ryr1b and raises the issue of whether amplification of the Ca2+ signals occurs in this tissue and, if so, how is this achieved. Clearly there is still much to understand about the maturation of depolarization-induced Ca2+ signaling and its role during skeletal muscle differentiation in vivo.
The expression of ryr genes in the developing nervous and cardiovascular systems
Our data has revealed that there are two ryr2 genes, ryr2a and ryr2b, which are exclusively expressed in either the developing nervous system or cardiac tissue, respectively. Studies in mammals revealed that RyR are expressed in the developing brain and that RyR-mediated Ca2+ signals may have a role in neuronal differentiation and neurite outgrowth [42–44]. All three ryr genes are expressed within the embryonic mouse brain; however, from postnatal day 7 onwards ryr2 becomes in the major isoform . The postnatal changes in RyR expression in mouse brain correlate with a period of neuronal differentiation and may therefore be important in establishing [Ca2+]i homeostasis in maturing neurons. In zebrafish ryr2a expression is localized to specific regions of the developing brains. In these regions ryr2a is likely to regulate neuronal Ca2+ signaling and therefore play a role in CNS development.
In mature cardiac muscle, Ca2+ signals are generated by CICR via the activation of VOCs and the cardiac RyR (RyR2). Knockout mice which do not express ryr2 initially display spontaneous rhythmic contractions of the heart at embryonic day 9 (E9) but no heart beat by day 10 (E10) . Furthermore, RyR-mediated Ca2+ release does not play a significant role in the [Ca2+]i changes observed within the heart of new born rats . Thus in mammals it appears that the RyR2 does not contribute to the onset of contractile activity at very early embryonic stages, but is important for the subsequent maturation and development of the heart in vivo. The zebrafish cardiac ryr gene (ryr2b) is expressed exclusively in the developing heart tissue (precardiac mesoderm) from 14 hpf, 8 hours prior to the onset of cardiac contraction at 22 hpf, and may well contribute to early cardiac development. Investigation of ryr2 function during mammalian development is complicated by the fact that a single ryr2 gene is expressed in several tissues . The sub-functionalisation of the ryr2a and ryr2b genes in the zebrafish embryo provides an excellent system to study individual receptor function in neuronal and cardiac tissues during vertebrate development.
This study has provided a comprehensive overview of the spatial and temporal expression of the ryr gene family in developing zebrafish embryos. This family of Ca2+-release channels are expressed predominantly in developing skeletal, cardiac and neuronal tissue, supportive of the view that RyRs function is relevant to later development events, such as tissue differentiation. In addition, the study has also revealed that maternal ryr mRNA is present in the very early embryo, suggestive of a function for this receptor prior to organogenesis. Ryanodine receptors have been implicated in human disease and the zebrafish is an important vertebrate developmental model which will facilitate work in this area. Future work will explore the function of RyR-regulated Ca2+ signal pathways during zebrafish embryogenesis.
Sutko JL, Airey JA: Ryanodine receptor Ca2+ release channels: does diversity in form equal diversity in function?. Physiol Rev. 1996, 76: 1027-71.
Betzenhauser MJ, Marks AR: Ryanodine receptor channelopathies. Pflugers Arch. 460: 467-80.
Takeshima H, Iino M, Takekura H, Nishi M, Kuno J, Minowa O, Takano H, Noda T: Excitation-contraction uncoupling and muscular degeneration in mice lacking functional skeletal muscle ryanodine-receptor gene. Nature. 1994, 369: 556-9. 10.1038/369556a0.
Takeshima H, Komazaki S, Hirose K, Nishi M, Noda T, Iino M: Embryonic lethality and abnormal cardiac myocytes in mice lacking ryanodine receptor type 2. EMBO J. 1998, 17: 3309-16. 10.1093/emboj/17.12.3309.
Takeshima H, Ikemoto T, Nishi M, Nishiyama N, Shimuta M, Sugitani Y, Kuno J, Saito I, Saito H, Endo M, et al: Generation and characterization of mutant mice lacking ryanodine receptor type 3. J Biol Chem. 1996, 271: 19649-52. 10.1074/jbc.271.33.19649.
Bertocchini F, Ovitt CE, Conti A, Barone V, Scholer HR, Bottinelli R, Reggiani C, Sorrentino V: Requirement for the ryanodine receptor type 3 for efficient contraction in neonatal skeletal muscles. EMBO Journal. 1997, 16: 6956-6963. 10.1093/emboj/16.23.6956.
Rosemblit N, Moschella MC, Ondriasa E, Gutstein DE, Ondrias K, Marks AR: Intracellular calcium release channel expression during embryogenesis. Dev Biol. 1999, 206: 163-77. 10.1006/dbio.1998.9120.
SE Webb, AL Miller: Calcium signalling during embryonic development. Nat Rev Mol Cell Biol. 2003, 4: 539-51. 10.1038/nrm1149.
Slusarski DC, Pelegri F: Calcium signaling in vertebrate embryonic patterning and morphogenesis. Dev Biol. 2007, 307: 1-13. 10.1016/j.ydbio.2007.04.043.
Ashworth R: Approaches to measuring calcium in zebrafish: focus on neuronal development. Cell Calcium. 2004, 35: 393-402. 10.1016/j.ceca.2004.01.002.
Brennan C, Mangoli M, Dyer CE, Ashworth R: Acetylcholine and calcium signalling regulates muscle fibre formation in the zebrafish embryo. J Cell Sci. 2005, 118: 5181-90. 10.1242/jcs.02625.
Chi NC, Shaw RM, Jungblut B, Huisken J, Ferrer T, Arnaout R, Scott I, Beis D, Xiao T, Baier H, et al: Genetic and physiologic dissection of the vertebrate cardiac conduction system. PLoS Biol. 2008, 6: e109-10.1371/journal.pbio.0060109.
Hirata H, Watanabe T, Hatakeyama J, Sprague SM, Saint-Amant L, Nagashima A, Cui WW, Zhou W, Kuwada JY: Zebrafish relatively relaxed mutants have a ryanodine receptor defect, show slow swimming and provide a model of multi-minicore disease. Development. 2007, 134: 2771-81. 10.1242/dev.004531.
Jurynec MJ, Xia R, Mackrill JJ, Gunther D, Crawford T, Flanigan KM, Abramson JJ, MT Howard, DJ Grunwald: Selenoprotein N is required for ryanodine receptor calcium release channel activity in human and zebrafish muscle. Proc Natl Acad Sci USA. 2008, 105: 12485-90. 10.1073/pnas.0806015105.
Zimprich F, Ashworth R, Bolsover SR: Real-time measurements of calcium dynamics in neurons developing in situ within zebrafish embryos. Pflugers Arch-Eur J Physiol. 1998, 436: 489-493. 10.1007/s004240050662.
Kimmel CB, Ballard WW, Kimmel SR, Ullman B, Schilling TF: Stages of Embryonic Development of the Zebrafish. Dev Dyn. 1995, 203: 253-310. 10.1002/aja.1002030302.
Schulte-Merker S, Ho RK, Herrmann BG, C Nusslein-Volhard: The protein product of the zebrafish homologue of the mouse T gene is expressed in nuclei of the germ ring and the notochord of the early embryo. Development. 1992, 116: 1021-32.
Yelon D, Horne SA, Stainier DY: Restricted expression of cardiac myosin genes reveals regulated aspects of heart tube assembly in zebrafish. Dev Biol. 1999, 214: 23-37. 10.1006/dbio.1999.9406.
Weinberg ES, Allende ML, Kelly CS, Abdelhamid A, Murakami T, Andermann P, Doerre OG, Grunwald DJ, Riggleman B: Developmental regulation of zebrafish MyoD in wild-type, no tail and spadetail embryos. Development. 1996, 122: 271-80.
Lee KH, Xu Q, Breitbart RE: A new tinman-related gene, nkx2.7, anticipates the expression of nkx2.5 and nkx2.3 in zebrafish heart and pharyngeal endoderm. Dev Biol. 1996, 180: 722-31. 10.1006/dbio.1996.0341.
Xu Y, He J, Wang X, Lim TM, Gong Z: Asynchronous activation of 10 muscle-specific protein (MSP) genes during zebrafish somitogenesis. Dev Dyn. 2000, 219: 201-15. 10.1002/1097-0177(2000)9999:9999<::AID-DVDY1043>3.3.CO;2-9.
Ashworth R, Zimprich F, Bolsover SR: Buffering intracellular calcium disrupts motoneuron development in intact zebrafish embryos. Brain Res Dev Brain Res. 2001, 129: 169-79.
Westerfield M: The Zebrafish Book: A guide for the laboratory use of zebrafish (Brachydanio rerio). 1995, Univ of Oregon Press, Eugene, 3
Darbandi S, Franck JP: A comparative study of ryanodine receptor (RyR) gene expression levels in a basal ray-finned fish, bichir (Polypterus ornatipinnis) and the derived euteleost zebrafish (Danio rerio). Comp Biochem Physiol B Biochem Mol Biol. 2009, 154: 443-8. 10.1016/j.cbpb.2009.09.003.
Hamilton SL: Ryanodine receptors. Cell Calcium. 2005, 38: 253-60. 10.1016/j.ceca.2005.06.037.
Hamilton SL, Serysheva I: Ryanodine receptor structure: Progress and challenges. J Biol Chem. 2008
Zalk R, Lehnart SE, Marks AR: Modulation of the ryanodine receptor and intracellular calcium. Annu Rev Biochem. 2007, 76: 367-85. 10.1146/annurev.biochem.76.053105.094237.
Sorrentino V, Barone V, Rossi D: Intracellular Ca(2+) release channels in evolution. Curr Opin Genet Dev. 2000, 10: 662-7. 10.1016/S0959-437X(00)00139-8.
Ponting CP: Novel repeats in ryanodine and IP3 receptors and protein O-mannosyltransferases. Trends Biochem Sci. 2000, 25: 48-50.
Ponting C, Schultz J, Bork P: SPRY domains in ryanodine receptors (Ca(2+)-release channels). Trends Biochem Sci. 1997, 22: 193-4. 10.1016/S0968-0004(97)01049-9.
Cui Y, Tae HS, Norris NC, Karunasekara Y, Pouliquin P, Board PG, Dulhunty AF, Casarotto MG: A dihydropyridine receptor alpha(1s) loop region critical for skeletal muscle contraction is intrinsically unstructured and binds to a SPRY domain of the type 1 ryanodine receptor. Int J Biochem Cell Biol. 2008
Fessenden JD, Feng W, Pessah IN, Allen PD: Mutational analysis of putative calcium binding motifs within the skeletal ryanodine receptor isoform, RyR1. J Biol Chem. 2004, 279: 53028-35. 10.1074/jbc.M411136200.
Du GG, MacLennan DH: Topology and Transmembrane Organisation of Ryanodine Receptors. Ryanodine Receptors: structure, function and dysfunction in clinical disease. 2005, Edited by ZHT Wehrens, AR Marks: Springer
Du GG, Sandhu B, Khanna VK, Guo XH, MacLennan DH: Topology of the Ca2+ release channel of skeletal muscle sarcoplasmic reticulum (RyR1). Proc Natl Acad Sci USA. 2002, 99: 16725-30. 10.1073/pnas.012688999.
Schredelseker J, Shrivastav M, Dayal A, Grabner M: Non-Ca2 + −conducting Ca2+ channels in fish skeletal muscle excitation-contraction coupling. Proc Natl Acad Sci USA. 2010, 107: 5658-63. 10.1073/pnas.0912153107.
Heyer V, Lux A, Alunni V, Degrave A, Seiliez I, Kirchner J, Parkhill JP, Thisse C: Spatial and temporal expression of the zebrafish genome by large-scale in situ hybridization screening. Methods Cell Biol. 2004, 77: 505-19.
Chen JN, Fishman MC: Zebrafish tinman homolog demarcates the heart field and initiates myocardial differentiation. Development. 1996, 122: 3809-16.
Franck JP, Morrissette J, Keen JE, Londraville RL, Beamsley M, Block BA: Cloning and characterization of fiber type-specific ryanodine receptor isoforms in skeletal muscles of fish. Am J Physiol. 1998, 275: C401-15.
Koulen P, Janowitz T, Johenning FW, Ehrlich BE: Characterization of the Calcium-release Channel/Ryanodine Receptor from Zebrafish Skeletal Muscleb. Journal of Membrane Biology. 2001, 183: 155-163. 10.1007/s00232-001-0063-8.
Ashworth R, Devogelaere B, Fabes J, Tunwell RE, Koh KR, De Smedt H, Patel S: Molecular and functional characterization of inositol trisphosphate receptors during early zebrafish development. J Biol Chem. 2007, 282: 13984-93. 10.1074/jbc.M700940200.
Sorrentino V, Reggiani C: Expression of the ryanodine receptor type 3 in skeletal muscle. A new partner in excitation-contraction coupling?. Trends Cardiovasc Med. 1999, 9: 54-61. 10.1016/S1050-1738(99)00003-1.
Faure AV, Grunwald D, Moutin MJ, Hilly M, Mauger JP, Marty I, De M, Waard M, Albrieux M: Developmental expression of the calcium release channels during early neurogenesis of the mouse cerebral cortex. Eur J Neurosci. 2001, 14: 1613-22. 10.1046/j.0953-816x.2001.01786.x.
Ooashi N, Futatsugi A, Yoshihara F, Mikoshiba K, Kamiguchi H: Cell adhesion molecules regulate Ca2 + -mediated steering of growth cones via cyclic AMP and ryanodine receptor type 3. J Cell Biol. 2005, 170: 1159-67. 10.1083/jcb.200503157.
Lee SM, Lee JW, Song YS, Hwang DY, Kim YK, Nam SY, Kim DJ, YW Yun, Yoon DY, Hong JT: Ryanodine receptor-mediated interference of neuronal cell differentiation by presenilin 2 mutation. J Neurosci Res. 2005, 82: 542-50. 10.1002/jnr.20655.
Mori F, Fukaya M, Abe H, Wakabayashi K, Watanabe M: Developmental changes in expression of the three ryanodine receptor mRNAs in the mouse brain. Neurosci Lett. 2000, 285: 57-60. 10.1016/S0304-3940(00)01046-6.
Perez CG, Copello JA, Li Y, Karko KL, Gomez L, Ramos-Franco J, Fill M, Escobar AL, Mejia-Alvarez R: Ryanodine receptor function in newborn rat heart. Am J Physiol Heart Circ Physiol. 2005, 288: H2527-40. 10.1152/ajpheart.00188.2004.
The work was in part funded by a MRC grant (G0700216) to Dr Ashworth. Houdini Ho Tin Wu funded by QMUL college studentship. The antibodies 34 C, F59 and MF20 were developed by J. Airey/J. Sutko, F.E. Stockdale and D.A. Fischman respectively and obtained from Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by the University of Iowa, Department of Biology, Iowa City, IA 52242. We thank Debbie Goode and Paul Piccinelli for their advice on the bioinformatics and Heather Callaway in the QMUL zebrafish facility. myhz1 and nkx2.5 plasmid DNA were received as a gift from Dr Yaniv Hintis (Prof. Simon Hughes Lab) and cmlc2 plasmid DNA was received as a gift from Ms Ana Filipa C. Simões (Prof. Roger Patient's Lab).
The authors declare that they have no competing interests.
H.H.T.W performed all the experimental work and contributed to the writing of the manuscript. C.B. supervised and provided input on the design of the experiments. R.A. designed and supervised the experiments and wrote the manuscript. All authors read and approved the final manuscript.