ABSTRACT

The development of a vascular network is essential to nourish tissues and sustain organ function throughout life. Endothelial cells (ECs) are the building blocks of blood vessels, yet our understanding of EC specification remains incomplete. Zebrafish cloche/npas4l mutants have been used broadly as an avascular model, but little is known about the molecular mechanisms of action of the Npas4l transcription factor. Here, to identify its direct and indirect target genes, we have combined complementary genome-wide approaches, including transcriptome analyses and chromatin immunoprecipitation. The cross-analysis of these datasets indicates that Npas4l functions as a master regulator by directly inducing a group of transcription factor genes that are crucial for hematoendothelial specification, such as etv2, tal1 and lmo2. We also identified new targets of Npas4l and investigated the function of a subset of them using the CRISPR/Cas9 technology. Phenotypic characterization of tspan18b mutants reveals a novel player in developmental angiogenesis, confirming the reliability of the datasets generated. Collectively, these data represent a useful resource for future studies aimed to better understand EC fate determination and vascular development.

INTRODUCTION

The cardiovascular system is one of the first vertebrate organs to form and function, and it is crucial for the delivery of oxygen and nutrients, and the removal of waste products from developing and mature organs (Marcelo et al., 2013). Endothelial cells (ECs) constitute the building blocks of the vasculature, lining the luminal side of all vessels. EC specification is a crucial event for the development of the circulatory system, and the lack of a vascular system leads to early embryonic lethality in most animals (Ferrara et al., 1996). In zebrafish, EC precursors were identified in the ventral region of the embryo as early as shield stage (Vogeli et al., 2006). During gastrulation, a subset of ventral mesodermal cells differentiates into angioblasts: the progenitor of ECs. During somitogenesis, the angioblasts migrate from the lateral plate mesoderm (LPM) towards the midline of the embryo, where they coalesce to form the axial vessels: the dorsal aorta and the cardinal vein (Jin et al., 2007; Herbert et al., 2009). Sequential transcriptional waves orchestrate EC differentiation starting from 6 h post fertilization (hpf) in zebrafish, and a number of transcription factor genes have been described to be involved in this process, including cloche/npas4l (Stainier et al., 1995; Reischauer et al., 2016), fli1 (Craig et al., 2015), etv2 (Pham et al., 2007), lmo2 (Weiss et al., 2012), sox7 (Hermkens et al., 2015) and tal1 (Bussmann et al., 2007). Importantly, cloche mutants exhibit the most severe impairment in EC specification, lacking most endothelial and hematopoietic tissues (Stainier et al., 1995), and they fail to activate the EC specification program (Liao et al., 1997; Ho et al., 1999).

The role of cloche in EC specification was determined to be cell-autonomous (Parker and Stainier, 1999). For more than two decades, the cloche mutant has been employed as an avascular model to study the impact of the vasculature on the development of other organs, including the retina (Dhakal et al., 2015), the pancreas (Field et al., 2003a) and the liver (Field et al., 2003b). As the cloche locus is situated in a telomeric region, the isolation of the gene was challenging, and only recently was cloche found to encode a bHLH-PAS transcription factor (Reischauer et al., 2016). The Cloche protein was thereafter called Npas4l due to its closest homology, in terms of amino acid sequence, to mammalian Npas4 (Reischauer et al., 2016). Interestingly, npas4l is present in fish, reptiles and birds, but is missing in mammals, raising questions about its mammalian functional orthologue and the evolution of endothelial cell differentiation (Reischauer et al., 2016).

In this study, we identified genes downstream of Npas4l. We generated novel loss- and gain-of-function models for npas4l, and combined transcriptomic data with chromatin immunoprecipitation (ChIP) of Npas4l to elucidate its molecular mechanism of action. Our data indicate that Npas4l drives EC differentiation by binding to the promoters and inducing the expression of crucial transcription factor genes, including etv2, tal1 and lmo2. Furthermore, the analyses performed led to the identification of tspan18b, a potential new player in vascular development, which we show is a target of Npas4l. Using CRISPR/Cas9 technology, we generated a tspan18b mutant that we show exhibits early angiogenic defects, confirming the reliability of the datasets generated. Collectively, these data represent a valuable resource of Npas4l target genes that may play a role in vascular and hematopoietic development.

RESULTS AND DISCUSSION

Npas4l promotes endothelial cell specification by directly inducing angioblast genes

In order to gain insight into the gene regulatory network controlled by Npas4l in vivo, we first conducted an overexpression experiment by injecting npas4l mRNA into one-cell stage zebrafish embryos. Following npas4l overexpression, we performed microarray analysis to identify transcripts upregulated at 30 and 95% epiboly (Fig. 1A, Figs S1A and S2, Table S1). Gene set enrichment analysis (GSEA) showed that npas4l was sufficient to drive the expression of a number of genes involved in blood vessel morphogenesis (Fig. 1B), consistent with previous data (Reischauer et al., 2016). As early as 4 hpf (30% epiboly), we detected the induction of angioblast and endothelial markers, including etv2, tal1 and lmo2 (Fig. S2 and Table 1). Notably, the expression levels of these angioblast markers were previously shown to be decreased in npas4l mutants, as assessed by in situ hybridization (Thompson et al., 1998; Sumanas et al., 2005; Reischauer et al., 2016). These and other angioblast-specific genes were upregulated at both stages (Fig. 1C). Additionally, we identified zgc:122979, a gene encoding a member of the heat shock protein 40 family, which was induced at 30 and 95% epiboly (Fig. S1B). In situ hybridization analysis revealed that zgc:122979 is expressed in the LPM at the 5-somite stage (11.5 hpf) (Fig. S1C). Notably, genes expressed in mature endothelium, such as kdrl, cdh5, she and clec14a, which have previously been reported to function downstream of Etv2 (Gomez et al., 2012), were substantially induced at 95% but not at 30% epiboly (Fig. 1C).

Fig. 1.

Npas4l promotes endothelial cell specification by directly regulating angioblast genes. (A) Schematic representation of the overexpression experiment followed by microarray analysis. (B) Gene set enrichment analysis (GSEA) performed on genes at 95% epiboly. (C) Venn diagram depicting the genes upregulated more than threefold at 30% and 95% epiboly following npas4l overexpression. (D) Genomic distribution of regions bound by Npas4l. (E) Integrative genomics viewer (IGV) visualization of Npas4l occupancy at selected loci: tal1, etv2 and lmo2. (F) De novo binding motif identification of Npas4l and comparison with previously reported binding motifs of other bHLH-PAS transcription factors.

Fig. 1.

Npas4l promotes endothelial cell specification by directly regulating angioblast genes. (A) Schematic representation of the overexpression experiment followed by microarray analysis. (B) Gene set enrichment analysis (GSEA) performed on genes at 95% epiboly. (C) Venn diagram depicting the genes upregulated more than threefold at 30% and 95% epiboly following npas4l overexpression. (D) Genomic distribution of regions bound by Npas4l. (E) Integrative genomics viewer (IGV) visualization of Npas4l occupancy at selected loci: tal1, etv2 and lmo2. (F) De novo binding motif identification of Npas4l and comparison with previously reported binding motifs of other bHLH-PAS transcription factors.

Table 1.

Overview of selected genes regulated by Npas4l

Overview of selected genes regulated by Npas4l
Overview of selected genes regulated by Npas4l

The transcriptional analysis performed at two different time points provides insight into the temporal dynamics of the transcriptional cascade initiated by Npas4l. We hypothesized that genes upregulated at 30% epiboly may reflect direct targets, while those upregulated at 95% epiboly represent indirect, or secondary, targets. In order to identify the direct downstream targets of Npas4l, we performed a genome-wide chromatin immunoprecipitation sequencing analysis (ChIP-Seq). Owing to the lack of commercially available antibodies for Npas4l, we used an HA-tagged version of the protein that was able to induce expression of etv2 at similar levels to wild-type Npas4l (Fig. S1D). We found that Npas4l-HA binds a total of 42 genomic regions, 19 within 5 kb from a transcription start site (TSS) (Fig. S1E), six in intragenic regions and the remaining 17 in intergenic regions (Fig. 1D and Table S1). Next, we performed a cross-analysis of genes whose expression was increased at least twofold by npas4l overexpression and genes whose TSS is present within 100 kb of an Npas4l-binding site. This led to the discovery of 15 genes that are likely to be direct targets of Npas4l, including etv2, tal1, lmo2, egfl7, gata2a and tspan18b (Fig. 1E and Fig. S1F). We also extrapolated the binding motif of Npas4l using the MEME Suite online tool (http://meme-suite.org), and identified the consensus sequence TCGTGA (Fig. 1F). This binding motif is consistent with the non-palindromic binding sites reported for other bHLH-PAS transcription factors, including NPAS4 and HIF1α, characterized by the core sequence CGTG (Schodel et al., 2011; Lo and Matthews, 2012). In particular, the TCGTG site was found to be the target of Dysfusion, the Drosophila orthologue of Npas4l (Jiang and Crews, 2007), and the same sequence was described to be the target of the neuronal-specific activity-dependent NPAS4 in mouse (Sorensen et al., 2016). Intriguingly, Tsang et al. (2017) reported that HIF1α binds to the ETV2 promoter in human embryonic stem cells, and that hypoxia promotes ETV2 expression via HIF1α. Considering the similarity of the HIF1α motif to that of Npas4l (Fig. 1F), this bHLH-PAS transcription factor may, at least in part, have taken over npas4l function in mammals.

Transcriptomic and epigenomic analyses of npas4l mutants

To better understand the molecular mechanisms of Npas4l function, we analysed the transcriptome and epigenome of npas4l mutants. Using CRISPR/Cas9 technology, we generated a new npas4l mutant allele with a lesion in its second exon, which encodes the bHLH domain (Fig. S3). TgBAC(etv2:EGFP); npas4l−/− exhibit a strong phenotype, comparable with the previously described m39 (Stainier et al., 1995) and s5 (Liao et al., 2000) alleles, with the advantage that this new allele (bns297) can be easily genotyped using high-resolution melt analysis (Samarut et al., 2016). Previous studies had analysed the transcriptome of npas4l/cloche mutants at the 15- and 18-somite stage (Qian et al., 2005; Sumanas et al., 2005). However, in order to focus on early transcriptional changes, we performed RNA-sequencing (RNA-seq) analysis of mutant embryos and their wild-type siblings at the six-somite stage (Fig. 2A and Table S1). tal1, etv2 and lmo2 expression was strongly downregulated in npas4l mutant embryos compared with wild-type siblings (Fig. 2B,C, Table 1). The expression levels of hematopoietic genes, including gfi1aa and gata1a, were also decreased in the mutants, although these genes were not induced by npas4l overexpression (Fig. 2B). Gene set enrichment analysis identified a number of genes involved in vascular and hematopoietic development downregulated in npas4l mutants at the six-somite stage, confirming its role in the early specification of endothelium and blood (Fig. S4A). Surprisingly, we found that genes involved in neural function were upregulated in npas4l mutants, suggesting that angioblasts, or npas4l itself, may have a role in repressing neuronal gene expression at the six-somite stage (Fig. S4B). Murine NPAS4 is known to regulate synaptic function (Spiegel et al., 2014), and npas4a expression has previously been described to be restricted to the zebrafish brain, similar to mammalian Npas4 (Klaric et al., 2014). We have previously reported that npas4l is expressed specifically in the LPM during early somitogenesis using in situ hybridization (Reischauer et al., 2016). Nevertheless, neural progenitors might express levels of npas4l too low to detect by in situ hybridization. Interestingly, a subset of genes regulated by npas4l, including gata2a and tal1, were found to be expressed during zebrafish development not only in EC progenitors, but also in subsets of neurons (Andrzejczuk et al., 2018). Additionally, mesodermal cells may acquire a neural fate in the absence of npas4l function, similarly to what has been reported in the Tbx6 knockout mouse (Chapman and Papaioannou, 1998).

Fig. 2.

Transcriptomic and epigenomic analyses of npas4l mutants. (A) Schematic representation of the RNA-seq experiment. Green lines depict angioblasts, marked by Tg(etv2:EGFP) expression. Tg(etv2:EGFP) expression was used to identify mutant embryos. (B) Venn diagram showing the number of genes downregulated more than twofold in npas4l mutants and the top 50 genes upregulated at 95% epiboly following npas4l overexpression. (C) IGV visualization of selected targets of Npas4l, exhibiting a significant reduction in expression levels in npas4l mutants. (D) Schematic representation of the ATAC-seq experiment. Green line depicts angioblasts, marked by Tg(etv2:EGFP) expression. Tg(etv2:EGFP) expression was used to identify mutant embryos. (E) IGV visualization of accessible chromatin at the lmo2 locus in one-somite stage wild-type and npas4l mutant embryos, combined with ChIP-seq data. The regions that appear to be less accessible in mutants are outlined in red. (F) IGV visualization of accessible chromatin at the etv2 locus in one-somite stage wild-type and npas4l mutant embryos, combined with ChIP-seq data.

Fig. 2.

Transcriptomic and epigenomic analyses of npas4l mutants. (A) Schematic representation of the RNA-seq experiment. Green lines depict angioblasts, marked by Tg(etv2:EGFP) expression. Tg(etv2:EGFP) expression was used to identify mutant embryos. (B) Venn diagram showing the number of genes downregulated more than twofold in npas4l mutants and the top 50 genes upregulated at 95% epiboly following npas4l overexpression. (C) IGV visualization of selected targets of Npas4l, exhibiting a significant reduction in expression levels in npas4l mutants. (D) Schematic representation of the ATAC-seq experiment. Green line depicts angioblasts, marked by Tg(etv2:EGFP) expression. Tg(etv2:EGFP) expression was used to identify mutant embryos. (E) IGV visualization of accessible chromatin at the lmo2 locus in one-somite stage wild-type and npas4l mutant embryos, combined with ChIP-seq data. The regions that appear to be less accessible in mutants are outlined in red. (F) IGV visualization of accessible chromatin at the etv2 locus in one-somite stage wild-type and npas4l mutant embryos, combined with ChIP-seq data.

Given the striking phenotype of npas4l mutants and the ability of Npas4l to potently activate its target genes as early as 4 hpf (Fig. 1C), we wanted to test whether Npas4l acts as a pioneer transcription factor. Pioneer transcription factors bind to condensed chromatin and have the ability to increase accessibility, enabling other factors to access their target sites (Iwafuchi-Doi and Zaret, 2014). To investigate whether Npas4l acts as a pioneer factor, we assessed chromatin accessibility in npas4l mutants, using the assay for transposase-accessible chromatin coupled to next-generation sequencing (ATAC-seq) (Buenrostro et al., 2015; Doganli et al., 2017). We incrossed npas4l+/− fish and genotyped single embryos at the one-somite stage. Subsequently, we performed ATAC-Seq on mutants and wild-type siblings (Fig. 2D and Table S1). We compared these ATAC-seq data with the Npas4l ChIP-Seq results, and in a small subset of Npas4l targets, chromatin accessibility was reduced, as in the case of lmo2, although the affected regions do not appear to interact directly with Npas4l (Fig. 2E). For the most part, however, chromatin accessibility at Npas4l-binding sites was not affected in npas4l mutants compared with wild-type (Fig. 2F). The present data thus indicate that Npas4l binds to open chromatin and promotes the expression of its target genes. Importantly, Npas4l might regulate chromatin state specifically in angioblasts, and this effect might be diluted in the whole-embryo approach used here.

npas4l overexpression at tailbud stage promotes the expression of etv2 but not cdh5

To test whether specific genes are induced by npas4l at later stages, we generated and used a new HOTcre line (Hesselson et al., 2009) (Fig. S5A). Considering the expression pattern of etv2, tal1 and lmo2, we induced npas4l expression at tailbud (TB) stage, which is concomitant with the appearance of the first angioblast markers (Fig. S5B). We performed qPCR at different time points to determine when direct targets were induced, but secondary targets were not affected. At 1 h post heat shock (hph), etv2 upregulation was detectable while expression of cdh5, a known target of Etv2, was not significantly upregulated (Fig. S5C). We performed RNA-seq analysis at 1 hph (Table S1), and could detect etv2 expression increasing by more than twofold (Table 1). Notably, tal1 and lmo2 expression levels were also upregulated after heat shock, but due to their high level of expression in wild type, the induction of expression of these genes was below twofold (Table 1).

tspan18b as a novel vascular development gene

Following the generation and analysis of the individual datasets, we combined the obtained data in order to identify new Npas4l target genes. The resulting Venn diagram includes the overexpression experiment, the transcriptome of npas4l mutants at the six-somite stage and the ChIP-Seq data, considering all genes whose TSS is found within a 100 kb window from at least one Npas4l-binding site. As the Venn diagram highlights (Fig. 3A), using a threshold of twofold change, etv2 is the only gene consistently present in all datasets. Notably, lmo2, tal1 and egfl7, despite Npas4l binding in their promoters, are not induced by npas4l overexpression at TB stage, owing to their physiologically high expression levels at the stage of sample collection. According to our ChIP-Seq data, Npas4l does not bind to the sox7 promoter, although sox7 is present in all other datasets and has previously been described to act downstream of etv2 (Wong et al., 2009). Our analysis aimed to identify novel genes and, intriguingly, a previously undescribed gene, si:dkey-121a11.3, was detected in most of the datasets, although Npas4l does not appear to bind its proximal promoter (Table 1). We therefore focused our attention on tspan18b, a gene induced by Npas4l and with an Npas4l-binding site in its promoter (Fig. 3B). tspan18 belongs to the tetraspanin gene family, and it has been reported to be expressed in the developing vasculature in chick (Fairchild and Gammill, 2013) and mouse (Scialdone et al., 2016). Here, we performed in situ hybridization to examine tspan18b expression. Although tspan18b expression was not limited to the blood vessels, it was reduced in the axial vasculature of npas4l mutants at 16 and 24 hpf (Fig. 3C). To study its function in zebrafish, we generated a mutant line using the CRISPR/Cas9 technology by targeting exon 4, and recovered an allele with a complex indel predicted to lead to a premature stop codon after a 74 amino acid long missense segment (Fig. S6). When crossed to the Tg(fli1a:EGFP) background (Lawson and Weinstein, 2002), we observed defects in intersegmental vessel formation in tspan18b mutants at 36 hpf (Fig. 3D). The early vasculature of tspan18b mutants at 14 and 24 hpf was not distinguishable from that of wild-type siblings, suggesting that vasculogenesis was not affected (Fig. S7A and B). Notably, the phenotype at 36 hpf was observed to be partially penetrant (Fig. 3E). It seems unlikely that this partial penetrance is due to transcriptional adaptation (Rossi et al., 2015; El-Brolosy et al., 2019), as tspan18b morphants do not exhibit a more severe intersegmental vessel phenotype than that observed in tspan18b mutants (Fig. S8). However, other compensatory mechanisms may be at play, and additional studies will be required to test this hypothesis.

Fig. 3.

tspan18b as a novel vascular development gene. (A) Venn diagram illustrating the cross-analysis of the different datasets used to identify genes downstream of Npas4l. (B) IGV visualization of Npas4l occupancy at the tspan18b locus. (C) tspan18b in situ hybridization in 16 and 24 hpf npas4l+/+ and npas4l−/− embryos. (D) Maximal intensity projections of confocal z-stacks of 36 hpf Tg(fli1a:EGFP) wild-type sibling and tspan18b mutant embryos. (E) Quantification of the intersegmental vessels connected to the dorsal longitudinal anastomotic vessel in tspan18b mutants and wild-type siblings. Each dot represents the number of complete vessels in one embryo. n=24 embryos from three independent clutches. Data are mean±s.d. **P<0.01, t-test. Scale bars: 100 μm in C; 50 μm in D.

Fig. 3.

tspan18b as a novel vascular development gene. (A) Venn diagram illustrating the cross-analysis of the different datasets used to identify genes downstream of Npas4l. (B) IGV visualization of Npas4l occupancy at the tspan18b locus. (C) tspan18b in situ hybridization in 16 and 24 hpf npas4l+/+ and npas4l−/− embryos. (D) Maximal intensity projections of confocal z-stacks of 36 hpf Tg(fli1a:EGFP) wild-type sibling and tspan18b mutant embryos. (E) Quantification of the intersegmental vessels connected to the dorsal longitudinal anastomotic vessel in tspan18b mutants and wild-type siblings. Each dot represents the number of complete vessels in one embryo. n=24 embryos from three independent clutches. Data are mean±s.d. **P<0.01, t-test. Scale bars: 100 μm in C; 50 μm in D.

In conclusion, our results present a collection of genome-wide data that allow for the identification of genes acting downstream of Npas4l in vivo. We foresee that these data and their ensuing analyses will be a valuable resource for the scientific community, especially those who aim to investigate not only EC specification, but also hematopoietic and vascular development. In addition, the present study points to molecular mechanisms underlying Npas4l function by revealing its binding motif and at least some of its likely direct targets. Altogether, these analyses also led to the identification and characterization of tspan18b, a novel regulator of vascular development in zebrafish, and future studies will be required to further investigate its role.

MATERIALS AND METHODS

Zebrafish handling

Zebrafish husbandry was performed under standard conditions in accordance with institutional (MPG) and national ethical and animal welfare guidelines. Zebrafish were maintained and embryos were obtained and raised under standard conditions (https://zfin.org/zf_info/zfbook/zfbk.html). Tg(fli1a:EGFP)y1 (Lawson and Weinstein, 2002) and TgBAC(etv2:EGFP)ci1 (Wong et al., 2009) fish were used in this study.

Expression analysis

RT-qPCR was performed in a CFX Connect Real-Time System (Bio-Rad). RNA was isolated using TRIzol (Life Technologies) and cDNA was synthesized using the Maxima First Strand cDNA Synthesis Kit for RT-qPCR, with dsDNase (Thermo Fisher Scientific), according to manufacturer's instructions. Plots show the results of three different experiments (three pools of embryos per developmental stage or treatment) and each pool consisted of 5-10 embryos. Primers were designed using primer3 (primer3.ut.ee) and gapdh was used for normalization. Primers used were: npas4l forward, 5′-GTACGTTCTCAGACACATACAG-3′; npas4l reverse, 5′-CTACAGGATTGTGCTCTACAG-3′; gapdh forward, 5′-TACCATGTGACCAGCTTGAC-3′; gapdh reverse, 5′-AGAAACCTGCCAAGTATGATGAG-3′; etv2 forward, 5′-CGAGGTTCTGGTAGGTTTGAG-3′; etv2 reverse, 5′-GCACAAAGGTCATGTTCTCAC-3′; cdh5 forward, 5′-TTCGGAGGAATATGTGCTGG-3′; cdh5 reverse, 5′-GATACAGAGAAGGATGGCGA-3′.

Microarray

RNA was isolated using TRIzol and cDNA was synthesized using the Maxima First Strand cDNA Synthesis Kit for RT-qPCR, with dsDNase. After purification with RNA Clean and Concentrator kit (Zymo Research), the microarray analysis was performed by Oaklabs (Hennigsdorf, Germany). A 8×60K zebrafish expression array (XS-5090; Agilent 60-mer SurePrint Technology) analysis was performed according to manufacturer's protocol.

Mutant generation by CRISPR/Cas9

pT7-gRNA and pT3TS-nlsCas9nls vectors were purchased from Addgene (www.addgene.org/46759/; www.addgene.org/46757/). gRNAs were designed using the CHOPCHOP online tool (chopchop.cbu.uib.no). The gRNA used for npas4lbns297 mutant generation was 5′-CCGCGCCTTAGATGCTCCTT-3′ (Fig. S3); the gRNA used for tspan18bns299 mutant generation was 5′-ACGCCCGTGAAGAGAAGA-3′ (Fig. S6). Oligonucleotides were annealed in a thermo block at 95°C for 5 min followed by a slow cooling at room temperature and cloned into the gRNA plasmid between BsmBI sites. All constructs were verified by sequencing. To make gRNA, the template DNA was linearized using BamHI digestion and purified using a QIAprep column (Qiagen). The gRNA was generated by in vitro transcription using a T7 RNA polymerase MEGA short script T7 kit (Life Technologies). After in vitro transcription, the gRNA (∼100 nucleotides long) was purified using an RNA Clean and Concentrator kit. To make nlsCas9nls RNA (Addgene), the template DNA was linearized by XbaI digestion and purified using a QIAprep column. The capped nlsCas9nls mRNA was synthesized using the mMESSAGE mMACHINE T3 kit (Life Technologies) and purified using an RNA Clean and Concentrator kit. nlsCas9nls mRNA (50 pg) and gRNA (50 pg) were co-injected into the cell at the one-cell stage. For further details on the high-resolution melt analysis, see supplementary Materials and Methods.

Heat shock treatments

Fish embryos raised at 28°C were subject to a 37°C heat shock for 1 h by replacing the egg water with pre-warmed (37°C) egg water starting at different time points, and then keeping them in a 37°C incubator.

Microinjection of morpholinos

One-cell stage embryos were injected with 2 or 4 ng of an ATG morpholino against tspan18b (sequence: 5′-CGAGCCCCATGAGTTCTTTGATGTA-3′) or with a standard control morpholino (sequence: 5′-CCTCTTACCTCAGTTACAATTTATA-3′) (Gene Tools).

Whole-mount in situ hybridization

The probes were generated using the following primers: zgc:122979-ISH forward, 5′-CTGGACTCAGTTCGGCGTCAA-3′; zgc:122979-ISH reverse, 5′-TAATACGACTCACTATAGCTGAAGTGGGCGTGTTTCTT-3′; tspan18b-ISH forward, 5′-AGATTGTGGCAGCAAACCCT-3′; tspan18b-ISH reverse; 5′-TAATACGACTCACTATAGGGAGATTCCCCTGAACAGACACATGG-3′. T7 polymerase (Promega) was used for transcription and digoxigenin labelling of probes (Roche). For whole-mount in situ hybridization, embryos were fixed in 4% paraformaldehyde overnight at 4°C, subsequently dehydrated in methanol and stored at −20°C. On the first day, embryos were rehydrated to PBS/0.1% Tween-20 and then digested in 10 μg/ml proteinase K (Roche) for 45 s followed by fixation in 4% PFA. Embryos were pre-incubated with hybridization buffer (Thisse and Thisse, 2008) at 70°C for 3 h and then incubated with DIG-labelled RNA antisense probes at 70°C overnight. The next day, after washing as described (Thisse and Thisse, 2008), the embryos were incubated with alkaline phosphatase-conjugated anti-digoxigenin antibody (Roche) at 4°C overnight. On the last day, after washing, the signal was visualized with NBT-BCIP staining solution (Roche) (Thisse and Thisse, 2008).

Imaging

Zebrafish embryos were immobilized with 0.015% tricaine and mounted in 1% low melting agarose (GeneMate). Bright-field images were acquired using a Nikon SMZ25, and confocal images were acquired using a Zeiss LSM800 Observer with a 20×/0.8 M27 Plan-Apochromat objective lens or a Zeiss LSM700 with a 10×/0.3 Plan-Neofluar objective lens.

Overexpression experiments

An HA-tagged version of npas4l was cloned into pCS2+. Between 25 and 50 pg of npas4l mRNA was injected into AB embryos at the one-cell stage. For Cre recombinase expression, the cre-coding sequence (Hesselson et al., 2009) was cloned into pCS2+, transcribed into mRNA using the SP6 mMessage mMachine Kit and purified using an RNA Clean and Concentrator kit (Zymo Research). The Tg(hsp70l:loxP-STOP-loxP-npas4l-GGGGGLSRS-P2A-mCherry)bns298 line (Fig. S5) was generated as described previously (Hesselson et al., 2009). Between 20 and 40 pg of cre mRNA was injected into transgenic embryos at the one-cell stage. Heat shock was performed at the tailbud stage as aforementioned, and uninjected embryos were used as controls. Total RNA was isolated from 10 cre-injected embryos and 10 uninjected sibling embryos using the miRNeasy Micro Kit (Qiagen). For exclusion of genomic DNA contamination, the samples were treated with an on-column DNase digestion (DNase-Free DNase Set, Qiagen).

Quantification of intersegmental vessels

To quantify intersegmental vessel (ISV) formation, we counted the number of ISVs that reach the dorsal region of the neural tube in an eight-somite long trunk area above the yolk extension in Tg(fli1a:EGFP) wild-type and tspan18−/− embryos.

Statistical analysis

Statistical analysis was performed using the GraphPad software. Data presented in bar graphs represent mean±s.e.m. or ±s.d. P values were calculated using Student's t-test for single comparisons of normally distributed data (**P<0.01; ns, no significant changes observed).

ATAC-sequencing

We collected single embryos (one- to two-somite stage; npas4l−/− and wild-type siblings) and dechorionated them using pronase. We homogenized the samples using a mortar and pestle. We centrifuged the samples at 500 g for 10 min at 4°C. After removing the supernatant, we added 51 μl of ice-cold PBS (one microliter was used for genotyping) and we centrifuged the samples again at 500 g for 10 min at 4°C. The supernatant was removed and 50 μl of cold lysis buffer, prepared as described previously (Doganli et al., 2017), was added. Samples were centrifuged at 500 g at 4°C for 10 min. The pellet was resuspended in 25 µl TD buffer, 2.5 µl Tn5 (Illumina Nextera DNA Sample Preparation Kit), 0.5 µl 10% NP-40 and 22 µl water. The cell/Tn5 mixture was incubated at 37°C for 30 min with occasional snap mixing. Transposase treatment was followed by a 30-min incubation at 50°C together with 500 mM EDTA (pH 8.0) for optimal recovery of digested DNA fragments. For neutralization of EDTA, 100 µl of 50 mM MgCl2 was added, followed by purification of the DNA fragments by MinElute PCR Purification Kit (Qiagen). Amplification of library together with indexing was performed as described elsewhere (Buenrostro et al., 2015). Libraries were mixed in equimolar ratios and sequenced on a NextSeq500 platform using V2 chemistry with 2×38 bp paired end mode. See supplementary Materials and Methods for further details.

Chromatin immunoprecipitation and sequencing

We deyolked 3000 embryos as described previously (Link et al., 2006). Cells were lysed as described previously (Bogdanovic et al., 2013). Chromatin was sonicated using a probe sonicator (SONOPULS, Bendelin) with a MS72 micro tip. The sonication was performed with 20% amplitude with 12×3 cycles of 10 s on, 50 s off with 12 min on ice between each cycle. After sonication, 120 μl of 10% Triton X-100 was added. The size of sonicated chromatin was assessed by gel electrophoresis with fragment sizes between 100 and 300 bp. ChIP was conducted as previously described (Bogdanovic et al., 2013), using Protein G Dynabeads (Invitrogen) and anti-HA antibody (Abcam, ab9110). For the elution, we resuspended the beads in 100 μl of elution buffer [50 mM Tris (pH 8.0), 10 mM EDTA, and 1% SDS] at 65°C for 15 min and reversal of crosslinks was performed overnight at 65°C with shaking at 1000 rpm. A total of 200 μl of TE buffer was added, and samples were incubated with RNaseA (final concentration 0.2 µg/ml) at 37°C for 2 h. Then, proteinase K was added (final concentration 0.2 µg/ml) and samples were incubated at 55°C for 2 h. An equal amount (400 μl) of phenol:chloroform:isoamyl alcohol (25:24:1) was added and the aqueous phase extracted after centrifugation at 15,500 g for 5 min. DNA was purified from the aqueous phase using the Qiagen MinElute PCR purification column according to manufacturer's protocol. ChIP samples were quantified by HS DNA Qubit measurement (Thermo), and 10 ng DNA from Input and ChIP samples were used as input for TruSeq ChIP Sample Prep following manufacturer's protocol until ligation of adapters and cleanup. Ligated products were not separated by gel but were used directly in PCR amplification followed by double-sided bead cleanup (0.85×+0.75×). Insert size and library integrity were tested on a LabChip Gx Touch 24 (Perkin Elmer) and quantified by a Qubit HS DNA assay. Libraries were sequenced on a Nextseq500 with V2 chemistry resulting in a minimum of 25M reads. See supplementary Materials and Methods for further details.

RNA-sequencing of npas4l mutants

For RNA-seq, total RNA was isolated from 10 npas4l−/− embryos and 10 wild-type sibling embryos using the miRNeasy Micro Kit. npas4l−/− and wild-type siblings were sorted based on their phenotype, by analysing TgBAC(etv2:EGFP) expression in the LPM. For exclusion of genomic DNA contamination, the samples were treated with an on-column DNase digestion (DNase-Free DNase Set). See supplementary Materials and Methods for further details.

Acknowledgements

We thank Ryota Matsuoka, Kenny Mattonet, Oliver Stone and Jason Lai for helpful discussions, as well as for sharing reagents and protocols.

Footnotes

Author contributions

Conceptualization: M.M., S.R., D.Y.R.S.; Methodology: M.M., A.B., C.G., F.L., N.F., S.G., C.K.; Software: M.M., A.B., S.G., C.K.; Validation: M.M., A.B., F.L., N.F.; Formal analysis: M.M., A.B., C.G., N.F., S.G., C.K.; Investigation: M.M., A.B., C.G., N.F., S.R.; Data curation: M.M., A.B.; Writing - original draft: M.M.; Writing - review & editing: M.M., A.B., C.G., S.R., D.Y.R.S.; Visualization: M.M., A.B., C.G., F.L., D.Y.R.S.; Supervision: A.B., S.R., D.Y.R.S.; Project administration: D.Y.R.S.; Funding acquisition: D.Y.R.S.

Funding

These studies were supported in part by funds from the Max Planck Society and the Deutsche Forschungsgemeinschaft, SFB1213, project B01.

Data availability

All genome-wide data, including RNA-Seq, ChIP-Seq and ATAC-Seq have been deposited in GEO under accession number GSE130202.

References

Andrzejczuk
,
L. A.
,
Banerjee
,
S.
,
England
,
S. J.
,
Voufo
,
C.
,
Kamara
,
K.
and
Lewis
,
K. E.
(
2018
).
Tal1, Gata2a, and Gata3 have distinct functions in the development of V2b and cerebrospinal fluid-contacting KA spinal neurons
.
Front. Neurosci.
12
,
170
.
Buenrostro
,
J. D.
,
Wu
,
B.
,
Chang
,
H. Y.
and
Greenleaf
,
W. J.
(
2015
).
ATAC-seq: a method for assaying chromatin accessibility genome-wide
.
Curr. Protoc. Mol. Biol.
109
,
21.29.1
-
21.29.9
.
Bussmann
,
J.
,
Bakkers
,
J.
and
Schulte-Merker
,
S.
(
2007
).
Early endocardial morphogenesis requires Scl/Tal1
.
PLoS Genet.
3
,
e140
.
Chapman
,
D. L.
and
Papaioannou
,
V. E.
(
1998
).
Three neural tubes in mouse embryos with mutations in the T-box gene Tbx6
.
Nature
391
,
695
-
697
.
Craig
,
M. P.
,
Grajevskaja
,
V.
,
Liao
,
H.-K.
,
Balciuniene
,
J.
,
Ekker
,
S. C.
,
Park
,
J.-S.
,
Essner
,
J. J.
,
Balciunas
,
D.
and
Sumanas
,
S.
(
2015
).
Etv2 and fli1b function together as key regulators of vasculogenesis and angiogenesis
.
Arterioscler. Thromb. Vasc. Biol.
35
,
865
-
876
.
Dhakal
,
S.
,
Stevens
,
C. B.
,
Sebbagh
,
M.
,
Weiss
,
O.
,
Frey
,
R. A.
,
Adamson
,
S.
,
Shelden
,
E. A.
,
Inbal
,
A.
and
Stenkamp
,
D. L.
(
2015
).
Abnormal retinal development in Cloche mutant zebrafish
.
Dev. Dyn.
244
,
1439
-
1455
.
Doganli
,
C.
,
Sandoval
,
M.
,
Thomas
,
S.
and
Hart
,
D.
(
2017
).
Assay for transposase-accessible chromatin with high-throughput sequencing (ATAC-Seq) protocol for zebrafish embryos
.
Methods Mol. Biol.
1507
,
59
-
66
.
El-Brolosy
,
M. A.
,
Kontarakis
,
Z.
,
Rossi
,
A.
,
Kuenne
,
C.
,
Günther
,
S.
,
Fukuda
,
N.
,
Kikhi
,
K.
,
Boezio
,
G. L. M.
,
Takacs
,
C. M.
,
Lai
,
S.-L.
, et al. 
(
2019
).
Genetic compensation triggered by mutant mRNA degradation
.
Nature
568
,
193
-
197
.
Fairchild
,
C. L.
and
Gammill
,
L. S.
(
2013
).
Tetraspanin18 is a FoxD3-responsive antagonist of cranial neural crest epithelial-to-mesenchymal transition that maintains cadherin-6B protein
.
J. Cell Sci.
126
,
1464
-
1476
.
Ferrara
,
N.
,
Carver-Moore
,
K.
,
Chen
,
H.
,
Dowd
,
M.
,
Lu
,
L.
,
O'Shea
,
K. S.
,
Powell-Braxton
,
L.
,
Hillan
,
K. J.
and
Moore
,
M. W.
(
1996
).
Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene
.
Nature
380
,
439
-
442
.
Field
,
H. A.
,
Si Dong
,
P. D.
,
Beis
,
D.
and
Stainier
,
D. Y. R.
(
2003a
).
Formation of the digestive system in zebrafish. II. Pancreas morphogenesis
.
Dev. Biol.
261
,
197
-
208
.
Field
,
H. A.
,
Ober
,
E. A.
,
Roeser
,
T.
and
Stainier
,
D. Y. R.
(
2003b
).
Formation of the digestive system in zebrafish. I. Liver morphogenesis
.
Dev. Biol.
253
,
279
-
290
.
Gomez
,
G.
,
Lee
,
J.-H.
,
Veldman
,
M. B.
,
Lu
,
J.
,
Xiao
,
X.
and
Lin
,
S.
(
2012
).
Identification of vascular and hematopoietic genes downstream of ETSRP by deep sequencing in zebrafish
.
PLoS ONE
7
,
e31658
.
Herbert
,
S. P.
,
Huisken
,
J.
,
Kim
,
T. N.
,
Feldman
,
M. E.
,
Houseman
,
B. T.
,
Wang
,
R. A.
,
Shokat
,
K. M.
and
Stainier
,
D. Y. R.
(
2009
).
Arterial/venous segregation by selective cell sprouting: an alternative mode of blood vessel formation
.
Science
326
,
294
-
298
.
Hermkens
,
D. M. A.
,
van Impel
,
A.
,
Urasaki
,
A.
,
Bussmann
,
J.
,
Duckers
,
H. J.
and
Schulte-Merker
,
S.
(
2015
).
Sox7 controls arterial specification in conjunction with hey2 and efnb2 function
.
Development
142
,
1695
-
1704
.
Hesselson
,
D.
,
Anderson
,
R. M.
,
Beinat
,
M.
and
Stainier
,
D. Y. R.
(
2009
).
Distinct populations of quiescent and proliferative pancreatic beta-cells identified by HOTcre mediated labeling
.
Proc. Natl. Acad. Sci. USA
106
,
14896
-
14901
.
Ho
,
C.-Y.
,
Houart
,
C.
,
Wilson
,
S. W.
and
Stainier
,
D. Y. R.
(
1999
).
A role for the extraembryonic yolk syncytial layer in patterning the zebrafish embryo suggested by properties of the hex gene
.
Curr. Biol.
9
,
1131
-
1134
.
Iwafuchi-Doi
,
M.
and
Zaret
,
K. S.
(
2014
).
Pioneer transcription factors in cell reprogramming
.
Genes Dev.
28
,
2679
-
2692
.
Jiang
,
L.
and
Crews
,
S. T.
(
2007
).
Transcriptional specificity of Drosophila dysfusion and the control of tracheal fusion cell gene expression
.
J. Biol. Chem.
282
,
28659
-
28668
.
Jin
,
S.-W.
,
Herzog
,
W.
,
Santoro
,
M. M.
,
Mitchell
,
T. S.
,
Frantsve
,
J.
,
Jungblut
,
B.
,
Beis
,
D.
,
Scott
,
I. C.
,
D'Amico
,
L. A.
,
Ober
,
E. A.
, et al. 
(
2007
).
A transgene-assisted genetic screen identifies essential regulators of vascular development in vertebrate embryos
.
Dev. Biol.
307
,
29
-
42
.
Klaric
,
T.
,
Lardelli
,
M.
,
Key
,
B.
,
Koblar
,
S.
and
Lewis
,
M.
(
2014
).
Activity-dependent expression of neuronal PAS domain-containing protein 4 (npas4a) in the developing zebrafish brain
.
Front. Neuroanat.
8
,
148
.
Lawson
,
N. D.
and
Weinstein
,
B. M.
(
2002
).
In vivo imaging of embryonic vascular development using transgenic zebrafish
.
Dev. Biol.
248
,
307
-
318
.
Liao
,
W.
,
Bisgrove
,
B. W.
,
Sawyer
,
H.
,
Hug
,
B.
,
Bell
,
B.
,
Peters
,
K.
,
Grunwald
,
D. J.
and
Stainier
,
D. Y.
(
1997
).
The zebrafish gene cloche acts upstream of a flk-1 homologue to regulate endothelial cell differentiation
.
Development
124
,
381
-
389
.
Liao
,
W.
,
Ho
,
C. Y.
,
Yan
,
Y. L.
,
Postlethwait
,
J.
and
Stainier
,
D. Y.
(
2000
).
Hhex and scl function in parallel to regulate early endothelial and blood differentiation in zebrafish
.
Development
127
,
4303
-
4313
.
Link
,
V.
,
Shevchenko
,
A.
and
Heisenberg
C. P.
(
2006
).
Proteomics of early zebrafish embryos
.
BMC Dev. Biol. 6, 1
.
Lo
,
R.
and
Matthews
,
J.
(
2012
).
High-resolution genome-wide mapping of AHR and ARNT binding sites by ChIP-Seq
.
Toxicol. Sci.
130
,
349
-
361
.
Marcelo
,
K. L.
,
Goldie
,
L. C.
and
Hirschi
,
K. K.
(
2013
).
Regulation of endothelial cell differentiation and specification
.
Circ. Res.
112
,
1272
-
1287
.
Parker
,
L.
and
Stainier
,
D. Y.
(
1999
).
Cell-autonomous and non-autonomous requirements for the zebrafish gene cloche in hematopoiesis
.
Development
126
,
2643
-
2651
.
Pham
,
V. N.
,
Lawson
,
N. D.
,
Mugford
,
J. W.
,
Dye
,
L.
,
Castranova
,
D.
,
Lo
,
B.
and
Weinstein
,
B. M.
(
2007
).
Combinatorial function of ETS transcription factors in the developing vasculature
.
Dev. Biol.
303
,
772
-
783
.
Qian
,
F.
,
Zhen
,
F.
,
Ong
,
C.
,
Jin
,
S.-W.
,
Meng Soo
,
H.
,
Stainier
,
D. Y. R.
,
Lin
,
S.
,
Peng
,
J.
and
Wen
,
Z.
(
2005
).
Microarray analysis of zebrafish cloche mutant using amplified cDNA and identification of potential downstream target genes
.
Dev. Dyn.
233
,
1163
-
1172
.
Reischauer
,
S.
,
Stone
,
O. A.
,
Villasenor
,
A.
,
Chi
,
N.
,
Jin
,
S.-W.
,
Martin
,
M.
,
Lee
,
M. T.
,
Fukuda
,
N.
,
Marass
,
M.
,
Witty
,
A.
, et al. 
(
2016
).
Cloche is a bHLH-PAS transcription factor that drives haemato-vascular specification
.
Nature
535
,
294
-
298
.
Rossi
,
A.
,
Kontarakis
,
Z.
,
Gerri
,
C.
,
Nolte
,
H.
,
Hölper
,
S.
,
Krüger
,
M.
and
Stainier
,
D. Y. R.
(
2015
).
Genetic compensation induced by deleterious mutations but not gene knockdowns
.
Nature
524
,
230
-
233
.
Samarut
,
E.
,
Lissouba
,
A.
and
Drapeau
,
P.
(
2016
).
A simplified method for identifying early CRISPR-induced indels in zebrafish embryos using High Resolution Melting analysis
.
BMC Genomics
17
,
547
.
Schodel
,
J.
,
Oikonomopoulos
,
S.
,
Ragoussis
,
J.
,
Pugh
,
C. W.
,
Ratcliffe
,
P. J.
and
Mole
,
D. R.
(
2011
).
High-resolution genome-wide mapping of HIF-binding sites by ChIP-seq
.
Blood
117
,
e207
-
e217
.
Scialdone
,
A.
,
Tanaka
,
Y.
,
Jawaid
,
W.
,
Moignard
,
V.
,
Wilson
,
N. K.
,
Macaulay
,
I. C.
,
Marioni
,
J. C.
and
Göttgens
,
B.
(
2016
).
Resolving early mesoderm diversification through single-cell expression profiling
.
Nature
535
,
289
-
293
.
Sorensen
,
A. T.
,
Cooper
,
Y. A.
,
Baratta
,
M. V.
,
Weng
,
F.-J.
,
Zhang
,
Y.
,
Ramamoorthi
,
K.
,
Fropf
,
R.
,
LaVerriere
,
E.
,
Xue
,
J.
,
Young
,
A.
, et al. 
(
2016
).
A robust activity marking system for exploring active neuronal ensembles
.
eLife
5
,
e13918
.
Spiegel
,
I.
,
Mardinly
,
A. R.
,
Gabel
,
H. W.
,
Bazinet
,
J. E.
,
Couch
,
C. H.
,
Tzeng
,
C. P.
,
Harmin
,
D. A.
and
Greenberg
,
M. E.
(
2014
).
Npas4 regulates excitatory-inhibitory balance within neural circuits through cell-type-specific gene programs
.
Cell
157
,
1216
-
1229
.
Stainier
,
D. Y.
,
Weinstein
,
B. M.
,
Detrich
,
H. W.
, III
,
Zon
,
L. I.
and
Fishman
,
M. C.
(
1995
).
Cloche, an early acting zebrafish gene, is required by both the endothelial and hematopoietic lineages
.
Development
121
,
3141
-
3150
.
Sumanas
,
S.
,
Jorniak
,
T.
and
Lin
,
S.
(
2005
).
Identification of novel vascular endothelial-specific genes by the microarray analysis of the zebrafish cloche mutants
.
Blood
106
,
534
-
541
.
Thisse
,
C.
and
Thisse
,
B.
(
2008
).
High-resolution in situ hybridization to whole-mount zebrafish embryos
.
Nat. Protoc.
3
,
59
-
69
.
Thompson
,
M. A.
,
Ranson
,
D. G.
,
Pratt
,
S. J.
,
MacLennan
,
H.
,
Kieran
,
W. M.
,
Detrich
,
H. W.
,
Vail
,
B.
,
Huber
,
T. L.
,
Paw
,
B.
,
Brownlie
,
A. J.
, et al. 
(
1998
).
The cloche and spadetail genes differentially affect hematopoiesis and vasculogenesis
.
Dev. Biol.
197
,
248
-
269
.
Tsang
,
K. M.
,
Hyun
,
J. S.
,
Cheng
,
K. T.
,
Vargas
,
M.
,
Mehta
,
D.
,
Ushio-Fukai
,
M.
,
Zou
,
L.
,
Pajcini
,
K. V.
,
Rehman
,
J.
and
Malik
,
A. B.
(
2017
).
Embryonic stem cell differentiation to functional arterial endothelial cells through sequential activation of ETV2 and NOTCH1 signaling by HIF1alpha
.
Stem Cell Rep.
9
,
796
-
806
.
Vogeli
,
K. M.
,
Jin
,
S.-W.
,
Martin
,
G. R.
and
Stainier
,
D. Y. R.
(
2006
).
A common progenitor for haematopoietic and endothelial lineages in the zebrafish gastrula
.
Nature
443
,
337
-
339
.
Weiss
,
O.
,
Kaufman
,
R.
,
Michaeli
,
N.
and
Inbal
,
A.
(
2012
).
Abnormal vasculature interferes with optic fissure closure in lmo2 mutant zebrafish embryos
.
Dev. Biol.
369
,
191
-
198
.
Wong
,
K. S.
,
Proulx
,
K.
,
Rost
,
M. S.
and
Sumanas
,
S.
(
2009
).
Identification of vasculature-specific genes by microarray analysis of Etsrp/Etv2 overexpressing zebrafish embryos
.
Dev. Dyn.
238
,
1836
-
1850
.

Competing interests

The authors declare no competing or financial interests.

Supplementary information