HDAC1-mediated repression of the retinoic acid-responsive gene ripply3 promotes second heart field development

Abstract

Coordinated transcriptional and epigenetic mechanisms that direct development of the later differentiating second heart field (SHF) progenitors remain largely unknown. Here, we show that a novel zebrafish histone deacetylase 1 (hdac1) mutant allele cardiac really gone (crg) has a deficit of ventricular cardiomyocytes (VCs) and smooth muscle within the outflow tract (OFT) due to both cell and non-cell autonomous loss in SHF progenitor proliferation. Cyp26-deficient embryos, which have increased retinoic acid (RA) levels, have similar defects in SHF-derived OFT development. We found that nkx2.5+ progenitors from Hdac1 and Cyp26-deficient embryos have ectopic expression of ripply3, a transcriptional co-repressor of T-box transcription factors that is normally restricted to the posterior pharyngeal endoderm. Furthermore, the ripply3 expression domain is expanded anteriorly into the posterior nkx2.5+ progenitor domain in crg mutants. Importantly, excess ripply3 is sufficient to repress VC development, while genetic depletion of Ripply3 and Tbx1 in crg mutants can partially restore VC number. We find that the epigenetic signature at RA response elements (RAREs) that can associate with Hdac1 and RA receptors (RARs) becomes indicative of transcriptional activation in crg mutants. Our study highlights that transcriptional repression via the epigenetic regulator Hdac1 facilitates OFT development through directly preventing expression of the RA-responsive gene ripply3 within SHF progenitors.

About the Authors:

Yuntao Charlie Song

Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

Affiliations Division of Molecular Cardiovascular Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, United States of America, Molecular and Developmental Biology Graduate Program, University of Cincinnati, Cincinnati, OH, United States of America

ORCID logo http://orcid.org/0000-0003-0218-8601

Tracy E. Dohn

Roles Data curation, Formal analysis, Funding acquisition, Investigation, Validation, Visualization, Writing – review & editing

Affiliations Division of Molecular Cardiovascular Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, United States of America, Molecular and Developmental Biology Graduate Program, University of Cincinnati, Cincinnati, OH, United States of America

ORCID logo http://orcid.org/0000-0001-7044-4350

Ariel B. Rydeen

Roles Data curation, Formal analysis, Funding acquisition, Investigation, Writing – review & editing

Affiliations Division of Molecular Cardiovascular Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, United States of America, Molecular and Developmental Biology Graduate Program, University of Cincinnati, Cincinnati, OH, United States of America

ORCID logo http://orcid.org/0000-0002-8865-6420

Alex V. Nechiporuk

Roles Funding acquisition, Resources

Affiliation: Department of Cell and Developmental Biology, Oregon Health & Science University, Portland, OR, United States of America

Joshua S. Waxman

Roles Conceptualization, Formal analysis, Funding acquisition, Supervision, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliations Division of Molecular Cardiovascular Biology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, United States of America, Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, United States of America

ORCID logo http://orcid.org/0000-0002-8132-487X

Introduction

The ultimate size of all vertebrate hearts is determined through a continuous contribution of differentiating cardiac progenitors, which are often described as the first and second heart field. Earlier-differentiating cardiac progenitors from the first heart field (FHF) migrate to the midline from the anterior lateral plate mesoderm, begin to differentiate into cardiomyocytes (CMs), and form the heart tube. Subsequently, later-differentiating cardiac progenitors from the second heart field (SHF), which lie in the adjacent dorsal and medial pharyngeal mesoderm, contribute to the growth of the heart through adding CMs, as well as smooth muscle and endothelial cells, to the poles [1–9]. In the 4-chambered heart of birds and mammals, the later differentiating cells of the SHF predominantly contribute to both of the atria, the right ventricle, and smooth muscle and endothelial cells of the outflow tract (OFT) [10]. In the comparatively simple two-chamber zebrafish heart, the SHF contributes to the OFT, which includes CMs at the arterial pole of the single ventricle as well as smooth muscle and endothelial cells within the bulbous arteriosus, and a small percentage of cells at the venous pole of the atrium [11–14]. Importantly, despite the differences in chamber number between fish, birds and mammals, conserved signals and transcription factors, including Bmp [15], Wnt [16, 17], Hedgehog (HH) [18, 19], Fgf [20, 21], and retinoic acid (RA) [22–24], regulate the allocation and deployment of cells from the SHF. Although we have gained insight into the roles of these molecular players in guiding SHF development, epigenetic mechanisms that influence their transcriptional inputs during SHF development remain poorly understood.

Significant epigenetic regulation that effects transcription is through modification of histones [25]. One group of these epigenetic modifiers is histone deacetylases (HDACs). Class I HDACs, consisting of HDACs1-3 and HDAC8, are primarily thought to be transcriptional co-repressors, which deacetylate histone tails and consequently causes compaction of chromatin [26]. Despite HDACs of this class being ubiquitously expressed during development, they have been shown to have tissue-specific requirements in development and disease [27]. For instance, while global HDAC1 knockout (KO) mice are early embryonic lethal, largely due to broad proliferation defects [28], tissue-specific deletion within CMs does not produce overt defects, which is thought to be due to redundancy with the nearly identical HDAC2. The requirement of HDAC2 alone may be somewhat controversial. One report indicates HDAC2 KO mice have no overt defects [29]. However, another study indicated that HDAC2 KO mice die shortly after birth with enlarged hearts from increased CM proliferation due to hyperacetylation directly on the transcription factor Gata4 [30]. Despite the controversy over HDAC2 alone, functions of HDAC1 and HDAC2 do appear to be redundant in the mammalian heart. Conditional double HDAC1 and HDAC2 KO mice in the heart using the αMHC:Cre transgene display postnatal defects in cardiac morphogenesis, growth, and contractility that contribute to cardiac arrhythmias and severe ventricular dilation [29]. Unlike mammals, zebrafish do not have a Hdac2 ortholog, suggesting that in zebrafish Hdac1 may perform functions of both Hdac1 and Hdac2 in mammals. Although a previous study suggested that zebrafish hdac1 mutants may have CM differentiation defects [31], the mechanisms underlying this defect are not understood. Therefore, if HDAC1 homologs regulate aspects of early vertebrate heart development, and specifically SHF development, remains unknown.

HDACs are unable to bind to DNA directly [27]. Thus, their function as epigenetic regulators of transcriptional repression depends on interactions with transcription factors and repressive complexes [27]. Retinoic acid receptors (RARs) are one of the transcription factors involved in recruiting HDACs to DNA [32–37]. At the transcriptional level, the prevailing model is that in the absence of RA the RARs repress transcription of their targets at RA response elements (RAREs) through interaction with transcriptional co-repressors, including HDAC1 [33, 34, 38, 39]. When RA binds to RARs, it induces conformational changes that shed HDACs in favor of transcriptional activators [32, 40, 41]. RA is a potent teratogen [42], with excess embryonic RA signaling perturbing numerous aspects of heart development [43, 44], including preventing SHF accrual to the OFT [24, 45]. Despite this prevailing transcriptional model and recent examples demonstrating alternative mechanisms of ligand-mediate transcriptional repression in the somites [46], there are few examples of requisite RAR-mediated transcriptional repression during development and none that are known to regulate heart development.

In this study, we demonstrate that the zebrafish mutant cardiac really gone (crg) is a novel hdac1 loss-of-function allele. Interestingly, crg mutants have smaller hearts due to a loss of SHF-derived ventricular cardiomyocytes (VCs) and smooth muscle within the OFT, which is similar to embryos deficient for the RA-degrading Cyp26 enzymes [24]. We identified that nkx2.5+ cells from both Hdac1 and Cyp26-deficient embryos have ectopic expression of ripply3, a transcriptional co-repressor of T-box (Tbx) transcription factors that is ordinarily restricted to the posterior pharyngeal endoderm. We find excess Ripply3 is sufficient to reduce VC number, while depletion of Ripply3 or its binding partner Tbx1 partially restores VC number in crg mutants. Furthermore, consistent with the ectopic expression of ripply3, epigenetic markers within the ripply3 promoter are switched to a state that favors transcriptional activation in crg mutants. Altogether, our study reveals that the epigenetic regulator Hdac1 facilitates OFT development through limiting the anterior encroachment of the RA-responsive gene ripply3 into posterior SHF progenitors.

Results

Crg mutants display defects in early SHF development

Through a N-ethyl N-nitrososurea (ENU)-induced mutagenesis screen for recessive alleles [47], we identified a mutant we named cardiac really gone (crg) that has overtly smaller hearts coupled with pericardial edema, as well as developmental defects including reduced pigmentation, shortened yolk extension, and a curved body axis (Fig 1A and 1B). Examining hearts with immunohistochemistry (IHC) revealed that in crg mutants the hearts are more linear, the size of the ventricle is reduced, and the atria are more bulbous at 48 hours post-fertilization (hpf) compared to wild-type (WT) sibling embryos (Fig 1C and 1D). Interestingly, counting CMs using the myl7:DsRed-NLS transgene [48] revealed that the number of VCs was reduced in crg mutants compared to their WT sibling embryos at 36 hpf through 72 hpf (Fig 1E–1G). However, despite the smaller, bulbous morphology of the atria, the number of atrial cardiomyocytes (ACs) was not significantly affected in crg mutants (Fig 1E–1G). Therefore, crg mutants have a specific deficit in VCs within the developing hearts of crg mutants, which is not caused by developmental delay.

[Figure omitted. See PDF.]

Fig 1. Crg mutants have a specific deficit in VCs.

(A-B) WT sibling and crg mutants at 48 hpf. Lateral views with anterior to the left. Arrow in B indicates pericardial edema. (C-D) Hearts from WT sibling and crg mutant myl7:NLS-DsRed2 embryos at 48 hpf. Frontal views. Purple alone indicates ventricle. Yellow indicates atrium. Arrows indicate arterial pole of the ventricle. (E-G) Quantification of CMs in the atria and ventricles of WT sibling and crg mutant myl7:DsRed2-NLS embryos at 36, 48, and 72 hpf. For 36 hpf, n = 14 for WT and crg mutants. For 48 hpf, n = 20 for WT and crg mutants. For 72 hpf, n = 17 for WT and crg mutants. Asterisk in all graphs indicates p<0.05 as determined by Student’s t-test. Error bars for all graphs indicate s.e.m.

https://doi.org/10.1371/journal.pgen.1008165.g001

To determine when the heart defects in crg mutants first arise, we examined the expression of CM specification and differentiation at earlier embryonic stages. By in situ hybridization (ISH), we found that in crg mutants neither specification of cardiac progenitors (indicated by hand2 and gata4) was affected at the 8 somite (s) stage nor was early differentiation of CMs (indicated by myl7) affected at the 20s stage (S1 Fig), suggesting early cardiac specification and FHF development are not dramatically affected. Therefore, we next quantified expression of myl7 and the SHF marker ltbp3 [1] at later stages of heart development with reverse transcription quantitative PCR (RT-qPCR). We found that at both 36 and 48 hpf myl7 and ltbp3 expression were reduced (Fig 2A and 2B). Moreover, ltbp3 expression was absent from the developing OFT of crg mutants at 30 hpf using two-color fluorescent in situ hybridization (FISH) (Fig 2C–2D”). In contrast to ltbp3, we found that the SHF marker mef2cb appeared to have increased expression in progenitor cells adjacent to the arterial pole in crg mutants at 30 hpf (S2 Fig), although quantitatively we did not find a difference in its expression at 36 hpf in crg mutants compared to WT siblings (S2 Fig). Together, these results imply that later-differentiating SHF-derived populations may be lost in crg mutants.

[Figure omitted. See PDF.]

Fig 2. SHF markers of the arterial pole are reduced in crg mutants.

(A,B) RT-qPCR for the pan-cardiac differentiation marker myl7 and SHF marker ltbp3 from embryos at 36 and 48 hpf. (C-D”) Two-color FISH for myl7 and ltbp3 in WT sibling and crg mutant embryos at 30 hpf. Brackets in C and D indicate presence and absence of ltbp3 at the arterial pole of WT and crg mutant hearts, respectively. n = 5 WT and n = 5 crg mutants embryos were examined.

https://doi.org/10.1371/journal.pgen.1008165.g002

To directly quantify the number of later-differentiating VCs in embryos, we used the established myl7:NLS-Kikgr transgene [2], which expresses a nuclear photo-convertible KikGR protein in differentiated CMs. Indeed, following conversion of cardiac NLS-KikGR from green to red (pseudo-colored as yellow and purple) at 30 hpf, we found that there was an ~50% reduction in the number of later-accruing green+ (yellow+)/red-(purple-) VCs within the OFT of crg mutants (Fig 3A–3C). Similar trends were obtained using the myl7:Kaede transgene [4] and photoconverting at 36 hpf (S3 Fig). Next, because the SHF gives rise to smooth muscle in addition to VCs, we analyzed smooth muscle within the OFT using IHC for Elastin b (Elnb) and MHC (CMs and skeletal muscle). Elnb+ cells of the OFT were not found in crg mutants at 72 hpf (Fig 3D and 3E). Consistent with the Elnb IHC, crg mutant hearts never stained for DAF-2DA (Fig 3F and 3G), which labels functional smooth muscle [49], further supporting these smooth muscle cells fail to differentiate in crg mutants. Altogether, these results support that impaired SHF development produces VC and smooth muscle defects with the arterial ventricle and OFT of crg mutants.

[Figure omitted. See PDF.]

Fig 3. SHF-derived VC and smooth muscle development is impaired in the crg mutants.

(A-B”) Representative images of hearts from photoconverted WT sibling and crg mutant myl7:NLS-KikGR embryos at 48 hpf. The arterial poles (brackets) are to the right. (C) Quantification of later-differentiating VCs (Yellow+/Purple- cells) (n = 10 for WT and crg mutants). (D,E) Confocal images of IHC for MHC and Elnb in WT and crg mutant embryos at 72 hpf. n = 10 WT and n = 10 crg mutants embryos examined. (F,G) Confocal images of DAF-2DA staining coupled with IHC for Vmhc in WT sibling and crg mutant embryos at 96 hpf. n = 10 WT and n = 10 crg mutants embryos examined. Images in D-G are frontal views with anterior up. Arrows indicate Elnb and DAF-2DA staining of the bulbous arteriosus.

https://doi.org/10.1371/journal.pgen.1008165.g003

Crg mutants contain a novel hdac1 loss-of-function allele

To identify the gene affected in crg mutants, we conducted traditional positional cloning with simple sequence length polymorphisms (SSLPs), which narrowed down the site of the mutation to within an ~0.5Mb region on Chr19 containing a cluster of 6 genes (Fig 4A). Subsequent RNA-seq on WT sibling and crg mutant embryos identified a mutation (T>A) in the splicing donor site of hdac1’s exon 7 linked to the crg phenotype (Fig 4B). The splice-site mutation appears have two consequences on hdac1 transcripts: it causes either the skipping of exon 7 or retention of both introns 6 and 7 (Fig 4B and 4C). The two possible outcomes for hdac1 transcripts in crg mutants respectively predicted the potential for proteins to be made with either a 32 amino acid deletion within the histone deacetylase domain or to be severely truncated from going out of frame after amino acid 213 (Fig 4D). However, Western Blot showed that neither WT Hdac1 protein nor smaller predicted proteins were detected in crg mutants (Fig 4E and S4 Fig). Thus, our data support that inappropriate splicing of hdac1 transcripts in the crg allele leads to a loss of Hdac1 protein. As further confirmation that the crg allele affects hdac1, the overt phenotype of crg mutants is similar to other previously reported hdac1 mutant alleles [50–52]. Additionally, injection with a previously verified hdac1 morpholino oligonucleotide (MO), which inhibits Hdac1 translation (S4 Fig), and treatment with the HDAC inhibitor Trichostatin A (TSA) both produced overt embryo body and cardiac defects equivalent to crg mutants (S5 Fig). Importantly, counting the number of CMs in Hdac1-depleted and TSA-treated embryos indicated a specific deficit of VCs in their hearts (S5 Fig), consistent with what we found in crg mutants. Therefore, these data show that crg is a novel loss-of-function hdac1 mutant allele and support that Hdac1 loss results in specific deficit of VCs within hearts.

[Figure omitted. See PDF.]

Fig 4. Crg mutants contain a loss of function mutation in hdac1.

(A) Summary of the region containing the crg mutation from positional cloning. Fractions indicate recombinants for SSLPs. (B) Reads from RNA-seq data indicating the mutation and effect on the hdac1 exon 7. Boxes indicate mutation with T to A (green bar) change affecting splice donor site. (C) Schematic indicating the skipping of exon 7 or the retention of intron 6 and 7 found in hdac1 transcripts from crg mutants. (D) Predicted consequences on the proteins generated from the improper splice forms of the hdac1 transcripts, if they were translated. The red bar indicates the 32 amino acid deletion predicted to occur from the transcript that skips exon 7 in crg mutants. The gray bar indicates a 9 amino acid extension after going out of frame at amino acid 213 for the hdac1 transcript that retains introns 6 and 7 in crg mutants. (E) Western blot for Hdac1 protein in WT sibling and crg mutants.

https://doi.org/10.1371/journal.pgen.1008165.g004

Hdac1 has cell and cell non-autonomous requirements promoting proliferation of SHF progenitors

Having found that Hdac1 loss underlies a failure to accrue a significant portion of SHF-derived cells in the OFT of crg mutants, we next sought to understand if SHF progenitors were affected in crg mutants. We assessed the SHF progenitors using IHC co-staining for Nkx2.5, which marks both SHF progenitors and differentiated CMs, and MHC, which marks differentiated CMs (Fig 5A–5B”). Counting the number of Nkx2.5+/MHC- cells adjacent to the OFT at 33 hpf revealed that there are fewer SHF progenitors in crg mutants (Fig 5C). Proliferation is reduced in some HDAC1-dependent developmental contexts [28, 53–55]. Therefore, we assessed if reduced cell proliferation may contribute to the diminished number of Nkx2.5+ SHF progenitors by co-staining these embryos for the mitotic marker phospho-Histone H3 (PHH3) (Fig 5A–5B”“). Indeed, we found that the percentage of Nkx2.5+/MHC-/PHH3+ cells was reduced in crg mutants (Fig 5D). Furthermore, in isolated nkx2.5:ZsYellow+ cells sorted at 33 hpf, Hdac1-depleted embryos had a dramatic increase in expression of the cell cycle inhibitor cdkn1a/p21 [56] (Fig 5E). However, while these sorted cells contain SHF progenitors, a caveat is that they likely also contain differentiated cardiomyocytes and pharyngeal arch progenitors [57, 58].

[Figure omitted. See PDF.]

Fig 5. Hdac1 is required for the proliferation of SHF progenitors.

(A-B”‘) Confocal images of IHC for hearts and Nkx2.5+ SHF progenitors in WT sibling and crg mutant embryos at 33 hpf. Nkx2.5+ (green), PHH3 (blue) and MHC (purple). Outline in A and B indicates Nkx2.5+/MHC- SHF. Arrows indicate Nkx2.5+/MHC-/PHH3+ cells. Yellow arrow indicates Nkx2.5+/MHC-/pHH3+ cell of the higher magnification inset. (A”“and B”“) Schematic indicating IHC from A and B. Green indicates Nkx2.5+/MHC- cells. Blue indicates Nkx2.5+/MHC-/PHH3+ cells. Purple indicates MHC+ cells. Anterior is up in A-B”“. (C) Quantification of SHF progenitors (Nkx2.5+/MHC-). (D) Percentage of PHH3+ SHF progenitors. For C and D, n = 14 for WT and n = 15 crg mutants. (E) RT-qPCR for cdkn1a from sorted nkx2.5:ZsYellow+ cells.

https://doi.org/10.1371/journal.pgen.1008165.g005

Given hdac1 is expressed ubiquitously in early zebrafish embryos [52], we next wanted to determine the cells that require Hdac1 for VC development. To assess cell-autonomy, we first performed blastula cell transplantation experiments with donor cells from WT(control) and Hdac1-depleted myl7:GFP embryos that were placed into WT hosts (Fig 6A). The myl7:GFP transgene was used so that CMs will unambiguously be marked with GFP in host embryos (Fig 6A). Both the frequency of donor GFP+ CM incorporation and the number of donor GFP+ CMs in the host atria and ventricles were quantified. When transplanting cells into WT hosts, we found that the frequency of finding donor GFP+ CMs was not changed between hosts receiving donor cells from WT or Hdac1-depleted embryos (Fig 6B), suggesting that Hdac1 loss does not affect early CM specification. However, the average number of GFP+ CMs found in the ventricles, but not the atria, was significantly reduced from Hdac1-depleted donor cells compared to WT donor cells (Fig 6C). In complementary transplantation experiments where WT donor cells were placed into WT or Hdac1-depleted hosts (Fig 6A), there also was not a statistical change in the frequency of donor GFP+ CMs contributing to either set of host embryos (Fig 6D). However, WT donor GFP+ CMs demonstrated a decreased frequency of contribution to the ventricles when placed into Hdac1-depleted hosts (Fig 6E). Together, these results are consistent with Hdac1 having both cell-autonomous and cell non-autonomous requirements in promoting the proliferation of SHF progenitors.

[Figure omitted. See PDF.]

Fig 6. Hdac1 is required both cell and non-cell autonomously to promote VCs.

(A) Schematic of the blastula cell transplantation strategy. Vmhc (purple) indicates ventricle. Arrows indicate myl7:GFP+ cells (yellow). (B) Frequency of host WT embryos with WT and Hdac1-depleted donor GFP+ CMs at 48 hpf. n = 245 WT donor into WT host transplants; n = 244 Hdac1 deficient into WT host transplants. (C) Quantification of donor CMs (GFP+) found in host embryos hearts from transplants (n = 19 for WT in WT and n = 18 Hdac1-depleted in WT). (D) Frequency of host WT and Hdac1-depleted embryos with WT donor GFP+ CMs at 48 hpf. n = 169 WT donor into WT host transplants; n = 177 WT donor into Hdac1-depleted host transplants. (E) Quantification of donor CMs (GFP+) found in host embryos hearts from transplants (n = 18 for WT in WT and n = 12 WT in Hdac1-depleted). Fisher exact test was used to compare significance of frequencies in B and D.

https://doi.org/10.1371/journal.pgen.1008165.g006

Ripply3 is an effector of Hdac1 in cardiac progenitors

RARs are one of the transcription factors that can recruit Hdac1 to chromatin, which it does in the absence of RA ligand [32, 38, 59, 60]. Our recent studies indicate that Cyp26-deficient (loss of both Cyp26a1 and Cyp26c1) embryos, which have ectopic levels of RA, display smaller hearts with a loss of SHF-derived VCs similar to crg embryos [24]. Given the established relationship between RARs and Hdac1, we postulated that Hdac1- and Cyp26-deficiency may result in the de-repression of common effector genes that contribute to SHF-derived OFT defects. To identify candidate genes with increased expression in both Hdac1- and Cyp26-deficient embryos, we examined trends of gene expression in RNA-seq analysis of whole embryos at 48 hpf and fluorescence-activated cell sorting (FACS)-isolated nkx2.5:ZsYellow+ cells at 28 hpf (S6 Fig). One gene we chose for further analysis is ripply3, as previous studies in other models have shown its expression is responsive to RA [61, 62]. Therefore, we subsequently confirmed it has increased expression in both Hdac1- and Cyp26-deficient embryos and isolated nkx2.5:ZsYellow+ cells at these same stages using RT-qPCR (Fig 7A–7D). Additionally, RT-qPCR for SHF marker genes ltbp3 and mef2cb in these isolated nkx2.5:ZsYellow+ cells showed the same trends that was found when analyzing their expression in mutants (S8 Fig, Fig 2 and S2 Fig)

[Figure omitted. See PDF.]

Fig 7. Ripply3 is expanded anteriorly into nkx2.5+ cells in crg mutants.

(A-D) RT-qPCR for ripply3 expression in whole embryos at 48 hpf and sorted nkx2.5:ZsYellow+ cells at 28 hpf. (E-H”) Confocal images of two-color FISH for nkx2.5 and ripply3 in WT and crg mutant embryos at 18 and 24 hpf. Images are dorsal views with anterior up. Insets in F-H indicate lateral views of the confocal images. Bracket in E indicates space between posterior nkx2.5 and anterior ripply3 domains. Arrow in F indicates border between nkx2.5 in pharyngeal mesoderm and ripply3 in pharyngeal endoderm. Brackets with arrows in G and H indicate overlap in nkx2.5 and ripply3 domains in crg mutant embryos. n = 22 WT and n = 5 crg mutants embryos for 18 hpf and n = 19 WT and n = 11 crg mutants embryos for 24 hpf examined. Scale bars in E and G are 50 μm. Scale bars in F and H are 100 μm.

https://doi.org/10.1371/journal.pgen.1008165.g007

We were intrigued by ripply3 (aka Down syndrome critical region 6 (DSRC6)) because it was originally identified as a gene within the portion of chromosome 21 that when supernumerary is associated with Down syndrome [63], which includes a high incidence of congenital heart defects [64]. Mechanistically, previous studies suggest Ripply proteins interact with T-box transcription factors and promote their function as transcriptional repressors [61, 62, 65–67]. Furthermore, Ripply3 KO mice have OFT and pharyngeal endoderm defects [62], with the latter thought to be from increased expression of Tbx1 target genes. Ripply3 in mice and Xenopus is restricted to pharyngeal ectoderm and endoderm [61, 67]. Therefore, we examined ripply3 expression in zebrafish and simultaneously compared its localization to nkx2.5 expression using two-color FISH from the 18s stage through 36 hpf. Similar to Ripply3 in mice and Xenopus [68], we found that by the 18s stage, its expression in zebrafish embryos was bilateral and posterior relative to the majority of nkx2.5 expression in the forming cardiac cone (Fig 7E–7E”). By 24 hpf, ripply3 resolves to being localized within the posterior pharyngeal endoderm adjacent to nkx2.5 expression within the pharyngeal mesoderm (Fig 7F–7F’), which will give rise to the posterior pharyngeal arch arteries. Ripply3 maintained this expression in the posterior pharyngeal endoderm adjacent to the developing posterior arch arteries through 36 hpf (S7 Fig).

Intriguingly, at the 18s stage ripply3 expression was shifted anteriorly in crg mutants (Fig 7G–7G”), which is reminiscent to posteriorization of tissues that occurs from excessive RA signaling [69]. Moreover, while the nkx2.5 and ripply3 expression domains were normally relatively far apart (Fig 7E–7E”), the posterior nkx2.5 and anterior ripply3 domains overlapped in crg mutants at the 18s stage (Fig 7G–7G”). At 24 hpf, ripply3 expression overlapped with the posterior nkx2.5+ cells (Fig 7H–7H”), which at this stage demarcates the developing pharyngeal arch arteries. At subsequent stages, the posterior expression of nkx2.5 was significantly diminished in crg mutants compared to WT siblings (S7 Fig). Although we were not able to consistently detect ripply3 expression within the nkx2.5+ domain with FISH at later stages in crg mutants (S7 Fig), this is likely due to the lower sensitively of ISH compared to RNA-seq and RT-qPCR. Thus, ripply3 expression is shifted anteriorly and overlaps with the posterior border of later-differentiating nkx2.5+ cardiac cells in crg mutants.

Because we found ectopic expression of ripply3 in crg mutants, we next determined if excess ripply3 is sufficient to induce similar VC deficits as crg mutants using an inducible transgenic line Tg(hsp70l:GFP-ripply3) that we generated (S9 Fig). Transgenic GFP-ripply3 embryos heat-shocked by the tailbud stage (10 hpf) displayed a truncated tail identical to ripply3 mRNA-injected embryos, supporting that the induced GFP-tagged Ripply3 is functional (S9 Fig). Since our data support that VC deficits in crg mutants may arise during the later phase of CM differentiation, we induced ripply3 expression at the 20s stage, when the heart has started to differentiate and form a cone [70] and approximately when we observed overlap in the cardiac progenitor and ripply3+ fields in crg mutants. Although the transgenic embryos were not overtly affected following GFP-ripply3 induction, we found the hearts were more linear with a reduction in VCs (Fig 8A–8E). Interestingly, ripply3 mRNA injection or GFP-ripply3 induction at the tailbud stage also produced a specific reduction in VCs (S9 Fig). Therefore, our data indicate that excess Ripply3 is sufficient to specifically inhibit the development of later-differentiating VCs.

[Figure omitted. See PDF.]

Fig 8. Loss of ripply3 in crg mutants partially restores VC number.

(A,B) Control and transgenic hsp70l:GFP-ripply3 embryos at 48 hpf following heat-shock at 20 hpf. (C,D) Hearts from control and hsp70l:GFP-ripply3 transgenic embryos with the myl7:DsRed2-NLS transgene at 48 hpf following heat-shock at 20 hpf. Purple alone indicates ventricle. Yellow indicates atrium. Arrows indicate arterial pole of the ventricle. (E) Quantification of CMs in control and hsp70l:GFP-ripply3 transgenic embryos at 48 hpf following GFP-ripply3 induction at 20 hpf (n = 18 for control, n = 29 for GFP-ripply3+). (F) Quantification of VCs in crgwt+het; ripply3wt+het, crgwt+het; ripply3-/-, crg-/-; ripply3wt+het, and crg-/-; ripply3-/- embryos at 48 hpf (n = 37 for crgwt+het; ripply3wt+het, n = 10 for crgwt+het; ripply3-/-, n = 48 for crg-/-; ripply3wt+het, n = 18 for crg-/-; ripply3-/-). (G) Quantification of VCs in WT (crgwt+het—uninjected) sibling, crgwt+het–Tbx1 depleted, crg-/-—uninjected, and crg-/-—Tbx1-depleted embryos at 48 hpf (n = 11 for crgwt+het—control uninjected, n = 10 for crgwt+het–Tbx1-depleted, n = 10 for crg-/-—control uninjected, n = 10 for crg-/-—Tbx1-depleted). For all CM quantification, embryos contained the myl7:DsRed2-NLS transgene.

https://doi.org/10.1371/journal.pgen.1008165.g008

Next, to determine if excess ripply3 contributes to the VC defects found in crg mutants, we engineered ripply3 mutants using CRISPR-Cas9 [71]. The ripply3 mutant allele we used deletes 61 bp within the 5’-untranslated region and 1st exon, which encompasses the translational start codon (S10 Fig). This mutation likely leads to non-sense mediated decay or fails to be transcribed, as we could no longer detect expression in the pharyngeal endoderm or via RT-qPCR in ripply3 mutants (S10 Fig). Unlike the ripply3 mutant mice, which have cardiac and pharyngeal defects [67], we found that ripply3 mutant zebrafish had no overt phenotypes and were homozygous viable. However, assessing CM number in ripply3; crg double mutant embryos, we found that the number of VCs is partially restored in crg mutants when ripply3 is lost (Fig 8F), supporting that excess ripply3 contributes to the loss of VCs in crg mutants. We next assessed if the VC defects in crg mutants also require its potential transcriptional partner Tbx1, since Ripply3 promotes transcriptional repression by Tbx1 [72] and Tbx1 has conserved requirements promoting proliferation of SHF progenitors [73–76]. To test this hypothesis, we injected a suboptimal dose of a previously verified tbx1 MO that phenocopies van gogh/tbx1 mutants [77] into embryos derived from adult crg+/- carriers. Use of the suboptimal dose helped overcome recently characterized early specification defects within the anterior lateral plate mesoderm of van gogh/tbx1 mutants [58]. Embryos were sorted at 48 hpf and the number of VCs quantified, as above. Similar to the ripply3; crg double mutants, we found a partial restoration of VCs in Tbx1-depleted crg mutants (Fig 8G). Manipulation of ripply3 expression did not affect tbx1 or cyp26a1 expression (S11 Fig). Together, these experiments indicate excess ripply3 contributes to the ventricular OFT defects in crg mutants, potentially through promoting transcriptional repression by Tbx1.

Repression of ripply3 in crg mutants requires RAREs

We next sought to gain insight into how Hdac1 and RA signaling may be regulating ripply3 expression. Previous analysis has suggested that RAREs are typically found within 5’kb promoter regions upstream of the transcriptional start site (TSS) of responsive genes [38, 78, 79]. Furthermore, according to the UCSC genome browser (Zv9/danRer7), the 5’kb upstream of the ripply3 TSS contains epigenetic marks indicative of poised DNA [38, 80–82]. Using NHR scan (www.cisreg.ca) [83] to identify putative RAREs, we found direct repeat 1 (DR1) and DR4 sites within this region (Fig 9A). Although DR4 sites are atypical RAR binding elements [84], electrophoretic mobility shift assays (EMSAs) indicated that RARs can bind both these RAREs in vitro (Fig 9B). We were not able to find ChIP-grade antibodies to the endogenous RARs. Therefore, to determine if RARs can bind these sites in vivo, we performed ChIP-PCR for GFP-VP16-RARab following heat-shock at 24hpf using our hsp70l:GFP-VP16-RARab transgenic line [85]. We found the RARs can bind these sites in vivo (Fig 9C). Importantly, ChIP-PCR for Hdac1 demonstrated that it no longer associated with these RAREs in crg mutants (Fig 9D). To determine if the epigenetic milieu of the ripply3 promoter was changed in crg mutants, we examined the transcriptional activation mark H3K27ac [86] and transcriptional repressive mark H3K27me3 [87] at these RAREs. ChIP-PCR revealed that the RAREs were now associated with transcriptional activation marks and lost transcriptional repression marks in crg mutants (Fig 9E and 9F). One possibility for this observation was that ectopic RA directly promoted transcriptional activation of ripply3 from these cis-regulatory elements. However, in luciferase assays, these elements were not able to activate expression following RA treatment (S12 Fig), instead supporting the idea that RARs may be required to repress ripply3 expression while some other factor(s) are necessary to activate ripply3 expression. Consistent with this hypothesis, we found that deleting regions containing either of the RAREs in crg mutants using CRISPR-Cas9 led to a loss of ectopic ripply3 expression at 36 hpf (Fig 9G and 9H), while deletion of both these sites dramatically reduced the ectopic ripply3 expression (Fig 9H). Injection of control gRNAs to a region ~100kb 3’ to the ripply3 promoter did not affect expression (Fig 9H). Altogether, these data suggest that HDAC1 promotes a transcriptionally repressive environment within the ripply3 promoter that prevents other factors from inducing ripply3 expression.

[Figure omitted. See PDF.]

Fig 9. Hdac1 promotes transcriptional repression at RAREs within the ripply3 promoter.

(A) Schematic of DR1 and DR4 RAREs within the ripply3 promoter. (B) EMSAs with RARs for the ripply3 DR1 and DR4 sites. The positive control is a DR5 site found in promoters of other direct targets [133]. (C) ChIP-qPCR for induced GFP-VP16-RARab at the DR1 and DR4 sites. (D-F) ChIP-qPCR for HDAC1, H3K27ac, H3K27me3 at the ripply3 DR1 and DR4 sites in WT sibling and crg mutants. Fold enrichment for C-F was normalized versus IgG pull-down. (G) Deletions of the ripply3 promoter DR1 and DR4 sites using multiplexed gRNAs and Cas9. Asterisks indicate deletions, which were confirmed with Sanger sequencing. (H) RT-qPCR for ripply3 expression in WT sibling and crg mutant embryos with ripply3 DR1 and DR4 promoter deletions at 36 hpf. The control gRNAs target a region ~100kb away from the ripply3 promoter. (I) Model depicting Hdac1 and RAR function regulating ripply3 in SHF development.

https://doi.org/10.1371/journal.pgen.1008165.g009

Discussion

HDACs control epigenetic modifications to histones that direct gene expression necessary for the proper development of many organs. Previous studies have not illuminated requirements for HDAC1 and/or HDAC2 in SHF development [29, 30]. In contrast to HDAC1 and HDAC2 in mice, deletion of HDAC3 specifically within SHF progenitors using an Isl1-Cre results in a spectrum of OFT defects, including double outlet right ventricle and semilunar valve malformations [88]. Despite zebrafish Hdac1 and murine HDAC3 both being Class I HDACs that promote SHF development, it is not clear they are functionally equivalent. OFT defects in HDAC3 SHF KO mice are in part due to increased TGF-β signaling [88, 89]. Additionally, HDAC3 is required for neural crest-derived smooth muscle that septates the distal OFT in mice [89]. In zebrafish, cardiac neural crest cells contribute to ~10% of the cardiomyocytes as well as smooth muscle within the OFT [90–92]. Although zebrafish hdac1 mutants have impaired neural crest development [93], we have not yet examined the requirement of Hdac1 in cardiac neural crest. Therefore, we cannot rule out that some of the cardiac and smooth muscle deficits we observe are due to requirements within neural crest-derived tissues. However, in contrast to what is observed with SHF-specific HDAC3 KOs in mice, our results suggest that in zebrafish Hdac1 loss impairs SHF progenitor proliferation. Interestingly, TGF-β signaling has been shown to promote proliferation of the SHF progenitors [13], which correlates with the loss of ltbp3 we observe at the arterial pole of hearts in crg mutants. Although global HDAC1-null mice are early embryonic lethal from decreased proliferation affecting many tissues [28] and endodermal progenitors exhibit reduced proliferation in zebrafish hdac1 mutants [94], loss of proliferation is not a universal consequence of HDAC1. For example, retina defects in zebrafish hdac1 mutants are due to increased Wnt signaling-dependent proliferation [51]. Therefore, our results showing Hdac1 promotes proliferation of SHF progenitors in zebrafish extends our understanding of the developmental contexts by which this conserved family of epigenetic regulators function in vertebrates.

RARs are one of the transcription factors that recruit HDACs to impart epigenetic modifications and control chromatin accessibility at specific loci [38]. Although proper RA signaling is critical during many stages of heart development, little progress has been made regarding the RA-dependent transcriptional mechanisms used to direct heart development. The canonical model for RAR-mediated transcription implies that RARs can act as transcriptional repressors in the absence of RA and transcriptional activators in the presence of RA [32]. However, recent studies suggest that the RAR transcriptional mechanisms are more versatile. Furthermore, while RARs can complex with HDACs through association with polycomb repressive complex 2 (PRC2) proteins [34, 95, 96], there is limited data suggesting that RARs function as transcriptional repressors in developmental contexts in the absence of ligand [97]. Comparing transcriptomic and ChIP-seq data in cultured cells has indicated that most RAREs are associated with ligand-induced transcriptionally activated targets [38, 98] and that RARs have more flexibility to bind RAREs with highly variable spacing (DR0 through DR8) [98–101]. Our study complements the previous in vitro analysis and provides an in vivo example that emphasizes the flexibility of RARs as we find they can bind atypical RAREs, in particular at a DR4.

RA-mediated induction of Ripply3 is likely conserved in vertebrates [61, 62]. Although we still need to decipher the precise transcriptional mechanism by which RA promotes ectopic ripply3 expression, our results suggest that in contrast to the canonical model of RAR-mediated transcription, RA may not activate ripply3 expression. Instead, our data provide evidence for a de-repressive transcriptional model whereby excess RA signaling or loss of HDAC1 at these RAREs relieves transcriptional repression of the promoter to allow currently unknown factor(s) to induce ectopic ripply3 (Fig 9I). In stem cells, unliganded RARs associated with DNA were found to interact with atypical DR0 and DR1 RAREs [98]. While the connection between RA signaling and HDACs has not been reported previously with respect to vertebrate SHF development, it is interesting to consider that both SHF-specific KOs of RARs leads to excess TGF-β that impinges on proper OFT development reminiscent of HDAC3 KOs [88, 102]. Conversely, our results suggest that loss of Hdac1 and excess RA-signaling may impair TGF-β signaling [103]. However, the transcriptional mechanisms by which RA signaling represses TGF-β signaling in mammalian SHF development have not been explored. Therefore, our study reveals insight into the convergence of RAR- and HDAC1-mediated transcriptional repression necessary for SHF development, which may have broad implications for understanding fundamental RA-mediated transcriptional mechanisms.

That ripply3 expression needs to be restricted to the pharyngeal endoderm for proper OFT development highlights the conserved, intimate relationship between the cardiac and pharyngeal progenitor fields in vertebrates [104–106]. Tbx1 is at the top of a conserved hierarchy of genes that regulates both SHF and pharyngeal development in vertebrates, in part through promoting proliferation and repressing RA signaling [73, 74, 107–110]. Loss of Tbx1 is associated with DiGeorge Syndrome in humans [111], which is characterized by OFT and craniofacial defects. In mice and Xenopus, Ripply3 is restricted to pharyngeal ectoderm and endoderm [67]. Previous work has shown transcriptionally repressive interactions between Ripply3 and Tbx1 are necessary for posterior pharyngeal development [67]. While Ripply3 KOs also have OFT defects [67], the mechanisms underlying these defects have not been elucidated. Although zebrafish ripply3 has conserved expression within the posterior pharyngeal endoderm, in contrast to mice, we were surprised to find that ripply3 is not overtly required for OFT development in zebrafish, or the development of other organs, as they are viable. Thus, the requirement for Ripply3 within the OFT could reflect a function that arose is mammals. Despite the overt lack of requirement in pharyngeal endoderm of zebrafish, our data investigating cardiac defects in crg mutants emphasize that it is critical to restrict ripply3 expression to the posterior endoderm as an expansion of ripply3 expression into adjacent mesodermal cells inhibits VC production. We propose that SHF defects in crg mutants are in part caused by ectopic Ripply3 forcing Tbx1 to function as a transcriptional repressor within these cardiac progenitor cells. As RIPPLY3 (DSCR6) is one of the genes in the supernumerary region of chromosome 21 [64], we also speculate its ectopic expression in this context could be one contributing factor to the high incidence of CHDs in children with Down syndrome [112]. Altogether, our data suggest epigenetic repression is necessary to compartmentalize ripply3 expression to the pharyngeal endoderm as its ectopic expression may impinge on Tbx1 function within SHF progenitors.

Overall, our study reveals insight into how the intersection of HDAC-dependent epigenetic regulation and transcriptional repression by RA signaling promote proper vertebrate OFT development. Because HDAC1 interacts with many other transcription factors and molecular complexes [113], we anticipate that it also regulates SHF development through RA-independent mechanisms. Similarly, we postulate that a combination of direct transcriptional activation and indirect repressive mechanisms, independent of HDAC1 function, contribute to RA-induced teratogenesis and OFT defects. Illuminating how these epigenetic and transcriptional mechanisms are integrated to control gene regulatory networks within SHF progenitors and adjacent organ fields will provide insight into the molecular etiology of developmental syndromes that include congenital heart defects in humans.

Materials and methods

Ethics statement

All zebrafish husbandry and experiments were performed as outlined in approved IACUC protocols at the Cincinnati Children’s Hospital Medical Center and Oregon Health and Science University.

Zebrafish husbandry, transgenic and mutant lines

Adult zebrafish (Danio rerio) were raised and maintained under standard laboratory conditions (Westerfield, 2000). Zebrafish transgenic lines used were: Tg(–5.1myl7:DsRed-NLS)f2 [48], Tg(hsp70l:GFP-VP16-RARab)ci1007 [85], Tg(myl7:NLS-KikGR)hsc6 [2], TgBAC(−36nkx2.5:ZsYellow)fb7 [1], Tg(myl7:Kaede)sd22 [4], and Tg(hs70l:GFP-Ripply3)ci1011. The Tg(hs70l:GFP-Ripply3) transgenic line used was created using standard Gateway cloning methods and Tol2 mediated transgenesis [114]. The GFP-ripply3 fusion construct was made using PCR. Gateway cloning was used to place the hsp70l 5’-entry clone [114], GFP-Ripply3 middle entry clone, and polyA 3’entry plasmids [114] into the pDEST-Tol2 P2a;α-cry:DsRed plasmid [115]. WT AB/TU embryos were injected with 75 pg of the hsp70l:GFP-ripply3 plasmid and 25 pg Tol2 mRNA. The F1 progeny of F0 founder fish found to have red eyes were raised. The F1 line selected produced offspring with ~50% containing the transgene, indicative of a single insertion, as well as GFP, and phenotypes equivalent to ripply3 mRNA injection following heat-shock by the end of gastrulation.

Zebrafish mutant lines used were ripply3ci1010 and crgnl18. Crg mutants were identified in an ongoing ENU screen. Crg mutants were genotyped using primers crg-HphI-F2 and crg-g-R1 (303 bp product) followed by digestion with the restriction enzyme HphI, which cuts the mutant allele. Primers sequences are listed in S1 Table.

CRISPR-Cas9 generated mutants and deletions

The ripply3ci1010mutant allele was created with CRISPR-Cas9 [71, 116, 117]. Guide RNAs (gRNAs) used were designed using ChopChop (http://chopchop.cbu.uib.no). The gRNA templates were generated using PCR similar to what has been described [71]. The gRNAs were generated using a MEGAshortscript T7 kit (Life Technologies; AM1354). Ripply3 gRNA (150pg) and Cas9 RNA (300ng) [118] were injected into embryos at the one-cell stage. Carriers of ripply3 mutations were identified using PCR on DNA from pooled progeny. Mutations in the F0 and F1 fish were selected for and confirmed through Sanger sequencing. PCR with the primers ripply3-t1-f1 and ripply3-t2-r1 were used for genotyping ripply3 mutants. The WT ripply3 product is 384 bp and the ripply3 mutant allele product is 323 bp. Primers sequences used to genotype the ripply3 mutant are listed in S1 Table.

For deletion of DR sites within the ripply3 promoter, gRNAs were injected (150 pg/gRNA for a pair; 75 pg/gRNA for the 2 pairs) along with EnGen Cas9 NLS protein (6μM) (New England Biolabs; M0646M). gRNAs and Cas9 protein were diluted to the indicated concentrations in a total volume of 5μl Ultrapure water (ThermoFisher; 1097715). Embryos were injected with 1nl of Cas9/gRNA solution. Co-injection of the gRNAs produced deletions of the predicted size (Fig 9G). Deletion of the target sites with the gRNAs was confirmed by Sanger sequencing. A gRNA pair that targets a genomic region ~100kb 3’ on Chr10 from the ripply3 promoter was used as a control for RT-qPCR analysis of ripply3 expression. Primers sequences used to confirm the DR-site deletions and for control gRNAs are listed in S1 Table.

MO injections

Zebrafish embryos were injected at the one-cell stage with MOs at the following doses: hdac1 MO 1 ng [52]; cyp26a1 MO1 2 ng, cyp26a1 MO2 1 ng, and cyp26c1 MO 6 ng (injected together) [119]; and tbx1 MO 1 ng [77]. To counteract non-specific MO-induced cell death, 1 ng p53 MO was used in all injections [120].

mRNA synthesis and injection

The coding sequence of ripply3 was cloned into pCS2p+DEST (a version of the reported pCS2-Dest vector [121] that we modified to have a Pst1 restriction site and corrected T7 primer sequence) using Gateway methods [121]. Ripply3 mRNA was synthesized using Sp6 Message Machine (Ambion). Embryos were injected with 200 pg ripply3 mRNA at the one-cell stage.

TSA treatments

TSA (Sigma; 8552) stock (10 mM DMSO) was diluted to a final concentration of 1 μM in embryo water beginning at 6 hpf. Embryos were incubated until 24 hpf in TSA. Samples were fixed and analyzed by in situ hybridization and immunostaining at respective stages.

Western blotting analysis of protein samples

Western blot assay was performed as described previously [122]. Primary antibodies used were: anti-zebrafish Hdac1 (GeneTex; 124499) and anti-alpha-Tubulin (Sigma; 6199). According to the manufacturer, the anti-zebrafish Hdac1 antibody was generated to amino acids 297–468 of zebrafish Hdac1. Secondary antibodies used were: IRDye 680LT Donkey anti-Rabbit IgG(H+L) (Licor; 92568023) and IRDye 800CW Donkey anti-Mouse IgG(H+L) (Licor; 92532212). Antibodies used are listed on S2 Table.

ISH and FISH

ISH and two-color FISH were performed as reported previously [123, 124]. Probes used were: hand2 (ZDB-GENE-000511-1), gata4 (ZDB-GENE-980526-476), ltbp3 (ZDB-GENE-060526-130), mef2cb (ZDB-GENE-040901-7), myl7 (formerly called cmlc2; ZDB-GENE-991019), nkx2.5 (ZDB-GENE-980526-321), and ripply3 (ZDB-GENE-060113-3). Plasmid used for ripply3 probe was a gift of Kazunori Okada and Shinji Takada. For standard ISH, embryos were imaged using a Zeiss M2BioV12 stereomicroscope. For FISH, embryos were imaged using a Nikon A1 inverted confocal microscope. Embryos were genotyped following imaging.

IHC and CM counting

IHC and counting of cardiomyocytes were performed as previously described [125]. Primary antibodies used were rabbit polyclonal anti-DsRed2 1:1000 (Clontech; 632496), mouse monoclonal anti-Sarcomeric myosin (MHC) 1:10 (MF20; University of Iowa Developmental Studies Hybridoma Bank), mouse monoclonal anti-Atrial myosin heavy chain (AMHC) 1:10 (S46; University of Iowa Developmental Studies Hybridoma Bank), rabbit polyclonal anti-Nkx2.5 1:250 (GeneTex; 128357), mouse monoclonal anti-PHH3 1:500 (Abcam, ab14955), anti-zebrafish Ventricular myosin heavy chain (Vmhc), and anti-zebrafish Elastin b (Elnb). Affinity purified rabbit antibodies to zebrafish Vmhc and Elnb were generated by YenZym (www.YenZym.com) to peptides KSRDVSSKKGHDQE (amino acids 1925–1938) and PGAGYQQQYPGFGGPGAGGPGS (amino acids 1958–1979) of the respective proteins. Secondary antibodies used were goat anti-chicken IgG-FITC (Southern Biotech; 6100–02), goat anti-mouse IgG1-TRITC (Southern Biotech; 1070–02), goat anti-mouse IgG1 FITC (Southern Biotech; 1070–02), goat anti-rabbit IgG-TRITC (Southern Biotech; 4050–03), goat anti-rabbit IgG-FITC (Southern Biotech; 4050–02), goat anti-mouse IgG1-DyLight 405 (BioLegend; Poly24091) and goat anti-mouse IgG2b-TRITC (Southern Biotech; 1090–03) were all used at 1:100. For counting of Nkx2.5+ cells and colocalization of Nkx2.5 and PHH3 embryos were imaged using a Nikon A1 confocal microscope. Counting was performed with ImageJ [126]. Colocalization was determined with the aid of Imaris (bitplane.com). All embryos were genotyped by tail clipping. Images were pseudo-colored using ImageJ [126].

DAF-2DA staining

DAF-2DA was used to stain the smooth muscle of OFT, similar to what has been described [49]. Embryos were transferred to embryo water with 20 γM DAF-2DA (EMD Millipore, 251505-M) at 80 hpf and incubated at 28.5°C in the dark until 96 hpf. Embryos were then harvested and immunostaining was performed as described above with the anti-zebrafish Vmhc antibody to mark the ventricle. The hearts of stained embryos were imaged using a Nikon A1 confocal microscope.

Photoconversion assays

Photoconversion assays for CM differentiation were performed similar to what has been previously described [24]. Embryos from adult crg carrier crosses containing the myl7:NLS-KikGr or myl7:Kaede transgenes were exposed to UV at 30 hpf for 30 min on a Zeiss M2BioV12 fluorescent stereomicroscope. Embryos were sorted into WT and crg mutant embryos at 48 hpf. Embryos were gently compressed with a footed coverslip and their hearts imaged using a Nikon A1 confocal microscope. ImageJ was used to mark and count the number of nuclei in green-only later differentiating CMs in myl7:NLS-KikGr transgenic hearts and the area of green-only later differentiating CMs in myl7:Kaede transgenic hearts. Images were pseudo-colored using ImageJ [126].

Blastula cell transplantation

To assess cellular autonomy, blastula cell transplantation was performed as previously described [85, 119]. Tg(myl7:EGFP) donor embryos were injected at the one-cell stage with Cascade blue-dextran (Invitrogen) alone or with the hdac1 MO. At the sphere stage, 15–20 donor cells were transplanted in the margin of WT (control) and/or hdac1-deficient host embryos. Host embryos were then grown to 48 hpf and scored for their frequency and contribution to atria and ventricles using a Zeiss M2BioV12 fluorescent stereomicroscope.

FACS

The preparation of samples from Tg(nkx2.5:ZsYellow) for FACS was performed as described previously [127]. Zombie Aqua (Biolegend) was used to label dead cells. FACS was performed in the CCHMC FACS Core using a BD FACSAria II at 488nm and 561nm. GFP+/Blue- cells were collected for RNA isolation.

RT-qPCR

cDNA was prepared from whole embryos as previously described [119, 128]. RT-qPCR using SYBR green PCR master mix (Applied Biosystems) was performed under standard PCR conditions in Bio-Rad CFX PCR machine. Relative expression levels of myl7, ltbp3, ripply3 and cdkn1a were standardized using β-actin and the 2−ΔΔCT Livak Method.

RNA-seq preparation and analysis

Embryos were lysed in Trizol then RNA was isolated using the PureLink RNA Micro Kit (Invitrogen). RNA was submitted to the CCMHC Sequencing core for library preparation. RNA from isolated cells was collected and amplified using a Single Cell RNA purification kit (Norgen Biotek; 51800). RNA was submitted to the CCHMC Gene Expression Core for library preparation. Samples were sequenced in the CCHMC Sequencing core using an Illumina High-Seq 2500 at a depth of >20M reads. Single-end stranded sequencing was performed on whole embryos. Paired-end sequencing was performed on samples from sorted cells. Sequences were aligned to the zebrafish genome (Zv9/danRer7) using Strand-NGS. Differential expression from single conditions of each of the samples was compared using DE-seq. Genes with >2 fold increased expression in Hdac- and Cyp26-deficient conditions relative to control conditions were identified based on manual annotation and clustering of genes with similar expression profiles using the Gene Expression tool kit in Strand NGS. P-values for individual gene expression were not considered for the analysis as they are not informative for statistical analysis of single conditions. The sequencing data have been deposited in GEO (accession # GSE126747).

EMSA

EMSAs were performed as described previously [129]. Oligonucleotides were designed containing the ripply3 DR1 site (AGGTCAGAGGTCA), the ripply3 DR4 (AGTTCCTCAGGGGTCA) site, and a previously reported Cyp26a1 DR5 site [130]. A complementary oligonucleotide was designed with a 5’ LI-COR IRDye 700 (IDT). Sequences for oligos are listed in S1 Table. The oligonucleotides were annealed and the ends filled with Klenow (New England Biolabs). Zebrafish myc-rarab was cloned into pCS2+MT. Zebrafish RXRba was cloned into pCS2p+. Proteins for EMSA were made using the TnT SP6 Quick Coupled Transcription/Translation System (Promega). Protein samples were gently mixed with LI-COR tagged probes and incubated at room temperature for 20 minutes. 4% polyacrylamide gels were run for 2 hours at 150 V. Gels were imaged using an Odyssey CLx LI-COR imager.

ChIP

ChIP was conducted essentially as previously described with 150 embryos per condition [131]. Antibodies used for IPs were: anti-GFP (Abcam; ab290), HDAC1 (Abcam; ab41407), H3K27ac (Abcam; ab4729), and H3K27me3 (Millipore; 7449). All IPs were performed using 1:100 dilutions of the antibodies. The IP’d DNA was analyzed with qPCR described above. Fold enrichment of ripply3-DR1 and ripply3-DR4 were standardized to that of negative control samples (IgG) (Southern Biotech; 617001). Primer sequences for IPs are listed in S1 Table.

Cell culture and dual luciferase assay

Fragments of the ripply3 promoter containing the DR1 (-3890 to -4759 upstream of the TSS) and DR4 (-2241 to -2968 upstream of the TSS) sites were inserted into the EcoRV site of pGL4.23 plasmids (Promega). The 12XRARE-tk-pGL3 vector and luciferase assays were reported and performed in HEK293 cells co-transfected with zebrafish RARab as described previously [132]. Primers for cloning the ripply3 fragments are listed in S1 Table.

Statistical analysis

All statistical analysis unless indicated was carried out using two-tailed Student's t-test with p<0.05 considered to be statistically significant.

Supporting information

S1 Fig. CM specification and early differentiation are overtly unaffected in crg mutants.

(A-D) ISH for CM specification markers gata4 and hand2 at the 8s stage. n = 5 WT and n = 5 crg mutants embryos examined. (E,F) ISH for the CM differentiation marker myl7 at the 20s stage. n = 10 WT and n = 5 crg mutants embryos examined. Views are dorsal with anterior up.

https://doi.org/10.1371/journal.pgen.1008165.s001

(TIF)

S2 Fig. The SHF marker mef2cb has increased expression adjacent to differentiated CMs at the arterial pole of the heart in crg mutants.

(A-B”) Two-color FISH for myl7 and mef2cb in WT sibling and crg mutant embryos at 30 hpf. Brackets in A and B indicate mef2cb at the arterial pole of in WT and crg mutant hearts, respectively. n = 5 WT and n = 5 crg mutants embryos examined. (C) RT-qPCR for the SHF marker mef2cb from embryos at 36 hpf.

https://doi.org/10.1371/journal.pgen.1008165.s002

(TIF)

S3 Fig. Later-differentiating VCs are reduced in crg mutants.

(A-B”) Images of hearts from photoconverted WT sibling and crg mutant myl7:Kaede embryos at 48 hpf following photoconversion at 36 hpf. The arterial poles (brackets) are to the right. (C) Quantification of the area of later-differentiating VCs (Yellow+/Purple-). (n = 11 for WT and crg mutants).

https://doi.org/10.1371/journal.pgen.1008165.s003

(TIF)

S4 Fig. Whole western blot for Hdac1 in hdac1 MO-depleted and crg embryos.

Both the hdac1 MO and crg mutants show a loss of the predicted WT Hdac1 proteins (black arrow). We do not observe the smaller protein with a 32 bp deletion predicted from the transcript analysis in RNA-seq, suggesting that this protein is not generated in the crg mutants. Since the antibody used for Western was generated to the C-terminal, it would not recognize the severely truncate protein, if that were made in crg mutants. However, a lower band (white arrow) recognized by the antibody, which is potentially a smaller Hdac1 isoform that is present in the WT/Ctrl samples, is also lost in the Hdac1-depleted and crg mutant embryos.

https://doi.org/10.1371/journal.pgen.1008165.s004

(TIF)

S5 Fig. Hdac1 depletion and treatment with TSA produce similar phenotypes as crg mutants.

(A,B) Control and Hdac1-depleted embryos at 48 hpf. Lateral views with anterior to the left. (C,D) Hearts from control and Hdac1-depleted myl7:NLS-DsRed2 embryos at 48 hpf. Frontal views. Purple alone indicates ventricle. Yellow indicates atrium. Arrows indicate arterial pole of the ventricle. (E) Quantification of CMs in the atria and ventricles of control and Hdac1-depleted myl7:NLS-DsRed2 embryos at 48 hpf (n = 19 for control and Hdac1-depleted embryos). (F,G) Control and TSA-treated embryos at 48 hpf. Lateral views with anterior to the left. (H,I) Hearts from control and TSA-treated myl7:NLS-DsRed2 embryos at 48 hpf. Frontal views. Purple alone indicates ventricle. Yellow indicates atrium. Arrows indicate arterial pole of the ventricle. (J) Quantification of CMs in the atria and ventricles of control and TSA-treated myl7:NLS-DsRed2 embryos at 48 hpf (n = 14 for both control and TSA-treated embryos).

https://doi.org/10.1371/journal.pgen.1008165.s005

(TIF)

S6 Fig. Differential expression analysis for whole embryos and sorted Nkx2.5:ZsYellow+ cells.

Heat-maps of genes found to have increased expression from RNA-seq of Hdac1- and Cyp26-depleted embryos at 48 hpf and sorted nkx2.5:ZsYellow+ cells at 33 hpf. Scale represents fold-change (log2) of normalized values after DE-seq analysis in Strand NGS. Genes represented showed similar trend of increased expression.

https://doi.org/10.1371/journal.pgen.1008165.s006

(TIF)

S7 Fig. Expression of nkx2.5 and ripply3 in WT and crg mutants.

(A-D”) Confocal images of two-color FISH for nkx2.5 and ripply3 in WT and crg mutant embryos at 30 and 36 hpf. Images are dorsal views with anterior up. Insets in A-D indicate lateral views of the confocal images. n = 10 WT and n = 9 crg mutants embryos for 30 hpf and n = 4 WT and n = 4 crg mutants embryos for 36 hpf examined. Scale bar is 100 μm.

https://doi.org/10.1371/journal.pgen.1008165.s007

(TIF)

S8 Fig. SHF marker gene expression in Hdac1-depleted nkx2.5:ZsYellow+ cells.

(A) RT-qPCR for ltbp3 from sorted nkx2.5+ cells at 28 hpf. (B) RT-qPCR for mef2cb from sorted nkx2.5+ cells at 28 hpf.

https://doi.org/10.1371/journal.pgen.1008165.s008

(TIF)

S9 Fig. Generation and validation of the heat-shock inducible GFP-ripply3 transgenic line.

(A) Schematic of the heat-shock inducible hsp70l:GFP-ripply3 transgene. (B-E) Heat-shock induction of GFP-ripply3 at 24 hpf. Control embryos are heat-shocked non-transgenic siblings. (F,H) Control and hsp70l:GFP-ripply3 embryos at 48 hpf following heat-shock at 10 hpf. Lateral views with anterior to the left. (G,I) Hearts from control and hsp70l:GFP-ripply3; myl7:NLS-DsRed2 embryos at 48 hpf following heat-shock at 10 hpf. Frontal views. Purple alone indicates ventricle. Yellow indicates atrium. Arrows indicate arterial pole of the ventricle. (J) Quantification of CMs from control and hsp70l:GFP-ripply3; myl7:NLS-DsRed2 embryos at 48 hpf following heat-shock at 10 hpf (n = 25 for control and GFP-ripply3+). (K,M) Control and ripply3 mRNA-injected embryos at 48 hpf. Lateral views with anterior to the left. (L,N) Hearts from control and ripply3 mRNA-injected myl7:NLS-DsRed2 embryos at 48 hpf. Frontal views. Purple alone indicates ventricle. Yellow indicates atrium. Arrows indicate arterial pole of the ventricle. (O) Quantification of CMs from control and ripply3 mRNA-injected myl7:NLS-DsRed2 embryos at 48 hpf (n = 21 for control and ripply3 mRNA-injected embryos).

https://doi.org/10.1371/journal.pgen.1008165.s009

(TIF)

S10 Fig. Generation of ripply3 mutants.

(A) Schematic of the ripply3 exons and the gRNA targeting exon of the ripply3 gene. (B) The ripply3 mutant allele used deletes 61bp including the start codon. (C,D) WT sibling and ripply3 mutants at 48 hpf. (E,F) Ripply3 transcripts are not detectable in ripply3 mutants at 30 hpf. Arrows indicate posterior pharyngeal region where ripply3 is expressed in WT embryos. Views are dorsal with anterior up. (G) RT-qPCR for ripply3 in ripply3 mutants at 36 hpf indicates the transcripts are essentially undetectable. Primers used do not bind within the deleted region. (H) Quantification of atrial CMs in crgwt+het; ripply3wt+het, crgwt+het; ripply3-/-, crg-/-; ripply3wt+het, and crg-/-; ripply3-/- embryos at 48 hpf (n = 37 for crgwt+het; ripply3wt+het, n = 10 for crgwt+het; ripply3-/-, n = 48 for crg-/-; ripply3wt+het, n = 18 for crg-/-; ripply3-/-).

https://doi.org/10.1371/journal.pgen.1008165.s010

(TIF)

S11 Fig. Modulation of ripply3 expression does not affect tbx1 or cyp26a1 expression.

(A-C) RT-qPCR for tbx1 and cyp26a1 from embryos at 24 and 36 hpf following GFP-Ripply3 induction at the 20s stage and ripply3 mutants.

https://doi.org/10.1371/journal.pgen.1008165.s011

(TIF)

S12 Fig. The DR sites within the ripply3 promoter do not respond to RA treatment.

Dual luciferase assays in HEK293 cells for the ripply3 DR1 and DR4 sites. pGL3-12XRARE-tk (positive control), pGL-4.23 (empty vector–negative control), pGL-4.23-ripply3-DR1, and pGL-4.23-ripply3-DR4.

https://doi.org/10.1371/journal.pgen.1008165.s012

(TIF)

S1 Table. Primers sequences used.

Blue sequences indicate the target sequence for the gRNAs.

https://doi.org/10.1371/journal.pgen.1008165.s013

(PDF)

S2 Table. Antibodies used.

https://doi.org/10.1371/journal.pgen.1008165.s014

(PDF)

Acknowledgments

We thank members of the Waxman lab for feedback regarding experiments and the manuscript.

Citation: Song YC, Dohn TE, Rydeen AB, Nechiporuk AV, Waxman JS (2019) HDAC1-mediated repression of the retinoic acid-responsive gene ripply3 promotes second heart field development. PLoS Genet 15(5): e1008165. https://doi.org/10.1371/journal.pgen.1008165

References

1. Zhou Y, Cashman TJ, Nevis KR, Obregon P, Carney SA, Liu Y, et al. Latent TGF-beta binding protein 3 identifies a second heart field in zebrafish. Nature. 2011;474(7353):645–8. Epub 2011/05/31. pmid:21623370.

2. Lazic S, Scott IC. Mef2cb regulates late myocardial cell addition from a second heart field-like population of progenitors in zebrafish. Dev Biol. 2011;354(1):123–33. Epub 2011/04/07. pmid:21466801.

3. Hami D, Grimes AC, Tsai HJ, Kirby ML. Zebrafish cardiac development requires a conserved secondary heart field. Development. 2011;138(11):2389–98. Epub 2011/05/12. pmid:21558385; PubMed Central PMCID: PMC3091499.

4. de Pater E, Clijsters L, Marques SR, Lin YF, Garavito-Aguilar ZV, Yelon D, et al. Distinct phases of cardiomyocyte differentiation regulate growth of the zebrafish heart. Development. 2009;136(10):1633–41. Epub 2009/04/28. 136/10/1633 [pii] pmid:19395641; PubMed Central PMCID: PMC2673760.

5. Sawada H, Rateri DL, Moorleghen JJ, Majesky MW, Daugherty A. Smooth Muscle Cells Derived From Second Heart Field and Cardiac Neural Crest Reside in Spatially Distinct Domains in the Media of the Ascending Aorta-Brief Report. Arterioscler Thromb Vasc Biol. 2017;37(9):1722–6. Epub 2017/06/29. pmid:28663257; PubMed Central PMCID: PMCPMC5570666.

6. Guner-Ataman B, Paffett-Lugassy N, Adams MS, Nevis KR, Jahangiri L, Obregon P, et al. Zebrafish second heart field development relies on progenitor specification in anterior lateral plate mesoderm and nkx2.5 function. Development. 2013;140(6):1353–63. pmid:23444361; PubMed Central PMCID: PMCPMC3585667.

7. Felker A, Prummel KD, Merks AM, Mickoleit M, Brombacher EC, Huisken J, et al. Continuous addition of progenitors forms the cardiac ventricle in zebrafish. Nat Commun. 2018;9(1):2001. Epub 2018/05/21. pmid:29784942; PubMed Central PMCID: PMCPMC5962599.

8. Witzel HR, Cheedipudi S, Gao R, Stainier DY, Dobreva GD. Isl2b regulates anterior second heart field development in zebrafish. Sci Rep. 2017;7:41043. Epub 2017/01/20. pmid:28106108; PubMed Central PMCID: PMCPMC5247716.

9. Paffett-Lugassy N, Novikov N, Jeffrey S, Abrial M, Guner-Ataman B, Sakthivel S, et al. Unique developmental trajectories and genetic regulation of ventricular and outflow tract progenitors in the zebrafish second heart field. Development. 2017;144(24):4616–24. Epub 2017/10/23. pmid:29061637; PubMed Central PMCID: PMCPMC5769620.

10. Buckingham M, Meilhac S, Zaffran S. Building the mammalian heart from two sources of myocardial cells. Nat Rev Genet. 2005;6(11):826–35. pmid:16304598.

11. Grimes AC, Durán AC, Sans-Coma V, Hami D, Santoro MM, Torres M. Phylogeny informs ontogeny: a proposed common theme in the arterial pole of the vertebrate heart. Evol Dev. 2010;12(6):552–67. pmid:21040422; PubMed Central PMCID: PMCPMC4180519.

12. Galli D, Domínguez JN, Zaffran S, Munk A, Brown NA, Buckingham ME. Atrial myocardium derives from the posterior region of the second heart field, which acquires left-right identity as Pitx2c is expressed. Development. 2008;135(6):1157–67. Epub 2008/02/13. pmid:18272591.

13. Zhou Y, Cashman TJ, Nevis KR, Obregon P, Carney SA, Liu Y, et al. Latent TGF-β binding protein 3 identifies a second heart field in zebrafish. Nature. 2011;474(7353):645–8. Epub 2011/05/29. pmid:21623370; PubMed Central PMCID: PMCPMC3319150.

14. Keegan BR, Meyer D, Yelon D. Organization of cardiac chamber progenitors in the zebrafish blastula. Development. 2004;131(13):3081–91. Epub 2004/06/02. pmid:15175246.

15. Yang L, Cai CL, Lin L, Qyang Y, Chung C, Monteiro RM, et al. Isl1Cre reveals a common Bmp pathway in heart and limb development. Development. 2006;133(8):1575–85. pmid:16556916.

16. Cohen ED, Wang Z, Lepore JJ, Lu MM, Taketo MM, Epstein DJ, et al. Wnt/beta-catenin signaling promotes expansion of Isl-1-positive cardiac progenitor cells through regulation of FGF signaling. J Clin Invest. 2007;117(7):1794–804. Epub 2007/07/04. pmid:17607356; PubMed Central PMCID: PMC1891000.

17. Qyang Y, Martin-Puig S, Chiravuri M, Chen S, Xu H, Bu L, et al. The renewal and differentiation of Isl1+ cardiovascular progenitors are controlled by a Wnt/beta-catenin pathway. Cell Stem Cell. 2007;1(2):165–79. Epub 2008/03/29. S1934-5909(07)00059-8 [pii] pmid:18371348.

18. Dyer LA, Kirby ML. Sonic hedgehog maintains proliferation in secondary heart field progenitors and is required for normal arterial pole formation. Dev Biol. 2009;330(2):305–17. Epub 2009/04/08. pmid:19361493; PubMed Central PMCID: PMCPMC2810612.

19. Washington Smoak I, Byrd NA, Abu-Issa R, Goddeeris MM, Anderson R, Morris J, et al. Sonic hedgehog is required for cardiac outflow tract and neural crest cell development. Dev Biol. 2005;283(2):357–72. pmid:15936751.

20. Zhang J, Lin Y, Zhang Y, Lan Y, Lin C, Moon AM, et al. Frs2alpha-deficiency in cardiac progenitors disrupts a subset of FGF signals required for outflow tract morphogenesis. Development. 2008;135(21):3611–22. Epub 2008/10/02. pmid:18832393; PubMed Central PMCID: PMCPMC3262991.

21. Park EJ, Watanabe Y, Smyth G, Miyagawa-Tomita S, Meyers E, Klingensmith J, et al. An FGF autocrine loop initiated in second heart field mesoderm regulates morphogenesis at the arterial pole of the heart. Development. 2008;135(21):3599–610. Epub 2008/10/04. pmid:18832392; PubMed Central PMCID: PMC2771204.

22. Ryckebusch L, Wang Z, Bertrand N, Lin SC, Chi X, Schwartz R, et al. Retinoic acid deficiency alters second heart field formation. Proc Natl Acad Sci U S A. 2008;105(8):2913–8. 0712344105 [pii] pmid:18287057.

23. Sirbu IO, Zhao X, Duester G. Retinoic acid controls heart anteroposterior patterning by down-regulating Isl1 through the Fgf8 pathway. Dev Dyn. 2008;237(6):1627–35. pmid:18498088.

24. Rydeen AB, Waxman JS. Cyp26 Enzymes Facilitate Second Heart Field Progenitor Addition and Maintenance of Ventricular Integrity. PLoS Biol. 2016;14(11):e2000504. pmid:27893754; PubMed Central PMCID: PMCPMC5125711.

25. Strahl BD, Allis CD. The language of covalent histone modifications. Nature. 2000;403(6765):41–5. pmid:10638745.

26. Chang CP, Bruneau BG. Epigenetics and cardiovascular development. Annu Rev Physiol. 2012;74:41–68. Epub 2011/10/24. pmid:22035349.

27. Haberland M, Montgomery RL, Olson EN. The many roles of histone deacetylases in development and physiology: implications for disease and therapy. Nat Rev Genet. 2009;10(1):32–42. Epub 2008/12/10. pmid:19065135; PubMed Central PMCID: PMCPMC3215088.

28. Lagger G, O'Carroll D, Rembold M, Khier H, Tischler J, Weitzer G, et al. Essential function of histone deacetylase 1 in proliferation control and CDK inhibitor repression. EMBO J. 2002;21(11):2672–81. pmid:12032080; PubMed Central PMCID: PMCPMC126040.

29. Montgomery RL, Davis CA, Potthoff MJ, Haberland M, Fielitz J, Qi X, et al. Histone deacetylases 1 and 2 redundantly regulate cardiac morphogenesis, growth, and contractility. Genes Dev. 2007;21(14):1790–802. pmid:17639084; PubMed Central PMCID: PMCPMC1920173.

30. Trivedi CM, Zhu W, Wang Q, Jia C, Kee HJ, Li L, et al. Hopx and Hdac2 interact to modulate Gata4 acetylation and embryonic cardiac myocyte proliferation. Dev Cell. 2010;19(3):450–9. pmid:20833366; PubMed Central PMCID: PMCPMC2947937.

31. Nambiar RM, Henion PD. Sequential antagonism of early and late Wnt-signaling by zebrafish colgate promotes dorsal and anterior fates. Dev Biol. 2004;267(1):165–80. pmid:14975724.

32. Cunningham TJ, Duester G. Mechanisms of retinoic acid signalling and its roles in organ and limb development. Nat Rev Mol Cell Biol. 2015;16(2):110–23. Epub 2015/01/07. pmid:25560970; PubMed Central PMCID: PMCPMC4636111.

33. Heinzel T, Lavinsky RM, Mullen TM, Söderstrom M, Laherty CD, Torchia J, et al. A complex containing N-CoR, mSin3 and histone deacetylase mediates transcriptional repression. Nature. 1997;387(6628):43–8. pmid:9139820.

34. Nagy L, Kao HY, Chakravarti D, Lin RJ, Hassig CA, Ayer DE, et al. Nuclear receptor repression mediated by a complex containing SMRT, mSin3A, and histone deacetylase. Cell. 1997;89(3):373–80. pmid:9150137.

35. Lin RJ, Nagy L, Inoue S, Shao W, Miller WH, Evans RM. Role of the histone deacetylase complex in acute promyelocytic leukaemia. Nature. 1998;391(6669):811–4. pmid:9486654.

36. Ferrara FF, Fazi F, Bianchini A, Padula F, Gelmetti V, Minucci S, et al. Histone deacetylase-targeted treatment restores retinoic acid signaling and differentiation in acute myeloid leukemia. Cancer Res. 2001;61(1):2–7. pmid:11196162.

37. Alland L, Muhle R, Hou H, Potes J, Chin L, Schreiber-Agus N, et al. Role for N-CoR and histone deacetylase in Sin3-mediated transcriptional repression. Nature. 1997;387(6628):49–55. pmid:9139821.

38. Urvalek AM, Gudas LJ. Retinoic acid and histone deacetylases regulate epigenetic changes in embryonic stem cells. J Biol Chem. 2014;289(28):19519–30. pmid:24821725; PubMed Central PMCID: PMCPMC4094062.

39. Vilhais-Neto GC, Fournier M, Plassat JL, Sardiu ME, Saraf A, Garnier JM, et al. The WHHERE coactivator complex is required for retinoic acid-dependent regulation of embryonic symmetry. Nat Commun. 2017;8(1):728. Epub 2017/09/28. pmid:28959017; PubMed Central PMCID: PMCPMC5620087.

40. Bajpe PK, Heynen GJ, Mittempergher L, Grernrum W, de Rink IA, Nijkamp W, et al. The corepressor CTBP2 is a coactivator of retinoic acid receptor/retinoid X receptor in retinoic acid signaling. Mol Cell Biol. 2013;33(16):3343–53. Epub 2013/06/17. pmid:23775127; PubMed Central PMCID: PMCPMC3753894.

41. Klein ES, Wang JW, Khalifa B, Gavigan SA, Chandraratna RA. Recruitment of nuclear receptor corepressor and coactivator to the retinoic acid receptor by retinoid ligands. Influence of DNA-heterodimer interactions. J Biol Chem. 2000;275(25):19401–8. pmid:10777502.

42. Lee LM, Leung CY, Tang WW, Choi HL, Leung YC, McCaffery PJ, et al. A paradoxical teratogenic mechanism for retinoic acid. Proc Natl Acad Sci U S A. 2012;109(34):13668–73. Epub 2012/08/06. pmid:22869719; PubMed Central PMCID: PMCPMC3427051.

43. Pan J, Baker KM. Retinoic acid and the heart. Vitam Horm. 2007;75:257–83. pmid:17368319.

44. Lammer EJ, Chen DT, Hoar RM, Agnish ND, Benke PJ, Braun JT, et al. Retinoic acid embryopathy. N Engl J Med. 1985;313(14):837–41. pmid:3162101.

45. Sakabe M, Kokubo H, Nakajima Y, Saga Y. Ectopic retinoic acid signaling affects outflow tract cushion development through suppression of the myocardial Tbx2-Tgfβ2 pathway. Development. 2012;139(2):385–95. pmid:22186728.

46. Kumar S, Duester G. Retinoic acid controls body axis extension by directly repressing Fgf8 transcription. Development. 2014;141(15):2972–7. Epub 2014/07/24. pmid:25053430; PubMed Central PMCID: PMCPMC4197666.

47. Patton EE, Zon LI. The art and design of genetic screens: zebrafish. Nat Rev Genet. 2001;2(12):956–66. pmid:11733748.

48. Mably JD, Mohideen MA, Burns CG, Chen JN, Fishman MC. heart of glass regulates the concentric growth of the heart in zebrafish. Curr Biol. 2003;13(24):2138–47. Epub 2003/12/19. pmid:14680629.

49. Grimes AC, Stadt HA, Shepherd IT, Kirby ML. Solving an enigma: arterial pole development in the zebrafish heart. Dev Biol. 2006;290(2):265–76. Epub 2006/01/10. pmid:16405941.

50. Noël ES, Casal-Sueiro A, Busch-Nentwich E, Verkade H, Dong PD, Stemple DL, et al. Organ-specific requirements for Hdac1 in liver and pancreas formation. Dev Biol. 2008;322(2):237–50. Epub 2008/07/21. pmid:18687323; PubMed Central PMCID: PMCPMC3710974.

51. Stadler JA, Shkumatava A, Norton WH, Rau MJ, Geisler R, Fischer S, et al. Histone deacetylase 1 is required for cell cycle exit and differentiation in the zebrafish retina. Dev Dyn. 2005;233(3):883–9. pmid:15895391.

52. Yamaguchi M, Tonou-Fujimori N, Komori A, Maeda R, Nojima Y, Li H, et al. Histone deacetylase 1 regulates retinal neurogenesis in zebrafish by suppressing Wnt and Notch signaling pathways. Development. 2005;132(13):3027–43. pmid:15944187.

53. Yamaguchi T, Cubizolles F, Zhang Y, Reichert N, Kohler H, Seiser C, et al. Histone deacetylases 1 and 2 act in concert to promote the G1-to-S progression. Genes Dev. 2010;24(5):455–69. pmid:20194438; PubMed Central PMCID: PMCPMC2827841.

54. Senese S, Zaragoza K, Minardi S, Muradore I, Ronzoni S, Passafaro A, et al. Role for histone deacetylase 1 in human tumor cell proliferation. Mol Cell Biol. 2007;27(13):4784–95. Epub 2007/04/30. pmid:17470557; PubMed Central PMCID: PMCPMC1951481.

55. Tao Y, Ruan H, Guo X, Li L, Shen W. HDAC1 regulates the proliferation of radial glial cells in the developing Xenopus tectum. PLoS One. 2015;10(3):e0120118. Epub 2015/03/19. pmid:25789466; PubMed Central PMCID: PMCPMC4366096.

56. Dutto I, Tillhon M, Prosperi E. Assessing Cell Cycle Independent Function of the CDK Inhibitor p21(CDKN1A) in DNA Repair. Methods Mol Biol. 2016;1336:123–39. pmid:26231713.

57. Chen Z, Huang W, Dahme T, Rottbauer W, Ackerman MJ, Xu X. Depletion of zebrafish essential and regulatory myosin light chains reduces cardiac function through distinct mechanisms. Cardiovasc Res. 2008;79(1):97–108. Epub 2008/03/14. pmid:18343897; PubMed Central PMCID: PMCPMC2724891.

58. Guner-Ataman B, González-Rosa JM, Shah HN, Butty VL, Jeffrey S, Abrial M, et al. Failed Progenitor Specification Underlies the Cardiopharyngeal Phenotypes in a Zebrafish Model of 22q11.2 Deletion Syndrome. Cell Rep. 2018;24(5):1342–54.e5. pmid:30067987; PubMed Central PMCID: PMCPMC6261257.

59. McGrane MM. Vitamin A regulation of gene expression: molecular mechanism of a prototype gene. J Nutr Biochem. 2007;18(8):497–508. Epub 2007/02/22. pmid:17320364.

60. Bastien J, Rochette-Egly C. Nuclear retinoid receptors and the transcription of retinoid-target genes. Gene. 2004;328:1–16. pmid:15019979.

61. Li HY, Grifone R, Saquet A, Carron C, Shi DL. The Xenopus homologue of Down syndrome critical region protein 6 drives dorsoanterior gene expression and embryonic axis formation by antagonising polycomb group proteins. Development. 2013;140(24):4903–13. pmid:24301465.

62. Janesick A, Shiotsugu J, Taketani M, Blumberg B. RIPPLY3 is a retinoic acid-inducible repressor required for setting the borders of the pre-placodal ectoderm. Development. 2012;139(6):1213–24. pmid:22354841; PubMed Central PMCID: PMCPMC3283127.

63. Delabar JM, Theophile D, Rahmani Z, Chettouh Z, Blouin JL, Prieur M, et al. Molecular mapping of twenty-four features of Down syndrome on chromosome 21. Eur J Hum Genet. 1993;1(2):114–24. pmid:8055322.

64. Shibuya K, Kudoh J, Minoshima S, Kawasaki K, Asakawa S, Shimizu N. Isolation of two novel genes, DSCR5 and DSCR6, from Down syndrome critical region on human chromosome 21q22.2. Biochem Biophys Res Commun. 2000;271(3):693–8. pmid:10814524.

65. Okubo T, Takada S. Pharyngeal arch deficiencies affect taste bud development in the circumvallate papilla with aberrant glossopharyngeal nerve formation. Dev Dyn. 2015;244(7):874–87. Epub 2015/05/23. pmid:25997579.

66. Osipovich AB, Long Q, Manduchi E, Gangula R, Hipkens SB, Schneider J, et al. Insm1 promotes endocrine cell differentiation by modulating the expression of a network of genes that includes Neurog3 and Ripply3. Development. 2014;141(15):2939–49. pmid:25053427; PubMed Central PMCID: PMCPMC4197673.

67. Okubo T, Kawamura A, Takahashi J, Yagi H, Morishima M, Matsuoka R, et al. Ripply3, a Tbx1 repressor, is required for development of the pharyngeal apparatus and its derivatives in mice. Development. 2011;138(2):339–48. pmid:21177346.

68. Hitachi K, Danno H, Tazumi S, Aihara Y, Uchiyama H, Okabayashi K, et al. The Xenopus Bowline/Ripply family proteins negatively regulate the transcriptional activity of T-box transcription factors. Int J Dev Biol. 2009;53(4):631–9. pmid:19247927.

69. Waxman JS, Yelon D. Increased Hox activity mimics the teratogenic effects of excess retinoic acid signaling. Dev Dyn. 2009;238(5):1207–13. pmid:19384962; PubMed Central PMCID: PMCPMC2739864.

70. Liu J, Stainier DY. Zebrafish in the study of early cardiac development. Circ Res. 2012;110(6):870–4. pmid:22427324; PubMed Central PMCID: PMCPMC3329892.

71. Talbot JC, Amacher SL. A streamlined CRISPR pipeline to reliably generate zebrafish frameshifting alleles. Zebrafish. 2014;11(6):583–5. pmid:25470533; PubMed Central PMCID: PMCPMC4248249.

72. Kawamura A, Koshida S, Takada S. Activator-to-repressor conversion of T-box transcription factors by the Ripply family of Groucho/TLE-associated mediators. Mol Cell Biol. 2008;28(10):3236–44. Epub 2008/03/10. pmid:18332117; PubMed Central PMCID: PMCPMC2423164.

73. Nevis K, Obregon P, Walsh C, Guner-Ataman B, Burns CG, Burns CE. Tbx1 is required for second heart field proliferation in zebrafish. Dev Dyn. 2013;242(5):550–9. pmid:23335360; PubMed Central PMCID: PMCPMC3676967.

74. Rana MS, Théveniau-Ruissy M, De Bono C, Mesbah K, Francou A, Rammah M, et al. Tbx1 coordinates addition of posterior second heart field progenitor cells to the arterial and venous poles of the heart. Circ Res. 2014;115(9):790–9. Epub 2014/09/04. pmid:25190705.

75. Chen L, Fulcoli FG, Tang S, Baldini A. Tbx1 regulates proliferation and differentiation of multipotent heart progenitors. Circ Res. 2009;105(9):842–51. Epub 2009/09/10. pmid:19745164; PubMed Central PMCID: PMCPMC2796444.

76. Chen L, Fulcoli FG, Ferrentino R, Martucciello S, Illingworth EA, Baldini A. Transcriptional control in cardiac progenitors: Tbx1 interacts with the BAF chromatin remodeling complex and regulates Wnt5a. PLoS Genet. 2012;8(3):e1002571. Epub 2012/03/15. pmid:22438823; PubMed Central PMCID: PMCPMC3305383.

77. Lin CY, Chen WT, Lee HC, Yang PH, Yang HJ, Tsai HJ. The transcription factor Six1a plays an essential role in the craniofacial myogenesis of zebrafish. Dev Biol. 2009;331(2):152–66. Epub 2009/05/05. pmid:19409884.

78. Niederreither K, Dollé P. Retinoic acid in development: towards an integrated view. Nat Rev Genet. 2008;9(7):541–53. Epub 2008/06/10. pmid:18542081.

79. Duester G. Retinoic acid synthesis and signaling during early organogenesis. Cell. 2008;134(6):921–31. pmid:18805086; PubMed Central PMCID: PMCPMC2632951.

80. de la Calle Mustienes E, Gómez-Skarmeta JL, Bogdanović O. Genome-wide epigenetic cross-talk between DNA methylation and H3K27me3 in zebrafish embryos. Genom Data. 2015;6:7–9. Epub 2015/07/22. pmid:26697317; PubMed Central PMCID: PMCPMC4664660.

81. Kashyap V, Gudas LJ. Epigenetic regulatory mechanisms distinguish retinoic acid-mediated transcriptional responses in stem cells and fibroblasts. J Biol Chem. 2010;285(19):14534–48. Epub 2010/03/15. pmid:20231276; PubMed Central PMCID: PMCPMC2863236.

82. Calo E, Wysocka J. Modification of enhancer chromatin: what, how, and why? Mol Cell. 2013;49(5):825–37. pmid:23473601; PubMed Central PMCID: PMCPMC3857148.

83. Predki PF, Zamble D, Sarkar B, Giguere V. Ordered binding of retinoic acid and retinoid-X receptors to asymmetric response elements involves determinants adjacent to the DNA-binding domain. Mol Endocrinol. 1994;8(1):31–9. Epub 1994/01/01. pmid:8152429.

84. Balmer JE, Blomhoff R. Gene expression regulation by retinoic acid. J Lipid Res. 2002;43(11):1773–808. pmid:12401878.

85. Rydeen A, Voisin N, D'Aniello E, Ravisankar P, Devignes CS, Waxman JS. Excessive feedback of Cyp26a1 promotes cell non-autonomous loss of retinoic acid signaling. Dev Biol. 2015;405(1):47–55. Epub 2015/06/24. pmid:26116175; PubMed Central PMCID: PMCPMC4529768.

86. Creyghton MP, Cheng AW, Welstead GG, Kooistra T, Carey BW, Steine EJ, et al. Histone H3K27ac separates active from poised enhancers and predicts developmental state. Proc Natl Acad Sci U S A. 2010;107(50):21931–6. Epub 2010/11/24. pmid:21106759; PubMed Central PMCID: PMCPMC3003124.

87. Kuzmichev A, Nishioka K, Erdjument-Bromage H, Tempst P, Reinberg D. Histone methyltransferase activity associated with a human multiprotein complex containing the Enhancer of Zeste protein. Genes Dev. 2002;16(22):2893–905. pmid:12435631; PubMed Central PMCID: PMCPMC187479.

88. Lewandowski SL, Janardhan HP, Trivedi CM. Histone Deacetylase 3 Coordinates Deacetylase-independent Epigenetic Silencing of Transforming Growth Factor-beta1 (TGF-beta1) to Orchestrate Second Heart Field Development. J Biol Chem. 2015;290(45):27067–89. Epub 2015/10/01. pmid:26420484; PubMed Central PMCID: PMCPMC4646383.

89. Singh N, Trivedi CM, Lu M, Mullican SE, Lazar MA, Epstein JA. Histone deacetylase 3 regulates smooth muscle differentiation in neural crest cells and development of the cardiac outflow tract. Circ Res. 2011;109(11):1240–9. Epub 2011/09/29. pmid:21959220; PubMed Central PMCID: PMCPMC3225257.

90. Cavanaugh AM, Huang J, Chen JN. Two developmentally distinct populations of neural crest cells contribute to the zebrafish heart. Dev Biol. 2015;404(2):103–12. Epub 2015/06/19. pmid:26086691; PubMed Central PMCID: PMCPMC4515179.

91. Li YX, Zdanowicz M, Young L, Kumiski D, Leatherbury L, Kirby ML. Cardiac neural crest in zebrafish embryos contributes to myocardial cell lineage and early heart function. Dev Dyn. 2003;226(3):540–50. pmid:12619138.

92. Sato M, Yost HJ. Cardiac neural crest contributes to cardiomyogenesis in zebrafish. Dev Biol. 2003;257(1):127–39. pmid:12710962.

93. Ignatius MS, Unal Eroglu A, Malireddy S, Gallagher G, Nambiar RM, Henion PD. Distinct functional and temporal requirements for zebrafish Hdac1 during neural crest-derived craniofacial and peripheral neuron development. PLoS One. 2013;8(5):e63218. Epub 2013/05/07. pmid:23667588; PubMed Central PMCID: PMCPMC3646935.

94. Wang Y, Tian Y, Morley MP, Lu MM, Demayo FJ, Olson EN, et al. Development and regeneration of Sox2+ endoderm progenitors are regulated by a Hdac1/2-Bmp4/Rb1 regulatory pathway. Dev Cell. 2013;24(4):345–58. pmid:23449471; PubMed Central PMCID: PMCPMC3586315.

95. Jepsen K, Hermanson O, Onami TM, Gleiberman AS, Lunyak V, McEvilly RJ, et al. Combinatorial roles of the nuclear receptor corepressor in transcription and development. Cell. 2000;102(6):753–63. pmid:11030619.

96. Gillespie RF, Gudas LJ. Retinoid regulated association of transcriptional co-regulators and the polycomb group protein SUZ12 with the retinoic acid response elements of Hoxa1, RARbeta(2), and Cyp26A1 in F9 embryonal carcinoma cells. J Mol Biol. 2007;372(2):298–316. Epub 2007/07/03. pmid:17663992; PubMed Central PMCID: PMCPMC2972191.

97. Koide T, Downes M, Chandraratna RA, Blumberg B, Umesono K. Active repression of RAR signaling is required for head formation. Genes Dev. 2001;15(16):2111–21. pmid:11511542; PubMed Central PMCID: PMCPMC312762.

98. Mahony S, Mazzoni EO, McCuine S, Young RA, Wichterle H, Gifford DK. Ligand-dependent dynamics of retinoic acid receptor binding during early neurogenesis. Genome Biol. 2011;12(1):R2. Epub 2011/01/13. pmid:21232103; PubMed Central PMCID: PMCPMC3091300.

99. Delacroix L, Moutier E, Altobelli G, Legras S, Poch O, Choukrallah MA, et al. Cell-specific interaction of retinoic acid receptors with target genes in mouse embryonic fibroblasts and embryonic stem cells. Mol Cell Biol. 2010;30(1):231–44. pmid:19884340; PubMed Central PMCID: PMCPMC2798310.

100. Moutier E, Ye T, Choukrallah MA, Urban S, Osz J, Chatagnon A, et al. Retinoic acid receptors recognize the mouse genome through binding elements with diverse spacing and topology. J Biol Chem. 2012;287(31):26328–41. Epub 2012/06/01. pmid:22661711; PubMed Central PMCID: PMCPMC3406717.

101. Chatagnon A, Veber P, Morin V, Bedo J, Triqueneaux G, Sémon M, et al. RAR/RXR binding dynamics distinguish pluripotency from differentiation associated cis-regulatory elements. Nucleic Acids Res. 2015;43(10):4833–54. Epub 2015/04/20. pmid:25897113; PubMed Central PMCID: PMCPMC4446430.

102. Lewandowski SL, Janardhan HP, Smee KM, Bachman M, Sun Z, Lazar MA, et al. Histone deacetylase 3 modulates Tbx5 activity to regulate early cardiogenesis. Hum Mol Genet. 2014;23(14):3801–9. Epub 2014/02/23. pmid:24565863; PubMed Central PMCID: PMCPMC4065153.

103. Li P, Pashmforoush M, Sucov HM. Retinoic acid regulates differentiation of the secondary heart field and TGFbeta-mediated outflow tract septation. Dev Cell. 2010;18(3):480–5. pmid:20230754; PubMed Central PMCID: PMCPMC2841063.

104. Mandal A, Holowiecki A, Song YC, Waxman JS. Wnt signaling balances specification of the cardiac and pharyngeal muscle fields. Mech Dev. 2017;143:32–41. Epub 2017/01/10. pmid:28087459; PubMed Central PMCID: PMCPMC5344754.

105. Wang W, Razy-Krajka F, Siu E, Ketcham A, Christiaen L. NK4 antagonizes Tbx1/10 to promote cardiac versus pharyngeal muscle fate in the ascidian second heart field. PLoS Biol. 2013;11(12):e1001725. Epub 2013/12/03. pmid:24311985; PubMed Central PMCID: PMCPMC3849182.

106. Stolfi A, Gainous TB, Young JJ, Mori A, Levine M, Christiaen L. Early chordate origins of the vertebrate second heart field. Science. 2010;329(5991):565–8. pmid:20671188; PubMed Central PMCID: PMCPMC4970750.

107. Guo C, Sun Y, Zhou B, Adam RM, Li X, Pu WT, et al. A Tbx1-Six1/Eya1-Fgf8 genetic pathway controls mammalian cardiovascular and craniofacial morphogenesis. J Clin Invest. 2011;121(4):1585–95. pmid:21364285; PubMed Central PMCID: PMCPMC3069777.

108. Roberts C, Ivins SM, James CT, Scambler PJ. Retinoic acid down-regulates Tbx1 expression in vivo and in vitro. Dev Dyn. 2005;232(4):928–38. pmid:15736167.

109. Roberts C, Ivins S, Cook AC, Baldini A, Scambler PJ. Cyp26 genes a1, b1 and c1 are down-regulated in Tbx1 null mice and inhibition of Cyp26 enzyme function produces a phenocopy of DiGeorge Syndrome in the chick. Hum Mol Genet. 2006;15(23):3394–410. Epub 2006/10/17. pmid:17047027.

110. Okano J, Sakai Y, Shiota K. Retinoic acid down-regulates Tbx1 expression and induces abnormal differentiation of tongue muscles in fetal mice. Dev Dyn. 2008;237(10):3059–70. pmid:18816858.

111. Papangeli I, Scambler P. The 22q11 deletion: DiGeorge and velocardiofacial syndromes and the role of TBX1. Wiley Interdiscip Rev Dev Biol. 2013;2(3):393–403. Epub 2012/06/19. pmid:23799583.

112. Benhaourech S, Drighil A, Hammiri AE. Congenital heart disease and Down syndrome: various aspects of a confirmed association. Cardiovasc J Afr. 2016;27(5):287–90. pmid:27805241; PubMed Central PMCID: PMCPMC5370349.

113. Seto E, Yoshida M. Erasers of histone acetylation: the histone deacetylase enzymes. Cold Spring Harb Perspect Biol. 2014;6(4):a018713. Epub 2014/04/01. pmid:24691964; PubMed Central PMCID: PMCPMC3970420.

114. Kwan KM, Fujimoto E, Grabher C, Mangum BD, Hardy ME, Campbell DS, et al. The Tol2kit: a multisite gateway-based construction kit for Tol2 transposon transgenesis constructs. Dev Dyn. 2007;236(11):3088–99. pmid:17937395.

115. Mandal A, Rydeen A, Anderson J, Sorrell MR, Zygmunt T, Torres-Vázquez J, et al. Transgenic retinoic acid sensor lines in zebrafish indicate regions of available embryonic retinoic acid. Dev Dyn. 2013;242(8):989–1000. Epub 2013/06/18. pmid:23703807; PubMed Central PMCID: PMCPMC3771353.

116. Hruscha A, Krawitz P, Rechenberg A, Heinrich V, Hecht J, Haass C, et al. Efficient CRISPR/Cas9 genome editing with low off-target effects in zebrafish. Development. 2013;140(24):4982–7. Epub 2013/11/20. pmid:24257628.

117. Gagnon JA, Valen E, Thyme SB, Huang P, Akhmetova L, Ahkmetova L, et al. Efficient mutagenesis by Cas9 protein-mediated oligonucleotide insertion and large-scale assessment of single-guide RNAs. PLoS One. 2014;9(5):e98186. Epub 2014/05/29. pmid:24873830; PubMed Central PMCID: PMCPMC4038517.

118. Hwang WY, Fu Y, Reyon D, Maeder ML, Tsai SQ, Sander JD, et al. Efficient genome editing in zebrafish using a CRISPR-Cas system. Nat Biotechnol. 2013;31(3):227–9. Epub 2013/01/29. pmid:23360964; PubMed Central PMCID: PMCPMC3686313.

119. Rydeen AB, Waxman JS. Cyp26 enzymes are required to balance the cardiac and vascular lineages within the anterior lateral plate mesoderm. Development. 2014;141(8):1638–48. pmid:24667328; PubMed Central PMCID: PMCPMC3978838.

120. Robu ME, Larson JD, Nasevicius A, Beiraghi S, Brenner C, Farber SA, et al. p53 activation by knockdown technologies. PLoS Genet. 2007;3(5):e78. Epub 2007/04/10. pmid:17530925; PubMed Central PMCID: PMCPMC1877875.

121. Villefranc JA, Amigo J, Lawson ND. Gateway compatible vectors for analysis of gene function in the zebrafish. Dev Dyn. 2007;236(11):3077–87. pmid:17948311; PubMed Central PMCID: PMCPMC4518551.

122. Duong TB, Ravisankar P, Song YC, Gafranek JT, Rydeen AB, Dohn TE, et al. Nr2f1a balances atrial chamber and atrioventricular canal size via BMP signaling-independent and -dependent mechanisms. Dev Biol. 2018;434(1):7–14. Epub 2017/11/20. pmid:29157563; PubMed Central PMCID: PMCPMC5785549.

123. Oxtoby E, Jowett T. Cloning of the zebrafish krox-20 gene (krx-20) and its expression during hindbrain development. Nucleic Acids Res. 1993;21(5):1087–95. pmid:8464695; PubMed Central PMCID: PMCPMC309267.

124. Brend T, Holley SA. Zebrafish whole mount high-resolution double fluorescent in situ hybridization. J Vis Exp. 2009;(25). Epub 2009/03/25. pmid:19322135; PubMed Central PMCID: PMCPMC2789764.

125. Waxman JS, Keegan BR, Roberts RW, Poss KD, Yelon D. Hoxb5b acts downstream of retinoic acid signaling in the forelimb field to restrict heart field potential in zebrafish. Dev Cell. 2008;15(6):923–34. pmid:19081079; PubMed Central PMCID: PMCPMC2752051.

126. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012;9(7):671–5. pmid:22930834; PubMed Central PMCID: PMCPMC5554542.

127. Lee Y, Manegold JE, Kim AD, Pouget C, Stachura DL, Clements WK, et al. FGF signalling specifies haematopoietic stem cells through its regulation of somitic Notch signalling. Nat Commun. 2014;5:5583. Epub 2014/11/27. pmid:25428693; PubMed Central PMCID: PMCPMC4271318.

128. D'Aniello E, Rydeen AB, Anderson JL, Mandal A, Waxman JS. Depletion of retinoic acid receptors initiates a novel positive feedback mechanism that promotes teratogenic increases in retinoic acid. PLoS Genet. 2013;9(8):e1003689. Epub 2013/08/08. pmid:23990796; PubMed Central PMCID: PMCPMC3750112.

129. Li-Kroeger D, Cook TA, Gebelein B. Integration of an abdominal Hox complex with Pax2 yields cell-specific EGF secretion from Drosophila sensory precursor cells. Development. 2012;139(9):1611–9. Epub 2012/03/21. pmid:22438572; PubMed Central PMCID: PMCPMC3317967.

130. Li J, Hu P, Li K, Zhao Q. Identification and characterization of a novel retinoic acid response element in zebrafish cyp26a1 promoter. Anat Rec (Hoboken). 2012;295(2):268–77. Epub 2011/12/23. pmid:22190473.

131. Bogdanović O, Fernández-Miñán A, Tena JJ, de la Calle-Mustienes E, Gómez-Skarmeta JL. The developmental epigenomics toolbox: ChIP-seq and MethylCap-seq profiling of early zebrafish embryos. Methods. 2013;62(3):207–15. pmid:23624103.

132. Waxman JS, Yelon D. Zebrafish retinoic acid receptors function as context-dependent transcriptional activators. Dev Biol. 2011;352(1):128–40. pmid:21276787; PubMed Central PMCID: PMCPMC3207040.

133. Loudig O, Babichuk C, White J, Abu-Abed S, Mueller C, Petkovich M. Cytochrome P450RAI(CYP26) promoter: a distinct composite retinoic acid response element underlies the complex regulation of retinoic acid metabolism. Mol Endocrinol. 2000;14(9):1483–97. pmid:10976925.