Developmental Biology 354 (2011) 123–133

Contents lists available at ScienceDirect

Developmental Biology

j ourna l homepage: www.e lsev ie r.com/deve lopmenta lb io logy

Mef2cb regulates late myocardial cell addition from a second heart field-likepopulation of progenitors in zebrafish

Savo Lazic a,b, Ian C. Scott a,b,c,⁎a Department of Molecular Genetics, University of Toronto, Toronto, Ontario, Canada M5S 1A8b Program in Developmental and Stem Cell Biology, The Hospital for Sick Children, Toronto, Ontario, Canada M5G 1X8c Richard Lewar Centre for Excellence in Cardiovascular Research, Toronto, Ontario, Canada M5S 1A8

⁎ Corresponding author at: The Hospital for Sick ChildRoom 11–307, Toronto, Ontario, Canada M5G 1L7. Fax:

E-mail address: ian.scott@sickkids.ca (I.C. Scott).

0012-1606/$ – see front matter © 2011 Elsevier Inc. Aldoi:10.1016/j.ydbio.2011.03.028

a b s t r a c t

a r t i c l e i n f o

Article history:Received for publication 17 February 2011Revised 25 March 2011Accepted 28 March 2011Available online 3 April 2011

Keywords:Mef2cHeart developmentSecond heart fieldZebrafish

Two populations of cells, termed the first and second heart field, drive heart growth during chick and mousedevelopment. The zebrafish has become a powerful model for vertebrate heart development, partly due to theevolutionary conservation of developmental pathways in this process. Here we provide evidence that thezebrafish possesses a conserved homolog to the murine second heart field. We developed a photoconversionassay to observe and quantify the dynamic late addition of myocardial cells to the zebrafish arterial pole. Wedefine an extra-cardiac region immediately posterior to the arterial pole, which we term the late ventricularregion. The late ventricular region has cardiogenic properties, expressing myocardial markers such as vmhc andnkx2.5, but does not express a full complement of differentiated cardiomyocytemarkers, lackingmyl7 expression.We show thatmef2cb, a zebrafish homolog of themouse second heart fieldmarkerMef2c, is expressed in the lateventricular region, and is necessary for latemyocardial addition to the arterial pole. FGF signaling after heart coneformation is necessary formef2cb expression, the establishment of the late ventricular region, and latemyocardialaddition to the arterial pole. Our study demonstrates that zebrafish heart growth shows more similarities tomurineheart growth thanpreviously thought. Further, as congenital heart disease is often associatedwith defectsin second heart field development, the embryological and genetic advantages of the zebrafish model can beapplied to study the vertebrate second heart field.

ren, 101 College Street, TMDT+1 416 813 8456.

l rights reserved.

© 2011 Elsevier Inc. All rights reserved.

Introduction

The heart is the first functional organ to form in the vertebrateembryo. Bilateral populations of cardiac progenitors fuse at themidline to form the linear heart tube, which later loops to form thetwo-, three-, or four-chambered heart (Srivastava, 2006). Growth andmorphogenesis proceed while the heart tube beats, putting noveldevelopmental constraints on cell migration, proliferation, anddifferentiation. As heart growth has been primarily studied interrestrial vertebrates, such as the mouse and chick, the evolutionaryconservation of heart growth has not been closely studied.

The study of cardiac progenitors in vertebrate heart developmentwas greatly advanced by the discovery of Nkx2.5, a homologue of theDrosophila tinman gene, whose mutation leads to loss of the fly heartequivalent, the dorsal vessel (Bodmer, 1993). Vertebrate Nkx2.5expression marks the bilateral cardiac progenitors (Chen and Fishman,1996; Lints et al., 1993), and it was at first assumed that these initialNkx2.5-positive cells represented all progenitors of the future myocar-dium. Themore recent discovery of the secondheart field (SHF) in chick

(Mjaatvedt et al., 2001; Waldo et al., 2001) and mouse (Buckinghamet al., 2005; Kelly et al., 2001) has highlighted the complexity of earlyheart development. At the early cardiac crescent stage, the first heartfield (FHF) is found lateral and ventral to the SHF and expresses heartdifferentiationmarkers before the SHF (Cai et al., 2003). SHF cells, foundin splanchnic mesoderm located dorsal to the midline heart, are apopulation of cardiacprogenitors thatmigrate into theheart tube after ithas been formed. Lineage tracing experiments have suggested thatdiscrete populations of heart progenitors contribute to restrictedregions of the heart (de la Cruz et al., 1977; Meilhac et al., 2004). TheSHF forms the majority of the right ventricle and the entirety of theinflow and outflow tracts of the heart. The FHF forms the initial linearheart tube and forms the entire left ventricle. Both the FHF and SHFcontribute cells to the atria. FGF,Wnt, BMP, and Hh signaling have beenshown to affect the proliferation, differentiation,migration, and survivalof SHF cells (Rochais et al., 2009). SHF regulators, such as Isl1 (Cai et al.,2003) and Mef2c, have been identified. The right ventricle and outflowtract are missing in Isl1 mutants, which is phenocopied by deletion ofMef2c (Lin et al., 1997), a direct transcriptional target of Isl1 (Dodouet al., 2004). Deficiencies in SHF contribution during heart developmenthave been shown to cause congenital heart defect (CHD)-likephenotypes in mouse models, including mispatterning of the outflowtract and septal defects (Dyer andKirby, 2009;Horsthuis et al., 2009). As

124 S. Lazic, I.C. Scott / Developmental Biology 354 (2011) 123–133

CHDs affect up to 1% of live births, a greater understanding of how cellsof the SHF are recruited to and differentiate in the heart is essential.

The early embryonic origins of the FHF and SHF are still poorlyunderstood (Meilhac et al., 2004). Further, the evolutionary origin ofthe SHF is unclear. Isl1 is expressed in Xenopus laevis (Brade et al.,2007) and Drosophila (Mann et al., 2009) embryos and is required forheart and dorsal vessel development, respectively. In the zebrafish,recent work has shown that formation of the heart tube is followed bylater phases of cardiomyocyte differentiation, with isl1 and FGFsignaling regulating addition to the venous and arterial poles of theheart, respectively (de Pater et al., 2009). It remains an open questionif an SHF-like population is conserved in the simpler two-chamberedheart of zebrafish. As defects in the SHF are key to many CHDs, this is acritical issue in considering the use of zebrafish to study thesediseases. The conservation of mouse SHF markers and regulators inzebrafish heart development would constitute strong evidence for amore ancient origin of the vertebrate SHF.

In this study,we employed a photoconvertible fluorescent protein tostudy the dynamic addition of cardiomyocytes to the arterial pole of thezebrafish heart. As previously reported (de Pater et al., 2009), weobserved early and late phases of myocardial cell addition. To furthercharacterize late myocardial cell addition, we measured cell additionbetween 24 hours post-fertilization (hpf) and 72 hpf. The majority ofaddition to the arterial pole occurs prior to 36 hpf, with low additionoccurring up to 2.5 days post-fertilization (dpf). We show that the cellscontributing to the early heart tube and the later addition to the arterialpole are derived from the same pre-gastrula region of the embryo,suggesting a shared progenitor. We discover the existence of acardiogenic region posterior and adjacent to the arterial pole, whichwe term the late ventricular region. We identify mef2cb as a novelregulator of zebrafishearly and late heart development that is expressedin the late ventricular region and show that it is necessary for latemyocardial addition.We show that FGF signaling after 20 hpf is requiredfor the establishment of the late ventricular region and the latemyocardial addition. Our data provide strong evidence that the latecell addition to the arterial pole of the zebrafishheart is analogous to themouse SHF, in particular the Mef2c-dependent anterior heart fieldpopulation. The ability to monitor with relative ease the dynamics ofheart cell addition in the zebrafish therefore represents a novel methodto study the regulation and development of the vertebrate SHF.

Materials and methods

Transgenic zebrafish lines

An nlsKikGR cassette was made by subcloning the KikGR codingsequence (Tsutsui et al., 2005) downstream of a nuclear localizationsignal. Zebrafish myl7 (900 bp) and kdrl (7 kb) promoter elements(Beis et al., 2005; Huang et al., 2003) were subcloned upstream of annlsKikGR cassette in between the minimal Tol2 transposon arms(Urasaki et al., 2006) in a pBluescript backbone vector. Stable Tg(myl7:nlsKikGR)hsc6 and Tg(kdrl:nlsKikGR)hsc7 zebrafish lines were madeusing standard Tol2 transgenesis approaches (Kawakami, 2005). Themyl7:EGFPtwu34 line used has been previously described (Huang et al.,2003).

Embryo maintenance, transplantation and microinjection

Zebrafish embryos were grown at 28 °C in embryo medium aspreviously described (Westerfield, 1993). Standard techniques wereused for transplantation and microinjection approaches. For rescueexperiments, morpholino-resistant mef2cb mRNA was created usingthe mMESSAGE mMACHINE kit (Applied Biosystems). Mutant mef2cbmRNA bearing five silent substitutions in the morpholino-bindingsequence (5′-ATGGGCCGGAAGAAAATTCAGATCAC-3′, where boldletters denote mutated positions) was subcloned into pCS2+ for in

vitro transcription of mRNA. mRNA (20 pg) per embryo was injectedfor rescue. nlsKikGR mRNA was created using the mMESSAGEmMACHINE kit (Applied Biosystems). Transplantation was performedusing 1) myl7:nlsKikGR transgenic donor embryos and wildtype hostembryos and 2) myl7:EGFP+tetramethylrhodamine dextran(10,000 MW, Invitrogen) transgenic donor embryos and wildtype hostembryos. At sphere stage (4 hpf), 10–20 cells were transplanted fromthe donor embryo to the margin of host embryos. For myl7:nlsKikGRtransplants, UV illumination of transplant embryos was performed at24 hpf and embryos were imaged at 48 hpf.

Photoconversion and image analysis

Photoconversion was carried out on a Zeiss Lumar V12 stereomi-croscope using fluorescent light passed through the DAPI filter (Zeiss,485049) until all green fluorescence was lost in the heart (approx-imately 1–2 min). Subsequent imaging of the heart was performed ona Zeiss Axiovert 200M Spinning Disk Confocal. To mount the embryosfor imaging, embryos were fixed in 4% PFA for 5 min and washedthree times in PBS. Two to five embryos were placed on a glass slideand covered gently by a cover slip. Image analysis was performedusing the Velocity 5 Software Suite (Improvision). Ventricular cellswere counted using the Point Tool. The green channel was turnedoff periodically to ensure that cells counted did not contain low levelsof red fluorescence.

Localized photoconversion

Embryos were injected with 100 pg of nlsKikGR mRNA at the one-cell stage. At 24 hpf, localized photoconversion was performed on theZeiss Axio Imager M1 microscope, using the fluorescence diaphragmto restrict the area of photoconversion to the distal tail. Each embryowas grown in the dark and imaged under identical exposure and gainsettings at 24, 48, and 72 hpf on a Leica M205 FA microscope.

Morpholinos

Morpholinos were purchased from Gene Tools (Oregon, USA). Themef2cbmorpholinowasdesigned against theATG site ofmef2cb to blocktranslation (5′-CTGAATCTTTTTTCTCCCCATTGTC-3′, the translationalstart site is underlined). Injection of 0.5 ng of morpholino at the one-cell stage yielded a heart-specific phenotype.

Chemical inhibition of FGF signaling

SU5402 was purchased from Tocris Biosciences (Missouri, USA).Embryos were treated either with 10 μM SU5402 in 1% DMSO or in 1%DMSO alone from 20 to 30 hpf. Inhibitor was removed by threewashes with fresh embryo medium.

RNA in situ hybridization

RNA in situ hybridization was performed as previously described(Thisse and Thisse, 2008) using riboprobes specific for myl7/cmlc2,vmhc, amhc, and nkx2.5 (Chen and Fishman, 1996; Yelon et al., 1999).For mef2cb riboprobe, 800 bp of mef2cb sequence was amplified (seprimer 5′-CACGGATTATGGATGAACGCAACAGA-3′ and as primer 5′-CCAGTGATTGCGCAGACTGAGAGTTG-3′) and cloned into the pGEM-TEasy Vector (Promega). Plasmids were linearized and DIG-labeledprobes made using a DIG RNA Labeling Kit (Roche). In situ imageswere captured on a Leica M205 FA microscope.

Bulbus arteriosus staining

The bulbus arteriosus was visualized using the NO indicator DAF-2DA as previously described (Grimes et al., 2006). Live embryos were

125S. Lazic, I.C. Scott / Developmental Biology 354 (2011) 123–133

incubated overnight in 10 μM DAF-2DA (EMD Chemicals, NJ, USA)diluted in eggwater adjusted to pH 7.0.

BrdU analysis

myl7:EGFP embryos were incubated on ice for 1 h in 10 mM BrdU(Sigma, cat# B9285)+15% DMSO in E3 at 24 and 28 hpf. BrdU wasremoved by three washes of E3 and embryos were grown to 48 hpf.After overnight fixation in 4% PFA, embryos were dehydrated inmethanol. Permeabilization was achieved by a 15 min incubation atroom temperature in 10 μg/ml of ProK and a 1 h incubation at roomtemperature in 2 N HCl. Mouse anti-BrdU (BD Biosciences, cat#347580) and rabbit anti-EGFP (Torrey Pines Biolabs Inc., cat# TP401)in conjunction with Alexa Fluor 568 anti-mouse (Invitrogen, cat# A-11004) and Alexa Fluor 488 anti-rabbit (Invitrogen, cat# A-11008)antibodies were used to visualize proliferating cells and EGFP-positivecardiomyocytes, respectively. BrdU positive cells in the arterial polewere counted using the Velocity 5 Software Suite (Improvision).Special care was taken to count only BrdU positive cells in the sameoptical plane as the myocardial cells.

Results

Late addition of myocardial cells to the zebrafish arterial pole of the heart

To observe the addition of myocardial cells to the beating hearttube, we exploited the properties of the photoconvertible fluorescentprotein KikGR (Tsutsui et al., 2005). KikGR fluoresces brightly, withspectral characteristics similar to EGFP. A brief exposure to UV lightpermanently photoconverts KikGR by cleavage of a Histidine side-chain in the chromophore. This causes a shift in fluorescence

Fig. 1. Photoconversion assay showed late myocardial addition to the arterial pole. (A). ScheUV light at 24 hpf (and later time points), permanently photoconverting nlsKikGR to a red flumarked by red fluorescence additionally expressed newly synthesized green fluorescing nlsKnlsKikGR. (B–B”) Representative image of a heart photoconverted at 24 hpf and imaged at 4not marked by red fluorescence (B’). (C) Cell counting of early and late ventricular myocardiaNumber of green-only cells observed decreased with later UV exposure, whereas yellow cellnlsKikGR line, marking endothelial cells. UV photoconversion at 28 hpf and imaging at 48 hpthe arterial pole (D”). Green channel (B and D), red channel (B’ and D’), and overlay (B” an

characteristics to a spectrum more similar to RFP. Using Tol2-mediated transgenesis, we created transgenic zebrafish expressingnuclear localized KikGR (nlsKikGR) in differentiated cardiomyocytesunder the regulation of the myl7 (cmlc2) promoter (myl7:nlsKikGR).Earliest nlsKikGR photoconversion was observable at 22 hpf. To studylate myocardial addition, hearts were exposed to UV at discrete timepoints between 24 and 38 hpf, marking differentiated heart cellsexpressing nlsKikGR with a red fluorescent signal. At 48 hpf, cellsmarked in this way will express both the red and green (as myl7-positive cells continuously produce new nlsKikGR) fluorescentversions of nlsKikGR, whereas cardiomyocytes that differentiateafter UV exposure will only fluoresce green (Fig. 1A). To ensure thatred fluorescence persisted for at least 24 h, nlsKikGR mRNA wasinjected into the one-cell stage embryo. At 24 hpf, a small region inthe posterior tail was photoconverted (Fig. S1A) and fluorescence wasmonitored over the next 2 days. Red fluorescence persisted for at least48 h (Fig. S1A’, A”), during which the labeled cells spread over a largerarea. To test for red fluorescence persistence in our myl7:nlsKikGRtransgenic embryos, 24 hpf hearts were photoconverted and imaged at48 hpf (Fig. S1B) and 72 hpf (Fig. S1C). Red fluorescence (Fig. S1B’, C’)remained readily detectable at both time points.

Late addition of myocardial cells was seen to occur primarily at thearterial (ventricular outflow) pole of the heart (Fig. 1B), with noaddition observed in the venous (atrial inflow) pole. In rare instances,we observed single green nuclei in the ventricle distal to the arterialpole (Fig. S2A, arrowhead). Given the rarity of this non-pole addition,we have focused our current analysis on the arterial pole. We firstcarried out time-course experiments to characterize the period overwhich arterial pole addition of cardiomyocytes occurs. The majority oflate cardiomyocyte addition to the arterial pole, as observed at 48 hpf,occurred with UV exposure at 24 hpf (42±2 cells, n=8) to 30 hpf

matic representation of myl7:nlsKikGR photoconversion assay. Hearts were exposed toorescent version andmarking differentiated myocardial cells. At 48 hpf, myocardial cellsikGR (yellow in overlay), while new myocardial cells only expressed green fluorescing8 hpf. White arrowhead indicates a population of green-only cells (B and B”) that werel nuclei. Hearts were exposed to progressively later UV exposures and imaged at 48 hpf.numbers increased. Data is shown as mean±SEM (D–D”) Representative image of kdrl:f showed a population of green-only endothelial cells (compare D to D’, white arrow) atd D”). Scale bar represents 50 μm.

126 S. Lazic, I.C. Scott / Developmental Biology 354 (2011) 123–133

(33±4 cells, n=5), with fewer cells added with UV exposure at34 hpf (14±2 cells, n=5) and 38 hpf (6±1 cells, n=4) (Fig. 1C).Concurrent with the decrease of green-only heart cells, the number ofyellow cells showed a gradual increase. UV exposure between 48 and56 hpf showed that three to six green-only cells are added to thearterial pole (data not shown). To further refine the time frame of latemyocardial addition, we performed the photoconversion assay over ashorter developmental window. UV exposure of embryos at 23 hpfand imaging at 28 hpf showed late myocardial addition to the arterialpole (Fig. S3A). Further, UV exposure at 28 hpf and imaging at32 hpf resulted in late myocardial addition at the arterial pole as well(Fig. S3B). A common feature of green-only cells at 48 hpf is that faintgreen fluorescence, which represents the myocardial cells that havemost recently turned on the myl7 promoter, is observed at the distaltip of the arterial pole (Fig. S2B, arrow). Our photoconversion resultsshow a gradual and continuous myocardial cell addition to the arterialpole, with the newest myocardial cells appearing at the distal tip ofthe heart tube.

Fig. 2. Contribution of wildtype andmef2cbmorphant cells to early and late myocardial po(A–A”) and late (arrowhead) and early (B–B”) myocardial addition (quantified in E). Momyocardial addition (C–C”). The average number of heart cells found in theWT to WT or M(A”, B”, and C”). Data shown as mean±SEM; **Pb0.01. Scale bar represents 50 μm.

To test if late endocardial cell addition occurs as well, we createdtransgenic fish expressing nlsKikGR under control of the endothelialcell-specific kdrl promoter (Choi et al., 2007). Robust photoconversionwas observed at 28 hpf, with no late endocardial cell addition in theventricle evident at 48 hpf (Fig. 1D). Unexpectedly, endothelial cells invessels immediately adjacent to the arterial pole were evident thatexclusively showed green fluorescence, demonstrating late endothe-lial addition adjacent to the arterial pole (Fig. 1D, arrow).

Embryonic origins of early and late cardiomyocytes in the ventricle

We next wished to examine if late myocardial cells of the arterialpole shared a common embryonic origin with earlier differentiatingcells of the heart tube. Prior to the onset of gastrulation, myocardialcell progenitors are localized to a specific region of the embryonicmargin (Stainier et al., 1993). At 4 hpf, cells transplanted into thisregion can take on a later myocardial fate. To investigate the pre-gastrula origin of late myocardial addition, we transplanted 10–20

pulations. WT cells transplanted to the WT host margin contributed to both early onlyrphant cell (0.25 ng) transplanted to the WT host margin contributed only to earlyO toWT transplants (D). Green channel (A–C), red channel (A’, B’, and C’), and overlay

127S. Lazic, I.C. Scott / Developmental Biology 354 (2011) 123–133

donor cells from Tg(myl7:nlsKikGR) embryos into the margin ofwildtype embryos at sphere stage (4 hpf). Hosts were exposed to UVlight at 24 hpf and imaged at 48 hpf. Three outcomeswere expected in agiven embryo where donor cells form cardiomyocytes: all transplantedcells fluoresce yellow (only early contribution), all transplanted cellsfluoresce green (only late contribution), or transplanted cells fluorescegreen and yellow (both early and late contribution). Out of 15transplants with myocardial contribution, 8 hearts showed only earlycontribution (Fig. 2A and E) and 7 hearts showed both early and latecontribution at 48 hpf (Fig. 2B and E). No exclusive contribution ofgreen-only cells was observed in any embryo (Fig. 2E). These resultsdemonstrate that early and late myocardial cells arise from a similarregion in the pre-gastrula embryo, andmay potentially share a commonprogenitor.

Expression of vmhc, nkx2.5 and mef2cb marks a novel cardiogenic regionat the arterial pole of the heart

To identify potential markers of cells added to the arterial poleprior to the onset of myl7 expression, we examined expression ofcandidate cardiac genes from 24 to 30 hpf. Late addition of myocardialcells occurred at the arterial pole, leading us to hypothesize that thecardiogenic source of late myocardial cells was located immediatelyposterior to the ventricle. At 30 hpf, vmhc transcripts localized to anextra-cardiac region posterior to and continuous with the ventricle,which we termed the late ventricular region (Fig. 3B, arrowhead).vmhc is expressed in the late ventricular region by 26 hpf (Fig. S4C,arrowhead), with no late ventricular region expression seen at 20 hpf(Fig. S4A). Potential early late ventricular region staining could beobserved at 24 hpf (Fig. S4B, arrowhead). Interestingly, from 26 to

Fig. 3. The late ventricular regionwasmarked by knownmyocardialmarkers nkx2.5 and vmhc, aregion (arrowhead), and bilateral pharyngeal domain. The late ventricular region was markedprogenitors at 17 hpf (E), in the heart cone (F), and heart tube (G). Expression becamegraduallywas co-expressedwith vmhc (compare I to J). Probing formef2cb andmyl7 expression at the sam

30 hpf vmhc expression marked what appeared to be “streams” ofcells extending from the arterial pole of the heart (Fig. S4C–E). Inaddition to vmhc, nkx2.5 expression was also observable in the lateventricular region, the heart tube (Chen and Fishman, 1996) and in aposterior bilateral pattern in the pharyngeal region (Fig. 3A).

To examine if the late ventricular region harbors similarities to theSHF, we looked at the expression of zebrafish homologs of mouse SHFmarkers. We found that mef2cb was expressed in the late ventricularregion at 30 hpf (Fig. 3D and I–L). mef2cb transcripts were found inbilateral stripes in the heart-forming region of the anterior lateralplate mesoderm (ALPM) at 17 hpf (Fig. 3E), and later in the in theheart cone at 20 hpf (Fig. 3F). As the heart cone elongated (Fig. 3G),mef2cb expression became localized to the posterior portion of theventricle (Fig. 3D), and by 32 hpf, faint expression was seen in thearterial pole (Fig. 3H). Expression ofmef2ca, the paralog ofmef2cb, hasbeen reported in myocardial progenitors at 16 hpf (Ticho et al., 1996),with mef2ca morphants having a “strung out” heart phenotype(Ghosh et al., 2009). At 20 hpf, we found thatmef2cawas expressed inthe heart cone (Fig. S5A), but, unlikemef2cb, also remained expressedthroughout the heart at 30 hpf (Fig. S5B and C).

To confirm mef2cb expression outside the myocardium, wecombined mef2cb and myl7 probes and performed RNA in situhybridization. Expression ofmyl7 did not extend to the late ventricularregion (Fig. 3C and K), whereas late ventricular region staining wasevident whenmef2cb probe was included (Fig. 3L). Combiningmef2cband vmhc (Fig. 3J) probes, in contrast, mirrored the expression patternseenwith vmhc alone (Fig. 3I), arguing thatmef2cb is co-expressed in asubset of vmhc-positive cells found immediately adjacent to theventricle. These results suggest the existence of a novel cardiogenicregion immediately posterior to the arterial pole of the heart.

s well asmef2cb. (A) nkx2.5 transcript was expressed in the heart tube, the late ventricularby vmhc (B), but not by myl7 (C).mef2cb was expressed in the presumed bilateral heartlocalized to the posterior ventricle (D) and became down-regulated by 32 hpf (H).mef2cbe time showed thatmef2cb is expressed outside the heart tube at 30 hpf (compare K to L).

128 S. Lazic, I.C. Scott / Developmental Biology 354 (2011) 123–133

mef2cb is necessary for ventricular development and late myocardialaddition

To test the function of mef2cb in zebrafish heart development, weinjected 0.5 ng of morpholino targeting mef2cb translation. Morpho-lino injected (morphant) embryos were indistinguishable fromwildtype controls until 24 hpf, when the morphant heart failed toinitiate beating. At 48 hpf, a cardiac edema was the only observedmorphant morphological defect (compare Fig. 4A and B), with readily

Fig. 4.mef2cbmorphants displayed defects in early and late myocardial addition. Compared tpooling (B). For (C–H) ntl probe was used as a midline reference. mef2cb expression was uventriclewas lost (compare C” to D”).Morphantmef2cb expression at 17 hpfwas expended poexpression (C). myl7 expression (E vs. F) and vmhc expression (G vs. H) was reduced in morsignificantly anterior to ntl expression, whereas wildtype expression of both genes (E and Gsmaller and lacked the late ventricular region (compare I to I’). The in situ datawas corroboratto controls (J). A decrease in both late and early myocardial addition was observed in mef2c

visible structures such as eyes, fins, and ears developing normally.Compared to wildtype hearts (Fig. S6B), morphant hearts beat at areduced rate and were either strung out (Fig. S6C) or showed aninflated atrium and collapsed ventricle (Fig. S6D). Injection of 20 pg ofmRNA encoding a morpholino-resistant form of mef2cb reduced theseverity of the cardiac phenotype in morphants (Fig. S6A), with themajority of the embryos having a clearly defined atrium and collapsedventricle. Injection of higher concentrations of mef2cb mRNA provedto be toxic. Partial rescue observed with mef2cb RNA injection, in

oWT controls (A), 48 hpfmef2cbmorphant embryos displayed a heart edema and bloodp-regulated in morphants at 17 hpf (compare C to D) and localization to the posteriorsterior to ntl expression (D),whereaswildtypemef2cb expression is found anterior to ntlphants at all stages tested. At 17 hpf, morphant expression of myl7 (F) and vmhc (H) is) is found immediately anterior to ntl expression. At 30 hpf, morphant ventricles wereed by photoconversion assay, inwhich themorphant ventricle (J’)was smaller comparedb morphants (K). Data shown as mean±SEM; **Pb0.01. Scale bar represents 50 μm.

Fig. 5. Cardiac contribution of mef2cb morphant cells. (A) Morphant transplantscontributed to fewer hearts than wildtype transplants, but the difference was notsignificant (P=0.12). Within an individual heart, morphant cells contributed fewercells (C) when compared to wildtype controls (B). Data shown as mean±SEM. Scalebar represents 100 μm.

129S. Lazic, I.C. Scott / Developmental Biology 354 (2011) 123–133

addition to the specific effect on heart development alone, stronglysuggests that the mef2cb MO phenotype is specifically due toinhibition of mef2cb.

Surprisingly, mef2cb expression was dramatically expanded inmorphants at 17 hpf (compare Fig. 4C and D), with a slight expansionalso apparent at 20 hpf (compare Fig. 4C’–D’). Using ntl as a midline(notochord) marker, we observed that mef2cb expression extendedfurther posteriorly in morphant embryos. Interestingly, the restrictionof mef2cb transcripts to the arterial pole at 26 hpf was lost inmorphants, with expression observed throughout the ventricle(Fig. 4D”). Concurrent with the expansion of mef2cb expression,staining levels for myl7 and vmhc were reduced at 17 hpf (Fig. 4F andH, respectively), with a significant reduction of vmhc expressionevident at 20 hpf (Fig. 4H’) and 26 hpf (Fig. 4H”). In addition,myl7 andvmhc expression in morphants was found to be more anterior to ntlcompared to the wildtype expression. Expression of the atrial markermyh6 (amhc) was reduced in morphant embryos at 26 hpf (Fig. S6F);however, the size of the atrium appeared normal at 26 hpf (Fig. S6E).

We next examined the role of mef2cb in late myocardial addition tothe arterial pole of the heart. At 30 hpf, morphants showed a reducedventricle that lacked the late ventricular region (Fig. 4I’). Wildtype andmef2cb morphant embryos carrying the myl7:nlsKikGR transgene wereUV illuminated at 30 hpf, and myocardial cell numbers quantified at48 hpf. In morphant hearts (Fig. 4J’), there is a significant reduction inboth early (23±3 vs. 103±8; n=6 vs. n=5, respectively) and late celladdition (8±1 vs. 27±2; n=6 vs. n=5, respectively) as compared towildtype (Fig. 4J and K). Given the strong effect ofmef2cb depletion oninitial development of the ventricle, transplant experiments were usedto discern the effect ofmef2cb on late myocardial addition to a wildtypeventricle.

To test the ability of mef2cb morphant cells to contribute to heartand other tissues, donor cells of mef2cb morphant transgenic myl7:EGFP embryos injected with tetramethylrhodamine dextran weretransplanted into the embryonic margin of wildtype 4 hpf embryos.The general contribution of morphant cells was indistinguishablefrom that of wildtype control cells, with widespread contribution toskeletal muscle and neural tissue clearly evident (data not shown).Contribution of donor cells to the heart was confirmed by EGFPfluorescence. Morphant cells contributed to the myocardium in alower percentage of host embryos (11%±4%, n=270) than wildtypecontrols (20%±3%, n=270), but not to a statistically significantdegree (Fig. 5A, P=0.12). However, within individual hearts,morphant cells appeared to contribute fewer myocardial cells(Fig. 5C) as compared to the wildtype controls (Fig. 5B).

To quantify morphant cell contribution to the heart, we trans-planted wildtype and mef2cb morphant myl7:nlsKikGR cells intowildtype hosts at sphere stage and counted the number of donormyocardial cells at 48 hpf. Upon transplantation of 0.5 ng morphantcells, we were unable to observe nlsKikGR positive heart cells,possibly due to the weaker fluorescence of nlsKikGR compared toEGFP. To obtain increased morphant cell contribution to the ventricle,we repeated the above experiment using a lower dose (0.25 ng) ofmorpholino. Embryos injected with 0.25 ng mef2cb morpholinodisplayed a milder heart phenotype. The heart remained unlooped,but chamber inflation and blood flow were apparent. We observedheart contribution in the 0.25 ng morpholino transplantation exper-iments (6% of embryos). Morphant cells contributed significantlyfewer cells (14±2, n=14) to the heart when compared to wildtype(24±3, n=11) (Fig. 2D). To test if this decrease in morphant cellcontribution affected late myocardial addition, donor cells fromwildtype and mef2cb morphant (0.25 ng) transgenic myl7:nlsKikGRembryos were transplanted into the embryonic margin of wildtype4 hpf embryos. Subsequent to this, embryos were UV illuminated at24 hpf, and contribution of donor cells to the early and late ventricularpopulations was assessed at 48 hpf. Morphant cells contributedexclusively to the early myocardial population (Fig. 2C and E).

Together, our transplant data show that mef2cb morphant cellscontribute to the heart, albeit with significantly fewermyocardial cellsadded per heart than wildtype. Our results suggest mef2cb function isessential for addition of myocardial cells to the arterial pole of theheart.

The bulbus arteriosus can be distinguished after 48 hpf immedi-ately adjacent to the arterial pole and is thought to be akin the arterialtrunk(s) of terrestrial vertebrates (Grimes et al., 2010). We tested ifthe loss of late myocardial addition to the arterial pole disruptedformation of the bulbus arteriosus by staining with DAF-2DA, a NOindicator that labels smooth muscle cells in the outflow tract (Grimeset al., 2006), in 5 dpf control and morphant embryos. Compared towildtype (Fig. S7A) embryos, morphant embryos developed atruncated bulbus arteriosus (Fig. S7B). We then tested if late myo-cardial addition affects the late endothelial addition observed usingthe kdrl:nlsKikGR transgenic line. Late endothelial addition was notlost in morphants (Fig. S8B”) when compared to wildtype embryos(Fig. S8A”).

FGF signaling is required during late myocardial addition

We next wished to examine how signaling pathways known toregulate mouse and chick SHF development regulate expression ofgenes we have localized to the late ventricular region of the arterialpole of the heart. FGF signaling has been shown to regulate the latedifferentiation of cardiomyocytes at the arterial pole of the zebrafishheart (de Pater et al., 2009). A critical role for FGF signaling in murineSHF development has been shown (Park et al., 2008; Zhang et al.,2008). We therefore assayed whether perturbation of FGF signalingaffected late ventricular region gene expression (in particularmef2cb).We inhibited FGF signaling between 20 and 30 hpf using the chemicalinhibitor SU5402, which binds and blocks the FGFR catalytic domain(Sun et al., 1999). Inhibition of FGF signaling resulted in loss ofmef2cbexpression at 30 hpf (Fig. 6A’). Additionally, localization of vmhctranscript to the late ventricular region was lost (Fig. 6B’), whileexpression in the ventricle proper remained unaffected (Fig. 6B). We

Fig. 6. FGF signaling between 20 and 30 hpf was required for early and late myocardial addition. Chemical inhibition of FGF signaling between 20 and 30 hpf completely abolishedmef2cbexpression at 30 hpf (compare A to A’). At 30 hpf, the ventricle was smaller and the late ventricular region (arrowhead) was not seen in FGF inhibited embryos (compare B to B’).Photoconversion assay (D) between 30 and 48 hpf showed a smaller ventricle and less latemyocardial addition in FGF inhibited embryos (C’) compared to DMSO treated control (C). Datashown as mean±SEM; **Pb0.01. Scale bar represents 50 μm.

130 S. Lazic, I.C. Scott / Developmental Biology 354 (2011) 123–133

employed the photoconversion assay to observe the effect of late FGFinhibition on early and late myocardial cell addition. Followingphotoconversion at 30 hpf and imaging at 48 hpf, we observed amoderate but significant effect of FGF inhibition on early myocardialcells compared to controls (Fig. 6C’ and D, 80±3 vs. 104±3; n=6 vs.n=4, respectively). However, a much stronger effect of FGF inhibitionon late myocardial addition (7±2 vs. 24±1 in wildtype; n=6 vs.n=4, respectively) was apparent. Our data shows that FGF signalingbetween 20 and 30 hpf is necessary for the recruitment of latemyocardial cells and the establishment of vmhc andmef2cb expressionin the late ventricular region.

FGF signaling has been shown to be a key regulator of proliferationof murine SHF progenitor cells (Hutson et al., 2010; Tirosh-Finkelet al., 2010). To test if proliferation during the establishment of thelate ventricular region plays a role in late myocardial addition, weemployed 1-h BrdU pulses at 24 hpf (Fig. 7A) and 28 hpf (Fig. 7B) andsubsequently imaged embryos at 48 hpf. Defining the arterial pole asthe distal one quarter of the ventricle, we observed 8±2 and 6±1BrdU positive cells at 48 hpf following BrdU incubation at 24 and at 28hpf, respectively (Fig. 7C). This suggests proliferation occurs in futurelate myocardial cells during the period at which the late ventricularregion is being established.

Discussion

In this study, we have used a photoconvertible assay to monitorand quantify the dynamics and regulation of myocardial cell additionduring zebrafish heart development. We show that gradual myocar-dial addition occurs at the arterial pole during the second day ofdevelopment, primarily between 24 hpf and 34 hpf. Global myocar-dial cell proliferation is low during this developmental period (dePater et al., 2009), indicating that late addition of newly formedmyocardial cells is likely a major mechanism of zebrafish heartgrowth. Our work adds considerably to the previous demonstration oflate differentiating cardiomyocytes being added to the venous andarterial poles of the zebrafish heart (de Pater et al., 2009). Focusing onthe arterial pole, we have identified a population of cells adjacent tothe ventricle that expresses markers of early myocardial progenitors,including nkx2.5, vmhc andmef2cb. Expression of thesemarkers in this“late ventricular region” is highly correlated with late myocardialaddition to the arterial pole, as shown by perturbation of mef2cb andFGF. Either the loss ofmef2cb expression in the late ventricular regionor the expansion ofmef2cb expression into the ventricle disrupted latemyocardial addition, implicating mef2cb transcript localization as anecessary mechanism for late heart development. As Mef2c is a key

Fig. 7. Cell proliferation in the arterial pole.myl7:EGFP embryos were pulsed with BrdUfor 1 h at 24 hpf (A) and 28 hpf (B) and cell proliferation was assessed at 48 hpf. Themyocardium is visualized in the green channel and BrdU labeled cells are visualized inthe red channel. Proliferating cells in the arterial pole of the heart (arrowheads) werequantified (C). Scale bar represents 70 μm.

131S. Lazic, I.C. Scott / Developmental Biology 354 (2011) 123–133

regulator of SHF development in the mouse, our discovery of mef2cbas a key regulator of zebrafish arterial polemyocardial addition clearlysuggests that these cell populations share common features in themouse/chick and zebrafish. While we are currently unable to directlyshow that these cells migrate into the heart, the expression patterns ofmef2cb and vmhc (Fig. 3 and Fig. S4) and the addition of the newestmyocardial cells to the distal tip of the arterial pole (Fig. S2B) aresuggestive of a migratory stream of differentiating cells adjacent tothe ventricle. However, at present we cannot exclude the possibilitythat late myocardial cells are part of the early heart tube and undergobelated differentiation. We have for the first time examined theembryonic origins and developmental regulation of these latemyocardial cells. Combined transplantation and photoconversionapproaches showed that the late myocardial cells arise from thesameembryonic regionas the initialmyocardial progenitors of the hearttube. We tested the in vivo persistence of photoconverted redfluorescing nlsKikGR, showing that strong fluorescence can be observedfor at least 48 h after photoconversion. These properties allowed foreasy and quick observation of late myocardial addition. However, weacknowledge that our assay may overestimate the timing of latemyocardial addition due to the delay frommyl7 promoter activation towhen nlsKikGR is first detected.

We believe that this evidence, in conjunction with the previousdemonstration that zebrafish isl1 regulates late myocardial addition tothe venous but not arterial pole (de Pater et al., 2009), strongly supportsamodel inwhich amyocardial population analogous to the SHFexists inzebrafish. While it remains to be seen if mef2cb regulates myocardialaddition to the venous pole before mef2cb becomes localized to theventricle, the distinct roles of isl1 andmef2cb in zebrafishmay be due tothe anatomy of the 24 to 48 hpf heart. As the atrial chamber is over theyolk, distal from the splanchnic mesoderm, an earlier contribution ofcells during the heart cone stage may occur at the venous pole.Interestingly, we also observed rare late myocardial addition distal tothe arterial pole. Although the rarity of these cells precluded anydetailed functional analysis, zebrafish cardiac neural crest cells (SatoandYost, 2003) andmurine epicardium(Zhou et al., 2008) are known tocontribute to themyocardial populations andmaybe the source of these

cells. Given the distinct localization of the pre-gastrula neural crest andmyocardial progenitors in the pre-gastrula zebrafish embryo (Kimmelet al., 1990; Stainier et al., 1993), our transplantation results argueagainst a neural crest origin for late myocardial cells at the arterial pole.

Elegant retrospective clonal analysis in the mouse has suggestedthat while FHF and SHF cells may share a common progenitor, theselineages diverge very early in development, perhaps coincident withthe onset of gastrulation (Meilhac et al., 2004). We showed in thispresent work that early and late myocardial cells are derived from thesame pre-gastrula region of the embryo, suggesting these cells mayshare an early common cardiac progenitor. As multiple donor cellswere used for each transplant, we cannot at present definitivelydetermine if and when the progenitors of the late and early myo-cardial populations diverge. In mouse experiments a rough ratio of2 FHF progenitors to 1 SHF progenitor was estimated based on clonesize in the embryonic heart (Meilhac et al., 2004). It is interestingto note that our transplant experiments, while qualitative, similarlysuggest a rough 2:1 ratio of early to late cells, or at the very least agreater extent of early progenitors. Future work with more complexsingle cell transplants or recently described mosaic lineage tracingapproaches (Collins et al., 2010) will be required to address the pro-genitor origin of the early and late myocardial populations.

At 17 hpf, expression of mef2cb is found in a bilateral population ofcells in the anterior lateral plate mesoderm anterior to ntl expression.Interestingly, in morphants mef2cb expression extends posterior to ntlexpression, while the posterior extents of vmhc and myl7 expressiondomains are more anterior relative to ntl expression. This indicates thatmef2cb might play a role in the correct anterior–posterior localizationof myl7 and vmhc expression. Unfortunately, mef2cb morphant rescuewas incomplete, but did result in the improvement of heart chamberinflation. This is most likely due to the low quantity of rescue mRNAinjected, necessitated by the high toxicity of mef2cb mRNA. However,the rescue observed in conjunction with the highly specific morphantphenotype argues that themef2cbmorpholino phenotype is not due tooff-target effects.

Weobserveda strong reduction in early and latemyocardial additioninmef2cbmorphants, arguing thatmef2cbplays both early and late rolesin zebrafish heart development. Our mef2cb morphant transplant datashows that mef2cb morphant cells contribute fewer cells to the heart ,that mef2cb is necessary for late myocardial addition to the ventricle,and that defects in mef2cb morphant late myocardial addition are notsecondary to a reduced ventricle. Intriguingly, mef2cb expressionbecomes progressively restricted to the arterial pole and the lateventricular region, the late cardiogenic region expressing a subset ofventricular markers, such as nkx2.5 and vmhc, but lacking expression ofmyl7. The localized mef2cb expression has also been shown in Atlanticcod, suggesting that mef2cb localization is a conserved teleostcharacteristic (Torgersen et al., 2010).mef2cbmorphants lose localizedexpression in the arterial pole, suggesting that mef2cb, directly orindirectly, is involved in its own localization within the heart. Theprecise nature of mef2cb function remains elusive. mef2cb morphantsdisplay an expansion ofmef2cb transcript at 17 hpf, strongly suggestingnegative mef2cb autoregulation. In mouse, HDAC9 has been implicatedin the negative auto-regulation of Mef2c (Haberland et al., 2007). Ourtransplantdata suggests thatmef2cb activity is further required for heartprogenitor recruitment, survival or later differentiation, consistent withthe recent demonstration that MEF2C can drive reprogramming offibroblasts to cardiomyocytes in vitro (Ieda et al., 2010). Taken together,our data is consistent with mef2cb being required to promote theterminal differentiation of late and early myocardial progenitors tocardiomyocytes. It will be of great interest to determine how mef2cbexpression is regulated, and also how the activity ofMef2cb is regulatedin the developing zebrafish heart.

While not a focus of this study, our data indicates that theendocardium does not grow by cell addition during the second day ofdevelopment. We hypothesize that the endocardium grows by the

132 S. Lazic, I.C. Scott / Developmental Biology 354 (2011) 123–133

proliferation of endocardial cells established during early heart develop-ment. Interestingly, we observed that new endothelial differentiationoccurs posterior to the arterial pole. This endothelial addition is notperturbed in mef2cb morphants, suggesting that mef2cb does not play arole inendothelial differentiation and that this newaddition isnot relatedto the late ventricular region. This matches what has been observed inMef2c null mice, where endothelial cells are correctly specified anddifferentiated, but are unable to organize properly (Lin et al., 1998).

Previous studies (de Pater et al., 2009) have implicated FGF signalingduring the second day (24–48 hpf) of zebrafish development in latemyocardial addition. We wanted to further characterize this regulationusing our photoconversion assay and markers. We showed that FGFinhibition between 20 and 30 hpf was sufficient to strongly reduce latemyocardial addition. The moderate decrease in early myocardial cellsmight indicate a decrease in myocardial proliferation. Interestingly,mef2cb expression and late ventricular region localization of vmhc arecompletely lost upon FGF inhibition, further linking the late ventricularregion andmef2cb to late myocardial addition. This data further refinesthe window of FGF activity necessary for late myocardial addition andimplicates mef2cb as a target of FGF signaling. At 24 and 28 hpf, weobserved proliferation in cells localized in the arterial pole at 48 hpf. Assprouty4, expressed in response to FGF signaling, is expressedimmediately posterior to the ventricle at 24 hpf (de Pater et al., 2009),it will be interesting to see if and when FGF regulated proliferation hasan effect on late myocardial addition.

The existence of a SHF-like population of cells in the zebrafishembryo has important evolutionary implications. Given the criticalrole of the SHF in the patterning and morphogenesis of the four-chambered mouse and human heart, it has been an open question asto how the functionality of the SHF first arose. If a population of cellswith SHF-like characteristics was already present in the sharedancestor of teleosts and terrestrial vertebrates, it becomes easy toimagine scenarios where this pre-existing population of cells canbe readily recruited and utilized in novel ways to impart newfunctionality and structures to the heart. Zebrafish mef2cbmorphantshave a shrunken bulbus arteriosus, possibly linking late myocardialaddition to bulbus arteriosus development. This may represent anearly function of these SHF-like cells: linking the developing heart tothe circulatory system of the animal. The zebrafish model has becomea favoured tool to study heart development and disease in the pastdecade (Lieschke and Currie, 2007). Chemical genetic approachesin zebrafish models of heart disease, for example, hold promisefor discovering potential pharmaceutical agents (Peal et al., 2010).Given the critical role of the SHF in CHDs, the presence of a SHF-likepopulation in zebrafish validates the zebrafish model for theseapproaches. However, the characteristics of these cells relative tothe mammalian SHF clearly need to be considered for these studies.

Supplementarymaterials related to this article can be found onlineat doi:10.1016/j.ydbio.2011.03.028.

Acknowledgments

The Tg(kdrl:nlsKikGR) transgenic line was generated by Nana Bit-Avragim. We thank Angela Morley for expert zebrafish care, andmembers of the Scott lab for thoughtful discussions and feedback onthis project. We thank Maria Bergen for copyediting. Funding for thisproject was provided by the SickKids Research Training Centre(studentship to S.L.) and the Canadian Institutes for Health Research(studentship funding to S.L. and grant funding to I.C.S.).

References

Beis, D., Bartman, T., Jin, S.W., Scott, I.C., D'Amico, L.A., Ober, E.A., Verkade, H., Frantsve, J.,Field, H.A., Wehman, A., Baier, H., Tallafuss, A., Bally-Cuif, L., Chen, J.N., Stainier, D.Y.,Jungblut, B., 2005. Genetic and cellular analyses of zebrafish atrioventricular cushionand valve development. Development 132, 4193–4204.

Bodmer, R., 1993. The gene tinman is required for specification of the heart and visceralmuscles in Drosophila. Development 118, 719–729.

Brade, T., Gessert, S., Kuhl, M., Pandur, P., 2007. The amphibian second heart field:Xenopus islet-1 is required for cardiovascular development. Dev. Biol. 311,297–310.

Buckingham, M., Meilhac, S., Zaffran, S., 2005. Building the mammalian heart from twosources of myocardial cells. Nat. Rev. Genet. 6, 826–835.

Cai, C.L., Liang, X., Shi, Y., Chu, P.H., Pfaff, S.L., Chen, J., Evans, S., 2003. Isl1 identifies acardiac progenitor population that proliferates prior to differentiation andcontributes a majority of cells to the heart. Dev. Cell 5, 877–889.

Chen, J.N., Fishman, M.C., 1996. Zebrafish tinman homolog demarcates the heart fieldand initiates myocardial differentiation. Development 122, 3809–3816.

Choi, J., Dong, L., Ahn, J., Dao, D., Hammerschmidt, M., Chen, J.N., 2007. FoxH1 negativelymodulates flk1 gene expression and vascular formation in zebrafish. Dev. Biol. 304,735–744.

Collins, R.T., Linker, C., Lewis, J., 2010. MAZe: a tool for mosaic analysis of gene functionin zebrafish. Nat. Methods 7, 219–223.

de la Cruz, M.V., Sanchez Gomez, C., Arteaga, M.M., Arguello, C., 1977. Experimentalstudy of the development of the truncus and the conus in the chick embryo. J. Anat.123, 661–686.

de Pater, E., Clijsters, L., Marques, S.R., Lin, Y.F., Garavito-Aguilar, Z.V., Yelon, D., Bakkers,J., 2009. Distinct phases of cardiomyocyte differentiation regulate growth of thezebrafish heart. Development 136, 1633–1641.

Dodou, E., Verzi, M.P., Anderson, J.P., Xu, S.M., Black, B.L., 2004. Mef2c is a directtranscriptional target of ISL1 and GATA factors in the anterior heart field duringmouse embryonic development. Development 131, 3931–3942.

Dyer, L.A., Kirby, M.L., 2009. The role of secondary heart field in cardiac development.Dev. Biol. 336, 137–144.

Ghosh, T.K., Song, F.F., Packham, E.A., Buxton, S., Robinson, T.E., Ronksley, J., Self, T.,Bonser, A.J., Brook, J.D., 2009. Physical interaction between TBX5 and MEF2C isrequired for early heart development. Mol. Cell. Biol. 29, 2205–2218.

Grimes, A.C., Stadt, H.A., Shepherd, I.T., Kirby, M.L., 2006. Solving an enigma: arterialpole development in the zebrafish heart. Dev. Biol. 290, 265–276.

Grimes, A.C., Duran, A.C., Sans-Coma, V., Hami, D., Santoro, M.M., Torres, M., 2010.Phylogeny informs ontogeny: a proposed common theme in the arterial pole of thevertebrate heart. Evol. Dev. 12, 552–567.

Haberland, M., Arnold, M.A., McAnally, J., Phan, D., Kim, Y., Olson, E.N., 2007. Regulationof HDAC9 gene expression by MEF2 establishes a negative-feedback loop in thetranscriptional circuitry of muscle differentiation. Mol. Cell. Biol. 27, 518–525.

Horsthuis, T., Christoffels, V.M., Anderson, R.H., Moorman, A.F., 2009. Can recentinsights into cardiac development improve our understanding of congenitallymalformed hearts? Clin. Anat. 22, 4–20.

Huang, C.J., Tu, C.T., Hsiao, C.D., Hsieh, F.J., Tsai, H.J., 2003. Germ-line transmission of amyocardium-specific GFP transgene reveals critical regulatory elements in thecardiac myosin light chain 2 promoter of zebrafish. Dev. Dyn. 228, 30–40.

Hutson, M.R., Zeng, X.L., Kim, A.J., Antoon, E., Harward, S., Kirby, M.L., 2010. Arterial poleprogenitors interpret opposing FGF/BMP signals to proliferate or differentiate.Development 137, 3001–3011.

Ieda, M., Fu, J.D., Delgado-Olguin, P., Vedantham, V., Hayashi, Y., Bruneau, B.G.,Srivastava, D., 2010. Direct reprogramming of fibroblasts into functional cardio-myocytes by defined factors. Cell 142, 375–386.

Kawakami, K., 2005. Transposon tools and methods in zebrafish. Dev. Dyn. 234,244–254.

Kelly, R.G., Brown, N.A., Buckingham, M.E., 2001. The arterial pole of the mouse heartforms from Fgf10-expressing cells in pharyngeal mesoderm. Dev. Cell 1, 435–440.

Kimmel, C.B., Warga, R.M., Schilling, T.F., 1990. Origin and organization of the zebrafishfate map. Development 108, 581–594.

Lieschke, G.J., Currie, P.D., 2007. Animal models of human disease: zebrafish swim intoview. Nat. Rev. Genet. 8, 353–367.

Lin, Q., Schwarz, J., Bucana, C., Olson, E.N., 1997. Control of mouse cardiacmorphogenesis and myogenesis by transcription factor MEF2C. Science 276,1404–1407.

Lin, Q., Lu, J., Yanagisawa, H., Webb, R., Lyons, G.E., Richardson, J.A., Olson, E.N., 1998.Requirement of the MADS-box transcription factor MEF2C for vascular develop-ment. Development 125, 4565–4574.

Lints, T.J., Parsons, L.M., Hartley, L., Lyons, I., Harvey, R.P., 1993. Nkx-2.5: a novel murinehomeobox gene expressed in early heart progenitor cells and their myogenicdescendants. Development 119, 969.

Mann, T., Bodmer, R., Pandur, P., 2009. The Drosophila homolog of vertebrate Islet1 is akey component in early cardiogenesis. Development 136, 317–326.

Meilhac, S.M., Esner, M., Kelly, R.G., Nicolas, J.F., Buckingham, M.E., 2004. The clonalorigin of myocardial cells in different regions of the embryonic mouse heart. Dev.Cell 6, 685–698.

Mjaatvedt, C.H., Nakaoka, T., Moreno-Rodriguez, R., Norris, R.A., Kern, M.J., Eisenberg,C.A., Turner, D., Markwald, R.R., 2001. The outflow tract of the heart is recruitedfrom a novel heart-forming field. Dev. Biol. 238, 97–109.

Park, E.J., Watanabe, Y., Smyth, G., Miyagawa-Tomita, S., Meyers, E., Klingensmith, J.,Camenisch, T., Buckingham, M., Moon, A.M., 2008. An FGF autocrine loop initiatedin second heart field mesoderm regulates morphogenesis at the arterial pole of theheart. Development 135, 3599–3610.

Peal, D.S., Peterson, R.T., Milan, D., 2010. Small molecule screening in zebrafish.J. Cardiovasc. Transl. Res. 3, 454–460.

Rochais, F., Mesbah, K., Kelly, R.G., 2009. Signaling pathways controlling second heartfield development. Circ. Res. 104, 933–942.

Sato, M., Yost, H.J., 2003. Cardiac neural crest contributes to cardiomyogenesis inzebrafish. Dev. Biol. 257, 127–139.

133S. Lazic, I.C. Scott / Developmental Biology 354 (2011) 123–133

Srivastava, D., 2006. Making or breaking the heart: from lineage determination tomorphogenesis. Cell 126, 1037–1048.

Stainier, D.Y., Lee, R.K., Fishman,M.C., 1993. Cardiovascular development in the zebrafish. I.Myocardial fate map and heart tube formation. Development 119, 31–40.

Sun, L., Tran, N., Liang, C., Tang, F., Rice, A., Schreck, R., Waltz, K., Shawver, L.K.,McMahon, G., Tang, C., 1999. Design, synthesis, and evaluations of substituted 3-[(3- or 4-carboxyethylpyrrol-2-yl)methylidenyl]indolin-2-ones as inhibitors ofVEGF, FGF, and PDGF receptor tyrosine kinases. J. Med. Chem. 42, 5120–5130.

Thisse, C., Thisse, B., 2008. High-resolution in situ hybridization to whole-mountzebrafish embryos. Nat. Protoc. 3, 59–69.

Ticho, B.S., Stainier, D.Y., Fishman, M.C., Breitbart, R.E., 1996. Three zebrafish MEF2genes delineate somitic and cardiac muscle development in wild-type and mutantembryos. Mech. Dev. 59, 205–218.

Tirosh-Finkel, L., Zeisel, A., Brodt-Ivenshitz, M., Shamai, A., Yao, Z., Seger, R., Domany, E.,Tzahor, E., 2010. BMP-mediated inhibition of FGF signaling promotes cardiomyo-cyte differentiation of anterior heart field progenitors. Development 137,2989–3000.

Torgersen, J.S., Takle, H., Andersen, O., 2010. Differential spatial expression of mef2paralogs during cardiac development in Atlantic cod (Gadus morhua). Comp.Biochem. Physiol. B Biochem. Mol. Biol. 158, 181–187.

Tsutsui, H., Karasawa, S., Shimizu, H., Nukina, N., Miyawaki, A., 2005. Semi-rationalengineering of a coral fluorescent protein into an efficient highlighter. EMBO Rep. 6,233–238.

Urasaki, A., Morvan, G., Kawakami, K., 2006. Functional dissection of the Tol2 transposableelement identified the minimal cis-sequence and a highly repetitive sequence in thesubterminal region essential for transposition. Genetics 174, 639–649.

Waldo, K.L., Kumiski, D.H., Wallis, K.T., Stadt, H.A., Hutson, M.R., Platt, D.H., Kirby, M.L.,2001. Conotruncal myocardium arises from a secondary heart field. Development128, 3179–3188.

Westerfield, M., 1993. The Zebrafish Book: A Guide for the Laboratory Use of ZebrafishDanio (Brachydanio) rerio. University of Oregon Press, Oregon.

Yelon, D., Horne, S.A., Stainier, D.Y., 1999. Restricted expression of cardiac myosin genesreveals regulated aspects of heart tube assembly in zebrafish. Dev. Biol. 214, 23–37.

Zhang, J., Lin, Y., Zhang, Y., Lan, Y., Lin, C., Moon, A.M., Schwartz, R.J., Martin, J.F.,Wang, F., 2008. Frs2{alpha}-deficiency in cardiac progenitors disrupts a subset ofFGF signals required for outflow tract morphogenesis. Development 135,3611–3622.

Zhou, B., Ma, Q., Rajagopal, S., Wu, S.M., Domian, I., Rivera-Feliciano, J., Jiang, D., vonGise, A., Ikeda, S., Chien, K.R., Pu, W.T., 2008. Epicardial progenitors contribute tothe cardiomyocyte lineage in the developing heart. Nature 454, 109–113.