Msx1 and Msx2 regulate survival of secondary heart field precursors and post-migratory proliferation...

08 (2007) 421–437www.elsevier.com/developmentalbiology

Developmental Biology 3

Msx1 and Msx2 regulate survival of secondary heart field precursors andpost-migratory proliferation of cardiac neural crest in the outflow tract

Yi-Hui Chen, Mamoru Ishii, Jingjing Sun, Henry M. Sucov, Robert E. Maxson Jr. ⁎

Department of Biochemistry and Molecular Biology, Norris Comprehensive Cancer Center and Hospital,University of Southern California Keck School of Medicine, 1441 Eastlake Avenue, Los Angeles, CA 90033, USA

Received for publication 20 December 2006; revised 21 May 2007; accepted 29 May 2007Available online 4 June 2007

Abstract

Msx1 and Msx2 are highly conserved, Nk-related homeodomain transcription factors that are essential for a variety of tissue–tissue interactionsduring vertebrate organogenesis. Here we show that combined deficiencies of Msx1 and Msx2 cause conotruncal anomalies associated withmalalignment of the cardiac outflow tract (OFT). Msx1 and Msx2 play dual roles in outflow tract morphogenesis by both protecting secondaryheart field (SHF) precursors against apoptosis and inhibiting excessive proliferation of cardiac neural crest, endothelial and myocardial cells in theconotruncal cushions. During incorporation of SHF precursors into the OFT myocardium, ectopic apoptosis in the Msx1−/−; Msx2−/− mutantSHF is associated with reduced expression of Hand1 and Hand2, which from work on Hand1 and Hand2 mutants may be functionally importantin the inhibition of apoptosis in Msx1/2 mutants. Later during aorticopulmonary septation, excessive proliferation in the OFT cushionmesenchyme and myocardium of Msx1−/−; Msx2−/− mutants is associated with premature down-regulation of p27KIP1, an inhibitor of cyclin-dependent kinases. Diminished accretion of SHF precursors to the elongating OFT myocardium and excessive accumulation of mesenchymal cellsin the conotruncal cushions may work together to perturb the rotation of the truncus arteriosus, leading to OFT malalignment defects includingdouble-outlet right ventricle, overriding aorta and pulmonary stenosis.© 2007 Elsevier Inc. All rights reserved.

Keywords: Msx1, Msx2; Aorticopulmonary septation; Apoptosis; Cardiac neural crest; Conotruncus; Myocardium; Outflow tract malalignment; Proliferation;Secondary heart field

Introduction

The heart is composed of several cell types with distinctlineage origins. These include cells of the primary andsecondary heart fields, which contribute to the myocardium,and cardiac neural crest (NC) cells, which contribute to theaorticopulmonary septum and membranous ventricular septum(Mikawa, 1999). Perturbation of interactions between thesecardiac progenitors can result in congenital heart malforma-tions, the most predominant of which are impaired alignmentand septation of the conotruncal region, or the outflow tract(OFT) (Hoffman and Kaplan, 2002). The cardiac OFT is anintricate structure formed by a series of complex morphogeneticprocesses, including elongation, rotation, alignment and septa-

⁎ Corresponding author. Fax: +1 323 865 0098.E-mail address: [email protected] (R.E. Maxson).

0012-1606/$ - see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.ydbio.2007.05.037

tion. OFT elongation requires accretion of extracardiac cells,including myocardial precursors from the secondary heart field(SHF) and NC cells, to the primitive heart tube, which iscomposed of myocardium and endocardium originating fromthe primary heart field (Kirby, 2002; Restivo et al., 2006). SHFprecursors migrate progressively through the aortic sac andbecome incorporated into the OFT at different times duringcardiac looping (Cai et al., 2003; Kelly et al., 2001; Mjaatvedt etal., 2001; Waldo et al., 2001). NC cells are required forlengthening of the OFT prior to cardiac looping as well asrotation of the OFTafter looping (Bajolle et al., 2006; Restivo etal., 2006; Yelbuz et al., 2002). Elongation of the OFT is aprerequisite for correct cardiac looping as well as completerotation and proper alignment during aorticopulmonaryseptation.

As NC cells colonize the OFT mesenchyme to build up aspiral configuration of the septating conotruncal cushions

mailto:[email protected]

http://dx.doi.org/10.1016/j.ydbio.2007.05.037

Table 1Outflow tract and vascular malformations in Msx1−/−; Msx2−/− doublemutants a

Embryonumber

Age atanalysis

Outflow tractphenotype

Associatedvascular defects

1 E14.5 DORV Pulmonary stenosis2 E14.5 DORV

(transposition type)RERSCA, right dAo

3 E14.5 Overriding aorta4 E15.5 DORV

(transposition type)5 E15.5 DORV Right dAo6 E16.5 Tetralogy of Fallot Pulmonary atresia7 E16.5 PTA-A4 Hypoplastic left

4th aortic arch8 E16.5 DORV

(transposition type)a dAo, dorsal aorta; DORV, double-outlet right ventricle; PTA-A4, persistent

truncus arteriosus type A4 (see text for details); RERSCA, retroesophageal rightsubclavian artery.

422 Y.-H. Chen et al. / Developmental Biology 308 (2007) 421–437

(Fananapazir and Kaufman, 1988; Waldo et al., 1999), the SHF-derived myocardium of the distal OFT must undergo a completerotation to ensure the alignment of the outflow septum anlagewith the interventricular septum (Lomonico et al., 1986; Yasuiet al., 1997). A dramatic increase in the rate of growth of thedistal OFT relative to the heart facilitates the counterclockwiserotation of the great arteries (Lomonico et al., 1986). Incompleteor arrested rotation of the OFT at different developmental stagescauses different OFT malalignment defects including transposi-tion of the great arteries (TGA), double-outlet right ventricle(DORV) and tetralogy of Fallot (TOF) (Bostrom and Hutchins,1988; Lomonico et al., 1986).

Cell proliferation, differentiation and apoptosis in thedeveloping OFT myocardium and cushions are finely tunedto ensure timely remodeling of the OFT (Barbosky et al.,2006; Cai et al., 2003; Kioussi et al., 2002; Martinsen, 2005;Risebro et al., 2006; Trinh et al., 2005; Watanabe et al., 1998).To achieve the exquisite regulation of OFT remodeling,different cell populations in the developing OFT interact anddevelop coordinately through a complex network of signalingmolecules and transcription factors (Brand, 2003; Hutson andKirby, 2003; Restivo et al., 2006; Srivastava, 2006; Srivastavaand Olson, 2000). Although this network is understood inbroad outline, a detailed picture of its components andfunction is lacking.

Msx1 andMsx2, members of the highly conserved Nk familyof homeodomain transcription factors, play important roles intissue–tissue interactions and organogenesis during vertebratedevelopment (Ishii et al., 2003, 2005; Maxson et al., 2003).Single mutants ofMsx1 andMsx2 as well asMsx1−/−; Msx2−/−double mutant mice all exhibit impaired development of cranialneural crest-derived structures (Ishii et al., 2005; Satokata et al.,2000; Satokata and Maas, 1994). Defects in double mutants aremuch more severe than in single mutants (Ishii et al., 2005;Satokata et al., 2000; Satokata and Maas, 1994). Interestingly,while neither Msx1−/− nor Msx2−/− mutant mice have anycardiac defect (Satokata et al., 2000; Satokata and Maas, 1994),Msx1−/−; Msx2−/− double mutant mice display a broad range ofOFT malformations including DORV, TOF and persistenttruncus arteriosus (PTA) (Ishii et al., 2005 and this study),suggesting functional redundancy between Msx1 and Msx2during OFT morphogenesis.

Here we address the roles of Msx1 and Msx2 in OFTmorphogenesis by elucidating the mechanistic basis of OFTmalformations in Msx1−/−; Msx2−/− mutant mice. We foundthat a single allele of Msx1 or Msx2 is sufficient forcardiogenesis. Msx1 or Msx2 is required for the survival of asubpopulation of SHF precursors at the time of migration intothe OFT myocardium. An Msx gene is also required torestrict the expansion of the post-looping OFT, possibly bymaintaining a high level of p27KIP1, an inhibitor of cyclin-dependent kinases, during early aorticopulmonary septation.Our study thus uncovers two distinct roles for Msx1 andMsx2 in OFT morphogenesis, one in the survival of asubpopulation of OFT myocardial precursors in the SHF andthe other in the post-migratory expansion of the conotruncalregion.

Materials and methods

Mouse strains and genotyping

All mice used in this study were maintained in a mixed genetic backgroundof BALB/c and CD-1. Msx1−/− and Msx2−/− single knockout and Msx1−/−;Msx2−/− double knockout mice were described previously (Ishii et al., 2005;Satokata et al., 2000; Satokata and Maas, 1994). The noon copulation plug wascounted as embryonic day 0.5 (E0.5). Genomic DNA was extracted from yolksac (embryos) or tails (postnatal mice) for genotyping. PCR primers andconditions for Msx1 and Msx2 knockout alleles as well as the Wnt1-Cre andR26R transgenes were as described (Jiang et al., 2000; Satokata et al., 2000;Satokata and Maas, 1994).

Cryostat sectioning and section in situ hybridization

E10.5 embryos were prefixed overnight at 4 °C in 4% paraformaldehyde(PFA) in phosphate-buffered saline (PBS) followed by cryopreservation insucrose solutions with increasing concentrations and freezing in OCT (Jiang etal., 2000). Cryostat sections were cut at 10-μm thickness and RNA in situhybridization was performed using the In Situ Hybridization kit (BioChain) withthe protocol modified according to Dijkman et al. (www.roche-applied-science.com/PROD_INF/MANUALS/InSitu/pdf/ISH_181-188.pdf). All RNA probeswere generated as described: Isl1 (Cai et al., 2003), Mef2c (von Both et al.,2004), Hand1 (McFadden et al., 2005), Hand2 (McFadden et al., 2005) andPitx2 (Lu et al., 1999).

LacZ staining

Whole embryos were prefixed for 30 min at room temperature in 4% PFA inPBS and stained with X-gal as described (Jiang et al., 2000). Stained embryoswere postfixed in 4% PFA in PBS overnight at 4 °C, dehydrated through gradedethanol and then embedded in paraffin. Sections were cut at 5-μm thickness,fixed in acetone and counter-stained with nuclear fast red.

Immunostaining and detection of cell proliferation

Cryostat sections were immunostained with the following primaryantibodies: mouse monoclonal anti-α-smooth muscle actin (1:800, Sigma),mouse monoclonal anti-AP-2α (1:4, Developmental Studies Hybridoma Bank)(Zhang et al., 1996), rabbit polyclonal anti-β-galactosidase (1:100, MPBiomedicals), biotinylated goat polyclonal anti-BMP-2/4 (1:100, R&DSystems), mouse monoclonal anti-bromodeoxyuridine (BrdU) (1:100, Sigma),rabbit polyclonal anti-cleaved Caspase 3 (1:100, R&D Systems), mouse

http://www.roche-applied-science.com/PROD_INF/MANUALS/InSitu/pdf/ISH_181-188.pdf

http://www.roche-applied-science.com/PROD_INF/MANUALS/InSitu/pdf/ISH_181-188.pdf

423Y.-H. Chen et al. / Developmental Biology 308 (2007) 421–437

monoclonal anti-Connexin 43 (1:50, Chemicon), goat polyclonal anti-p21CIP1

(1:100, Santa Cruz Biotechnology), mouse monoclonal anti-p27KIP1 (1:100,Zymed), rabbit polyclonal anti-phospho-Smad1/5/8 (1:50, Cell Signaling),guinea pig polyclonal anti-Pitx2 (1:200, a kind gift from Dr. Kioussi) (Kioussi etal., 2002) and mouse anti-proliferating cell nuclear antigen (PCNA) (in theZymed PCNA staining kit).

Immunostaining using anti-PCNA or anti-BrdU antibodies was carriedout to detect cell proliferation. BrdU (Sigma) in 1× PBS was injectedintraperitoneally into the pregnant mouse (200 μg/g of body weight) 2 hprior to embryo harvesting. Harvested embryos were fixed, washed and

Fig. 1. Cardiac outflow tract defects in Msx1/2 double mutant embryos. (A–C) A seembryo illustrate the normal alignment of the pulmonary trunk with the right ventricleventricle by a normal interventricular septum continuous with the conotruncal tissuTable 1) ordered rostral to caudal in the G-D-E sequence indicate persisting right-spulmonary trunk from the right ventricle (double-outlet right ventricle; compare paneThe white arrow indicates the connection between the ascending aorta and the right vventricular septal defect (F) and double-outlet right ventricle with pulmonary stenosseverely reduced size of the pulmonary trunk, which may result from the obstructionembryo 7 reveals the origination of the common truncus arteriosus from the right vecardiac outflow tract. Ao, ascending aorta; ct, conotruncal tissue; IVS, interventricpulmonary trunk; RA and RV, right atrium and ventricle; rdAo, right-sided dorsal aor(for A–C), D (for D, E, G), F (for F, I) and H are 0.5 mm.

cryosectioned as described above followed by immunostaining with theanti-BrdU antibody (Sigma). The Quantity-One program (Bio-Rad) wasused to count cells from all sections of an OFT per embryo (on average 4–5 sections for embryos at E10, 6–7 sections at E11, and 8 sections atE12). CNC cells were labeled by the Wnt1-Cre/R26R two-componenttransgenic reporter and were detected by the presence of β-galactosidase(Jiang et al., 2000). CNC cell proliferation indices were represented by thepercentage of BrdU+/β-galactosidase+ double positive cells in the wholepopulation of β-galactosidase+ cells (i.e., CNC cells). Littermates of eachMsx1−/−; Msx2−/− mutant embryo were used as the controls, and no

ries of transverse sections (ordered rostral to caudal) of one control (Msx1+/−)and the ascending aorta with the left ventricle, which is separated from the righte. (D, E, G) Transverse sections of one Msx1/2 mutant embryo (number 5 inided dorsal aorta (G) as well as the common origin of the ascending aorta andls D and E with panels A and B), although the valves of each vessel are distinct.entricle. (F, I) Transverse sections ofMsx1/2 mutant embryo 1 (Table 1) displayis (I). Note the normal morphology and size of the ascending aorta in spite of aby the aorta (I). (H) Malaligned persistent truncus arteriosus in Msx1/2 mutant

ntricle, illustrating combination of the malalignment and septation defects in theular septum; LA and LV, left atrium and ventricle; pa, pulmonary arteries; PT,ta; TA, truncus arteriosus; VSD, ventricular septal defect. Scale bars in panels A


discernable variation in either gene expression or cell proliferation/survival/differentiation was detected between the controls from the same litter andwith different genotypes.

Results

Cardiac outflow tract defects in mice with combineddeficiencies of Msx1 and Msx2

In our previous study, we noted the presence of conotruncalabnormalities in a small number of Msx1−/−; Msx2−/− doublemutant embryos, but did not evaluate these in detail (Ishii et al.,2005). We have now analyzed eight embryos in total (Table 1).Five Msx1/2 mutant embryos had both OFTs originatingcompletely from the right ventricle (i.e., DORV). Thisphenotype occurs when the OFT does not lengthen sufficientlyto allow the juxtaposition of the OFT with the septatingventricular chambers.

Of these five embryos, in two the aorta and pulmonary trunkwere oriented in a nearly side by side fashion (conventionalDORV; Figs. 1D, E) whereas in the other three the aortic valvewas located left and anterior to the pulmonary valve(transposition type DORV (Bostrom and Hutchins, 1988);data not shown). One embryo of the eight had an overridingaorta, which, compared to DORV, is a less severe manifestationof incomplete OFT elongation.

One embryo exhibited tetralogy of Fallot (TOF). Thepulmonary trunk had regressed due to atresia, and bothpulmonary arteries were supplied retrograde from a normalductus arteriosus originating from the dorsal aorta (data notshown). TOF results from obstruction of the early OFT such thatthe pulmonary channel does not form, and in this case, theobstruction was complete: only a single outflow vessel wasapparently leaving the heart. The diagnosis of TOF as comparedto truncus arteriosus was possible because of the retrogradeorigin of the pulmonary arteries from the dorsal aorta.

In the remaining embryo with PTA, there was a hypoplasticleft 4th aortic arch artery (PTA type A4 (Van Praagh and VanPraagh, 1965)). Interestingly, the PTA was also malalignedsince the truncus originated predominantly from the rightventricle (Fig. 1H).

All eight Msx1/2 null mutants had a membranousventricular septal defect (VSD) (Fig. 1F) due to malorientationof the OFT (conus) septum and subsequent imperfect fusionwith the interventricular (conoventricular) septum (Bostrom

Fig. 2. Section in situ hybridization to analyze marker gene expression in theMsx1/2(D–F) in the pharyngeal endoderm and mesoderm, splanchnic and somatic mesodermthe OFTand right ventricle in wild-type embryos at E9.5 (A, D), E10.5 (B, E) and E11in a subpopulation of myocardial cells in the right ventricle at E10.5 and almost noMexpression in the right ventricular myocardium at E10.5 and E11.5: black arrows inbetween Msx1/2 null mutants and littermate controls in the pharyngeal endoderm aAsterisks indicate Isl1 expression in the motor neurons (Pfaff et al., 1996). (O–V) Sigthe pharyngeal mesoderm (triangle arrowheads), splanchnic and somatic mesoderm (nHand1 only) inMsx1/2 null mutants compared with littermate controls at both E9.5 (Oin the splanchnic and somatic mesoderm (notched arrowheads) compared with the phand E10.5 (T, V). AoS, aortic sac; APS, aorticopulmonary septum; ba, branchial (phventricle; OFT, outflow tract; PM, pharyngeal mesoderm; RA and RV, right atrium a0.2 mm in panels C, F, H (for G–J), L (for K–N) and T (for S–V); 0.1 mm in pane

and Hutchins, 1988; Yasui et al., 1997). Thus, our analysesreveal that the absence of both Msx1 and Msx2 gene productsresults in a range of OFT malformations, all consistent withproblems in OFT elongation and alignment.

Histological analyses revealed normal cardiac developmentin Msx1−/− and Msx2−/− single mutants (n=22) and in Msx1–Msx2 homozygous–heterozygous compound mutants (n=20).Although Msx1−/− mutant pups did not survive to term due tocleft palate, mutant mice of the other genotypes includingMsx1+/−, Msx2+/−, Msx1+/−; Msx2+/−, Msx2−/− andMsx1+/−; Msx2−/− have all been raised to adulthood withoutany apparent cardiac manifestation. Therefore, our analysesindicate that a single allele of either Msx1 or Msx2 is sufficientto support normal cardiogenesis.

Msx1 and Msx2 are required for survival of a subpopulation ofsecondary heart field cells whose normal fate is to beincorporated into the outflow tract myocardium

Since the OFT malalignment defects in Msx1/2 null mutantsare consistent with a disruption of the SHF (Hutson and Kirby,2003; Ward et al., 2005), we sought to address the roles ofMsx1and Msx2 in SHF development. We first examined theexpression patterns of Msx1 and Msx2 in the SHF and OFTbetween embryonic days 9.5 and 11.5 (E9.5–E11.5), spanningthe phases of cardiac looping and aorticopulmonary septation.The patterns of Msx1 and Msx2 expression largely overlappedin most sites, including pharyngeal endoderm and mesoderm,splanchnic mesoderm, somatic mesoderm, epicardium as wellas subpopulations of the OFT myocardium, endocardium andcushion mesenchyme (Figs. 2A–F). This overlap supports thehypothesis that Msx1 and Msx2 are functionally redundant inOFT morphogenesis. In the right ventricle, however, Msx1expression was confined to a subpopulation of myocardial cellsat E10.5 and was significantly reduced at E11.5 compared withMsx2, which spanned a broader range in the ventricularmyocardium (Figs. 2B, C, E, F).

We showed previously that Msx1 and Msx2 are required forthe survival of NC subpopulations in the region of thetrigeminal ganglion and in the first pharyngeal arch (Ishii etal., 2005). To investigate whether Msx1 and Msx2 promote thesurvival of cardiac progenitors, we performed immunostainingfor cleaved Caspase 3 in transverse sections of E9.0–E9.5Msx1−/−; Msx2−/− double mutants and control littermates atthe levels of the second and third pharyngeal arches.

mutant SHF and OFT. (A–F) Overlapping expression ofMsx1 (A–C) andMsx2, epicardium, cardiac neural crest as well as the endocardium and myocardium of.5 (C, F). Black arrows in panels B and C denote the absence ofMsx1 expressionsx1 expression in the right ventricular myocardium at E11.5 (compare withMsx2panels E and F). (G–N) Comparable expression of Isl1 (G–J) and Mef2c (K–N)nd mesoderm, splanchnic and somatic mesoderm as well as OFT myocardium.nificantly decreased expression of Hand1 (O, P, S, T) and Hand2 (Q, R, U, V) inotched arrowheads), cardiac neural crest, OFT myocardium and epicardium (for–R) and E10.5 (S–V). Note a greater degree of reduction ofHand1/2 expressionaryngeal mesoderm (triangle arrowheads) in Msx1/2 mutants at both E9.5 (P, R)aryngeal) arch; CV, common ventricle; FG, foregut; LA and LV, left atrium andnd ventricle; SoM, somatic mesoderm; SpM, splanchnic mesoderm. Scale bars:ls D (for A, D), P (for O, P) and R (for Q, R).


While we did not detect cleaved Caspase 3-positive cells inthe control SHF at E9.0 or E9.5 (red signals in Figs. 3A, C, E,G; green signals in Figs. 3I, K), we did observe apoptotic cellsin the Msx1/2 null mutant SHF. The signal, most extensive atE9.0, spanned distal OFT myocardium, somatic mesodermand a subpopulation of splanchnic mesoderm adjacent topharyngeal mesoderm (green signals in Figs. 3J, L; red signalsin Figs. 3B, D, F, H).

Cai and colleagues (2003) documented extensive ectopicapoptosis of SHF precursors in embryos deficient for the LIMhomeobox transcription factor, Isl1, at the same developmentalstages (Cai et al., 2003). However, apoptotic cells in Isl1 nullmutants were confined to SHF subpopulations distinct fromthose in Msx1/2 null mutants: they were located in pharyngealendoderm and ectoderm as well as in a subpopulation ofsplanchnic mesoderm adjacent to pharyngeal endoderm (Caiet al., 2003).

Consistent with these results, we did not detect a changein the expression of Isl1 or its downstream effector, Mef2c,

Fig. 3. Ectopic apoptosis of a subpopulation of Pitx2-expressing cells in the Msx1/2 mstaining (blue fluorescence). Panels C, D, G, H, K, L, O and P are enlarged views of thJ, M and N, respectively. (A–L) Immunofluorescent staining for cleaved Caspase 3mesoderm at both E9.0 (A–H) and E9.5 (I–L), with more apoptotic cells present at E(green FITC fluorescence) and cleaved Caspase 3 (red Rhodamine fluorescence) revemyocardium of theMsx1/2 null mutant but not in the control (orange merged signalsH). White arrows in panels B and F indicate apoptotic cells in the right SHF and disDouble immunostaining for proliferating cell nuclear antigen (PCNA; red Rhodaminectopic apoptosis in theMsx1/2mutant splanchnic and somatic mesoderm (indicated bpanel J) but no discernable change of cell proliferation in the double mutant SHF or Ogreatly reduced number of Pitx2-expressing cells in the pharyngeal mesoderm, left soMsx1/2 null mutant (indicated by the white arrow and notched arrowheads in panelforegut; LV, left ventricle; OFT, outflow tract; RV, right ventricle; SoM, somatic mesE (for E, F) and I (for I, J); 0.2 mm in panel M (for M, N).

an MADS box transcription factor (Dodou et al., 2004), inMsx1−/−; Msx2−/− mutant embryos (Figs. 2G–N). We alsoanalyzed cell proliferation in the Msx1/2 mutant SHF and OFTat the same developmental stages and did not detect anysignificant change (red signals in Figs. 3I, J).

To confirm that the apoptotic cells in the Msx1/2 mutantsplanchnic mesoderm are indeed SHF precursors whose normalfate is to be added to the elongating OFT, we performed doubleimmunostaining with anti-Pitx2 and anti-cleaved Caspase 3antibodies. Pitx2 is expressed in the left SHF. Its function isrequired for the development of SHF but not cardiac NC cells(Ai et al., 2006; Liu et al., 2002).

We found that apoptotic cells in left splanchnic mesoderm ofMsx1/2 null mutants were all positive for Pitx2 (Figs. 3B, D, F,H). Immunostaining at E10.5 revealed dramatically reducednumbers of Pitx2-positive cells in the pharyngeal mesoderm,left somatic and splanchnic mesoderm as well as the left distalOFT myocardium of Msx1/2 double mutants (Figs. 3M–P anddata not shown), indicating that Pitx2-expressing cells in these

utant SHF and distal OFT. (A–P) All nuclei were visualized by DAPI counter-e boxes focusing on the splanchnic and somatic mesoderm in panels A, B, E, F, I,demonstrates ectopic apoptosis in the Msx1/2 mutant splanchnic and somatic9.0 (compare panels D and H with L). (A–H) Double immunostaining for Pitx2als that Pitx2-expressing cells undergo apoptosis in the left SHF and distal OFTindicated by white notched arrowheads in panels B and F; see also panels D andtal OFT myocardium of Msx1/2 null mutants, which do not express Pitx2. (I–L)e fluorescence) and cleaved Caspase 3 (green FITC fluorescence) demonstratesy the white arrow on the right side and by notched arrowheads on the left side inFT. (M–P) Pitx2 immunostaining (green FITC fluorescence) at E10.5 reveals a

matic and splanchnic mesoderm as well as the left distal OFT myocardium of theN) compared with the control (M). AoS, aortic sac; CV, common ventricle; FG,oderm; SpM, splanchnic mesoderm. Scale bars: 0.1 mm in panels A (for A, B),

Fig. 4. Altered looping configuration of Msx1/2 double mutant hearts. (A–D)Lateral views of control and Msx1/2 null mutant hearts at E9.5 reveal that thedextral cardiac looping in double mutants is less convoluted compared withcontrol hearts (indicated by white dashed curves), leading to diminished spacebetween inflow and outflow limbs in double mutants. In addition, the width ofthe common ventricle (indicated by red dashed lines) is relatively narrow inMsx1/2 null mutants compared with controls probably because of ventraldisplacement of the ventricle resulting from a shortened and cranially displacedOFT. (E–H) Ventrolateral views of control and Msx1/2 null mutant hearts atE10.5 demonstrate more obvious cranial displacement of the right ventricle andventral displacement of the left ventricle, concomitant with an apparent cranialand rightward displacement of the cardiac apex (red asterisks). The red dashedcurves indicate the direction of displacement of the whole heart in Msx1/2mutants resulting from a shortened and cranially displaced OFT. ba, branchial(pharyngeal) arch; CV, common ventricle; LV, left ventricle; M, mandible; OFT,outflow tract; RV, right ventricle; T, tail. Scale bars in panels A (for A–D), E (forE, F) and G (for G, H) are 0.2 mm.


regions died at an earlier developmental stage. In contrast, thelevel of Pitx2 in the Msx1/2 mutant right ventricular myo-cardium was normal (Figs. 3M, N), in agreement with thepossibility that myocardial cells in the Msx1/2 mutant proximalOFT and common ventricle were not dying at E9.0 or E9.5(Figs. 3B, F, J). Therefore, Msx1 or Msx2 is required for thesurvival of a subpopulation of SHF precursors that areultimately incorporated into the myocardium of the OFT butnot of the right ventricle.

To identify potential Msx1/2 downstream targets that mightplay a role in inhibiting ectopic apoptosis of cardiogenicprecursors, we first focused on the expression of Hand genes inMsx1/2 null mutants. Hand genes were of interest for tworeasons. First, the expression patterns of Hand1 and Hand2overlap with those of Msx1 and Msx2 in pharyngeal arches andthe heart (McFadden et al., 2005; Togi et al., 2004; Yanagisawaet al., 2003). Second, a Hand-related gene was shown to inhibitapoptosis of cardioblasts and pericardial cells in Drosophila(Han et al., 2006).

We found that both Hand1 and Hand2 exhibited signifi-cantly reduced expression in the Msx1/2 mutant somaticmesoderm, SHF and OFT (Figs. 2O–V). Interestingly, ectopicapoptosis in the Msx1/2 mutant SHF was predominant in theregions with a greater reduction of Hand1 and Hand2expression (compare Figs. 2P, R with Figs. 3B, F, J).

Since ectopic apoptosis of OFT myocardial precursors mightbe associated with a shortened OFT, we next asked if there wasany cardiac looping defect inMsx1−/−; Msx2−/−mutant hearts.All Msx1/2 null mutants analyzed at E9.5 (n=9) displayedcorrect dextral cardiac looping, and there was no significantalteration in the length of the distal OFT (Anderson et al., 2003)(data not shown). Nonetheless, in comparison with controlhearts, Msx1/2 mutant hearts appeared less convoluted and hada reduced space of the inner curvature between the inflow andoutflow limbs, features that were demonstrated in embryos witha shortened OFT (Yelbuz et al., 2002) (Figs. 4A–D). Inaddition, the Msx1/2 mutant ventricles were slightly displacedventrally and were relatively narrow when compared with thecontrol ventricles. This set of features was also characteristic ofa shortened OFT (Yelbuz et al., 2002) (Figs. 4A–D).

At E10.5, there was an even more severe cranial displace-ment of the OFT and ventral displacement of ventricles inMsx1/2 null mutants, which was accompanied by an apparentrightward turn of the mutant cardiac apex (Figs. 4E–H).Therefore, reduced accretion of SHF precursors into theMsx1/2mutant OFT was associated with an altered looping configura-tion that could contribute to an abnormal alignment of the OFT.

Msx1 and Msx2 negatively regulate post-migratory cellproliferation in the outflow tract during aorticopulmonaryseptation

Previously we documented a delay in NC migration intothe pharyngeal region of E9.5 Msx1/2 null mutants (Ishii etal., 2005). We also showed that overexpression of Msx2 isassociated with a reduced contribution of NC to the OFT ofthe Splotch mutant mouse (Conway et al., 1997; Kwang et

al., 2002). To further investigate the effect of loss of functionof Msx1 and Msx2 on the contribution of NC to the septatingOFT, we traced the colonization of NC cells in the Msx1/2mutant conotruncal regions during aorticopulmonary septationusing the Wnt1-Cre/R26R system. Transverse sections across


E10.5 conotruncal regions revealed comparable numbers ofNC cells colonizing the OFT cushions of Msx1/2 null mutantsand littermate controls (Figs. 5A, B). However, at E12.5,significantly increased numbers of NC cells were evident inthe OFT cushion mesenchyme of Msx1/2 null mutants,indicating a dramatic expansion of the NC population inconotruncal cushions after its migration into the OFT wascompleted (Figs. 5C–J).

To assess whether reduced apoptosis or increased prolifera-tion contributes to the overpopulation of post-migratory cardiacNC cells, we performed TUNEL and BrdU double labeling. Wedetected only sparse apoptotic signals and no significantdifference in the number of apoptotic cells between Msx1/2null mutants and littermate controls (n=4; Figs. 6A–F). Incontrast, BrdU labeling indices were significantly higher inMsx1/2 null mutants within the OFT region (n=4, Pb0.05,Student's t test; Figs. 6A–F, M; and Table 2). Following theelevated proliferation, the total number of cells was increasedsignificantly within the Msx1/2 mutant OFT at E12.0 (n=4,Pb0.05, Student's t test; Table 2). In the SHF and 4–6thpharyngeal arches, however, the level of cell proliferation wascomparable between controls and double mutants at alldevelopmental stages examined (Figs. 3I, J and 6A–F, anddata not shown). Therefore, aberrantly elevated cell prolifera-tion in Msx1/2 null mutant hearts was confined to the OFT.

To further determine whether cardiac NC cells in the Msx1/2 mutant OFT undergo excessive post-migratory proliferation,we performed double immunostaining to colocalize lacZ-expressing NC cells and BrdU-labeled proliferating cells inembryos carrying both the Wnt1-Cre and R26R reportertransgenes. In Msx1/2 mutant OFTs, the increased proliferationindices were statistically significant for NC cells at alldevelopmental stages examined (n=4, Pb0.05, Student'st test; Figs. 6G–L, N, and Table 2), but were statisticallysignificant for non-NC cells at E10.0 and E11.0 only (n=4,Pb0.05 at E10.0 and E11.0, PN0.05 at E12.0, Student'st test; Figs. 6G–L, O, and Table 2). In agreement with thedifferentially increased cell proliferation, we observed sig-nificantly increased total cell numbers for the NC populationbut not for non-NC populations in Msx1/2 mutant OFTs atE12.0 (n=4, Pb0.05 for NC cells, PN0.05 for non-NC cells,Student's t test; Table 2). The results of these quantitativeanalyses are consistent with our observation of excessive NCcells in the Msx1/2 mutant conotruncal cushions at E12.5(Figs. 5C–L). Taken together, our findings indicate (i) thatbetween E10.0 and E12.0 Msx1 and Msx2 negatively regulatecell proliferation in the septating OFT, and (ii) that this effect ispredominantly on the cardiac neural crest.

Fig. 5. Excessive post-migratory expansion of cardiac NC in theMsx1/2 double mutaconotruncal region by LacZ staining on transverse sections of mouse embryos carryamounts of LacZ-expressing cardiac NC cells colonize the control and Msx1/2 doubexcess of cardiac NC cells accumulates in the conotruncal tissue separating the ascendpanels C–F and panels G–J are from two different pairs of Msx1/2 double mutants apanel H, the excessive accumulation of mesenchymal cells surrounding the pulmonpulmonary OFT and lead to pulmonary stenosis. Ao, ascending aorta; AoS, aortic sacatrium and ventricle; OFT, outflow tract; PT, pulmonary trunk; RA and RV, right atriumC–F) and G (for G–J).

Since Msx2 was shown to regulate the differentiation ofmesoangioblast stem cells into smooth muscle cells (Brunelliet al., 2004), and since some neural crest cells normallydifferentiate into smooth muscle cells prior to the formation ofthe aorticopulmonary septum (Beall and Rosenquist, 1990), wetested whether differentiation of cardiac NC cells into smoothmuscle cells was perturbed in the Msx1−/−; Msx2−/− mutantOFT. At both E11.5 and E12.5, immunostaining for α-smoothmuscle actin (α-SMA) revealed a normal level of α-SMAsurrounding the left and right 4th arch arteries and the ductusarteriosus as well as in the OFT of Msx1−/−; Msx2−/− mutants(Figs. 7A–D). These results indicate that cardiac NC cellsdifferentiated normally into smooth muscle cells in Msx1/2 nullmutants.

Premature down-regulation of p27KIP1 and increased Bmp2/4signaling in Msx1−/−; Msx2−/− mutant outflow tracts duringaorticopulmonary septation

To unravel the molecular mechanisms leading to increasedcell proliferation in Msx1−/−; Msx2−/− double mutant OFTs,we asked whether deficiencies of Msx1 and Msx2 alter theactivity of specific growth control genes in the septating OFT.Previous studies have demonstrated that both Msx1 and Msx2regulate the G1/S transition of the cell cycle in culturedmesenchymal and epithelial progenitor cells by activatingCyclin D1 expression and Cdk4 activity (Hu et al., 2001), whileMsx1 also down-regulates the expression of p19INK4d, aninhibitor of cyclin-dependent kinases (CDKs), in the developingdental mesenchyme (Han et al., 2003a). Because the INK familyof CDK inhibitors (CKIs) is not expressed in the developingheart (Koh et al., 1998; Quelle et al., 1995), we asked whetherthe CIP/KIP family of CKIs was regulated by Msx1 and Msx2in the cardiac OFT.

Immunostaining revealed that p21CIP1 levels were similar incontrol and Msx1/2 null mutant OFTs from E10.0 to E12.0(Figs. 8A–F and data not shown), whereas p27KIP1 levels weresignificantly lower in Msx1/2 mutant OFTs at both E10.0 andE11.0 (n=4, Pb0.05, Student's t test; Figs. 8G–J). We alsonoticed a much greater reduction in p27KIP1 levels in NC cellsthan in non-NC cells inMsx1/2 mutant OFTs (n=4, Pb0.01 forNC cells, 0.01bPb0.05 for non-NC cells, Student's t test; Figs.8M, N), corresponding to our observation of a greater increaseof proliferation rates in NC cells than in non-NC cells (Figs. 6N,O). Interestingly, in control OFTs, the p27KIP1 level declineddramatically between E11.0 and E12.0, consistent with anincrease of proliferation rates from E11.0 to E12.0 (Figs. 6M–Oand Table 2).

nt conotruncal cushions. (A–J) Demonstration of cardiac NC colonization in theing the Wnt1-Cre and R26R reporter transgenes. (A, B) At E10.5, comparablele mutant OFT cushions (indicated by black arrows). (C–J) At E12.5, a massiveing aorta and the right ventricle inMsx1/2 null mutants (indicated by red arrows;nd littermate controls) and disrupts conotruncal alignment and organization. Inary OFT will likely obstruct the connection between the right ventricle and the; CNC, cardiac neural crest; DA, ductus arteriosus; FG, foregut; LA and LV, leftand ventricle. Scale bars: 0.2 mm in panel A (for A, B); 0.4 mm in panels C (for

Fig. 6. Increased post-migratory cell proliferation and unaltered cell survival in the Msx1/2 double mutant OFT during aorticopulmonary septation. All nuclei were visualized by DAPI counter-staining (bluefluorescence). Panels M–O are statistical charts of proliferation indices obtained from data analyses in panels A–L. (A–F) BrdU (red Rhodamine fluorescence) and TUNEL (green FITC fluorescence) double labeling tolocalize both proliferating and apoptotic cells in the transverse sections of control and Msx1/2 mutant OFTs between E10.0 and E12.0. TUNEL-positive signals are sparse in both control and Msx1/2 mutant sections,indicating low levels of cell apoptosis during these developmental stages. Compared with control OFTs, however, there is a dramatic increase in the number of BrdU-positive cells inMsx1/2mutant OFTs (white arrows)at both E10.0 (A, B) and E11.0 (C, D), while the increase at E12.0 (E, F) is milder but still reaches statistical significance (panel M; n=4, Pb0.02 at E10.0, Pb0.01 at E11.0 and Pb0.05 at E12.0, Student's t test; seealso Table 1). (G–L) BrdU (red Rhodamine signals) and β-galactosidase (green FITC signals) double immunostaining to localize proliferating cardiac NC cells (yellow merged signals). The percentages of BrdU-positiveNC cells (yellow) and BrdU-positive non-NC cells (red) are both significantly higher inMsx1/2mutant OFTs compared with control OFTs at E10.0 (G, H) and E11.0 (I, J) (white arrows), whereas at E12.0 (K, L), onlythe percentage of BrdU-positive NC cells is significantly higher in theMsx1/2mutant OFT (panels N and O; n=4, Pb0.05 for NC cells at all developmental stages, whereas Pb0.05 for non-NC cells at E10.0 and E11.0but not E12.0, Student's t test; see also Table 1). Notched arrowheads in panels A–L indicate comparable levels of cell proliferation in the pharyngeal, splanchnic and somatic mesoderm between controls and Msx1/2double mutants. AoS, aortic sac; CNC, cardiac neural crest; LA and LV, left atrium and ventricle; RA and RV, right atrium and ventricle. Scale bars in panels B (for A, B), D (for C, D), F (for E, F), H (for G, H), J (for I, J)and L (for K, L) are 0.1 mm.

430Y.-H

.Chen

etal.

/Developm

entalBiology

308(2007)

421–437

Table 2Summary of total cell numbers and proliferation indices in control and Msx1−/−; Msx2−/− double mutant outflow tracts between E10.0 and E12.0 a

Developmentalstage

Cellpopulationin theOFT

Controls Msx1−/−; Msx2−/− double mutants

Averagetotal cellnumber

AverageBrdU+ cellnumber

Averageproliferationindex (%)

Averagetotal cellnumber

AverageBrdU+ cellnumber

Averageproliferationindex (%)

E10.0 All cells 1561 555 35.55 1668 (P=0.448) 928 55.64% (P=0.014)NC cells 998 341 34.17 1047 (P=0.415) 611 58.36% (P=0.008)Non-NC cells b 563 214 38.01 621 (P=0.502) 317 51.05% (P=0.023)

E11.0 All cells 7117 2646 37.18 7564 (P=0.216) 4767 63.02% (P=0.009)NC cells 4985 1768 35.47 5316 (P=0.109) 3503 65.90% (P=0.001)Non-NC cells b 2132 878 41.18 2248 (P=0.323) 1264 56.23% (P=0.044)

E12.0 All cells 10410 5868 56.37 13,227 (P=0.039) 8618 65.15% (P=0.046)NC cells 7445 4293 57.66 10,750 (P=0.021) 7205 67.02% (P=0.033)Non-NC cells b 2965 1575 53.12 2477 (P=0.245) 1413 57.04% (P=0.214)

a All P values were calculated using Student's t test. Significant P values (Pb0.05) are in bold face.b Non-NC cells include myocardial and endocardial cells in the outflow tract.


Our findings demonstrate that loss of Msx1 and Msx2 resultsin premature down-regulation of p27KIP1 levels in the septatingOFT before E12.0. This is correlated with significantly elevatedcell proliferation in the Msx1/2 mutant OFT between E10.0 andE11.0, which could cause excessive accumulation of cells in theconotruncal cushions by E12.0.

We did not detect any significant difference in p27KIP1 levelsbetweenMsx1/2 null mutants and their littermate controls in theSHF and 4–6th pharyngeal arches (Figs. 8G–L and data notshown), consistent with our failure to detect a difference in

Fig. 7. Normal differentiation of cardiac NC cells into smooth muscle cells inMsx1/2 double mutant hearts and vasculatures. (A–D) Immunostaining for α-smooth muscle actin (α-SMA) reveals that, at both E11.5 (A, B) and E12.5(C, D), numbers of α-SMA-positive cells are comparable between Msx1/2mutants and littermate controls surrounding the vasculature (the left and rightfourth arch arteries at E11.5, and the ductus arteriosus, aorta and pulmonarytrunk at E12.5, as indicated by notched arrowheads), at the distal end of the OFT,where the aorticopulmonary septum lies (indicated by triangle arrowheads), aswell as the proximal conotruncal cushions, where the OFT valves and theintervalvular fibrous connection will form (indicated by black arrows). Ao,aorta; APS, aorticopulmonary septum; DA, ductus arteriosus; dAo, dorsal aorta;L4, left 4th arch artery; LA and LV, left atrium and ventricle; OFT, outflow tract;R4, right 4th arch artery; RA and RV, right atrium and ventricle. Scale bars inpanels B (for A, B) and D (for C, D) are 0.2 mm.

proliferation rates between Msx1/2 mutants and controls in theSHF and pharyngeal arches (Figs. 3I, J and 6A–F, and data notshown). We also noticed a relatively low number of p27KIP1-expressing cells in the 4–6th pharyngeal arches between E10.0and E12.0 (Figs. 8G–L and data not shown), which suggeststhat p27KIP1 functions primarily within the OFT duringaorticopulmonary septation.

Because no cardiovascular defects have been reported inp27KIP1 null mutant mice (Fero et al., 1996; Kiyokawa et al.,1996; Nakayama et al., 1996), we reasoned that other genes mayact together with p27KIP1 to regulate post-migratory cellproliferation in the OFT. Bone morphogenetic proteins (Bmps)are potential candidates since they have been implicated in OFTmorphogenesis (Cai et al., 2003; Delot et al., 2003; Kaartinen etal., 2004; Kim et al., 2001; Liu et al., 2004; Stottmann et al.,2004), and Bmp2, Bmp4 and Bmp7 have all been described asputative downstream targets ofMsx1 orMsx2 (Han et al., 2003b;Ishii et al., 2005; Wu et al., 2003; Zhang et al., 2002, 2003).

Immunostaining for Bmp2/4 and phosphorylated Smad1/5/8revealed a statistically significant elevation of Bmp signaling inMsx1−/−; Msx2−/− mutant OFTs compared with control OFTsat both E10.5 and E11.5 (n=3, Pb0.05, Student's t test; Figs.9A–J). As Bmp4 is known to play a role in promoting localproliferation of NC cells after they reached the OFT (Liu et al.,2004), it is possible that upregulation of Bmp2/4 signaling anddown-regulation of p27KIP1 work together to facilitate excessiveproliferation in theMsx1/2mutant conotruncal cushions (Fig. 10).

Discussion

Here we show that Msx1 and Msx2 exhibit overlappingfunctions in both early and late stages of the morphogenesis ofthe cardiac outflow tract. Msx1 and Msx2 seem to displaycomplete functional redundancy in cardiogenesis as evidencedby the sufficiency of a single Msx allele for normal heartdevelopment. In addition,Msx1 andMsx2 exert different effects(inhibit apoptosis and proliferation) on distinct cell populations(SHF precursors and cardiac NC cells) during early and lateOFT morphogenesis. Our findings demonstrate requirementsfor Msx1 and Msx2 in two major aspects of OFT development,

Fig. 8. Premature down-regulation of the CDK inhibitor p27KIP1 in the Msx1/2 mutant OFT during aorticopulmonary septation. (A–F) Immunostaining reveals comparable levels of the CDK inhibitor p21CIP1 in theseptating OFT betweenMsx1/2 double mutants and littermate controls at all developmental stages examined: E10.0 (A, B), E11.0 (C, D) and E12.0 (E, F). (G–L) Double immunostaining to colocalize p27KIP1-positivecells (red Rhodamine fluorescence) and LacZ-expressing cardiac NC cells (green FITC fluorescence) demonstrates greatly reduced numbers of p27KIP1-positive NC cells (yellow merged signals) in the Msx1/2 doublemutant OFTcushions compared with the control at both E10.0 (G, H) and E11.0 (I, J) (indicated by white arrows; n=4, Pb0.01, Student's t test). Numbers of p27KIP1-positive non-NC cells also significantly decreased inMsx1/2 mutant OFTs at both E10.0 (G, H) and E11.0 (I, J) (n=4, 0.01bPb0.05, Student's t test). On the other hand, fewer p27KIP1-positive cells are present in pharyngeal and somatic mesoderm in both controls andMsx1/2 null mutants (indicated by triangle arrowheads). (K, L) At E12.0, the number of p27KIP1-positive cells dramatically reduced in the control OFT (in both NC and non-NC cells) and led to comparable levels ofp27KIP1 between the control OFT and Msx1/2 mutant OFT (indicated by notched arrowheads). (M, N) Statistical charts showing average percentages of p27KIP1-positive cells in cardiac NC and non-NC cells in theseptating OFT between E10.0 and E12.0. (O) A schematic model illustrating that decreased p27KIP1 levels inMsx1/2 null mutant OFTcushions permit more G1/S transitions and promote cell proliferation. AoS, aortic sac;CNC, cardiac neural crest; FG, foregut; LV, left ventricle; RA and RV, right atrium and ventricle. Scale bars in panels B (for A, B), D (for C, D), F (for E, F), H (for G, H), J (for I, J) and L (for K, L) are 0.1 mm.

432Y.-H

.Chen

etal.

/Developm

entalBiology

308(2007)

421–437

Fig. 9. Elevated Bmp2/4 signaling in the Msx1/2 double mutant OFT during aorticopulmonary septation. Immunostaining for Bmp2/4 (green FITC fluorescence)(A–D) and phosphorylated Smad1/5/8 (F–I) reveals significantly increased numbers of Bmp2/4-positive and phospho-Smad1/5/8-positive cells inMsx1/2 null mutantOFTs compared with control OFTs at both E10.5 (A, B, F, G) and E11.5 (C, D, H, I), with a much greater difference at E11.5 (for Bmp2/4: n=3, 0.01bPb0.05 atE10.5, Pb0.01 at E11.5; for phospho-Smad1/5/8: n=3, 0.005bPb0.01 at E10.5, Pb0.005 at E11.5; Student's t test). White arrows indicate increased numbers ofBmp2/4-positive and phospho-Smad1/5/8-positive cells in the Msx1/2 mutant conotruncal cushions (B, D, G, I). Notched arrowheads indicate increased Bmp2/4-positive signals in the double mutant somatic mesoderm at E10.5 (B) as well as increased Bmp2/4-positive and phospho-Smad1/5/8-positive signals in the pharyngealmesoderm, splanchnic mesoderm and myocardium of the OFT and right ventricle in the Msx1/2 mutant at E11.5 (D, I). (E, J) Statistical charts displaying averagepercentages of Bmp2/4-positive and phospho-Smad1/5/8-positive cells in the control and Msx1/2 mutant OFTs at E10.5 and E11.5. AoS, aortic sac; FG, foregut; LAand LV, left atrium and ventricle; OFT, outflow tract; RA and RV, right atrium and ventricle; SoM, somatic mesoderm; SpM, splanchnic mesoderm. Scale bars in panelsB (for A, B), D (for C, D), G (for F, G) and I (for H, I) are 0.1 mm.


one in survival of SHF precursor cells during cardiac loopingand the other in proliferation of OFTcushion mesenchymal cellsduring aorticopulmonary septation (Fig. 10).

Roles of Msx1 and Msx2 in the survival and differentiation ofsecondary heart field precursors and in the elongation of theoutflow tract

All Msx1−/−; Msx2−/− double mutant mice that weexamined had a malaligned OFT (Table 1), consistent with

defects in the elongation or spiral rotation of the OFT (Bajolleet al., 2006; Yelbuz et al., 2002). In Msx1/2 null mutants,increased apoptosis in the SHF was associated with thereduced accretion of SHF cells into the lengthening OFT,similar to the consequence of ablating SHF precursors inchick embryos, which resulted in a shortened OFT and thusled to dextroposed aorta or pulmonary atresia/stenosis (Wardet al., 2005). The presence of Pitx2-expressing cells in theright ventricle and conal region but not the truncal region ofMsx1/2 mutant hearts at E10.5 (Figs. 3M, N) suggests that

Fig. 10. Summary of the roles of Msx1 and Msx2 in OFT morphogenesis.Dashed arrows and lines indicate upregulation or down-regulation of geneexpression or protein levels. Previous studies have shown that: (1) Bmp4 down-regulates Pitx2 expression in the early lateral plate mesoderm, and Bmp4 andPitx2 act antagonistically in the pharyngeal arch mesoderm (Brand, 2003;Branford et al., 2000; Liu et al., 2003; St Amand et al., 2000); (2) Pitx2 specifiesa left cardiac lineage of cells (Campione et al., 2001; Franco and Campione,2003); (3) Hand genes are required for survival, differentiation and migration ofmyocardial precursors (Han et al., 2006; Trinh et al., 2005; Yelon et al., 2000);(4) Bmp4 is required for Pitx2 expression in the OFT (Liu et al., 2004); (5) Pitx2promotes cell type-specific proliferation (Kioussi et al., 2002); (6) p27KIP1

inhibits cell proliferation and p27KIP1-deficient mice exhibit hyperplasia ofmultiple organs and tissues, including the heart (Nakayama et al., 1996;Poolman et al., 1999).


one function of Msx1 and Msx2 in the SHF is to maintain thesurvival of a subset of SHF precursors that becomeincorporated into the distal (truncal) OFT late in cardiaclooping. A flow of migratory SHF precursors along the pathfrom splanchnic mesoderm adjacent to pharyngeal endodermand mesoderm through the aortic sac gives rise to distinctregions of the OFT at different times of development (Cai etal., 2003; Kelly et al., 2001; Mjaatvedt et al., 2001; Waldo etal., 2001).

Isl1 expression is required to maintain the survival of asubset of SHF cells that become incorporated into the OFTearlyin its development (Cai et al., 2003). Our data indicate thatMsx1 and Msx2 are necessary for the survival of anotherdistinct subset of SHF cells that are incorporated into the OFT ata later stage. These two subpopulations of SHF precursors arelocalized in the medial and lateral splanchnic mesoderm, asshown by labeling of apoptotic cells (Figs. 3A, B, E, F, I, J andCai et al., 2003). It is noteworthy, however, that not all SHFcells within the expression domain of Isl1 orMsx1/2 underwentectopic apoptosis. Why this specific subset of SHF cells is moresensitive to Msx function than other SHF cells is unclear, but islikely a result of the combinatorial action of a group of geneswithin this subset of SHF cells. Our findings raise the intriguingpossibility that Hand1/2 genes may be a part of this group:differences between Msx1/2 mutant SHF subpopulations intheir susceptibility to apoptosis appear to be associated withdifferential down-regulation of Hand1 and Hand2 in distinctregions of the Msx1/2 mutant SHF.

Hand1/2 may also function together with Msx1/2 in laterstages of OFT morphogenesis. Inactivation of Hand1 inNkx2.5-expressing cells, which constitute most of the primaryheart field and a subset of SHF cells (Waldo et al., 2001), resultsin OFT malalignment defects (McFadden et al., 2005). SinceHand1 and Hand2 play pivotal roles in the differentiation ofmyocardial precursors and cardiomyocytes (Risebro et al.,2006; Trinh et al., 2005), reduced Hand1/2 expression inMsx1/2 mutant hearts may also perturb the differentiation of SHFprecursors and myocardial cells in the OFT, disrupting OFTmorphogenesis. Against this possibility is our finding that theexpression of the smooth muscle marker, α-SMA, was notdetectably altered in Msx1/2 mutant OFTs (Fig 7). Despite thisresult, further studies analyzing the expression patterns andlocalization of adherens junctional proteins will be required todetermine whether formation of epithelia during myocardialdifferentiation is affected by the loss of Msx1 and Msx2function (Trinh et al., 2005).

Roles of Msx1 and Msx2 in left–right patterning and rotation ofthe outflow myocardium

That left–right patterning is important for the correctrotation and alignment of the OFT myocardium follows fromfindings (i) that deficiencies in laterality genes such as Pitx2cimpair the rotation of the OFT (Bajolle et al., 2006; Gaioet al., 1999; Liu et al., 2002; Oh and Li, 1997), and (ii) thatdisruptions in the incorporation of right SHF precursors intothe left side of preseptation OFT myocardium result inpulmonary stenosis/atresia (Ward et al., 2005). Our resultsshow that in Msx1/2 mutants, a reduction in the number ofPitx2-expressing cells in left splanchnic mesoderm (due toapoptosis) resulted in accretion of fewer Pitx2-expressing cellsto the left side of the distal OFT myocardium by E10.5 (Fig.3). The dramatically reduced number of Pitx2-expressing cellson the left side of the Msx1/2 mutant OFT myocardium wouldlikely diminish left side-specific, Pitx2-mediated signaling(such as Nodal-Cfc signaling (Bamforth et al., 2004; Linask et


al., 2003; Shen et al., 1997)). This may result in insufficientleft–right signaling for the complete spiral rotation of the OFT.Angular analyses of the aortic-to-pulmonary valve axes haveattributed the malaligned OFT in DORV and tetralogy of Fallotto arrests in normal OFT rotation at certain stages ofembryogenesis (Bostrom and Hutchins, 1988; Lomonico etal., 1988). Similarly, a low level of Pitx2-mediated left–rightsignaling in the Msx1/2 mutant OFT may only be sufficient tosupport normal rotation of the septating OFT to a specificdevelopmental stage, leaving the OFT in an intermediateangular position.

Msx1 and Msx2 regulate post-migratory expansion of cardiacneural crest cells as well as endothelial and myocardial cells inthe outflow tract during aorticopulmonary septation

Although in previous work, we observed a delay in NCmigration into the Msx1/2 mutant pharyngeal region at E9.5(Ishii et al., 2005), our findings in this study indicate that normalnumbers of cardiac NC cells colonize OFT cushions of Msx1/2double mutants at E10.5 (Figs. 5A, B). Moreover, our findingsreveal a substantial excess of cardiac NC cells in E12.5 Msx1/2mutant conotruncal cushions (Figs. 5C–J). An increase in theabundance of cardiac NC cells in the conotruncal region hasbeen described in mice overexpressing Connexin 43. Thesemice displayed obstructions of pulmonary outflow tracts similarto the phenotype of pulmonary stenosis or atresia in Msx1−/−;Msx2−/− double mutants (Huang et al., 1998b; Lo et al., 1999;Lo and Wessels, 1998). However, accumulation of excessivecardiac NC cells in the OFTs of Connexin 43 overexpressingmice resulted not from increased proliferation as occurs inMsx1/2 mutants, but from increased rates of NC migration(Huang et al., 1998a).

The previous finding of increased cell proliferation in theventricular myocardium of Connexin 43 overexpressing micesuggested that changes in the abundance or functional activityof cardiac NC cells might perturb the balance of cytokines thatact on cardiomyocyte proliferation (Huang et al., 1998a; Lo etal., 1999). Similarly, the increased proliferation rates in Msx1/2mutant cardiomyocytes in the OFT may be secondary to theincrease in cardiac NC cell proliferation. Potential candidatesfor such local cytokines include Bmp2 and Bmp4, whichexhibited elevated levels of signaling in Msx1/2 mutant OFTs,but unaltered patterns of spatial expression (Fig. 9 and data notshown). A role for Bmp4 in regulating local cell proliferation inthe septating OFT has been demonstrated in mutant micedeficient for Bmp4 in the myocardium (Nkx2.5cre; Bmp4 n/f).These mice displayed significantly reduced cell proliferation inthe conotruncal cushions at E11.5 (Liu et al., 2004). We proposethat increased Bmp2/4 signaling may collaborate with reducedp27KIP1 levels to promote local cell proliferation in Msx1/2mutant OFTs. Furthermore, the resultant excessive accumula-tion of mesenchymal cells in Msx1/2 mutant conotruncalcushions may disrupt the spiral configuration required to formthe spiral aorticopulmonary septum (Fananapazir and Kaufman,1988) and may thus contribute to OFT malalignment andobstruction.

Acknowledgments

This work was supported by NIH grants DE12941 andDE12450 to RM. We would like to thank Bibha Choudhary forher technical assistance and thank Dr. Chrissa Kioussi for thekind gift of anti-Pitx2 antibody. We are also grateful to thefollowing people for cRNA probes: Drs. Sylvia Evans, James F.Martin, Eric N. Olson and Jeffrey L. Wrana.

References

Ai, D., Liu, W., Ma, L., Dong, F., Lu, M.F., Wang, D., Verzi, M.P., Cai, C., Gage,P.J., Evans, S., Black, B.L., Brown, N.A., Martin, J.F., 2006. Pitx2 regulatescardiac left–right asymmetry by patterning second cardiac lineage-derivedmyocardium. Dev. Biol. 296, 437–449.

Anderson, R.H., Webb, S., Brown, N.A., Lamers, W., Moorman, A., 2003.Development of the heart: (3) formation of the ventricular outflow tracts,arterial valves, and intrapericardial arterial trunks. Heart 89, 1110–1118.

Bajolle, F., Zaffran, S., Kelly, R.G., Hadchouel, J., Bonnet, D., Brown, N.A.,Buckingham, M.E., 2006. Rotation of the myocardial wall of the outflowtract is implicated in the normal positioning of the great arteries. Circ. Res.98, 421–428.

Bamforth, S.D., Braganca, J., Farthing, C.R., Schneider, J.E., Broadbent, C.,Michell, A.C., Clarke, K., Neubauer, S., Norris, D., Brown, N.A., Anderson,R.H., Bhattacharya, S., 2004. Cited2 controls left–right patterning and heartdevelopment through a Nodal–Pitx2c pathway. Nat. Genet. 36, 1189–1196.

Barbosky, L., Lawrence, D.K., Karunamuni, G., Wikenheiser, J.C., Doughman,Y.Q., Visconti, R.P., Burch, J.B., Watanabe, M., 2006. Apoptosis in thedeveloping mouse heart. Dev. Dyn. 235, 2592–2602.

Beall, A.C., Rosenquist, T.H., 1990. Smooth muscle cells of neural crest originform the aorticopulmonary septum in the avian embryo. Anat. Rec. 226,360–366.

Bostrom, M.P., Hutchins, G.M., 1988. Arrested rotation of the outflow tract mayexplain double-outlet right ventricle. Circulation 77, 1258–1265.

Brand, T., 2003. Heart development: molecular insights into cardiac specificationand early morphogenesis. Dev. Biol. 258, 1–19.

Branford, W.W., Essner, J.J., Yost, H.J., 2000. Regulation of gut and heart left–right asymmetry by context-dependent interactions between Xenopus leftyand BMP4 signaling. Dev. Biol. 223, 291–306.

Brunelli, S., Tagliafico, E., De Angelis, F.G., Tonlorenzi, R., Baesso, S., Ferrari,S., Niinobe, M., Yoshikawa, K., Schwartz, R.J., Bozzoni, I., Ferrari, S.,Cossu, G., 2004. Msx2 and necdin combined activities are required forsmooth muscle differentiation in mesoangioblast stem cells. Circ. Res. 94,1571–1578.

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

Campione, M., Ros, M.A., Icardo, J.M., Piedra, E., Christoffels, V.M.,Schweickert, A., Blum, M., Franco, D., Moorman, A.F., 2001. Pitx2expression defines a left cardiac lineage of cells: evidence for atrial andventricular molecular isomerism in the iv/iv mice. Dev. Biol. 231, 252–264.

Conway, S.J., Henderson, D.J., Kirby, M.L., Anderson, R.H., Copp, A.J., 1997.Development of a lethal congenital heart defect in the splotch (Pax3) mutantmouse. Cardiovasc. Res. 36, 163–173.

Delot, E.C., Bahamonde, M.E., Zhao, M., Lyons, K.M., 2003. BMP signaling isrequired for septation of the outflow tract of the mammalian heart.Development 130, 209–220.

Dodou, E., Verzi, M.P., Anderson, J.P., Xu, S.M., Black, B.L., 2004. Mef2c is adirect transcriptional target of ISL1 and GATA factors in the anterior heartfield during mouse embryonic development. Development 131, 3931–3942.

Fananapazir, K., Kaufman, M.H., 1988. Observations on the development of theaortico-pulmonary spiral septum in the mouse. J. Anat. 158, 157–172.

Fero, M.L., Rivkin, M., Tasch, M., Porter, P., Carow, C.E., Firpo, E., Polyak, K.,Tsai, L.H., Broudy, V., Perlmutter, R.M., Kaushansky, K., Roberts, J.M.,1996. A syndrome of multiorgan hyperplasia with features of gigantism,


tumorigenesis, and female sterility in p27(Kip1)-deficient mice. Cell 85,733–744.

Franco, D., Campione, M., 2003. The role of Pitx2 during cardiac development.Linking left–right signaling and congenital heart diseases. TrendsCardiovasc. Med. 13, 157–163.

Gaio, U., Schweickert, A., Fischer, A., Garratt, A.N., Muller, T., Ozcelik, C.,Lankes, W., Strehle, M., Britsch, S., Blum, M., Birchmeier, C., 1999. A roleof the cryptic gene in the correct establishment of the left–right axis. Curr.Biol. 9, 1339–1342.

Han, J., Ito, Y., Yeo, J.Y., Sucov, H.M., Maas, R., Chai, Y., 2003a. Cranial neuralcrest-derived mesenchymal proliferation is regulated by Msx1-mediated p19(INK4d) expression during odontogenesis. Dev. Biol. 261, 183–196.

Han, M., Yang, X., Farrington, J.E., Muneoka, K., 2003b. Digit regeneration isregulated by Msx1 and BMP4 in fetal mice. Development 130,5123–5132.

Han, Z., Yi, P., Li, X., Olson, E.N., 2006. Hand, an evolutionarily conservedbHLH transcription factor required for Drosophila cardiogenesis andhematopoiesis. Development 133, 1175–1182.

Hoffman, J.I., Kaplan, S., 2002. The incidence of congenital heart disease.J. Am. Coll. Cardiol. 39, 1890–1900.

Hu, G., Lee, H., Price, S.M., Shen, M.M., Abate-Shen, C., 2001. Msx homeoboxgenes inhibit differentiation through upregulation of cyclin D1. Development128, 2373–2384.

Huang, G.Y., Cooper, E.S., Waldo, K., Kirby, M.L., Gilula, N.B., Lo, C.W.,1998a. Gap junction-mediated cell–cell communication modulates mouseneural crest migration. J. Cell Biol. 143, 1725–1734.

Huang, G.Y., Wessels, A., Smith, B.R., Linask, K.K., Ewart, J.L., Lo, C.W.,1998b. Alteration in connexin 43 gap junction gene dosage impairsconotruncal heart development. Dev. Biol. 198, 32–44.

Hutson, M.R., Kirby, M.L., 2003. Neural crest and cardiovascular development:a 20-year perspective. Birth Defects Res. C Embryo Today 69, 2–13.

Ishii, M., Merrill, A.E., Chan, Y.S., Gitelman, I., Rice, D.P., Sucov, H.M.,Maxson Jr., R.E., 2003. Msx2 and Twist cooperatively control thedevelopment of the neural crest-derived skeletogenic mesenchyme of themurine skull vault. Development 130, 6131–6142.

Ishii, M., Han, J., Yen, H.Y., Sucov, H.M., Chai, Y., Maxson Jr., R.E., 2005.Combined deficiencies of Msx1 and Msx2 cause impaired patterning andsurvival of the cranial neural crest. Development 132, 4937–4950.

Jiang, X., Rowitch, D.H., Soriano, P., McMahon, A.P., Sucov, H.M., 2000. Fateof the mammalian cardiac neural crest. Development 127, 1607–1616.

Kaartinen, V., Dudas, M., Nagy, A., Sridurongrit, S., Lu, M.M., Epstein, J.A.,2004. Cardiac outflow tract defects in mice lacking ALK2 in neural crestcells. Development 131, 3481–3490.

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

Kim, R.Y., Robertson, E.J., Solloway, M.J., 2001. Bmp6 and Bmp7 are requiredfor cushion formation and septation in the developing mouse heart. Dev.Biol. 235, 449–466.

Kioussi, C., Briata, P., Baek, S.H., Rose, D.W., Hamblet, N.S., Herman, T.,Ohgi, K.A., Lin, C., Gleiberman, A., Wang, J., Brault, V., Ruiz-Lozano, P.,Nguyen, H.D., Kemler, R., Glass, C.K., Wynshaw-Boris, A., Rosenfeld,M.G., 2002. Identification of a Wnt/Dvl/beta-Catenin→Pitx2 pathwaymediating cell-type-specific proliferation during development. Cell 111,673–685.

Kirby, M.L., 2002. Molecular embryogenesis of the heart. Pediatr. Dev. Pathol.5, 516–543.

Kiyokawa, H., Kineman, R.D., Manova-Todorova, K.O., Soares, V.C.,Hoffman, E.S., Ono, M., Khanam, D., Hayday, A.C., Frohman, L.A.,Koff, A., 1996. Enhanced growth of mice lacking the cyclin-dependentkinase inhibitor function of p27(Kip1). Cell 85, 721–732.

Koh, K.N., Kang, M.J., Frith-Terhune, A., Park, S.K., Kim, I., Lee, C.O., Koh,G.Y., 1998. Persistent and heterogeneous expression of the cyclin-dependentkinase inhibitor, p27KIP1, in rat hearts during development. J. Mol. Cell.Cardiol. 30, 463–474.

Kwang, S.J., Brugger, S.M., Lazik, A., Merrill, A.E., Wu, L.Y., Liu, Y.H., Ishii,M., Sangiorgi, F.O., Rauchman, M., Sucov, H.M., Maas, R.L., Maxson Jr.,R.E., 2002. Msx2 is an immediate downstream effector of Pax3 in the

development of the murine cardiac neural crest. Development 129,527–538.

Linask, K.K., Han, M.D., Linask, K.L., Schlange, T., Brand, T., 2003. Effects ofantisense misexpression of CFC on downstream flectin protein expressionduring heart looping. Dev. Dyn. 228, 217–230.

Liu, C., Liu, W., Palie, J., Lu, M.F., Brown, N.A., Martin, J.F., 2002. Pitx2cpatterns anterior myocardium and aortic arch vessels and is required forlocal cell movement into atrioventricular cushions. Development 129,5081–5091.

Liu, W., Selever, J., Lu, M.F., Martin, J.F., 2003. Genetic dissection of Pitx2 incraniofacial development uncovers new functions in branchial archmorphogenesis, late aspects of tooth morphogenesis and cell migration.Development 130, 6375–6385.

Liu, W., Selever, J., Wang, D., Lu, M.F., Moses, K.A., Schwartz, R.J., Martin,J.F., 2004. Bmp4 signaling is required for outflow-tract septation andbranchial-arch artery remodeling. Proc. Natl. Acad. Sci. U. S. A. 101,4489–4494.

Lo, C.W., Wessels, A., 1998. Cx43 gap junctions in cardiac development.Trends Cardiovasc. Med. 8, 264–269.

Lo, C.W., Waldo, K.L., Kirby, M.L., 1999. Gap junction communication and themodulation of cardiac neural crest cells. Trends Cardiovasc. Med. 9, 63–69.

Lomonico, M.P., Moore, G.W., Hutchins, G.M., 1986. Rotation of the junctionof the outflow tract and great arteries in the embryonic human heart. Anat.Rec. 216, 544–549.

Lomonico, M.P., Bostrom, M.P., Moore, G.W., Hutchins, G.M., 1988. Arrestedrotation of the outflow tract may explain tetralogy of Fallot and transpositionof the great arteries. Pediatr. Pathol. 8, 267–281.

Lu, M.F., Pressman, C., Dyer, R., Johnson, R.L., Martin, J.F., 1999. Function ofRieger syndrome gene in left–right asymmetry and craniofacialdevelopment. Nature 401, 276–278.

Martinsen, B.J., 2005. Reference guide to the stages of chick heart embryology.Dev. Dyn. 233, 1217–1237.

Maxson, R., Ishii, M., Merrill, A., 2003. Msx genes in organogenesis and humandisease. In: Lufkin, T. (Ed.), Advances in Developmental Biology andBiochemistry, vol. 13. Elsevier B.V., Amsterdam, pp. 43–68.

McFadden, D.G., Barbosa, A.C., Richardson, J.A., Schneider, M.D., Srivastava,D., Olson, E.N., 2005. The Hand1 and Hand2 transcription factors regulateexpansion of the embryonic cardiac ventricles in a gene dosage-dependentmanner. Development 132, 189–201.

Mikawa, T., 1999. Cardiac lineage. In: Harvey, R.P., Rosenthal, N. (Eds.), HeartDevelopment. Academic Press, San Diego, pp. 19–33.

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

Nakayama, K., Ishida, N., Shirane, M., Inomata, A., Inoue, T., Shishido, N.,Horii, I., Loh, D.Y., Nakayama, K., 1996. Mice lacking p27(Kip1) displayincreased body size, multiple organ hyperplasia, retinal dysplasia, andpituitary tumors. Cell 85, 707–720.

Oh, S.P., Li, E., 1997. The signaling pathway mediated by the type IIB activinreceptor controls axial patterning and lateral asymmetry in the mouse. GenesDev. 11, 1812–1826.

Pfaff, S.L., Mendelsohn, M., Stewart, C.L., Edlund, T., Jessell, T.M., 1996.Requirement for LIM homeobox gene Isl1 in motor neuron generationreveals a motor neuron-dependent step in interneuron differentiation. Cell84, 309–320.

Poolman, R.A., Li, J.M., Durand, B., Brooks, G., 1999. Altered expression ofcell cycle proteins and prolonged duration of cardiac myocyte hyperplasia inp27KIP1 knockout mice. Circ. Res. 85, 117–127.

Quelle, D.E., Ashmun, R.A., Hannon, G.J., Rehberger, P.A., Trono, D., Richter,K.H., Walker, C., Beach, D., Sherr, C.J., Serrano, M., 1995. Cloning andcharacterization of murine p16INK4a and p15INK4b genes. Oncogene 11,635–645.

Restivo, A., Piacentini, G., Placidi, S., Saffirio, C., Marino, B., 2006. Cardiacoutflow tract: a review of some embryogenetic aspects of the conotruncalregion of the heart. Anat. Rec. A Discov. Mol. Cell Evol. Biol. 288,936–943.

Risebro, C.A., Smart, N., Dupays, L., Breckenridge, R., Mohun, T.J., Riley, P.R.,


2006. Hand1 regulates cardiomyocyte proliferation versus differentiation inthe developing heart. Development 133, 4595–4606.

Satokata, I., Maas, R., 1994. Msx1 deficient mice exhibit cleft palate andabnormalities of craniofacial and tooth development. Nat. Genet. 6,348–356.

Satokata, I., Ma, L., Ohshima, H., Bei, M., Woo, I., Nishizawa, K., Maeda, T.,Takano, Y., Uchiyama, M., Heaney, S., Peters, H., Tang, Z., Maxson, R.,Maas, R., 2000. Msx2 deficiency in mice causes pleiotropic defects in bonegrowth and ectodermal organ formation. Nat. Genet. 24, 391–395.

Shen, M.M., Wang, H., Leder, P., 1997. A differential display strategy identifiesCryptic, a novel EGF-related gene expressed in the axial and lateralmesoderm during mouse gastrulation. Development 124, 429–442.

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

Srivastava, D., Olson, E.N., 2000. A genetic blueprint for cardiac development.Nature 407, 221–226.

St Amand, T.R., Zhang, Y., Semina, E.V., Zhao, X., Hu, Y., Nguyen, L., Murray,J.C., Chen, Y., 2000. Antagonistic signals between BMP4 and FGF8 definethe expression of Pitx1 and Pitx2 in mouse tooth-forming anlage. Dev. Biol.217, 323–332.

Stottmann, R.W., Choi, M., Mishina, Y., Meyers, E.N., Klingensmith, J., 2004.BMP receptor IA is required in mammalian neural crest cells fordevelopment of the cardiac outflow tract and ventricular myocardium.Development 131, 2205–2218.

Togi, K., Kawamoto, T., Yamauchi, R., Yoshida, Y., Kita, T., Tanaka, M.,2004. Role of Hand1/eHAND in the dorso-ventral patterning andinterventricular septum formation in the embryonic heart. Mol. Cell.Biol. 24, 4627–4635.

Trinh, L.A., Yelon, D., Stainier, D.Y., 2005. Hand2 regulates epithelialformation during myocardial differentiation. Curr. Biol. 15, 441–446.

Van Praagh, R., Van Praagh, S., 1965. The anatomy of commonaorticopulmonary trunk (truncus arteriosus communis) and its embryologicimplications. A study of 57 necropsy cases. Am. J. Cardiol. 16, 406–425.

von Both, I., Silvestri, C., Erdemir, T., Lickert, H., Walls, J.R., Henkelman,R.M., Rossant, J., Harvey, R.P., Attisano, L., Wrana, J.L., 2004. Foxh1 isessential for development of the anterior heart field. Dev. Cell 7, 331–345.

Waldo, K.L., Lo, C.W., Kirby, M.L., 1999. Connexin 43 expression reflectsneural crest patterns during cardiovascular development. Dev. Biol. 208,307–323.

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 secondaryheart field. Development 128, 3179–3188.

Ward, C., Stadt, H., Hutson, M., Kirby, M.L., 2005. Ablation of the secondaryheart field leads to tetralogy of Fallot and pulmonary atresia. Dev. Biol. 284,72–83.

Watanabe, M., Choudhry, A., Berlan, M., Singal, A., Siwik, E., Mohr, S., Fisher,S.A., 1998. Developmental remodeling and shortening of the cardiacoutflow tract involves myocyte programmed cell death. Development 125,3809–3820.

Wu, L.Y., Li, M., Hinton, D.R., Guo, L., Jiang, S., Wang, J.T., Zeng, A., Xie,J.B., Snead, M., Shuler, C., Maxson Jr., R.E., Liu, Y.H., 2003.Microphthalmia resulting from MSX2-induced apoptosis in the opticvesicle. Invest. Ophthalmol. Visual Sci. 44, 2404–2412.

Yanagisawa, H., Clouthier, D.E., Richardson, J.A., Charite, J., Olson, E.N.,2003. Targeted deletion of a branchial arch-specific enhancer revealsa role of dHAND in craniofacial development. Development 130,1069–1078.

Yasui, H., Nakazawa, M., Morishima, M., Ando, M., Takao, A., Aikawa, E.,1997. Cardiac outflow tract septation process in the mouse model oftransposition of the great arteries. Teratology 55, 353–363.

Yelbuz, T.M., Waldo, K.L., Kumiski, D.H., Stadt, H.A., Wolfe, R.R.,Leatherbury, L., Kirby, M.L., 2002. Shortened outflow tract leads toaltered cardiac looping after neural crest ablation. Circulation 106, 504–510.

Yelon, D., Ticho, B., Halpern, M.E., Ruvinsky, I., Ho, R.K., Silver, L.M.,Stainier, D.Y., 2000. The bHLH transcription factor hand2 plays parallelroles in zebrafish heart and pectoral fin development. Development 127,2573–2582.

Zhang, J., Hagopian-Donaldson, S., Serbedzija, G., Elsemore, J., Plehn-Dujowich, D., McMahon, A.P., Flavell, R.A., Williams, T., 1996. Neuraltube, skeletal and body wall defects in mice lacking transcription factorAP-2. Nature 381, 238–241.

Zhang, Z., Song, Y., Zhao, X., Zhang, X., Fermin, C., Chen, Y., 2002. Rescue ofcleft palate in Msx1-deficient mice by transgenic Bmp4 reveals a network ofBMP and Shh signaling in the regulation of mammalian palatogenesis.Development 129, 4135–4146.

Zhang, Z., Zhang, X., Avniel, W.A., Song, Y., Jones, S.M., Jones, T.A., Fermin,C., Chen, Y., 2003. Malleal processus brevis is dispensable for normalhearing in mice. Dev. Dyn. 227, 69–77.

Msx1 and Msx2 regulate survival of secondary heart field precursors and post-migratory proliferation...

Documents

Transcript of Msx1 and Msx2 regulate survival of secondary heart field precursors and post-migratory proliferation...