Germline mutation in mice of the retinoic acid receptor gene RXRα results in a proliferative failure of cardiomyocytes, which leads to an underdeveloped ventricular chamber and midgestation lethality. Mutation of the cell cycle regulator N-myc gene also leads to an apparently identical phenotype. In this study, we demonstrate by chimera analysis that the cardiomyocyte phenotype in RXRα−/−embryos is a non-cell-autonomous phenotype. In chimeric embryos made with embryonic stem cells lacking RXRα, cardiomyocytes deficient in RXRα develop normally and contribute to the ventricular chamber wall in a normal manner. Because the ventricular hypoplastic phenotype reemerges in highly chimeric embryos, we conclude that RXRα functions in a non-myocyte lineage of the heart to induce cardiomyocyte proliferation and accumulation, in a manner that is quantitatively sensitive. We further show that RXRα is not epistatic to N-myc, and that RXRα and N-myc regulate convergent obligate pathways of cardiomyocyte maturation.

Most aspects of vertebrate organogenesis begin with the assembly of cells of different lineage into a tissue primordium. Thereafter, a progressive series of inductive interactions between these lineages results in morphogenesis and maturation into a differentiated and functional organ. In the developing heart, cells of the cardiomyocyte and endothelial (endocardial) lineages are segregated within the cardiac crescent of anterior lateral mesoderm prior to the assembly of a beating heart. Cells which form the epicardium (the outer layer of the heart) and tissue of the aorta and pulmonary outflow vessels then migrate to the heart from the sinus venosus and neural crest, respectively. These lineages are assembled in the mouse heart by approximately embryonic day 10.5 (E10.5). Thereafter, each domain of the developing heart undergoes further maturation. Morphogenesis of the ventricular chamber involves proliferation and accumulation of cardiomyocytes in the ventricular chamber wall, resulting in a thickened muscular ‘compact zone’, which is essential for pumping blood through the rapidly growing embryo.

The vitamin A derivative retinoic acid (RA) is a critical signaling molecule in numerous developmental and physiological processes, including several aspects of cardiovascular morphogenesis. The biological effects of RA are mediated by members of the nuclear receptor family of ligand-dependent transcription factors (Evans, 1988). Germline mutation (Sucov et al., 1994; Kastner et al., 1994) in mice of the retinoic acid receptor gene RXRα results in a prominent and completely penetrant ventricular chamber phenotype. In RXRα−/−embryos, all lineages of the developing heart are present and normally organized up to E11.5. However, further proliferation of the myocardium in the compact zone of the ventricular chamber wall fails to occur, so that by E14.5 a persistently thin-walled and hypoplastic ventricle remains. Some 50 years ago, nutritional studies established that expansion of the compact zone requires vitamin A (Wilson and Warkany, 1949), which is now recognized as a requirement for provision of RA as a signaling molecule and which is received at least in part by RXRα. The RXRα−/−phenotype is embryonic lethal around E15.5 because of insufficient cardiac performance (Dyson et al., 1995).

Mutation of several other genes, including the cell-cycle regulator N-myc (Moens et al., 1993), results in a similar, if not identical, hypoplastic ventricular chamber phenotype. Although this phenotype is ultimately a proliferative deficiency of cardiomyocytes, it cannot be assumed that the function of any of the genes that affect this process is required in cardiomyocytes. RXRα is ubiquitously expressed prior to E15.5 (Mangelsdorf et al., 1992; Dolle et al., 1994) and, while fetal (unpublished observations) and neonatal (Zhou et al., 1995) cardiomyocytes are able to respond to RA signaling (as evidenced by transcriptional activation of reporter genes), almost all cell types have this same capacity (Sucov et al., 1990). Another lineage of the heart could in principle respond to RA signaling through RXRα and in a secondary process then inductively direct cardiomyocyte proliferation and accumulation. RXRα, N-myc and other genes might function within or external to the ventricular cardiomyocyte lineage and possibly regulate pathways of cardiac development that are epistatic or convergent.

In this study, we have undertaken a direct investigation of the site of RXRα action. We show in chimeric embryos, made by introduction of RXRα−/−embryonic stem (ES) cells into wild-type recipient blastocysts, that cardiomyocytes that lack RXRα proliferate normally in the compact zone of the ventricular chamber wall. Thus, RXRα functions in a lineage external to the cardiomyocyte population and the RXRα−/−cardiomyocyte phenotype is cell non-autonomous. This same conclusion has been reached by a completely different methodology, as described in the accompanying paper (Chen et al., 1998). We furthermore demonstrate that RXRα and N-myc regulate convergent obligate pathways of cardiomyocyte maturation.

Derivation of embryonic stem cell lines

Mice bearing the ROSA-26 transgene (Friedrich and Soriano, 1991) were obtained from Jackson Laboratories at the 6th generation of breeding into the C57Bl/6 background. These were crossed to RXRα−/+ partners that had been inbred on the C57Bl/6 background at least seven generations, and transgenic male heterozygotes were used in matings with nontransgenic heterozygous females from the same strain background. Females were superovulated prior to mating and morula isolated from the oviduct-uterine junction at E2.5. Following overnight culture, blastocysts were plated on a layer of mitotically inactive primary embryonic fibroblast feeder cells in conventional ES cell media supplemented with 1000 units/ml LIF. Blastocyst outgrowths were picked and expanded, and then genotyped and characterized for the presence of the ROSA-26 transgene.

Production and analysis of chimeric embryos

Morula were isolated from matings of wild-type CD-1 (ICR) mice at E2.5 and cocultured with ES cells essentially as described (Wood et al., 1993). For low percentage chimeras, ES cells were plated at a concentration of 1.5×105 cells/ml; for highly chimeric embryos, the ES cell concentration was raised to 5×105 cells/ml. Coculture was for 3.5 hours, after which embryos were removed and allowed to incubate overnight and viable blastocysts were implanted into pseudopregnant ICR females the following day. Embryos were isolated at midgestation, with chimeric embryos identified initially on the basis of eye pigmentation chimerism. Generally, hearts were isolated, fixed and stained in X-gal by standard procedures, then paraffin embedded, sectioned at 3 μm thickness and counterstained with nuclear fast red. For determination of percentage embryonic chimerism, several pieces of noncardiac tissue were removed, genomic DNA isolated as above and analyzed by Southern blotting with probe A as described previously (Sucov et al., 1994). Chimerism was also estimated by degree of eye pigmentation mosaicism and by visual estimation of per cent of heart tissue that stained in X-gal, which were the only available means of evaluating chimeric embryos made with wild-type ES cells.

Myc transgenic crosses

The RSV-c-myc transgenic line (Swain et al., 1987) was provided by Dr Judith Swain and was crossed into the RXRα mutant background to generate transgenic male heterozygotes. These were crossed with RXRα−/+ females (this direction of cross is necessary because the myc transgene is imprinted and is only expressed when paternally inherited), with matings either allowed to go to term or interrupted at E14.5 to isolate embryos. The presence of the myc transgene was determined by dot blotting.

Chimeric analysis of RXRα−/−cardiomyocytes

We elected to address directly the issue of whether RXRα functions in a cardiomyocyte cell autonomous manner by chimera analysis. In this approach, chimeric embryos are made by introduction of embryonic stem (ES) cells lacking RXRα (or control wild-type ES cells) into wild-type recipient blastocysts. When these embryos are allowed to develop to midgestation, the behavior of cardiomyocytes lacking RXRα (i.e., those derived from the RXRα−/−ES cells) can be observed.

New ES cell lines were established by blastocyst outgrowths obtained from crosses of RXRα−/+adult mice. For purposes of identifying ES-derived cells in the midgestation embryo, the ubiquitously expressed ROSA-26 β-galactosidase transgene (Friedrich and Soriano, 1991) was crossed into the RXRα−/+line in prior matings and ES cell lines were screened for the presence of this marker gene. Three RXRα−/−ES cell lines and three wild-type cell lines, all carrying the ROSA-26 transgene, were isolated and used in this study. All RXRα−/−ES cells grew normally and had normal ES cell morphology. Chimeric embryos were made by aggregation, reimplanted into recipient females and allowed to develop to various stages of midgestation. A total of 19 chimeric embryos made with wild-type ES cells, and 15 chimeric embryos made with RXRα−/−ES cells, were evaluated in this study.

Previous studies using retroviral labeling in chick embryos (Mikawa et al., 1992) indicate that ventricular cardiomyocytes proliferate clonally and coherently. We initially employed conditions that resulted in chimeric embryos with a low contribution of ES cells, such that the clonal behavior of individual cardiomyocyte precursors could be determined. In Fig. 1A, a section through the heart of a chimeric embryo at E12.5 is shown in which the β-galactosidase-positive cells are wild type; two independent clones of myocardium that are ES cell-derived are evident, distributed in the ventricular wall (which at this stage is still thin) and in the trabeculated layer of myocardium that projects into the ventricular lumen. A chimeric embryo at E12.5 made with RXRα−/−ES cells is shown in Fig. 1B; a single clone is seen, distributed both in the chamber wall and in the trabecular layer. The number of cells per patch is comparable in wild-type and RXRα−/−chimeric heart tissue from several different embryos. Thus, embryonic precursor cells lacking RXRα are normally allocated at gastrulation to the cardiomyocyte lineage and proliferate in an apparently normal manner through E12.5.

Fig. 1.

Chimera analysis of RXRα cell autonomy. Chimeric embryos made with wild-type (A,D) or RXRα−/− (B,C,E) ES cells were isolated at E12.5 (A,B), E14.5 (C) and E16.5 (D,E). ES-derived cells are X-gal positive (blue) and wild-type host-derived cells are counterstained red. In wild-type embryos, at E12.5, the compact zone (cz) of the ventricular chamber wall is approximately 2-3 cell diameters thick, increases to 10-15 cell diameters by E14.5 and enlarges even more by E16.5. The ability of RXRα−/− cardiomyocytes to contribute to the compact zone in E14.5 (C) and E16.5 (E) embryos is readily apparent. F depicts a hematoxylin/eosin-stained section through a conventional RXRα−/− embryo at E14.5, to illustrate the hypoplastic ventricular chamber wall phenotype. All images are at the same magnification. Scale bar in A, 100 μm. Other abbreviations: t, trabecular myocardium; e, epicardium.

Fig. 1.

Chimera analysis of RXRα cell autonomy. Chimeric embryos made with wild-type (A,D) or RXRα−/− (B,C,E) ES cells were isolated at E12.5 (A,B), E14.5 (C) and E16.5 (D,E). ES-derived cells are X-gal positive (blue) and wild-type host-derived cells are counterstained red. In wild-type embryos, at E12.5, the compact zone (cz) of the ventricular chamber wall is approximately 2-3 cell diameters thick, increases to 10-15 cell diameters by E14.5 and enlarges even more by E16.5. The ability of RXRα−/− cardiomyocytes to contribute to the compact zone in E14.5 (C) and E16.5 (E) embryos is readily apparent. F depicts a hematoxylin/eosin-stained section through a conventional RXRα−/− embryo at E14.5, to illustrate the hypoplastic ventricular chamber wall phenotype. All images are at the same magnification. Scale bar in A, 100 μm. Other abbreviations: t, trabecular myocardium; e, epicardium.

Expansion of the compact zone of the ventricular chamber is initiated around E12.5, with the chamber wall thickening progressively with advancing gestational age; this is the process that fails in RXRα−/−embryos. Chimeric embryos were isolated at E14.5 and at E16.5 and stained to reveal ES cell-derived cardiomyocytes. In embryos with a low to moderate level of chimerism, patches of myocardium were obtained with RXRα−/−ES cells, which appeared indistinguishable from the wild-type myocardium around them (Fig. 1C) and which appeared comparable in cell number to those observed when wild-type ES cells were used to make chimeras. An example of a highly chimeric heart at E16.5 is shown in Fig. 1E. This embryo, estimated to be approximately 50% derived from RXRα−/−ES cells overall, was normally developed and did not demonstrate any signs of impending heart failure. In sections, a substantial contribution of RXRα−/−cardiomyocytes was seen throughout the chamber wall. Therefore, in chimeric embryos, cardiomyocytes lacking RXRα are able to proliferate in the ventricular chamber wall and form a normal compact zone. This is in direct contrast to the behavior of this lineage in an embryo that is completely deficient in RXRα (Fig. 1F) and demonstrates that the RXRα mutant ventricular phenotype is non-cell autonomous.

RXRα function is required in a quantitatively sensitive manner

In chimeric embryos, all cell lineages are chimeric, unless the RXRα mutation prevents contribution to certain tissues. As the per cent contribution of RXRα−/−ES cells in a chimeric embryo increases, the RXRα−/−ventricular hypoplastic phenotype should re-emerge. In this study, we isolated three chimeric embryos at E16.5 that had died in utero 1 or 2 days earlier (i.e., around E14.5-15.5) and which exhibited underdevelopment of the ventricles (Fig. 2). The overall ES cell contribution to these embryos was 75-95%. As noted above, embryos with 50% chimerism or less appeared normal. Three highly chimeric embryos made with wild-type ES cells (two of approximately 75% contribution of ES cell progeny, one of over 90%) and isolated at E16.5 were also normal, indicating that the aggregation procedure or a high contribution of ES cells per se do not cause dismorphic development. Thus, while RXRα gene function is not required in a cell autonomous manner for normal cardiomyocyte ontogeny, ventricular maturation requires RXRα in another embryonic lineage in a manner that is quantitatively sensitive.

Fig. 2.

Compact zone deficiency in highly chimeric embryos. A section through the ventricular chamber of a chimeric embryo in which over 75% of the cells lacked RXRα is shown. This embryo was isolated at E16.5 and had been dead for approximately 1 or 2 days, accounting for the histolyzed appearance of this tissue. Nonetheless, the general organization of the epicardium (e), compact zone (cz) and trabecular myocardium (t) is still apparent. Scale bar, 100 μm.

Fig. 2.

Compact zone deficiency in highly chimeric embryos. A section through the ventricular chamber of a chimeric embryo in which over 75% of the cells lacked RXRα is shown. This embryo was isolated at E16.5 and had been dead for approximately 1 or 2 days, accounting for the histolyzed appearance of this tissue. Nonetheless, the general organization of the epicardium (e), compact zone (cz) and trabecular myocardium (t) is still apparent. Scale bar, 100 μm.

RXRα and N-myc regulate independent required steps of cardiogenesis

Mutation of the N-myc gene results in a cardiac phenotype that is indistinguishable from the RXRα−/−phenotype. Unlike RXRα, which is broadly expressed throughout the heart, the N-myc gene is restricted in expression to the myocardium and is more abundantly expressed in the compact zone of the outer ventricular chamber wall (Moens et al., 1993). A transgene that expresses the functionally related c-myc protein specifically in cardiomyocytes (Swain et al., 1987) prevents the cardiac defects seen in N-myc mutant embryos (J. Rossant, personal communication). Thus, N-myc functions in a cell autonomous manner in the cardiomyocyte lineage.

Because RXRα functions outside the cardiomyocyte lineage, it is possible that N-myc and RXRα are epistatically related, with N-myc being downstream and more proximal to regulation of cardiomyocyte proliferation. If so, transgenic myc expression might rescue RXRα deficiency as it does for N-myc deficiency. We crossed the myc-expressing transgene into the RXRα background and mated myc-transgenic, RXRα heterozygous males with RXRα−/+females. As shown in Table 1, of 43 pups that reached full term in these matings, none were homozygous for the RXRα mutation. Of 23 embryos isolated at E14.5, six were RXRα homozygous, of which two were also positive for the myc transgene. All six were visibly edematous and presumed near death, and there was no obvious distinction between transgenic and nontransgenic embryos in the severity of their appearance. In histological sections (not shown), transgenic RXRα−/−embryos displayed compact zone hypoplasia comparable to nontransgenic RXRα−/−embryos and comparable to our previous reports on this phenotype. Thus, expression of myc does not prevent the onset of the myocardial hypoplasia in RXRα−/−embryos. N-myc is either genetically upstream of RXRα or each gene regulates independent pathways of cardiomyocyte development that converge at a common downstream point.

Table 1.

Distribution of progeny from RXRα−/−, RSV-c-myc TG+ × RXRα−/+matings

Distribution of progeny from RXRα−/−, RSV-c-myc TG+ × RXRα−/+matings
Distribution of progeny from RXRα−/−, RSV-c-myc TG+ × RXRα−/+matings

We have demonstrated in this paper by chimeric analysis that proliferation and accumulation of cardiomyocytes in the developing compact zone does not directly require RXRα function. An identical conclusion was reached in the accompanying paper (Chen et al., 1998) using a cre/lox recombination strategy. Although there are technical considerations with any single approach, together the various strategies complement each other. For example, the normal behavior of RXRα−/−cardiomyocytes in a chimeric context might still be compatible with a cardiomyocyte function for RXRα, if there is a second compensatory RXRα-independent signal secreted in a paracrine manner by nearby wild-type cardiomyocytes. However, the demonstration that cardiomyocyte-specific mutation of the RXRα gene does not lead to the RXRα phenotype (Chen et al., 1998) negates this alternative explanation. Similarly, the normal phenotype of embryos in which the RXRα gene is mutated selectively in cardiomyocytes might still be compatible with a cardiomyocyte function for RXRα, if recombination does not occur until after the time at which RXRα function is required. However, in chimeric embryos, RXRα-deficient cardiomyocytes derive from cells that lack RXRα from the blastocyst stage. The RXRα ventricular deficiency is therefore conclusively demonstrated to be a non-cell-autonomous phenotype.

This conclusion explains several unpublished observations that we have made previously: (i) transgenic expression of wild-type RXRα protein in ventricular cardiomyocytes from the myosin light chain 2v promoter (Ross et al., 1996) does not prevent the onset of the RXRα−/−ventricular phenotype; (ii) an apparently normal number of ventricular cardiomyocytes are recovered from differentiating embryoid bodies derived from aggregated RXRα-deficient ES cells and (iii) ventricular cardiomyocytes isolated from E13.5 RXRα−/−embryos and grown in culture as single cells are able to traverse the cell cycle in an apparently normal manner. There are experimental caveats associated with these studies that could trivially explain each observation. Nonetheless, when taken at face value, these observations are only compatible with a mechanism of action for RXRα in a non-myocyte cell population, as we show definitively in this paper by chimera analysis.

It is of obvious interest to determine the lineage in which RXRα function is required. The relatively few non-myocyte lineages of the heart include the endocardium (cardiac endothelium), a single cell layer lining the interior of the heart, which originates within the cardiogenic crescent prior to formation of definitive heart tissue (Cohen-Gould and Mikawa, 1996); the epicardium, a cell layer lining the outside of the heart, which migrates to surround the heart from the sinus venosus region (Viragh and Challice, 1981); loose connective tissue that occupies the subepicardial space adjacent to the myocardium and is believed to originate with the epicardium; and neural crest (Kirby and Waldo, 1990), which migrate into the conotruncal (outflow) region of the heart but which may also influence ventricular development. It is also possible that the requirement for RXRα occurs outside of the heart, either in the surrounding visceral pericardium or in a distant tissue, which in an endocrine manner secretes a signal that directs ventricular cardiomyocyte development. The approaches described here and in the accompanying paper that we have applied to the cardiomyocyte lineage can also be used to explore the role of RXRα in these other lineages.

Highly chimeric embryos, in which approximately 75% or greater of the cells lacked RXRα, were affected by ventricular hypoplasia and were dead upon recovery, although embryos of approximately 50% or less chimerism were normal. These observations indicate that, no matter where RXRα function is required, the absence of RXRα does not prevent cells from contributing to this lineage in chimeric embryos or prevent it from forming in RXRα-/-embryos. This behavior is in contrast to genes that affect initial lineage allocation or specification.

For example, the Flk1 gene is essential for formation of endothelium in a cell autonomous manner and, even in heavily chimeric embryos made with Flk1−/−ES cells, the endothelium is completely composed of wild-type cells (Shalaby et al., 1997). The role of RXRα therefore appears to be not one of early lineage determination, but rather part of a later event that involves retinoic-acid-induced transcriptional regulation in an already-present tissue. Furthermore, the observation that an approximately 25% or less contribution of wild-type cells to chimeric embryos is unable to support compact zone formation suggests that there is a quantitative threshold of RXRα-mediated transcriptional regulation which is necessary for ventricular maturation.

It is not possible at the moment to distinguish the nature of the interaction between the cardiomyocyte population and the lineage in which RXRα is functional. The relatively large patches of normal RXRα-deficient myocardium seen in chimeric embryos at E14.5 and beyond strongly suggest that a diffusible factor is involved. This could be a growth factor, or could be a secreted component of the extracellular matrix. It is less likely that physical cell-cell contacts are involved, in that most cardiomyocytes are in contact only with other cardiomyocytes, although this possibility cannot yet be disproven.

Several other gene mutations have been established in mice that are similar, if not identical, to the RXRα−/−ventricular phenotype. These include the transcription factors N-myc (Moens et al., 1993), TEF-1 (Chen et al., 1994) and WT-1 (Kreidberg et al., 1993), the cell surface receptor gp130 (Yoshida et al., 1996), and the cell surface receptor kinase βARK-1 (Jaber et al., 1996). With the exception of N-myc, which as described above is established to act autonomously in the cardiomyocyte lineage, little is known of the nature of the defects seen in the other mutant backgrounds. It is interesting that WT-1 is reported (Armstrong et al., 1993) to be expressed in the epicardium and not in the myocardium, which might also suggest a non-autonomous role for this gene.

A critical aspect of future efforts in understanding the nature of the RXRα phenotype will be to define RXRα function in the context of other genes that also regulate ventricular morphogenesis. We have demonstrated in this paper that N-myc is not the downstream target through which RXRα function is manifest. It would be surprising if N-myc is upstream of RXRα, in that myc functions in cardiomyocyte proliferation and this is the ultimate process that is compromised in both RXRα and N-myc mutant backgrounds. N-myc could be in one of multiple obligate pathways that are regulated by RXRα, or N-myc and RXRα could regulate independent required pathways that ultimately converge at a common downstream point in the regulation of cardiomyocyte proliferation. The assignment of RXRα, N-myc and other genes that regulate ventricular maturation to their tissue of action and to epistatic pathways of action will define the nature of the inductive events that underlie cardiac ventricular morphogenesis.

We are indebted to Dr Ronald Evans for support during the early phase of this work. We thank Dr Judith Swain for provision of the RSV-c-myc transgenic line of mice, Dr Janet Rossant for communication of unpublished results, and Dr Jeffrey Mann for technical advice. This work was supported by a Grant-in-Aid from the Greater Los Angeles Affiliate of the American Heart Association and a Research Grant from the Robert E. and May R. Wright Foundation to H. M. S.

Armstrong
,
J. F.
,
Pritchard-Jones
,
K.
,
Bickmore
,
W. A.
,
Hastie
,
N. D.
and
Bard
,
J. B.
(
1993
).
The expression of the Wilms’ tumour gene, WT1, in the developing mammalian embryo
.
Mech Dev
40
,
85
97
.
Chen
,
J.
,
Kubalak
,
S. W.
and
Chien
,
K. R.
(
1998
).
Ventricular muscle-restricted targeting of the RXRα gene reveals a non-cell-autonomous requirement in cardiac chamber morphogenesis
.
Development
125
,
1943
1949
.
Chen
,
Z.
,
Friedrich
,
G. A.
and
Soriano
,
P.
(
1994
).
Transcriptional enhancer factor 1 disruption by a retroviral gene trap leads to heart defects and embryonic lethality in mice
.
Genes Dev.
8
,
2293
2301
.
Cohen-Gould
,
L.
and
Mikawa
,
T.
(
1996
).
The fate diversity of mesodermal cells within the heart field during chicken early embryogenesis
.
Dev. Biol.
177
,
265
273
.
Dolle
,
P.
,
Fraulob
,
V.
,
Kastner
,
P.
and
Chambon
,
P.
(
1994
).
Developmental expression of murine retinoid X receptor (RXR) genes
.
Mech. Dev.
45
,
91
104
.
Dyson
,
E.
,
Sucov
,
H. M.
,
Kubalak
,
S. W.
,
Schmid-Schonbein
,
G. W.
,
DeLano
,
F. A.
,
Evans
,
R. M.
,
Ross
,
J.
and
Chien
,
K. R.
(
1995
).
Atrial-like phenotype is associated with embryonic ventricular failure in RXRα-/-mice
.
Proc. Natn. Acad. Sci. USA
92
,
7386
7390
.
Evans
,
R. M.
(
1988
).
The steroid and thyroid hormone receptor superfamily
.
Science
240
,
889
895
.
Friedrich
,
G.
and
Soriano
,
P.
(
1991
).
Promoter traps in embryonic stem cells: a genetic screen to identify and mutate developmental genes in mice
.
Genes Dev.
5
,
1513
1523
.
Jaber
,
M.
,
Koch
,
W. J.
,
Rockman
,
H.
,
Smith
,
B.
,
Bond
,
R. A.
,
Sulik
,
K. K.
,
Ross
,
J. J.
,
Lefkowitz
,
R. J.
,
Caron
,
M. G.
and
Giros
,
B.
(
1996
).
Essential role of beta-adrenergic receptor kinase 1 in cardiac development and function
.
Proc. Natl. Acad. Sci. USA
93
,
12974
12979
.
Kastner
,
P.
,
Grondona
,
J. M.
,
Mark
,
M.
,
Gansmuller
,
A.
,
LeMeur
,
M.
,
Decimo
,
D.
,
Vonesch
,
J. L.
,
Dolle
,
P.
and
Chambon
,
P.
(
1994
).
Genetic analysis of RXRα developmental function: convergence of RXR and RAR signaling pathways in heart and eye morphogenesis
.
Cell
78
,
987
1003
.
Kirby
,
M. L.
and
Waldo
,
K. L.
(
1990
).
Role of neural crest in congenital heart disease
.
Circulation
82
,
332
340
.
Kreidberg
,
J. A.
,
Sariola
,
H.
,
Loring
,
J. M.
,
Maeda
,
M.
,
Pelletier
,
J.
,
Housman
,
D.
and
Jaenisch
,
R.
(
1993
).
WT-1 is required for early kidney development
.
Cell
74
,
679
691
.
Mangelsdorf
,
D. J.
,
Borgmeyer
,
U.
,
Heyman
,
R. A.
,
Zhou
,
J. Y.
,
Ong
,
E. S.
,
Oro
,
A. E.
,
Kakizuka
,
A.
and
Evans
,
R. M.
(
1992
).
Characterization of three RXR genes that mediate the action of 9-cis retinoic acid
.
Genes Dev.
6
,
329
344
.
Mikawa
,
T.
,
Borisov
,
A.
,
Brown
,
A. M.
and
Fischman
,
D. A.
(
1992
).
Clonal analysis of cardiac morphogenesis in the chicken embryo using a replication-defective retrovirus: I. Formation of the ventricular myocardium
.
Dev. Dyn.
193
,
11
23
.
Moens
,
C. B.
,
Stanton
,
B. R.
,
Parada
,
L. F.
and
Rossant
,
J.
(
1993
).
Defects in heart and lung development in compound heterozygotes for two different targeted mutations at the N-myc locus
.
Development
119
,
485
499
.
Ross
,
R. S.
,
Navankasattusas
,
S.
,
Harvey
,
R. P.
and
Chien
,
K. R.
(
1996
).
An HF-1a/HF-1b/MEF-2 combinatorial element confers cardiac ventricular specificity and established an anterior-posterior gradient of expression
.
Development
122
,
1799
1809
.
Shalaby
,
F.
,
Ho
,
J.
,
Stanford
,
W. L.
,
Fischer
,
K. D.
,
Schuh
,
A. C.
,
Schwartz
,
L.
,
Bernstein
,
A.
and
Rossant
,
J.
(
1997
).
A requirement for Flk1 in primitive and definitive hematopoiesis and vasculogenesis
.
Cell
89
,
981
990
.
Sucov
,
H. M.
,
Dyson
,
E.
,
Gumeringer
,
C. L.
,
Price
,
J.
,
Chien
,
K. R.
and
Evans
,
R. M.
(
1994
).
RXRα mutant mice establish a genetic basis for vitamin A signaling in heart morphogenesis
.
Genes Dev.
8
,
1007
1018
.
Sucov
,
H. M.
,
Murakami
,
K. K.
and
Evans
,
R. M.
(
1990
).
Characterization of an autoregulated response element in the mouse retinoic acid receptor type β gene
.
Proc. Natl. Acad. Sci. USA
87
,
5392
5396
.
Swain
,
J. L.
,
Stewart
,
T. A.
and
Leder
,
P.
(
1987
).
Parental legacy determines methylation and expression of an autosomal transgene: a molecular mechanism for parental imprinting
.
Cell
50
,
719
727
.
Viragh
,
S.
and
Challice
,
C. E.
(
1981
).
The origin of the epicardium and the embryonic myocardial circulation in the mouse
.
Anat. Rec.
201
,
157
168
.
Wilson
,
J. G.
and
Warkany
,
J.
(
1949
).
Aortic arch and cardiac anomalies in the offspring of vitamin A deficient rats
.
Am. J. Anat.
85
,
113
155
.
Wood
,
S. A.
,
Pascoe
,
W. S.
,
Schmidt
,
C.
,
Kemler
,
R.
,
Evans
,
M. J.
and
Allen
,
N. D.
(
1993
).
Simple and efficient production of embryonic stem cell-embryo chimeras by coculture
.
Proc. Natl. Acad. Sci. USA
90
,
4582
4585
.
Yoshida
,
K.
,
Taga
,
T.
,
Saito
,
M.
,
Suematsu
,
S.
,
Kumanogoh
,
A.
,
Tanaka
,
T.
,
Fujiwara
,
H.
,
Hirata
,
M.
,
Yamagami
,
T.
,
Nakahata
,
T.
,
Hirabayashi
,
T.
,
Yoneda
,
Y.
,
Tanaka
,
K.
,
Wang
,
W. Z.
,
Mori
,
C.
,
Shiota
,
K.
,
Yoshida
,
N.
and
Kishimoto
,
T.
(
1996
).
Targeted disruption of gp130, a common signal transducer for the interleukin 6 family of cytokines, leads to myocardial and hematological disorders
.
Proc. Natl. Acad. Sci. USA
93
,
407
411
.
Zhou
,
M. D.
,
Sucov
,
H. M.
,
Evans
,
R. M.
and
Chien
,
K. R.
(
1995
).
Retinoid-dependent pathways suppress myocardial cell hypertrophy
.
Proc. Natl. Acad. Sci. USA
92
,
7391
7395
.