Advanced Cardiac Morphogenesis Does Not Require Heart Tube Fusion

See allHide authors and affiliations

Science  10 Sep 2004:
Vol. 305, Issue 5690, pp. 1619-1622
DOI: 10.1126/science.1098674


The bilateral cardiac mesoderm migrates from the lateral region of the embryo to the ventral midline, where it fuses to form the primitive heart tube. It is generally accepted that migration and fusion are essential for subsequent stages of cardiac morphogenesis. We present evidence that, in Foxp4 mutant embryonic mice, each bilateral heart-forming region is capable of developing into a highly differentiated four-chambered mammalian heart in the absence of midline fusion. These data demonstrate that left-right chamber specification, cardiac looping, septation, cardiac myocyte differentiation, and endocardial cushion formation are preprogrammed in the precardiac mesoderm and do not require midline positional identity or heart tube fusion.

Although the molecular mechanisms underlying cardiac myocyte differentiation have been extensively examined and initial molecular pathways have been identified, much less is understood about how specified myocardial cells form the primitive heart tube and the four-chambered mammalian heart. This complex morphological process has a major impact on human health, because cardiovascular defects account for a substantial percentage of neonatal congenital disease. The bilateral precardiac mesoderm forms at the anterior pole of the vertebrate embryo (1, 2). How this mesoderm migrates to the ventral midline to form the single heart tube in the embryo is not well understood, although contributions from anterior foregut endoderm have been implicated. Moreover, whether this fusion event is required for cardiac differentiation and morphogenesis is not known, although some markers of the cardiac myocyte lineage that become spatially restricted in later development (such as eHAND, MLC2a, and MLC2v) are expressed throughout the early precardiac mesoderm (1, 2). The heart is also the first organ to exhibit left-right asymmetry. Whether this asymmetry is preprogrammed into the precardiac mesoderm or whether it is acquired coincident with midline fusion is unknown. Most genetic models of defective heart tube fusion, also known as cardia bifida, are characterized by bilateral regions of specified cardiac mesoderm that express cardiac-specific genes but fail to progress through later stages of cardiac morphogenesis (35). In addition, most murine models of cardia bifida exhibit additional extracardiac defects in body pattern formation, including severe defects in ventral morphogenesis and embryonic turning (3, 6). These data have led to a working model in which ventral midline fusion of the bilateral cardiac primordia is essential for subsequent cardiac development and morphogenesis, especially later aspects of chamber identity, left-right cardiac asymmetry, and looping morphogenesis (1, 7).

We have previously cloned and characterized Foxp4, a member of the Fox gene family that is expressed in multiple tissues, including the lung, gut, and brain, in the developing mouse embryo (8, 9). Foxp4 is expressed in early foregut endoderm and later in development in lung and hindgut (8). To generate a mutant allele of Foxp4, we replaced two exons encoding the forkhead DNA binding domain with the neomycin resistance cassette (fig. S1, A and B). Proper gene targeting was confirmed by Southern blotting and polymerase chain reaction (fig. S1, D and E). Immunohistochemistry with a Foxp4-specific polyclonal antibody confirmed that mutant embryos no longer expressed Foxp4 protein, suggesting that we had generated a null allele (fig. S1C). Heterozygous embryos were fertile and exhibited no obvious defects (10). The majority of homozygous embryos died around embryonic day 12.5 (E12.5) (table S1).

Histological analysis from E8.5 to E12.5 revealed the development of two complete hearts in Foxp4 mutant embryos (Fig. 1B). This was apparent as early as E8.5, when midline fusion of the bilateral cardiac primordia has normally occurred (Fig. 1, C and D). The two hearts in Foxp4 mutants were positioned bilaterally, suggesting a lack of proper migration of the precardiac primordia to the midline (Fig. 1, C to H). Foxp4 mutants exhibited grossly normal ventral morphogenesis and embryonic turning (Fig. 1, A to H), suggesting that cardia bifida was not due to secondary defects in these processes, as has been observed in other mouse models (36, 11). Upon sacrifice (E8.5 to E12.5), each of the two bilateral hearts was beating at approximately the same rate as in wild-type embryos, although they were asynchronous (10).

Fig. 1.

Loss of Foxp4 results in the formation of two hearts with extensive differentiation. (A) Wild-type and (B) Foxp4 mutant embryos at E11.5 were processed for whole-mount in situ hybridization with a probe for cTNI to detect cardiac myocardium (arrows and dotted line). FL, forelimb. (C to H) H+E–stained sections of wild-type [(C), (E), and (G)] and Foxp4 mutant [(D), (F), and (H)] embryos at E8.5 [(C) and (D)], E11.5 [(E) and (F)], and E12.5 [(G) and (H)], showing two well-developed hearts in the mutant embryos. H, heart; NT, neural tube. (I) Foxp4 mutant hearts have two ventricles, expressing (J) eHAND and (K) dHAND properly, i.e., in the left and right ventricle, respectively. Panels (I), (J), and (K) are adjacent sections of the same heart; bright interluminal fluorescence in (K) is due to autofluorescence from red blood cells. RV, right ventricle; LV, left ventricle; VS, ventricular septum. (L to N) Foxp4 mutant hearts exhibit formation of (L) endocardial cushions (arrow), (M) two atria, and (N) the endocardium (arrow). The atria are dorsal and anterior to the ventricles, and the neural tube, forelimbs, and eyes are positioned correctly in Foxp4 mutants. a, atria; v, ventricle. Scale bars, (C) and (D), 200 μm; (E) to (H), 400 μm; (I) to (M), 100 μm; (N), 50 μm.

Hematoxylin and eosin (H+E) staining showed that each heart in Foxp4 mutants had distinct atria and ventricles (Fig. 1, I and J). Proper chamber septation and left-right chamber specification was evident from the expression of eHAND and dHAND (Fig. 1, J and K). Endocardial cushions were also present (Fig. 1L). In both wild-type and Foxp4 mutant hearts, MLC2a and MLC2v were expressed in the atria and ventricles, respectively (Fig. 2, A to D), while plexin D1 was expressed in the developing endocardium (Fig. 2E) (12, 13). These data support the conclusion that bilateral heart tube fusion is not required for cell or chamber specification in the heart. Furthermore, the position of the atria dorsal and cranial to the ventricles suggests that looping morphogenesis occurs in the bilateral hearts of Foxp4 mutants despite the absence of midline fusion (Fig. 2, B and D).

Fig. 2.

Marker gene analysis demonstrates proper chamber and cell-type differentiation in Foxp4 mutant hearts. In situ hybridization to detect (A and B) MLC2v,(C and D) MLC2a, (F) ANF, and (G) N-myc expression and (E) double immunofluorescent staining to detect plexin D1 (green, arrow) and myosin-MF20 (red) protein expression. Analysis was performed on wild-type [(A) and (C)] and Foxp4 mutant [(B), (D), and (E) to (G)] embryos at E12.5. Scale bars, (A) to (D), 400 μm; (E), 50 μm, (F) and (G), 100 μm.

A hallmark of ventricular myocyte differentiation and maturation is the generation of compact and trabecular myocardium. Compact myocardium lies in the outer region of the ventricular wall and is more proliferative and less mature than trabecular myocardium (14). Development of distinct trabecular and compact myocardium is thought to enhance contractility and compartmentalization of oxygenated and unoxygenated blood before septation (14). The formation of trabecular myocardium is essential for cardiac function as demonstrated in neuregulin knockout mice that lack trabecular myocardium and die at E10.5 because of heart failure (15). Compact myocardium can be distinguished by expression of N-myc, whereas trabecular myocardium can be distinguished by expression of atrial naturetic factor (ANF) (16). To determine whether proper development of compact and trabecular myocardium occurred in Foxp4 mutant hearts, we performed in situ hybridizations to assess N-myc and ANF expression. N-myc and ANF were expressed in Foxp4 mutant hearts in the same pattern as in wild-type hearts, with expression of N-myc observed in compact myocardium and ANF expression observed in trabecular myocardium (Fig. 2, F and G) (15, 16). Taken together, these data demonstrate that Foxp4 mutants develop two hearts with proper chamber formation and normal trabecular and compact myocardial development.

The heart is the first organ to display asymmetry during development (17, 18). Along with other lateral-plate mesoderm derivatives, the heart expresses Pitx2, a bicoid-related homeodomain transcription factor, only on its left side; Pitx2 mutants exhibit multiple embryonic defects, including defective cardiac positioning after looping (19). At E10.5, Foxp4 mutants have two visible hearts that were positioned bilateral to the ventral midline (Fig. 3, A to D). Pitx2 expression was observed in the left heart of Foxp4 mutant embryos, whereas expression was not observed in the right heart (Fig. 3E). These data suggest that embryonic left-right asymmetry was retained in Foxp4 mutants. To determine whether cardiac specific left-right asymmetry was retained in Foxp4 mutant hearts, we performed in situ hybridization to determine expression of the basic helix-loop-helix (bHLH) transcription factors eHAND and dHAND and of FGF10, a member of the fibroblast growth factor family. eHAND is normally expressed primarily in the left side of the developing heart, whereas dHAND and FGF10 are expressed primarily on the right side of the developing heart (20, 21). We observed eHAND expression in the left ventricle of both Foxp4 mutant hearts (Fig. 3, D and E), whereas dHAND and FGF10 were expressed on the right side of Foxp4 mutant hearts, including the right ventricle (Fig. 3, F and G, and fig. S2). These data demonstrate that development of cardiac left-right asymmetry does not require heart tube fusion and that embryonic and cardiac asymmetry are specified by distinct mechanisms. These data demand a reconsideration of the currently accepted models of cardiac development and morphogenesis to accommodate the observation that precardiac mesoderm is capable of left-right asymmetrical chamber development, looping, myocyte differentiation, and endocardial cushion formation in the absence of midline heart tube fusion.

Fig. 3.

Left-right asymmetry is retained in Foxp4 mutant embryos. (A) At E10.5, Foxp4 mutants clearly have two hearts bilateral to the ventral midline. R, right heart; L, left heart. (B) H+E–stained section of E10.5 Foxp4 mutant. (C) In situ hybridization for Pitx2 demonstrates that only the left heart in Foxp4 mutant embryos is Pitx2 positive. (D to G) In situ hybridization using [(D) and (E)] eHAND, (F) dHAND, and (G) FGF10 probes to demonstrate that eHAND expression is observed in the left side (left ventricle) of Foxp4 mutant hearts [(E) is a higher magnification of (D)], whereas dHAND and FGF10 is expressed in the right side (right ventricle) of Foxp4 mutant hearts. Scale bars, (B) to (D), 300 μm; (E) to (G), 200 μm.

Our previous studies have shown that Foxp4 mRNA expression in the heart at E10.5 cannot be demonstrated by in situ hybridization, although it can be demonstrated in the adult heart by Northern blot analysis (8). However, abundant levels of Foxp4 protein expression in the embryo are observed in the anterior foregut endoderm (fig. S3F). Immunohistochemistry shows that Foxp4 protein expression is not observed in cardiac myocytes at E9.5 to E14.5 (fig. S2, A to C). In contrast, expression is observed in the epicardium and endocardium at these times (fig. S3, A to C).

Histological analysis identified defects in anterior foregut endoderm development in Foxp4 mutants. As shown in Fig. 1D, at E8.5, the anterior foregut was open in mutant embryos, whereas in wild-type embryos it was a closed tube. At E10.5, the most anterior aspects of the foregut remained open in Foxp4 mutants, whereas more posterior regions were closed (Fig. 4, A to C). By E11.5, the foregut had closed in Foxp4 mutants in the anterior region, but failed to separate into the esophagus and trachea; however, it did give rise to endodermally derived tissues, including the lung and liver (Fig. 4, D to I). At E11.5, terminal deoxynucleotidyl transferase–mediated deoxyuridine triphosphate nick end labeling assays revealed that anterior endoderm was highly apoptotic (fig. S4, A and B), and by E12.5, this region of the foregut had degraded, leaving a large open cavity in the embryo (Fig. 4, J to O). However, foregut endoderm was properly specified as demonstrated by expression of sonic hedgehog (SHH) and Foxa2 (fig. S4, C to H) (10). Together, these data suggest that Foxp4 mutants exhibit a delay in anterior foregut closure and cell death–mediated loss of anterior foregut endoderm after closure. We hypothesize that the defects in anterior foregut development may be responsible for cardia bifida in Foxp4 mutants. Although these defects led to a delay in foregut closure in Foxp4 mutants, the neural tube, limb buds, and head structures formed normally, indicating that the vast majority of ventral morphogenetic processes were unperturbed (Figs. 1,2,3).

Fig. 4.

Anterior foregut development in Foxp4 mutant embryos. (A to C) The anterior region of the foregut remains open at E10.5 in Foxp4 mutants, whereas the posterior region is closed (arrow). (D to I) In wild-type embryos at E11.5, the (D) anterior foregut is closed (green arrow) and has bifurcated into the [(F) and (H)] esophagus (red arrow) and trachea (blue arrow). In Foxp4 mutants at E11.5, the (E) anterior region of the foregut has closed but does not bifurcate into the [(G) and (I)] esophagus and trachea (green arrow). LU, lung. (J to O) At E12.5, wild-type embryos continue to show development of (J) the foregut (green arrow) including (L) the trachea (blue arrow) and esophagus (red arrow). [(K), (M), and (O)] By E12.5, the anterior foregut endoderm in Foxp4 mutants has degraded, leaving a large cavity in this region of the embryo (black arrowheads). HG, hindgut. Scale bars, (A) to (M), 100 μm; (N) to (O), 200 μm.

Defects in anterior foregut endoderm development have previously been associated with cardia bifida. Mutations in several zebrafish genes expressed in the foregut endoderm, including casanova (sox32) and faust (GATA5), result in cardia bifida, and these mutants display severe defects or complete lack of foregut endoderm development (22, 23). In mice lacking GATA4, anterior foregut endoderm development is defective, and GATA4-null embryos display cardia bifida (3, 4). Thus, the correlation between foregut defects and cardia bifida phenotypes is very strong. However, there are distinct differences between these mutants and Foxp4 mutants. GATA4 mutants have severe defects in ventral morphogenesis and lack proper cardiac chamber development (3, 4). Moreover, zebrafish cardiac morphogenesis is distinctly different from that in mammals, in that there are only two chambers and the heart does not loop or septate, but instead forms a serial connection between the single atria and ventricle. Although other models of cardia bifida have been reported, including Mesp1 and furin mutant mice, these embryos either die too early to examine the complex morphogenetic processes required for late-stage cardiac development or they exhibit other severe defects in general developmental processes such as embryonic turning (5, 6, 11). Thus, none of the cardia bifida mutants described previously has permitted the analysis of complex cardiac development in sufficient detail to determine whether the later stages of cardiac morphogenesis require heart tube fusion.

Foxp4 mutant embryos demonstrate that bilateral heart tube migration and fusion are not required for extensive cardiac development, including chamber formation, ventricular myocyte differentiation, looping, endocardial cushion formation, and development of cardiac left-right asymmetry. These data indicate a higher degree of preprogramming in the bilateral precardiac mesoderm than has been appreciated and indicate that a redefinition of the current model of cardiac development is required (1).

Supporting Online Material

Materials and Methods

SOM Text

Figs. S1 to S4

Table S1

References and Notes

References and Notes

View Abstract

Navigate This Article