SEARCH

SEARCH BY CITATION

Keywords:

  • heart;
  • looping;
  • extracellular matrix;
  • flectin;
  • laterality genes;
  • left–right;
  • CFC;
  • Pitx2;
  • Nodal;
  • asymmetry;
  • dorsal mesocardium;
  • midline;
  • plakoglobin;
  • secondary heart field;
  • chick embryo

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Dextral looping of the heart is regulated on multiple levels. In humans, mutations of the genes CFC and Pitx2/RIEG result in laterality-associated cardiac anomalies. In animal models, a common read-out after the misexpression of laterality genes is heart looping direction. Missing in these studies is how laterality genes impact on downstream morphogenetic processes to coordinate heart looping. Previously, we showed that Pitx2 indirectly regulates flectin protein by regulating the timing of flectin expression in one heart field versus the other (Linask et al. [2002] Dev. Biol. 246:407–417). To address this question further we used a reported loss-of-function approach to interfere with chick CFC expression (Schlange et al. [2001] Dev. Biol. 234:376–389) and assaying for flectin expression during looping. Antisense CFC treatment results in abnormal heart looping or no looping. Our results show that regardless of the sidedness of downstream Pitx2 expression, it is the sidedness of predominant flectin protein expression in the extracellular matrix of the dorsal mesocardial folds and splanchnic mesoderm apposed to the foregut wall that is associated directly with looping direction. Thus, Pitx2 can be experimentally uncoupled from heart looping. The flectin asymmetry continues to be maintained in the secondary heart field during looping. Developmental Dynamics 228:217–230, 2003. © 2003 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Upstream of the rightward looping observed during normal vertebrate heart morphogenesis are genes that are expressed in a distinct left–right manner in the embryo. Differences in the laterality pathways exist between chick and mouse (Schneider et al., 1999). Despite differences, the upstream laterality genes, Nodal, Lefty, Cryptic (the latter is a member of the EGF-CFC family of proteins) have been implicated in L/R axis determination, organ positioning, and in the direction of heart looping in all vertebrates analyzed. Members of the Nodal family in each vertebrate species analyzed (mouse Nodal; chick cNR1; frog Xnr1; zebrafish cyclops) are expressed unilaterally in the left, but not right, lateral plate mesoderm and, thus, are considered to confer “left-sidedness” in the embryo (Schier and Shen, 1999). In turn, the left-sided expression of Nodal leads downstream to the left-sided Pitx2 expression in the lateral plate and subsequently is in the left side of the straight heart tube (recently reviewed in Mercola and Levin, 2001). Thus, Pitx2 has been considered to be at the interface of the upstream L/R specifying pathway with morphoregulation of heart looping. The present experiments analyzing Pitx2 expression and flectin localization, however, suggest that Pitx2 expression and looping are independent events that normally are linked in development, but can be experimentally uncoupled (Patel et al., 1999; Schlange et al., 2001). This interpretation is also consistent with the observation that looping morphogenesis is not affected in Pitx2 null mutant mice (Kitamura et al., 1999; Lin et al., 1999; Lu et al., 1999). Rather, it is the sidedness of predominant flectin protein expression in the extracellular matrix of the dorsal mesocardial folds that is associated directly with looping direction at stages 10 to 11. In subsequent stages, the asymmetry of flectin continues to be maintained in the secondary heart field, as additional segments are added to the cephalic end of the tubular heart and the looping bend deepens.

Left–right (L/R) axis abnormalities are common in human development occurring at a frequency of 1 in 8,500 live births (Bamford et al., 2000). Mutations in laterality genes are associated with congenital defects in humans. For example, loss-of-function mutations in the EGF-CFC gene CFC1 are associated with left–right laterality defects such as heterotaxia, dextrocardia, transposition of the great arteries, and/or other laterality defects as polysplenia, L/R isomerism of the lungs, or stomach (Bamford et al., 2000). Mutations in the Pitx2/RIEG homeobox gene underlie another human syndrome known as Rieger (or Axenfeld-Rieger) syndrome (Semina et al., 1996). This is an autosomal dominant disorder resulting in haploinsufficiency and leads to anomalies that include cardiac, as well as ocular, craniofacial, and umbilical abnormalities (Semina et al., 1996; Lin et al., 1999; Lu et al., 1999). Defining how the upstream laterality genes impact on downstream events of heart looping is necessary for understanding the possible underlying mechanisms of heterotaxia or randomized organ positioning and of other human congenital heart defects that have a laterality-associated underpinning.

Members of the EGF-CFC family of proteins recently have been shown to be essential cofactors for Nodal signaling in the upstream laterality pathway in vertebrates (Shen and Schier, 2000). In contrast to the mouse which has two EGF-CFC genes which control anterior/posterior and left–right axis formation, only a single cDNA species, chick CFC, is reported in the avian genome (Colas and Schoenwolf, 2000; Schlange et al., 2001). The spatiotemporal pattern of chick CFC expression recently reported shows that it is asymmetrically expressed in the left side of Hensen's node between Hamburger and Hamilton (HH) stage 5 and 7 (Schlange et al., 2001). Symmetric expression domains were present in lateral plate mesoderm, the notochord, and in the prechordal plate. Antisense oligodeoxynucleotide treatment resulted in bilateral expression of Nodal, Pitx2, Nkx3.2, and Caronte due to a transient loss of Lefty 1 expression in the midline (Schlange et al., 2001). As a consequence, antisense treatment resulted in a high incidence of abnormal heart looping.

In a previous study in which we misexpressed Pitx2, it was observed that cardiac midline information appeared to be important for looping (Linask et al., 2002). As this importance had been noted by others also (Bisgrove et al., 2000), we undertook the present study to modulate specifically CFC gene expression to have not only the ability to modulate a midline gene, Lefty 1, but also to affect Pitx2 expression in a more “physiological” manner (Schlange et al., 2001). As previously, to follow effects on a marker, extracellular matrix (ECM), protein that is involved in the looping process, we assayed for the asymmetry pattern of flectin protein expression in the heart ECM (Linask et al., 2002). Based on the experimental manipulations of CFC gene expression, the resultant changes in Pitx2 gene expression, and flectin protein expression patterns, we propose that flectin localization patterns in the ECM are more directly associated with heart looping direction than is Pitx2 expression. Our analyses, as was suggested previously, highlighted also the importance of the dorsal mesocardium and the splanchnic mesoderm in association with the midline of the ventral floor of the foregut (Linask et al., 2002; Taber, 2001; see also Taber et al., 1995). These results have implications for mutations in human laterality genes and the downstream molecular changes that may result during development of the heart and possibly of other organ systems in which these proteins as flectin are expressed.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Importance of Flectin in the Dorsal Mesocardium and Splanchnic Mesoderm for Looping

Reported genetic misexpression studies in the chick usually end with a read-out of heart looping direction. We decided to take the analyses further downstream to address whether the expression of the specific matrix molecule flectin, that appears associated with the looping process, is changed by misexpression of upstream laterality genes during randomization of looping direction. Taking together (1) the asymmetric extracellular matrix characteristics of the chick heart that we have reported earlier (Tsuda et al., 1996; Linask et al., 2002), (2) the lack of a role for the cardiac jelly in looping in the rat embryo (Baldwin and Solursh, 1989), (3) biomechanical considerations (Taber et al., 1995), and (4) perturbation studies of flectin resulting in randomization of looping direction (Linask et al., 2002), these findings suggest that flectin in the dorsal mesocardial folds at the cardiac midline is important for establishing looping direction. To establish the validity of flectin within the dorsal region as being important to heart looping, we carried out two sets of experiments: (1) We treated chick embryos with hyaluronidase to remove extracellular matrix from the cardiac jelly and from the basal lamina of the myocardium, followed by flectin immunolocalization. (2) We treated chick embryos with flectin antibody that we previously had shown randomizes heart looping (Linask et al., 2002) and immunolocalized hLAMP-1, an extracellular matrix protein that has been shown to localize to the cardiac ECM during the looping stages (Sinning, 1997). Expression of hLAMP1 in both right- and left-looping hearts was then characterized and compared with that of flectin in the CFC misexpression experiments that resulted also in right- and left-looping hearts.

Hyaluronidase experiments using chick embryos.

Previous experiments in rat embryos demonstrated that even though degradation of hyaluronic acid in the ECM of the rat heart essentially removes the extracellular matrix of the cardiac jelly, the directionality and normal looping of the heart is maintained (Nakamura and Manasek, 1978; Baldwin and Solursh, 1989). We used hyaluronidase degradation here with avian embryos, as a means to remove the extensive ECM in the chick cardiac jelly and in the basal lamina of the myocardium to address whether flectin still remains localized anywhere within the heart ECM and, if so, where. The rationale was that any remaining localization of flectin in the ECM will provide information as to the region that may be important for the observed normal looping. Stage 5 chick embryos were incubated in the presence of hyaluronidase and after a 20- to 22-hr incubation period, the embryos had reached stages 12 to 14, a time period in which heart looping is well under way. As previously reported for the rat embryo, the rightward looping heart tube has collapsed into a flat structure. Despite the extensive removal of the cardiac jelly and the ECM of the myocardial basal lamina, flectin remains expressed in the left dorsal mesocardial fold and in the left splanchnic mesoderm that is closely apposed to the ventral floor of the foregut (Fig. 1). Although some flectin is also expressed within the right dorsal mesocardial fold, it is not as extensive as in the left. Only small areas of flectin are apparent in the myocardial wall. The accompanying diagram is to aid in understanding the dorsal mesocardial (DM) folds/splanchnic mesoderm (SM) areas of the collapsed heart. In summary, this experiment documents that flectin remains localized in the left dorsal mesocardium and in the left splanchnic mesoderm adjacent to the ventral floor of the foregut, after extensive removal of matrix molecules in a heart that continues to loop in the normal direction. Hence, this region was highlighted as an important area for heart looping and to be analyzed further in our CFC misexpression studies.

thumbnail image

Figure 1. Hyaluronidase-treated embryo immunostained for flectin. A: The heart has noticeably collapsed and is flattened. Although the cardiac jelly has been degraded, flectin continues to show asymmetric localization in the myocardial wall (MYO), primarily on the left side (L) in the left dorsal mesocardium (DM) and splanchnic mesoderm (SM; large arrows) now flattened against the floor of the foregut (FG). The small arrow points to limited flectin localization within right fold of the dorsal mesocardium. Little, if any, flectin is apparent in splanchnic mesoderm on right. B: Lower magnification for orientation of boxed-in region shown at higher magnification in A. The diagram depicts the relationships and foldings of the various tissues seen. NT, neural tube; EN, endocardium. Scale bars = 60 μm in A, 100 μm in B.

Download figure to PowerPoint

Localization of hLAMP-1 extracellular matrix protein in right- and left-looping hearts.

In this set of experiments, we determined whether the localization of another ECM molecule that is present during heart looping would be altered upon randomization of heart looping. Heart specific lectin-associated matrix protein-1 (hLAMP-1; Sinning, 1997) has been localized to the extracellular matrix of the looping chick heart and is present also in the extensive basal lamina of the chick embryo (Fig. 2). We incubated stage 5 chick embryos in the presence of flectin monoclonal antibody that we demonstrated previously results in both normally right-looping and abnormally left-looping hearts (Linask et al., 2002). Control untreated hearts (Fig. 2A), and flectin antibody-perturbed right- (Fig. 2B) and left- looping (Fig. 2C) hearts were then immunostained with an antibody for hLAMP-1. As seen in the control embryo, hLAMP-1 localizes to the cardiac jelly ECM and the ECM of the basal lamina and to an attachment region of the endocardium to the midline of the ventral foregut floor (shown in Fig. 2A). hLAMP-1 is not expressed, however, within the myocardium, in the dorsal mesocardium, or splanchnic mesoderm adjacent to the foregut wall. In the flectin perturbed right-looping (Fig. 2B) and left-looping (Fig. 2C) hearts, the ECM appears somewhat diminished. However, the hLAMP-1 expression pattern does not change and continues to be expressed in the same manner in the basal lamina and cardiac jelly of the heart, regardless of sidedness of looping. This finding is in contrast to the modulation of the specific asymmetric expression pattern of flectin within the dorsal mesocardium/splanchnic mesoderm that correlates closely with the change in looping direction reported previously (Tsuda et al., 1996; Linask et al., 2002). Note also in the left-looping heart, the foregut is displaced well to the left of the embryo midline, as defined by the position of the notochord. The heart is positioned relative to the midline of the ventral floor of the foregut and not relative to the embryo midline.

thumbnail image

Figure 2. Immunolocalization of hLAMP-1 in the control, untreated, heart of a stage 12 chick embryo, after a 22-hr incubation period in culture (A) and in flectin antibody perturbed right- (B) and left-looping (C) hearts. A: hLAMP-1 localizes primarily to the basal lamina of most developing tissues. In the heart, it is detectable also in the cardiac jelly, in association with the endocardium, in the basal lamina of the myocardium, and in the extracellular matrix (ECM) between the splanchnic mesoderm and the apposing wall of the foregut (FG). Little, if any, hLAMP-1 is detectable within the myocardium, in the dorsal mesocardium (see arrowheads), or splanchnic mesoderm. In flectin antibody perturbed hearts, less cardiac jelly was apparent in both right-looping (B) and left-looping (C) hearts. There was no major change relative to the controls in localization of hLAMP-1, as it continues to be expressed in the basal laminae, endocardium, and cardiac jelly. No expression of hLAMP is observed in the dorsal mesocardial folds (arrowheads) or in the splanchnic mesoderm as was seen with flectin. Note in C of the left-looping heart that the FG is displaced laterally far to the left of the embryo midline as defined by the notochord (N). hLAMP-1 is also highly expressed in the ECM in association with endocardial cells seen in this region adjoining the ventral foregut midline (white arrow points to the hLAMP-1 in this midregion near foregut floor). NT, neural tube. Scale bars = 50 μm in A (applies to A,C), 100 μm in B.

Download figure to PowerPoint

CFC Misexpression

CFC misexpression was carried out in the present study because it has two principal effects: it transiently represses Lefty-1 expression at the midline and leads downstream to bilateral expression of Pitx2 (Schlange et al., 2001). Results of chick CFC misexpression using antisense oligonucleotides on heart looping, Pitx2 expression, and flectin localization are shown in Figure 3A–O, Figure 4A,B, and Figure 5A–D, in Tables 1 and 2, and a control heart in Figure 6. Forty-three anterior halves of the antisense-oligonucleotide–treated embryos were immunostained for flectin as a cardiac extracellular matrix asymmetry marker, and 31 of these embryos were sectioned anteriorly to posteriorly through the heart (Fig. 3). Posterior part of each embryo cut just below the heart was analyzed for Pitx2 mRNA expression by whole-mount in situ hybridization. The Pitx2c in situ hybridization and flectin protein patterns are indicated for all the different classes of embryos examined (see Tables 1, 2). The relationship of the heart to the mid-region of the ventral floor of the foregut and to the embryo midline was noted also in the sectioned embryos. Representative embryos are shown in the figures.

thumbnail image

Figure 3. Misexpression of CFC and resultant flectin localization patterns. Ventral view of hearts is shown in whole-mounts. Right (R) and left (L) sides are the same for all panels. In all rows of panels corresponding to different embryos, the left panel shows the in situ hybridization pattern for Pitx2 mRNA expression, the center panel shows flectin localization in the whole-mount, and the right panel shows a section through the same heart shown in the middle panel. A–C: Heart is looping to the right as seen in embryo in B. In this embryo, Pitx2 mRNA expression was on the left (green arrows, A). This heart when sectioned displays predominant localization of flectin in left dorsal mesocardial fold and in adjacent myocardial wall (in C, large green arrow). D–F: Heart is looping to the right (E) within an embryo showing bilateral Pitx2 localization (D, green arrows). When sectioned, the heart shows slightly more flectin on left side (F). Note in F and in O that the midline of the foregut (FG) is displaced more to the right or left of the embryo midline, as defined by the notochord's (N) position than is normally seen (compare with foregut localization in C, I, and L). The heart develops, however, in relation to the midline of the ventral floor of the foregut in all embryos analyzed. G–I: Heart shows a definite leftward loop. This embryo also showed a bilateral Pitx2 localization (G, green arrows). A predominant right-sided flectin localization pattern is now apparent (I, larger green arrows. Small arrows show comparable region on other side of tube with little flectin expression detectable). J–L: In this whole-mount, the heart appears unlooped (K), while Pitx2 pattern was left-sided (J) as is normally observed for Pitx2 expression. The heart sectioned displayed a symmetric expression of flectin in the dorsal mesocardial folds and adjacent myocardial wall. This finding is consistent with earlier observations of hearts that do not loop. M–O: Heart in the whole-mount appeared to be not looping (seen in N). The Pitx2 expression pattern was bilateral (see green arrows in M). This heart sectioned showed a relatively low level of intensity for flectin protein expression, as well as a symmetrical pattern of flectin within the mesocardial fold regions and adjacent myocardium (O, arrows). Scale bars = 100 μm in C (applies to C,F,I,L), 50 μm in O, 200 μm in E (applies to B,E,H,K,N).

Download figure to PowerPoint

thumbnail image

Figure 4. This CFC-treated embryo displayed an abnormally, leftward, looping heart with a normal, left-sided Pitx2 mRNA expression pattern. A: As was expected with leftward looping hearts, the sectioned heart showed a predominant right-sided flectin expression pattern (“painted” yellow). The rectangle demarcates the region in which the area painted was measured. B: A similar area was defined in the red on the left side, and the painted region was measured. The areas were expressed as left–right ratios and were defined as asymmetry quotients (AQ) to obtain relative measurements of flectin localization. FG, foregut; R, right; HT, heart.

Download figure to PowerPoint

thumbnail image

Figure 5. Some hearts were described as S-shaped (A shows the whole-mount of one such heart). These embryos showed normal, left-sided Pitx2 expression. Heart in A is shown sectioned in B–D. The anterior region has normal rightward bending with predominant left-sided flectin localization. The flectin localization area defined in red relative to yellow was measured anteriorly (B,C), as well as posteriorly (D). The more posterior heart regions of this same heart express little flectin and show no extracellular matrix predominance, and, as a result, this part of the tube does not show bending and appears in whole-mounts to be S-shaped (A). L, left; R, right; Ant, anterior; AIP, anterior intestinal portal; N, notochord; FG, foregut; MYO, myocardial wall. Scale bar in D = 100 μm in A, 50 μm in D (applies to B,C).

Download figure to PowerPoint

Table 1. CFC-Treated Embryos
SitusPitx2 expressionNumber (n)Embryos sectioned
R-loopLeft-sided137
UnloopedLeft-sided126
R-loopBilateral33
UnloopedBilateral33
L-loopBilateral33
R-loopWeak staining33
L-loopLeft-sided11
S-shapedLeft-sided55
Table 2. Asymmetry Quotients (AQ) of CFC Antisense-Oligonucleotide–Treated Embryos
SitusFlectin AQaFigure no.Pitx2 expression
  • a

    An AQ (L/R) quotient >1 = more flectin on the left; AQ = 1, symmetrical expression; AQ < 1, more flectin is on the right.

  • b

    First quotient in column is for embryo shown in figures.

  • c

    S-shaped hearts: right loop/Ant, anterior part of heart; Pos, posterior part of same heart.

Right4.6b; 3.3; 2.33CLeft-sided
Right1.1; 1.4; 1.13FBilateral
Left0.2; 0.8; 0.73IBilateral
Left0.64A,BLeft-sided
Unlooped1.0; 1.03LLeft-sided
Unlooped0.9; 1.0; 0.93OBilateral
S-shaped   
 R-loop/Antc1.7; 1.25B,CLeft-sided
 No loop/Posc1.0; 0.85DLeft-sided
thumbnail image

Figure 6. A control embryo treated with scrambled oligonucleotides displays rightward looping. The area defined in yellow (A) relative to the area defined in red (B) was determined to have an asymmetry quotient of 2.2, consistent with rightward looping and more flectin in the left dorsal mesocardial fold and adjacent splanchnic mesoderm than on the right. FG, foregut; En, endocardium. Scale bar = 50 μm in B (applies to A–B).

Download figure to PowerPoint

In Table 2, the first asymmetry quotient (AQ) given depicts the asymmetry quotient that was calculated for the embryo shown in the figures. The additional AQs represent quotients that were calculated for flectin localization in hearts of embryos within the same class but are not shown separately in the figures. However, they are the same with regard to Pitx2 expression and direction of heart looping. A statistical analysis was not attempted due to reasons given in Experimental Procedures section.

Figure 3A–C shows anterior and posterior halves of a CFC antisense-oligonucleotide–treated embryo with a rightward looping heart. This embryo displayed a left-sided Pitx2 mRNA expression pattern (Fig. 3A, green arrows). The anterior half of the embryo shown in Figure 3A was immunostained for flectin (ventral view shown of whole-mount in Fig. 3B: heart wall is outlined in yellow) and the same heart sectioned (Fig. 3C). A dominant, left-sided, flectin localization is apparent in the left dorsal mesocardial fold (Fig. 3C, large green arrow) and in the outer curvature of the heart. The same pattern is seen also in control and untreated embryos (not shown). The L/R asymmetry quotient equals 4.6, which is consistent with a predominant left-sided flectin expression (see Experimental Procedures section and Fig. 4).

Figure 3D–F shows an embryo in which the heart looped to the right, but Pitx2 expression was bilateral (Fig. 3D, green arrows). In sections through the heart, greater localization of flectin is seen in left, outer curvature of the myocardial wall (Fig. 3F, large green arrows) than in comparison to the same regions of the right myocardial wall (smaller green arrows). Although flectin is present in the dorsal mesocardial fold on the right side, it is almost absent where the bend actually occurs. In contrast to the heart in the Figure 3A, there is now Pitx2 present also on the right side. Although the detectable flectin matrix difference is small in this section of an anterior region of the heart, there is more flectin on left (AQ = 1.1) and a rightward bend has occurred. The higher flectin expression seen on the dorsal side of the foregut is often, but not always, observed. Flectin is also observed in the extraembryonic splanchnic membrane covering the heart and also is seen at high levels in this embryo. It should be noted in this embryo that the midline of the ventral floor of the foregut (FG) has shifted more than is normally observed to the right of the embryo midline (as defined by notochord, N, location). A similar shift was also seen earlier, when Pitx2c was directly misexpressed (Linask et al., 2002). The heart develops, however, in relation to the ventral foregut midline, not embryo midline, as defined by the position of the notochord.

In an abnormally, leftward looping heart shown in Figure 3G–I, bilateral Pitx2 expression was also observed in this embryo. When this heart was sectioned, a predominance of flectin was observed in the right dorsal mesocardial fold and right outer curvature of the heart in comparison to the left (AQ = 0.2). The asymmetry within the dorsal mesocardial area of this abnormally looping heart is seen clearly in the section in Figure 3I.

In hearts that were classified as “unlooped” based on whole-mount observation (Fig. 3K,N), the Pitx2 mRNA expression differed: in embryo 3K, Pitx2 was left-sided only (in situ hybridization pattern for Pitx2 mRNA expression shown in Fig. 3J). This heart, when sectioned (Fig. 3L), showed relatively equal intensity for flectin in the left and right myocardium and dorsal mesocardium (AQ = 1.0). The heart tube did not show any distinct looping direction. The unlooped heart (shown in Fig. 3N) displayed bilateral Pitx2 expression (in situ hybridization in Fig. 3M). This heart also displayed relatively symmetric, although in contrast to heart in Figure 3L, little flectin expression in the two sides of the tubular heart (Fig. 3O) and in the dorsal mesocardial folds (AQ = 0.9). Thus, these results provide additional evidence that regardless of Pitx2 expression being left only or bilateral, the asymmetric predominance of flectin, or sometimes symmetric expression, in the dorsal mesocardial folds and splanchnic mesoderm, as well as in the myocardial wall, is associated directly with the direction of heart looping, or no looping, respectively. Note also in these embryos that the midline of the ventral floor of the foregut (FG) may be slightly to the left or right of the embryo midline, as defined by the location of the notochord (N). As seen with embryonic hearts in all panels, the heart and dorsal mesocardial folds develop in relation to the midline of the ventral floor of the foregut, not the embryo midline. Thus, the positioning of the foregut, and specifically the ventral floor, defines the position of the cardiac midline. For all 31 embryos analyzed, the results were similar to these shown.

Among the CFC antisense-oligonucleotide–treated embryos one embryo only showed abnormal leftward looping, but a normal, left-sided Pitx2 expression. Based upon the above results, one would predict that this heart when sectioned would have a predominant right-sided flectin expression. This was found to be the case (Fig. 4A,B). This figure also shows how we obtained the AQ indicated above. In Figure 4A, brighter intensity values were painted in yellow for the right side and red on the left. The areas “painted” within the same fixed region defined by the same-sized rectangular box for each side were then calculated in numbers of pixels (3 × 3) and are expressed as an asymmetry quotient L/R obtained from dividing the means calculated for the two sides. The AQ for the heart region in this embryo equals 0.6, consistent with a leftward looping direction taken.

In embryos in which looping is described as being S-shaped (see also Schlange et al., 2001), the anterior region is beginning to loop to the right, but the posterior region is seemingly bending to the left or is unlooped, thus giving an S-shape to the heart. (See whole-mount of S-shaped heart in Fig. 5A. AIP designates the anterior intestinal portal. Arrow shows plane of anteroposterior axis). In one such heart, the anterior region shows normal rightward bending with predominant left-sided flectin localization (AQ = 1.7; Fig. 5B,C). The more posterior heart regions of the same heart tube, however, express little flectin as yet and show no ECM predominance. This finding correlates with the part of the cardiac tube that does not show bending (AQ = 1; Fig. 5D). Thus, the heart appears in whole-mounts to be S-shaped. In some embryos in the S-shaped heart group, a slight left predominance of expression was apparent. One would expect these hearts to continue bending in the rightward direction as development continues.

Figure 6A,B shows a control heart of an embryo treated with a scrambled sequence of oligonucleotides. This embryo had developed to a slightly older stage than seen in previous figures. Flectin AQ based on measurements in the dorsal mesocardial folds was 2.2, consistent with hearts that loop to the right and display left-sided Pitx2 expression.

Relationship to Secondary Heart Field and Pharyngeal Region

Once looping direction is specified in anterior regions within chick stages 9/10 and becomes detectable at stage 11, heart morphogenesis proceeds directly into the next stage with the addition of new segments to the outflow region from the secondary heart field, as well as to the caudal inflow region beginning with stage 12 (de la Cruz et al., 1991; Waldo et al., 2001). The secondary heart field encompasses the splanchnic mesoderm that is adjacent to the ventral floor of the gut tube (caudal pharyngeal region) and is continuous with the dorsal mesocardial folds. At stage 12, flectin continues to be asymmetrically expressed in the secondary heart field, primarily in the left splanchnic mesoderm adjacent to the lateral wall of the pharyngeal foregut and in the left dorsal mesocardial fold (see arrows in Fig. 7A and in Fig. 7B of same region shown at higher magnification). The dorsal mesocardial region, as we specified above, continues to be closely associated with the endoderm of the ventral floor of the foregut, specifically encompassing this mid-region of the foregut floor. A strand of endocardial cells pass through the sleeve of the dorsal mesocardium to associate with the gut tube cells at the midline of the ventral pharynx.

thumbnail image

Figure 7. Stage 12 normal embryo immunostained for flectin. A: For orientation, A shows a section at lower magnification through the heart and the secondary heart field. Note the left (L) and right (R) dorsal mesocardial areas and the splanchnic mesoderm adjacent to ventral foregut (FG) endoderm continue to show asymmetry in that more flectin is seen in left side than in the right (compare regions indicated by arrows). B: At higher magnification showing flectin immunolocalization, as well as the alignment of embryo midline as designated by the notochord (N) with the ventral floor of the foregut endoderm with which the endocardial cells (EN) associate as a cord of cells in between the dorsal mesocardial folds. NT, neural tube. Scale bars = 100 μm in A, 50 μm in B.

Download figure to PowerPoint

From our previous studies on N-cadherin/β-catenin localization during heart development, we noted that catenins are involved in compartment formation in the early embryo (Linask et al., 1992, 1997). We now document that β-catenin–like plakoglobin defines a compartment of cells within the foregut floor endoderm at stages 11 and 12. A distinct compartment of cells within the foregut endoderm is delineated by plakoglobin-mediated cell to cell junctions (see arrows in Fig. 8A, showing a stage 11 embryo and Fig. 8B, a stage 12 embryo). The arrowheads in Figure 8B designate the lateral boundaries of the compartment. It is this ventral endoderm compartment with which the endocardial cells (EN) associate and that the dorsal mesocardium/splanchnic mesoderm encompasses at the cardiac midline. If Pitx2 or flectin is misexpressed, often the position of the foregut is seen to be more lateral to the left or right than normal (see Fig. 2; also Linask et al., 2002). The position of the heart is defined in all of our analyses by its association with the foregut endoderm compartment.

thumbnail image

Figure 8. Localization of plakoglobin in stage 11 (A) and stage 12 (B) embryos. Similar region as shown above. Expression of plakoglobin is detectable within cell–cell junctions (see arrows) of a compartment of endoderm cells at the mid-region of the ventral foregut floor. Arrowheads in B designate the lateral boundary of pharyngeal endoderm compartment at stage 12. This compartment may be a signaling component during development of the secondary heart field into additional cardiac segments. This addition of segments occurs during subsequent looping, right after looping direction has been specified. DM, dorsal mesocardial fold; EN, endocardial cells; FG, foregut; N, notochord; NT, neural tube. Scale bar = 50 μm in A (applies to A,B).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

In an earlier analyses of the asymmetric expression of the extracellular matrix (ECM) protein flectin, we demonstrated the persistence of a left-sided dominance from its first detection at stage 7+/8 in the heart forming lateral plate mesoderm and later in the myocardium and cardiac jelly throughout heart looping stages (Tsuda et al., 1996, 1998). Flectin (also known as F-22 antigen) is an extracellular matrix molecule of Mr 250 × 103 that was first isolated from the extracellular matrix of the interphotoreceptor layer of the developing eye in the 10-day chick embryo (Mieziewska et al., 1994a, b). Flectin cDNA has been sequenced, and its characterization will be published independently by others. A class of matrix proteins, the matricellular proteins with which flectin appears to have similarities functionally, have the ability to interact with multiple cell-surface receptors, cytokines, growth factors, proteases, and structural proteins (Yang et al., 2000). Flectin may also have a possible nuclear regulatory role in some tissues (unpublished observations). Thus, these types of matrix proteins may play a role as adaptors and effectors of cell–matrix interactions. The present focus of our analysis was to determine whether misexpression of the CFC gene is interfaced with morphologic regulation of heart looping using flectin as a marker for an asymmetrically expressed protein in the heart.

Morphoregulation of Heart Looping

On the organ level, the first break in embryonic symmetry begins with the heart bending to the right. Embryonic heart looping begins after the fusion of the left and right epithelial cardiac compartments to form a straight, single-chambered, tubular structure. The biophysical reason why the heart loops to the right appeared related to an asymmetrical extracellular matrix expression in the left heart field relative to the right (Tsuda et al., 1996, 1998; Smith et al., 1997; Linask et al., 2002). The complexity of the looping phenomenon is illustrated by several levels of regulation which, when perturbed, disrupt normal looping directionality. Perturbation of different upstream left/right axis specifying regulatory genes, extracellular matrix molecules, embryonic flexure, or cytoskeletal molecules, all disrupt the normal direction of looping (Itasaki et al., 1991; Mercola and Levin, 2001; Linask et al., 2002).

Our interest in laterality gene expression and directionality of heart looping results from the identification of an extracellular matrix molecule, flectin, that is expressed first in the left heart field in the chick at the three- to five-somite stage and then subsequently is expressed in the right (Tsuda et al., 1996; Linask et al., 2002). This left–right delay in the timing of flectin expression creates a detectable physical asymmetry of the heart tube before looping. This finding is diagrammatically depicted in Figure 9. In a stage 10–11 embryo, after the two bilateral cardiac areas have fused at the midline, flectin is expressed asymmetrically within the dorsal mesocardial folds, splanchnic mesoderm adjacent to the midline of the ventral floor of the foregut and within the myocardium. When flectin protein interactions are perturbed using antibodies, directionality of heart looping is to the left (30%) as well as to the right (35%), or no looping occurs (Linask et al., 2002). Another extracellular matrix molecule hLAMP-1 that is present during the looping stages (Sinning, 1997), shows a different pattern of expression than does flectin in the heart, although in some regions they colocalize. The expression of hLAMP-1 is not altered in abnormally left-looping hearts, suggesting that flectin has a role distinct from hLAMP-1 in heart looping. As the heart bends to the right at stage 12, flectin remains localized asymmetrically within the dorsal mesocardial folds and the splanchnic mesoderm adjacent to the ventral floor of the foregut, as well as in the myocardium. The splanchnic mesoderm near the foregut wall at stage 12 is referred to as the secondary heart field, and flectin continues to be consistently expressed at higher levels in the left side of this field and in the left dorsal mesocardium (Tsuda et al., 1996; Linask et al., 2002).

thumbnail image

Figure 9. Diagram of relationship of flectin expression to key regulatory molecules in normal left–right (L/R) axis determination in the normal sequence of events (A) and under conditions of misexpression of CFC using antisense oligonucleotides (B). The midline of the embryo is indicated by a line, with the left side of the embryo to the readers' left in this dorsal view of the embryo. Chick CFC is required for Nodal maintenance. Nodal expression results in the activation of Pitx2c in the left LPM. Pitx2c activates an unknown molecule X that affects the timing of flectin expression in the left LPM. Probably Nodal can affect this putative factor “X” independent of Pitx2c, as looping and Pitx2c expression can be experimentally uncoupled. This signal is relayed across the midline to initiate flectin synthesis also on the right. The delay in L/R timing leads to flectin protein asymmetry. The absence of CFC by antisense oligonucleotide exposure (B) leads to bilateral Nodal and Pitx2c expression. It is suggested that flectin-dominant sidedness is dependent on the concentrations of Nodal (or Pitx2) achieved in either the left or right lateral plate, which determines the timing of flectin activation in the left and right sides. AQ, asymmetry quotient; LPM, lateral plate mesoderm.

Download figure to PowerPoint

The most noticeable result of the upstream modulation of CFC, as well as Pitx2c misexpression in the chick, was the downstream perturbation of the normal pattern of flectin extracellular matrix asymmetry in the developing heart and specifically in the dorsal mesocardial folds/splanchnic mesoderm. The perturbed flectin asymmetry appears to be directly associated with the observed randomization of heart looping. Because human RIEG/Pitx2 mutations result in haploinsufficiency, our analyses suggest that these effects are dependent on threshold concentrations of Pitx2 approximately during stages 7 to 8 in the left heart field which, in turn, may affect a parallel pathway that leads downstream to flectin expression in the heart (see diagram, Fig. 9). If Pitx2 is misexpressed bilaterally, the timing of flectin is determined by the relative concentrations of Pitx2 in the bilateral fields. The resultant left–right threshold concentrations of flectin that are synthesized may determine the direction of heart looping. As is detectable from the results, even minor differences in matrix asymmetries can have a large impact on normal morphogenetic processes to result in organ-level anomalies. Further biophysical and biochemical experiments are ongoing to fully understand how dorsal mesocardial/splanchnic mesoderm matrix asymmetry, but not of the cardiac jelly/myocardium, imparts direction to the bending of the heart tube.

CFC modulation of Nodal/Pitx2c expression and its impact on flectin's predominance on one side of the heart tube in comparison to the other side suggests that Nodal/Pitx2c indirectly by means of a factor “X,” as shown in Figure 9, modulates a pathway that regulates flectin expression. An indirect effect is suggested, because flectin is expressed normally in both left and right lateral plate mesoderm, while Nodal/Pitx2 is expressed only in the left. However, it is the timing of flectin protein synthesis in the left lateral mesoderm that results in its asymmetry (Linask et al., 2002). This timing means that Pitx2c's role in these events involves either activating an intermediary molecule or derepressing a repressor molecule in the left lateral mesoderm. This signal may subsequently be relayed in some manner across the midline to the right heart field. Involvement of a transforming growth factor-beta (TGF-β) family member, as Lefty-1, in this midline relay is a possibility. Presently, Lefty-1 at the midline is thought to be acting primarily as a midline inhibitor antagonizing bilateral Nodal expression (Meno et al., 1998; Rodriguez Esteban et al., 1999). The antagonism of Nodal appears to be the main effect in our studies also, thus allowing for a bilateral Pitx2 expression pattern. A Lefty-1 involvement also in the “relay” of activation of flectin expression across the midline, however, seems plausible because many TGF-β family members are known to regulate synthesis of extracellular matrix molecules. For example, Lefty has been shown to have an active role in extracellular matrix remodeling in association with a fibrosarcoma model (Mason et al., 2001).

Modulation of flectin expression by several experimental approaches, including a loss-of-function antisense approach to misexpress CFC reported here, and second, a loss-of function and overexpression of Pitx2 reported earlier (Linask et al., 2002), have suggested that cardiac looping is regulated by multiple signaling inputs and parallel pathways that cross-talk (see also Tsuda et al., 1998, for discussion in relation to mouse). This study extends the earlier suggestions of cross-talk between parallel pathways and that Pitx2 expression can be experimentally uncoupled from looping (Patel et al., 1999; Schlange et al., 2001). Flectin matrix asymmetry within the dorsal mesocardium/splanchnic mesoderm and myocardium, however, is an apparent common endpoint for these parallel pathways in regulating looping direction.

An Importance for the Midline of the Ventral Floor of the Pharynx and Secondary Heart Field

An importance for the dorsal mesocardium at the cardiac midline during looping had been suggested upon biomechanical considerations (Taber et al., 1995; reviewed in Taber, 2001). In addition, an importance for the left dorsal mesocardial folds and the splanchnic mesoderm region also became evident from hyaluronidase degradation of the cardiac jelly in the chick. In the hyaluronidase-treated chick embryos, hearts looped normally and flectin continued to show a normal asymmetric, predominant left-localization in the dorsal mesocardium, in the adjacent splanchnic mesoderm, and in the myocardium. Molecules within the dorsal mesocardial region and cardiac midline structures, including the positioning of the ventral floor of the foregut, are of morphoregulatory importance for not only heart looping but also for postlooping heart development (manuscript in preparation). As shown here, within this mid-region of the ventral endoderm, a compartment of cells can be defined based on their cell–cell associations that are mediated by plakoglobin. Plakoglobin and β-catenin are distinct molecules but closely related structurally and functionally that associate with cadherins at the cell membrane (Peifer et al., 1992; Wheelock et al., 1996). β-Catenin/plakoglobin are generally associated with compartment development in embryos (Fagotto and Gumbiner, 1994). We demonstrated earlier that N-cadherin/β-catenin delineate the cardiac compartments bilaterally (Linask, 1992a, b). Such compartments are defined in the embryo to coordinate cell function, whether in inductive events or in tissue formation (Gurdon, 1987; Gurdon et al., 1993). By stage 14 this dorsal region begins to express Gata-4, Nkx 2.5, and FGF-8, possibly involved in inducing myocardialization of the splanchnic mesoderm of the secondary heart field, as it is incorporated into the heart tube (Waldo et al., 2001). Other recent studies also suggested that such an endodermal foregut signaling center must exist (Smith et al., 2000) and that it relates to heart development (Lickert et al., 2002).

As discussed above, the ventral floor of the foregut appears not only to have an important signaling role during the pre- and early-looping stages shown here, but also during late looping beginning with stage 12 with the development of the secondary heart field (see also Waldo et al., 2001). The secondary heart field is encompassed by the dorsal mesocardial folds and extends into the splanchnic mesoderm adjacent to the ventral floor of the foregut. These regions continue to show an asymmetric expression of flectin into stage 12 (Fig. 7A,B) and beyond. At earlier stage 9, it is this anterior region that determines the direction of looping, as based upon our experimental manipulations of flectin (Linask et al., 2002). Looping direction is easily discernible by stage 11. Another 5–8 hr later (stages 12–13), this area is designated the secondary heart field (de la Cruz et al., 1991; Markwald et al., 1998; Waldo et al., 2001). At stage 12, as based on marking experiments, new cephalic cardiac segments begin to add on to the initial ventricular region of the straight heart tube (de la Cruz et al., 1991). By stage 14, signaling within the secondary heart field is facilitated possibly by BMP-2 and FGF-8 in the splanchnic mesoderm adjacent to the foregut wall (Waldo et al., 2001) and by GATA-4 in the ventral foregut endoderm compartment that also expressed GATA-4 at earlier time-points, i.e., in the bilateral anterior endoderm underlying the splanchnic precardiac mesoderm. Thus, due to inductive signaling in this area, new segments are added on to the straight heart tube to form the outflow tract myocardium. It is suggested that, by the addition of new cardiac segments, this process enables the heart bend to deepen. In addition, this newly formed myocardium will continue to show left–right asymmetry of flectin, as flectin continues to be expressed asymmetrically within the secondary heart field throughout looping stages.

It becomes understandable why looping has been considered a key aspect of heart morphogenesis and that any minor perturbation of this process can result in severe congenital malformations. Given the complexities of matrix composition and molecular interactions and that some of these matrix molecules can serve as cell signaling molecules, it enables us to begin to envisage how molecular asymmetries can regulate not only biochemical parameters of the organ, but also biophysical. Mutations within laterality genes CFC, Pitx2, and other genes that affect any of the described features and the matrix asymmetries around the midline can result in anomalies of looping. Abnormal looping eventually results in the misalignment of the subsequent development of cardiac chambers, valves, and septa, as well as blood vessels entering the heart, leading to a spectrum of congenital anomalies.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Chick Embryos

Fertilized chick eggs (Charles River, Sulzfeld, Germany) were incubated until embryos reached stage 5 in a humidified incubator at 38°C. Staging is according to Hamburger and Hamilton (1951).

Immunohistochemistry and Microscopy

Embryos were fixed in Histochoice (Amresco, Sohon, OH) and processed for immunohistochemistry as previously described (Linask and Tsuda, 2000). Araldite sections were analyzed by using a Nikon Optiphot II fluorescence microscope with appropriate epifluorescence filters. Digitized images were captured by a high-resolution Princeton Micromax CCD camera and were analyzed by using Image-Pro Plus (Media Cybernetics, Silver Spring, MD) or Metamorph (Universal Imaging Corporation, Downington, PA) software.

Antibodies

The flectin (previously named F22) mouse monoclonal antibody was generously provided by Dr. Nancy Philp, Thomas Jefferson University (Philadelphia, PA). The hLAMP-1 antibody was generously given to us by Dr. Allan Sinning, University of Mississippi Medical Center, Jackson. Mouse monoclonal antibody against plakoglobin was obtained from Biogenesis. Cy-3–conjugated secondary antibody (Jackson Immunochemicals) was used to visualize primary antibody localization in the embryo.

Antisense Experiments

The antisense oligonucleotide (antisense ODN; Eurogentec, Cologne) used in this study was targeted against nucleotide (nt) 234-253 of chick CFC cDNA (antisense1: 5′-A*T*G*TTTTCGCCAGAA*C*A*T-3′; the asterisks indicate phosphorothioate modification of the phosphate backbone). The control ODN corresponded to the reverse sequence of antisense1 (control1: 5′-T*A*C*AAAAGCGGTCTT*G*T*A-3′). The ODNs were applied in a final concentration of 20 μM in a 20% (w/v) solution of Pluronic F-127 (Sigma, Steinheim, Germany) in Pannett-Compton saline. Approximately 2 μl of ODN-containing solution of Pluronic F-127 was applied to the node region of the embryos at room temperature by using a cold glass capillary. The solution of Pluronic F-127 forms a gel on the surface of the embryo within a few seconds, thereby limiting the area of exposure to the ODNs to the region around Hensen's node and the forming notochord (for further detail, see Fig. 4 in Schlange et al., 2001). Embryos were placed into New culture (New, 1955) and were subsequently incubated overnight until they reached HH stage 10–11. The embryos were then fixed and cut in half just below the heart. The anterior half was processed for flectin protein localization using immunohistochemistry; the posterior half for Pitx2 mRNA localization using in situ hybridization according to a previously described protocol (Schlange et al., 2001).

Forty-three embryos were treated with CFC antisense oligonucleotides and were processed for immunohistochemistry for flectin and for Pitx2 expression by in situ hybridization. Thirty-one of these embryos were sectioned as shown in Table 1 and were then analyzed for flectin. Table 1 tabulates the numbers of sectioned embryos within each class that were analyzed. Classes were based upon direction of heart looping and Pitx2c expression. Not all embryos showing normal looping direction and normal left-sided Pitx2c expression were sectioned, as the cardiac flectin pattern in this class has been well documented previously (Tsuda et al., 1996; Linask et al., 2002) and was observed to be the same in all embryos sectioned here as well. Also half of left-sided, unlooped hearts were sectioned (6 of 12), because all that were sectioned showed similar results, as had been observed in previous experiments. All hearts of embryos in the remaining classes were analyzed. Representative localization patterns for Pitx2c mRNA and for cardiac flectin for each class depicted in Table 1 are shown in the accompanying figures.

Hyaluronidase and Flectin Antibody Perturbation Experiments

Hyaluronidase: Experimental embryos at stage 5 were set up in hyaluronidase-containing medium for a 22- to 24-hr incubation period. The hyaluronidase from Streptomyces hyaluronlyticus (Calbiochem, LaJolla, CA) was reconstituted in Simm's balanced salt solution and was used at 21.3 TRU and 10.65 TRU in 2:2:1 medium. Both concentrations produced similar results. Flectin antibody perturbation was carried out as previously described (Linask et al., 2002).

Microscopy and Photography

After incubations, embryos were processed for immunohistochemistry, embedded in Araldite, and embryos were serially sectioned at 2–6 microns through the heart. A Nikon Optiphot fluorescence microscope interfaced with a Princeton MicroMax cooled CCD digitizing camera was used to obtain images.

Computer-Assisted Image Analysis

To provide relative quantitation of results, the images of sections through the hearts were analyzed for pixel area (3 × 3) within a region of the dorsal mesocardium and adjacent myocardial wall that falls within a specified threshold range that corresponds to regions of bright fluorescence intensities (see Figs. 4–6). A rectangular box was drawn on the image, delineating the region within which pixels were to be counted. This box was kept constant for left and right side measurements made on the same sample. Within this region, an upper threshold in range of 125–255 levels of gray was specified based on viewing an overlay on the image showing the area being defined. Pixels that fall within the intensity values of the defined range were colored in red on the left side; yellow on the right side. The means of area measurements for left (L) and right (R) were calculated, and the resulting AQ was obtained by dividing the L mean measurement by the R. For AQ > 1, the heart is looping normally to the right; if between 0.2 and 0.8, generally the heart loops abnormally to the left. If between 0.8–1.0, generally the heart was not looping. Statistics was not carried out on these results, as considerable variability is inherent in these types of experiments. Experimental batches of embryos differed in general intensities of immunofluorescence, although somite numbers may be the same. The developmental stage of embryonic hearts at end of experiment may not be exactly the same; and also the exact levels at which sections were compared along the anteroposterior gradient of heart development from one embryo to the next most likely differ. These ratio ranges, however, helped to define the detectable pattern of asymmetrical protein expression that began to emerge when comparing left to right regions in the same section of a looping heart of an experimental antisense oligonucleotide–treated heart to that of a control.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

In addition to the coauthors, we thank Dr. Robert Nagele, University of Medicine and Dentistry of New Jersey-SOM, for the critical reading of this manuscript. We also thank Dr. Sinning for providing the hLAMP-1 antibody. K.K.L. was funded by an American Heart Association Established Investigator Grant and the NIH, T.B. was funded by the Deutsche Forschungsgemeinschaft.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  • Baldwin HS, Solursh M. 1989. Degradation of hyaluronic acid does not prevent looping of the mammalian heart in situ. Dev Biol 136: 555559.
  • Bamford RN, Roessler E, Burdine RD, Saplakoglu U, dela Cruz J, Splitt M, Towbin J, Bowers P, Marino B, Schier AF, Shen MM, Muenke M, Casey B. 2000. Loss-of-function mutations in the EGF-CFC gene CFC1 are associated with human left-right laterality defects. Nat Genet 26: 365369.
  • Bisgrove BW, Essner JJ, Yost HJ. 2000. Multiple pathways in the midline regulate concordant brain, heart, and gut left-right asymmetry. Development 127: 35673579.
  • Colas J, Schoenwolf GC. 2000. Substractive hybridization identifies chick-cripto, a novel EGF-CFC ortholog expressed during gastrulation, neurulation and early cardiogenesis. Gene 255: 205217.
  • de la Cruz MV, Sanchez Gomez C, Cayre R. 1991. The developmental components of the ventricles: their significance in congenital cardiac malformations. Cardiol Young 1: 123128.
  • Fagotto F, Gumbiner BM. 1994. β-catenin localization during Xenopus embryogenesis: accumulation at tissue and somite boundaries. Development 120: 36673679.
  • Gurdon JB. 1987. Embryonic induction-molecular prospects. Development 99: 285306.
  • Gurdon JB, Lemainre P, Kato K. 1993. Community effects and related phenomena in development. Cell 75: 831834.
  • Hamburger V, Hamilton HL. 1951. A series of normal stages in the development of the chick embryo. J Morphol 88: 4992.
  • Itasaki N, Nakamura H, Sumida H, Yasuda M. 1991. Actin bundles on the right side in the caudal part of the heart tube play a role in dextro-looping in the embryonic chick heart. Anat Embryol (Berl) 183: 2939.
  • Kitamura K, Miura H, Miyagawa-Tomita S, Yanazawa M, Katoh-Fukui Y, Suzuki R, Ohuchi H, Suehiro A, Motegi Y, Nakahara Y, Kondo S, Yokoyama M. 1999. Mouse Pitx2 deficiency leads to anomalies of the ventral body wall, heart, extra- and periocular mesoderm and right pulmonary isomerism. Development 126: 57495758.
  • Lickert H, Lutsch S, Kanzler B, Tamai Y, Taketo MM, Kemler R. 2002. Formation of multiple hearts in mice following deletion of β-catenin in the embryonic mesoderm. Dev Cell 3: 171181.
  • Lin CR, Kioussi C, O'Connell S, Briata P, Szeto D, Liu R, Izpisua-Belmonte JC, Rosenfeld MG. 1999. Pitx2 regulates lung asymmetry, cardiac positioning and pituitary and tooth morphogenesis. Nature 401: 279282.
  • Linask KK. 1992a. N-cadherin localization in the early heart development and polar expression of Na, K-ATPase, and integrin during pericardial coelom formation and epithelialization of the differentiating myocardium. Dev Biol 151: 213224.
  • Linask KK. 1992b. Regulatory role of cell adhesion molecules in early heart development. In: BellairsR, SandersEJ, LashJW, editors. Regulatory role of cell adhesion molecules in early heart development. New York: Plenum Publishing Corporation. p 301313.
  • Linask KK, Tsuda T. 2000. Application of plastic embedding for sectioning whole-mount immunostained early vertebrate embryos. In: TuanRS, LoCW, editors. Application of plastic embedding for sectioning whole-mount immunostained early vertebrate embryos. Vol. 1. Totowa, NJ: Humana. p 165173.
  • Linask K, Gui YH, Rasheed R, Kwon L. 1992. Pattern development during pericardial coelom formation and specification of the cardiomyoctye cell population by N-cadherin and the Drosophila armadillo protein homologue in the early chick embryo. Mol Biol Cell 3 (Suppl): 206A.
  • Linask KK, Knudsen KA, Gui YH. 1997. N-cadherin-catenin interation: Necessary component of cardiac cell compartmentalization during early vertebrate heart development. Dev Biol 185: 148164.
  • Linask KK, Yu X, Chen YP, Han MD. 2002. Directionality of heart looping: effects of Pitx2c misexpression on flectin asymmetry and midline structures. Dev Biol 246: 407417.
  • Lu MF, Pressman C, Dyer R, Johnson RL, Martin JF. 1999. Function of Rieger syndrome gene in left-right asymmetry and craniofacial development. Nature 401: 276278.
  • Markwald RR, Trusk T, Moreno-Rodriguez R. 1998. Formation and septation of the tubular heart: integrating the dynamics of morphology with emerging molecular concepts. In: de la CruzMV, MarkwaldRR, editors. Formation and septation of the tubular heart: integrating the dynamics of morphology with emerging molecular concepts. Boston: Birkhauser. p 4384.
  • Mason JM, Xu HP, Rao SK, Leask A, Barcia M, Shan J, Stephenson R, Tabibzadeh S. 2001. Lefty contributes to the remodeling of extracellular matrix my inhibition of CTGF and collagen mRNA expression and increased proteolytic activity in a fibrosarcoma model. J Biol Chem 277: 407415.
  • Meno C, Shimono A, Saijoh Y, Yashiro K, Mochida K, Ohishi S, Noji S, Kondoh H, Hamada H. 1998. lefty-1 is required for left-right determination as a regulator of lefty-2 and nodal. Cell 94: 287297.
  • Mercola M, Levin M. 2001. Left-right asymmetry determination in vertebrates. Annu Rev Cell Dev Biol 17: 779805.
  • Mieziewska K, Szel A, Van Veen T, Aguirre GD, Philp N. 1994a. Redistribution of insoluble interphotoreceptor matrix components during photoreceptor differentiation in the mouse retina. J Comp Neurol 345: 115124.
  • Mieziewska KE, Devenny J, van Veen T, Aguirre GD, Philp N. 1994b. Characterization of a developmentally regulated component of ocular extracellular matrix that is evolutionarily conserved. Invest Ophthalmol Vis Sci 35: 1608.
  • Nakamura A, Manasek FJ. 1978. Experimental studies of the shape and structure of isolated cardiac jelly. J Embryol Exp Morph 43: 167183.
  • New DAT. 1955. A new technique for the cultivation of the chick embryo in vitro. J Embryol Exp Morphol 3: 326331.
  • Patel K, Isaac A, Cooke J. 1999. Nodal signaling and the roles of the transcription factors SnR and Pitx2 in vertebrate left-right asymmetry. Curr Biol 9: 609612.
  • Peifer M, McCrea PD, Green KJ, Wieschaus E, Gumbiner BM. 1992. The vertebrate adhesive junction proteins β-catenin and plakoglobin and the Drosophila segment polarity gene armadillo form a multigene family with similar properties. J Cell Biol 118: 681691.
  • Rodriguez Esteban C, Capdevila J, Economides AN, Pascual J, Ortiz A, Izpisua Belmonte JC. 1999. The novel Cer-like protein Caronte mediates the establishment of embryonic left-right asymmetry. Nature 401: 243251.
  • Schier AF, Shen MM. 1999. Nodal signalling in vertebrate development. Nature 403: 385389.
  • Schlange T, Schnipkoweit I, Andree B, Ebert A, Zile MH, Arnold HH, Brand T. 2001. Chick CFC controls Lefty 1 expression in the embryonic midline and nodal expression in the lateral plate. Dev Biol 234: 376389.
  • Schneider A, Mijalski T, Schlange T, Dai W, Overbeek P, Arnold HH, Brand T. 1999. The homeobox gene NKX3.2 is a target of left-right signaling and is expressed on opposite sides in chick and mouse embryos. Curr Biol 9: 911914.
  • Semina EV, Reiter RS, Leysens NJ, Alward MLW, Small KW, Datson NA, Siegel-Bartelt J, Bierke-Nelson D, Bitoun P, Zabel BU. 1996. Cloning and characterization of a novel bicoid-related homeobox transcription factor gene RIEG, involved in Rieger syndrome. Nat Genet 14: 392399.
  • Shen MM, Schier A. 2000. The EGF-CFC gene family in vertebrate development. Trends Genet 16: 303309.
  • Sinning A. 1997. Partial purification of HLAMP-1 provides direct evidence for the multicomponent nature of the particulate matrix associated with cardiac mesenchyme formation. J Cell Biochem 66: 112122.
  • Smith DM, Nielsen C, Tabin CJ, Roberts DJ. 2000. Roles of BMP signaling and Nkx2.5 in patterning at the chick midgut-foregut boundary. Development 127: 36713681.
  • Smith SM, Dickman ED, Thompson RP, Sinning AR, Wunsch AM, Markwald RR. 1997. Retinoic acid directs cardiac laterality and the expression of early markers of precardiac asymmetry. Dev Biol 182: 162171.
  • Taber LA. 2001. Biomechanics of cardiovascular development. Annu Rev Biomed Eng 3: 125.
  • Taber LA, Lin IE, Clark E. 1995. Mechanics of cardiac looping. Dev Dyn 203: 4250.
  • Tsuda T, Philp N, Zile MH, Linask KK. 1996. Left-right asymmetric localization of flectin in the extracellular matrix during heart looping. Dev Biol 173: 3950.
  • Tsuda T, Majumder K, Linask KK. 1998. Differential expression of flectin in the extracellular matrix and left-right asymmetry in mouse embryonic heart during looping stages. Dev Genet 23: 203214.
  • Waldo KL, Kumiski DH, Wallis KT, Stadt HA, Hutson MR, Platt DH, Kirby ML. 2001. Conotruncal myocardium arises from a secondary heart field. Development 128: 31793188.
  • Wheelock MJ, Knudsen KA, Johnson KR. 1996. Membrane-cytoskeleton interactions with cadherin cell adhesion proteins: roles of catenins as linker proteins. Curr Top Membr 43: 169185.
  • Yang Z, Kyriakides TR, Bornstein P. 2000. Matricellular proteins as modulators of cell-matrix interactions: adhesive defect in thrombospondin 2-null fibroblasts is a consequence of increased levels of matrix metalloproteinase-2. Mol Biol Cell 11: 33533364.