SEARCH

SEARCH BY CITATION

Keywords:

  • programmed cell death;
  • cardiogenesis;
  • conotruncus;
  • bulbus;
  • ventricle;
  • hypoxia;
  • HIF-1

Abstract

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

Apoptosis occurs at high frequency in the myocardium of the developing avian cardiac outflow tract (OFT). Up- or down-regulating apoptosis results in defects resembling human conotruncal heart anomalies. This finding suggested that regulated levels of apoptosis are critical for normal morphogenesis of the four-chambered heart. Recent evidence supports an important role for hypoxia of the OFT myocardium in regulating cell death and vasculogenesis. The purpose of this study was to determine whether apoptosis in the outflow tract myocardium occurs in the mouse heart during developmental stages comparable to the avian heart and to determine whether differential hypoxia is also present at this site in the murine heart. Apoptosis was detected using a fluorescent vital dye, Lysotracker Red (LTR), in the OFT myocardium of the mouse starting at embryonic day (E) 12.5, peaking at E13.5–14.5, and declining thereafter to low or background levels by E18.5. In addition, high levels of apoptosis were detected in other cardiac regions, including the apices of the ventricles and along the interventricular sulcus. Apoptosis in the myocardium was detected by double-labeling with LTR and cardiomyocyte markers. Terminal deoxynucleotidyl transferase–mediated deoxyuridinetriphosphate nick end-labeling (TUNEL) and immunostaining for cleaved Caspase-3 were used to confirm the LTR results. At the peak of OFT apoptosis in the mouse, the OFT myocardium was relatively hypoxic, as indicated by specific and intense EF5 staining and HIF1α nuclear localization, and was surrounded by the developing vasculature as in the chicken embryo. These findings suggest that cardiomyocyte apoptosis is an evolutionarily conserved mechanism for normal morphogenesis of the outflow tract myocardium in avian and mammalian species. Developmental Dynamics 235:2592–2602, 2006. © 2006 Wiley-Liss, Inc.


INTRODUCTION

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

Many human congenital heart defects involve the region of the heart that connects the aorta and pulmonary artery with the appropriate ventricular chambers. This region has been termed the conotruncus or outflow tract (Pexieder, 1995). Defects of the conotruncus are often associated with high morbidity and mortality (Marino et al., 2001; Goldmuntz, 2001; Hoffman and Kaplan, 2002). This region is defective in the embryos or neonates of several mouse lines in which various genes have been knocked out, overexpressed, or mutated (e.g., Lo et al., 1999, Towbin and Belmont, 2000; Epstein, 2001; Yin et al., 2002). Despite a great deal of investigation, we are in the early phases of understanding the cellular and molecular mechanisms that regulate morphogenesis of this critical region.

We define the embryonic outflow tract (OFT) as the smooth-walled region between the trabeculated ventricle and the aortic sac. It is a tubular structure lined by cardiac muscle on the outside surface of the heart and a single sheet of epithelial cells, the endocardial epithelium, on the luminal side, with a layer of connective tissue in between termed “cardiac jelly” that is populated at later stages by mesenchymal cells. In avian and mammalian species with four-chambered hearts, this simple tubular structure undergoes complex morphogenesis during the last stages of cardiac septation (reviewed in Webb et al., 2003). The events include septation of the OFT lumen, shortening, realignment of the roots of the pulmonary and aortic vessels, and differentiation of the valves of the aortic and pulmonary artery. During this same time, the epicardium grows over the myocardial surface of the heart, including the OFT (Manner et al., 2001). Within this epicardium, a vascular plexus forms around the base of the OFT, and some of these vessels eventually connect to the aortic lumen (Bogers et al., 1989; Waldo et al., 1990). The embryonic OFT tissues are believed to contribute to the valves of the aorta and pulmonary artery and tissues below the valves, including the infundibulum of the mature right ventricle (de la Cruz et al., 1977). Below the aortic valve is fibrous tissue, whereas under the pulmonary artery is the smooth-walled, striated muscle-lined infundibular chamber (also called the conus) demarcated by the crista supraventricularis (parietal band), the septal band, and moderator band (for review, see Anderson et al., 2003).

The outflow tract myocardium is the site of high levels of apoptosis in the chicken and quail embryo during the time that it undergoes dramatic morphogenesis (Pexieder, 1972, 1975; Hurle and Ojeda, 1979; Watanabe et al., 1998; Cheng et al., 2002; Rothenberg et al., 2002; Schaefer et al., 2004). Perturbation of the timing and levels of apoptosis at this site causes misalignment of the great vessels with the ventricular chambers (Watanabe et al., 2001; Sallee et al., 2004; Sugishita et al., 2004a, b; reviewed in Rothenberg et al., 2003; Sugishita et al., 2004c). These results suggest that a tightly regulated level of apoptosis is critical for normal OFT morphogenesis and that genetic or environmental factors that perturb apoptosis in this region may result in cardiac defects. Mammalian hearts (rat, mouse, and human) have been reported to undergo outflow tract morphogenesis resembling that described for chicken embryos, with shortening or “regression,” rotation, and septation (Goor, 1972; Ya et al., 1998; Anderson et al., 2003). Cell death in the endocardium was reported many years ago (Pexieder, 1975) and investigated in chicken (Keyes and Sanders, 1999, 2002) and mouse hearts (Zhao and Rivkees, 2000). Apoptosis has been detected by terminal deoxynucleotidyl transferase–mediated deoxyuridinetriphosphate nick end-labeling (TUNEL) in the mammalian heart, including a low frequency within the OFT myocardium (Zhao and Rivkees, 2000; Kubalak et al., 2002; Sharma et al., 2004). However, apoptosis in the mammalian OFT myocardium has not been the focus of study.

The purpose of this study was to determine whether apoptosis could be a universal mechanism for morphogenesis of four-chambered hearts, with particular emphasis on the OFT during septation. A rapid and sensitive method of assaying whole-mount preparations of embryonic mouse hearts for apoptosis was used to screen the stages of interest and to focus the analysis to the most relevant stages. Our hypothesis was that cardiomyocytes within the mammalian OFT myocardium, similar to the cardiomyocytes within the avian OFT myocardium, undergo apoptosis with a comparable spatiotemporal pattern that is correlated with hypoxia of the OFT myocardium and peritruncal vascular development.

RESULTS

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

LTR Staining

Lysotracker Red (LTR) was used to detect apoptotic sites by observations of whole-mount mouse embryo hearts by fluorescence stereomicroscopy. This technique has been used in the past to detect apoptosis in chicken (Schaefer et al., 2004) and rodent embryos (Zucker et al., 1998; Price et al., 2003). External surfaces of mouse embryo hearts from stages embryonic day (E) 10.5 to E18.5 were observed. No LTR staining was present at E10.5 (not shown). A few LTR staining particles were evident over the entire heart from as early as E11.5 and E12.0 but were not distinctly above background levels in the OFT until E12.5 (Fig. 1). Differential LTR staining of the OFT was observed at stage E12.5, peaked at E13.5–E14.5, and decreased in subsequent stages, with a few LTR-positive particles present as late as E18.5 (Fig. 1). At these stages, the staining appeared in a distinct band of positive particles running across the base of the ventral OFT where it connects with the ventricle similar to that observed for chicken embryos at comparable stages (Fig. 1). At E17.5–E18.5, some embryo hearts exhibited LTR staining in the OFT and others did not (23 negative out of 30). Ventricular LTR staining was most abundant in the apices and the interventricular sulcus and remained high in those ventricular regions throughout the stages studied (E10.5–E18.5). Another LTR-positive region was the proximal aorta at E13.5, but not in E15.5 embryos.

thumbnail image

Figure 1. Lysotracker Red (LTR) staining of whole-mouse embryo hearts of stages embryonic day (E) 11.5–E17.5. A: At E11.5, there was only background LTR staining. B,C,I: Starting at E12.5 (B), and more distinctly by E13.5 (C,I), LTR particles accumulated in the outflow tract (OFT) myocardium and interventricular groove (arrows) at above-background levels. D,F: The LTR-positive staining of the OFT decreased after E13.5–E14.5 but remained above background levels through E16.5 (D) and in some E17.5 embryos (F). E: The ventricular apex regions continued to be LTR-positive as well at E16.5. G: Other E17.5 hearts showed no LTR staining in the OFT, although they were positive for LTR staining in other embryonic tissues. H: The LTR staining pattern was similar in the stage 30 chicken embryo heart (H) and the E13.5 mouse embryo heart (I) OFT myocardium (arrow in I, and top arrow in C). In both species the LTR particles ringed the base of the OFT. In the mouse embryo heart, LTR staining was prominent in the myocardium at the apex of the ventricles and followed portions of the interventricular groove (ivg). ao, aorta; ivg, interventricular groove; ivs, intraventricular septum; pa, pulmonary artery; ra, right atrium; la, left atrium; rv, right ventricle; lv, left ventricle. Scale bars = 500 μm in A (applies to A–C), 500 μm in E (applies to D–G), bar in H applies to I.

Download figure to PowerPoint

To determine the tissue location of the LTR-positive cells, paraffin sections of the LTR-stained hearts were co-stained with striated myocyte marker Sr-1, an antibody to sarcomeric actin (Fig. 2). This analysis revealed that the LTR staining of the OFT was within the myocardium. One cell type that is abundant in the OFT at the peak of cell death is endothelial cells. To determine to what extent this cell type might be undergoing cell death, intact embryonic mouse hearts at E13.5 were co-stained with the endothelial marker anti–platelet endothelial cell adhesion molecule-1 (PECAM) and LTR. Confocal microscopy of these hearts showed that there was an extensive anastomosing network of PECAM-1–positive (CD31) or CD34-positive vessels surrounding the base of the OFT at the same level as the LTR positive staining (Fig. 3). There was little overlap, however, in staining for PECAM-1 or CD34 and LTR, indicating that, although a few endothelial cells were dying, these cells were unlikely to account for the bulk of the LTR staining in the OFT myocardium.

thumbnail image

Figure 2. A–C: Sagittal sections through an embryonic day (E) 13.5 mouse embryo heart costained with cardiomyocyte marker Sr-1 (A,B) and Lysotracker Red (LTR, C). The LTR particles were present in the myocardium of the outflow tract (OFT). B and C are higher magnifications of A. ao, aorta; at, atria; pa, pulmonary artery; v, ventricle. Scale bar = 100 μm in C.

Download figure to PowerPoint

thumbnail image

Figure 3. Confocal microscopy of intact embryonic day (E) 13.5 mouse hearts double-labeled with Lysotracker Red (LTR, red) and anti–platelet endothelial cell adhesion molecule-1 (PECAM, green) at the level of the outflow tract (OFT). The anti-PECAM labeled the endothelial component of the peritruncal anastomoses. In the projected z-series at the top of the right ventricle and OFT, the pulmonary artery (pa) was optically sectioned longitudinally and a portion of the aortic endothelium (ao) was visible in cross-section at the top of the photomicrograph. The same region was taken at different magnifications: ×100 (A), ×200 (B), ×400 (C). Few LTR particles colocalized (yellow, arrow in C) with PECAM-positive regions.

Download figure to PowerPoint

Other Markers for Apoptosis

A well-accepted histological assay for apoptosis, TUNEL staining, identifies concentrations of DNA fragments in sections (Gavrieli et al., 1992). TUNEL-positive particles were found in the same pattern in sections as the LTR-positive particles in the OFT (not shown) and ventricles (Fig. 5).

thumbnail image

Figure 5. Terminal deoxynucleotidyl transferase–mediated deoxyuridinetriphosphate nick end-labeling (TUNEL) staining and cleaved Caspase-3–positive and Sr-1–positive ventricular cardiomyocytes. The TUNEL staining pattern in frontal sections of the embryonic day (E) 14.5 mouse embryo heart was similar to the pattern of Lysotracker Red (LTR) staining. A,B: TUNEL+ particles (stained brown at arrowheads in A) were concentrated in the myocardium of the ventricular apex and interventricular groove (circle in B). Cleaved Caspase-3 was detected by indirect immunofluorescent labeling (green) in cardiomyocytes at the ventricular lumen in sections of a different embryo at the same stage. C,D: These sections were coimmunostained for Sr-1 (red) in the trabeculae of the left ventricle. Some of the costained regions (yellow) are indicated by arrows. ivs, interventricular septum; lv, left ventricle; rv, right ventricle. Scale bars = 200 μm in B, 10 μm in D.

Download figure to PowerPoint

Neither TUNEL nor LTR staining was adequate to definitively determine the cell type undergoing apoptosis in the mouse embryo myocardium, because both techniques labeled apoptotic bodies of debris as well as apoptotic cells (for review of apoptosis techniques, see Watanabe et al., 2002). We and others have documented that cardiomyocytes themselves phagocytose apoptotic debris in the chicken OFT myocardium (Hurle et al., 1977, 1978; Watanabe et al., 2001; Rothenberg et al., 2002). Therefore, cells positive for markers of cardiomyocytes that are TUNEL-positive or LTR-positive could be healthy cardiomyocytes ingesting apoptotic debris rather than dying cardiomyocytes. For this reason, the antibody to the cleaved Caspase3 was used to identify cells earlier during apoptosis. Caspase 3 is a protease that initiates the terminal cascade of apoptosis and is activated by cleavage in most forms of apoptosis (reviewed in Boatright and Salvesen, 2003). We also showed in a previous study that embryo exposure to peptide inhibitors of Caspase 3 caused OFT defects (Watanabe et al., 2001). The pattern of anti-cleaved Caspase 3 staining in the E13.5 and E14.5 heart sections corresponded with the LTR and TUNEL pattern. Colocalization of cardiomyocyte markers and cleaved Caspase 3 was detected in cells in the OFT myocardium by standard fluorescence microscopy (Fig. 4) and laser scanning confocal microscopy (Supplementary Figure S1, which can be viewed at http://www.interscience.wiley.com/jpages/1058-8388/suppmat). The cells of the trabeculae lining the E13.5–E14.5 right ventricle were also immunopositive for both cardiomyocyte markers and cleaved Caspase 3 (Fig. 5). This finding provided evidence that many dying cells at these sites were apoptotic cardiomyocytes. As for nonmyocardial cells, mesenchyme between the aortic and pulmonary outflow (Supplementary Figure S2) were found to be positive for cleaved Caspase 3 and subnuclear-sized 4′,6-diamidine-2-phenylidole-dihydrochloride (DAPI) particles.

thumbnail image

Figure 4. Cleaved Caspase 3 in the outflow tract (OFT) myocardium. A–D: Sagittal sections through the embryonic day (E) 13.5 mouse outflow tract were stained with 4′,6-diamidine-2-phenylidole-dihydrochloride (DAPI, A) and costained with MF20 (B,C) and anti-titin (C,D) and anti-cleaved Caspase 3 (green). The merged image in C of the cardiomyocyte markers and anti–cleaved Caspase 3 shows colocalization (yellow, arrows). at, atria; pa, pulmonary artery.

Download figure to PowerPoint

cGATA6(D6) Reporter Staining Indicated Shortening of the OFT Myocardium

In chicken embryos, high levels of apoptosis of OFT cardiomyocytes occurred at the same stages as did shortening of the OFT myocardium. Shortening of the OFT myocardium is easily marked in the avian system using adenovirus vectors. Because viral labeling of the OFT is much more difficult in mouse compared with chicken embryos, we used a different tactic. β-Galatosidase reporter gene expression driven by a distal segment of the enhancer upstream of the chicken GATA-6 (cGATA-6/D6) marks mouse OFT myocardial tissue starting at E10.5 (Davis et al., 2000). This transgenic mouse line cGATA6(D6) was used to determine the changes in OFT myocardium during the stages before and during LTR staining. To assess whether the OFT decreases in length during stages of OFT apoptosis, we stained whole-mount OFTs of cGATA6(D6) embryonic mice from E11 and E15 and found that the most intensely stained portion of the OFT myocardium undergoes a dramatic reduction in height during this time period (Fig. 6). Sections of these embryonic hearts revealed that the staining was within the OFT myocardium and ran the length of the OFT myocardium from the trabeculated ventricle to the level of the pulmonary and aortic valve (Fig. 7).

thumbnail image

Figure 6. Mouse embryo hearts of the GATA6 mouse line stained with x-gal/β-galactosidase. The most intense staining (blue–black) from embryonic day (E) 10.5 to E15.5 indicated the outflow tract (OFT) myocardium. Frontal (anterior) views of the heart from cGata6-lacz mouse embryos stained using x-gal. A: At stage E10.5, the curved OFT myocardium is the most intensely stained region of the heart. B: By stage E12.5, the staining showed that the length of the OFT has decreased while the OFT increased in width. By stages E13.5 and E15.5, the stained OFT is a fraction of its original length. A–C: Images were taken at the same magnification. D: Half the magnification of the other panels. Panel B was taken with a green fluorescence filter to enhance the differential blue staining of the OFT myocardium. Scale bar = 100 μm in B, 200 μm in D.

Download figure to PowerPoint

thumbnail image

Figure 7. Sections of mouse hearts that were stained for x-gal. A,B: The most intense x-gal–positive staining was in the outflow tract (OFT) myocardium, starting where the trabeculated ventricle ends and running along the length of the OFT until it connects to the great vessels in sagittal sections of embryonic day (E) 10 (A) and E12.5 (B) and in frontal sections of the E15 hearts (C,D). Arrows indicate the beginning and end of the most intense x-gal staining in the OFT myocardium. C: The arrows indicate x-gal staining adjacent to the pulmonary artery valve leaflets. D: The arrows indicate x-gal staining adjacent to the aortic valve leaflets. avj, atrioventricular junction; rv, right ventricle; lt, left ventricle; pa, pulmonary artery; ao, aorta; avj, atrioventricular junction. All panels were taken at the same magnification. Scale bar = 0.5 mm.

Download figure to PowerPoint

Hypoxia Markers in the OFT at the Peak of LTR Staining

In chicken embryos, OFT cardiomyocyte apoptosis occurred at stages when the OFT was hypoxic as measured by the hypoxia indicator EF5 and anti-HIF1α nuclear localized staining. When tissue hypoxia was perturbed by hyperoxic conditions, cell death levels were reduced, indicating that normally occurring levels of tissue hypoxia may be important for initiating or regulating cardiomyocyte apoptosis. To determine whether a similar relationship existed between apoptosis and relative hypoxia in the mouse OFT, we assayed for EF5 staining and HIF-1α nuclear localization at the peak of LTR staining in sections of the mouse embryo heart that included the OFT. EF5 is a metabolic marker for hypoxia that binds to cellular macromolecules in cells experiencing hypoxia (see review by Koch, 2002) and is detected by immunostaining. In cultured cells, EF5 binding has a linear relationship with oxygen levels below 10%. Hypoxia-inducible factor-1α (HIF-1α) is the oxygen-sensitive heterodimer component of HIF-1. HIF-1α is degraded under normoxic conditions but is stabilized under hypoxic conditions, accumulates in the nucleus, and regulates the transcription of several genes, including those that control apoptosis, survival, vascular development, and cellular metabolism (for review, see Semenza, 2000, 2003). Strong EF5 staining and HIF-1α nuclear localization were observed within the mouse OFT myocardium (Figs. 8, 9) as we observed in the chicken OFT (Sugishita et al., 2004a, b). EF5 staining was also present between the aortic and pulmonary OFT at the level of and below the valves (Fig. 8) where we also observed immunostaining for cleaved Caspase 3 in the mesenchyme (Supplementary Figure S2).

thumbnail image

Figure 8. EF5 immunostaining of mouse outflow tract (OFT) tissues. A–C: Sagittal frozen sections through the embryonic day (E) 13.5 mouse OFT under transmitted light (A), stained with EF5 and anti–EF5-Cy3 (B, red), or exposed to EF5 with no anti–EF5-Cy3 (C, Control). The positive EF5 staining (red) was observed in the myocardial layer of the OFT (between white arrows) and not in the ventricular myocardium. There was also intense staining in the mesenchyme between the pulmonary artery (pa) and the aorta (ao). vent, ventricular lumen; oft, lumen of outflow tract. Scale bar = 200 μm in A.

Download figure to PowerPoint

thumbnail image

Figure 9. HIF-1α staining of the mouse outflow tract (OFT) myocardium. B,E: Sagittal frozen sections of embryonic day (E) 13.5 mouse embryo hearts were stained with anti-HIF1α with TSA-fluorescein isothiocyanate amplification. Nuclei of cells in the OFT myocardium (B, arrow) (my) were intensely positive for anti-HIF1α immunostaining, whereas nuclei in the ventricular wall (E) were negative as were the endocardial (en) and epicardial cells. C,F: The 4′,6-diamidine-2-phenylidole-dihydrochloride (DAPI) staining was used to indicate all nuclei (blue) in the same sections. A,D: The identical sections viewed under the red filter exhibited the faintly autofluorescent myocardium (yellow dotted line). Red blood cells autofluoresced under both red and green filters and were of a higher intensity than the anti-HIF1α nuclear immunostaining. Scale bar = 200 μm in E (applies to all panels).

Download figure to PowerPoint

DISCUSSION

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

Apoptosis in the Mouse OFT Myocardium

In this study, LTR staining and other apoptosis assays were used to show that apoptosis is occurring at specific sites within the embryonic mouse myocardium and that it is the cardiomyocytes that are undergoing apoptosis. Patterns of staining similar to LTR staining were observed by using other apoptosis markers that use different principles, TUNEL and immunostaining for cleaved (active) Caspase 3, supporting the validity and usefulness of LTR in detecting apoptotic cells in the embryonic mouse heart at stages spanning the progression of OFT morphogenesis and cardiac septation. Compared with TUNEL staining and immunostaining for cleaved Caspase 3 in sections, LTR staining revealed more clearly a distinct region of apoptosis in the OFT in both whole-mounted hearts and in sections. With other apoptosis detection techniques such as TUNEL, only a few apoptotic cells are identified per section, thus requiring reconstruction and/or stacking the results from many sections to determine whether there is a higher than background level of apoptosis in a particular compartment such as the trabecular compartment of the ventricles (Cheng et al., 2002) or the OFT myocardium (Sharma et al., 2004). LTR in whole-mounts appears to be a more sensitive method for detecting apoptotic regions than other more commonly used assays.

We found that apoptosis occurs in the mouse OFT myocardium during a similar phase in cardiac development as in the chicken OFT. The marker for apoptosis, particulate LTR staining, appeared in the mouse proximal OFT at stage E12.5. The LTR particles increased in number at that site until E13–E14.5 and then decreased by E16.5, although persisting at above-background levels until later stages (E17.5–E18.5). LTR particles were found in the OFT under both the developing aorta and the pulmonary artery as we previously observed in avian species (Schaefer et al., 2004). In the embryonic chick heart, the OFT myocardium stains for apoptosis markers (TUNEL, LTR) between stages 26 and 35 with a peak between stages 27 and 32 (Watanabe et al., 1998; Schaefer et al., 2004). Staining at stage 29 is widely distributed throughout the OFT myocardium, but by stage 32, the level of apoptosis begins to decrease. The stages of apoptosis in both the mouse and avian OFT myocardium corresponded to the period when the OFT myocardium is reducing in length, separating into the aortic and pulmonary outflows, and when the aorta and pulmonary artery are establishing their mature relationship with the left and right ventricles, respectively (Vuillemin and Pexieder, 1988; Kaufman, 1992).

We made several novel findings in this study when compared with previous studies of mouse OFT apoptosis where TUNEL was used as the assay (Zhao and Rivkees, 2000; Kubalak et al., 2002; Sharma et al., 2004): (1) whole-mount LTR staining showed that the level of apoptosis in the OFT myocardium is particularly high but in a tight band at the base of the OFT that appeared to be a low frequency of apoptosis by TUNEL staining in sections; (2) the apoptotic cells in the OFT myocardium were cardiomyocytes; and (3) the peak of OFT apoptosis occurs coincident with a decrease in the height of the OFT myocardium, intense hypoxia indicator EF5 staining within the OFT myocardium, nuclear localization of HIF-1a within the myocardium, and differentiation of a primitive network of vasculature around the apoptotic OFT.

Outflow Tract Shortening

Whether the mammalian OFT undergoes shortening during septation has been a point of discussion for many years. In chicken embryos, it is possible to specifically label the OFT myocardium by injecting a replication-defective adenovirus with a reporter gene expression being driven by the cytomegaloviral (CMV) promoter into the pericardial space. During stages 25–32, when the OFT was septating, the length of the labeled OFT myocardium was reduced dramatically as determined by observing whole-mounted hearts (Watanabe et al., 1998). When apoptosis was specifically inhibited at that site, more of the labeled OFT myocardium persisted, and there was conotruncal misalignment (Watanabe et al., 2001). For mouse embryos, such adenoviral injections would be difficult to perform. As an alternative method to follow the OFT myocardium, we analyzed the reporter gene-labeled OFT of a transgenic mouse line. GATA-6 is a transcription factor that is expressed from an early stage in heart development in vertebrate species (reviewed in Charron and Nemer, 1999; Pikkarainen et al., 2004). The β-galactosidase reporter gene expression driven by a distal segment of the enhancer upstream of the chicken GATA-6 (cGATA-6/D6) allowed intense marking of mouse OFT myocardial tissue starting at E10.5 (Davis et al., 2000). At subsequent stages, the region that stained was reduced in length so that the originally stained tubular structure at E10.5 became ring-like by E15.5. Although the extent of this transgene expression becomes restricted in a posterior to anterior pattern at early stages (before E10.5), the sections showed that the extent of the OFT myocardium from the level of the valves of the aorta and pulmonary artery to the trabeculated ventricles was most intensely stained during the stages (E11–E15) that were relevant to this study. Therefore, our data support the absolute shortening of the OFT myocardium in the mouse embryo coinciding with the peak of apoptosis in the OFT myocardium, similar to what we reported for the chicken embryo (Watanabe et al., 1998).

Comparison of Mouse and Chicken Apoptosis

Our studies show that both mouse and chicken cardiogenesis involve apoptosis and relative hypoxia within the OFT myocardium. In both species, the LTR staining was on the ventral surface along a band across the proximal OFT, and the hypoxia markers EF5 staining and HIF-1α nuclear localization labeled a similar region. Differences in the level and pattern of apoptosis were noted between the species. An obvious difference was that the number of LTR-positive particles observed in the whole-mount embryonic mouse hearts at any stage was less than observed in the avian OFT. Potential explanations for this difference are that in mouse (1) there is less apoptosis altogether, (2) apoptosis during OFT shortening in mouse occurs at lower levels over a longer period of time, and (3) the removal of apoptotic bodies is more rapid in mammalian systems so that the number of apoptotic bodies evident at any one time point is lower. Another possibility is suggested by the results of previous investigations using supravital dye staining in rat (Pexieder, 1975). These investigators found that the stained cellular granules in rat embryo cardiac tissues were smaller than that observed in the chicken embryos. By using the magnifications of the stereomicroscope (×100), we may be detecting only the larger of LTR particles present in the mouse OFT myocardium.

In contrast to the avian hearts, the mouse heart exhibited much more intense staining with several apoptosis markers at stages E13.5–E15.5 in the ventricles. Ventricular cell death as detected by LTR and TUNEL was concentrated at the apical portion of the ventricles and along the interventricular sulcus/septum (IVS) in the periseptation stages studied (this study and Sharma et al., 2004). Chicken embryos also have apoptosis in the IVS (Cheng et al., 2002; Pexieder, 1972, 1975). The death ligand receptor Fas has been detected in the embryonic myocardium with its receptor FasL in specific regions of the heart, the IVS (Sharma et al., 2004; Eralp et al., 2005), and endothelial cells within the OFT myocardium. Furthermore, forced expression of FasL resulted in inappropriately high levels of embryonic cardiomyocyte cell death (Sallee et al., 2004). We speculate that these sites of apoptosis may reflect tissue remodeling related to the invasion of FasL-positive vessel precursors and vessels of the IVS.

In quail, the stages when invasion of the epicardial cells occurs is correlated with apoptosis within the OFT myocardium. Furthermore, if the epicardial cells are delayed in their coverage of the heart, apoptosis is inhibited (Rothenberg et al., 2002). The peak of LTR staining in the OFT myocardium of the mouse (E13.5–E14.5) occurs after epicardial coverage, which is first detected at E9 and completely covers the heart and the OFT by E11 (Viragh and Challice, 1981; Komiyama et al., 1987).

Apoptosis and Hypoxic Tissues

Our previous findings using the chicken embryo model correlated the peak of apoptosis in the OFT myocardium with the peak of hypoxia as measured with the hypoxia indicator EF5 (Sugishita et al., 2004a). At the same stage, HIF-1α nuclear localization was also increased in the myocardium within the OFT, indicating that the hypoxia was high enough and the cells competent to stabilize and translocate HIF-1α to the nucleus. It has been proposed that a gradient of hypoxia across the ventricular wall corresponds with the HIF-1–responsive VEGF gene expression both in transcript and protein levels and endothelial tube formation (Tomanek et al., 1999) and that the perturbation of hypoxia can lead to abnormal coronary vascular development (Ivnitski-Steele et al., 2004). Our most recent findings for EF5 and HIF-1α staining in the chicken embryo heart (Wikenheiser et al., 2006) suggest that other regions in addition to the OFT myocardium are hypoxic and correlate with regions that have been shown to exhibit apoptosis in chicken and mouse embryos such as the IVS (Cheng et al., 2002; Sharma et al., 2004). Furthermore, ambient hypoxia enhances the expression of the HIF-1–regulated gene VEGF in avian systems (Tomanek et al., 2003). Our studies show a correlation in the mouse embryo OFT between myocardial apoptosis, tissue hypoxia, HIF-1α nuclear localization, and a primitive network of vasculature around the OFT (diagrammed in Shaefer et al., 2004; Sugishita et al., 2004c). These findings suggest that conclusions that we have come to regarding hypoxia-regulated mechanisms in chicken embryo cardiogenesis (Sugishita et al., 2004a–c) may have relevance to mammalian systems.

Summary

In previous studies, we experimentally connected microenvironmental hypoxia with apoptosis levels in the chicken OFT myocardium and OFT morphogenesis. In mouse as in chicken, the embryonic OFT myocardium undergoes shortening and apoptosis at stages of OFT septation, suggesting that apoptosis is also an important mechanism for OFT morphogenesis in the mouse. In both species, the OFT myocardium is relatively hypoxic, and HIF-1α nuclear localization is initiated. This finding suggests that the OFT myocardium may be subject to differential regulation by hypoxia and HIF-1 at these stages. The relationship between microenvironmental tissue hypoxia, apoptosis, and OFT morphogenesis has been made in avian systems, and based on our current results, it appears likely that this connection exists in mammalian systems as well.

EXPERIMENTAL PROCEDURES

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

Animals

Mating pairs of C57/Bl6 mice (Jackson Laboratories) were put together in the late afternoon, and the females were checked for plugs the following morning. The middle of the day of plug detection was designated E0.5. Staging was confirmed by comparison with the external features of embryos (Kaufman, 1992). Litter size was 5–12, with an average size of 8, and a minimum of two litters was analyzed per stage. For stages E13.5 and E14.5, more than eight litters were analyzed per stage. Animals were used in accordance with IACUC procedures.

Apoptosis Assays

Lysotracker Red (LTR).

The staged embryos, with the chest wall removed to expose the hearts, were incubated in 5 μM Lysotracker Red in warm Dulbecco's phosphate buffered saline (PBS) at 37°C for 20–30 min. The embryos were fixed at room temperature for 2–4 hr in 4% paraformaldehyde or in some specimens 10% formalin (Ted Pella, Inc., Redding, CA), and observed under the fluorescence stereomicroscope (Leica MZFLIII). The sections were observed under the standard upright fluorescence microscope (Leica DMLB). Images were captured with a digital camera (SPOT RT), and brightness and contrast were adjusted with SPOT software or Adobe Photoshop 6.0 or 7.0.

Hearts from only those embryos with bright LTR staining in the limbs and body wall, the positive control tissues, were analyzed. These regions were shown previously to be suitable for positive control tissues because they are highly Nile Blue Sulfate–positive (vital dye) and TUNEL-positive in chicken embryos (Zou and Niswander, 1996; Yokouchi et al., 1996) and LTR-positive in rodent embryos at a comparable stage (Price et al., 2003). The presence of intense endocardial staining indicated that the LTR penetrated through the mouse cardiac tissues during the whole-mount staining procedure. This LTR pattern did not change whether the hearts remained within the thorax or were excised and incubated in LTR solution for staining.

Embryo hearts that were fixed immediately, processed and stained in sections with the TUNEL technique exhibited the same pattern of staining as those from embryos stained with LTR before TUNEL staining (data not shown). This similarity indicated that the 45 min of processing during the LTR staining did not illicit apoptosis in the cardiac tissues. Although the pattern of staining was the same, LTR stained more particles than TUNEL did, as we found in our previous results using chicken embryo hearts (Schaefer et al., 2004).

Autofluorescence of the red blood cells (RBCs) were distinguished from the LTR staining because (1) RBCs fluoresce under both the red and green filter sets (N2.1,L5), whereas the LTR staining was only positive with the red filter (N2.1); (2) the RBCs are larger and more homogeneously sized than are the LTR-positive particles; and (3) the RBC fluorescence is less intense than that of the LTR-positive particles. The sections were counterstained with MF20, anti-titin, or Sr-1 (see below) to identify cardiomyocytes and to delineate the myocardium.

TUNEL.

TUNEL staining was carried out according to the manufacturer's protocol (Apoptag-Peroxidase in situ apoptosis detection kit; Intergen, Purchase, NY) as previously described (Watanabe et al., 1998) in paraffin-embedded tissue sections.

Immunostaining for Cardiomyocyte Markers

Monoclonal antibody Alpha-Sr-1 (DAKO) is a mouse IgM against rabbit sarcomeric actin that we used to delineate the myocardium in sections. It is known to recognize sarcomeric actin in many species. The selective staining of the mouse myocardium in this study suggests that it recognizes the sarcomeric actin in mouse as well.

MF20 (ascites fluid, 10× partially purified supernatant, supernatant) mouse monoclonal antibody against myosin heavy chain was developed by Dr. D. Fishman and obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by The University of Iowa, Department of Biological Sciences (Iowa City, IA).

Anti–cleaved Caspase 3 rabbit polyclonal antibody against human and mouse cleaved Caspase 3 was obtained from R&D Systems (catalog no. AF835; Minneapolis, MN). Caspase-3 is activated by cleavage and serves as an earlier marker for apoptosis than TUNEL or LTR. An advantage to this apoptosis assay is that it will detect dying cells before cardiomyocyte markers, such as MF20, are degraded. Cryosections were double labeled with anti–Caspase-3 and MF20 to localize and identify the apoptotic cells as cardiomyocytes. A Leica DM IRE2 laser scanning confocal microscope was used to detect the cleaved Caspase-3–positive cells in the myocardium that were MF20-positive and, therefore, dying cardiomyocytes. Sections were magnified at ×63 using oil immersion. The filters used were N2.1 and I3. Z-sections were 1 μm or less.

Hypoxia Markers

EF5 staining of mouse embryos was performed as previously described (Ryan et al., 1998). Briefly, pregnant mice at the appropriate stage of gestation were anesthetized (using a mixture of acepromazine, atropine sulfate, Ketamine HCl), and the tail-vein was injected with 0.2–0.3 ml of the EF5 solution (10 mM solution in 0.9% saline) warmed to 37°C. The pregnant mice were killed, and the embryos were dissected, fixed in 4% paraformaldehyde in PBS for 2 hr, and prepared for cryosectioning in the sagittal plane. Rabbit polyclonal antibody raised against the C-terminal region of HIF-1α was used to stain cryostat sections of mouse hearts using the protocol of Sugishita et al. (2004b).

Immunostaining for Endothelial Markers

PECAM-1 or CD34 were used to identify endothelial cells in intact E13.5 mouse hearts that were also stained with LTR. Embryos were fixed in 4% buffered paraformaldehyde, permeabilized in 0.02% Triton X-100, labeled with anti–PECAM-1 (B&D-Pharmingen, San Diego, CA) or anti-CD34 [anti-moCD34;RAM(34) B&D Pharmingen], and immunolocalized with fluorescein isothiocyanate–conjugated donkey anti-rat IgG (Jackson Immunoresearch, West Grove, PA). The intact hearts were mounted in anti-photobleaching medium and analyzed on a Bio-Rad MRC-1024 (Bio-Rad, Microscopy Division, Cambridge, MA) laser scanning confocal microscope (Zeiss 40× Plan neofluor objective, 0.90 NA, oil immersion). Optical sections were stored on an IBM workstation and z-projections were assembled using Image J 1.31v (National Institutes of Health, Bethesda, MD) and processed using Adobe Photoshop 7.0 (Adobe Systems, Inc., San Jose, CA). Observations were made on Z-sections less than 1 micron to demonstrate colocalization.

Transgenic Mice

Timed-mated cGATA6 D6 mice (cG6/lacZ-D) were fixed, developed in x-gal solution for 1–12 hr (depending on the stage) as previously described (Davis et al., 2000), and photographed as intact hearts. They were then sectioned and observed for histological detail. The most intense x-gal staining was detected in the OFT myocardium during the stages studied.

Acknowledgements

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

The authors thank Erica Sieverding, Florence Rothenberg, MaryAnn Pendergast, Hironi Makita, Christopher Drake, Sapna Shah, Yasuyuki Sugishita, Steven A. Fisher, and Dr. Burch's laboratory personnel for advice and technical assistance at critical phases of the study. We also thank Sarah Richer for technical assistance on the laser confocal microscope of the Rainbow Babies and Children's Pediatric Imaging Center. We thank Dr. Faton Agani for generously providing advice and the HIF-1α antibody. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

REFERENCES

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

Supporting Information

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

The Supplementary Material referred to in this article can be found at http://www.interscience.wiley.com/jpages/1058-8388/suppmat .

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.