Hypoxia-responsive signaling regulates the apoptosis-dependent remodeling of the embryonic avian cardiac outflow tract

Developmental Biology 273 (2004) 285 – 296 www.elsevier.com/locate/ydbio

Hypoxia-responsive signaling regulates the apoptosis-dependent remodeling of the embryonic avian cardiac outflow tract Yasuyuki Sugishita a, David W. Leifer a, Faton Agani b, Michiko Watanabe c, Steven A. Fisher a,* a

Department of Medicine, Case Western Reserve University, Cleveland, OH 44106, USA Department of Anatomy, Case Western Reserve University, Cleveland, OH 44106, USA c Department of Pediatrics, Case Western Reserve University, Cleveland, OH 44106, USA b

Received for publication 16 April 2004, revised 19 May 2004, accepted 25 May 2004 Avaialable online 20 July 2004

Abstract We proposed a model in which myocardial hypoxia triggers the apoptosis-dependent remodeling of the avian outflow tract (OFT) in the transition of the embryo to a dual circulation. In this study, we examined hypoxia-dependent signaling in cardiomyocyte apoptosis and outflow tract remodeling. The hypoxia-inducible transcription factor HIF-1a was specifically present in the nuclei of OFT cardiomyocytes from stages 25 – 32, the period of hypoxia-dependent OFT remodeling. HIF-1a expression was sensitive to changes in ambient oxygen concentrations, while its dimerization partner HIF-1h was constitutively expressed. There was not a simple relationship between HIF-1a expression and apoptosis. Apoptotic cardiomyocytes were detected in HIF-1a-positive and -negative regions, and a hypoxic stimulus sufficient to induce nuclear accumulation of HIF-1a did not induce cardiomyocyte apoptosis. The hypoxia-dependent expression of the vascular endothelial growth factor receptor (VEGFR2) in the distal OFT myocardium may be protective as cardiomyocyte apoptosis in the early stages (25 – 30) of OFT remodeling was absent from this region. Furthermore, recombinant adenoviral-mediated expression of dominant negative Akt, an inhibitor of tyrosine kinase receptor signaling, augmented cardiomyocyte apoptosis in the OFT and constitutively active Akt suppressed it. Adenovirus-mediated forced expression of VEGF165 induced conotruncal malformation such as double outlet right ventricle (DORV) and ventricular septal defect (VSD), similar to defects observed when apoptosis-dependent remodeling of the OFT was specifically targeted. We conclude that normal developmental remodeling of the embryonic avian cardiac OFT involves hypoxia/HIF-1-dependent signaling and cardiomyocyte apoptosis. Autocrine signaling through VEGF/VEGFR2 and Akt provides survival signals for the hypoxic OFT cardiomyocytes, and regulated VEGF signaling is required for the normal development of the OFT. D 2004 Elsevier Inc. All rights reserved. Keywords: Akt; Apoptosis; Flk-1/KDR; Heart; Hypoxia-inducible factor; Vascular endothelial growth factor; Remodeling

Introduction Birds and mammals develop a dual series circulation, in which deoxygenated blood transits the pulmonary circuit and oxygenated blood transits the systemic circuit to more efficiently deliver oxygen to the peripheral tissues. The transition from a single to dual circulation during development is a complex process involving multiple cell types. The embryonic cardiac outflow tract (OFT), the portion of the embryonic heart connecting the developing ventricles to

* Corresponding author. Department of Medicine, Case Western Reserve University, 422 Biomedical Research Building, 2109 Adelbert Road, Cleveland, OH 44106. Fax: +1-216-368-0507. E-mail address: [email protected] (S.A. Fisher). 0012-1606/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ydbio.2004.05.036

the aortic sac, plays a key role in the establishment of the dual circulation (reviewed in Pexieder et al., 1995; Rothenberg et al., 2003; Webb et al., 2003). In birds and mammals, OFT myocardium is added from a ‘secondary (anterior) heart-forming field’ from splanchnic mesoderm beneath the caudal pharynx after the formation of the primitive heart tube from the primary heart-forming field. (Kelly et al., 2001; Mjaatvedt et al., 2001; Waldo et al., 2001). The OFT subsequently undergoes a complex remodeling process in the transition to a dual circulation. The OFT is divided into pulmonary and systemic circuits by septation, a process dependent upon invasion of cells migrating from the neural crest (Kirby et al., 1983). The OFT also shortens and rotates in order for the pulmonary artery to connect to the right ventricle in a position anterior and to the right of the aorta’s connection to the left ventricle (De La Cruz et al., 1977;

286

Y. Sugishita et al. / Developmental Biology 273 (2004) 285–296

Thompson et al., 1987; Watanabe et al., 1998). The shortening and rotation of the avian OFT require the spatially and temporally regulated elimination of OFT cardiomyocytes by programmed cell death (apoptosis) between stages 25 and 32 (Hamburger and Hamilton, 1951) of chicken development (Sallee et al., 2004; Watanabe et al., 1998, 2001). We have proposed that regional (OFT) myocardial hypoxia is a trigger for the apoptosis-dependent remodeling of the OFT based on (1) increased binding of the nitroimidazole EF5, an indicator of tissue hypoxia (Evans et al., 1997), to the avian OFT myocardium during the stages of OFT remodeling (25 – 32); (2) expression of the hypoxia-responsive vascular endothelial growth factor receptor (VEGFR2) by the OFT myocardium; (3) approximately coincident migration of endothelial progenitor cells to the OFT and vasculogenesis; and (4) incubation of eggs in hyperoxic air attenuates the hypoxic stimulus, diminishes the prevalence of apoptosis, and perturbs OFT remodeling (Sugishita et al., 2004). However, the specific role of tissue hypoxia in the remodeling of the embryonic cardiac OFT remains to be established. Tissue hypoxia may elicit a broad range of responses, many of which are dependent upon the hypoxia-inducible transcription factors (HIFs) (reviewed in Semenza et al., 1999). The HIFs are heterodimeric transcription factors consisting of inducible (HIFa-1,2,3) and constitutive [HIF-1h, also known as arylhydrocarbon receptor nuclear translocator (ARNT) subunits, each of which contains basic helix– loop – helix and PAS (Per –Arnt–Sim, named for the founding members)] domains. The activity of HIF is largely determined by the induction of the HIF-a subunit in response to hypoxia. With tissue hypoxia, the oxygendependent proline hydroxylation of HIF-1a is inhibited (reviewed in Kaelin, 2002; Masson and Ratcliffe, 2003), allowing nuclear accumulation of HIF1a and transcription of HIF-dependent target genes. The role of HIFs in development has been examined by gene inactivation in mice. Mice lacking HIF-1a or HIF-1h have early embryonic lethality (Kotch et al., 1999; Maltepe et al., 1997), suggesting that these transcription factors are essential for the development of the cardiovascular system. Because of the early embryonic lethality, the specific role of HIFs in the subsequent development of the cardiovascular system and, in particular, in the remodeling of the OFT has not been addressed. In the current study, we sought to determine (1) if the expression of the HIF subunits localizes to the developing cardiac tissues that we have proposed are hypoxic based on EF5 binding; (2) if HIF expression is oxygensensitive in the developing heart; and (3) the relationship between hypoxia-dependent gene expression and apoptosis during heart development, particularly in regards to the remodeling of the OFT. Our results suggest complexity in hypoxia/HIF-1 signaling in regulating cardiomyocyte apoptosis in the OFT and the necessity of regulated VEGF/ VEGFR2 signaling for OFT remodeling in the transition to the dual series circulation of warm blooded animals.

Materials and methods Animals and exposure to hypoxia and hyperoxia Fertile White Leghorn (Gallus gallus) chicken eggs were obtained from Squire Valleevue Farm (Cleveland, OH). Fertile quail (Cortunix Cartunex Japonica) eggs were obtained from G.Q.F. Manufacturing Co. (Savannah, GA). Eggs were incubated in a humidified room air incubator at 38jC to the appropriate stages for each experiment [generally Hamburger-Hamilton (HH) stages 18– 32 (ED 3 –8)] (Hamburger and Hamilton, 1951). For exposure to hyperoxia, eggs were then placed in a sealed chamber (Modular Incubator Chamber, Billups-Rothenberg, Inc., Del Mar, CA) with water, and the chamber was set in a humidified incubator (Circulated Air Incubator Model 1250, G.Q.F. Manufacturing Co.). The sealed chamber was flushed with 95% O2/5% CO2 for 10 min to completely exchange the air, and the inlet and outlet tubes were subsequently clamped. The chamber was flushed with oxygen gas on a daily basis to maintain hyperoxia. Control eggs were incubated in the same chamber in room air. For exposure to hypoxia, eggs were incubated in a Water Jacketed CO2 Incubator (Model 3130, Forma Scientific, Inc., Marietta, OH) in which oxygen concentration was digitally set and controlled by the infusion of nitrogen gas. Immunohistochemistry Embryos were fixed in fresh 4% paraformaldehyde at 4jC for the appropriate time (1 –2 h) for each stage and cryosectioned at 10-A thickness. The primary antibodies used in this study were polyclonal rabbit anti-HIF-1a antibody (provided by Dr. F. Agani) at 1:200 dilution, polyclonal rabbit anti-HIF-1h antibody used at 6 Ag/ml (kindly provided by Dr. R. Pollenz), mouse anti-VEGFR2 IgG1 (kindly provided by Dr. Eichmann) without dilution, rabbit antiactive caspase-3 antibody (R and D Systems, Minneapolis, MN) at 1:200 dilution, and mouse anti-HA antibody (Roche, Indianapolis, IN) at 0.2 Ag/ml. Antiactive caspase-3 antibody and anti-HA antibody were detected with an appropriate secondary antibody conjugated with Texas Red (Southern Biotechnology Associates, Inc., Birmingham, AL) or Alexa Fluor 488 (Molecular Probes, Eugene, OR), respectively. Anti-HIF-1a antibody, anti-HIF-1h antibody, and antiVEGFR2 antibody were detected with goat antirabbit or antimouse IgG secondary antibody conjugated with biotin (Vector, Burlingame, CA), and the signal was amplified with TSA and fluorescein isothiocyanate (FITC) (Perkin Elmer, Boston, MA) as per the manufacturer’s instructions. Selected sections stained with anti-HIF-1a antibody were also stained with phalloidin conjugated with Alexa Fluor 568 (Molecular Probes) to delineate myocardium, or 4V,6-diamindino-2-phenylindole (DAPI) (Sigma– Aldrich, St. Louis, MO) to stain the nuclei. Sections were observed with a fluorescent microscope (Leica DM LB, Leica Microsystems, Wetzlar, Germany) and images captured with SPOT RT camera and

Y. Sugishita et al. / Developmental Biology 273 (2004) 285–296

software v3.1 (Diagnostic Instruments, Inc., Sterling Heights, MI). Apoptosis was detected by LysoTracker Red (LTR; Molecular Probes) staining as previously described (Schaefer et al., 2004). All images were modified with Adobe Photoshop v.7 to optimize the signal. For comparisons between samples, sections were stained and images captured and modified in parallel.

287

1014 (Ad-dnAkt and Ad-caAkt) or 1012 (Ad-VEGF and AdCMVGFP) pfu/ml. Each recombinant adenovirus was applied in a volume of 500 nl to the surface of HH stages 18 – 19 chick or quail embryo hearts as previously described (Fisher and Watanabe, 1996). After an additional incubation appropriate for each experiment, the embryos were harvested for analyses.

Western blots Results Tissues were microdissected from cardiac OFT of chick embryos at stages 23, 26, 29, and 33 and homogenized in lysis buffer consisting of 20 mM Tris, 150 mM NaCl, 1% Triton X-100, 10% (v/v) Sigma Protease inhibitor cocktail at a final pH of 7.5. Protein concentrations were determined by the bicinchoninic acid (BCA) (Pierce, Rockford, IL) method. Nuclear and cytoplasmic fractions were prepared from whole hearts of stage 25 chick embryos using the Nuclear Extraction kit (Active Motif, Carlsbad, CA). Samples were loaded at 10 – 15 Ag per lane onto a 4 – 12% gradient polyacrylamide gel and run for 2.5 h and subsequently transferred to polyvinylidene difluoride (PVDF) membrane. The membrane was blocked with 5% nonfat dry milk in Tris-buffered saline with 1% Tween (TBST) for 1 h. AntiHIF-1a primary antibody was applied at a dilution of 1:500 and allowed to incubate at 4jC overnight. Antirabbit IgG secondary antibody conjugated with peroxidase (Affinity Bioreagents, Golden, CO) was applied at a dilution of 1:1000 and allowed to incubate for 1 h at room temperature. Signals were detected with the enhanced chemiluminescence detection system (ECL) (Amersham Biosciences, Piscataway, NJ). Recombinant replication-defective adenovirus Recombinant adenoviruses expressing dominant negative mutant Akt (Ad-dnAkt) or constitutively active Akt (Ad-caAkt) under the control of the cytomegalovirus (CMV) promoter/enhancer were kindly provided by Dr. Kenneth Walsh (Boston University). An HA tag is fused in frame to the N termini of the Akt coding sequences in both of the adenoviruses. The Ad-dnAkt virus expresses the mutant Akt [K179A, T308A, S473A] that cannot be activated by phosphorylation, and the Ad-caAkt construct has the c-Src myristoylation sequence fused in frame to the N terminus of the HA-Akt (wild-type) coding sequence (Fujio and Walsh, 1999). A recombinant adenovirus expressing human VEGF165 under the control of the CMV promoter/ enhancer (AdVEGF-A165) was obtained from Pittsburgh Experimental Gene Therapy Vector Core Facility, University of Pittsburgh. The recombinant adenovirus expressing green fluorescent protein (GFP) under the control of the CMV promoter/enhancer (AdCMVGFP) was used as a control. Recombinant adenoviruses were amplified in 293 cells, purified and titered as previously described (Fisher and Watanabe, 1996). Titers were approximately

Hypoxia-inducible factor in the embryonic chick heart We first examined expression of HIF-1a in the OFT myocardium of the stage 30 (ED 7) embryonic chick heart, the stage at which EF5 binding (hypoxia) and cardiomyocyte apoptosis are maximal (Sugishita et al., 2004; Watanabe et al., 1998). HIF-1a protein was detected in the proximal and distal region of the OFT myocardium of the stage 30 chick (n = 3) (Fig. 1A). Costaining with HIF-1a and phalloidin –Alexa Fluor 568, which binds to filamentous actin and thus identifies cardiomyocytes, showed colocalization, indicating that HIF-1a was within the myocardial compartment of the OFT (Fig. 1B). HIF-1a was also detected in the adjacent OFT endocardial cushion mesenchymal cells. No signal was detected when the primary antibody was omitted (n = 2) (Fig. 1C) or when the antiHIF-1a antibody was preincubated with competing peptide (n = 2) (data not shown). Costaining with DAPI (blue, Fig. 1E) and overlaying of images showed that the HIF-1a localized to the nuclei (Fig. 1F, light blue nuclei). Thus, HIF-1a is present in the nuclei of cardiomyocytes and mesenchymal cells of the stage 30 OFT, the stage when OFT myocardial EF5 binding (hypoxia) is most intense (Sugishita et al., 2004). We next examined the stage dependence of HIF-1a expression. HIF-1a was not detected in the heart at stage 22, just before the apoptosis-dependent remodeling of the OFT (data not shown). Nuclear accumulation of HIF-1a was detected at stage 25 throughout the OFT myocardium (Fig. 2A), as well as at stages 27, 30, and 32 (n = 2 for each stage) (Figs. 2B– D), encompassing the period when OFT myocardium is EF5-positive and apoptosis-dependent OFT remodeling occurs (stages 25– 32, ED5 – 8) (Sugishita et al., 2004; Watanabe et al., 1998). HIF-1a was not detected ubiquitously in the nuclei of the ventricular and atrial myocardium (not shown) but was detected in a few specific sites at specific stages of development. HIF-1a-positive nuclei were present in the atrial myocardium adjacent to the OFT in what is referred to as the inner curvature of the heart (Fig. 2A). At stage 30, HIF1a-positive nuclei were detected along the outer edge of the ventricular free wall near the apex adjacent to the epicardium (n = 3) (Figs. 3A – C). HIF-1a-positive nuclei were also evident in the myocardium of the right ventricle immediately adjacent to the OFT of the stage 32 embryo (Fig. 2D). Thus, HIF-1a

288

Y. Sugishita et al. / Developmental Biology 273 (2004) 285–296

Fig. 1. HIF-1a in the stage 30 chick embryo OFT. HIF-1a protein was detected with an anti-HIF-1a antibody and TSA amplification of the fluorescein isothiocyanate (FITC) signal (Perkin Elmer) as described in Materials and methods. HIF-1a was detected throughout the proximal and distal regions of OFT myocardium (A, arrows). (B) Costaining for HIF-1a (green) and phalloidin – Alexa Fluor 568 (red), which binds f-actin and delineates the myocardium. HIF1a (green) colocalized with the phalloidin staining of the myocardium (red) in the OFT (arrow) and is also present in the adjacent endocardial cushion mesenchyme. (C) No signal is present when the primary antibody for HIF-1a is omitted. (D, E, F) Region indicated by the rectangle in Panel A at higher magnification showing HIF-1a staining (D), 4V,6-diamindino-2-phenylindole (DAPI) staining of nuclei (E), and the overlay of images (F) to show nuclear localization of HIF-1a (light blue nuclei). Images in this and subsequent figures were captured with Spot RT digital camera and modified with Adobe Photoshop software to optimize the signal. Images within a figure were processed in parallel and captured at identical exposures. Bars, 200 Am in A – C, 50 Am in D – F.

accumulates in the nuclei of stage 25– 32 OFT myocardium correlating with the relative tissue oxygen concentrations as assessed by EF5 binding. In contrast to the regulated expression of HIF-1a, HIF1h was detected throughout the stages 25 and 30 embryonic chick heart (Figs. 4A –D). We confirmed that the HIF-1h was localized to nuclei by costaining with DAPI (data not

shown). Thus, the nuclear accumulation of HIF-1h protein appears to be constitutive, as it does not localize to the hypoxic regions of the embryonic chick heart. Effects of altering ambient oxygen concentrations on HIF-1a expression We next determined if the expression of HIF-1a in the embryonic heart was sensitive to oxygen concentrations by exposing the embryos within the intact egg to hypoxic or hyperoxic air using protocols that we previously showed to result in increased or decreased EF5 binding, respectively (Sugishita et al., 2004). Incubation of stage 24 eggs in 7.5% O2 for 7 h induced nuclear accumulation of HIF-1a in myocardial and endocardial cells throughout the heart (n = 2) (Figs. 5A –D). Incubation of stage 25 embryos in 95% O2/5% CO2 for 1 day abolished the HIF-1a nuclear accumulation observed in OFT myocardium of stage-matched embryos incubated under normoxic conditions (n = 2) (Figs.

Fig. 2. HIF-1a in the OFT during remodeling (stages 25 – 32). Immunohistochemistry for HIF-1a was performed in stages 25 (ED 5) to 32 (ED 8) chick embryos, the period when OFT myocardium is hypoxic and cardiomyocyte apoptosis occurs. (A) HIF-1a was present in OFT myocardium at stage 25, the stage when the EF5 signal indicating hypoxia is faintly detected (Sugishita et al., 2004) and is especially intense in the inner curvature of the OFT. HIF-1a continued to be nuclear-localized in the OFT myocardium at stages 27 (ED 6), 30 (ED 7), and 32 (ED 8) (B – D). Bars, 200 Am.

Fig. 3. HIF-1a in the stage 30 chick ventricle. (A) HIF-1a protein was detected in the outermost region of the ventricular free wall adjacent to the epicardium. (B, C) Magnified picture of the region within the rectangle in (A), and showing HIF-1a (B).

Y. Sugishita et al. / Developmental Biology 273 (2004) 285–296

289

These results were corroborated by analysis of protein expression by Western blotting. A 120-kDa band corresponding to HIF-1a protein was detected in stages 26 – 33 OFT homogenates (Fig. 6). HIF-1a was detected in the nuclear and not the cytoplasmic fraction of embryonic heart extracts and significantly up-regulated by hypoxia. Relationship of HIF-1a expression to OFT cardiomyocyte apoptosis

Fig. 4. HIF-1h in the embryonic chick heart. HIF-1h protein was detected with an anti-HIF-1h antibody and TSA amplification of the FITC signal as described in Materials and methods. HIF-1h was in detected cells of the OFT (A, C), ventricles (B, D), and atria (not shown) of stages 25 (A, B) and 30 (C, D) embryonic chick heart. HIF-1h localized to the nuclei by DAPI costaining (not shown). Bars, 50 Am.

5E, F). Thus, the nuclear accumulation of HIF-1a in the OFT myocardium is associated with tissue hypoxia and modulated by ambient oxygen concentrations.

Fig. 5. HIF-1a responds to changes in ambient oxygen concentrations. HIF1a in a stage 24 chick embryo incubated in (A, C) normoxic (21% O2) and (B, D) hypoxic air (7.5% O2 for 7 h). Nuclear accumulation of HIF-1a was undetectable in stage 24 control OFT and ventricle and was induced by hypoxia in cells of the OFT and ventricle. (E) HIF-1a is present in the stage 27 OFT myocardium and (F) is suppressed by incubation in hyperoxic air (95% O2/5% CO2 for 24 h). In F, the autofluorescence of red blood cells is seen within the lumen of the ventricle and OFT. Bars, 100 Am.

As a first step in defining the relationship between HIF1a expression and OFT cardiomyocyte apoptosis, we determined if they colocalized during heart development. We used the anti-HIF-1a antibody and LysoTracker Red (LTR), or the antiactive caspase-3 antibody, to colocalize apoptosis and HIF-1a in the hearts of stages 29– 32 chick embryos. HIF-1a was detected in nuclei throughout the OFT myocardium of the stage 29 embryo. There was particularly strong staining for HIF-1a in the dorsal portion of the OFT in the region referred to as the inner curvature. Apoptotic cardiomyocytes in the stage 29 embryo, detected by either LTR (Fig. 7) or antiactive caspase-3 antibody (data not shown), were concentrated in the proximal OFT myocardium (n = 2), as we have shown previously (Schaefer et al., 2004; Watanabe et al., 1998). Few or no apoptotic cardiomyocytes were detected in the distal OFT myocardium or the dorsal region of the OFT of the stage 29 embryo. Thus, there is overlap of nuclear HIF-1a and cardiomyocyte apoptosis in the proximal OFT, while nuclear HIF-1a is also detected in the distal and dorsal OFT myocardium, where there is minimal apoptosis at the earlier stages of OFT remodeling (25 – 30). To more directly test the role of hypoxic induction of HIF-1a in cardiomyocyte apoptosis, intact stage 24 eggs were incubated under hypoxic conditions (7.5% oxygen for 5 h) that induced HIF-1a nuclear accumulation in the heart. These embryos examined in whole mount showed increased LTR staining in the OFT but not in the remainder of the heart (n = 5 experimental, n = 4 control embryos; Figs. 8A, B). In sections, it was apparent that this increased LTR

Fig. 6. Detection of HIF-1a protein by Western blot. Nuclear extract from deferoxamine mesylate-treated Hepa cells (positive control, Con), homogenates of embryonic chick OFTs from stages 23, 26, 29, and 33, and nuclear and cytoplasmic extracts of stage 25 chick hearts incubated in normoxia (21% O2) or hypoxia (7.5% O2, 7 h) were collected and analyzed by Western blotting as described in Materials and methods. Ten micrograms of protein was loaded from each sample. Specific bands for HIF-1a were detected at 120 kDa. Additional bands were not observed. HIF-1a is expressed most intensely in the stages 26 and 29 OFT, localizes to the nuclear fraction, and is induced by hypoxia.

290

Y. Sugishita et al. / Developmental Biology 273 (2004) 285–296

death of the myocardium and nonviable embryos (data not shown). VEGFR2 and cardiomyocyte apoptosis

Fig. 7. Costaining for HIF-1a and LysoTracker Red (LTR), a marker of apoptosis, in the stage 29 chick embryo. HIF-1a was detected in both proximal and distal regions of the OFT myocardium and in both the ventral and dorsal OFT myocardium. LTR, which accumulates in the lysosomes of dying cardiomyocytes (Schaefer et al., 2004), was abundant only in the proximal-ventral OFT myocardium at this stage, as we previously reported (Schaefer et al., 2004). Similar results were obtained at earlier stages of development (not shown). Bars, 200 Am.

signal localized to the cushion mesenchyme of the OFT, not the myocardium (Figs. 8C, D). In addition, the LTR signal was not above background in the ventricular myocardium of hypoxic embryos (Fig. 8B), although HIF-1a accumulated in the nuclei of the ventricular myocytes under these conditions. Therefore, hypoxia or HIF-1a nuclear accumulation by itself is insufficient for inducing cardiomyocyte apoptosis in the developing embryo. We also used other hypoxia exposure protocols, including a variety of hypoxia/ normoxia regimens for up to 24 h, but in no instance did we observe significant increases in apoptosis in the OFT or ventricular myocardium. Short exposures to more intense hypoxia, for example, 3 h at 5% oxygen, resulted in necrotic

Hypoxia signaling through HIF-1a may have pro- or antiapoptotic effects depending on the context (reviewed in Wenger, 2002, and Discussion). We previously showed that VEGFR2 is expressed in cardiomyocytes in the distal OFT myocardium of the stage 30 embryo and that the expression is oxygen-sensitive (Sugishita et al., 2004). This raised the possibility that expression of VEGFR2 in the OFT myocardium is a hypoxia-dependent cytoprotective response. As a first step in defining the relationship between VEGFR2 expression and OFT cardiomyocyte apoptosis, we examined their dynamic patterns of expression during heart development. At stage 25, VEGFR2 protein was weakly detected in the OFT myocardium and was abundant in the endocardium/endothelium and in the distal OFT at the site of formation of the aorticopulmonic septum (Fig. 9A). The intensity of VEGFR2 signal in the OFT myocardium increased at subsequent stages and was most intense in the myocardium of the distal OFT (Figs. 9B –D). The signal was maximal at stage 30 and diminished at stage 32 (Figs. 9C, D). At stages 30 and 32, the signal for VEGFR2 was also detected in the epicardium at the site of vessel formation (Figs. 9C, D; arrowheads) and within the endocardial cushions as they are shaped into valves. We next colocalized VEGFR2 expression with markers of apoptosis to determine if there may be a relationship between the two. At stages 29 and 30, VEGFR2 was

Fig. 8. Effect of ambient hypoxia on cardiomyocyte apoptosis in the OFT. Chick embryos at stages 24 – 25 were incubated in normoxic (21% O2) or hypoxic (7.5% O2 for 5 h) air and stained with LTR. The number of LTR particles observed in whole mount is increased in the OFT in embryos incubated in (B) hypoxia as compared with (A) normoxia. In section, it is evident that the increased LTR in the embryos incubated in hypoxic air is in the cushion mesenchyme of the OFT (compare D with C). There was no significant difference between the two groups in LTR in the OFT myocardium (C, D). Bars, 200 Am.

Y. Sugishita et al. / Developmental Biology 273 (2004) 285–296

Fig. 9. Stage dependence of VEGFR2 expression in the chick OFT. VEGFR2 expression was examined in stages (A) 25, (B) 27, (C) 30, and (D) 32 chicken embryos. The intensity of VEGFR2 staining in the OFT myocardium had a stage dependence similar to that of EF5 binding (hypoxia) (Sugishita et al., 2004). It was first detected at stage 25, maximal at stage 30 and diminished by stage 32, and most prominent in the distal OFT myocardium. At stage 25, VEGFR2 was also present in the forming aorticopulmonic septum (A, arrowheads). At stages 30 and 32, VEGFR2 was also detected in endothelial cells within the epicardium at the apparent sites of vessel formation and on the downstream surface of the stage 32 forming valve leaflets (C, D, arrowheads). At all stages, VEGFR2 was detected in the endothelium lining the heart and great vessels. Bars, 500 Am.

expressed primarily in the distal region of OFT myocardium, while LTR was predominantly in the proximal region of the OFT myocardium (Figs. 10A, B; arrowheads). Two days later, at stage 31, LTR signals were detected throughout the OFT myocardium, including the distal region (Fig. 10C; arrowheads), where VEGFR2 expression was observed. At stage 32, the OFT myocardium had shortened, and VEGFR2 signals and LTR staining, while still present, were markedly diminished (Fig. 10D). The same results were observed by costaining with anti-VEGFR2 and antiactive caspase-3 antibodies (data not shown). Thus, in the early stages of OFT remodeling (25 –30), cardiomyocyte apoptosis is excluded from the distal region of the OFT where VEGFR2 expression is maximal. Subsequently, after several more days of development, the more distal OFT cardiomyocytes also succumb to a death-inducing stimulus, while the surviving OFT cardiomyocytes form the subpulmonic infundibulum (Watanabe et al., 1998). The PI3K/Akt pathway mediates the cell-protective effect of VEGF/VEGFR signaling (reviewed in Wick et al., 2002). To test the role of VEGF receptor/tyrosine kinase signaling in survival of OFT cardiomyocytes, stages 18 –19 embryonic hearts were transduced in ovo with recombinant replication-defective adenovirus expressing dominant negative (AddnAkt) or constitutively active (AdcaAkt) Akt (Fujio and Walsh, 1999). Staining for the HA tag showed that mutant Akt was selectively expressed in the OFT

291

myocardium (Fig. 11A), as we have previously shown for the expression of other transgenes in this model (Fisher and Watanabe, 1996; Watanabe et al., 1998). No difference in the frequency of apoptosis was observed between control and AddnAkt or AdcaAkt transduced hearts 1– 2 days after incubation (stages 22 –25, n = 12 AdAkt, n = 6 control embryos, not shown). Three days after adenovirus delivery (stages 28 –30), there were significantly more LTR particles throughout the OFT of the Ad-dnAkt embryos as compared to control embryos (n = 7 Ad-dnAkt vs. n = 4 control, Figs. 11B, C). Examination of these hearts in section revealed that the increased apoptosis was confined to the OFT myocardium (data not shown). Three days after the delivery of the Ad-caAkt adenovirus (stages 28 –30), there were significantly fewer LTR particles in the OFT as compared to control embryos (n = 4 each, Fig. 11D). Thus, forced expression of dominant negative or constitutively active Akt protein specifically affects OFT cardiomyocyte apoptosis during the stages of hypoxia-dependent apoptotic remodeling of the OFT, suggesting that VEGF signaling through VEGFR2 and Akt may provide a survival signal to the OFT cardiomyocytes (see Discussion). Forced expression of VEGF165 causes OFT defects To further test the role of hypoxia-regulated gene expression in OFT remodeling, we used recombinant adenovirus to force the expression of the VEGF ligand (VEGF-A165 isoform). We used quail embryos for these

Fig. 10. Costaining for VEGFR2 and LysoTracker Red in the OFT. Localization of VEGFR2 and cardiomyocyte apoptosis in OFT myocardium was examined in stages (A) 29, (B) 30, (C) 31, and (D) 32 chick embryos. (A, B) At stages 29 and 30, VEGFR2-positive myocardium was found mainly in the distal region of OFT myocardium (arrows), while LysoTracker Red (LTR) signals, a marker of apoptotic cardiomyocytes, were localized in the proximal region of OFT myocardium (arrowheads). At stage 31, LTR signals were observed along the entire length of the OFT myocardium (arrowheads) and overlapped with the distal region of OFT expressing VEGFR2 expression. At stage 32, OFT myocardium shortened, and VEGFR2 and LTR signals were diminished (D). Bars, 200 Am.

292

Y. Sugishita et al. / Developmental Biology 273 (2004) 285–296

thinning was also observed. Thus, hypoxia-dependent responses that include regulated expression of VEGF and VEGFR signaling are required for the normal developmental remodeling of the OFT in the transition to a dual circulation.

Discussion

Fig. 11. Effect of dominant negative and constitutively active Akt on OFT cardiomyocyte apoptosis. Stage 18 chick embryos were infected with recombinant replication-defective adenoviruses expressing dominant negative or constitutively active Akt as described in Materials and methods. (A) The hemagglutinin tag on the adenoviral-expressed Akt is detected by immunohistochemistry exclusively in the OFT of the stage 27 chick embryo. We have previously shown that this method of transgene delivery targets expression to the OFT myocardium. (B – D) Prevalence of apoptosis was detected by LTR staining and observed in intact (whole mount) hearts. (B) A control stage 30 chick embryo shows apoptosis concentrated at the base of the OFT (arrow). (C) A stage 30 chick embryo heart transduced with adenovirus expressing dominant negative mutant Akt. The number of LTR particles was significantly increased throughout the OFT (arrow) as compared with control embryos (D) A stage 30 chick embryo transduced with adenovirus expressing constitutively active Akt. The number of LTR particles was significantly diminished in OFT (arrow) compared with control embryos. Bars, 200 Am in A, 500 Am in B – D. OFT indicates outflow tract; RV, right ventricle; LV, left ventricle.

experiments as an antibody (QH1) is available to specifically identify endothelial progenitors in quail only (Pardanaud et al., 1987), and apoptosis-dependent remodeling of the OFT is similar to that of the chicken (Rothenberg et al., 2002). Forced expression of VEGF-A165 in the embryonic quail heart OFT myocardium had no significant effect on the prevalence of apoptosis during OFT remodeling (data not shown) but did have a significant effect on OFT morphogenesis (Fig. 12). In control embryos, the rotation and shortening of the OFT are nearly complete by stage 31, so that the pulmonary artery connects to the right ventricle in a position anterior and slightly rightward of the connection of the aorta to the left ventricle (Figs. 12A –C). In 9 of 14 embryos transduced with AdVEGF-A165, both aorta and pulmonary artery connected to right ventricle, showing double outlet right ventricle (DORV) morphology (Figs. 12D – I). In some instances, there was also a large ventricular septal defect (VSD) (Figs. 12G – I). The OFT cushions appeared hypercellular, and in some embryos, ventricular

We previously proposed a model in which myocardial hypoxia triggers the apoptosis-dependent remodeling of the avian OFT in the transition to a dual circulation (Sugishita et al., 2004). In this study, we demonstrate that HIF-1a specifically accumulates in the nuclei of OFT cardiomyocytes from stages 25 – 32, the stages of increased EF5 binding and apoptosis-dependent remodeling of the OFT (Sugishita et al., 2004; Watanabe et al., 1998). HIF-1a expression and EF5 binding are suppressed by increased ambient oxygen concentrations, supporting the contention that these are markers of OFT myocardial hypoxia during the phase of OFT remodeling (stages 25 –32). Nuclear HIF1a protein was induced in cells throughout the heart by incubation at 7.5% oxygen (approximately 57 mm Hg), further indicating that nuclear accumulation of HIF-1a in the embryonic heart is a hypoxic response. This concentration of oxygen is much higher than those that are reported to induce HIF-1a in cultured cells (reviewed in Semenza et al., 1999), suggesting that the working embryonic heart may be more sensitive than are cultured cells to oxygen deprivation. HIF-1a protein was also detected in tissues where we did not previously detect increased EF5 binding, such as the stage 30 endocardial cushion mesenchyme adjacent to the myocardium and the stage 30 ventricular myocardium at the apex adjacent to the epicardium. The comparison between EF5 binding and abundance of HIF-1a is complicated by the likely differing sensitivity and dynamic range of the assays depending upon the antibody used and method of detection. Thus, it is possible that the cells that are HIF-1apositive but were not observed to have increased EF5 binding are hypoxic relative to other cells but are less hypoxic than the OFT myocardium. Using signal amplification to detect EF5 binding has revealed regions of apparent tissue hypoxia in the embryonic heart that was not evident in our initial analysis (unpublished data). Alternatively, it is possible that HIF-1a nuclear accumulation in some of the cells of the embryonic heart is a response to other signals, for example, cytokines, which also may stabilize HIF-1a protein (reviewed in Bruick, 2003; Wenger, 2002). In contrast to the hypoxia-dependent accumulation of HIF-1a in the embryonic heart, the nuclearlocalization of its dimerization partner HIF-1h (Arnt) was constitutive. An earlier study suggested that nuclear accumulation of HIF-1h protein is spatially and temporally restricted during heart development (Walker et al., 1997). In this study, we used a signal amplification method to detect HIF-1h (and HIF-1a) which may obscure modest

Y. Sugishita et al. / Developmental Biology 273 (2004) 285–296

293

Fig. 12. The effect forced expression of VEGF-A165 on the development of the heart and OFT. Stage 18 quail embryo hearts were transduced with recombinant replication-defective adenovirus expressing VEGF-A165 or GFP as a control. Sections were prepared from stage 31 control (AdGFP) and AdVEGF-A165 hearts. Representative frontal sections, from anterior to posterior, are shown for (A – C) control and (D – I) two different AdVEGF-A165 hearts. In the control embryos, the aorta and pulmonary artery were connected to left and right ventricles, respectively. The RVOT is anterior to the Ao and tilts to the left while the LVOT course from left to right. In the embryos with AdVEGF-A165, both the aorta and pulmonary artery connect to the right ventricle [double outlet right ventricle (DORV)], and the spiral relationship of the two outlets is absent. (G – I) A large VSD is also observed. Some of the AdVEGF-hearts also have thinning of the ventricular myocardium. Bars, 200 Am. PA indicates pulmonary artery; Ao, aorta; RV, right ventricle; LV, left ventricle; RA, right atrium; LA, left atrium; VSD, ventricular septal defect.

differences in protein expression, although the HIF-1h subunit in most other systems is constitutively expressed. How might hypoxia regulate the apoptosis-dependent remodeling of the embryonic cardiac outflow tract? Hypoxia signaling through HIF-1-dependent or -independent transcriptional mechanisms may induce the expression of proor antiapoptotic genes, depending on the cell type and context (Alvarez-Tejado et al., 2001; Dong and Wang, 2004; Dong et al., 2001; Kubasiak et al., 2002; Regula et al., 2002; Sowter et al., 2001). Supporting the complexity of the relationship between HIF expression and cell death/ survival in vivo, we found HIF1a expression in regions of

the OFT myocardium where apoptosis was both prevalent as well as sparse. In contrast, myocardial expression of VEGFR2 was restricted to the more distal portion of the OFT myocardium where during the early stages of OFT remodeling (stages 25– 30, ED5 – 7) cardiomyocyte apoptosis is not prevalent. As the VEGFR2 gene (and VEGF gene) contains hypoxia-responsive elements and in many instances are regulated by hypoxia/HIFs (Elvert et al., 2003; Kappel et al., 1999), this suggests the hypothesis that hypoxia (HIF)-dependent induction of VEGFR2 in the OFT myocardium protects the cardiomyocytes from potential death-inducing signals. An autocrine protective effect of

294

Y. Sugishita et al. / Developmental Biology 273 (2004) 285–296

VEGFR signaling through Akt has been demonstrated in other models of hypoxia and ischemia (Alvarez-Tejado et al., 2001; Jin et al., 2000; Ogunshola et al., 2002; Rissanen et al., 2002; Wick et al., 2002) reviewed in Shiojima and Walsh (2002). Using several commercially available antibodies, we were not able to detect activation (phosphorylation) of endogenous VEGFR2 or of its downstream effector, the kinase Akt in situ (data not shown). However, we used a loss-of-function approach and showed that forced expression of dominant negative Akt (kinase-defective) significantly increased apoptosis in the OFT cardiomyocytes. This response appeared to be specific to the hypoxic challenge as it was only observed approximately 3 days after viral transduction (stages 28– 30 embryos), the stage of maximal hypoxic stress in the OFT myocardium. In contrast, in this system, transgene expression is maximal at 24 –48 h after transduction with recombinant adenovirus (Fisher and Watanabe, 1996), and apoptosis is induced as early as 12 – 24 h (stages 20– 21) after forced expression of a death receptor (Fas) ligand (Sallee et al., 2004). This suggests that the delayed apoptotic response to dnAkt is specific to the deathinducing stimulus, which we propose is hypoxia, between stages 25 and 32. We also used gain-of-function approach by expressing constitutively active (myristylated) Akt to show a reduction in the prevalence of apoptosis in the stages 28 – 30 OFT cardiomyocytes. These experiments provide strong evidence that the kinase Akt provides a survival signal to the OFT cardiomyocytes. Akt signaling is not specific to VEGFR, and it remains to be shown that Akt is activated by VEGFR in these cells during normal development. A number of targets of Akt survival signaling have been identified, including the Bcl-2 family member Bad, Ask (apoptosis signal regulating kinase), and Flip (Fas ligand inhibitory peptide) (reviewed in Lawlor and Alessi, 2001), and it will be of interest to examine these potential targets in the developing OFT. Does hypoxia induce the programmed cell death of OFT cardiomyocytes during normal development? The coincidence of EF5 binding, an indicator of tissue hypoxia, and stage-dependent apoptosis is suggestive of a relationship between the two. There also is overlap between the expression of HIF-1a in the proximal OFT myocardium and apoptosis, further suggesting a relationship between hypoxia and apoptosis. However, a hypoxic stimulus that was sufficient to induce nuclear accumulation of HIF-1a throughout the heart did not result in apoptosis in either the OFT myocardium or cardiomyocytes in other chambers of the heart. These results could be interpreted as showing that tissue hypoxia (and nuclear accumulation of HIF-1a) by itself is not sufficient to induce apoptosis of cardiomyocytes. This would be akin to in vivo studies of the adult myocardium where hypoxia (ischemia) plus reperfusion are more effective at inducing apoptosis than hypoxia (ischemia) alone (reviewed in Gottlieb, 2003; Regula et al., 2003). Similarly with in vitro studies of isolated cardiomyocytes, hypoxia plus acidosis or metabolic inhibition are required to

induce apoptosis (Bishopric et al., 2001; Dong and Wang, 2004; Webster et al., 1999). One important distinction between this study and the in vitro experiments is in the degree of hypoxia. In vitro experiments are done at near anoxic conditions (less than 1% oxygen) for prolonged periods (24 – 48 h). In the working embryonic heart, in ovo exposure to 5% oxygen for 3 h resulted in myocardial necrosis and cessation of the circulation. It is thus problematic to extrapolate from in vitro to in vivo studies, although it is possible that different results would be obtained with regional myocardial hypoxia as opposed to the global hypoxia of the working heart in ovo used in this study. It is possible that the hypoxic stimulus is required at the specific stage in development during which it normally occurs and that it is the response to the stimulus at these specific stages (25 –32) that triggers apoptosis-dependent remodeling of the OFT. In this scenario, precocious exposure to hypoxia would not induce apoptosis. One such response to tissue hypoxia could be the recruitment of endothelial progenitors cells to the OFT and vasculogenesis (‘‘developmental reperfusion’’), a process that is coincident with apoptosis-dependent remodeling of the OFT (Rothenberg et al., 2002; Sugishita et al., 2004). Support for this scenario is provided by our previous experiments which showed that delaying the epicardial coverage of the OFT myocardium, thereby inhibiting proendothelial cell invasion of the OFT myocardium, significantly reduced the prevalence of apoptosis (Rothenberg et al., 2002). Other studies have shown that hypoxia/ischemia up-regulates the expression of death ligands/factors in cardiomyocytes(Tanaka et al., 1994) and endothelial cells (Scarabelli et al., 2001); in this scenario, the invading endothelial progenitor cells would trigger OFT cardiomyocyte apoptosis through the death receptor pathway. We have not yet identified an endogenous death ligand in the OFT but have shown that forced expression of a death ligand is a potent killer of embryonic cardiomyocytes in ovo (Sallee et al., 2004). Hypoxia/HIF-1 may also induce the transcription of proapoptotic Bcl-2 family members such as BNIP3, Nix, and Noxa in cardiomyocytes and nonmuscle cells (Bruick, 2000; Kim et al., 2004; Kubasiak et al., 2002; Lee et al., 2004; Regula et al., 2002; Yussman et al., 2002) that trigger cell death through the mitochondrial pathway. The pattern of expression of these proteins in relation to developmental apoptosis remains to be investigated. Our previous study showed that prolonged exposure to hyperoxia results in morphologic defects of the OFT (Sugishita et al., 2004) that phenocopy the effects of inhibitors of apoptosis (Watanabe et al., 2001). This suggests that perturbations in the remodeling of the OFT in the transition to a dual circulation may involve oxygen-sensitive mechanisms. In the current study, we showed that forced expression of a hypoxia-responsive gene, VEGF-A165, caused similar conal defects, DORV with large VSD, as well as ventricular thinning. These defects are similar to those previously reported in chick embryos injected with

Y. Sugishita et al. / Developmental Biology 273 (2004) 285–296

VEGF protein (Feucht et al., 1997) or mice that overexpress VEGF (Miquerol et al., 2000; Stalmans et al., 2003). In the current study, an effect of VEGF165 on OFT cardiomyocyte apoptosis was not observed, and the cell or molecular mechanism by which overexpression of VEGF leads to OFT defects remains to be determined. VEGF has also been proposed to stimulate the proliferation of endocardial cushion cells and inhibit their transformation to a mesenchymal phenotype (Dor et al., 2001, 2003) which could also contribute to the observed phenotype. In conclusion, hypoxia-dependent responses are critical to the normal developmental remodeling of the OFT in the transition of the avian embryo to a dual circulation. The survival of OFT cardiomyocytes that eventually form the subpulmonic infundibulum involves expression of VEGFR2 and Akt signaling. Whether hypoxia and HIF-1 induce OFT cardiomyocyte apoptosis through transcriptional responses and a death receptor and/or mitochondrial pathway requires further study. It will also be of interest to determine if tissue hypoxia during the remodeling of the OFT myocardium renders this tissue susceptible to toxic effects of teratogens that are associated with congenital human conotruncal heart defects.

Acknowledgments Anti-HIF-1h antibody was a generous gift from Dr. R. Pollenz. Anti-VEGFR2 antibody was a generous gift from Dr. A. Eichmann. Ad-dnAkt and Ad-caAkt were provided by Dr. K. Walsh. The authors thank Yong-Qiu Doughman for excellent technical assistance. This work was supported by NIH Grant 1 R01 HL 65314-01 (S.A.F.) and AHA Ohio Valley Grant-in-Aid (M.W.). Y.S. is supported by an AHA Ohio Valley Postdoctoral Fellowship and Japan Heart Foundation/Bayer Yakuhin Research Grant Abroad.

Appendix A Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ydbio.2004. 05.036.

References Alvarez-Tejado, M., Naranjo-Suarez, S., Jimenez, C., Carrera, A.C., Landazuri, M.O., del Peso, L., 2001. Hypoxia induces the activation of the phosphatidylinositol 3-kinase/Akt cell survival pathway in PC12 cells. Protective role in apoptosis. J. Biol. Chem. 276, 22368 – 22374. Bishopric, N.H., Andreka, P., Slepak, T., Webster, K.A., 2001. Molecular mechanisms of apoptosis in the cardiac myocyte. Curr. Opin. Pharmacol. 1, 141 – 150. Bruick, R.K., 2000. Expression of the gene encoding the proapoptotic nip3 protein is induced by hypoxia. Proc. Natl. Acad. Sci. U. S. A. 97, 9082 – 9087. Bruick, R.K., 2003. Oxygen sensing in the hypoxic response pathway:

295

regulation of the hypoxia-inducible transcription factor. Genes Dev. 17, 2614 – 2623. De La Cruz, M.V., Gomez, C.S., Arteaga, M.M., Arguello, C., 1977. Experimental study of the development of the truncus and the conus in the chick embryo. J. Anat. 123, 661 – 686. Dong, Z., Wang, J., 2004. Hypoxia selection of death-resistant cells: a role for Bcl-XL. J. Biol. Chem. 279, 9215 – 9221. Dong, Z., Venkatachalam, M.A., Wang, J., Patel, Y., Saikumar, P., Semenza, G.L., Force, T., Nishiyama, J., 2001. Up-regulation of apoptosis inhibitory protein IAP-2 by hypoxia. HIF-1-independent mechanisms. J. Biol. Chem. 276, 18702 – 18709. Dor, Y., Camenisch, T.D., Itin, A., Fishman, G.I., McDonald, J.A., Carmeliet, P., Keshet, E., 2001. A novel role for VEGF in endocardial cushion formation and its potential contribution to congenital heart defects. Development 128, 1531 – 1538. Dor, Y., Klewer, S.E., McDonald, J.A., Keshet, E., Camenisch, T.D., 2003. VEGF modulates early heart valve formation. Anat. Rec. 271A, 202 – 208. Elvert, G., Kappel, A., Heidenreich, R., Englmeier, U., Lanz, S., Acker, T., Rauter, M., Plate, K., Sieweke, M., Breier, G., Flamme, I., 2003. Cooperative interaction of hypoxia-inducible factor-2alpha (HIF-2alpha ) and Ets-1 in the transcriptional activation of vascular endothelial growth factor receptors-2 (Flk-1). J. Biol. Chem. 278, 7520 – 7530. Evans, S.M., Bergeron, M., Ferriero, D.M., Sharp, F.R., Hermeking, H., Kitsis, R.N., Geenen, D.L., Bialik, S., Lord, E.M., Koch, C.J., 1997. Imaging hypoxia in diseased tissues. Adv. Exp. Med. Biol. 428, 595 – 603. Feucht, M., Christ, B., Wilting, J., 1997. VEGF induces cardiovascular malformation and embryonic lethality. Am. J. Pathol. 151, 1407 – 1416. Fisher, S.A., Watanabe, M., 1996. Expression of exogenous protein and analysis of morphogenesis in the developing chicken heart using adenoviral vector. Cardiovasc. Res. 31, E86 – E95. Fujio, Y., Walsh, K., 1999. Akt mediates cytoprotection of endothelial cells by vascular endothelial growth factor in an anchorage-dependent manner. J. Biol. Chem. 274, 16349 – 16354. Gottlieb, R.A., 2003. Mitochondrial signaling in apoptosis: mitochondrial daggers to the breaking heart. Basic Res. Cardiol. 98, 242 – 249. Hamburger, V., Hamilton, J.L., 1951. A series of normal stages in the development of the chick embryo. J. Morphol. 88, 49 – 92. Jin, K.L., Mao, X.O., Greenberg, D.A., 2000. Vascular endothelial growth factor: direct neuroprotective effect in in vitro ischemia. Proc. Natl. Acad. Sci. 97, 10242 – 10247. Kaelin Jr., W.G., 2002. How oxygen makes its presence felt. Dev. Genes 16, 1441 – 1445. Kappel, A., Ronicke, V., Damert, A., Flamme, I., Risau, W., Breier, G., 1999. Identification of vascular endothelial growth factor (VEGF) receptor-2 (Flk-1) promoter/enhancer sequences sufficient for angioblast and endothelial cell-specific transcription in transgenic mice. Blood 93, 4284 – 4292. Kelly, R.G., Brown, N.A., Buckingham, M.E., 2001. The arterial pole of the mouse heart forms from Fgf10-expressing cells in pharyngeal mesoderm. Dev. Cell 1, 435 – 440. Kim, J.Y., Ahn, H.J., Ryu, J.H., Suk, K., Park, J.H., 2004. BH3-only protein Noxa is a mediator of hypoxic cell death induced by hypoxiainducible factor 1{alpha}. J. Exp. Med. 199, 113 – 124. Kirby, M.L., Gale, T.F., Stewart, D.E., 1983. Neural crest cells contribute to normal aorticopulmonary septation. Science 220, 1059 – 1061. Kotch, L.E., Iyer, N.V., Laughner, E., Semenza, G.L., 1999. Defective vascularization of HIF-1alpha-null embryos is not associated with VEGF deficiency but with mesenchymal cell death. Dev. Biol. 209, 254 – 267. Kubasiak, L.A., Hernandez, O.M., Bishopric, N.H., Webster, K.A., 2002. Hypoxia and acidosis activate cardiac myocyte death through the Bcl-2 family protein BNIP3. Proc. Natl. Acad. Sci. 99, 12825 – 12830. Lawlor, M.A., Alessi, D.R., 2001. PKB/Akt: a key mediator of cell proliferation, survival and insulin responses? J. Cell Sci. 114, 2903 – 2910. Lee, M.J., Kim, J.Y., Suk, K., Park, J.H., 2004. Identification of the hypoxia-

296

Y. Sugishita et al. / Developmental Biology 273 (2004) 285–296

inducible factor 1{alpha}-responsive HGTD-P gene as a mediator in the mitochondrial apoptotic pathway. Mol. Cell. Biol. 24, 3918 – 3927. Maltepe, E., Schmidt, J.V., Baunoch, D., Bradfield, C.A., Simon, M.C., 1997. Abnormal angiogenesis and responses to glucose and oxygen deprivation in mice lacking the protein arnt. Lett. Nat. 386, 403 – 407. Masson, N., Ratcliffe, P.J., 2003. HIF prolyl and asparaginyl hydroxylases in the biological response to intracellular O(2) levels. J. Cell Sci. 116, 3041 – 3049. Miquerol, L., Langille, B.L., Nagy, A., 2000. Embryonic development is disrupted by modest increases in vascular endothelial growth factor gene expression. Development 127, 3941 – 3946. Mjaatvedt, C.H., Nakaoka, T., Moreno-Rodriguez, R., Norris, R.A., Kern, M.J., Eisenberg, C.A., Turner, D., Markwald, R.R., 2001. The outflow tract of the heart is recruited from a novel heart-forming field. Dev. Biol. 238, 97 – 109. Ogunshola, O.O., Antic, A., Donoghue, M.J., Fan, S.Y., Kim, H., Stewart, W.B., Madri, J.A., Ment, L.R., 2002. Paracrine and autocrine functions of neuronal vascular endothelial growth factor (VEGF) in the central nervous system. J. Biol. Chem. 277, 11410 – 11415. Pardanaud, L., Altmann, C., Kitos, P., Dieterlen-Lievre, F., Buck, C.A., 1987. Vasculogenesis in the early quail blastodisc as studied with a monoclonal antibody recognizing endothelial cells. Development 100, 339 – 349. Pexieder, T., 1995. The conotruncus and its septation at the advent of the molecular biology era. In: Clark, E.B., Markwald, R.R., Takao, A. (Eds.), Developmental Mechanisms of Heart Disease. Futura Publishing Co, Inc., Armonk, NY, pp. 227 – 248. Regula, K.M., Ens, K., Kirshenbaum, L.A., 2002. Inducible expression of BNIP3 provokes mitochondrial defects and hypoxia-mediated cell death of ventricular myocytes. Circ. Res. 91, 226 – 231. Regula, K.M., Ens, K., Kirshenbaum, L.A., 2003. Mitochondria-assisted cell suicide: a license to kill. J. Mol. Cell. Cardiol. 35, 559 – 567. Rissanen, T.T., Vajanto, I., Hiltunen, M.O., Rutanen, J., Kettunen, M.I., Niemi, M., Leppanen, P., Turunen, M.P., Markkanen, J.E., Arve, K., Alhava, E., Kauppinen, R.A., Yla-Herttuala, S., 2002. Expression of vascular endothelial growth factor and vascular endothelial growth factor receptor-2 (KDR/Flk-1) in ischemic skeletal muscle and its regeneration. Am. J. Pathol. 160, 1393 – 1403. Rothenberg, F., Hitomi, M., Fisher, S.A., Watanabe, M., 2002. Initiation of apoptosis in the developing avian outflow tract myocardium. Dev. Dyn. 223, 469 – 482. Rothenberg, F., Fisher, S.A., Watanabe, M., 2003. Sculpting the cardiac outflow tract. Birth Defects Res., Part C. Embryol. Today 69, 38 – 45. Sallee, D., Qiu, Y., Liu, J., Watanabe, M., Fisher, S.A., 2004. Fas ligand gene transfer to the embryonic heart induces programmed cell death and outflow tract defects. Dev. Biol. 267, 309 – 319. Scarabelli, T., Stephanou, A., Rayment, N., Pasini, E., Comini, L., Curello, S., Ferrari, R., Knight, R., Latchman, D., 2001. Apoptosis of endothelial cells precedes myocyte cell apoptosis in ischemia/reperfusion injury. Circulation 104, 253 – 256. Schaefer, K.S., Doughman, Y.Q., Fisher, S.A., Watanabe, M., 2004. Dynamic patterns of apoptosis in the developing chicken heart. Dev. Dyn. 229, 489 – 499. Semenza, G.L., Agani, F., Iyer, N., Kotch, L., Laughner, E., Leung, S., Yu, A., 1999. Regulation of cardiovascular development and physiology by hypoxia-inducible factor 1. Ann. N. Y. Acad. Sci. 874, 262 – 268.

Shiojima, I., Walsh, K., 2002. Role of Akt signaling in vascular homeostasis and angiogenesis. Circ. Res. 90, 1243 – 1250. Sowter, H.M., Ratcliffe, P.J., Watson, P., Greenberg, A.H., Harris, A.L., 2001. HIF-1-dependent regulation of hypoxic induction of the cell death factors BNIP3 and NIX in human tumors. Cancer Res. 61, 6669 – 6673. Stalmans, I., Lambrechts, D., De Smet, F., Jansen, S., Wang, J., Maity, S., Kneer, P., von der, O.M., Swillen, A., Maes, C., Gewillig, M., Molin, D.G., Hellings, P., Boetel, T., Haardt, M., Compernolle, V., Dewerchin, M., Plaisance, S., Vlietinck, R., Emanuel, B., Gittenberger-de Groot, A.C., Scambler, P., Morrow, B., Driscol, D.A., Moons, L., Esguerra, C.V., Carmeliet, G., Behn-Krappa, A., Devriendt, K., Collen, D., Conway, S.J., Carmeliet, P., 2003. VEGF: a modifier of the del22q11 (DiGeorge) syndrome? Nat. Med. 9, 173 – 182. Sugishita, Y., Watanabe, M., Fisher, S.A., 2004. Role of myocardial hypoxia in the remodeling of the embryonic avian cardiac outflow tract. Dev. Biol. 267, 294 – 308. Tanaka, M., Ito, H., Adachi, S., Akimoto, H., Nishikawa, T., Kasajima, T., Marumo, F., Hiroe, M., 1994. Hypoxia induces apoptosis with enhanced expression of Fas antigen messenger RNA in cultured neonatal rat cardiomyocytes. Circ. Res. 75, 426 – 433. Thompson, R.P., Abercrombie, V., Wong, M., 1987. Morphogenesis of the truncus arteriosus of the chick embryo heart: movements of autoradiographic tattoos during septation. Anat. Rec. 218, 434 – 440. Waldo, K.L., Kumiski, D.H., Wallis, K.T., Stadt, H.A., Hutson, M.R., Platt, D.H., Kirby, M.L., 2001. Conotruncal myocardium arises from a secondary heart field. Development 128, 3179 – 3188. Walker, M.K., Pollenz, R.S., Smith, S.M., 1997. Expression of the aryl hydrocarbon receptor (ahr) and ahr nuclear translocator during chick cardiogenesis is consistent with 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced heart defects. Toxicol. Appl. Pharmacol. 143, 407 – 419. Watanabe, M., Choudhry, A., Berlan, M., Singal, A., Siwik, E., Mohr, S., Fisher, S.A., 1998. Developmental re-modeling and shortening of the cardiac outflow tract involves myocyte programmed cell death. Development 125, 3809 – 3820. Watanabe, M., Jafri, A., Fisher, S.A., 2001. Apoptosis is required for the proper formation of the ventriculo-arterial connections. Dev. Biol. 240, 274 – 288. Webb, S., Qayyum, S.R., Anderson, R.H., Lamers, W.H., Richardson, M.K., 2003. Septation and separation within the outflow tract of the developing heart. J. Anat. 202, 327 – 342. Webster, K.A., Discher, D.J., Kaiser, S., Hernandez, O., Sato, B., Bishopric, N.H., 1999. Hypoxia-activated apoptosis of cardiac myocytes requires reoxygenation or a pH shift and is independent of p53. J. Clin. Invest. 104, 239 – 252. Wenger, R.H., 2002. Cellular adaptation to hypoxia: O2-sensing protein hydroxylases, hypoxia-inducible transcription factors, and O2-regulated gene expression. FASEB J. 16, 1151 – 1162. Wick, A., Wick, W., Waltenberger, J., Weller, M., Dichgans, J., Schulz, J.B., 2002. Neuroprotection by hypoxic preconditioning requires sequential activation of vascular endothelial growth factor receptor and Akt. J. Neuroscience 22, 6401 – 6407. Yussman, M.G., Toyokawa, T., Odley, A., Lynch, R.A., Wu, G., Colbert, M.C., Aronow, B.J., Lorenz, J.N., Dorn, G.W., 2002. Mitochondrial death protein Nix is induced in cardiac hypertrophy and triggers apoptotic cardiomyopathy. Nat. Med. 8, 725 – 730.