Apoptosis and Regulation of Bax and Bcl-X Proteins During Human Neonatal Vascular Remodeling
Abstract—To verify that apoptosis is one of the possible mechanisms of neonatal vascular remodeling during the transition from fetal to neonatal circulation, we assayed for apoptosis and evaluated the expression of apoptosis-regulatory proteins in umbilical vessel versus ascending aorta, ductus arteriosus (DA) versus adjacent pulmonary artery and aorta, or aorta versus its branching arteries. Twenty-two umbilical cords (UCs), 6 DAs with adjacent aortas and pulmonary arteries, and 4 aortic arches with their branching great arteries were obtained from neonates. Smooth muscle cell (SMC) apoptosis in umbilical vessels was identified in all UCs. The expressions of Bax and Bcl-X were stronger in umbilical artery than in the neonatal aorta, but Bcl-2 was weak in both arteries in immunohistochemistry. In the immunoblot analysis of UCs, the expression of the proapoptotic short isoform of Bcl-X was stronger than in other tissue, and caspase-3 was selectively activated, whereas it was not in the other components of the cardiovascular system. In contrast, the expression patterns of the FasAg and Fas ligand were similar in umbilical artery and aorta. Regulation of Bcl-2 family proteins was also observed in other vascular sites at which SMCs undergo apoptosis on hemodynamic changes during birth, such as the DA and the branching points of the great arteries from the aortic arch. Apoptosis is involved in the regression of human umbilical vessels and the DA and in the remodeling of the branching great arteries during the neonatal period, when Bcl-2 family proteins are likely to play a key role.
Presented in part at the 71st Scientific Sessions of the American Heart Association, Dallas, Tex, November 10, 1998, and published in abstract form (Circulation. 1998;98[suppl I]:I-331).
- Received June 10, 1999.
- Accepted October 8, 1999.
Apoptosis is a form of programmed cell death that plays a vital role in the maintenance of cell homeostasis in mature organisms, in morphogenesis during development, and in the pathogenesis of various disorders.1 During perinatal morphogenesis of the cardiovascular system, apoptosis is known to be involved in the molding and shaping of the conduction system, including the atrioventricular node and His bundle,2 and the remodeling of the vasculature according to perinatal hemodynamic changes such as the regression of the abdominal aorta.3
The 2 most important changes in the circulation system at birth are the closure of umbilical vessels and the ductus arteriosus. In previous studies of ours and others, it has been demonstrated that certain unique features of umbilical vessels and the ductus arteriosus may contribute to closure after birth. First, they are both arterial systems that close at birth. Second, they contract when exposed to oxygen.4 5 Third, they constrict sensitively to endothelin-1.6 7 Fourth, they are composed of smooth muscle cells with adult phenotype, even during the neonatal period.8 In addition to these physiological properties and phenotypical characteristics, other mechanisms, such as apoptosis, could be involved in the closure of these vessels, a process that is the most dramatic example of vascular remodeling at birth. It has recently been reported that apoptosis of smooth muscle cells was found in areas of cytolytic necrosis in the media of the human neonatal ductus,9 and another study suggested that apoptosis occurred in the postnatal intra-abdominal umbilical artery of neonatal lamb.3 However, no reports have examined apoptosis-regulatory protein expression during apoptosis or neonatal vascular remodeling in humans.
To verify apoptosis and its mechanisms that may be involved in the closure of the human umbilical artery and the ductus arteriosus and in the remodeling of the great arteries that branch off the aortic arch at birth, we investigated the occurrence of apoptosis and analyzed the expression pattern of its related genes. In particular, we examined the expression of the Bcl-2 gene families and the Fas ligand/FasAg, which constitute 2 independent pathways to apoptosis,10 11 and the activation of caspase-3, which has been found in both atherosclerotic plaques12 and vascular smooth muscle cells,13 in these specific vascular sites, comparing them with the pattern seen in neonatal human aorta or pulmonary artery.
Collection and Preparation of Tissue Samples
Twenty-two umbilical cords, with a gestational age at the time of delivery of between 34 and 42 weeks, were obtained within 10 minutes of birth; external influences were thus minimized. They were immediately frozen in liquid nitrogen for immunoblot analysis or fixed in 10% buffered formalin for in situ terminal deoxynucleotidyl transferase (TdT)–mediated dUTP nick end-labeling (TUNEL) and immunohistochemistry. As a control, 2 thoracic aortas from neonates who died of noncardiac causes 1 or 2 days after birth were obtained at autopsy and fixed in 10% buffered formalin for immunohistochemistry. Surgical specimens of stomach, colon, and lymph node were also obtained and prepared as described previously.
Six ductus arteriosus with adjacent aorta and pulmonary artery were obtained from autopsy cases of 1-day-old (n=4), 4-day-old, and 5-day-old neonates. Four aortic arches with branching great arteries were obtained from autopsy cases of 1-day-old (n=3) and 4-day-old neonates. These were fixed in 10% buffered formalin for TUNEL and immunohistochemistry.
All specimens were obtained with the approval of the Institutional Review Board of Seoul National University Hospital.
Formalin-fixed tissue sections were deparaffinized, rehydrated, and treated with proteinase K (20 μg/mL), and endogenous peroxidase activity was blocked with 3% hydrogen peroxide.
In situ end-labeling of apoptotic cells was performed as described in the protocol provided by the manufacturers of the in situ apoptosis detection kit (Oncor). After a brief equilibration, end-labeling with digoxigenin-11-dUTP by TdT enzyme in buffer was carried out for 1 hour at 37°C in a humidifying chamber. After treatment with stop/wash buffer, sections were incubated with anti-digoxigenin antibody–peroxidase conjugate, rinsed, and stained with diaminobenzidine tetrahydrochloride. Negative controls were incubated with PBS instead of TdT enzyme, and positive controls were treated with DNase 1. Sections were counterstained with Mayer’s hematoxylin and mounted.
Taq Polymerase–Based In Situ Ligation Assay
Because apoptotic nuclei are known to have DNA ends ligatable to labeled DNA fragments with single-base 3′ overhangs that could be obtained from polymerase chain reaction (PCR) with Taq polymerase, a 270-bp double-stranded DNA fragment was prepared by PCR using primers for the ACE gene and genomic DNA from patients with the DD genotype of ACE gene polymorphism as described in our previous study.14 To label this DNA fragment with digoxigenin, 16.6 μmol/L of digoxigenin-11-dUTP was added to the PCR reaction. Digoxigenin-labeled fragments were ligated to nicked DNA of apoptotic cells in tissue sections in situ with T4 DNA ligase (Boehringer Mannheim) as described in a previous study.15
After treatment with blocking solutions, sections were incubated with anti-digoxigenin antibody–peroxidase conjugate, rinsed, and stained with diaminobenzidine tetrahydrochloride. Negative controls were incubated with PBS instead of T4 ligase, and positive controls were treated with DNase 1. Sections were counterstained with Mayer’s hematoxylin and mounted.
Umbilical arteries and veins were dissected into 1×1×1-mm fragments, fixed in 2.5% glutaraldehyde (pH 7.3) containing 0.1 mol/L phosphate buffer for 6 hours at 4°C, and then postfixed with 1% osmium tetroxide in 0.1 mol/L phosphate buffer for 1 hour at 4°C. After dehydration, tissue was embedded in liquid epoxy resin that underwent polymerization. Semithin or ultrathin sections were obtained and stained with toluidine blue or uranyl acetate and lead citrate. Ultrathin sections were examined with a transmission electron microscope (TEM) (Hitachi H-600) at 50 kV.
After deparaffinization and rehydration, sections were blocked for endogenous peroxidase activity with 3% hydrogen peroxide and incubated with goat serum (DAKO A/S) to inhibit nonspecific background staining. The following were used as primary antibodies during the course of this work: rabbit polyclonal anti-human Bax-α antibody (Biochemistry, 1:100), mouse monoclonal anti-human Bcl-2 antibody (DAKO A/S, 1:20), rabbit polyclonal anti-human Bcl-XS/L (Santa Cruz, 1:50), rabbit polyclonal anti-human FasAg antibody, and Fas ligand antibody (Santa Cruz, 1:25). These were incubated for 1 hour at room temperature. Sections were further incubated with biotinylated goat anti-rabbit immunoglobulin (DAKO A/S, 1:1000) or biotinylated goat anti-mouse immunoglobulin (DAKO A/S, 1:1000) and with horseradish peroxidase–conjugated streptavidin. Peroxidase activity was demonstrated by exposing sections to diaminobenzidine tetrahydrochloride for 8 minutes.
After being crushed with a mortar and pestle, 200 mg of snap-frozen tissues of umbilical cords, as well as surgical specimens of colons and lymph nodes, and snap-frozen tissues of heart and aorta of a traffic accident victim as control tissues, were immediately suspended in a 1.5-mL lysis buffer containing 1 mmol/L Tris (pH 7.4), 15 mmol/L NaCl, 1% Triton X-100, 0.1% SDS, 0.5 mmol/L EDTA, 1 mmol/L PMSF, 0.3 U/mL aprotinin, and 50 μg/mL leupeptin for 30 minutes on ice. Samples were sonicated for 30 minutes until the supernatant became clear. After centrifugation, the supernatant was separated and the protein concentration measured with a Bio-Rad protein assay kit (Bio-Rad Laboratories). Aliquots containing 15 μg of protein were size-fractionated by SDS-PAGE (12% gel) and transferred to a nitrocellulose membrane. Blots were blocked with 10% skim milk in TBS-T (Tris-buffered saline, 0.2% Tween-20) for 12 hours and then incubated with primary antibody, either rabbit polyclonal anti-human Bax-α (Biochemistry, 1:100), mouse monoclonal anti-human Bcl-2 (DAKO A/S, 1:20), rabbit polyclonal anti-human Bcl-XS/L (Santa Cruz, 1:50), or rabbit polyclonal anti-human CPP32 (Santa Cruz, 1:1000) for 1 hour at room temperature. Blots were then incubated with horseradish peroxidase–conjugate donkey anti-rabbit immunoglobulin antibody or sheep anti-mouse immunoglobulin antibody (Amersham, 1:1000) for 1 hour at room temperature. For detection, blots were incubated with a mixture of detection reagents 1 and 2 (Amersham), and chemiluminescence was captured on film from 15 seconds to 15 minutes.
Occurrence of Apoptosis in Human Umbilical Vessel at Birth
In all umbilical cords examined, TUNEL-positive cells were present at birth in both umbilical arteries and veins. In umbilical arteries, ≈70% to 80% of intimal cells and 30% of medial cells were TUNEL-positive (Figure 1⇓, A, B, and E), as were ≈80% of endothelial cells and 30% of medial cells in umbilical veins (Figure 1⇓, C, D, and E). This pattern of apoptosis was observed in all immediately postnatal umbilical cords, regardless of their gestational periods at the time of delivery. In neonatal human aorta, however, TUNEL-positive cells were hardly visible (Figure 1⇓, E). This high frequency of apoptosis in umbilical vessels by TUNEL assay could also be observed by Taq polymerase–based in situ ligation assay, which is a more specific method to detect apoptotic cells (see Figure⇓ I; Figures I through VI can be found online at http://atvb.ahajournals.org/cgi/content/full/20/4/957/DC1).
We confirmed by TEM that these TUNEL-positive cells undergo apoptosis; this process occurred in smooth muscle cells and endothelial cells, in which we observed peripheral condensation of chromatin and cytoplasmic condensation as well as apoptotic bodies (Figure⇑ II, online). There was no inflammatory cell infiltration around the apoptotic cells. In 5 umbilical arteries, apoptotic index was evaluated by TEM. Among the observed nuclei of smooth muscle cells in intima, ≈20% showed features of apoptosis (see Figure⇑ II, online). In 5 umbilical veins evaluated by TEM, ≈30% of endothelial cells showed apoptotic features (Figure⇑ III, online). This low apoptotic index by TEM compared with that by TUNEL and in situ ligation assay suggests that cells destined to undergo apoptosis become TUNEL-positive before they manifest typical morphological features of apoptosis.
Mechanism of Apoptosis in Human Umbilical Vessel
As possible mechanisms of apoptosis in human umbilical vessels, we can think of at least 2 apoptotic death pathways: the first via mitochondria, regulated by members of the Bcl-2 group, and the other initiated by the interaction of death ligand with a receptor, such as Fas ligand with FasAg. Thus, we examined the expression of Bax/Bcl-2/Bcl-X, FasAg/Fas ligand, and the activation of caspase-3.
Immunohistochemical staining showed that the Bax protein, a proapoptotic gene product, was strongly expressed in intimal and medial smooth muscle cells in the umbilical arteries immediately after birth (Figure 2⇓, B), whereas it was weakly expressed in neonatal human aorta (Figure 2⇓, C). In contrast, the expression of Bcl-2 protein, an antiapoptotic gene product, was rarely observed in either intimal or medial smooth muscle cells in umbilical arteries (Figure 2⇓, E), which was similar to what was observed in neonatal human aorta (Figure 2⇓, F). This differential expression of Bax and Bcl-2 in umbilical vessels, strong Bax and weak Bcl-2, was confirmed by immunoblot analysis (Figure 2⇓, G). This unique pattern of stronger immunoreactivity to Bax than to Bcl-2 in umbilical arteries was found in all immediately postnatal umbilical cords, regardless of their gestational periods at birth. Another member of the Bcl-2 family, Bcl-X, was also expressed selectively in umbilical artery but not in neonatal aorta (Figure 3⇓, B, C, and D). With regard to the 2 isoforms of Bcl-X, the relative expression of the proapoptotic short isoform is stronger in umbilical cords than in the control tissue of lymph nodes (Figure 3⇓, E and F). However, there was no difference between umbilical artery and neonatal aorta in the expression of FasAg or Fas ligand, ie, strong expression of FasAg and negligible expression of Fas ligand (data not shown).
This result, the selective high expression of Bax and Bcl-X protein in umbilical artery rather than neonatal aorta, having been accepted, the activation of the downstream molecule caspase was examined. Immunoblot analysis revealed that caspase-3 is activated in the umbilical artery, whereas it is not activated in other components of the cardiovascular system (Figure 4⇓).
Mechanism of Apoptosis at Ductus Arteriosus and Branching Points of Great Arteries
We tested whether this “Bax-associated apoptosis” is also observed in other vascular sites that undergo dramatic hemodynamic changes at birth. We compared the results obtained from the ductus arteriosus and the adjacent aorta or pulmonary artery (Figure⇑ IV, online) and the aortic arch and the branching points of the great arteries (Figure⇑ IV, B, online) on the same slide and under the same conditions of staining. In the case of the hematoxylin-eosin–stained ductus, pyknotic nuclei were frequently observed, in contrast to the normal plump nuclei in neonatal aorta (Figure⇑ V, A and E, online). These pyknotic nuclei were all TUNEL-positive, whereas positive cells are absent from the aorta (Figure⇑ V, B and F, online). In high-power fields of TUNEL staining and electron microscopy, we confirmed the presence of apoptosis in the ductus arteriosus (data not shown), which was associated with the differential expression of Bax. In other words, Bax is expressed more strongly in ductus but weakly in the adjacent aorta or pulmonary artery (Figure⇑ V, C and G, online). Bcl-2 was expressed neither in ductus nor in adjacent arteries (Figure⇑ V, D and H, online).
The same Bax-associated apoptosis was also observed in the branching points of the great arteries that branch off the aortic arch of neonates, which is in contrast to that found in the aortic arch itself (Figure⇑ VI, online).
This study demonstrates that apoptosis occurs in the human umbilical vessels during delivery and that the ratio of immunoreactivity to Bax/Bcl-2 and the expression of Bcl-X is higher than those in neonatal human aorta during the same period. This pattern of expression of Bcl-2 gene families was associated with the activation of caspase-3 in umbilical cords. Similar Bax-associated apoptosis was also observed in other vascular sites that undergo dramatic hemodynamic changes during the perinatal period, for example, ductus arteriosus and the branching points of the great arteries. These findings suggest that apoptosis is involved in the closure and regression of the umbilical vessels and the ductus arteriosus and in the remodeling process of the great arteries that branch from the aortic arch. Furthermore, they indicate that the specific expression of Bax and Bcl-X may play a key role in the process of neonatal vascular remodeling.
Apoptosis plays a major role in the morphogenesis and remodeling of organs according to interior and exterior environmental changes during development, as well as in the pathogenesis of diseases. The role of apoptosis in the human vascular system has been investigated. These studies explored the role of apoptosis in vascular disease by demonstrating the occurrence of apoptosis in human atheroma and restenotic lesions16 17 18 and by analyzing the mechanism of apoptosis using human vascular smooth muscle cells from atherosclerotic plaques.19 The role of apoptosis in the remodeling or morphogenesis of vasculature during development, however, has not been extensively studied. The most dramatic example of vascular remodeling during normal development in humans is the closure and regression of the umbilical artery and ductus arteriosus. It was recently reported that apoptosis was found in the umbilical vessels of neonatal lambs3 and in the human neonatal ductus.8 In this study, we demonstrated that apoptosis is present in the human umbilical artery at birth, the branching points of the great arteries, and the ductus arteriosus. Therefore, we believe that in humans, apoptosis plays an important role in vascular remodeling during normal development as well as in the pathogenesis of atherosclerosis.
It has been suggested that Bcl-2 family members, including Bcl-2 itself and Bax, are mediators of apoptosis.11 20 21 The balance of proapoptotic Bax and antiapoptotic Bcl-2 is known to be important in determining whether cells die or survive.22 In this study, Bax expression was significantly higher in the umbilical artery, the ductus arteriosus, and the branching points of the great arteries than in the aorta, the pulmonary artery, or the aortic arch, whereas Bcl-2 expression did not differ significantly between these arteries. This differential expression of Bax/Bcl-2 may be the underlying mechanism of apoptosis in these vessels as they undergo remodeling during the perinatal period.
Furthermore, Bcl-X, another partner molecule of Bax,11 20 21 was also selectively expressed in the umbilical artery rather than in the neonatal aorta. In a previous study,23 the withdrawal of Bcl-X expression in immunohistochemistry was associated with apoptosis immediately after denudation injury to rat carotid artery; in that case, Bcl-X was thought to be antiapoptotic. However, the relative expression of the 2 isoforms of Bcl-X was not examined.23 In our study, Bcl-X expression was associated with apoptosis, and the relative expression of the proapoptotic short isoform was greater in umbilical vessels than in the control tissue. Bcl-X expression data in both this and a previous study suggest that the regulation and role of this gene may differ during normal development from that which occurs during response to injury in adulthood.
In contrast to Bcl-2 families, there was no difference between umbilical artery and neonatal aorta in the expression pattern of FasAg or Fas ligand, other well-known regulators of apoptosis in mammalian cells,24 25 including human vascular smooth muscle cells.26 Thus, we believe that the activation of caspase-3 in the umbilical vessel is regulated by Bcl-2 gene families rather than by the Fas ligand/FasAg system.
In our previous study8 and others,4 5 6 7 a number of unique distinguishing features of umbilical vessels and the ductus arteriosus were demonstrated. Current findings concerning the regulation of Bcl-2 family proteins can now be added as a fifth characteristic of umbilical vessels and the ductus arteriosus. The types of stimuli that cause apoptosis of the umbilical vessels, the ductus arteriosus, and the branching points of the great arteries during the perinatal period are unknown. It is possible that during normal embryonic development, apoptosis follows a predetermined timetable, as exemplified by the disappearance of the interdigital web and the differentiation of the neural system. It has been reported that a classic morphogen, such as retinoic acid, induces apoptosis during the neural differentiation of embryonal stem cells,27 which suggests that the apoptosis of a specific organ may be induced by morphogens during a specific developmental period. In the present study, however, we found no difference in either apoptosis or the expression pattern of Bax/Bcl-2 in umbilical vessels of different gestational periods at birth. Furthermore, loci of vasculature where apoptosis occurs share the common feature of being subject to dramatic hemodynamic changes during the perinatal period. This suggests another possibility: that at birth, apoptosis in these vessels may be triggered by the factors accompanying the transition from fetal to neonatal circulation. It has been reported that blood flow reduction,28 distension of blood vessels,23 or regulators of the extracellular matrix, such as matrix metalloproteinase29 and its inhibitor,30 can affect the apoptosis of vascular smooth muscle cells. These factors may also be involved in the triggering of apoptosis in the vascular sites that undergo the remodeling process at birth.
In conclusion, the present study indicates that apoptosis might be involved in the closure and regression of human umbilical vessels and the ductus arteriosus and in the remodeling process of the great arteries that branch off the aortic arch. This study also indicates that the relative expression of Bax, Bcl-2, and Bcl-X and the activation of caspase-3 may be involved in the regulation of apoptosis in these vascular loci.
This study was supported by a research grant from the Seoul National University Hospital.
Thompson CB. Apoptosis in the pathogenesis and treatment of disease. Science. 1995;267:1456–1462.
James TN. Normal and abnormal consequences of apoptosis in the human heart: from postnatal morphogenesis to paroxysmal arrhythmias. Circulation. 1994;90:556–573.
Cho A, Courtman DW, Langille BL. Apoptosis (programmed cell death) in arteries of the neonatal lamb. Circ Res. 1995;76:168–175.
Noel S, Cassin S. Maturation of contractile response of ductus arteriosus to oxygen and drugs. Am J Physiol. 1976;231:240–243.
Kim HS, Aikawa M, Kimura K, Kuro-o M, Nakahara K, Suzuki T, Katoh H, Okamoto E, Yazaki Y, Nagai R. Ductus arteriosus: advanced differentiation of smooth muscle cells demonstrated by myosin heavy chain isoform expression in rabbits. Circulation. 1993;88:1804–1810.
Slomp J, Gittenberger-de Groot AC, Glukhova MA, Van Musteren JC, Kockx MM, Schwartz SM, Koteliansky VE. Differentiation, dedifferentiation, and apoptosis of smooth muscle cells during the development of the human ductus arteriosus. Arterioscler Thromb Vasc Biol. 1997;17:1003–1009.
Haunstetter A, Izumo S. Apoptosis: basic mechanisms and implications for cardiovascular disease. Circ Res. 1998;82:1111–1129.
Mallat Z, Ohan J, Leseche G, Tedgui A. Colocalization of CPP-32 with apoptotic cells in human atherosclerotic plaques. Circulation. 1997;96:424–428.
Wang HE, Keiser JA. Molecular characterization of rabbit CPP32 and its function in vascular smooth muscle cell apoptosis. Am J Physiol. 1998;274:H1132–H1140.
Didenko VV, Hornsby PJ. Presence of double-strand breaks with single-base 3′ overhangs in cells undergoing apoptosis but not necrosis. J Cell Biol. 1996;135:1369–1376.
Isner JM, Kearney M, Bortman S, Passeri J. Apoptosis in human atherosclerosis and restenosis. Circulation. 1995;91:2703–2711.
Bennett MR, Littlewood TD, Schwartz SM, Weissberg PL. Increased sensitivity of human vascular smooth muscle cells from atherosclerotic plaques to p53-mediated apoptosis. Circ Res. 1997;81:591–599.
Adams JM, Cory S. The Bcl-2 protein family: arbiters of cell survival. Science. 1998;281:1322–1326.
MacLellan WR, Schneider MD. Death by design: programmed cell death in cardiovascular biology and disease. Circ Res. 1997;81:137–144.
Perlman H, Maillard L, Krasinski K, Walsh K. Evidence for the rapid onset of apoptosis in medial smooth muscle cells after balloon injury. Circulation. 1997;95:981–987.
Ashikenazi A, Dixit VM. Death receptors: signaling and modulation. Science. 1998;281:1305–1308.
Nagata S, Golstein P. The Fas death factor. Science. 1995;267:1449–1456.
Geng YJ, Henderson LE, Levesque EB, Muszynski M, Libby P. Fas is expressed in human atherosclerotic intima and promotes apoptosis of cytokine-primed human vascular smooth muscle cells. Arterioscler Thromb Vasc Biol. 1997;17:2200–2208.
Okazawa H, Shimizu J, Kamei M, Imafuku I, Hamada H, Kanazawa I. Bcl-2 inhibits retinoic acid-induced apoptosis during the neural differentiation of embryonal stem cells. J Cell Biol. 1996;132:955–968.
Cho A, Mitchell L, Koopmans D, Langille BL. Effects of changes in blood flow rate on cell death and cell proliferation in carotid arteries of immature rabbits. Circ Res. 1997;81:328–337.
Jones PL, Crack J, Rabinovitch M. Regulation of tenascin-C, a vascular smooth muscle cell survival factor that interacts with the alpha-v beta-3 integrin to promote epidermal growth factor receptor phosphorylation and growth. J Cell Biol. 1997;139:279–293.