Donate Help Contact The AHA Sign In Home
American Heart Association
Circulation
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Circulation. 1995;91:494-504

This Article
Right arrow Abstract Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Payne, R. M.
Right arrow Articles by Strauss, A. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Payne, R. M.
Right arrow Articles by Strauss, A. W.

(Circulation. 1995;91:494-504.)
© 1995 American Heart Association, Inc.

Articles

Toward a Molecular Understanding of Congenital Heart Disease

R. Mark Payne, MD; Mark C. Johnson, MD; James W. Grant, MD; Arnold W. Strauss, MD

From the Departments of Pediatrics and Molecular Biology and Pharmacology (A.W.S.), Washington University School of Medicine, St Louis, Mo.

Correspondence to Arnold W. Strauss, MD, Division of Pediatric Cardiology, Department of Pediatrics, Box 8116, St Louis Children's Hospital, One Children's Place, St Louis, MO 63110.


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowVasculopathies: Genetic Defects...
 down arrowAnimal Models of CHD...
 down arrowHuman CHD Secondary to...
 down arrowReferences
 
Background This review discusses the incidence and importance of congenital heart disease (CHD), the reasons that investigation of causative mechanisms for human CHD has been slow, and the limitations of the multifactorial theory for the etiology of CHD.

Methods and Results The molecular defects underlying three vasculopathies—Marfan's syndrome (fibrillin), supravalvar aortic stenosis, and Williams' syndrome (elastin)—and hereditary telangiectasia are presented to emphasize the role of microfibrils and extracellular matrix in the pathophysiology of these vascular defects. Animal models of CHD, including situs inversus, canine conotruncal malformations, and chick neural crest ablation, are examined to emphasize how such studies relate to human CHD, especially by pointing to single-gene defects for conotruncal malformations, candidate loci for situs inversus, and phenotypic variability caused by neural crest lesions. The crucial role of cardiac transcription factors in heart morphogenesis is emphasized by review of gene knockout studies of these factors, which cause fetal death secondary to heart maldevelopment. Several lines of evidence demonstrating genetic etiologies of human CHD are also presented, including the mapping of familial atrial septal defects, to prove that one anatomic type of CHD may be due to single-gene defects at different loci. Review of atrioventricular canal, both secondary to trisomy 21 and as an autosomal-dominant familial defect, reiterates this conclusion. The evidence that monosomy on chromosome 22 causes multiple types of CHD, including aortic arch and conotruncal defects as part of the CATCH-22 syndrome, is presented, with results supporting the idea that deletions at this site alone may cause 5% of surgically treated CHD.

Conclusions We conclude that (1) human CHD is frequently due to single-gene defects and that even sporadic defects may arise from a single-gene abnormality; (2) a common genetic defect may cause several apparently different forms of CHD; (3) elucidation of the genetic basis of CHD provides clues to normal cardiovascular developmental biology; (4) the same cardiac malformation can be caused by mutant genes at different loci; and (5) interactions of clinical investigators (cardiologists and cardiothoracic surgeons) with basic scientists should allow more rapid progress in defining the genetic basis of CHD.


Key Words: molecular biology • heart diseases • genes


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowVasculopathies: Genetic Defects...
 down arrowAnimal Models of CHD...
 down arrowHuman CHD Secondary to...
 down arrowReferences
 
Congenital heart malformations constitute a common cause of birth defects, with the prevalence of confirmed defects ranging from 5 to 10 per 1000 live births. The recent report of the American Heart Association Task Force on Children and Youth emphasized that cardiovascular disease affects more than 600 000 infants, children, and youths in the United States.1 This large number includes infants and children with congenital heart disease (CHD) (anatomic or structural abnormalities), inherited progressive disorders such as Marfan's syndrome (MFS) and hypertrophic cardiomyopathy, acquired disorders such as Kawasaki disease and rheumatic heart disease, and cardiac dysrhythmias or conduction abnormalities. Tremendous advances in the diagnosis and treatment of congenital heart malformations have occurred in the last five decades since the first surgical interventions for patent ductus arteriosus and coarctation. Lethal and severely morbid anomalies can be surgically treated with success rates of 90% for most lesions. However, significant mortality and universal morbidity still exist for complex lesions such as hypoplastic left heart and single ventricle.

Despite these advances in diagnosis and treatment, understanding of the developmental causes and etiologies of CHD, which is essential to prevention and more physiological treatments, has, until very recently, been limited. Several reasons exist. First, human cardiac development is not amenable to study or experimental manipulation, and the molecular tools to characterize animal models of CHD have not been available until recently. Thus, the developmental biology of the mammalian cardiovascular system has not been studied with a view toward understanding the genetic causes of CHD. Second, classification of CHD, based necessarily on anatomic pathology and not on developmental or embryological considerations, may have obscured the relatedness of different defects such as conotruncal malformations. Third, the pervading view that the causes of CHD are multifactorial, with most not due to single-gene defects, has probably impeded investigation. This view suggests that most CHD is sporadic and is based on earlier surveys of the incidence of familial heart disease, in which 8% of defects were thought to be due to chromosomal or single-gene defects, 2% due to environmental teratogens, and 90% "multifactorial."2 Multifactorial means that cardiac malformations are caused by the combined effects of one or more alleles at a number of loci interacting with stochastic and/or environmental factors.2 Such a complex multivariable model has limited the generation of testable hypotheses or focus on single genes crucial in causing congenital heart defects. However, recent clinical investigations have clearly demonstrated a much higher incidence of inherited CHD. This has resulted, in part, because more patients with severe defects have survived and then had children, which has allowed observations to be made. It has thus become increasingly apparent that CHD is frequently inherited and that single-gene defects can cause CHD.

As these clinical investigations have pointed toward genetic etiologies for CHD, significant technical advances and achievements in human molecular genetics have occurred and demonstrate the feasibility of investigating genetic defects in CHD. Specific advances that are altering our understanding of the etiology of congenital heart malformations include (1) the detection of chromosomal microdeletions and small translocations by high-resolution chromosomal analysis and fluorescence in-situ hybridization (FISH); (2) the delineation of the genetic defects of familial heart disease through linkage analysis and reverse genetics; and (3) advances in animal genetics and embryo manipulation, permitting detailed studies of animal models of CHD. Although many results from these new approaches are just emerging, some clear trends are apparent. First, single-gene defects are much more common than was previously thought. Second, phenotypic variations and incomplete penetrance may result from limited ascertainment of mild anatomic variants that have minimal pathophysiological consequences such as the occurrence of bicuspid aortic valves in families with hypoplastic left heart syndrome. Third, a limited number of altered developmental mechanisms such as neural crest migration may explain a relatively large amount of CHD with significantly different apparent phenotypes, such as truncus arteriosus and interrupted aortic arch in the DiGeorge or CATCH-22 syndromes.3 In addition, alleles of the genes controlling these developmental mechanisms most likely exist to explain variations in phenotype and the low inheritability of particular anatomic cardiac malformations.

This review focuses on recent data demonstrating single-gene causes of CHD and vasculopathies, such as MFS. Some examples of animal models of CHD are included to highlight information that affects our understanding of human CHD. Genetic causes of progressive or acquired cardiovascular disease, including arrhythmias, atherosclerosis, cardiomyopathies (dilated and hypertrophic), and hypertension are not included because they have been reviewed recently.4 A partial list of proven single-gene defects responsible for such diseases is provided in Table 1Down, however, to emphasize the emerging importance of genetics for all cardiovascular pathophysiology.


View this table:
[in this window]
[in a new window]
  		Table 1. Human Gene Defects Causing Cardiomyopathy and Arrhythmias


*    Vasculopathies: Genetic Defects of the Vascular Extracellular Matrix Proteins
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Vasculopathies: Genetic Defects...
down arrowAnimal Models of CHD...
 down arrowHuman CHD Secondary to...
 down arrowReferences
 
Table 2Down lists several inherited human vasculopathies that cause symptoms in the pediatric age group and may require evaluation and treatment by pediatric cardiologists and cardiovascular surgeons. The recent delineations of precise mutations at single-gene loci responsible for these diseases emphasize rapid recent progress.


View this table:
[in this window]
[in a new window]
  		Table 2. Human Single-Gene Defects Causing Progressive Pediatric Vasculopathies

Marfan's Syndrome
MFS is characterized by abnormalities in the skeletal, cardiovascular, and ocular systems5 and occurs with an incidence of 1 to 2 in 20 000 with an autosomal-dominant inheritance pattern with high penetrance. This obvious genetic pattern of inheritance made MFS an excellent target for study by positional cloning methods. Premature death occurs because of cardiovascular abnormalities involving progressive dilation of the aortic root associated with catastrophic aortic dissection or aortic insufficiency.5 The neonatal form of MFS is associated with a high mortality because of involvement of multiple cardiac valves, resulting in severe congestive heart failure.

The molecular genetic defects causing MFS lie in the fibrillin gene. The first evidence that fibrillin was abnormal in MFS was based on a deficiency in the production of fibrillin from skin fibroblasts of patients with MFS.6 Fibrillin, a 350-kD glycoprotein, is the major component of 10- to 12-nm extracellular microfibrils.7 The primary structure of fibrillin consists of repetitive elements, including 43 repeats of an epidermal growth factor (EGF)–like domain (which bind calcium), interspersed with 7 motifs that are cysteine-rich, transforming growth factor-ß–binding motifs located between the EGF-like domains.8 Genetic linkage analyses of families with MFS established that a strong linkage existed for markers of chromosome 15 and the disease,9 10 and fibrillin became a candidate gene.11 Analysis of mRNA and the corresponding genomic region performed by Dietz and coworkers12 demonstrated a single nucleotide conversion, resulting in a nonconservative amino acid substitution of an arginine for a proline in the fibrillin gene in two unrelated patients with sporadic MFS.

Since the initial exciting discoveries less than 5 years ago, at least 30 mutations in the fibrillin gene have been described in association with MFS.13 14 These include (1) nucleotide changes that result in nonconservative amino acid substitutions in the EGF-like domains and are predicted to alter calcium binding or disulfide bond formation; (2) intronic mutations resulting in abnormal splicing and exon skipping; (3) an exonic mutation with exonic skipping; and (4) deletion of a portion of the gene, resulting in premature termination and a shorter protein. The mechanism by which the abnormal fibrillin produced by these mutant genes causes the molecular phenotype of MFS is not known, although a significant dominant negative effect is suggested.

Correlations of clinical features of MFS and MFS-like syndromes are just beginning as genotypes become available in more families. To date, all families fitting the clinical criteria for MFS with cardiovascular disease have mutations in the fibrillin gene. With rare exception, each family screened has carried a different mutation, suggesting that spontaneous de novo mutations are common. A mutation in this fibrillin-1 gene in a four-generation family with autosomal-dominant ectopia lentis and skeletal abnormalities but without cardiovascular involvement has also been detected.15 Structural characterizations of fibrillin cDNAs provided evidence that more than one human fibrillin gene exists.16 One of the fibrillin genes is on chromosome 5 (termed Fib-5) and is linked to congenital contractural arachnodactyly (CCA).16 This disease shares with MFS the skeletal abnormalities of long extremities. These patients usually have joint contractures and do not have ectopia lentis, and although mitral valve prolapse is associated with CCA, aortic involvement has not been observed. Another MFS-like syndrome, with autosomal-dominant inheritance—mitral valve prolapse, aortic dilation, and skin and skeletal abnormalities (the MASS phenotype)—has been described.17 Analysis of three families with this syndrome has not demonstrated linkage to either of the fibrillin genes on chromosome 5 or 15. A fibrillin-1 mutation has been identified in a single patient with sporadic MASS phenotype. These results suggest that other MFS-like syndromes may be secondary to mutations in other fibrillin genes.

Fibrillin is part of a multiprotein complex of at least five other proteins present in the 10- to 12-nm microfibrils. Microfibril interactions with elastin, a major component of large-vessel vascular wall, are apparent. It seems very likely that mutations in these other microfibrillar proteins may also be a source of phenotypic changes involving the extracellular matrix and connective tissue. The recent discovery that elastin deficiency causes a vasculopathy (see below) emphasizes this point.

The MFS story is a model for investigation of dominantly inherited pediatric heart disease. Ongoing studies to define the mechanisms by which mutant fibrillin proteins alter the vascular wall emphasize the need for understanding pathobiology and molecular genetics of such diseases.

Supravalvular Aortic Stenosis
Recent genetic linkage data18 have demonstrated that alterations in a second major protein of the extracellular matrix, elastin, are the cause of left heart obstruction resulting from discrete narrowing of the ascending aorta above the level of the sinuses of Valsalva. This anatomic entity, supravalvular aortic stenosis (SVAS), occurs in three settings. First, families with an autosomal-dominant inheritance pattern of SVAS without other major phenotypic abnormalities and with high penetrance have been described for many years. Second, SVAS occurs as part of Williams' syndrome in association with stenoses of other major systemic arteries, the large pulmonary arteries, renovascular anomalies, a characteristic facial appearance, unusual behavior patterns, and developmental delay in some areas, but exceptional talent in others. Williams' syndrome is usually sporadic. Third, a sporadic form of SVAS that lacks other major phenotypic abnormalities exists. The anatomy and pathology of the cardiovascular lesion in Williams' syndrome and familial SVAS are similar, suggesting a common cardiovascular pathogenesis.19

The first evidence suggesting that the elastin gene is defective in SVAS was based on linkage analysis of two families with autosomal-dominant SVAS18 and a third family with a translocation of part of chromosome 620 to the elastin locus of chromosome 7 on one allele. The translocation interrupted the elastin gene in exon 28, near the 3' end of the gene and in the COOH-terminal coding region. The abnormal elastin generated by transcription of this elastin allele would be truncated, with the deleted portion containing sequences necessary for desmosine cross-links, an important disulfide bond, and the microfibril-associated glycoprotein binding site. This truncated protein might create a dominant negative effect on elastin fibril formation. Alternatively, the pathogenetic mechanism may simply be secondary to lowered synthesis of elastin, a gene dosage effect. In two other families mapping to this locus, large deletions within the elastin gene have also been documented,21 further supporting the concept that elastin deficiency secondary to monosomy at this locus is the cause of SVAS.

With the knowledge that familial SVAS and Williams' syndrome have similar aortic pathology, Ewart and coworkers22 explored the possibility that monosomy at the elastin locus (7q11.23) might also cause Williams' syndrome. Both by FISH and Southern blot data, these investigators then conclusively demonstrated hemizygosity, the presence of only a single allele at a gene locus, in many individuals with sporadic Williams' syndrome and in the rare families with dominant inheritance of the Williams' phenotype. Loss of heterozygosity with monosomy of this region has been confirmed in some Williams' syndrome patients by other investigators. Analysis of more than 100 Williams' syndrome patients shows deletions at this locus in more than 95% (M.T. Keating et al, personal communication, 1994). Through flanking cosmids and FISH, a large deletion at 7q11.23, encompassing the entire elastin gene and substantial flanking DNA, has been described in these patients.22 The results suggest that Williams' syndrome may represent the effect of not only the elastin locus but loss of contiguous genes in the submicroscopic deletion. The exciting prospect that mutations of nearby genes may cause the psychological, behavioral, and developmental abnormalities is real. These data strongly support the concept that pediatric heart disease may be part of a contiguous-gene syndrome.

Two important results of these studies are apparent. First, the cosmid markers for locus 7q11.23 are now available for FISH as a cytogenetic screening test to allow definitive diagnosis in young infants who may not have developed the obvious clinical features of Williams' syndrome. Second, the discovery that familial SVAS and sporadic Williams' syndrome are secondary to molecular defects at the same locus emphasizes a very important point. Sporadic disease, as the vasculopathy here and as in many instances of CHD, can be secondary to a single-gene defect. This concept, we believe, is relevant to many types of CHD and emphasizes that single-gene defects may not be obviously familial.

Hereditary Hemorrhagic Telangiectasia (Rendu-Osler-Weber Syndrome)
Hereditary telangiectasia is an autosomal-dominant vascular dysplasia. The clinical manifestations include skin, mucosal, and visceral telangiectases with recurrent hemorrhage. Multiple arteriovenous malformations of the gut may cause significant blood loss, and malformations in the lungs are particularly obvious because these fistulas produce cyanosis. Two recent studies have mapped the abnormal gene to human chromosome 9q.23 24 Isolation and characterization of the abnormal gene product will provide insight into genes essential for angiogenesis and normal vascular maturation.


*    Animal Models of CHD and Their Relation to Human Disease
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowVasculopathies: Genetic Defects...
 *Animal Models of CHD...
down arrowHuman CHD Secondary to...
 down arrowReferences
 
The occurrence of congenital heart defects in animals, either spontaneous or induced by experimental manipulation, provides opportunities to overcome inherent limitations encountered in studies of the etiologies and genetic aspects of human disease. Some animal species, especially mice, provide large litter size, controlled mating, complete ascertainment of the phenotype before intrauterine loss, relatively homogenous genetic backgrounds, and controlled breeding environments, which give particular experimental advantages. In addition, recent exciting technical advances in the manipulation of mammalian embryos provide the ability to perform germline manipulations to ablate, or "knock out," genes regulating cardiac morphogenesis or to rescue a defined genetic defect by replacing a defective or mutated gene with a normal transgene. We discuss below some animal experiments that affect our understanding of the molecular genetic basis of CHD, especially of monogenic etiologies.

Situs Inversus
Some of the most complex and severe forms of human CHD are associated with alterations in the normal asymmetrical pattern of thoracoabdominal viscera (situs inversus, situs ambiguous, and heterotaxy). Hummel and Chapman25 initially described mice with spontaneous situs inversus that were then inbred to create the iv/iv strain (situs inversus viscerum). The iv/iv phenotype is a recessive trait determined by a single-gene locus on the distal arm of mouse chromosome 12. Approximately 50% of the offspring of iv/iv mice are affected (situs defect) and 50% are normal, despite the fact that the animals are homozygous. The proposed explanation for 50% normal situs is that the normal iv/iv gene controls the development of normal body patterning and that absence of the normal gene permits random body patterning, including complete situs inversus, situs ambiguous, and situs solitus secondary to this loss of function mutation.26 This is particularly pertinent to human cardiac malformations in families in which the discordant nature of intrafamilial disease is a prominent feature. Formal linkage analysis identified the iv/iv locus on the distal arm of mouse chromosome 12, an area syntenic with human chromosome 17.27 Studies with chimeric embryos28 revealed that the maternal intrauterine environment does not influence outcome and that expression of the iv/iv phenotype is high if the heart and adjacent visceral yolk sac endoderm are derived mostly from iv/iv embryos, supporting the hypothesis that the defective gene in the iv/iv strain is involved with cell-to-cell interaction. As the molecular defect at this mouse locus is defined, the homologous human gene will be identified, and patients with situs inversus and heterotaxy can be screened for potential mutations.

Creation of transgenic mice necessitates insertion of the transgene into normal mouse DNA. If this usually random insertion interrupts the function of a normal gene, the resultant mutant phenotype will provide clues to the normal function of the interrupted gene. Two random insertional mutagenic events in transgenic mice associated with situs inversus have been described. The first ("legless") created limb deformities and situs inversus.29 The insertional site is on the distal part of chromosome 12, suggesting that it involves the iv/iv locus. Crossbreeding studies demonstrated that the legless strain complements the iv/iv strain, consistent with the legless defect being allelic to the iv/iv locus. The second insertional mutation, causing both a deletion and partial duplication on mouse chromosome 4,30 causes situs inversus with recessive inheritance. Thus, at least two loci in mice are responsible for the normal development of body asymmetry, providing conclusive evidence that determination of normal mammalian body and cardiac asymmetry requires different gene products at distinct loci.

In humans, three modes of inheritance of visceral heterotaxy have been noted. A large family pedigree with an X-linked inheritance pattern of heterotaxy and CHD was mapped by linkage studies to band Xq24-27.1 of the X chromosome.31 Most families with recurrent cases of situs inversus have had an autosomal-recessive pattern of inheritance. More recently, some families with inherited heterotaxia fit an autosomal-dominant mode of inheritance. Although most human heterotaxia with complex CHD is sporadic, the animal and human genetic data strongly suggest that these defects are due to mutations, either spontaneous or transmitted, in heterogeneous, but specific, genetic loci required for development of body asymmetry and cardiac morphogenesis.

Canine Conotruncal Malformations
A second animal model of CHD provides evidence that a single-gene defect causes conotruncal abnormalities. tetralogy of Fallot occurs with high frequency in an inbred dog, Keeshond hounds.32 Relatives of these animals have an increased incidence of subarterial ventricular septal defects. Detailed anatomic analysis demonstrated a spectrum of anatomy from subclinical abnormalities in the crista supraventricularis to severe conotruncal defects, including truncus arteriosus. It was initially believed that a threshold model explained this inheritance. In the threshold model, which some have proposed to explain human CHD,2 multiple genes contribute additively to create the conotruncal defect. However, recent breeding experiments with this inbred line demonstrate that the conotruncal defect is due to an inheritance pattern consistent with a single-gene defect.33 Microscopic analysis of offspring of matings of affected animals confirmed that the region of conotruncus has a deficiency of myocardium and conoseptal cushion tissue accompanied by an enlarged lumen. These reproducible abnormalities in myocardial growth lead to the defect, and mild subclinical defects are common in outbred populations. These subclinical defects led to underestimation of the heritability of the abnormality in the original reports. These painstaking animal breeding experiments confirm that a common type of CHD transmitted in a nonmendelian pattern in outbred animal populations is the result of a single-gene defect. The obvious analogy is that the apparent nonmendelian inheritance of human CHD may be similar, particularly because subclinical abnormalities in relatives of children with CHD, such as atrial septal aneurysms or closed ventricular septal defects, may escape detection.

Neural Crest in Heart Development
The association of abnormalities in neural crest derivatives in DiGeorge syndrome34 —thymic aplasia, hypoparathyroidism, craniofacial dysmorphology, and conotruncal or aortic arch defects (see below)—has focused attention on the role of the neural crest in CHD. Experimental manipulation of chicken embryos confirmed the role of the neural crest in heart malformations. In these studies, Kirby and coworkers35 36 37 performed ablations of specific regions of the neural crest. Lesions of the caudal portion of the cranial neural crest extending to the region above somite 5 and branchial arches 3 through 6 resulted in persistent truncus arteriosus, aplasia, or hypoplasia of the glands of the pharyngeal region and anomalous development of the great arteries. Smaller ablations resulted in dextroposition of the aorta (double-outlet right ventricle, tetralogy of Fallot, and overriding of the aorta). Inflow anomalies, including tricuspid atresia, straddling tricuspid valve, and double-inlet left ventricle, occurred with a reduced frequency (3% to 10%) in dextroposition defects. The fact that similar lesions produce various cardiac malformations emphasizes that a single genetic defect might cause a spectrum of CHD.

The role of particular genes in both neural crest and cardiac development is emphasized by two mutant mouse strains, with abnormalities in homeobox genes. The homeobox genes encode DNA-binding proteins that are highly conserved across species (including invertebrates). These proteins exert significant effects as transcriptional regulators of gene expression during cellular differentiation and development, and they specify body pattern formation. Ablation of one homeobox gene (HOX 1.5) by homologous recombination38 in embryonic mouse stem cells and breeding to obtain animals homozygous for loss of this gene caused the mice to develop a phenotype similar to DiGeorge syndrome, including absent thymus, craniofacial anomalies, absent parathyroid, and decreased thyroid tissue. These animals had CHD with bicuspid aortic valve, abnormal aortic arches, and cardiac dilation, but they did not have conotruncal defects. This example provides strong evidence that mutations in mammalian homeobox genes may underlie human CHD associated with segmentation abnormalities, as in aortic arch anomalies.

The second example of neural crest–based abnormalities in mice is the naturally occurring39 40 or radiation-induced mutant strain Splotch, with several mutant alleles on chromosome 1. The phenotype is a characteristic hair pattern in heterozygotes (white patch or splotch on the abdomen), neural tube defects (including spina bifida), defects in truncal neural crest derivatives, abnormal muscle development, and conotruncal malformations (including persistent truncus arteriosus). These mice exhibit a reduction in neural crest elements, abnormal migration of neural crest cells, and alterations in N-CAM, an extracellular matrix protein. Splotch mice have mutations in the PAX-3 gene, which is a member of the paired-box gene family. This gene is expressed in a segmental manner in Drosophila and has a highly conserved structure across species involving homeobox DNA-binding motifs. A mutation within the human homologue for the PAX-3 gene is the basis of Waardenburg syndrome (white hair patch, hearing loss, and aniridia). In mice, the severity of the phenotype correlates with the type of mutation. Large deletions, encompassing the entire PAX-3 gene and adjacent DNA, are associated with severe neural tube defects, truncus arteriosus, and early fetal demise. A PAX-3 point mutation creates a Splotch delayed phenotype, with limb muscular defects, spina bifida, overriding of the aorta, and death soon after birth. These Splotch mutations may be analogous to human contiguous-gene syndromes due to monosomy. That is, large deletions manifest a severe or extended and multifaceted phenotype, as in Williams' syndrome, whereas point mutations or small deletions may present with milder disease with a more restricted phenotype, as in dominant SVAS.

These animal data strongly support the hypothesis that the neural crest is a major determinant of normal cardiac morphogenesis. The discovery of other animal genes that control neural crest formation and other genes induced during neural crest cell migration to the conotruncus will define candidate genes, mutations of which cause CHD.

Mouse Cardiac Transcription Factor Gene Ablation Studies
The ability to ablate or knock out genes by homologous recombination with generation of mice deficient in both alleles of the targeted gene has provided an opportunity to examine mammalian developmental biology in a unique fashion. The results of these gene knockout experiments often provide surprising results. Four such recent ablation studies document severe disruption of cardiac morphogenesis. In the first, the neurofibromin gene was ablated.41 Neurofibromin was originally isolated as the mutated protein in neurofibromatosis, an autosomal-dominant disease characterized by abnormalities in neural crest tissues. Neurofibromin is a tumor-suppressor protein homologous to the inhibitors of ras, a regulator of cell proliferation. Neurofibromin is abundant in embryonic rodent heart and neuronal tissues. The hearts of embryos homozygous for the neurofibromin-1 ablation show hypoplasticity, disoriented myofibrils, a bulboventricular foramen, incomplete development of the atrioventricular valves, and incomplete conotruncal septation. Death from myocardial failure occurs in mid-gestation (day 11 to 12). The obvious and quite unexpected conclusion is that neurofibromin is essential for normal cardiac morphogenesis.

The second murine knockout42 associated with embryonic lethality secondary to cardiac anomalies was in the Wilms' tumor–associated gene (WT-1). WT-1 is a zinc-finger DNA-binding protein and tumor suppressor expressed in the developing metanephros (kidney). Germline mutations of WT-1 cause urogenital abnormalities. Cytogenetically visible deletions at this locus, 11p13, and point mutations of the WT-1 are associated with severe human malformations, but not in the heart. Homozygous WT-1–deleted mouse embryos by day 13.5 manifest severe edema, hemopericardium, and "thinning" of the myocardium. Because WT-1 protein is not detectable in embryonic heart, the mechanism by which this cardiac effect occurs is uncertain.

Retinoic acid and its metabolites are signals that trigger and modulate morphogenesis during vertebrate development (for review, see Reference 43). These retinoids are cardiac teratogens in chick fetuses.44 The mediators of response to retinoids are transcription factors, proteins of the nuclear hormone receptor superfamily, that bind these ligands. One subset of this family, the retinoid X receptors (RXR), binds 9-cis retinoic acid, and the RXR{alpha} subtype is abundantly expressed in fetal tissues, including heart. Mice homozygous for ablation of the RXR{alpha} gene, like those with neurofibromin deficiency, die at day 11 or 12 of cardiac failure secondary to a similar phenotype—myofibrillar disorganization of the trabecular portion of the left ventricle, chamber hypoplasia, and muscular ventricular septal defects.45 These results demonstrate the essential role of RXR{alpha} in cardiac development and morphogenesis.

Komuro and Izumo46 and Lints and colleagues47 recently isolated a mammalian homologue of a Drosophila homeobox gene that is expressed from the onset of cardiomyocyte differentiation and remains expressed at high levels in cardiac tissue. This cardiac-specific homeobox, denoted Csx or Nkx-2.5, is among the earliest transcription factors expressed in developing mouse heart. Lints and coworkers created mice homozygous for ablation of this gene that demonstrate early (day 9.5) embryonic lethality secondary to arrested cardiac development, with a thin-walled, poorly functioning, incompletely septated heart. Defects in a related protein in Drosophila, designated tinman and described by Bodmer,48 produce acardia, although the relationship of the fly circulation to the mammalian cardiovascular system is distant.

Other transcription factors associated with cardiomyocyte differentiation are GATA-449 and myocyte enhancing factor–2.50 Both are expressed early in cardiac embryogenesis and are crucial in activation of cardiac-specific genes. Their role in early cardiac development remains to be defined.

These gene knockout experiments emphasize the crucial role of transcription factors in normal mammalian cardiac morphogenesis. Undoubtedly, other transcription factors and members of DNA-binding protein families will be discovered in the near future and will have similar roles in heart development. It seems likely that mild mutations in some may cause CHD. Isolation of target genes that are regulated by these factors—such as adhesion molecules, extracellular matrix proteins, and growth factors—will provide additional candidate genes, mutations of which may cause CHD.


*    Human CHD Secondary to Known Single-Gene Defects
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowVasculopathies: Genetic Defects...
 up arrowAnimal Models of CHD...
 *Human CHD Secondary to...
down arrowReferences
 
Familial CHD
Numerous reports of multiple cases of CHD in families without detectable chromosomal defects have appeared.51 52 53 54 55 56 The anatomic types of CHD that may occur in families include secundum atrial septal defects55 ; tricuspid valve abnormalities,51 including Ebstein's anomaly; atrioventricular septal defects53 ; pulmonary valve stenosis; bicuspid aortic valve or aortic valve stenosis; and total anomalous pulmonary venous return.54 55 In most families, inheritance is autosomal-dominant, perhaps because this is most easily discerned by history. Although the vast majority of CHD of these anatomic phenotypes is sporadic, these reports in families provide an important lesson, perhaps analogous to the situation described above with SVAS and Williams' syndrome and underscored by the inherited canine conotruncal defects. That is, mutations in a single gene may indeed cause CHD, which is obvious in some cases because of familial inheritance, high penetrance, and severe phenotype but is less apparent in others because of spontaneous new mutations in the gene or because subclinical, inherited abnormalities are not discerned.

Familial Atrial Septal Defect and Holt-Oram Syndrome
Holt-Oram syndrome56 57 is inherited in an autosomal-dominant fashion and causes anomalies of the radial ray of the upper limbs and CHD, usually secundum atrial septal defects or ventricular septal defects and conduction abnormalities. Genetic mapping studies in two large families with severe phenotypes57 revealed linkage to the long arm of chromosome 12, at 12q2. As yet, the specific gene responsible has not been identified (C. Seidman, personal communication, 1994). Mapping of other hand-heart syndromes reveals nonlinkage to this locus, indicating that mutations in different single genes can cause these syndromes.

Mapping of several families with dominantly inherited secundum atrial septal defects without limb anomalies is currently being performed by the Seidman laboratory and our team. Nonlinkage to the Holt-Oram locus on chromosome 12 has been conclusively demonstrated, but the site of linkage remains elusive.

Nonetheless, these results emphasize that different single-gene defects can cause secundum atrial septal defects. We predict that the isolation and characterization of these genes will provide tools allowing study of sporadic atrial septal defects and that spontaneous mutations in these genes will be found.

Atrioventricular Canal (Septal) Defects
Atrioventricular canal (AVC) defects caused by trisomy 21 (Down's syndrome) account for the majority of congenital heart defects associated with chromosomal abnormalities.58 Elucidation of the genetics of Down's syndrome and AVC defects will show a mechanism by which aneuploidy leads to congenital defects. Moreover, this work may identify genes that influence development and fusion of the endocardial cushions and atrioventricular septum, a critical stage of cardiac morphogenesis.

The phenotype of Down's syndrome and the associated AVC defects are likely due to extra copies of genes on chromosome 21. The gene dosage theory postulates that genes on chromosome 21 in the aneuploid cell are expressed at a level 1.5 times greater than diploid cells.59 Overexpression of several genes that map to the Down's syndrome region of chromosome 21 has been demonstrated.60 61 This gene dosage theory is central to a computer simulation model in which the increased adhesiveness of cells from fetuses with trisomy 21 predicts deficiencies of cushion-to-cushion and cushion-to-septum fusion.62

The hypothesis that nonspecific effects of aneuploidy account for the congenital defects of Down's syndrome through amplified developmental instability63 is refuted by recent evidence from genetic mapping studies.64 65 66 67 Molecular studies of families with partial trisomy for chromosome 21 have allowed mapping of specific phenotypic features of Down's syndrome to distinct regions of chromosome 21. Although duplication of regions throughout chromosome 21 is associated with mental retardation, mapping studies from partial trisomy 21 families combined with the trisomy 16 (Ts16) mouse model have narrowed the critical region for endocardial cushion defects to a region of 9 Mb in length at 21q22.2-21q22.3.66 67 The Ts16 mouse model is based on a large conserved region between human chromosome 21 and mouse chromosome 16 and malformations in the Ts16 mouse similar to Down's syndrome, including AVC defects.59 68 Although the genes responsible for AVC defects in Down's syndrome have not been identified, genes that map to this region include the ETS-2 oncogene, the ETS-2–related gene, and collagen type VI, {alpha}-1 and {alpha}-2 genes.67 The collagen type VI genes coded for on chromosome 21 are coordinately regulated and expressed in the human fetal heart69 ; recently discovered polymorphisms in this cluster will assist in the analysis of the possible role of these genes in the development of AVC defects.70 The critical genes may regulate the expression of the transforming growth factor–ß proteins. Studies in the mouse and chicken implicate these proteins in the transformation of endothelial cells to mesenchymal cells, a critical step in the morphogenesis of the endocardial cushions.71 72

AVC occurs in association with other complex cardiac malformations in the heterotaxia syndromes. As noted above, mutations causing these defects probably lie in genes responsible for directing body asymmetry.

The occurrence of isolated AVC defects in individuals without other features of Down's syndrome and the association of AVC with other defects are consistent with genetic heterogeneity. Linkage analysis of several pedigrees with autosomal-dominant transmission of isolated AVC defects excludes the cardiac region of chromosome 21 as the cause of heart defects in these families.53 73 Moreover, the association of AVC defects with partial deletions on the short arm of chromosome 8 suggests the presence of another AVC critical region.74 These data suggest that AVC defects can be monogenetic in origin and indicate that genes distinct from the Down's syndrome critical region are involved in development of the endocardial cushions and atrioventricular septum.

Conotruncal Cardiac Defects and CATCH-22 Syndrome
The most dramatic and clinically important evidence for genetic etiologies of CHD is the recent discovery of the association of conotruncal cardiac defects with monosomy of a locus on chromosome 22 (Table 3Down). This association also provides another very important lesson for clinicians because it proves that a diverse group of clinical cardiac phenotypes have a common genetic etiology. In 1965, DiGeorge34 explained absence of parathyroids and thymus on the basis of common embryologic origin in a group of infants with immune defects. Later recognition of association of conotruncal cardiac defects or head and facial anomalies with absence of the thymus and parathyroids led to the concept of a developmental field defect involving the third and fourth pharyngeal pouches.75 76 Clinical studies of CHD76 in partial DiGeorge syndrome noted that aortic arch anomalies, including interrupted arch and right aortic arch, were also commonly associated with this developmental field defect. In DiGeorge syndrome, the typical CHD phenotypes76 are truncus arteriosus (25% to 30%) and interrupted aortic arch (25% to 40%), although tetralogy of Fallot (20%) and isolated ventricular septal defect (10%) also occur. Many patients have a right aortic arch (25%). Some (5% to 10%) have no CHD. More recently, patients with velo-cardio-facial syndrome (VCFS, Shprintzen syndrome, conotruncal anomaly face syndrome) were noted to share some manifestations with DiGeorge syndrome.77 78 CHD in VCFS79 80 is most commonly a ventricular septal defect (54%) or tetralogy of Fallot (20%), frequently with a right aortic arch. However, some VCFS patients do not have clinical CHD (14%) or have only a right aortic arch or mild anomalies of the origins of the arch vessels (11%). Thus, the cardiac manifestations of these two syndromes are quite different. In concert with these clinical observations, the laboratory studies discussed above concerning the association of neural crest lesions and these same cardiac defects emphasize the potential convergence of these clinical syndromes with neural crest anomalies.


View this table:
[in this window]
[in a new window]
  		Table 3. Cardiac Manifestations of the CATCH-22 Syndrome

The first clues pointing to the abnormal genetic locus in these patients began with study of patients with unbalanced translocations and partial monosomy of chromosome 2281 and with the observation that small deletions of chromosome 22q were present in about 10% of patients with DiGeorge syndrome. These results strongly supported the concept that monosomy of this region could cause CHD. For these reasons, Carey and coworkers82 in England and Emanuel and coworkers83 84 in Philadelphia used markers from this region to map the size of the deletion and to begin physical mapping of this locus. With this effort, numerous probes have been isolated and used in subsequent molecular analysis of patients with DiGeorge syndrome and VCFS. Initial gene dosage studies and more recent FISH analysis demonstrated, to the surprise of many, that a majority of these syndromic patients have deletions in chromosome 22q11. Current data suggest that about 80% to 95% of individuals with either syndrome have microdeletions. The association of unusual facial features, described as the conotruncal anomaly face,85 and hypernasal speech with pulmonary atresia and ventricular septal defect was noted in the past. Evaluation of patients with this constellation of findings demonstrates monosomy at this locus in most patients. These amazing results led several groups to analyze patients with the same cardiac lesions, ie, isolated tetralogy of Fallot, truncus arteriosus, or interrupted arch, but without any features of either DiGeorge syndrome or VCFS for this same deletion. With two different probes from this region, deletions were observed in 20% to 30% of such unselected, nonsyndromic patients.86 We have recently examined all our living patients with absent pulmonary valve syndrome or anomalous origin of the right pulmonary artery from the aorta (sometimes called hemitruncus). Although only three of these patients had facial anomalies, seven of nine individuals were monosomic at this locus (unpublished data). These results indicate that monosomy at this locus is very common in CHD patients, whether or not an obvious syndrome is noted.

The acronym CATCH-22 (cardiac defects, abnormal facies, thymic hypoplasia, cleft palate, and hypocalcemia from deletions in chromosome 22) was coined to encompass the spectrum of clinical manifestations in patients with deletions in this locus.87 Associated clinical features of CATCH-22 may include growth delay, learning disabilities, nasal speech, bronchospasm, psychiatric disorders, ocular disorders, and renal abnormalities.79 87 Parental deletions are found in approximately 25% of patients with CATCH-22, although variable expression within a family is well documented.87 The mechanism for de novo deletions may involve low-copy-number repetitive sequences that flank the CATCH-22 locus.88 The estimated minimal size of the deleted region is 0.5 million bp83 ; however, mapping studies with multiple probes suggest that the size of the deletion does not correlate with the phenotype.87 Together with the highly variable phenotypes observed, these results suggest that CATCH-22 may represent a contiguous-gene syndrome. As with Williams' syndrome and the elastin locus, the exciting prospect exists that genes responsible for the behavioral and psychiatric abnormalities in these patients may be isolated. In addition, identification and characterization of the genes deleted in these patients will help explain the embryological mechanism of conotruncal cardiac defects. Candidate genes being investigated include a zinc-finger gene, which codes for a putative transcription factor,89 and the TUPLE-1 gene, which codes for a protein that has homology with the yeast Tup1 transcriptional regulator.90 However, because of the large size of deletions, many genes are undoubtedly deleted. Genes on other chromosomes also must cause conotruncal defects because some patients with DiGeorge syndrome have chromosomal abnormalities at other sites, especially chromosome 10.91 Investigation of the roles of candidate genes will require ablation studies in mice. In the relatively near future, however, it is certain that the gene or genes required for normal human aortic arch and conotruncal development will be located.

The potential impact of the discovery of the CATCH-22 syndrome for the clinical practice of pediatric cardiology is tremendous. Although insufficient data are available, it is clear that among patients with supracristal VSDs, tetralogy of Fallot, pulmonary atresia with ventricular septal defect, absent pulmonary valve syndrome, interrupted aortic arch, truncus arteriosus, isolated right aortic arch, and rare arch anomalies (eg, aortic origin of the right pulmonary artery), as many as 20% to 90% may be monosomic at this locus. This suggests that among patients requiring surgery for CHD, 5% to 10% may have this single genetic abnormality. Because FISH probes for this locus are readily available, we believe that all patients with these diagnoses should be tested for the CATCH-22 genetic defect. We also urge that the designation CATCH-22 be universally adopted to designate monosomy at this locus, no matter what the syndromic phenotype of any particular patient. This will reduce confusion based on historical clinical descriptions of syndromes.

Conclusions
This review has amalgamated some of the vast recent laboratory and clinical investigations providing evidence that human CHD has a genetic origin that is rapidly becoming understandable in molecular terms. The results cited provide several important lessons for both the clinician caring for patients with CHD and the basic scientist investigating mechanisms of cardiac development and morphogenesis.

First, human CHD is frequently due to single-gene defects. Even sporadic defects may arise from a single-gene abnormality, as with Williams' syndrome and SVAS.

Second, clinically useful classifications of CHD must not interfere with critical thinking concerning similar genetic and pathophysiological causative mechanisms. As the CATCH-22 syndrome suggests, conotruncal malformations and aortic arch anomalies may share a common genetic defect.

Third, elucidation of the genetic basis of CHD will provide clues to normal cardiovascular developmental biology. As causative or candidate genes are isolated, ablation studies in mice will be essential in defining interactions with other genes to understand the mechanisms by which CHD becomes manifest. Isolation of contiguous genes responsible for the noncardiac manifestations, including behavioral disorders, will be an exciting by-product.

Fourth, mutations at a single locus cause multiple different cardiac phenotypes, as illustrated by the CATCH-22 and elastin loci. Although some of this variability may reflect abnormalities in contiguous genes, it is also likely that interactions with the embryonic environment and with other genes may explain this variability.

Fifth, the same cardiac malformation is caused by specific gene defects at different loci, indicating that several different, but single, gene defects may cause the same apparent phenotype. Examples include AVC, tetralogy of Fallot, and atrial septal defects.

Finally, the revolution in molecular genetics has clearly hit mammalian cardiovascular development. It is time for clinicians, including pediatric cardiologists, geneticists, cardiovascular surgeons, and adult cardiologists, to interact with basic scientists in the search for genes causing CHD and other cardiovascular disease.

Received October 19, 1994; accepted October 24, 1994.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowVasculopathies: Genetic Defects...
 up arrowAnimal Models of CHD...
 up arrowHuman CHD Secondary to...
 *References
 

  1. Moller JH, Allen HD, Clark EB, Dajani AS, Golden A, Hayman LL, Lauer RM, Marmer EL, McAnulty JH, Oparil S, Strauss AW, Taubert KA, Wagner A. Report of the task force on children and youth. Circulation. 1992;88:2479-2486. [Free Full Text]
  2. Nora JJ. Causes of CHD: old and new modes, mechanisms, and models. Am Heart J. 1993;125:1409-1418. [Medline] [Order article via Infotrieve]
  3. Hall JG. CATCH 22. J Med Genet. 1993;30:801-802. [Medline] [Order article via Infotrieve]
  4. Kelly DP, Strauss AW. Inherited cardiomyopathy: the spectrum of genetic etiologies. N Engl J Med. 1994;330:913-919. [Free Full Text]
  5. Pyeritz RE, McKusick VA. The Marfan syndrome: diagnosis and management. N Engl J Med. 1979;300:772-777. [Medline] [Order article via Infotrieve]
  6. Hollister DW, Godfrey M, Sakai LY, Pyeritz RE. Immunohistologic abnormalities of the microfibrillar-fiber system in the Marfan syndrome. N Engl J Med. 1990;323:152-159. [Abstract]
  7. Sakai LY, Keene DR, Engvall E. Fibrillin: a new 350 kD glycoprotein is a component of extracellular microfibrils. J Cell Biol. 1986;103:2499-2509. [Abstract/Free Full Text]
  8. Maslen CL, Corson GM, Maddox BK, Glanville RW, Sakai LY. Partial sequence of a candidate gene for the Marfan syndrome. Nature. 1991;325:334-337.
  9. Lee B, Godfrey M, Vitale E, Hori H, Mattei M-G, Sarfarazi M, Tsipouras P, Ramirez F, Hollister DW. Linkage of Marfan syndrome and a phenotypically related disorder to two different fibrillin genes. Nature. 1991;352:330-334. [Medline] [Order article via Infotrieve]
  10. Kainulainen K, Pulkkinen L, Savolainen A, Kaitila I, Peltonen L. Location on chromosome 15 of the gene defect causing Marfan syndrome. N Engl J Med. 1990;323:935-939. [Abstract]
  11. Dietz HC, Pyeritz RE, Hall BD, Cadle RG, Hamosh A, Schwartz J, Meyers DA, Francomano CA. The Marfan syndrome locus. Genomics. 1991;9:355-361. [Medline] [Order article via Infotrieve]
  12. Dietz HC, Cutting GR, Pyeritz RE, Maslen CL, Sakai LY, Corson GM, Puffenberger EG, Hamosh A, Nanthakumar EJ, Curristin SM, Stetten G, Meyers DA, Francomano CA. Marfan syndrome caused by a recurrent de novo missense mutation in the fibrillin gene. Nature. 1991;352:337-339. [Medline] [Order article via Infotrieve]
  13. Dietz HC, Valle D, Francomano CA, Kendzior RJ, Pyeritz RE, Cutting GR. The skipping of constitutive exons in vivo induced by nonsense mutations. Science. 1993;259:680-683. [Abstract]
  14. Dietz HC, McIntosh I, Sakai LY, Corson GM, Chalberg SC, Pyeritz RE, Francomano CA. Four novel FBN1 mutations: significance for mutant transcript level and EGF-like domain calcium binding in the pathogenesis of Marfan syndrome. Genomics. 1993;17:468-475. [Medline] [Order article via Infotrieve]
  15. Kainulainen K, Karttunen L, Puhakka L, Sakai L, Peltonen L. Mutations in the fibrillin gene responsible for dominant ectopia lentis and neonatal Marfan syndrome. Nat Genet. 1994;6:64-69. [Medline] [Order article via Infotrieve]
  16. Tsipouras P, Del Mastro R, Sarfarazi M, Lee B, Vitale E, Child AH, Godfrey M, Devereux RB, Hewett D, Steinmann B, Viljoen D, Sykes BC, Kilpatrick M, Ramirez F. Genetic linkage of the Marfan syndrome, ectopia lentis and congenital contractural arachnodactyly to the fibrillin genes on chromosomes 15 and 5. N Engl J Med. 1992;326:905-909. [Abstract]
  17. Boileau C, Jondeau G, Babron M-C, Coulon M, Alexandre J-A, Sakai L, Melki J, Delorme G, Dubourg O, Bonaiti-Pellie C, Bourdarias JP, Junien C. Autosomal dominant Marfan-like connective-tissue disorder with aortic dilation and skeletal anomalies not linked to the fibrillin genes. Am J Hum Genet. 1993;53:46-54. [Medline] [Order article via Infotrieve]
  18. Ewart AK, Morris CA, Ensing GJ, Loker J, Moore C, Leppert M, Keating M. A human vascular disorder, supravalvular aortic stenosis, maps to chromosome 7. Proc Natl Acad Sci U S A. 1993;90:3226-3230. [Abstract/Free Full Text]
  19. O'Connor WN, Davis JB, Geissler R, Cottrill CM, Noonan JA, Todd EP. Supravalvular aortic stenosis. Arch Pathol Lab Med. 1985;109:179-185. [Medline] [Order article via Infotrieve]
  20. Curran ME, Atkinson DL, Ewart AK, Morris CA, Leppert MF, Keating MT. The elastin gene is disrupted by a translocation associated with supravalvular aortic stenosis. Cell. 1993;73:159-168. [Medline] [Order article via Infotrieve]
  21. Ewart AK, Jin W, Atkinson D, Morris CA, Keating MT. SVAS associated with a deletion disrupting the elastin gene. J Clin Invest. 1994;93:1071-1077.
  22. Ewart AK, Morris CA, Atkinson D, Jin W, Sternes K, Spallone P, Stock AD, Leppert M, Keating MT. Hemizygosity at the elastin locus in a developmental disorder, Williams syndrome. Nat Genet. 1993;5:11-16. [Medline] [Order article via Infotrieve]
  23. Shovlin CL, Hughes JMB, Tuddenham EGD, Temperley I, Perembelon YFN, Scott J, Seidman CE, Seidman JG. A gene for hereditary hemorrhagic telangiectasia maps to chromosome 9q3. Nat Genet. 1994;6:205-209. [Medline] [Order article via Infotrieve]
  24. McDonald MT, Papenberg KA, Ghosh S, Glatfelter AA, Biesecker BB, Helmbold EA, Markel DS, Zolotor A, McKinnon WC, Vanderstoep JL, Jackson CE, Iannuzzo M, Collins FS, Boehnke M, Porteious ME, Guttmacker AE, Marchuk DA. A disease locus for hereditary hemorrhagic telangiectasia maps to chromosome 9q33-34. Nat Genet. 1994;6:197-204. [Medline] [Order article via Infotrieve]
  25. Hummel KP, Chapman DB. Visceral inversion and associated anomalies in the mouse. J Hered. 1959;50:9-13. [Free Full Text]
  26. Seo J-W, Brown NA, Ho SY, Anderson RH. Abnormal laterality and congenital cardiac anomalies: relations of visceral and cardiac morphologies in the iv/iv mouse. Circulation. 1992;86:642-650. [Abstract/Free Full Text]
  27. Brueckner M, D'Dustachio P, Horwich AL. Linkage mapping of a mouse gene, iv, that controls left-right asymmetry of the heart and viscera. Proc Natl Acad Sci U S A. 1989;86:5035-5038. [Abstract/Free Full Text]
  28. Brown NA, McCarthy A, Wolpert L. The development of handed asymmetry in aggregation chimeras of situs inversus mutant and wild-type mouse embryo. Development. 1990;110:949-954. [Abstract/Free Full Text]
  29. Singh G, Supp DM, Schreiner C, McNeish J, Merker H-J, Copeland NG, Jenkins NA, Potter SS, Scott W. Legless insertional mutation: morphological, molecular, and genetic characterization. Genes Dev. 1991;5:2245-2255. [Abstract/Free Full Text]
  30. Yokoyama T, Copeland NG, Jenkins NA, Montgomery CA, Elder FFB, Overbeek PA. Reversal of left-right asymmetry: a situs inversus mutation. Science. 1993;260:679-682. [Abstract/Free Full Text]
  31. Casey B, Devoto M, Jones KL, Ballabio A. Mapping a gene for familial situs abnormalities to human chromosome Xq24-q27.1. Nat Genet. 1993;5:403-407. [Medline] [Order article via Infotrieve]
  32. Patterson DF, Pyle RL, Van Mierop L, Melbin J, Olson M. Hereditary defects of the conotruncal septum in keeshond dogs: pathologic and genetic studies. Am J Cardiol. 1974;34:187-205. [Medline] [Order article via Infotrieve]
  33. Patterson DF, Pexieder T, Schnarr WR, Navratil T, Alaili R. A single major-gene defect underlying cardiac conotruncal malformations interferes with myocardial growth during embryonic development: studies in the CTD line of keeshond dogs. Am J Hum Genet. 1993;52:388-397. [Medline] [Order article via Infotrieve]
  34. DiGeorge AM. Discussions on a new concept of the cellular basis of immunology. J Pediatr. 1965;67:907-908.
  35. Kirby ML, Waldo KL. Role of neural crest in CHD. Circulation. 1990;82:332-340. [Free Full Text]
  36. Nishibatake M, Kirby ML, van Mierop LH. Pathogenesis of persistent truncus arteriosus and dextroposed aorta in the chick embryo after neural crest ablation. Circulation. 1987;75:255-264. [Abstract/Free Full Text]
  37. Leatherbury L, Gauldin HE, Waldo K, Kirby ML. Microcinephotography of the developing heart in neural crest-ablated chick embryos. Circulation. 1990;81:1047-1057.[Abstract/Free Full Text]
  38. Chisaka O, Capecchi MR. Regionally restricted developmental defects resulting from targeted disruption of the mouse homeobox gene hox-1.5. Nature. 1991;350:473-483. [Medline] [Order article via Infotrieve]
  39. Franz T. Persistent truncus arteriosus in the Splotch mutant mouse. Anat Embryol (Berl). 1989;180:457-464. [Medline] [Order article via Infotrieve]
  40. Epstein DJ, Vogan KJ, Trasler DG, Gros P. A mutation within intron 3 of the Pax-3 gene produces aberrantly spliced mRNA transcripts in the Splotch mouse mutant. Proc Natl Acad Sci U S A. 1993;90:532-536. [Abstract/Free Full Text]
  41. Brannon CI, Perkins AS, Vogel KS, Ratner NA, Nordlund ML, Reid SW, Buchberg AM, Jenkins NA, Parada LF, Copeland NG. Targeted disruption of the neurofibromatosis type-1 gene leads to developmental abnormalities in heart and various neural crest-derived tissues. Genes Dev. 1994;8:1019-1029. [Abstract/Free Full Text]
  42. Kreidberg JA, Sariola H, Loring JM, Maeda M, Pelletier J, Housman D, Jaenisch R. WT-1 is required for early kidney development. Cell. 1993;74:679-691. [Medline] [Order article via Infotrieve]
  43. Leid M, Kastner P, Chambon P. Multiplicity generates diversity in the retinoic acid signalling pathways. Trends Biochem Sci. 1992;17:427-433. [Medline] [Order article via Infotrieve]
  44. Broekhuizen MLA, Wladimiroff JW, Tibboel D, Poelmann RE, Wenicnk ACG, Gittenberger-de Groot AC. Induction of cardiac anomalies with all-trans retinoic acid in the chick embryo. Cardiol Young. 1992;2:311-317.
  45. Sucov HM, Dyson E, Gumeringer CL, Price J, Chien KR, Evans RM. RXR-{alpha} mutant mice establish a genetic basis for vitamin A signaling in heart morphogenesis. Genes Dev. 1994;8:1007-1018. [Abstract/Free Full Text]
  46. Komuro I, Izumo S. Csx: a murine homeobox-containing gene specifically expressed in the developing heart. Proc Natl Acad Sci U S A. 1993;90:8145-8149. [Abstract/Free Full Text]
  47. Lints TJ, Parsons LM, Hartley L, Lyons I, Harvey RP. Nkx-2.5: a novel murine homeobox gene expressed in early heart progenitor cells and their myogenic descendants. Development. 1993; 119:419-431.
  48. Bodmer R. The gene tinman is required for specification of the heart and visceral muscles in Drosophila. Development. 1993;118:719-729. [Abstract]
  49. Arceci RJ, King AA, Simon MC, Orkin SH, Wilson DB. Mouse GATA-4: a retinoic acid-inducible GATA-binding transcription factor expressed in endodermally derived tissues and heart. Mol Cell Biol. 1993;13:2235-2246. [Abstract/Free Full Text]
  50. Edmondson DG, Lyons GE, Martin JF, Olsen EN. Mef2 gene expression marks the cardiac and skeletal muscle lineages during mouse embryogenesis. Development. In press.
  51. Lo KS, Loventhal JP, Walton JA. Familial Ebstein's anomaly. Cardiology. 1979;64:246-255. [Medline] [Order article via Infotrieve]
  52. Whittemore R, Wells JA, Castellsague X. A second-generation study of 427 probands with CHD and their 837 children. J Am Coll Cardiol. 1994;23:1459-1467. [Abstract]
  53. Cousineau AJ, Lauer RM, Pierpont ME, Burns TL, Ardinger RH, Patil SR, Sheffield VC. Linkage analysis of autosomal dominant atrioventricular canal defects: exclusion of chromosome 21. Hum Genet. 1994;93:103-108. [Medline] [Order article via Infotrieve]
  54. Raisher BD, Dowton SB, Grant JW. Father and two children with total anomalous pulmonary venous return. Am J Hum Genet. 1991;40:105-106.
  55. Bleyl S, Ruttenberg HD, Carey JC, Ward K. Familial total anomalous pulmonary venous return. Am J Hum Genet. 1994; 52:462-466.
  56. Holt M, Oram S. Familial heart disease with skeletal malformations. Br Heart J. 1960;22:236-242.
  57. Basson CT, Cowley GS, Solomon SD, Weissman B, Poznanski AK, Traill TA, Seidman JG, Seidman CE. The clinical and genetic spectrum of the Holt-Oram syndrome (Hand-Heart syndrome). N Engl J Med. 1994;330:885-891. [Abstract/Free Full Text]
  58. Kramer HH, Majewski F, Trampisch HJ, Rammos S, Bourgeois M. Malformation patterns in children with congenital heart disease. Am J Dis Child. 1987;141:789-795. [Abstract]
  59. Holtzman DM, Epstein CJ. The molecular genetics of Down syndrome. Mol Genet Med. 1992;2:105-120. [Medline] [Order article via Infotrieve]
  60. Pash J, Popescu N, Matocha M, Rapoport S, Bustin M. Chromosomal protein HMG-14 gene maps to the Down syndrome region of human chromosome 21 and is overexpressed in mouse trisomy 16. Proc Natl Acad Sci U S A. 1990;87:3836-3840. [Abstract/Free Full Text]
  61. Minc-Golomb D, Knobler H, Groner Y. Gene dosage of CuZnSOD and Down's syndrome: diminished prostaglandin synthesis in human trisomy 21, transfected cells and transgenic mice. EMBO J. 1991;10:2119-2124. [Medline] [Order article via Infotrieve]
  62. Kurnit DM, Aldridge JF, Matsuoka R, Matthysse S. Increased adhesiveness of trisomy 21 cells and atrioventricular canal malformations in Down syndrome: a stochastic model. Am J Med Genet. 1985;20:385-399. [Medline] [Order article via Infotrieve]
  63. Shapiro BL. Amplified developmental instability in Down's syndrome. Ann Hum Genet. 1975;38:429-437. [Medline] [Order article via Infotrieve]
  64. Epstein CJ. Specificity versus nonspecificity in the pathogenesis of aneuploid phenotypes. Am J Med Genet. 1988;29:161-165. [Medline] [Order article via Infotrieve]
  65. Rahmani Z, Blouin JL, Creau-Goldberg N, Watkins PC, Mattei JF, Poissonnier M, Prieur M, Chettouh Z, Nicole A, Aurias A, Sinet PM, Delabar JM. Critical role of the D21S55 region on chromosome 21 in the pathogenesis of Down syndrome. Proc Natl Acad Sci U S A. 1989;86:5958-5962. [Abstract/Free Full Text]
  66. Korenberg JR, Kawashima H, Pulst SM, Ikeuchi T, Ogasawara N, Yamamoto K, Schonberg SA, West R, Allen L, Magenis E, Ikawa K, Taniguchi N, Epstein CJ. Molecular definition of a region of chromosome 21 that causes features of the Down syndrome phenotype. Am J Hum Genet. 1990;47:236-256. [Medline] [Order article via Infotrieve]
  67. Korenberg JR, Pulst SM, Gerwehr S. Advances in the understanding of chromosome 21 and Down syndrome. In: Lott IT, McCoy EE, eds. Down Syndrome: Advances in Medical Care. New York, NY: Wiley-Liss, Inc; 1992:3-12.
  68. Miyabara S, Gropp A, Winking H. Trisomy 16 in the mouse fetus associated with generalized edema and cardiovascular and urinary tract anomalies. Teratology. 1982;25:369-380. [Medline] [Order article via Infotrieve]
  69. Duff K, Williamson R, Richards SJ. Expression of genes encoding two chains of the collagen type VI molecule during human fetal heart development. Int J Cardiol. 1990;27:128-129. [Medline] [Order article via Infotrieve]
  70. Davies GE, Howard CM, Gorman LM, Farrer MJ, Holland AJ, Williamson R, Kessling AM. Polymorphisms and linkage disequilibrium in the COL6A1 and COL6A2 gene cluster: novel DNA polymorphisms in the region of a candidate gene for congenital heart defects in Down's syndrome. Hum Genet. 1993;90:521-525. [Medline] [Order article via Infotrieve]
  71. Potts JD, Vincent EB, Runyan RB, Weeks DL. Sense and antisense TGFß3 mRNA levels correlate with cardiac valve induction. Dev Dyn. 1992;193:340-345. [Medline] [Order article via Infotrieve]
  72. Dickson MC, Slager HG, Duffie E, Mummery CL, Akhurst RJ. RNA and protein localizations of TGFß2 in the early mouse embryo suggest an involvement in cardiac development. Development. 1993;117:625-639. [Abstract]
  73. Wilson L, Curtis A, Korenberg JR, Schipper RD, Allan L, Chenevis-Trench G, Stephenson A, Goodship J, Burn J. A large, dominant pedigree of atrioventricular septal defect (AVSD): exclusion from the Down syndrome critical region on chromosome 21. Am J Hum Genet. 1993;53:1262-1268. [Medline] [Order article via Infotrieve]
  74. Marino B, Reale A, Giannotti A, Digilio MC, Dallapiccola B. Nonrandom association of atrioventricular canal and del (8p) syndrome. Am J Med Genet. 1992;42:424-427. [Medline] [Order article via Infotrieve]
  75. Freedom RM, Rosen FS, Nadas AS. Congenital cardiovascular disease and anomalies of the third and fourth pharyngeal pouch. Circulation. 1972;46:165-172.