CHAPTER 55: MENDELIAN BASIS OF CONGENITAL AND OTHER CARDIOVASCULAR DISEASES

Genetic factors play a significant role in almost all cardiovascular disorders. Genetic defects are responsible for malformations of the heart and blood vessels, which account for the largest number of human birth defects. The estimated incidence of congenital heart disease is approximately 1% of all live births.¹ The prevalence is estimated to be 10-fold higher among stillbirths.² Genetic defects are responsible for familial cardiovascular disorders, such as cardiomyopathies, the long QT syndromes (LQTSs), as well as sporadic forms of such disorders. Over 3600 genes have been identified to date for single-gene disorders with Mendelian patterns of inheritance. Likewise, genetic factors predispose to common complex phenotypes, such as atherosclerosis and hypertension. Over 4000 loci have been mapped for complex phenotypes. Molecular genetic studies provide the opportunity to decipher the genetic basis and pathogenesis of many disorders, especially cardiovascular diseases. Genetic discoveries have the potential to lead to identification of new targets for the prevention and treatment of human diseases. This is best illustrated in the case of the PCSK9 gene, which was genetically mapped in 2003 as a cause of autosomal dominant hypercholesterolemia.³ Recently, the discovery led to subsequent development of proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors (evolocumab and alirocumab), which are highly effective for treatment of certain forms of hypercholerolemia.^4,5 Partial deciphering of the genetic basis of cardiovascular disease has also ushered in routine genetic testing for various diseases, leading to an accurate genetic-based diagnosis and intervention. Given the rapid pace of genetic discoveries, it is expected that genetic diagnosis and genetic-based interventions will become incorporated into standard practice of medicine in the future. Thus, this chapter provides an overview of the genetic principles, and discusses the genetic basis for cardiovascular disorders as well as the medical and ethical implications of genetic findings.

The Human Genome

A genome is the complete set of the genetic material of each cell or individual: the nuclear and the mitochondrial genomes. The nuclear genome is composed of 3.2 billion base pairs. There are four types of nucleotides in each genome, two of which are purine (adenine, guanine) and two are pyrimidine (cytosine, thymine). The nuclear genome is distributed among 22 pairs of double-stranded autosomal chromosomes and two sex chromosomes (XX in females and XY in males). The sizes of chromosomes vary, with chromosome 1 being the largest at ~250 million base pairs and chromosome 21 the smallest at 48 million base pairs. The nuclear genome contains approximately 20,000 protein-coding genes.⁶ A gene has a 5′ transcriptional regulatory region, exons (which contain the protein coding sequence), introns (which are the intervening regions between exons), and the 3′ untranslated region (Fig. 55–1). The nuclear genome also contains several thousands of non–protein-coding genes that get transcribed into noncoding ribonucleic acids (RNAs) including microRNAs and long-noncoding RNAs. About 1% of the genome, which is approximately 300 million base pairs, code for the proteins. More than 50% of the human nuclear genome is composed of repetitive deoxyribonucleic acid (DNA) elements. The functions of the repetitive elements and noncoding regions of the nuclear genome are largely unknown.

Transcription of the nuclear genome is pervasive, as the vast majority of the nuclear genome is transcribed into RNA.⁷ The primary transcripts of genes are spliced to exclude introns and produce the messenger RNAs (mRNAs). Splicing of each primary transcript is not uniform and often multiple splice variants are generated with one being the predominant isoform. Each unit of three bases, referred to as a codon, encodes a specific amino acid. There are 61 amino acid codons and 3 stop codons in the nuclear genome. Thus, each amino acid has multiple codons.

The second genome in each cell is the mitochondrial DNA (MtDNA). Mitochondrial DNA is a circular DNA, composed of 16,700 nucleotides. It contains 37 mitochondrial genes, which code for 13 protein-coding genes, two ribosomal RNAs, and 22 transfer RNAs. Each cell typically contains a large number of mitochondria and each mitochondrion might contain several copies of its genome. The codons are slightly different between nuclear and mitochondrial genomes, as there are 60 codons for amino acid and four stop codons in mtDNA. Likewise, transcription of mtDNA is a contiguous as opposed to discontinuous transcription of the nuclear genome.

Basis for Genetic Transmission

All hereditary information is transmitted through nuclear DNA. The gene is the basic hereditary unit. Each individual has two copies of each gene, called alleles. The genes are localized in a linear sequence along 23 pairs of chromosomes, comprised of 22 pairs of autosomes (chromosomes 1 to 22) and two sex chromosomes, X and Y. Females have two X chromosomes, whereas males carry one X and one Y chromosome. Each parent contributes one of each chromosome pair (the members of the pair are referred to as homologous chromosomes) and thus one copy of each gene. The site at which a gene is located on a particular chromosome is referred to as the genetic locus. A given gene always resides at the same genetic locus on a particular chromosome. Therefore, the loci on homologous chromosomes are identical. However, alleles residing at these loci may be identical or different, leading to homozygous (identical alleles) and heterozygous (two different alleles present at the locus) states.

The genetic information is encoded by the linear sequence of the four bases of the DNA. Translation of this genetic information into a protein is through a “translational code” passed on through mRNA. The transcribed mRNA serves as the template that determines sequence of the amino acids in the resulting polypeptide. Both autosomal chromosomes are usually transcribed into mRNA and translated into protein. However, expression of a gene can be restricted to specific cells and organs. The gene is regulated during the developmental stage because of regulation by cell- and tissue-specific transcription factors and epigenetic factors. In cells that carry two X chromosomes, only one X is active and the other X is silent after early embryogenesis.

Classification of Genetic Disorders

In general, DNA nucleotide sequences remain stable during transmission to offspring. Nonetheless, because of the rare error rate of DNA replication machinery, occasional base sequence changes do occur. These are referred to as sequence variants or polymorphisms or mutations. Mutations represent stable, heritable alterations in DNA. Mutations may occur in somatic and the germ cells. Somatic mutations, however, are not inheritable. A number of mutagenic factors—such as environmental agents (radiation and chemicals)—can induce mutations. Mutations can involve a visible alteration at the level of the chromosome (chromosomal abnormalities). This can result in the deletion or translocation of a portion of the chromosome, whereby several genes are often eliminated or altered. In contrast, mutations can be restricted to minor alterations in the DNA sequence, which vary from the substitution of a single nucleotide to that of the deletion or addition of multiple nucleotides. Thus, hereditary and congenital diseases are conventionally classified into three broad categories: (1) chromosomal abnormalities, (2) single-gene or monogenic disorders, and (3) polygenic disorders or complex traits.

Chromosomal Abnormalities

Each human cell has two copies of each chromosome (diploids). Chromosomes are double stranded and have two arms, referred to as the long, or “q,” and the short, or “p,” arms. The arms of the chromosomes meet at a primary constriction, referred to as the centromere. Mutations typically occur during DNA replication and meiosis when chromosomes separate. Mutations can involve large deletions, duplications, translocations, and rearrangements. This can result in aneuploidy: too few or too many chromosomes. Chromosomal abnormalities are relatively common during embryonic life and lead to spontaneous abortion, often during the first trimester of pregnancy. However, a significant number of fetuses with chromosomal abnormalities survive. Chromosomal aberrations, numerical or structural, occur in approximately 1 in 150 liveborn infants.⁸ Most diseases caused by chromosomal abnormalities are detected in the neonates or infants because of involvement of many genes causing distinctive phenotypes that are easily diagnosed on physical examination. Chromosomal abnormalities often lead to structural heart defects and are found in 5% to 13% of live born children with congenital heart disease.^1,2

The usual cause for gain of a chromosome is nondisjunction, resulting from failure of a homologous pair of chromosomes to separate during meiosis. When an additional copy of the chromosome is added during fertilization, three copies of the same chromosome (or only one copy) are found in the new zygote instead of the chromosome pair. Two of the most common chromosomal disorders causing heart disease in the adult, namely Down syndrome (trisomy 21) and Turner syndrome (XO), are both commonly caused by nondisjunction. Chromosomal rearrangements occur when a chromosome breaks and rejoins within itself incorrectly, which can result in an inversion of the genetic material. Inversion occurs when a chromosome breaks at two points and the intermediate segment reunites in inverted orientation. Typically, there is no apparent phenotype in persons carrying an inversion, unless it disrupts a gene or its regulatory regions. Isochromosomes are formed when one of the chromosome arms is lost and two long or short arms are joined to form a single chromosome. Chromosome duplications or gains of chromosomal material may also be associated with phenotypic abnormality, but most commonly, they cause no obvious aberration.

Chromosome deletions are large deletions (equal to or greater than 10⁶ base pairs) that commonly lead to loss of a large amount of DNA and loss or disruption of multiple genes. Consequently, a series of phenotypes in a single individual may be present as a result of interruptions in a series of genes within the loci of a single chromosome.

Alterations in chromosome structure could lead to structural variations, which are defined as changes such as inversion, duplication, deletion, and rearrangements. Each human genome also contains several thousand structural variations.^9,10 The size of the structural variations could vary from 2 base pairs, but is typically larger than 1 kbp, to several million base pairs. Insertion or deletion structural variations that change the copy number of the genes are commonly referred to as copy number variants (CNVs). Structural variants or CNVs are relatively common in the population, but do not necessarily cause a gross phenotype, as typically observed in chromosomal abnormalities. However, structural variations including CNVs play significant roles in susceptibility to complex phenotypes, such as developmental abnormalities associated with congenital heart defects and neurological disorders.

Single-Gene Disorders

A single-gene disorder is an inherited disease caused by a mutation in a single gene (monogenic). Only a small fraction of cardiovascular disorders is monogenic. Single-gene disorders that occur in families are characterized by Mendelian patterns of inheritance (Fig. 55–2). They are classified as autosomal dominant, autosomal recessive, or X-linked (dominant or recessive). The majority of monogenic diseases exhibit an autosomal dominant mode of inheritance. In an autosomal dominant disease, approximately half of the members of the family carry the causal mutation and are affected. Monogenic disorders with an autosomal recessive inheritance are caused by mutations in both copies of the gene. Therefore, in a given family, only 25% of the offspring exhibit the phenotype, 50% carry the mutation, and 25% are normal. In X-linked inheritance, males exhibit the disease and females are usually free of the phenotype but carry the mutation. However, if the mutation involves a major protein, the effect of the mutation may be dominant and females can exhibit the clinical phenotype. There is also no father-to-son transmission in X-linked inheritance. In diseases caused by mitochondrial DNA mutations, inheritance is from the mother (no male-to-male transmission), because mitochondrial DNA is predominantly inherited from the ovum.

The DNA mutation could give rise to a change in the corresponding amino acids of the encoded protein, and exerts its deleterious effects via functional alterations. A change in even one amino acid located in a critical domain of the protein can enhance the function (gain-of-function mutation) or impair the function (loss-of-function mutation), with a concomitant change in the clinical phenotype. On average, a mutation occurs every 1.2 × 10⁸ nucleotides during meiosis, resulting in about 30 to 50 de novo variants in each offspring.^11,12 Only mutations occurring in the gametes (male or female germ cells) are transmitted.

In single-gene disorders, a mutation in a single gene typically causes the disease. However, various factors including genetic variants other than the causal variant affect phenotypic expression of the disease. Genetic factors that influence phenotypic expression of the disease are called the modifier genetic variants. Modifier genes and the environmental factors are major determinants of phenotypic expression of a single-gene disorder. Typically, penetrance (proportion of the mutation carriers showing the phenotype) is less than complete and, given the age-dependent penetrance of single gene disorders, young mutation carriers might not express the disease. Occasionally, diseases that are conventionally known, as single-gene disorders are digenic or oligogenic in origin, as two or more mutations in the given gene or in multiple genes cause the phenotype.

Common Polygenic Disorders

Polygenic disorders such as coronary artery disease (CAD) and systemic arterial hypertension are caused by the stochastic and nonlinear interactions of the genetic and environmental factors. In this setting, the presence of a single variant is neither sufficient to cause a disease nor its absence prevents development of the disease. Essentially, all of the common disorders have a genetic predisposition, which is imparted by the genetic variants. Polygenic disorders account for the majority of the cardiovascular diseases, including atherosclerosis, hypertension, obesity, diabetes mellitus, and peripheral arterial disease. In polygenic diseases, multiple genes predispose to the condition, each contributing modestly. The genetic variants associated with complex diseases are typically single nucleotide variants (SNVs), which are also called single nucleotide polymorphisms (SNPs). SNVs are distributed throughout the human genome with a frequency of about 1 per 800 base pairs (bps).^13,14,15 Thus, each genome contains about 4 million SNVs. The vast majority of these variants have negligible biological effects and are clinically insignificant.^16,17 A fraction of SNVs are functional variants and contribute to the pathogenesis of common complex diseases. The effects of SNPs could confer protection against or susceptibility towards a complex phenotype. Technological advances since 2005 have provided the opportunity to perform genome-wide association studies (GWAS), which has led to the mapping of thousands of loci predisposing to polygenic diseases, including CAD (discussed later in this chapter).

Classification of Mutations

Most human diseases exhibit genetic heterogeneity, defined as being caused by different genes and mutations that result in the same phenotype. The heterogeneity may arise from multiple mutations in one gene (allele heterogeneity) or in two or more genes (locus heterogeneity). Within any one family, however, typically there is one causal gene and mutation in all affected members; uncommonly there are two different causal mutations or genes transmitted for the same disease in a given family. A good example is familial hypertrophic cardiomyopathy (HCM), which involves more than a dozen different genes (locus heterogeneity) with multiple mutations in each (allelic heterogeneity). In less than 10% of the cases, HCM is caused by double or triple mutations.^18,19

Mutations can involve a microscopically visible alteration, such as deletion or translocation of a portion of the chromosome (chromosomal abnormalities), or a minute change in the DNA sequence, such a change from adenine (A) to guanine (G), resulting in a change in the amino acid sequence. Mutations involving only a single nucleotide are known as point mutations and are responsible for 70% or more of all adult single-gene disorders (http://www.hgmd.cf.ac.uk/). A point mutation may be a substitution of one nucleotide for another, changing the amino acid sequence (missense mutation); or it may change from encoding an amino acid to become a stop codon, which will truncate the protein (truncated or nonsense mutation); or it may eliminate a stop codon so the protein is elongated (elongated mutant). Finally, it may change the codon without changing the amino acid sequence (synonymous mutation). All genes during transcription (synthesis of an RNA from a DNA template) and translation (synthesis of a protein from a mRNA template) are read from 5′ to 3′ orientation (see Fig. 55–1), with each triplet of bases (codon) coding for a specific amino acid. If a nucleotide is deleted (deletion) or an additional one is inserted (insertion), it will shift the reading frame. The resulting protein would be entirely different (frameshift mutation) and usually nonfunctional. If a purine nucleotide is substituted for a pyrimidine or vice versa, the mutation is referred to as a transversion. If purine or pyrimidine substitutes for another purine or pyrimidine, respectively, it is called a transition. Other mutations may result from the deletion or addition of several nucleotides. In one form of myotonic dystrophy, for example, a triplet repeat of several thousand nucleotides in length is inserted into the 3′ end of the gene. Another type of mutation is known as a gene conversion, where two genes interact and part of the nucleotide sequence of one gene becomes incorporated into that of the other. Mutations in genes exert their deleterious effects via a structural alteration of the protein that has functional consequences, as noted.

Genetic Penetrance and Expressivity

Penetrance is defined as the percentage of individuals within a family who have inherited the causal mutation and have one or more features of the disease. Penetrance is an all-or-none phenomenon. Any manifestation, however minute, indicates that the gene has penetrance in that individual. Nonpenetrance refers to lack of any observable phenotype. This feature is distinct from expressivity, which refers to the variable nature of the clinical phenotype, such as the severity. Thus, by definition, to have expressivity, the trait must be penetrant. Numerous genetic and environmental factors can affect expression of a gene, making it nearly impossible to determine which factor is most important in a specific individual or disease. Table 55–1 shows these factors.

Patterns of Inheritance

Inherited disorders caused by a single abnormal gene are transmitted to offspring in a predictable fashion termed Mendelian transmission. As previously noted, each individual has two copies of each gene, referred to as alleles, one transmitted from each parent. Per Mendel’s first law, each of the two alleles, located on separate chromosomes, segregates independently and is transmitted unchanged to offspring. Thus, the chance of inheriting the mother’s allele versus the father’s is 50%. Mendel’s second law states that genes on the same chromosome also assert themselves independently through the process of crossover between segments of chromosomes (discussed below). The greater the distance between two loci, the more likely they are to be separated during genetic transmission. Mutant genes located on any of the 22 autosomal pairs or the 2 sex chromosomes may produce phenotypes inherited by simple patterns classified as autosomal (dominant or recessive) or X-linked, respectively. The terms dominant inheritance and recessive inheritance refer to characteristics of the phenotype. Dominant inheritance implies that a person with one copy of a mutant allele and one copy of the normal allele develops the phenotype associated with the mutant allele. Recessive traits, on the other hand, require both alleles to be mutant in order to produce a phenotype.

Autosomal Dominant Inheritance

Dominant disorders are those exhibiting a phenotype in heterozygous individuals, as noted. Males and females are equally affected, and offspring of an affected heterozygote have a 50% chance of inheriting the mutant allele (see Fig 55–2). In a sporadic case, the mutation occurs de novo and in one of the germ lines of parents (typically sperm). By definition, it is absent in the somatic cells (cells other than the germ cells) of parents. Autosomal dominant inheritance can be misdiagnosed as sporadic if there is low expressivity in the phenotypically normal parent carrying the mutant allele or if extramarital paternity has occurred. Table 55–2 lists the features characteristic of autosomal dominant inheritance.

Autosomal Recessive Inheritance

Autosomal recessive phenotypes are clinically apparent when the patient carries two mutant alleles (ie, is homozygous) at the locus responsible for the disease (see Fig. 55–2). The disease-causing gene is found on one of the 22 autosomes. Thus, both males and females are equally affected. Clinical uniformity is typical and disease onset generally occurs early in life. Recessive disorders are more commonly diagnosed in childhood than are dominant diseases. On average, only 1 in 4 children (25%) will be affected (see Table 55–2).

X-Linked Inheritance

X-linked inherited disorders are caused by defects in genes located on the X chromosome. Because females have two X chromosomes, they may carry either one mutant allele (heterozygote) or two mutant alleles (homozygote). The trait may, therefore, display dominant or recessive expression. Males have a single X chromosome (and one Y chromosome). Consequently, a male is expected to display the full syndrome whenever he inherits the abnormal gene from his mother. Hence, the terms X-linked dominant and X-linked recessive apply only to the expression of the gene in females. Because a male must pass on his Y chromosome to all male offspring, no male-to-male transmission in X-linked disorders can occur. On the other hand, a male must contribute his one X chromosome to all daughters (see Fig. 55–2). All females receiving a mutant X chromosome are known as carriers, and those who become affected clinically with the disease are known as manifesting female carriers. Table 55–2 lists the characteristic features of X-linked inheritance. Examples of X-linked disorders of the heart include X-linked cardiomyopathy, Barth syndrome, and Duchenne, Becker, and Emery-Dreifuss muscular dystrophies.

Mitochondrial Inheritance

Spermatocytes contribute few or no mitochondria to the zygote. The entire mitochondrial DNA in an embryo is derived from the mitochondria already present in the cytoplasm of the oocyte. Thus, phenotypes caused by mitochondrial DNA mutations demonstrate only maternal inheritance (see Fig. 55–2). Table 55–2 lists the characteristic features of mitochondrial inheritance.

Chromosomal Mapping in Single-Gene Disorders

To locate a particular gene, one must first map the chromosomal locus, which requires knowledge of certain chromosomal landmarks, referred to as DNA markers. A DNA marker is a polymorphic sequence of DNA with a known chromosomal position, which can be detected by analyzing an individual’s DNA. Initial genetic linkage studies were performed using short tandem repeat (STR) DNA markers. In recent years, SNPs have been used for genetic linkage. The typical SNP linkage panel contains several thousands SNPs that are positioned throughout the genome with an average distance of approximately 0.5 Mbp. Genetic distance is measured in terms of centimorgans, named after the geneticist T.H. Morgan. One cM approximates 1 million bp (Mbp). DNA markers, like genes, have two alleles in a given individual and are transmitted to offspring according to Mendel’s law, with the individual being heterozygous or homozygous for that marker. If a marker is homozygous, it is not informative for genetic linkage, as one cannot track cosegregation of the maternal or paternal copy of the marker with the disease. When all of the markers are placed together on each chromosome and the genetic distance between them is estimated, a genetic map is produced.

Identification of a particular locus is made possible by showing that the causal gene of interest is in close proximity to a DNA marker on the same chromosome, a method referred to as genetic linkage analysis. A fundamental requirement for linkage analysis is a family in which the disease of interest is transmitted to offspring over at least two and preferably three generations. At least six affected individuals are required for analyzing cosegregation of DNA markers with inheritance of the disease, although a larger number of affected individuals would be preferable.²⁰

Family History and Evaluation

The most important part of an evaluation for genetic disease is the family history. It may give clues to the diagnosis of a particular phenotype and inheritance patterns within an individual family. For instance, an individual’s ethnic background may suggest the need for specific types of genetic screening, such as for hemoglobinopathies in individuals of African or Mediterranean ancestry or for Tay-Sachs disease in individuals of eastern European (Ashkenazi) Jewish ancestry. The individual with the medical problem who brought the family to the attention of the physician is referred to as the proband or propositus (proposita for females) or index case. Information should generally be collected on all individuals who are first-, second-, or third-degree relatives of the proband. A pedigree chart is then generated. This information should include medical problems and pregnancies. If relatives are deceased, the age at death and the cause of death should be recorded. With a pedigree chart and specific family information, general questions are asked, including whether other family members have the same or similar problems. Information is sought about various types of birth defects, early infant deaths, miscarriages, stillbirths, or other diseases or handicaps in the family. With some disorders, there may be a variability of a particular condition (ie, clinical heterogeneity), even within a family. A pregnancy history may provide information to support a possible teratogenic exposure. Pregnancy and family histories can then be used in conjunction with the findings on physical examination to derive a potential etiologic diagnosis and to plan for further diagnostic studies. The term etiologic diagnosis should suggest whether a specific cardiac defect is familial (by family history), genetic but not familial (sporadic), teratogenic (by pregnancy history), or multifactorial. Prognosis and recurrence risk are linked strongly to an accurate diagnosis and its probable etiology. In sum, accurate phenotypic characterization is essential for all genetic studies.

Concept of Genetic Linkage Analysis

Despite the independent assortment of chromosomes and genes during meiosis, genes (alleles) on two or more loci in close proximity to each other are often coinherited. Two loci that coinherited more than 50% of the time are considered genetically linked. To map the chromosomal locus responsible for a causal gene, DNA markers that are evenly distributed across the chromosomes are selected. DNA is collected from all members of a family (normal and affected) and genotyped for the selected markers. If a DNA marker is coinherited with the phenotype in the affected individuals, the chromosomal locus where the DNA marker resides is in close physical proximity to the locus of the causal gene. This is referred to as genetic linkage between the disease (causal gene) and the DNA marker. Figure 55–3 illustrates the concept of linkage analysis. The odds for and against linkage are calculated using computational algorithms. Linkage exists if the odds in favor of linkage are at least 1000:1. Commonly, the logarithm of the odds, referred to as the LOD score (log of the odds), is used and a LOD score of 3 or greater indicates linkage. A LOD score of –2 (ie, 10² or 100:1 odds against linkage) excludes the linkage. The likelihood of two genes being separated by recombination increases in proportion to the distance between them. The distance between a marker and a disease-causing gene when genetically linked is quite variable and may be anywhere from 1 to 50 Mbp, but is usually within 1 to 10 Mbp. Thus, the inherent resolution of genetic linkage analysis is not better than 1 million base pairs Mbp.

It is possible on the basis of linkage analysis alone to construct a chromosomal map of all of the DNA markers, with the distance between the various markers estimated in centimorgans. This is a complex calculation derived from the number of recombinations between the DNA markers during meiosis. The recombination frequency between two markers, two genes, or a gene and a marker is the ratio of the number of crossover events to the total number of meioses. The lower the recombination frequency between the locus of a DNA marker and that of a disease-causing gene, the closer those two are in physical distance on the chromosome. However, despite the close physical proximity of the loci of the DNA marker and the disease-causing gene, recombination still may occur. The extent to which recombination does occur reflects roughly the physical distance between the two loci. The recombination fraction (or theta) is used to develop a means of estimating the genetic distance (in centimorgans) between genetically linked loci. A recombination frequency or crossover of 1% between two loci, whether occupied by two genes or one gene and a DNA marker, reflects a physical distance of approximately 1 centimorgan between them. For a marker and a gene separated by 1 centimorgan, this means the chance of a crossover between them during meiosis is only 1%; thus, the chance of their being coinherited is 99%. This is a statistically derived genetic map, however, and the distances are only approximate.

Identification of the Gene and Causal Mutation

Once the chromosomal location of a gene has been mapped, the first technique in attempting to identify the gene is referred to as the positional candidate gene approach. There are about 20,000 genes in the genome. Known genes in the mapped chromosomal region are sequenced as candidate genes that contain the causative mutation. If none of the candidate genes in the region is shown to have a mutation that cosegregates with the disease, it may be necessary to clone the region. This approach is referred to as positional cloning, so named because a region is cloned knowing only its position relative to the genetically linked DNA markers. Positional cloning is usually unnecessary as most genes in the human genome have been mapped and identified. However, if attempted, it is necessary to reduce the region (containing the gene) to 1 centimorgan or less. It is often necessary to expand the family with the hope of finding crossovers such that DNA markers common to all affected would span only a short distance (< 1 centimorgan). The cloned genes, or DNA amplified by polymerase chain reaction (PCR), are then analyzed, commonly by direct sequencing, for the presence of the mutation. With the advent of massively parallel sequencing, often referred to as Next Generation Sequencing, typically the entire exome and even the genome is sequenced in all family members and the sequence variants are used in the linkage analysis.²¹ To strengthen the causality, the mutation must be shown to co-segregate with the disease and not with the unaffected members in the family. In addition, it is crucial to show that the variant is absent in large number of normal individuals with the same ethnic background and, hence, it is not a polymorphism. Finally, to support the causality, in vivo and in vitro functional studies are necessary. Table 55–3 summarizes the approach to chromosomal mapping of hereditary diseases by linkage analysis and subsequent isolation of the gene.

Genetic Counseling Principles

Genetic counseling should provide information about the diagnosis, possible etiology, and prognosis of a disease, as well as genetic-based interventions. In addition, psychosocial issues, reproductive options, and the availability of prenatal diagnosis should be discussed. Genetic counseling should be nondirective, providing information in a nonjudgmental, unbiased manner. The family should then be able to make decisions based on medical information in the context of their religious, moral, cultural, and social backgrounds and their financial situation. Although a genetic counselor may occasionally feel frustrated with a specific couple’s decision, an effective counselor does not let personal biases interfere with the counseling role. Conflicts leading to major ethical issues and disputes may arise, however, and may be particularly apparent regarding issues of nonpaternity, sex selection, pregnancy termination, and selective nontreatment of malformed infants. Couples have many potential reproductive options, but not all may be acceptable religiously or culturally. Nevertheless, potential options should be mentioned in a sensitive manner. A common misunderstanding among families in genetic counseling is the issue of prenatal diagnosis and its relationship to abortion. Prenatal diagnosis does not imply that a parent should or would terminate the pregnancy. In many circumstances the information from prenatal diagnosis may help to reassure a couple that their risk of having another handicapped child is, in fact, much lower than expected. Conversely, if defects are found, the subspecialist may use more diagnostic approaches to make rational decisions about medical management of the infant prior to or immediately after delivery.

The accelerated pace of progress in gene discovery, molecular medicine, and molecular diagnostics has begun to allow for improved genetic counseling and portends the possibility of genetic-based therapies.²² As knowledge about the genetic basis of disease grows, however, so does the potential for discriminatory health insurance policies to exclude individuals who are at risk for an illness or to charge prohibitively high rates on the basis of predetermined illness. For this reason, planners of the Human Genome Project recognized the need to protect individuals who volunteer for genetic study as well as those diagnosed by molecular methods in the future. Also for this reason, the National Institutes of Health–Department of Energy Working Group on Ethical, Legal, and Social Implications of the Human Genome Project was developed. After years of debate in Congress, the Genetic Information Non-discrimination Act (GINA) was enacted May 21, 2008. The bill is designed to protect Americans against discrimination based on genetic information for hiring or insurance.²³

Table 55–4 lists chromosomal defects that cause cardiovascular abnormalities. The most common chromosomal defects are described briefly below.

Down Syndrome

Down syndrome, or trisomy 21, is a major cause of congenital heart disease, with a characteristic set of facial and physical features. The incidence of Down syndrome is approximately 1 in 700 livebirths, affecting more than 350,000 individuals in the United States alone.^24,25,26 The risk of having a liveborn with Down syndrome increases with maternal age. It is estimated at 1 in 1000 at age 30 years and 10-fold higher at age 45 years.²⁷ The recurrence rate in the offspring is approximately 1%. Clinical manifestations include congenital anomalies of the heart and gastrointestinal tract, epicanthal folds, flattened facial profile, small and rounded ears, upslanted palpebral fissures, excess nuchal skin, and brachycephaly.²⁴ An increased risk of leukemia, immune system defects, and an Alzheimer-like dementia are associated with Down syndrome. Cardiac abnormalities are present in approximately half of cases.^{26,28,29,30,31} The most common cardiac abnormalities are atrioventricular canal defect and isolated ventricular septal defect (VSD), which occur in 45% and 35% of cases, respectively. The incidence of atrioventricular septal defect is about 1000-fold higher in patients with Down syndrome as compared to the general population.^29,30 Isolated secundum atrial septal defect (ASD) is present in 8% and tetralogy of Fallot in 5% of cases.^{26,28,29,30,31}

Down syndrome is caused by trisomy 21. It is full trisomy in 95%, chromosomal translocation in 2%, and mosaic in 3%. The vast majority of errors in meiosis leading to trisomy 21 are of maternal origin and occur during the first meiosis in two-thirds and during second meiosis in one-fifth of the cases. The exact causal genes responsible for the cardiovascular defects are unknown. The prevailing hypothesis implicates dosage-sensitive genes on chromosome 21 as the main mechanism. Likewise, multiple genes might act in concert to affect specific biological pathways and lead to cardiac phenotypes. However, the exact causal genes for cardiovascular anomalies in Down syndrome are unknown. Several Down syndrome critical regions have been mapped but not conclusively shown to be responsible for cardiac defects. Among the candidate genes are DSCAM, encoding the Down syndrome cell adhesion molecule, which is the only gene in one of the mapped critical regions that is expressed in the heart.³² Likewise, COL6A1, which is highly expressed at the atrioventricular cushions in the heart; CREDL1, which encodes a cell adhesion molecule; and RCAN1, a regulator of protein phosphatase calcineurin 1, are implicated in cardiac anomalies in Down syndrome.^33,34

Turner Syndrome

Turner syndrome is characterized by a constellation of findings that result from partial or complete monosomy of the X chromosome.^24,35 It is the most common chromosomal abnormality in females, with an incidence of 50 per 100,000 liveborn girls, which corresponds to approximately 2 million cases worldwide.³⁶ It is characterized by cardiovascular anomalies, short stature, low-set ears, excess nuchal skin, broad chest with widely spaced nipples, peripheral lymphedema, and ovarian dysgenesis.²⁴ Turner syndrome is associated with increased risk of diabetes mellitus, hypothyroidism, osteoporosis, aortic dissection, and hypertension. Cardiac abnormalities are common, with a prevalence estimated to be between 20% and 40%.³⁶ The most common cardiovascular abnormalities are bicuspid aortic valve, which is present in 20% to 30%, coarctation of aorta, present in 10% to 15% of the adult cases, and aortic dilatation and aneurysm, present in 10% to 40% of the cases.³⁶ The prevalence of these abnormalities is higher in children. Less common cardiovascular anomalies include aortic stenosis, systemic hypertension, mitral valve prolapse, conduction defects, partial anomalous venous drainage, and VSD. Women with Turner syndrome are more susceptible to aortic aneurysms and ischemic heart disease. Patients with Turner syndrome should undergo periodic cardiovascular evaluation, including 12-lead ECG and echocardiography.

Turner syndrome is caused by complete or partial absence of an X chromosome. The most common karyotype is monosomy X (45,X).^35,36 The classic 45X karyotype account for about half of the Turner cases. Approximately 5% to 10% of the cases have duplication of the long arm of one X (46,X,i[Xq]) and the rest have mosaicism. The pathogenesis of Turner syndrome is not fully understood. It likely entails haploinsufficiency of genes (located on the X chromosome) that, under normal conditions, escape inactivation. Inactivation of one copy of the X chromosome during early embryogenesis is partial and several genes escape inactivation. Specific genes that account for the cardiovascular phenotype in Turner syndrome are unknown. SHOX (short stature homeobox-containing gene) or PHOG (pseudoautosomal homeobox-containing osteogenic gene), which encode two isoforms of a homeodomain protein, are considered responsible for the short stature in Turner syndrome.^37,38 Hormone replacement therapy including administration of low-dose estrogen and growth hormone might be beneficial in attenuating metabolic abnormalities, ovarian failure, and growth retardation in children with Turner syndrome.³⁹

Edwards Syndrome

Edwards syndrome, or trisomy 18, is the second most common trisomy, with a prevalence of approximately 1 in 6000 to 8000 livebirths.⁴⁰ Its prevalence increases with increasing maternal age. The majority of the infants die within a couple of weeks and approximately 5% to 10% survive more than a year.⁴⁰ The major causes of death include central apnea, cardiac failure, and respiratory failure. The syndrome is characterized by anomalies of the heart and microcephaly with a prominent occiput, a narrow forehead, low-set and malformed ears, micrognathia, clefting of the lip and palate, clenched hand with overlapping digits, rocker-bottom feet, and various hernias.⁴⁰ Cardiovascular anomalies are present in approximately 90% of the cases.⁴¹ They include VSD, ASD, patent ductus arteriosus (PDA), pulmonary stenosis, tetralogy of Fallot, transposition of the great arteries, bicuspid aortic valve, dysplastic valves, and coarctation of the aorta.^41,42 Complex cardiac malformations occur in approximately 10% of cases. Full trisomy occurs in about 90%, chromosomal translocation in 3%, and mosaicism in 5% of cases. The causal gene(s) for the cardiovascular anomalies remain unknown.

Patau Syndrome

Patau syndrome, or trisomy 13, is a rare disorder with an incidence of 1 per 5000 to 1 in 20,000 livebirths and a high early mortality.^43,44 Approximately 50% of the affected infants die within the first month and 85% within first year of life.⁴⁴ Patau syndrome is characterized by cardiac, urogenital, craniofacial, and central nervous system anomalies. Specific anomalies include microcephaly with sloping forehead, microphthalmia, cleft lip and palate, overlapping fingers with postaxial polydactyly, renal abnormalities including polycystic kidney disease, and cutis aplasia. Cardiac abnormalities are present in approximately 80% of cases. They include VSD, ASD, PDA, pulmonary stenosis, coarctation of the aorta, dextrocardia, and truncus arteriosus.⁴⁴

Patau syndrome is caused by nondisjunction of chromosome 13 during meiosis in the vast majority of cases and rarely by translocation. Five percent of the cases are mosaic. The causal genes for cardiovascular anomalies in trisomy 13 are unknown.

22q11.2 Deletion, DiGeorge (Catch-22) and Velocardiofacial Syndrome

22q11.2 deletion syndrome, formerly known as DiGeorge and velocardiofacial syndromes, is an autosomal dominant congenital syndrome caused by hemizygous microdeletion of a large segment of the long arm of chromosome 22 (22q11). The deletion leads to anomalies of multiple organs including the heart and facial bones. The prevalence is approximately 1 in 4000, accounting for approximately 15% of all congenital heart defects.⁴⁵ The term CATCH-22 denotes cardiac, abnormal facies, thymic hypoplasia, cleft palate, hypocalcemia (as a result of parathyroid hypoplasia), and the 22nd chromosome. Cardiac anomalies are present in approximately two-thirds of the cases. A diverse array of congenital heart defects, specifically those involving the aortic arch, include tetralogy of Fallot, interrupted aortic arch, truncus arteriosus, and PDA. The 22q11.2 deletion syndrome accounts for approximately 15% of Tetralogy of Fallot cases.⁴⁶ Patients with velocardiofacial syndromes exhibit craniofacial anomalies, cleft palate, and a variety of cardiac abnormalities, such as aortic arch anomalies, tetralogy of Fallot, and VSD. Cardiac valves and the myocardium are usually spared.

DiGeorge syndrome is caused by microdeletion of approximately 3 Mbp of DNA encompassing approximately 30 genes. Genetic analysis in mouse and mutation analysis of the candidate genes in the region have led to identification of mutations in TBX1, which encodes a T-box transcription factor.⁴⁷ TBX1 is critical for embryogenesis of the second heart field and the aortic and pulmonary outflow tracts. Loss-of-function mutations in TBX1 result in haploinsufficiency. The downstream target genes of TBX1 and the pathways involved in the pathogenesis of cardiac phenotype are mostly unknown. Likewise, a common copy number variant involving SLC2A3 seems to be a genetic modifier of 22q11.2 deletion syndrome.⁴⁸

A recent whole exome sequencing of 184 individuals with 22q11.2 syndrome, including 89 cases with severe congenital heart defects, implicated genes involved in epigenetic regulation of gene expression through histone modifications in the pathogenesis of this syndrome.⁴⁹ Rare deleterious variants were enriched in JMJD1C, PREB1, MINA, and KDM7A, which encode proteins involved in demethylations of H3K9 and H3K27.⁴⁹ These findings are in accord with the recent large-scale sequencing projects implicating epigenetics in the pathogenesis of a garden-variety form of congenital heart disease, particularly congenital heart disease in conjunction with neurological abnormalities.^50,51

Another candidate gene is UFD1L, which encodes a protein involved in degradation of ubiquitinated proteins. However, its role in 22q11.2 deletion syndrome has not been established.

Congenital heart diseases often occur in isolation and are not part of complex phenotypes, as observed in chromosomal abnormalities. Recently, the causal genes for several congenital heart diseases have been identified. Preliminary studies depict a common theme in the pathogenesis of isolated congenital heart defects, which implicate deficiency of several transcriptional and epigenetic factors that regulate cardiac gene expression during embryogenesis. However, there is considerable phenotypic, locus, and allelic heterogeneity.

Supravalvular Aortic Stenosis

Supravalvular aortic stenosis is an autosomal dominant disease characterized by discrete narrowing of the ascending aorta above the level of the sinus of Valsalva. It commonly occurs as a phenotype of Williams syndrome (or Williams-Beuren syndrome) in conjunction with significant development delay in some, and exceptional talents in others, hypercalcemia, characteristic facial appearance, and stenosis of other major arteries.⁵² The prevalence of supravalvular aortic stenosis is estimated to be 1 in 25,000 livebirths.

The gene responsible for supravalvular aortic stenosis was initially mapped to chromosome 7q11.23 and subsequently identified as ELN, encoding elastin.^53,54 Point and deletion mutations in ELN gene are the main causes of isolated supravalvular aortic stenosis. Mutations result in elastin deficiency, which in the vascular system leads to inelasticity of the vessel wall and subsequent fibrosis as a result of an altered stress–strain relation (elastin arteriopathy). Thus, haploinsufficiency underlies the pathogenesis of supravalvular aortic stenosis.

Patients with Williams syndrome may exhibit additional cardiovascular phenotypes, including, but less commonly, pulmonary arterial stenosis, aortic and mitral valve abnormalities, and tetralogy of Fallot.⁵⁴ In 98% of cases of Williams syndrome the deletion mutation includes 1.5 Mbp of DNA comprising ELN and another 20 contiguous genes. Contribution of these genes to pathogenesis of specific phenotypes in Williams syndrome remains unknown.

Familial Atrial Septal Defect

ASD is among the most common congenital heart diseases, with an estimated incidence of 1 in 1000 livebirths.¹ ASD is usually sporadic. However, familial ASD with an autosomal dominant mode of inheritance also occurs. Individuals with ASD are commonly asymptomatic until the third or fourth decades, when they may present with signs and symptoms of right heart failure, pulmonary hypertension, and decreased exercise tolerance. Common symptoms are palpitations, commonly caused by supraventricular arrhythmias, and symptoms associated with pulmonary hypertension and right-sided volume overload resulting in left-to-right shunt. Uncorrected ASD can lead to heart failure, Eisenmenger syndrome, and premature death in the fourth or fifth decade of life.

The first gene identified for familial ASD is NKX2–5, which is the human homologue of Nkx2.5 in mouse and tinman in Drosophila melanogaster.⁵⁵ The gene is located on 5q35 and encodes NKX2-5, a predominantly cardiac-specific transcription factor that regulates expression of several cardiac genes and heart development in various organisms.^56,57 A multiplicity of mutations has been described in patients with secundum ASD and conduction defects. Mutations often result in haploinsufficiency and those located in the DNA-binding domain reduce the binding affinity of NKX2-5 for the promoter region of its target genes and, hence, reduced gene expression.⁵⁸ The spectrum of clinical phenotypes caused by mutations in NKX2-5 extends beyond secundum ASD and comprises VSDs, tetralogy of Fallot, subvalvular aortic stenosis, pulmonary atresia, and others.⁵⁹

The second causal gene for familial ASD with an autosomal dominant mode of inheritance is GATA4 on chromosome 8p22–23.⁶⁰ The mutations diminish DNA-binding affinity and transcriptional activity of GATA4 transcription factor, and block its physical interaction with TBX5, another transcription factor involved in the pathogenesis of congenital heart disease.⁶⁰

The third causal gene for familial ASD is MYH6, which is located in chromosome 14q12 and encodes myosin heavy chain 6.⁶¹ Missense mutations in MYH6 are rare causes of isolated ASD of ASD occurring in conjunction with atrial fibrillation. The MYH6 protein is expressed at high levels in atrial tissues and is a major component of the sarcomere.

Holt-Oram Syndrome

Holt-Oram syndrome is a rare autosomal dominant inherited disorder characterized by anomalies of the heart and upper extremities, hence the name hand–heart syndrome.^62,63 The most common congenital heart defects are ASD and VSD followed by conduction system abnormalities and atrial fibrillation. Less common cardiac abnormalities include truncus arteriosus, mitral valve defect, PDA, and tetralogy of Fallot. Anomalies of the upper limb vary from mild malformation of the carpal bones to phocomelia, but upper limb preaxial radial abnormalities are commonly present.

Mutations in TBX5 on chromosome 12q24, which codes for transcription factor TBX5, are responsible for the cardiac and skeletal abnormalities in Holt-Oram syndrome (Fig. 55–4).⁶⁴ A number of mutations have been described and most are nonsense, frameshift, or splice-junction abnormalities. The proposed molecular mechanism is haploinsufficiency, resulting in reduced expression level of TBX5. Haploinsufficiency because of truncation or frameshift mutations results in severe birth defects in the heart and hands, whereas point mutations predominantly affect either hand or heart development.⁶⁵ Mutations in the 5′ end of the gene exhibit a preponderance of cardiac abnormalities with mild skeletal abnormalities, and those in the 3′ end lead to severe skeletal and mild cardiac abnormalities.

Ellis-van Creveld Syndrome

Ellis-van Creveld syndrome is an autosomal recessive skeletal dysplasia, which is associated with congenital heart disease in the majority of cases.⁶⁶ Skeletal anomalies include short limbs, short ribs, postaxial polydactyly, and dysplastic nails and teeth.⁶⁶ ASD and common atrium are the typical cardiac anomalies present in two-thirds of the cases.⁶⁶ Several splice donor, truncation, and missense mutations in EVC1 and EVC2 (LBN) genes, located on chromosome 4p16, have been identified as causes of Ellis-van Creveld syndrome.⁶⁷ The genes are coexpressed in atrial septum primum.⁶⁸ However, the pathogenesis of Ellis-van Creveld syndrome remains unknown.

Familial Patent Ductus Arteriosus (Char Syndrome)

PDA can occur as a sole cardiac anomaly or in conjunction with other congenital heart disease. Familial PDA with an autosomal dominant inheritance has been described in patients with Char syndrome. Char syndrome is a congenital disease that was first described by Florence Char in 1978. It is characterized by a constellation of facial dysmorphism, fifth-finger middle phalangeal hypoplasia, and PDA. Variation of this syndrome is associated with bicuspid aortic valve, distinctive facial appearance, polydactyly, and fifth-finger clinodactyly. The predominant clinical features are those of PDA, which include symptoms and signs of left-heart failure and pulmonary hypertension.

The gene responsible for Char syndrome in two families was recently mapped to chromosome 6p12–21.⁶⁹ Subsequently mutations in the TFAP2B, which encodes a neural crest-related helix-span-helix transcription factor, were identified.⁷⁰ TFAP2B regulates cell growth, differentiation, and apoptosis.⁷¹

Noonan and Leopard Syndromes

Noonan syndrome is an uncommon autosomal dominant disorder (estimated prevalence 1:1000 to 1:2000 newborns) characterized by dysmorphic facial features, including hypertelorism and low-set ears, HCM, pulmonic stenosis, mental retardation, and bleeding disorders (Fig. 55–5).^72,73 Leopard syndrome (lentigines, electrocardiographic conduction abnormalities, ocular hypertelorism, pulmonic stenosis, abnormal genitalia, retardation of growth, and deafness) is an allelic variant of the Noonan syndrome. Pulmonic stenosis and HCM are the primary cardiac phenotypes. Others include atrioventricular septal defects, aortic coarctation, ASD, mitral valve defects, PDA, and fibroelastosis. Noonan syndrome is also seen in conjunction with cardiofaciocutaneous syndrome and other congenital abnormalities, such as neurofibromatosis. Noonan syndrome has a phenotype that resembles to Turner syndrome and hence it is often referred to as male Turner syndrome. However, Turner is a distinct genetic eitity caused by 45X chromosomal disorder.

Noonan syndrome is sporadic in half of the cases and an autosomal dominant disease in the other half. Several genes for Noonan syndrome have been identified and all are the components of the Ras/mitogen-activated protein kinase (MAPK) pathway (Table 55–5). PTPN11, encoding protein tyrosine phosphatase, nonreceptor type 11, is the most common gene responsible for autosomal dominant Noonan and Leopard syndromes.^74,75 Overall, mutations in PTPN11 are found in approximately one-half to two-thirds of the cases of Noonan syndrome. Mutations are typically gain-of-function mutations leading to increased signaling through Ras/MAPK pathway.

Mutations in SOS1, KRAS, SOS1, RAF1, NRAS, and SHOC2 are also responsible for Noonan syndrome.⁷³ SOS1 is probably the second most common causal gene for Noonan syndrome, accounting for about 15% of cases. SOS1 mutations are also gain-of-function mutations, leading to increased activity of the MAPK pathway.⁷³ Mutations in the other genes collectively account for about 10% of all Noonan syndrome cases.

Familial Myxoma Syndrome (Carney Complex)

Myxomas are the most common cardiac tumors. Myxomas are generally sporadic, but familial forms with an autosomal dominant mode of inheritance also occur in approximately 10% of cases, as a part of Carney complex.^76,77,78 Carney complex is characterized by the constellation of cardiac myxoma, endocrine disorders, and skin pigmentation.^77,78 LAMB (lentigines, atrial myxoma, mucocutaneous myxoma, blue nevi) and NAME (nevi, atrial myxoma, myxoid neurofibromata, ephelides) syndromes are considered variants of Carney complex. Atrial, ventricular, and skin myxomas, endocrine tumors and disorders, such as Cushing syndrome, and skin lesions, such as lentiginosis, are part of the phenotypic expression of Carney complex. These myxomas are commonly found in the right atrial-ventricular groove. Clinical features of atrial myxoma may include fever, arthralgia, dyspnea, diastolic rumble, tumor plop, and systemic embolisms (Fig. 55–6).⁷⁸

Carney complex exhibits locus heterogeneity, and at least two loci on chromosome 17q24 and 2p16 have been mapped.^78,79,80 The majority of familial cardiac myxomas (Carney complex) are caused by mutations in the PRKRA1A gene on chromosome 17q24.⁸¹ This gene encodes the α-regulatory subunit of cyclic adenosine monophosphate (cAMP)-dependent protein kinase. Frameshift mutations in PRKRA1A result in haploinsufficiency, which suggests that the PRKRA1A functions as a tumor-suppressor gene.⁷⁸ Recently, a missense mutation in the perinatal myosin heavy-chain gene (MYH8) was identified in members of a family with Carney complex and trismus-pseudocamptodactyly syndrome.⁸² In addition, PRKACA and PRKACB genes are also implicated in Carney Complex. However, their pathogenic role remains uncertain.

Situs Inversus

Situs inversus is a disorder of laterality leading to reversal of the asymmetric anatomic position of visceral organs. In situs inversus totalis, all visceral organs are reversed in a mirror-image manner (involving the chest and abdomen). It is part of primary ciliary dyskinesia (PCD), previously known as immotile cilia syndrome. Situs inversus might occur in conjunction with sinusitis and bronchiectasis, which is known as Kartagener syndrome. Most cases of situs inversus are sporadic. Autosomal recessive, autosomal dominant, and X-linked forms have been reported.⁸³

Situs inversus, as a component of PCD, such as that in Kartagener syndrome, is caused by defect in NODAL signaling. It typically exhibits an autosomal recessive inheritance and is caused by mutations in dynein axonemal intermediate chain 1 (DNAI1) on chromosome 9p13-p21, dynein axonemal heavy chain 5 (DNAH5) on chromosome 5p, and dynein axonemal heavy chain type 11 (DNAH11) on chromosome 7p21, genes that encode dyanein proteins.^83,84,85 Dyneins are large proteins with adenosine triphosphatase (ATPase) activity that interact with intermediary filaments to produce energy and motion.

Mutations in several other genes, including ZIC3, which encodes a zinc-finger protein of the cerebellum, are associated with laterality defects and situs inversus.⁸³

Alagille Syndrome (Arteriohepatic Dysplasia)

Alagille syndrome is an autosomal dominant disorder characterized by anomalies of the right side of the heart and developmental abnormalities of eyes, skeleton, and kidney. Cardiac abnormalities are present in approximately 70% of cases; the most common is diffuse pulmonary artery stenosis. Others include hypoplastic pulmonary circulation, pulmonary atresia, tetralogy of Fallot, coarctation of aorta, secundum ASD, PDA, and VSD.⁸⁶ The most common causal gene is the Jagged-1 gene (JAG1), located on chromosome 20p12.^87,88 Deletion or point mutations in JAG1 are found in approximately 90% of the patients with Alagille syndrome.^86,87,88 JAG1 is a cell-surface protein that is a ligand for the Notch receptor. The Notch intercellular signaling pathway mediates cell fate decisions during development. The proposed molecular mechanism is haploinsufficiency leading to defective cell adhesions. Recently, mutations in NOTCH2 were found in those who did not have JAG1 mutations.^86,89 Collectively, the findings indicate that Alagille syndrome is a disease of the Notch signaling pathway.

The term cardiomyopathy denotes an exclusive group of disorders in which the primary defect is in the myocardium, affecting cardiac myocyte structure and/or function. The primary defect, however, does not need to be exclusive to the heart. It can also involve other tissues and organs, as in cardiomyopathies arising from metabolic disorders and mitochondrial myopathies. Myocardial dysfunction can also occur because of systemic, infiltrative, restrictive, toxic, and endocrine disorders, coronary atherosclerosis, and valvular pathologies. In such conditions, the primary defect is not in the myocardium. Thus, myocardial involvement is considered secondary. In a sense, cardiac involvement in such disorders does not meet the pure definition of cardiomyopathy.

Cardiomyopathies are classified according to their phenotypic characteristics. The four common groups are hypertrophic, dilated, arrhythmogenic, and restrictive cardiomyopathy. Phenotypic classification, although clinically convenient and useful, does not necessarily reflect the molecular and genetic basis of cardiomyopathies.

Genetic Basis of Hypertrophic Cardiomyopathy

HCM is a relatively common autosomal dominant disease diagnosed clinically by the presence of unexplained cardiac hypertrophy.^90,91 Commonly, a left ventricular wall thickness of 13 mm or greater, in the absence of hypertension or valvular heart disease, is used to define HCM.⁹⁰ The prevalence of HCM is approximately 1 in 500 in young adults.⁹² It is likely higher in the elderly population because of age-dependent penetrance.

Cardiac hypertrophy, the clinical hallmark of HCM, is asymmetric in approximately two-thirds of the cases with predominant involvement of the interventricular septum (Fig. 55–7). Hence, the term asymmetric septal hypertrophy is used to describe this condition. Occasionally, hypertrophy is restricted to apex of the heart (apical HCM). Morphologically, the left ventricular cavity is small and the left ventricular ejection fraction, a measure of global systolic function, is increased. However, more sensitive indices of myocardial function show impaired contraction and relaxation secondary to “myocyte disarray.”^93,94 Diastolic function is commonly impaired, leading to an increased left ventricular end-diastolic pressure and thus frequently to symptoms of heart failure.

Patients with HCM exhibit protean clinical manifestations ranging from minimal or no symptoms to severe heart failure. The clinical manifestations often do not develop until the third or fourth decades of life, but the onset is variable. The majority of patients are asymptomatic or mildly symptomatic. Predominant symptoms include dyspnea, chest pain, palpitations, and/or syncope. Severe systolic heart failure is uncommon. It occurs in a small fraction of patients in whom the disease evolves into a dilated cardiomyopathy (DCM) phenotype. In contrast, cardiac diastolic function is usually impaired and left ventricular end-diastolic pressure is elevated. A dynamic left ventricular outflow is present in approximately 25% of the patients. It could contribute to mitral regurgitations and symptoms of heart failure. Left ventricular outflow tract obstruction is an important determinant of heart failure in patients with HCM.⁹⁵ Cardiac arrhythmias, in particular atrial fibrillation and nonsustained ventricular tachycardia, are relatively common and are associated with adverse clinical outcome.⁹⁶ Wolff-Parkinson-White (WPW) syndrome is present in about 5% of patients with HCM.⁹⁷ Its presence suggests the possibility of a phenocopy, typically a glycogen storage disease (discussed later in this chapter).⁹⁸

Unexplained syncope is a serious symptom.^99,100,101 It is often a result of serious cardiac arrhythmias and associated with an increased risk of sudden cardiac death (SCD).^99,101,102 HCM, although uncommon, is the most common cause of SCD in young, competitive athletes.¹⁰³ It accounts for almost half of all cases of SCD in athletes younger than 35 years of age in the United States.¹⁰³ SCD is often the first and tragic manifestation of HCM in young apparently healthy individuals. Table 55–6 lists the factors associated with an increased risk of SCD. Overall, in the assessment of the risk of SCD, the combination of all known risk factors should be considered. In the absence of major risk factors for SCD, HCM has a relatively benign course with an estimated annual mortality of about 1% in the adult population and even less in those who have undergone a defibrillator implantation.^91,104 Apical HCM is characterized by giant T-wave inversion in the precordial leads on the electrocardiogram. The overall prognosis of patients with apical HCM is similar to those with the garden-variety form of HCM.^91,105

The pathologic hallmark of HCM is cardiac myocyte disarray.¹⁰⁶ It is defined as maligned, distorted, and often short and hypertrophic myocytes oriented in different directions (Fig. 55–7). Myocyte disarray often comprises more than 20% of the ventricle, as opposed to < 5% of the myocardium in normal hearts.^107,108 It is more prominent in the interventricular septum, but commonly is found throughout the myocardium. Other pathologic features of HCM include myocyte hypertrophy, interstitial fibrosis, thickening of media of intramural coronary arteries and, on occasion, malpositioned mitral valve with elongated leaflets. In addition to cardiac hypertrophy and myocyte disarray, interstitial fibrosis is also associated with the risk of SCD, mortality, and morbidity in patients with HCM.^109,110 Late gadolinium enhancement is used as a surrogate marker for interstitial fibrosis and is a predictor of the risk of SCD in patients with HCM.^109,110 Other pathologic features of HCM include thickening of the media of intramural coronary arteries, abnormal positioning of the mitral valve apparatus, and elongated mitral valve leaflets.

Molecular Genetics of Hypertrophic Cardiomyopathy

HCM is a genetically heterogeneous disease with an autosomal dominant mode of inheritance. Approximately two-thirds of patients have a family history of HCM. In the remainder, the disease is sporadic. Familial and sporadic cases both are caused primarily by mutations in contractile sarcomere proteins.^90,91 In sporadic cases, mutations are de novo and could be transmitted to the offspring of the index cases.¹¹¹ Because hypertrophy is a common response of the heart to all forms of injury or stimuli, a phenotype of hypertrophy in the absence of an increased external load could also occur because of mutations in genes encoding proteins other than sarcomere proteins. As such, unexplained cardiac hypertrophy, which clinically denotes HCM, could also occur in storage disorders, metabolic disorders, mitochondrial diseases, and triplet repeat syndromes, as well as congenital heart diseases.^90,112 Although the gross phenotype is similar, the pathogenesis of HCM caused by different classes of mutant proteins, at least in part, could differ. Therefore such conditions are considered phenocopy (diseases mimicking HCM).

Causal Genes and Mutations

The pioneering works of Christine and Jonathan Seidman have led to elucidation of the molecular genetic basis of HCM. In 1990, an arginine-to-glutamine substitution at codon 403 (R403Q) in the β-myosin heavy chain (MHC) was identified as the first causal mutation.¹¹³ Since then, several hundred different mutations in more than a dozen genes encoding sarcomere proteins have been identified (Table 55–7). Consequently, HCM (excluding phenocopy conditions) is considered a disease of contractile sarcomere proteins (Fig. 55–8). Systematic screening of sarcomere genes suggests that mutations in MYHC and MYBPC3, which encode β-MHC and myosin-binding protein C (MYBPC3), respectively, are the most common causes of human HCM, together accounting for approximately half of all cases.^{114,115,116,117} Mutations in TNNT2 and TNNI3, encoding cardiac troponin T and I, respectively, are relatively uncommon, each accounting for approximately 3% to 5% of HCM cases.¹¹⁷ Thus, mutations in MYH7, MYBPC3, TNNT2, and TNNI3 collectively account for approximately 60% of all HCM cases. A small fraction of HCM cases are caused by mutations in genes encoding αtropomyosin (TPM1), titin (TTN), cardiac α-actin (ACTC), telethonin (TCAP), and essential and regulatory light chains (MYL3 and MYL2, respectively).⁹⁰ Rare mutations in several other genes coding for thick and think filament of the sarcomeres as well as the Z-disk proteins have been reported.^118,119 Likewise, rare mutations in non-sarcomere proteins, including calcium regulatory proteins, are associated with HCM.^118,119 Overall, the causal genes and mutations for approximately two-thirds of HCM cases have been identified. The remainder remain to be identified and/or might be caused by phenocopy conditions.

Molecular Genetics of HCM Phenocopy

A phenotypic feature of several diseases, particularly storage diseases, is cardiac hypertrophy mimicking HCM, and hence such conditions are referred to as HCM phenocopy. Despite phenotypic similarities, however, the distinction between true HCM and HCM phenocopy is important, as the pathogenesis of the two conditions differs. Table 55–8 provides a partial list of HCM phenocopy conditions. The prevalence of HCM phenocopy is not precisely known, but might comprise approximately 5% to 10% of the clinically diagnosed adult cases with HCM, and even more in children.^120,121,122 A prototypic example of HCM phenocopy is Fabry disease, an X-linked lysosomal storage disease.^121,122 Fabry disease is present in approximately 1% to 5% of cases with the clinical diagnosis of HCM in the adult population.^122,123 The causal gene is GLA on chromosome Xq22, which codes for lysosomal hydrolase α-galactosidase A (α-Gal A) protein.¹²⁴ The phenotype results from deficiency of α-Gal A, also known as ceramide trihexosidase. Deficiency of the enzyme results in deposits of glycosphingolipids in multiple organs, including the heart. The phenotype is characterized by angiokeratoma, renal insufficiency, proteinuria, neuropathy, transient ischemic attack, stroke, anemia, corneal deposits, and cardiac hypertrophy. Cardiac hypertrophy, which is often indistinguishable from the true HCM, is associated with high QRS voltage, conduction defects, cardiac arrhythmias, and SCD. Other cardiac phenotypes include valvular regurgitation, coronary artery disease, myocardial infarction, and aortic annular dilatation.^121,122,123 The disease predominantly affects males. Female carriers could exhibit a milder form. The diagnosis is established by measuring α-Gal A levels and activity in leukocytes. However, in many cases the diagnosis is difficult to establish, both at the genetic level as well as after measurement of α-Gal A enzymatic activity. Enzyme replacement therapy using human α-Gal A (agalsidase α) or recombinant human α-Gal A (agalsidase β) has been advocated with some evidence of improvement in progression and reversal of the disease.^{125,126,127,128}

Glycogen storage disease caused by mutations in the PRKAG2 gene is another HCM phenocopy.^{98,129,130,131,132} Cardiac hypertrophy results predominantly from storage of glycogen in myocytes but probably also from cardiac myocyte proliferation.¹³³ The gene encodes the γ₂ regulatory subunit of adenosine monophosphate (AMP)-activated protein kinase (AMPK), which is considered the energy biosensor of the cell. Mutations in PRKAG2 lead to cardiac hypertrophy, conduction defects, and WPW.^98,129,132

HCM phenocopy also occurs in trinucleotide repeat syndromes, a group of genetic disorders caused by expansion of naturally occurring trinucleotide repeats. HCM phenocopy occurs in Friedreich ataxia, an autosomal recessive neurodegenerative disease caused by expansion of GAA repeat sequences in the intron of FRDA.¹³⁴

HCM phenocopy also occurs in patients with Noonan syndrome. The phenotype is characterized by dysmorphic facial features, pulmonic stenosis, mental retardation, bleeding disorders, and cardiac hypertrophy, as discussed earlier.⁷³

Metabolic diseases also cause HCM phenocopy. Refsum disease, Pompe disease (glycogen storage disease type II), Danon disease, Niemann-Pick disease, Gaucher disease, hereditary hemochromatosis, and CD36 deficiency are examples of metabolic disorders that cause HCM phenocopy.^112,135 Defective mitochondrial oxidative phosphorylation pathways also cause HCM phenocopy. Kearns-Sayre syndrome is a mitochondrial disease characterized by a triad of progressive external ophthalmoplegia, pigmentary retinopathy, and cardiac conduction defects, and less frequently HCM phenocopy.^112,135,136

Genotype-Phenotype Correlation

A remarkable feature of HCM is the presence of considerable variability in its phenotypic expression, whether it is the degree of cardiac hypertrophy or the risk of SCD.^137,138 The molecular basis of such variability is not fully known. It is probably partly because of the diversity of the causal genes and mutations, which impart a spectrum of functional and structural defects. In addition, the presence of multiple mutations simultaneously present, detected in a small fraction of patients, is associated with a severe phenotype.^19,139 Environmental factors, such as competitive sports and exercise, could potentially contribute to the phenotypic expression of HCM. However, there is insufficient data to support their contributions to the phenotype.

Causal genes and mutations are the primary determinant of expressivity of cardiac phenotype, including the severity of hypertrophy and the risk of SCD. Initial genotype-phenotype correlation studies implied gene- and mutation-dependent phenotypic expression of HCM. For example, p.R403Q mutation was considered a high-risk mutation associated with increased risk of SCD (Fig. 55–9). Likewise, patients with mutations in MYH7 were shown to have more severe hypertrophy than patients with mutations in MYBPC3 or TNNT2.^{140,141,142,143,144} Likewise, the overall impression is that TNNT2 mutations, despite exhibiting a milder hypertrophic phenotype, were associated with a higher risk of SCD.^{140,145,146,147} Such generalizations, however, are not broadly applicable as the results of genotype–phenotype correlation studies in HCM are subject to a large number of confounding factors, including the small size of the families, small number of families with identical mutations, low frequency of each mutation, phenotypic variability, homozygosity for the causal mutations or compound mutations, and the influence of modifier genes and environmental factors. Collectively, the existing data indicate that mutations exhibit highly variable linical, electrocardiographic E, echocardiographic, and cardiac magnetic resonance imaging (MRI) manifestations, and no particular phenotype is mutation specific.

Modifier Genetic Factors

The presence of considerable phenotypic variability among affected members of different families with identical causal mutations emphasizes the significance of the genetic background to phenotypic expression of HCM. Genes other than the causal genes that affect the phenotype are referred to as the “modifier” genes. Unlike the causal genes, modifier genes are neither necessary nor sufficient to cause HCM.¹⁴⁸ However, they influence the severity of cardiac hypertrophy, risk of SCD, and expression of other cardiac phenotypes in HCM. DNA polymorphisms, including SNPs located in the coding or regulatory regions or splice junctions in genes involved in cardiac hypertrophy, are the prime candidates to modify phenotypic expression of HCM. The specific modifier genes in HCM are largely unknown. Five modifier loci have been mapped through genome-wide linkage studies and several genes have been implicated.^{149,150,151,152,153} The effect sizes of the modifier loci are considerable, as the loci in homozygous form could influence expression of cardiac hypertrophy markedly. Functional variants of genes coding for the components of the renin-angiotensin-aldosterone system are the most extensively studied candidates. ACE, encoding angiotensin-I converting enzyme 1 (ACE-1), was the first gene implicated as a modifier of cardiac phenotype, including severity of cardiac hypertrophy and the risk of SCD in human HCM.^152,153 Observational data show cardiac hypertrophy accelerates during puberty and adolescence in patients with HCM.¹⁵⁴ The findings are in accord with the role of growth factors in contributing to expression of cardiac hypertrophy.¹⁵⁵ Differential expression of four-and-a-half domains protein 1 (FHL1) in human HCM has been shown, and alternative 5′ start usage is considered a modifier of cardiac phenotype, including sex-dependent differences, in a mouse model of HCM.^156,157 Overall, the final phenotype in HCM is determined not only by the causal mutations but also by the effects of modifier genes, environmental factors, epigenetic and epistatic factors, and posttranscriptional and posttranslational modifications of the proteins.

Pathogenesis of Hypertrophic Cardiomyopathy

As the diversity of the causal mutations would suggest, there is no single initial defect that is common to all mutations (Table 55–9). The diversity of the clinical phenotypes, such as hypertrophic, dilated, or restrictive cardiomyopathy arising from mutations in the same gene further adds to the complexity of the pathogenesis. Topography of the causal mutation is likely to be important, as the initial defect is likely to be triggered by protein domain. Given that each sarcomere protein has multiple functions, mutations in different domains of the same protein could impart several initial defects.

The causal mutation initiates a series of molecular events, which begins with alteration of the molecular structure and function of the protein (see Table 55–9). Because the majority of mutant sarcomere proteins differ from the wild type only by a single amino acid (missense mutations), the mutant proteins typically incorporate into the sarcomere, albeit sometimes inefficiently. Following incorporation, mutant sarcomere proteins exert diverse functional defects, such as alterations in myofibrillar Ca²⁺ sensitivity and ATPase activity. Altered calcium sensitivity of the myofibrils seems to be a common biological effect of the sarcomere protein mutations.¹⁵⁸ Functional phenotypes lead to activation of secondary molecules, which are largely unknown, but expected to include activating of many intracellular signaling pathways. The secondary molecular phenotype mediates induction of the morphologic and histologic phenotypes. Accordingly, hypertrophy and fibrosis are considered secondary phenotypes because of activation of intermediary molecular phenotypes (see Fig. 55–10).

Many HCM mutations involve deletions or truncations that are considered null alleles because of the possible expression of unstable mRNA and proteins.^{114,116,117,158} Although there might be a partial allelic compensation (the wild type allele partly compensating the deficiency of the encoded protein), a truncation or a null mutation might lead to haploinsufficiency and alter stoichiometry of the sarcomere proteins, resulting in HCM. Regardless of the initial primary defect, cardiac hypertrophy, the clinical hallmark of HCM, is considered a compensatory phenotype resulting from upregulation and activation of various trophic and mitotic factors. The predominant involvement of the left ventricle and its frequent absence in the low-pressure right ventricle, despite equal expression of mutant sarcomere protein in both, suggest contribution of the environment, such as the loading conditions, to the development of hypertrophy. Furthermore, variation in hypertrophic response because of the genetic background, its absence early on in life, and its attenuation through pharmacologic interventions, at least in animal models, supports the secondary nature of hypertrophy. The primary impetus for hypertrophy is not well defined. It is likely to involve altered Ca²⁺ sensitivity of the myofilaments and activation of various calcium-dependent and independent signaling pathways in response to increased cell mechanical stress.

In keeping with the diversity of cardiac phenotypes in HCM, expression levels of a variety of genes, in response to the mutant protein, are altered. Expression of genes encoding contractile sarcomere proteins, cytoskeletal proteins, ion channels, intracellular signaling transducers, proteins maintaining the reduction–oxidation state, and those involved in transcriptional and translation machinery are changed.^156,159 Likewise, expression of a number of microRNAs (miRs) is disregulated and circulating levels of a number of miRs are increased in patients with HCM.^160,161,162 Plasma levels of miR-29a are associated with cardiac hypertrophy and fibrosis in patients with HCM.^160,163

Potential New Therapeutic Interventions

Current pharmacologic interventions in HCM are empiric without a firm evidence of efficacy in regression of hypertrophy, fibrosis, and disarray. Pharmacological therapy is restricted primarily to the use of beta-blockers and, on occasion, calcium channel blockers and antiarrhythmic drugs, as indicated. Nonpharmacological interventions include surgical septal myectomy (Morrow procedure) and catheter-based septal ablation, typically upon infusion of alcohol to the main septal branch of the left anterior decending coronary artery, and are very effective in reducing the left ventricular outflow tract obstruction, improving symptoms, and even possibly survival. However, septal myectomy, whether surgical or catheter-based, does not address the underlying pathology as hypertrophy, fibrosis, and disarray typically persist. Currently, there is no suitable method to correct the underlying genetic defect in humans. Therefore, the emphasis has been on prevention, reversal, and attenuation of the phenotype through pharmacologic interventions aimed at blockade of intermediary molecular phenotypes of genetic interventions in animal models. Recent studies have shown potential clinical usefulness of angiotensin II receptor blockers, beta-hydroxy-beta-methylglutaryl-coenzyme A (HMG-CoA) reductase inhibitors, and antioxidants in prevention, attenuation, and reversal of cardiac phenotypes in animal models of HCM.^{164,165,166,167,168,169} However, preliminary data in human patients with HCM have not shown significant benefits of statins.^170,171 Likewise, blockade of the renin-angiotensin-aldosterone system has shown only a modest, if any, beneficial effect.^{172,173,174,175,176} Preliminary studies suggest potential beneficial effects of pretreatment with diltiazem, an L-type calcium channel blocker, on echocardiographic indices of cardiac size and function in HCM mutation carriers.¹⁷⁷ Randomized clinical trials in human patients with HCM using N-acetylcysteine (HALT-HCM), eleclazine (LIBERTY-HCM), and several others are ongoing (https://clinicaltrials.gov).

Genetic Basis of Primary Dilated Cardiomyopathy

Primary DCM is a disease of the myocardium that manifests by dilatation of the left ventricle along with a gradual decline in contractility. The diagnosis is based on a left ventricular ejection fraction of less than 0.45 and a left ventricular end-diastolic diameter of > 2.7 cm/m². Primary DCM has a prevalence of 40 cases per 100,000 individuals and an incidence of 5 to 8 cases per 100,000 persons.¹⁷⁸ Patients with DCM are often asymptomatic in the early stages, but gradually develop symptoms and signs of heart failure, syncope, cardiac arrhythmias, and SCD. A significant number of affected relatives of patients with DCM are asymptomatic and are diagnosed for the first time on additional testing (such as an echocardiogram or electrocardiogram [ECG]).¹⁷⁹ A normal history and physical examination in a subject at risk, particularly in the early decades of life, does not exclude DCM. A family history of DCM is present in approximately half of all index cases with idiopathic DCM. In the remainder, DCM is considered sporadic. Familial DCM is commonly inherited as an autosomal dominant disease, which clinically manifests during the third and fourth decades of life.¹⁷⁹

An X-linked DCM is suspected when only male members of a family exhibit symptoms and signs of DCM and there is no male-to-male transmission. Three common forms of X-linked DCM have been identified: Duchenne/Becker muscular dystrophies, Emery-Dreifuss syndrome, and Barth syndrome. DCM also occurs in multi-organ disorders, such as mitochondrial DNA mutations, triplet repeat syndromes, and metabolic disorders.

Molecular Genetics of Dilated Cardiomyopathy

DCM is an extremely heterogeneous disease, as indicated by the heterogeneity of the mapped loci and genes for familial DCM (Table 55–10). The predominant mode of inheritance is autosomal dominant; however, autosomal recessive and X-linked DCM also occur. The most common causal gene for primary DCM is TTN, which encodes the giant sarcomere protein titin.¹⁸⁰ Several causal genes for autosomal dominant DCM have been identified. Several causal genes encode sarcomere proteins, which are also known to cause HCM. Thus, despite the contrasting phenotypes of HCM and DCM, mutations in sarcomere genes can cause either of the very different phenotypes. Because many of the known causal genes for DCM involve the myocyte cytoskeleton, DCM is mostly, but not exclusively, a disease of cytoskeletal proteins.

Causal Genes and Mutations

The gene encoding cardiac α-actin (ACTC) was the first causal gene identified for autosomal dominant DCM.¹⁸¹ Subsequently, mutations in genes encoding additional components of the sarcomere, namely MYH7, TNNT2, TTN, and TCAP, were found in patients with DCM. Using next-generation sequencing of a large DCM cohort, truncating mutations in TTN were found to be responsible for 25% of the cases of DCM.¹⁸⁰ Because mutations in ACTC, MYH7, and TNNT2 are also known to cause HCM, these findings point to the commonality of the genetic basis of DCM and HCM. The diversity of the phenotype may reflect the topography of the causal mutations on the protein as well as the genetic background of the individuals. The molecular basis of such extreme phenotypic diversity remains largely unknown.

Mutations in several cytoskeletal proteins, such as delta sarcoglycan, beta sarcoglycan, metavinculin, and dystrophin, are important causes of DCM.¹⁸² Mutations in alpha sarcoglycan (adhalin) cause an autosomal recessive form of DCM that occurs in conjunction with limb-girdle muscular dystrophy. Mutations in cysteine and glycine-rich protein 4 (CSRP3) and LIM domain-binding protein 3 (LDB3), members of muscle LIM proteins, also cause DCM.

Another intriguing category of genes causing DCM is genes encoding for ion channels, suggesting the presence of overlap syndromes between arrhythmias and heart failure. Mutations in ABCC9, which encodes the regulatory SUR2A subunit of the cardiac K(ATP) channel, as well as mutations SCN5A, which codes for a sodium channel and is a causal gene for Brugada syndrome, cause DCM.^183,184 A mutation in KCNQ1, encoding a potassium channel, is also associated with cardiac arrhythmias and DCM.¹⁸⁵ Cardiac arrhythmias are often the main presentation of this category of DCM and typically disproportionate to cardiac dysfunction. The term arrhythmogenic cardiomyopathy encompasses this group of DCM.

Mutations in the gene encoding lamin A/C (LMNA), which is a nuclear envelope protein, are important causes of DCM, typically in conjunction with conduction defect. LMNA interacts with large segments of chromatin. Consequently, LMNA mutations cause a diverse array of phenotypes including DCM, progressive conduction disease, atrial arrhythmias, SCD, muscular dystrophy, lipodystrophy, insulin resistance, and progeria.¹⁸⁶

A subgroup of DCM is caused by protein aggregation in the myocardium (proteotoxicity).^187,188 Mutations in the intermediary filament protein desmin and its associated protein alpha/B-crystallin cause DCM in part through protein aggregation in the myocardium.¹⁸⁸ Often, such mutations lead to a phenotype of cardiac and skeletal myopathy that is referred to as desmin-related myopathy.

Finally, mutations in RNA-binding motif protein 20 (RBM20), which regulates RNA spicing, cause DCM by influencing splicing of cardiac genes.^189,190 Collectively, these findings indicate the diversity of the causal genes responsible for DCM and render DCM genetically among the most heterogeneous diseases, with mutations affecting the integrity of the sarcomere and cytoskeleton being the main causes.

Genotype-Phenotype Correlation in Dilated Cardiomyopathy

Given the extreme genetic heterogeneity of DCM, the focus of genotype-phenotype correlation is on the presence of concomitant phenotype, cardiac or otherwise, that might offer clues to the underlying pathogenesis of the disease. Typically, mutations in genes encoding cardiac sarcomere proteins, such as MYH7, TNNT2, and TPM1 cause isolated forms of DCM without accompanying phenotypes, such as conduction defects, deafness, or skeletal myopathy. In contrast, mutations in the LMNA gene typically cause DCM in conjunction with progressive conduction defects, atrial arrhythmias, and SCD.¹⁸⁶ In addition, LMNA mutations also cause an autosomal-dominant form of Emery-Dreifuss syndrome.¹⁹¹ Mutations in desmin and alphaB-crystallin genes are commonly associated with skeletal myopathy as well as DCM with unique pathologic features, a phenotype referred to as the desmin-related myopathy.¹⁸⁷ Mutations in the dystrophin gene commonly lead to skeletal and cardiac myopathy. The severity of the myopathic phenotype is partly determined by the type of mutation. Those that are frame-shift mutations—for example, insertion or deletion of a single base—cause a severe form, whereas missense mutations often lead to a mild form of DCM and muscular dystrophy. Mutations in the 5′ region of the dystrophin gene are associated with DCM without skeletal involvement.¹⁹²

Cardiac involvement is quite common in triplet repeat syndromes and includes DCM, conduction disorders, and arrhythmia. Prevalence of cardiac involvement increases with advancing age, and approximately three-quarters of adult patients exhibit conduction defects, such as first-degree atrioventricular block and intraventricular conduction defects.¹⁹³ There is also a correlation between the severity of the disease and the severity of cardiac involvement and the number of CTG repeats.¹⁹³

Pathogenesis of Dilated Cardiomyopathy

A diverse array of mechanisms is likely to be involved in the pathogenesis of hereditary DCM. Among notable mechanisms are impaired sarcomere structure and function, cytoskeletal integrity and mechanical force transmission, and impaired ATPase activity.^182,194 Identification of mutations in MLP, LDB3, and TCAP emphasize the significance of the Z disk in maintaining normal cardiac function. Similarly, identification of mutations in the dystrophin-associated protein complex as causes of DCM signifies the role of sarcolemma in the pathogenesis of DCM. Mutations in the dystrophin gene lead to decreased expression levels of dystrophin, a major cytoskeletal protein in skeletal and cardiac muscles. Decreased levels of dystrophin are expected to impair mechanical coupling and myocyte shortening. Interestingly, using the CRISPR-Cas9 system to target the mutant dystrophin leads to increase dystrophin levels and improvement of muscle and cardiac phenotype.^{195,196,197,198}

Pathogenesis of DCM resulting from mutations in desmin and alphaB-crystallin involves deposition of desmin and alphaB-crystallin aggregates in the myocardium.¹⁸⁷ The molecular pathogenesis of DCM caused by mutations in lamin A/C or emerin remains largely unknown, but is likely to involve perturbed epigenetic regulation of gene expression as well as disruption of integrity of the cytoskeleton.¹⁸⁶ The pathogenesis of cardiomyopathies in patients with the triplet repeat syndromes is not fully known, but pertains to altered splicing of cardiac genes (multiple species of transcripts resulting from inclusion or exclusion of one or several coding exons).¹⁸⁶ It is evident that there is no unique mechanism, but multiple mechanisms are involved in the pathogenesis of DCM caused by various genes and mutations.

Genetic Basis of Arrhythmogenic Cardiomyopathy

Arrhythmogenic cardiomyopathy (AC), also known as arrhythmogenic right ventricular cardiomyopathy (ARVC), is an uncommon cardiomyopathy with characteristic clinical and pathologic features.^199,200,201 It is best defined as primary cardiomyopathy whose cardinal manifestation is cardiac arrhythmias, typically ventricular tachycardia, and it occurs disproportionate to cardiac dysfunction. This is contrast to ventricular tachycardia in DCM, which typically occurs in the presence of severely depressed cardiac function. The disease often has a “concealed” stage, which is characterized by minor ventricular arrhythmias and subtle pathologic findings. It is followed by symptomatic ventricular arrhythmias and gradual progression to right-heart failure and, finally, global cardiac failure. The prototypic form of AC is ARVC, which predominantly involves the right ventricle until in advanced stages, when the left ventricle is also involved.^201,202 The clinical phenotype comprises ventricular arrhythmias, primarily originating from the right ventricle, SCD, and heart failure. Involvement of the left ventricle typically portends an advanced stage and a poor prognosis.²⁰² In addition, left dominant AC is also described. The pathologic phenotype is characterized by the gradual replacement of the cardiac myocytes by adipocytes and fibrosis (Fig. 55–11).

Electrocardiographic features include the characteristic and yet uncommon epsilon wave, depolarization and repolarization abnormalities in the right precordial leads, and ventricular arrhythmias originating from the right ventricle.^203,204

AC is an important cause of SCD in young and apparently healthy individuals.^201,205,206 In the US population, it accounts for 3% to 5% of SCD associated with physical activity in the young athletes.¹⁰³ In some reports, AC was found in up to 25% of the cases of nontraumatic SCD.^207,208,209 Collectively, the data suggest ARVC is an important cause of SCD in young, competitive athletes.

Significant fatty infiltration of the myocardium could be present in normal individuals, particularly in the elderly. Cor adiposum (fatty infiltration of the myocardium) is distinguished from true ARVC by the absence of right ventricular myocardial thinning, myocyte atrophy and apoptosis, patchy fibroadipocytic replacement of myocytes, predominantly in the right ventricle, and lymphocytic myocarditis.²¹⁰ Right ventricular dilatation, fibrosis, myocyte atrophy, and excess adipocytes have been observed in patients with Becker muscular dystrophy, Emery-Dreifuss muscular dystrophy, and myotonic dystrophy. The distinction between muscular dystrophies and ARVC is usually not problematic because of the skeletal involvement in muscular dystrophies. Given the difficulty in the accurate diagnosis of ARVC, the Task Force committee recommends requirement of two major criteria, or one major and two minor criteria, or four minor criterial for the diagnosis of ARVC.²¹¹

Molecular Genetics of Arrhythmogenic Cardiomyopathy

Classic ARVC, which is a subtype of AC, is a primarily genetic disorder of intercalated disk proteins (Table 55–11). Mutations in several genes encoding protein constituents of intercalated disks, which include desmosomes, are established causes of ARVC. Mutations in DSP, JUP, PKP2, DSC2, and DSG2, encoding desmosomal proteins desmoplakin (DP), plakoglobin (PG), plakophilin 2 (PKP2), desmocollin 2, and desmoglein 2 (DSG2), are among the most common causes of ARVC (see Table 55–11). Mutations in PKP2 appear to be the most common causes of ARVC, accounting for approximately 25% of the cases. A significant number of PKP2 mutations cause a frameshift and hence are expected to lead to premature termination of the proteins. DSG2 and DSP genes each account for approximately 10% of the ARVC cases.

A subgroup of arrhythmogenic cardiomyopathy, because of expression of desmosome proteins in the skin and the heart, exhibits cardiocutaneous manifestations including keratoderma and woolly hair. The classic cardiocutaneous syndrome is Naxos disease (described first in a family from the island of Naxos in Greece), which is caused by a recessive truncating mutation in JUP, encoding junction protein plakoglobin.^212,213 Carvajal syndrome is another form of a cardiocutaneous syndrome wherein the left ventricle is the predominant site of involvement as opposed to the right ventricle in Naxos disease.^199,212 Carvajal syndrome is caused by mutations in the DSP gene.²¹²

Mutations in TMEM43 encoding transmembrane protein 43 are considered rare causes of classic ARVC.^214,215 The initially discovered mutation was considered fully penetrant and was associated with a high risk of lethal cardiac arrhythmias.²¹⁵ Not much is known about the structure and function of TMEM43 and the mechanisms by which it leads to AC, except that it appears to be a nuclear membrane protein with yet-to-be determined function.

ARVC also added to the diversity of phenotypic expression of the LMNA gene, which causes over a dozen distinct phenotypes.²¹⁶ Likewise, the gene encoding phospholamban (PLN) and Titin (TTN) are considered causal genes for AC, presenting as DCM and ventricular arrhythmias.^217,218

Point mutations in the 5′ untranslated region of the TGFB3 gene have been associated with ARVC.²¹⁹ However, the causal role of these variants in ARVC remains unsettled.

Idiopathic right ventricular outflow tract tachycardia and stress-induced (catecholaminergic) polymorphic ventricular tachycardia, caused by mutations in cardiac ryanodine receptor (RYR2), often present with arrhythmias resembling those in ARVC.^220,221 The absence of structural or histologic cardiac abnormalities suggests phenocopy and not true ARVC.

Pathogenesis of Arrhythmogenic Cardiomyopathy

The molecular pathogenesis of AC is largely unknown. The prevailing hypothesis implicates impaired mechanical integrity of myocyte-myocyte attachment as the instigator of activation of the mechanosensitive signaling pathways.^222,223 Among the pathways activated by impaired mechano-transduction is the Hippo pathway, which is activated in AC.²²⁴ Activation of the upstream molecules of Hippo pathways through cascade phosphorylation of the downstream molecules leads to phosphorylation and inactivation of the YAP molecules, which results in reduced gene expression through the TEAD transcription factor.²²⁴ YAP also interacts with beta-catenin, the effector of the canonical Wnt signaling pathway, leading to cytoplasmic sequestration of beta-catenin and reduced gene expression through the TCF7L2 transcription factor.^223,224 Additional mechanisms, such as partial translocation of membrane JUP to the nucleus, also further suppress gene expression through the canonical Wnt signaling.^223,225 Suppression of the canonical Wnt signaling in a subset of cardiac cells leads to their differentiation to adipocytes.

The heart is a cellular heterogeneous organ. The molecular events described above also occur in cells other than mature and terminally differentiated myocytes. The latter cells are not expected to transdifferentiate. Prevailing data suggest that a subset of cardiac mesenchymal cells, in addition to cardiac myocytes, express desmosome proteins and differentiate to adipocytes in AC.^225,226 The origin of fibrosis in AC and its molecular mechanisms are unknown.

Genetic Basis of Restrictive Cardiomyopathy

Restrictive cardiomyopathy (RCM) is a heart-muscle disease characterized by severely enlarged atria as a result of elevated right and left ventricular filling pressures, normal or reduced ventricular volumes, and usually preserved global systolic function.²²⁷ The clinical manifestations are those of heart failure, often with predominance of right-sided signs and symptoms. The age of onset of the disease is variable, and the prognosis is relatively poor. RCM can occur because of systemic infiltrative disorders, such as amyloidosis and sarcoidosis, and storage diseases such as Fabry disease.^227,228 Such disorders are also genetic in etiology, but their hemodynamic phenotype exhibits a restrictive physiology, and hence they phenocopy true RCM. Imaging, systemic tests, and pathology are needed to sometimes decipher these different forms of RCM.

Molecular Genetics of Restrictive Cardiomyopathy

RCM partly shares a genetic etiology with DCM and HCM, as mutations in the sarcomere gene could lead to either of the phenotypes.^{229,230,231,232,233} In addition, a subset of patients with HCM or DCM exhibits a predominantly restrictive hemodynamic physiology. Accordingly, mutations in MYH7, TNNT2, TNNI3, TTN, and MYPN cause RCM.^{229,232,233,234} Mutations in DES, encoding intermediary filament desmin, are also causes of familial RCM, occurring in conjunction with skeletal myopathy and atrioventricular conduction defects with an autosomal dominant pattern of inheritance.²³⁵ As discussed, the phenotype of restrictive hemodynamic physiology resembling RCM also occurs in various other conditions including Noonan syndrome, caused by mutations in the protein tyrosine phosphatase, nonreceptor type II.²³³

Genetic Basis of Cardiomyopathies in Triplet Repeat Syndromes

Cardiac involvement is common in neuromuscular disorders. A subset of neuromuscular disorders involves expansion of the naturally occurring GC-rich triplet repeats in genes and is referred to as triplet repeat syndromes.^{236,237,238,239} The group comprises more than 10 different diseases, including myotonic muscular dystrophy and Huntington disease. Cardiac involvement is common in several forms of triplet repeat syndromes and is a major determinant of morbidity and mortality.^238,239 The phenotype commonly includes DCM, HCM, conduction disorders, and arrhythmias. Average life expectancy of the affected individuals is about 30 to 40 years.

Genetic Basis of Cardiomyopathies in Myotonic Dystrophy

Myotonic dystrophy (DM) is an autosomal dominant disorder with highly variable penetrance.^238,239,240 The estimated prevalence of DM is approximately 1 in 8000 in the North American population.^238,239,240 It is the second most common muscular dystrophy after Duchenne muscular dystrophy. DM commonly manifests itself as progressive degeneration of muscles and myotonia, cardiomyopathy, conduction defects, male-pattern baldness, infertility, premature cataracts, and endocrine abnormalities.^238,239,240 Cardiomyopathy is a common phenotypic manifestation of myotonic DM.^238,239 Cardiac conduction defects, such as first-degree atrioventricular block and intraventricular conduction defects, are present in approximately three-quarters of adult patients.

Expansion of CTG (CUG in mRNA) repeats in the 3′ untranslated region of the DMPK gene, encoding dystrophica myotonica protein kinase, and is the underlying pathogenic mutation in myotonic dystrophy.²⁴¹ The number of CTG repeats in normal individuals varies between 5 and 37. It expands from 50 to more than several thousand in patients with DM.²⁴¹ The length of the CTG repeats often correlates with the severity of clinical phenotypes, including conduction defects and cardiomyopathy. Expansion of the repeats leads to folding of the RNA into hairpin-like structures and a number of biological events, including transcriptional, translational, and posttranslational changes. Among notable changes is sequestration of muscle blind-like protein-1 (MBNL1) and increased expression of ETR3-like factor-1 (CELF1), which are involved in regulating splicing of various genes, including genes encoding cardiac sarcomere proteins.^242,243,244

The second gene responsible for DM is zinc finger protein 9 (ZNF9), located on 3q21.²⁴⁵ Expansion of a CCTG tetranucleotide repeats in the first intron of ZNF9 leads to expression of abnormal RNA. The underlying mechanisms responsible for various phenotypic effects of the triplet repeat expansion in ZNF9 are less well understood.²⁴⁴

Genetic Basis of Cardiomyopathies in Friedreich Ataxia

Friedreich ataxia (FRDA) is an autosomal recessive neurodegenerative disease. It primarily involves the central and peripheral nervous system and less frequently manifests as cardiomyopathy and occasionally as diabetes mellitus.²⁴⁶ FRDA is caused by the expansion of the GAA trinucleotide repeats in intron 1 of FRDA.²⁴⁷ The encoded protein is frataxin, which is a soluble mitochondrial protein with 210 amino acids. Cardiac involvement can manifest as either DCM or HCM. The severity of clinical manifestations of FRDA also correlates with the size of the repeats. The pathogenesis of cardiomyopathies in FRDA is likely to involve impaired iron homeostasis, leading to iron accumulation in the mitochondria, and increased oxidative stress.²⁴⁶

Genetic Basis of Cardiomyopathies in Neuromuscular Disorders

Duchenne and Becker Muscular Dystrophies

The phenotype is characterized by progressive degeneration of skeletal and cardiac muscle function with an incidence of 1 in 3500 newborn males.²⁴⁸ It commonly manifests itself during the first or second decades of life in male patients as mild, but progressive, skeletal myopathy, early contractures, and cardiomyopathy. Female family members are commonly spared. However, they may exhibit a mild phenotype, typically late in life. Duchenne muscular dystrophy is a severe form and Becker muscular dystrophy a milder form of the disease. Patients with Becker muscular dystrophy typically show onset of muscular symptoms in the second or third decades of life. This is in contrast to patients with Duchenne muscular dustrophy, who typically exhibit neuromuscular symptoms within the first decade of life. Cardiac involvement includes progressive atrioventricular block, arrhythmia, loss of P-wave amplitude on the ECG, atrial standstill, DCM, akinesis/dyskinesis of the walls, particularly the posterobasal wall, and SCD.²⁴⁹ Often DCM is the primary feature of Duchenne and Becker muscular dystrophies. Approximately 90% of patients will eventually develop DCM. Death often occurs by the third decade of life.

The gene responsible for Duchenne and Becker muscular dystrophies is dystrophin, located on Xp21, which encodes a large cytoskeletal protein. Mutations, which are typically frame-shift variants, result in the absence of dystrophin.²⁵⁰ DCM can also occur in the absence of skeletal myopathy. Mutations leading to a frameshift induce a severe form, while missense mutations often lead to a mild form of the disease. Mutations in the 5′ region of the dystrophin gene can cause DCM without skeletal involvement.

The molecular mechanisms are diverse and include altered mechano-transduction and stretch-activated channels, calcium sensitivity, NO signaling, mitochondrial dysfunction, and oxidative stress, among others.²⁴⁹ A number of experimental approaches are used to induce expression of dystrophin; stop-codon read-through therapies, upregulation of utrophin, CRISPR-Cas9-mediated gene editing, and recombinant viruses have been used to increase expression of dystrophin and alleviate muscle pathology.²⁴⁹

Emery-Dreifuss Muscular Dystrophy

Emery-Dreifuss muscular dystrophy is an X-linked degenerative disorder characterized by mild, but progressive, skeletal and cardiac myopathy.²⁵¹ Clinical features include muscle weakness and atrophy, flexion deformities of the elbows, and mild pectus excavatum. Cardiac phenotypes include cardiomyopathy, arrhythmia, SCD, conduction defects, loss of P-wave amplitude on the ECG, and atrial standstill.

The first causal gene identified for Emery-Dreifuss muscular dystrophy is EMD, which encodes nuclear membrane protein emerin.²⁵² Likewise, mutations in LMNA, encoding nuclear membrane protein lamin A/C, also cause Emery-Dreifuss muscular dystrophy.²⁵³ Finally, mutations in FHL1, TMEM43, SYNE1, and SYNE2 have been associated with phenotypic variants of Emery-Dreifuss muscular dystrophy.^251,254

Barth Syndrome

DCM is a major phenotypic component of the Barth syndrome, an X-linked disorder characterized by skeletal and cardiac myopathy, neutropenia, and abnormal mitochondria. Barth syndrome is caused by mutations in TAZ encoding tafazzin (formerly known as G4.5), which is involved in fatty acid metabolism.²⁵⁵ The majority of the mutations are missense mutations, but frameshift and splice junction mutations have also been described. Mutations affect levels of unsaturated cardiolipin leading to mitochondrial dysfunction and oxidative stress, leading to muscle weakness and DCM.²⁵⁶

Genetic Basis of Cardiomyopathies in Storage Disorders

Cardiomyopathies observed in storage diseases encompass a group of disorders in which there is the primary metabolic abnormality in the heart (Table 55–12). This metabolic abnormality also may involve other organs; however, cardiac involvement is direct and not a consequence of secondary changes in other organs. Secondary involvement of the myocardium in systemic metabolic disorders is not included in this section. The phenotype typically resembles that of HCM or DCM caused by mutation in sarcomere proteins (phenocopy).

A prototype of storage cardiomyopathies is glycogen storage disease type II (glycogenosis type II or Pompe disease).²⁵⁷ Pompe disease is an autosomal recessive disorder caused by mutations in GAA gene, which encodes acid alpha-galactosidase. Mutations lead to deficiency of α-1,4-glucosidase (acid maltase), which degrades α-1,4 and α-1,6 linkages in glycogen, maltose, and isomaltose. Deficiency of the enzyme leads to storage of glycogen in lysosomal membranes. Phenotypic expression of Pompe disease includes HCM, DCM, conduction defects, and muscular hypotonia. A high-protein diet and recombinant acid α-glucosidase have been used effectively to treat this disorder. Enzyme replacement therapy with algucosidase alpha early in the course of the diseases has beneficial clinical effects.²⁵⁷

Fabry disease, which is a relatively common HCM phenocopy, was discussed under HCM. In brief, it is an X-linked lysosomal storage disease caused by mutations in GLA, which encodes α-galactosidase.^{121,122,123,124} Mutations lead to deficiency of α-galactosidase A and deposits of glycosphingolipids in multiple organs, including the heart. Enzyme replacement therapy with human agalsidase α or recombinant agalsidase β imparts beneficial effects.^{125,126,127,128}

Mutations in the gene encoding the AMP-activated γ2 noncatalytic subunit of protein kinase A (PRKAG2) cause a glycogen storage disease with phenotypic expression as HCM and WPW syndrome.^{98,129,130,131,132} AMP-activated kinase is a biosensor of the cellular energy state. Cardiac involvement varies from a predominant phenotype of preexcitation and conduction abnormalities to a predominant phenotype of cardiac hypertrophy.

Refsum disease is an autosomal recessive disorder characterized clinically by a tetrad of retinitis pigmentosa, peripheral neuropathy, cerebellar ataxia, and elevated protein levels in the cerebrospinal fluid.²⁵⁸ Cardiac involvement includes ECG abnormalities, which are common. Cardiac hypertrophy and heart failure are uncommon. Mutations in the gene encoding phytanoyl-coenzyme A (CoA) hydroxylase (PAHX or PHYH) are responsible for Refsum disease.^259,260 Mutations reduce the enzymatic activity and lead to accumulation of phytanic acid, an unusual branched-chain fatty acid in tissues and body fluids.²⁵⁸

The heart is also involved in patients with mucopolysaccharidosis, Niemann-Pick disease, and Gaucher disease. Cardiac phenotype entails cardiomyopathies, valvular disease, and obstructive coronary artery disease.²⁶¹ Table 55–13 lists various forms of mucopolysaccharidosis and the associated cardiovascular phenotypes.

Hereditary hemochromatosis is an autosomal recessive disease commonly caused by mutations in HFE gene on chromosome 6p12.3.²⁶² The encoded protein regulates iron absorption and leads to iron storage in various organs including liver and heart.²⁶² Cardiac involvement includes cardiac hypertrophy, heart failure, atrial fibrillation, and conduction defects.

Genetic Basis of Cardiomyopathies in Mitochondrial Disorders

Cardiomyopathies are common in patients with mitochondrial disorders. Mitochondrial cardiomyopathy exhibits a matrilineal transmission. Mitochondrial DNA (mtDNA) is a circular double-stranded genome of approximately 16.5 kb, encoding 13 polypeptides of the respiratory chain complexes I, III, IV, and V subunits; 2 ribosomal RNAs; and 22 tRNAs (transfer ribonucleic acids). Each mitochondrion has multiple copies of mtDNA, and each cell contains thousands of mitochondria. Nuclear genes encode for the vast majority of mitochondrial proteins that primarily regulate mitochondrial function. Thus, mutations in nuclear genes are the primary causes of mitochondrial myopathies. Examples include mutations in SCO2, FRDA, and ATP12 leading to HCM seen in patients with Leigh syndrome and Friedreich ataxia.²⁶³ Mutations in genes encoding mitochondrial oxidative phosphorylation pathways often result in a complex phenotype involving multiple organs, including the heart. Cardiac involvement can lead to hypertrophy as well as dilatation. Mitochondrial DNA mutations typically result in a significant degree of heteroplasmy, because of the presence of a very large number of mitochondria in each cell, and multiple copies of mtDNA in each mitochondria. Heteroplasmy increases over time as the mitochondria multiply. In general, the majority of the mtDNA must mutate in order to affect mitochondrial function and lead to a clinical phenotype.

Kearns-Sayre syndrome is a mitochondrial disease caused by sporadically occurring mutations in mtDNA. Kearns-Sayre syndrome is characterized by a triad of progressive external ophthalmoplegia, pigmentary retinopathy, and cardiac conduction defects.²⁶⁴ The classic cardiac abnormality in Kearns-Sayre syndrome is conduction defects; however, DCM and HCM are also rarely observed.²⁶⁴

l-Carnitine deficiency is a cause of mitochondrial myopathy resulting from mutations in nuclear DNA.²⁶⁵ The phenotype is characterized by skeletal myopathy, congestive heart failure, abnormalities of the central nervous system and liver, and rarely, HCM. Carnitine is an important component of fatty acid metabolism and is necessary for the entry of long-chain fatty acids into mitochondria. Mutations in the chromosomal gene encoding solute carrier family 22, member 5 (SLC22A5), also known as OCTN2 transporter, impair transport of carnitine to mitochondria and cause systemic carnitine deficiency.²⁶⁶ Similarly, mutations in genes encoding enzymes involved in the transfer and metabolism of carnitine can cause carnitine deficiency. The list includes carnitine mitochondrial carnitine palmitoyltransferase I (CATI or CPT-1), located in the outer mitochondrial membrane; carnitine-acylcarnitine translocase (SLC25A20), located in the inner membrane; and carnitine palmitoyl transferase 2 (CPTII).²⁶⁷ Mutations in genes coding for enzymes involved in carnitine biosynthesis, such as TMLHE and BBOX1, are also involved in carnitine deficiency.²⁶⁵

Mitochondria regulate fatty-acid oxidation as a major energy source in the heart. Mutations in proteins regulating fatty-acid beta-oxidation, such as CPTIII, MTP, and VLCAD, lead to lipid accumulation in cardiac muscle and heart failure, typically in conjunction with hypotonia, hypoglycemia, encephalopathy, hepatic dysfunction, skeletal myopathy, and SCD.²⁶⁷ Likewise, mutations in ABHD5 and PNPLA2, encoding alpha/beta-hydrolase domain-containing protein 5 and patatin-like phospholipase domain containing 2, respectively, lead to neutral lipid storage disease that occasionally involves the heart.²⁶⁷

Cardiac rhythm and conduction abnormalities could occur as the primary phenotypes of genetic disorders or secondary phenotypes resulting from genetic diseases that primarily affect structure of the heart. In general, cardiac arrhythmias and conduction defects result from abnormalities in three main families of proteins: contractile sarcomeric proteins, such as that in HCM; the cytoskeletal proteins, which are responsible for DCM; and the ion channels and their regulators, which are responsible for familial arrhythmias and conduction defects. As discussed earlier, there is significant phenotypic overlap as mutations in the same gene could cause a variety of cardiac rhythm and conduction disorders. This is best exemplified by mutations in sodium channel SCN5A, which could cause LQTS, Brugada syndrome, and familial conduction disease.²⁶⁸ Therefore, a simplistic classification of genetic disorders is considered preliminary and some of the key genetic findings used to risk-stratify for SCD or arrhythmias are based on studies in only few families. Table 55–14 summarizes the list of genetic disorders in which the primary phenotype is cardiac arrhythmias and conduction defects.

Ion Channelopathies as the Basis for Cardiac Arrhythmias and Conduction Defects

Ion channels, crucial units in cardiac excitability, are glycoproteins embedded in the membrane of the cardiac myocytes, which allow flux of ions in and out of the cell to modulate the electrical gradient. Many different ion channels and chaperone proteins, which maintain protein homeostasis, are activated in an orderly manner to give rise to the electrical current that will ultimately be responsible for the development of the myocyte excitability. This is a complex process and requires a very well controlled ionic balance to prevent arrhythmogenesis. Genetic defects in these proteins will be responsible for channelopathies, inherited diseases that may be associated with arrhythmias and/or sudden cardiac death.²⁶⁹

Brugada Syndrome and its Variants

Brugada syndrome is identified by a characteristic ECG pattern consisting of right bundle branch block and ST-segment elevation in V₁ to V₃ (Fig. 55–12) and sudden death at a young age. It was described originally in 1992 based on electrocardiographic pattern and occurrence of syncope or sudden death episodes in patients with a structurally normal heart.²⁷⁰ The episodes of syncope and sudden death are caused by fast polymorphic ventricular tachycardia.

Brugada syndrome usually manifests in subjects in the third or fourth decades of life, but may cause symptoms at any age, and even cause sudden infant death syndrome (SIDS).²⁷¹ In 1997, it was shown that Brugada syndrome was underlying some of the cases of SUDS (sudden unexpected death syndrome), which is prevalent in Southeast Asia.²⁷² SUDS is estimated to affect up to 1% of the population and it is the most common cause of death in young males in Thailand. Death often occurs at night and more commonly in male subjects (male-to-female ratio is 10:1). Electrocardiographically, the disease is identical to Brugada syndrome.²⁷³ As in Brugada syndrome, mutations in SCN5A have been found responsible for SUDS and biophysical data indicate a nonworking SCN5A or accelerated inactivation.

Genetics of the Brugada Syndrome

The α subunit of the cardiac sodium channel gene, SCN5A, was identified as the first causal gene for Brugada syndrome in 1998. SCN5A, which is responsible for phase 0 of the cardiac action potential, has been studied extensively during the past years. SCN5A was first cloned and characterized in 1995 and localized to 3p21. The gene is comprised of 28 exons that code for a 2016 amino-acid protein. It contains four homologous domains (DI to DIV), each of which contains six membrane-spanning segments (S1 to S6). In 1995, Wang and associates linked mutations in SCN5A to LQTS, a disease characterized by prolongation of the QT interval and sudden death at a young age.²⁷⁴ Subsequently, mutations in SCN5A were linked to idiopathic ventricular fibrillation, the Brugada syndrome, progressive conduction defect, SIDS, and SUDS. Important overlap exists between phenotypes, as it is not uncommon to observe several electrocardiographic features in a same family.²⁶⁸ These data suggest mutations in SCN5A cause variable phenotypic manifestations that span Brugada syndrome, long QT syndrome 3 (LQT3), and progressive conduction defects (see Fig. 55–12). Figure 55–13 shows the phenotypes arising from mutations in SCN5A.

In Brugada syndrome, SCN5A is responsible for 25% to 30% of familial cases. As in many other genetic disorders, Brugada syndrome also exhibits locus heterogeneity. Currently, more than 300 pathogenic variants have been identified in 19 genes (ABCC9, CACNA1C, CACNA2D1, CACNB2b, GPD1-L, HCN4, KCND3, KCNE3, KCNE5, KCNJ8, PKP2, RANGRF, SCN10A, SCN1B, SCN2B, SCN3B, SCN5A, SLMAP, and TRPM4) (Table 55–14), which affect sodium, potassium, and calcium currents.²⁷⁵ PKP2 mutations are also associated with Brugada syndrome, a finding that suggests structural proteins by decreasing I_Na and reducing the number of channels at the intercalated disc could be responsible for Brugada syndrome.²⁷⁶

Pathogenesis of Brugada Syndrome

Biophysical characterization of mutations in SCN5A suggests that mutations decrease the Na⁺ current availability by two main mechanisms: decreased expression of the mutant channel or acceleration of inactivation of the channel. In addition, the alteration in the ionic currents that worsen at higher temperatures has been implicated for certain mutations, such as the T1620M.²⁷⁷ The clinical relevance of this mechanism is corroborated by the observation of several cases of ventricular fibrillation during febrile illnesses in patients with Brugada syndrome.²⁷⁸ Compared to LQT3, the pathogenesis of Brugada syndrome could be considered a mirror image. Biophysical data indicate that LQT3 mutations cause a delayed inactivation of the channel, which is exactly the opposite as in Brugada syndrome, where there is an accelerated inactivation.²⁶⁸

How Na⁺ current alteration causes elevation of the ST segment in the right precordial leads is still a matter of controversy. Two main hypotheses are considered in the pathogenesis of Brugada syndrome: the repolarization versus depolarization hypotheses.²⁷⁹

In the repolarization hypothesis it is claimed that mutations in SCN5A in patients with Brugada syndrome will have decreased availability of sodium ions, which could shift the ionic balance in favor of transient outward potassium current (I_to) during phase 1 of the action potential. This disequilibrium will generate a shortening of the subepicardial action potential, with elevation of the ST segment in the right precordial leads, and cause a susceptibility to phase 2 reentry. This current alteration can be better seen in the right ventricular outflow tract (RVOT), as it seems that this is the area where the equilibrium is more critical.

On the other hand, the depolarization hypothesis suggests that the substrate for arrhythmogenicity is right ventricular conduction delay, caused by the genetic defects in SCN5A.²⁸⁰ Evidence towards the depolarization hypothesis has been accumulating in these past years with the detection of abnormal late potentials in the free wall of the RVOT epicardium, with the interstitial fibrosis without transmural repolarization differences, in an explanted heart from a patient with Brugada Syndrome, or with the increased fibrosis, more severe in aging males, affecting all cardiac regions, but especially the right ventricle in the SCN5A knock-out animal model.^281,282 Strong evidence favoring the depolarization hypothesis has been made available in recent years with a new form of treatment: epicardial ablation of the RVOT epicardium, which has resulted in disappearance of the ECG pattern, inducibility during electrophysiological studies, as well as decrease in event recurrence. This technique, only performed at present in patients with an implanted defibrillator, has the potential to significantly alter the current therapeutic approach in Brugada syndrome.^283,284

Genotype–Phenotype Correlation in Brugada Syndrome

Interestingly, limited data are available regarding the correlation between genotypes and phenotype in patients with Brugada syndrome. This may partly reflect very recent identification of the first causal gene, phenotypic variability of SCN5A, and allelic and locus heterogeneity of Brugada syndrome. It has been suggested that electrocardiographic parameters, such as longer conduction intervals (PQ and HV) on baseline ECG, could distinguish the carriers of sodium channel mutations from the noncarriers.²⁸⁵

In the past years, polymorphisms have been acquiring greater importance to explain certain phenotypes of genetic diseases. In the SCN5A locus, the common H558R polymorphism has been shown to restore (at least partially) the sodium current impaired by other simultaneous mutations causing either cardiac conduction disturbances or Brugada syndrome. Thus, this polymorphism seems to alter clinical phenotype by modulating the effect of nearby mutations.^{286,287,288,289} GWAS have shown that cardiac conduction may be modulated by genetic polymorphisms in SNC5A and SCN10A and influence risk of cardiac arrhythmia.²⁹⁰ However, the role of SCN10A as a causative gene in Brugada syndrome appears to be minimal.²⁹¹

Long QT Syndrome

LQTS is a disease of ventricular repolarization identified by the prolongation of the QT interval on ECG. The diagnosis of the disease is based on the presence of a QTc > 480 msec, a LQTS score > 3 (a scoring system ranking from 1 to 9, based on ECG findings, clinical history, and family history), or the presence of a confirmed pathogenic LQTS mutation, irrespective of QT duration.²⁹²

LQTS has a prevalence of 1/2500 to 1/5000 and it is characterized by syncopal episodes, malignant ventricular arrhythmias, and ventricular fibrillation. The majority of patients with the LQTS are asymptomatic. However, approximately one-third present with syncope or aborted malignant ventricular arrhythmias including torsades de pointes, which is the most typical ventricular arrhythmia in LQTS. SCD is relatively common. Prognosis of the symptomatic cases, if untreated is poor. The annual rate of SCD in untreated patients is estimated at approximately 0.9%.²⁹³ In syncope patients, the rate rises to 5%.²⁹⁴

The LQTS is either acquired, which is iatrogenic and commonly induced by drugs, or congenital. A common cause of the acquired disease is the use of medications such as antiarrhythmics, antidepressants, and phenothiazines (Table 55–15). In addition, electrolyte imbalance, such as hypokalemia, hypomagnesemia, and hypocalcemia, especially in the presence of predisposing medications, could cause LQTS.

Two patterns of inheritance have been described in the congenital LQTS: (1) autosomal recessive disease, described by Jervell and Lange Nielsen in 1957, which is associated with deafness; and (2) autosomal dominant disease, described by Romano and Ward, which is not associated with deafness and is more common than the recessive form.²⁹⁵

LQTS is caused by mutations in several genes, which are responsible for Na⁺, K⁺, and Ca²⁺ currents. The most common forms, however, are defects that cause K⁺ and Na⁺ current alterations. The pathogenesis of LQTS could be summarized as mutations in K⁺ channel resulting in inadequate opening and decreased potassium outward currents, while mutations in Na⁺ channels lead to inadequate closure of the channels and excessive sodium inward currents. The ensuing result is inadequate maintenance of electrical gradient (loss of function) during an action potential and prolongation of the QT interval. Pathogenesis of LQTS caused by CACNA1C mutations (Timothy syndrome) involves delayed channel closing, prolonged inward current through L-type calcium channels, and QT prolongation.

Autosomal Dominant Long QT Syndrome (Romano-Ward Syndrome)

In the past 25 years, at least 18 genes have been identified as being responsible for the LQTS (see Table 55–14). The known genes are responsible for 80% to 85% of cases. All encode proteins that are responsible for automaticity of the electrical activity in the cardiac cells. Mutations in these genes disrupt electric currents, alter cardiac action potential, and create a voltage gradient especially at the ventricular level, leading to reentrant arrhythmias. Three genes are responsible for the majority (60%-70%) of LQTS. These will be described in the next sections.

Long QT Syndrome 1

The causal gene for long QT syndrome 1 (LQT1) is KVLQT1 (or KCNQ1), which encodes a voltage-gated potassium channel α subunit and is strongly expressed in the heart. It consists of 16 exons spanning 400 kb, which form 6 transmembrane segments. It coassembles with the β subunit min K (KCNE1) to form the slow-activating potassium current I_Ks. Mutations in this gene disrupt the normal function of the protein, causing a decrease in the potassium current. Several mutations have been described in KCNQ1 to date.

Long QT Syndrome 2

The long QT syndrome 2 (LQT2) gene is KCNH2 (or HERG, “human ether-a-go-go related” gene), which was isolated in 1994 from hippocampus and named human ether-a-go-go related gene because of its homology to the Drosophila “ether-a-go-go” gene. The gene is localized on chromosome 7q35-q36. It contains 16 exons spanning approximately 55 Kb of genomic sequence. It encodes a protein that forms six transmembrane segments. The protein is responsible for the rapidly activating delayed rectifier potassium current I_Kr after coassembly with MIRP1 (KCNE2). As in the case of LQT1, mutations in HERG cause an abnormal protein with a resulting loss of potassium current.

Long QT Syndrome 3

The causal gene is SCN5A, located on chromosome 3, which encodes for the cardiac sodium channel. Mutations in SCN5A also cause Brugada syndrome and progressive conduction system disease, as discussed earlier. Electrophysiologic studies following expression of the mutant proteins in xenopus oocytes indicate a gain-of-function mutations evidenced by delayed inactivation and persistent sodium influx after phase 0 of the action potential.²⁶⁹

Other genes have been described, affecting sodium, potassium or calcium currents, but they are responsible for a low percentage of cases. In addition, some of these long QT cases will present with extracardiac features. The other genes are described in the next two sections.

Long QT Syndrome 7

Andersen syndrome is characterized by the constellation of periodic paralysis, cardiac arrhythmias, long QT, and dysmorphic features such as short stature, scoliosis, clinodactylism, hypertelorism, lowset or slanted ears, micrognathia, and broad forehead. The causal gene is KCNJ2, located on chromosome 17q23. It encodes the inward rectifier potassium channel Kir2.1, expressed in skeletal and cardiac muscles. Kir2.1 is a strong inward rectified channel that prevents passage of any current at potential greater than 40 mV. Electrophysiologic studies indicate the mutant protein exerts a dominant negative effect on Kir2.1 function with an ultimate decrease in potassium current.

Long QT Syndrome 8

Long QT syndrome 8 (LQT8), also known as Timothy syndrome, is characterized by the presence of facial dysmorphic features, syndactyly, small teeth, mental retardation, and severe QT prolongation. Mutations in CACNA1C, encoding for the α subunit of the L-type calcium channel, have been identified as responsible for the syndrome. The mutations cause a gain of function defect, increasing the inward current and prolonging the action potential.²⁹⁶

Other genes have been described affecting sodium, potassium, or calcium currents.

Autosomal Recessive Long QT Syndrome (Jervell and Lange-Nielsen Syndrome)

The autosomal recessive forms of the LQTS have been linked to mutations in the genes encoding I_Ks current, namely KCNQ1, KCNE1, and TRDN.²⁹⁷

For the LQTS recessive phenotype, which in KCNQ1 and KCNE1 is also associated with sensorineuronal deafness, the patients must inherit a mutation from both parents. Consequently, it is less common than the Romano-Ward syndrome, but is associated with a more malignant course and a longer QT interval. The phenotype could also arise in recessive forms when different mutations in the same gene are inherited from the parents (compound heterozygote).²⁹⁸

Management in Long QT Syndrome

There are several recommendations in patients with LQTS. Given the role of external factors in affecting the QT interval, patients with LQTS are advised to avoid QT-prolonging drugs, electrolyte abnormalities, and genotype-specific triggers for arrhythmias. On the other hand, all LQTS patients, irrespective of genotype and symptoms, are advised to be treated with beta-blockers. Silent genetic carriers have a risk of cardiac events estimated at 10% until 40 years of age; thus beta-blockers are also indicated in his population. Implantable cardioverter defibrillator (ICD) implantation should be considered in patients with previous cardiac arrest, in patients with syncopal episodes despite beta-blocker therapy, and in asymptomatic patients with a QTc > 500 msec. Left cardiac sympathetic denervation can be considered in patients who remain symptomatic. Finally, sodium channel blockers can be considered in patients with LQT3.

Genotype–Phenotype Correlation in LQT Syndrome

Given the availability of a large number of families with the LQTS, several genotype–phenotype correlation studies have been performed to identify the genetic determinants of triggering events, electrocardiographic phenotype, and response to therapy. The studies predominantly, encompass the three most common forms of LQTS—LQT1, LQT2, and LQT3—and have significant limitations inherent to genotype–phenotype correlation studies described earlier. Despite their limitations, characteristic features have emerged that could guide the analysis of the patients toward a specific genetic defect (Table 55–16).

Triggering Events

Characteristic features have emerged that help in part guide the analysis of patients towards a specific genetic defect. In general, individuals with LQT1 exhibit symptoms during physical activity, especially in swimming activities. Individuals with LQT3 are symptomatic during sleep.²⁹⁹

Electrocardiographic Phenotypes

LQT1 patients have a T wave that usually begins just after the QRS, becoming long and broad based. LQT2 patients have a small or notched T wave and LQT3 patients show a very late T wave with a prolonged ST segment.³⁰⁰

Clinical Phenotypes According to Genotype

Mutations also carry prognostic significance, and in all three groups (LQT1, 2, and 3) there is a correlation between cardiac events and the QT interval. In general, patients with LQT1 and LQT2 have a higher risk of cardiac events than patients with LQT3. The latter, despite having fewer events, have a relatively higher mortality, which indicates higher lethality of the events.³⁰¹

Biophysical Phenotypes

There have been few studies that have assessed the biophysical effect and location of the mutation and correlated it with the severity of the phenotype. Pore mutations in KCNH2 and transmembrane KCNQ1 mutations are associated with a higher incidence of events.^302,303

Response to Therapy

In addition, response to drug therapy seems to correlate with the genotype (class IIa). Beta-blockers are considered the first line of therapy in LQTS, despite previous work that indicated that LQT3 did not benefit as much from this medication. Preliminary data suggest LQT3 patients might benefit from Na⁺ channel blockers, such as mexiletine (class IIb).

Induced or Acquired Long QT Syndrome

Induced or acquired LQTS is iatrogenic, caused by a long list of medications (see Table 55–15) and electrolyte abnormalities. A large number of factors determine the risk of developing LQTS in an individual in response to drug therapy. They include bioavailability of the drug, the interaction with other medications that affect the same repolarizing current, and the presence of SNPs. SNPs play a major role in determining pharmacodynamics and pharmacokinetics of drugs and thus the risk of LQTS. The final effect on the repolarization will depend on the so-called repolarizing reserve, or the degree of alteration that the ionic currents can sustain before repolarization is compromised. Any combination of genetic and environmental factors (drug, electrolyte abnormalities) that decrease this “repolarization reserve” below a safe threshold will place the individual at risk of arrhythmia.³⁰⁴

Short QT Syndrome

The short QT syndrome (SQTS) is a disease characterized by the presence of shortening of the QT interval on ECG and clinically by episodes of syncope, paroxysmal atrial fibrillation, and/or life-threatening cardiac arrhythmias. SQTS usually affects young and healthy individuals with no structural heart disease. It may be present in sporadic cases, as well as in families. It was originally described in 2000.³⁰⁵ In 2003, a link was provided between the SQTS and familial sudden death with the first clinical report of two families with SQTS and a high incidence of SCD.³⁰⁶

Clinical Manifestations

Most patients with SQTS have a history of familial sudden death and/or atrial fibrillation, short refractory periods, and inducible ventricular fibrillation at electrophysiologic study. The age at onset of clinical manifestations could be extremely young. Malignant forms of SQTS responsible for neonatal SCD have been attributed to SIDS. The severity of the clinical manifestations of short QTs is highly variable, ranging from asymptomatic to atrial fibrillation, and recurrent syncope to sudden death.³⁰⁷

The characteristic sign for the disease is the presence of a very short QT interval on ECG. The T wave remains upright and the interval between the peak and end of the T wave is not prolonged. The appearance of a well-separated U wave has also been reported in several cases. It is difficult to define the normal QT interval, as the correcting equations have several limitations. Nevertheless, at a heart rate of 60 beats/min, the uncorrected QT interval is usually higher than 360 milliseconds. Clinical guidelines recommend making the diagnosis of SQTS with a QTc of < 340. In addition, SQTS is diagnosed in patients with a QTC < 360 ms and a confirmed pathogenic mutation, a family history of SQTS, and a family history of sudden death at < 40 years of age or after survival of ventricular tachycardia/ventricular fibrillation episode in the absence of structural heart disease.

Molecular Genetics and Electrophysiology

Genetics and biophysical analysis have provided important information regarding the pathophysiologic mechanisms that cause SQTS. SQTS, like the majority of primary familial electrical diseases, is caused by mutations in genes encoding cardiac ion channels. Three genes have so far been discovered, thereby proving that the disease is genetically heterogeneous. These genes will be described in the next sections.

Short QT Syndrome 1

KCNH2 expresses a protein that makes up the channel responsible for the rapidly activating outward K⁺ current involved in phase 3 repolarization. The protein is often referred to as HERG. The first genetic basis for the disease was obtained with the identification of two different missense mutations in the same residue in KCNH2 in three unrelated families. Both mutations resulted in the substitution of asparagine for lysine at codon 588 (N588K), an area at the outer mouth of the channel pore.³⁰⁸

Analysis of the current generated after transferring the mutated channels into human mammalian cells showed that the mutation abolished the inactivation of the channels, thus resulting in an increased developing current. The biophysical analysis therefore showed that the mutation induced a “gain of function” in I_Kr current, thus causing a shortening of the action potential.

The presence of paroxysmal atrial fibrillation in some affected individuals, and especially in one of the families as the only clinical manifestation, suggested that the increased heterogeneity would also be present at the atrial level and could be responsible for the arrhythmia.³⁰⁹

Short QT Syndrome 2

The KCNQ1 gene encodes a subunit of the proteins responsible for the slowly activating delayed outward K⁺ current. A missense mutation in this gene causing SQTS was first identified by Bellocq and colleagues in a 70-year-old individual who suffered ventricular fibrillation and had a QT interval of 290 ms after resuscitation. He was not inducible at electrophysiologic study and had no cardiac structural abnormalities.³¹⁰ A second mutation in KCNQ1 has been identified in several cases of fetal bradycardia, atrial fibrillation, and SQTS.³¹¹

Biophysical analysis showed that both mutations were leading to a gain of function in the outward current that explains the SQTS phenotype.

Short QT Syndrome 3

The third form of SQTS has been linked to mutations in the KCNJ2 gene, which codes for the channel protein responsible for the inward rectifier current (a current that moves inside the cells to stabilize the resting membrane potential). The proband and her father with the mutation both displayed short QT intervals of 315 and 320 ms, respectively. ECG recordings showed asymmetrical T waves with an abnormally rapid terminal phase. When expressed in Chinese hamster ovary cells, the mutated channels generated electrical currents that did not rectify (decreased) as much as the normal channels in their functional positive range of potentials (80 to 30 mV). Such a range of voltages corresponds to the very end of phase 3 repolarization and to phase 4. Simulation of the effects of the mutated channels on the morphology of the ventricular action potential showed a selective speeding of late repolarization, thus shortening significantly the action potential duration at 90% repolarization.³¹²

Finally, pathogenic variants in CACNA1C and CACNB2 have been associated with overlapping of Brugada Syndrome and a shortened QT interval.³¹³

Phenotype–Genotype Correlation

Robust phenotype–genotype correlation data are not yet available for SQTS. The disease is clinically highly heterogeneous, as indicated by the fact that, in the three families with the same mutation, a tremendous variation exists in symptoms and presentation.

Treatment Strategy of Short QT Syndrome

Preliminary data show that there may be effective pharmacologic therapy for this disease (class IIb). However, the high incidence of SCD warrants the implantation of an intracardiac cardioverter/defibrillator, especially in individuals with aborted sudden death.

Because QT shortening is likely caused by an increase in the outward current, it was suggested that blocking the current with class III antiarrhythmic drugs (known to increase the QT interval) could be a potential therapeutic approach for the treatment of SQTS (class IIb). Gaita and coworkers showed that treatment with quinidine prolonged the QT interval, decreased inducibility, and therefore had the potential to be an effective therapy for these patients (class IIb).³¹⁴ Clinical follow-up in one family with paroxysmal atrial fibrillation indicates that the episodes respond well to treatment with the class Ic agent propafenone (class IIb).

Progressive Familial Heart Block

Familial heart block is autosomal dominant progressive disease of cardiac conduction system characterized by initial development of bundle branch block and gradual progression to complete heart block. Two forms have been recognized. In type I, the onset is early and the disease is rapidly progressive. In type II, the onset is later in life and commonly the QRS complex is narrow and atrioventricular (AV) nodal block predominates. Clinical features of the disease include syncope, SCD, and Stokes-Adams attacks (sudden syncope related to paraoxysmal or chronic AV block, sinoatrial block, or paroxysmal tachycardia or fibrillation). A locus was identified in a large family of Portuguese descent on chromosome 19q13. The gene TRPM4 was identified in 2010 in three different families.³¹⁵ Immunohistochemistry showed that TRPM4 channel signal level was highest in Purkinje fibers, with mutations resulting in a gain of function. As discussed earlier, mutations in SCN5A have also been shown in some families with familial heart block. In addition, AV block in conjunction with congenital heart disease, such as ASD (NKX2.5 mutations), and DCM (lamin A/C mutations) have been described; these were discussed earlier.^316,317

Catecholaminergic Polymorphic Ventricular Tachycardia

Ryanodine receptors are responsible for release of calcium from the sarcoplasmic reticulum and are activated by the incoming calcium; therefore, they are Ca²⁺-activated Ca²⁺ channels.

Familial polymorphic ventricular tachycardia is an autosomal-dominant inherited disease with a mortality of approximately 30% by the age of 30 years. Phenotypically, it is characterized by runs of bidirectional and polymorphic ventricular tachycardia in response to vigorous exercise in the absence of evidence of structural myocardial disease. The prevalence of the disease is estimated at 1/10,000. An added difficulty in the diagnosis is that the basal electrocardiogram is usually normal, and patients have to be diagnosed with an electrocardiogram during exercise. The value of this test is very dependent on the families, as in some cases it does not reach a diagnosis of 50% of genetic carriers.³¹⁸

Currently, six genes have been associated with the disease, following an autosomal dominant (CALM1, CALM2, KCNJ2, and RyR2), and autosomal recessive (CASQ2 and TRDN) pattern of inheritance (see Table 55–1). A comprehensive genetic study identifies pathogenic variants in nearly 60% of all clinically diagnosed cases. The main gene responsible for catecholaminergic polymorphic ventricular tachycardia (CPVT) is RyR2 (CPVT type 1). It is responsible for nearly 50% of all cases. Thus, all other genes together are responsible for nearly 10% of CPVT cases.³¹⁹

Therapy in CPVT is based on lifestyle modifications (avoiding competitive sports, strenous exercise, and stressful environments), beta-blocker and flecainide therapy, and ICD implantation and left cardiac sympathetic denervation in those who remain symptomatic despite medical therapy.

Sick Sinus Syndrome

Sick sinus syndrome is characterized by the occurrence of sinus bradycardia, sinus arrest, and chronotropic incompetence. Sinus node dysfunction has been linked to loss of function mutations in HCN4, and in recessive form to SCN5A.³²⁰ HCN4 contributes to native f-channels in the sinoatrial node, the natural cardiac pacemaker region. In 2006, a loss of function defect in HCN4 was also linked to familial sinus bradycardia.³²¹

Familial Atrial Fibrillation

Atrial fibrillation is the most common sustained rhythm disorder or arrhythmia encountered in clinical practice, with a prevalence of 1% in the general population, which increases with age to about 6% in people over the age of 65 years. It is responsible for over one-third of all cardioembolic episodes.

Atrial fibrillation was not generally appreciated to be a familial inherited disorder. Atrial fibrillation was first reported as a familial form in 1943. In later years, published studies have shown several lines of evidence supporting the genetically determined heritable basis to atrial fibrillation. A study by Darbar and associates indicated that the familial form of the disease may have a higher prevalence than previously suspected and emphasized the importance of expanding genetics studies. Familial atrial fibrillation may be a monogenic disorder and is often identified as atrial fibrillation being present in many members of the same family. The first locus for familial atrial fibrillation was identified in 1997, in 10q22.³²² Several other loci and genes have been identified since then.

Atrial Fibrillation Associated with Genetic Alterations in Potassium Currents

Most of the single-gene mutations that have been discovered in families with atrial fibrillation cause cardiac K⁺-channel defects. The first gene responsible for familial atrial fibrillation, KCNQ1, was identified in 2003, linking the disease to an ion channelopathy.³²³ KCNQ1 encodes the pore-forming α-subunit of the cardiac slow delayed-rectifier (I_Ks) channel and its loss of function had been previously associated with LQTS. The analysis of KCNQ1 identified a missense mutation resulting in the amino acid change S140G. Electrophysiological studies revealed a gain of function in I_Ks current when the mutated channel was expressed with the β-subunits KCNE1 and KCNE2. This gain-of-function explained well the shortening of the action potential duration and effective refractory period, which are thought to be the culprits of the disease. In an interesting new finding, a gain-of-function mutation in codon 141, next to the one described in the above-mentioned family, was found responsible for a severe form of atrial fibrillation in utero and SQTS.³²⁴

Identification of mutations in KCNE2, KCNJ2, KCNE5, and KCNA5 further confirmed the role of genetic defects in potassium channels in the etiology of familial atrial fibrillation.^{325,326,327,328}

Atrial Fibrillation Associated with Genetic Alterations in Sodium Current

The α-subunit of the cardiac sodium channel gene, SCN5A, responsible for phase 0 of the cardiac action potential, has been studied extensively. Because of its key role in phase 0 of the cardiac action potential, SCN5A is involved in several primary arrhythmia syndromes. Gain-of-function mutations, mainly caused by the inability to inactivate, have been associated with LQTS type 3, and loss of function has been associated with Brugada syndrome, SIDS, Lev-Lenègre syndrome, SUDS, and DCM. Variants in SCN5A have been linked with lone and familial atrial fibrillation, and also associated with LQTS.⁴⁰³ In recent years, loss-of-function mutations in beta subunits, SCN1B, SCN2B, SCN3B have been identified in patients with atrial fibrillation. These findings further support the hypothesis that decreased sodium current enhances atrial fibrillation susceptibility.^329,330

Atrial Fibrillation Associated with Genetic Mutations in Genes Other Than Ion Channel Genes

Atrial natriuretic peptide precursor (NPPA) encodes atrial natriuretic peptide (ANP). ANP modulates ionic currents in cardiac myocytes and can play a role in shortening of the atrial conduction time, which could be a potential substrate for atrial reentrant arrhythmias. In 2008, Hodgson-Zingman and coworkers identified a frame-shift mutation in NPPA in a large family with atrial fibrillation.³³¹

Zhang and associates identified a new gene associated with a clinical phenotype characterized by a neonatal onset, with autosomal recessive inheritance.³³² They identified a mutation in NUP155, which encode a member of the nucleoporins. Although still unknown, the mechanism by which NUP155 may be associated with atrial fibrillation could be related to modulation of calcium-handling proteins and ion channel and expression of its possible target genes, like HSP70. The mutation was associated with inhibition of both export of HSP70 mRNA and nuclear import of Hsp70 protein, indicating that loss of function of NUP155 caused the disease by altering mRNA and protein transport. This gene is located in 5p13 and had been also associated with sudden death in the family.

Somatic Mutations in Atrial Fibrillation

In 2006, three missense mutations in GJA5 were identified in atrial tissue specimens from patients with idiopathic atrial fibrillation.³³³ GJA5 encodes the gap junction protein connexin 40, which is involved in electrical coupling and its disruption may cause atrial arrhythmias. In addition, atrial tissue genetic mosaicism was detected in a patient with a loss-of-function Cx43 mutation associated with atrial fibrillation.³³⁴

Familial Wolff-Parkinson-White Syndrome

Familial WPW syndrome is a rare syndrome with autosomal dominant mode of inheritance. It occurs in isolation or in conjunction with other disorders, such as HCM and Pompe disease. It is characterized by evidence of preexcitation on electrocardiogram, palpitation, and syncope as a result of supraventricular arrhythmias. The phenotype of WPW in conjunction with HCM and conduction defect was found in patients with mutations in PRKAG2, as discussed earlier. WPW has also been reported in patients with Pompe disease caused by mutations in α-1,4-glucosidase, in patients with HCM caused by mutations in TNNI3 and MYBPC3, and in Leber hereditary optic neuropathy, which is caused by mutations in mitochondrial DNA.³³⁵

The results of modifying lifestyle and drug intervention have documented the importance and the effect of both primary and secondary prevention in common polygenic disorders such as CAD and related traits. Drug therapy alone has been shown repeatedly in multiple clinical trials to decrease the incidence of clinical events from CAD by at least 30% to 40%.^336,337,338 Despite the reduction, cardiovascular disease remains the number one cause of death for both men and women in the United States.³³⁹ Continued health-care burden of CAD mandates new approaches to diagnosis and treatment of patient with CAD, including addressing the genetic basis of this complex phenotype. It is estimated that at least 50% of susceptibility for CAD is the result of a genetic component.³⁴⁰ Furthermore, most of the risk factors, such as hypertension, hypercholesterolemia, and obesity, have an inherent significant genetic component. Identification of the genes that predispose to CAD and related traits may be necessary for comprehensive prevention of these traits and CAD. Secondly, elucidation of the molecular pathways through which these genes mediate their effects should provide targets for development of new therapies. This is best exemplified by the case of PCSK9, which was discovered through genetic linkage analysis in a family with autosomal dominant hypercholesterolemia.³ The discovery led within a period of ~10 years to development of monoclonocal antibodies that target the PCSK9 protein and lower plasma low-density lipoprotein cholesterol (LDL-C) levels markedly.^4,5

Heritability of Coronary Artery Disease

CAD is a heritable trait influenced by a number of environmental factors. Heritability is defined as the proportion of inter-population variation in susceptibility to a disease that is caused by genetic factors. Twin studies clearly illustrate the marked influence of genetic factors in susceptibility to CAD and death from coronary heart disease.³⁴¹ Likewise, family studies show that those with a family history of heart disease in first-degree relatives are at a two- to threefold higher risk of premature CAD.^342,343,344 Early-onset CAD has a greater genetic component, with a heritability estimate attributed to nonlipid risk factors of greater than 50%.³⁴⁵ There is also a clustering of susceptibility to CAD in families that have risk factors associated with abnormalities such as lipid metabolism, hypertension, diabetes, and obesity, indicating a genetic basis for these conditions and risk factors.^346,347

Approach to Genetic Studies of Complex Diseases Including CAD

Common polygenic disorders, such as CAD, are in part caused by interactions of a large number of genetic variants and the environmental factors, each contributing modestly to the risk of CAD. The most commonly used approach to identify the susceptibility variants is referred to as a GWAS. A GWAS commonly has a case-control study design comprised of several thousand individuals mapped for several hundred-thousand to millions of SNVs spanning the genome.^17,348 Population allele frequencies are calculated in each group and compared between cases and controls to identify variants associated with the phenotype. Because a very large number of SNVs are analyzed at a much higher significance threshold, typically a P value of 5×10^-8 is set to reduce the high incidence of false-positive associations. A very large sample size of the GWAS not only reduces the chance of spurious association resulting from inadequate population stratification, but also enables detecting modest effect sizes of the variants.

GWAS by design analyzes known variants and typically those that are common, defined as a population frequency of 1% or greater. In addition, GWAS typically identifies variants that are in linkage disequilibrium with the true risk allele. Moreover, the effect sizes of the common variants are rather modest, typically accounting for a very small fraction of the risk.^17,348 To complement these shortcomings, deep sequencing approaches are used to identify the rare variants, true risk variants, and those that might potentially impart larger effect sizes.

GWAS has led to identification of over 100 susceptibility loci for CAD, the majority of which has been validated (http://www.ebi.ac.uk/gwas/home). However, collectively, loci identified through GWAS explain less than 10% of heritability of CAD, indicating a large number of yet-to-be-identified loci and genetic factors.³⁴⁰ The so-called missing heritability might be in part inherent to the design of GWAS, which analyzes only the common variants and the small effect of the risk variants. Sequencing-based approaches would enable identification and analysis of rare susceptibility variants, which typically exert larger effect sizes, and may be in part responsible for susceptibility for CAD.³⁴⁹ The “missing heritability,” however, might have several other components, including overestimation of the heritable etiology of coronary atherosclerosis and the presence of stochastic and typically nonlinear interactions among multiple etiological determinants including genetics, genomics, and environmental factors.³⁴⁹ It is important to note that the modest effect size of the risk alleles mapped for CAD through GWAS do not detract from the power of the discoveries in providing new insights into the pathogenesis of CAD and hence identifying new preventive and therapeutic targets. For example, variants in HMGCR, which encodes HMG coenzyme A reductase, the main target of statins, shift plasma LDL-C levels only modestly and yet statins are the cornerstone of treatment of high plasma LDL-C levels.

Not surprisingly, GWAS have identified SNVs in several genes involved in cholesterol biosynthesis and metabolism as susceptibility variants for CAD.^177,350 GWAS studies of CAD have also led to identification of a large number of genes previously not implicated in the pathogenesis of CAD (Table 55–17). The discoveries have the potential to lead to identification of novel molecular mechanisms involved in the pathogenesis of CAD, and hence offer the opportunity to develop novel therapies. However, the task of identifying the true risk allele is quite tedious because of the complexity of the genomic structure, which also includes long-distance interactions among the variants.

Among the notable discoveries are the 9p21 and 1p13 loci, which are among the most intensely studied GWAS loci for CAD.^{351,352,353,354,355} The 9p21 locus expresses the long noncoding RNA ANRIL, which targets cell cycle regulator CDKN2B and hence is referred to as CDKN2B-AS1. The variant allele at the 9p21 locus affects increased expression of CDKN2B-AS1, which in turn targets cell cycle regulators and leads to enhanced proliferation and reduced senescence of smooth muscle cells.³⁵⁶ Likewise, SNVs at the 9p21 locus disrupt a STAT1 enhancer and impair response to interferon gamma.³⁵⁷

The 1p13 locus contains several genes including CELSR2, PSRC1, MYBPHL, and SORT1.^354,355 The latter, which encodes Sortilin 1, has emerged as the prime susceptibility gene for coronary atherosclerosis. The underlying mechanism remains inadequately understood, but likely relates to the influence of Sortilin 1 on hepatic apolipoprotein B (apoB) secretion, LDL-C catabolism in the liver, and possibly expression of the inflammatory markers.^358,359 The risk variant generates a binding site for the CEBPA adipogenic transcription factor on the 5′ regulatory region of SORT1 gene, enhancing its expression.^358,360 The discovery of SORT1 as a new susceptibility gene for atherosclerosis has rendered it as a potential target for hypercholesterolemia and therefore modification and treatment.

Plasma levels of lipids are complex traits determined by interactions of a large number of SNVs and the environmental factors, including diet. Over 50 GWAS have been reported along with identification of several hundred loci associated with plasma levels of LDL-C, high-density lipoprotein cholesterol (HDL-C), triglycerides, and various apolipoproteins.^{177,361,362,363,364} A complete searchable list of the GWAS related to lipid disorders is found in the GWAS catalog web page (http://www.ebi.ac.uk/gwas/home).

As is common to GWAS of complex traits, SNVs identified through GWAS influence plasma levels of cholesterol, triglycerides, and lipoproteins only modestly. For example, SNVs associated with plasma HDL-C level shift the level by 1 mg/dL or less, and those associated with plasma LDL-C typically by less than 5 mg/dL.^{177,361,362,363,364} The modest effect sizes of the GWAS findings, however, do not mitigate the clinical impact of discovering the genetic basis of lipid disorders through GWAS. The clinical significance of the GWAS discoveries relates to their potentials in leading to identification of novel preventive and therapeutic targets. The point is illustrated for the effect sizes of the common variants in HMGCR, encoding HMG coenzyme A reductase, the target of statins, which shift plasma LDL-C by less than 5 mg/dL and yet inhibition of this enzyme with statins is the mainstay of treatment of dyslipidemias. SORT1 is an example of discovery of a novel gene for LDL-C through GWAS, which was discussed earlier.^354,355

Monogenic forms of dyslipidemias are described briefly next.

Familial Hypercholesterolemia

Familial hypercholesterolemia is an autosomal dominant disorder with a prevalence of 1 in 500 in the mild form and in 1 in 100,000 people in its severe form.^365,366 Familial hypercholesterolemia is characterized by severely elevated plasma levels of LDL-C (type IIa hyperlipidemia) and premature atherosclerosis.³⁶⁶ Plasma levels of total cholesterol are in the range of 300 to 400 mg/dL in affected heterozygous individuals, and greater than 500 mg/dL in homozygous subjects. The affected individuals develop severe atherosclerosis involving multiple vascular territories, tendon xanthomata, and corneal arcus. Subjects homozygous for the causal mutations exhibit clinical atherosclerosis in the first or second decades of life and heterozygous subjects in the fourth or fifth decades. These patients often suffer from ischemic symptoms and/or cardiac events requiring revascularization procedures very early in life.³⁶⁶

A common causal gene is LDLR, which is located on chromosome 19p13.^367,368,369 It encodes LDL-C receptors. More than 1000 point, deletion, and splice mutations in LDLR have been identified in patients with familial hypercholesterolemia. Approximately 60% of the mutations are missense mutations, 20% are minor rearrangements, 13% are major rearrangements, and 7% are splice-junction mutations. Mutations cause familial hypercholesterolemia by perturbing the function of LDL-C receptors.^367,368,369 Mutations could affect synthesis and targeting to the cell membrane, binding of the receptor to LDL-C, internalization of the receptor following binding to LDL-C, and recycling of the receptors. The ensuing biologic effect is impaired removal of apoB and apoE from the circulation. Mutations affect LDL-C receptor function to variable degrees, leading to variable clinical manifestations. In general, there is an inverse correlation between plasma levels of LDL-C and the level of residual LDL receptor activity. Mutations that completely inactivate the receptors lead to severe premature atherosclerosis in childhood. Frameshift mutations by markedly altering the structure of the LDL-C receptors cause severe phenotype. In contrast, mutations that partially inactivate the receptors cause mild-to-moderate hypercholesterolemia. Thus, the development and severity of coronary atherosclerosis in part vary according to causal mutations, that is, residual LDL-C receptor activity. Genetic background, diet, environmental factors, and epigenetic factors are also likely to contribute to the phenotype.

The second causal gene for familial hypercholesterolemia is the APOB gene, located on chromosome 2q24. Mutations in APOB gene lead to familial defective apolipoprotein B (FDB). The causal mutation involves amino acid 3500 in > 99% of the cases, with the predominant mutation being R3500Q and, rarely, R3500W. Mutations decrease the affinity of LDL receptors for apoB, which results in accumulation of very low-density lipoprotein cholesterol (VLDL-C) and LDL-C in the plasma and blood vessels. Phenotypically, it is similar to heterozygous familial hypercholesterolemia and includes premature CAD and tendon xanthoma, in addition to elevated plasma levels of LDL-C. However, the LDL receptor activity is normal in these individuals. FDB is a relatively common disorder, with an estimated frequency of 1 in 1000 people, but the prevalence varies worldwide. The frequency of the mutation varies in different regions of the world.

The third causal gene for autosomal dominant hypercholesterolemia was mapped in 2003 and identified as PCSK9, which encodes PCSK9, also known as neural apoptosis-regulated convertase (NARC1).³ Pathogenesis of the phenotype involves enhanced degradation of the LDL receptors by PCSK9.³⁷⁰ Population genetic studies have extended the clinical spectrum of SNVs in PCSK9 including identification of gain- and loss-of-function mutations in PCSK9 influencing plasma levels of LDL-C and the clinical effects.^{370,371,372,373,374} Since the discovery of PCSK9 and partial elucidation of the responsible mechanisms, monoclonocal antibodies that target this molecule have been developed and have been shown to be extremely effective in lowering not only plasma LDL-C levels but also cardiovascular outcomes.^4,5,375

More recently, STAP1I, located on 4p13 and encoding signal transducing adaptor family member 1, has been identified as a plausible fourth gene for familial hypercholesterolemia. The phenotype seems to be a milder form of familial hypercholesterolemia.^376,377

Autosomal Recessive Hypercholesterolemia

Autosomal recessive hypercholesterolemia is a rare disease with a phenotype similar to familial hypercholesterolemia. The causal gene is LDLRAP1 (also known as ARH), which encodes an adaptor protein that binds to LDL receptors and hence is called LDL receptor adaptor protein 1 (LDLRAP1).^378,379 The protein contains a phosphotyrosine-binding domain that interacts with the cytoplasmic tail of LDL receptors. The pathogenesis of the phenotype involves impaired clearance of plasma LDL-C, despite normal activity of LDL receptors.³⁷⁹

Hypobetalipoproteinemia

Hypobetalipoproteinemia, or abetalipoproteinemia, is a rare disease characterized by extremely low plasma levels of apoB, total cholesterol, and LDL-C; HDL-C levels are high and atherosclerosis is very uncommon.³⁸⁰ The phenotype often presents in childhood with failure to thrive, fat malabsorption, celiac disease, vitamin A and E deficiency, ataxia, demyelination of the central nervous system, and low plasma LDL-C levels, underscoring the importance of the need of cholesterol for cell function and growth. Sporadic and familial cases with an autosomal dominant mode of inheritance have been reported. A causal gene is MTTP on chromosome 4q22–24, which encodes the microsomal triglyceride transfer protein.^381,382 MTTP is a heterodimer of a unique large subunit and the protein disulfide isomerase, which catalyzes the transport of triglyceride, cholesteryl ester, and phospholipid from phospholipid surfaces. Mutations encode truncated nonfunctional protein, thus leading to very low levels of apoB, LDL-C, and total cholesterol.³⁸⁰

Familial hypobetalipoproteinemia is also caused by loss-of-function heterozygous mutations in PCSK9, which was discussed earlier. It also can arise because of truncation mutations in APOB. It is also a rare disorder characterized by very low plasma levels of LDL-C and total cholesterol.³⁸³ Another form of familial hypobetalipoproteinemia is the chylomicron retention disease, which is an autosomal dominant disease characterized by the selective absence of apoB-48.^380,383 Chylomicrons are absent in plasma of the affected individuals after a fat-containing meal. The phenotype is characterized by steatorrhea, growth retardation, malnutrition, and acanthocytosis. The causal gene is SAR1B (SARA2) located on chromosome 5q31. The encoded protein is involved in intracellular trafficking of proteins in COP-coated vesicles. Finally, recessive mutations in ANGPTL3, encoding for angiopoietein-like 3, cause familial combined hyperlipidemia, characterized by reduced levels of plasma LDL-C, HDL-C, and triglycerides.³⁸⁴

Fish-Eye Disease

Fish-eye disease is a rare autosomal dominant condition caused by deficiency of lecithin:cholesterol acyltransferase (LCAT).³⁸⁵ The LCAT gene is located on chromosome 16q22.1 and codes for a protein involved in the synthesis from pre-apoA1 and conversion of HDL₃ to HDL₂ cholesterol. Deficiency of LCAT leads to premature coronary atherosclerosis, proteinuria, anemia, renal failure, and corneal opacification.^385,386

Tangier Disease

Tangier disease is an autosomal codominant disease characterized by the virtual absence of HDL-C and very low plasma levels of apoA1.³⁸⁷ Deposition of cholesteryl esters results in characteristic hypertrophic orange-colored tonsils, hepatosplenomegaly, premature myocardial infarction, and stroke. Other clinical features might include thrombocytopenia, anemia, gastrointestinal disorders, and corneal opacities.³⁸⁷ Mutations in the ATP-binding cassette transporter (ABCA1) gene cause Tangier disease and familial hypoalphalipoproteinemia, its allelic variant.^388,389 The ABCA1 gene is located on chromosome 9q31 and codes for a protein of 2261 amino acids. ABCA1 is a transmembrane protein with 12 transmembrane domains. It acts as a flippase (transmembrane lipid transporter protein) at the plasma membrane, stimulating cholesterol and phospholipid efflux to apoA1 and HDL-C.³⁹⁰ Normally, ABCA1 transports free cholesterol to the extracellular space, where it binds to apoA1 synthesized by the liver and forms nascent HDL particles from VLDL. In the absence of ABCA1, free cholesterol is not transported extracellularly and lipid-poor apoA1 rapidly degrades. Common polymorphisms in the ABCA1 gene have been associated with coronary atherosclerosis in the general population.^391,392

The predominant form of systemic arterial hypertension is essential hypertension, which accounts for 95% of all cases. Hypertension is a complex phenotype caused by the interactions of multiple genes and environmental factors. GWAS have led to identification of over 100 genes for essential systolic and diastolic hypertension (http://www.ebi.ac.uk/gwas/). Among the genes identified by GWAS as regulators of systemic arterial blood pressure are UMOD, which encodes uromodulin, and NPR3, coding for natriuretic peptide receptor-3 and targeting natriuretic peptides for cleavage.^393,394,395 Each variant, as is the case for most if not all complex traits, exerts a very modest effect on the blood pressure, typically within the error of measurement. Nevertheless, as discussed earlier, the impact is on identification of novel mechanisms and novel preventive and therapeutic targets. In this chapter, we will discuss monogenic forms of hypertension. Systemic arterial hypertension will be elaborated in another chapter.³⁹⁶

Glucocorticoid-Remediable Aldosteronism

Glucocorticoid-remediable aldosteronism is a rare autosomal dominant disorder and the first described familial form of hyperaldosteronism.³⁹⁶ It is caused by a chimeric mutation that joins the promoter region of the 11β-hydroxylase (CYP11B1) gene to the coding region of the aldosterone synthase (CYP11B2) gene.³⁹⁷ The new chimeric gene, located on chromosome 8q24, lacks the negative feedback regulation imparted by angiotensin II. The promoter of the fusion gene, which is made up of the 5′ fragment of the CYP11B1 gene, is responsive to adrenocorticotropic hormone. Thus, expression of aldosterone remains unchecked, and excess aldosterone synthesis leads to the retention of sodium and salt and fluid and subsequent hypertension. Glucocorticoid-remediable aldosteronism responds to treatment with glucocorticoids, which suppress the production of adrenocorticotropic hormone. Alternatively, treatment with mineralocorticoid receptor blockers also controls the hypertension.

Apparent Mineralocorticoid Excess

Apparent mineralocorticoid excess is a rare autosomal recessive disease of peripheral metabolism of cortisol.³⁹⁶ Clinical manifestations of apparent mineralocorticoid excess, in addition to hypertension, include hypokalemia, low plasma renin activity, and responsiveness to spironolactone. There are two types of apparent mineralocorticoid excess, defined on the basis of severity of the biochemical phenotype. Both clinical variants are caused by mutations in the HSD11B2 gene on chromosome 16q22, which encodes 11β-hydroxysteroid dehydrogenase II.³⁹⁸ The enzyme is responsible for the peripheral conversion of biologically active cortisol to inactive cortisone. Point mutations in HSD11B2 reduce or abolish the activity of 11β-hydroxysteroid dehydrogenase in the conversion of cortisol to cortisone. Thus, cortisol accumulates, leading to retention of salt and fluid through activation of mineralocorticoid receptors and hypertension. Accordingly, patients with apparent mineralocorticoid excess respond to blockade of mineralocorticoid receptors.

Liddle Syndrome

Liddle syndrome is a rare autosomal dominant disease characterized by hypertension, hypokalemic metabolic alkalosis, low plasma renin activity, and suppressed aldosterone secretion. The phenotype usually develops early in life and hypertension is frequently severe. The first gene identified was the SCNN1B, located on locus 16p12, which encodes the β subunit of the amiloride-sensitive Na⁺ channel.³⁹⁹ The renal epithelial Na⁺ channel has three subunits: α, β, and γ. Subsequently mutations in the γ subunit of epithelial sodium channels were also identified (SCNN1G).⁴⁰⁰ The mutations activate the channel (gain-of-function mutations) and lead to sodium retention and hypertension.

Pseudohypoaldosteronism Type II

Pseudohypoaldosteronism type II, also known as the Gordon hyperkalemia–hypertension syndrome, is a rare autosomal dominant disorder characterized by hypertension and hyperkalemia early in life, mild hyperchloremia, metabolic acidosis, and suppressed plasma renin activity. Two causal genes for pseudohypoaldosteronism type II—WNK4 on chromosome 17q21 and WNK1 on chromosome 12p—have been identified. WNK4 and WNK1 encode serine threonine kinases expressed in the distal nephron.⁴⁰¹ Missense and deletion mutations exert a gain-of-function effect, increasing the expression levels of the proteins in the kidney and leading to increased renal salt reabsorption and reduced renal K⁺ excretion. Recently KLHL3, coding for Kelch-like 3, and CUL3, encoding Cullin 3 (CUL3) also have been identified as causal genes for hypertension and electrolyte abnormalities.⁴⁰²

Familial Primary Pulmonary Hypertension

Primary pulmonary hypertension (PPH) is diagnosed when mean resting pulmonary artery blood pressure is greater than 25 mm Hg in the absence of known secondary causes such as lung disease or pulmonary venous congestion secondary to heart failure.⁴⁰³ Dyspnea is the most common symptom. It is the first symptom in 60% of the cases. Other clinical manifestations include exercise intolerance, fatigue, cyanosis, syncope, and SCD.⁴⁰³ Clinical manifestations often start in the third and fourth decades of life and are about twice as common in females. Median survival in those with established PPH is about 3 years. Prevalence is 1 to 2 per 1,000,000 individuals.⁴⁰³

PPH is a familial disease with an autosomal dominant mode of inheritance in 5% to 10% of the cases. PPH is a genetically heterogenous disease. Mutations in bone morphogenic protein receptor type II (BMPR2), mapped to chromosome 2q31–33, are responsible for approximately 50% of the familial PPH and 10% to 15% of the sporadic cases.^404,405 The spectrum of mutations includes nonsense and frameshift mutations expected to produce dysfunctional protein.^406,407,408 Bone morphogenic protein (BMP) receptor is a cell-surface receptor that belongs to the transforming growth factor β (TGFβ) family. Binding of ligands to BMPR2 activates signaling through Smad molecules.

Mutations in ACVRL1, which codes for a type I receptor of the TGF-β family, cause an autosomal-dominant vascular disorder characterized by pulmonary hypertension, hereditary hemorrhagic telangiectasia, and visceral arteriovenous malformations.⁴⁰⁹ Likewise, mutations in SMAD9, ENG (encoding endoglin), CAV1 (encoding caveolin 1), KCNA5 and KCNK3 (both encoding potassium channels), and EIF2AK4 (coding for eukaryotic translation initiation factor 2 alpha kinase 4), are causes of PPH.^{406,407,408,409}

The pathogenesis of pulmonary hypertension caused by mutations in the TGFβ1 pathway includes haploinsufficiency, wherein defective TGFβ signaling via Smad molecules results in proliferation of smooth muscle cells and reduced apoptosis. Other mechanisms include altered resting membrane potential in pulmonary vascular cells (KCNA5 and KCNK3 mutations) and angiogenesis in response to cellular stress (EIF2AK4 mutations).^406,407

Marfan Syndrome

Marfan syndrome is a primary disorder of connective tissue characterized by cardiovascular, ocular, and skeletal abnormalities.^410,411 There is significant variability in the clinical manifestations of Marfan syndrome, but the predominant features are progressive dilatation of the aortic root, aortic aneurysm, dissection, and aortic and mitral valve regurgitation. The estimated incidence of Marfan syndrome is 1 per 5000 population. The age of onset of clinical manifestations of Marfan syndrome is variable, but cardiac phenotypes commonly occur in the third or fourth decades of life. Aortic dissection is the leading cause of premature death in patients with Marfan syndrome. In addition to cardiovascular abnormalities, marfanoid habitus (increased height, disproportionately long limbs and digits), lens dislocation or subluxation, arachnodactyly, thoracic abnormalities, and increased joint laxity are common clinical features.

Marfan syndrome is an autosomal dominant disease that exhibits locus and allelic heterozygosity. The first causal gene to be identified is FBN1, which is located on 15q15.23 and encodes fibrillin.^411,412,413 Fibrillin is a cysteine-rich protein with a molecular mass of 350 kDa; it is the major component of extracellular microfibrils in both elastic and non-elastic connective tissues. Several hundred nonrecurring unique mutations in FBN1 have been described that encompass missense, nonsense, and deletion mutations, as well as abnormal splicing or exon skipping.⁴¹¹ Mutations are spread throughout most of the gene, and the frequency of each particular mutation is relatively low, which makes screening for mutations tedious.

The Loeys-Dietz syndrome is an autosomal dominant aortic-aneurysm syndrome characterized by the triad of arterial tortuosity and aneurysms, hypertelorism, and bifid uvula or cleft palate.⁴¹⁴ Loeys-Dietz syndrome is caused by mutations in the TGFβR1 and TGFβR2.^410,413

There is a significant variability in the phenotypic expression of Marfan syndrome. The phenotypic variability may be partly caused by locus and allelic heterogeneity and partly by the effect of modifier genes and perhaps environmental factors. The clinical spectrum varies from ectopia lentis in the absence of any other phenotype to neonatal Marfan syndrome and premature death, often within the first 2 years of life. Mutations inducing premature termination of the protein result in approximately a 50% reduction in the level of fibrillin and more frequent ocular manifestations.

Contractural arachnodactyly is characterized by Marfan-like appearance, severe kyphoscoliosis, generalized osteopenia, flexion contractures of the fingers, abnormally shaped ears, and less frequently, mitral regurgitation and congenital heart disease. Point mutations in the FBN2 gene have been described as causes of contractual arachnodactyly.⁴¹⁵ FBN2 mutations clustered in limited regions alter amino acids in the calcium-binding consensus sequence in the epidermal growth factor (EGF)-like domains. Mutations affect either the conserved cysteine residues or residues of the calcium-binding consensus sequence of the cbEGF motifs, and often result in premature termination of the protein, which therefore results in the phenotype just described.

The pathogenesis of Marfan syndrome entails decreased expression levels of the fibrillin protein, and reduced deposition of fibrillin in vascular adventitia, which results in weakening of the adventitia and aortic aneurysm formation as well as impaired mechano-transduction.

Ehlers-Danlos Syndrome

Ehlers-Danlos syndrome (EDS), a relatively uncommon disorder, encompasses a group of conditions characterized by increased elasticity of the skin and connective tissue diseases.⁴¹⁶ The classic form of EDS is characterized by joint hypermotility and fragile, bruisable skin that heals with peculiar “cigarette-paper” scars. Other clinical features include translucent elasticity of the skin, mitral valve prolapse, spontaneous rupture and aneurysm of large arteries, particularly the aorta and its branches, kyphoscoliosis, atrophic scars, and hematomas in the joint areas, especially in the knees and elbows. In the severe form, spontaneous aneurysmal rupture of large arteries and even rupture of the large and small intestines is common. In the benign form, the only manifestations may be hyperextensibility of joints and easy bruisability. The age of onset of clinical manifestations is variable and ranges from childhood to late adulthood.

EDS has several different forms that are inherited in three different patterns of transmission—autosomal dominant, autosomal recessive, and X-linked recessive.⁴¹⁶ Cardiovascular abnormalities are more common in forms I and IV and include congenital malformations, such as tetralogy of Fallot, ASDs, and valvular abnormalities such as mitral and tricuspid valve prolapse.

EDS is genetically an exceedingly heterogenous disease.⁴¹⁷ Ehlers-Danlos type IV, caused by mutations in COL3A1, is considered the most malignant form because of susceptibility to spontaneous rupture of the bowel and large arteries as well as a high incidence of pregnancy-related complications.⁴¹⁸ Cardiac manifestations include aortic and coronary artery aneurysms with a high incidence of rupture. Mutations in various collagen genes such as COL5A1/COL5A2, COL1A1, COL1A2, COL3A1, COL1A1, COL1A1/COL1A2, and COL1A1/COL1A2 are described in patients with EDS.⁴¹⁷ Likewise, mutations in genes encoding for the components of the extracellular matrix proteins or that target such proteins, such as ADAMTS2, TNXB, PLOD1, CHST14, SLC39A13, ZNF469, and PRDM5, have been reported.⁴¹⁷

Cutis Laxa

Cutis laxa comprises a heterogeneous group of acquired and genetic disorders characterized by redundant, wrinkled, loose, sagging skin that slowly returns to normal after stretching. Cardiac manifestations include pulmonic stenosis, aortic aneurysms, and right-sided heart failure. Vessels are very tortuous, resembling corkscrews on the angiogram.

Autosomal dominant, autosomal recessive, and X-linked forms have been described, and mutations in the elastin gene (ELN) have been identified in the autosomal dominant form.⁴¹⁹ A homozygous missense mutation in the fibulin-5 (FBLN5) gene has been identified as responsible for the recessive form.⁴²⁰

Pseudoxanthoma Elasticum

Pseudoxanthoma elasticum is a genetic disorder characterized by dermatologic, ocular, and cardiovascular abnormalities resulting from degeneration of the elastic fibers.⁴²¹ Manifestations include pseudoxanthoma, especially in areas of the neck and axillae, angioid streaks in the optic fundus, and gastrointestinal hemorrhagic and occlusive disease. Cardiovascular abnormalities include calcification of the peripheral arteries, with resulting intermittent claudication, CAD, mitral valve prolapse, and hypertension.⁴²¹

Recently, mutations in the ATP-binding cassette (ABC) transporter gene (ABCC6) on chromosome 16p13 were identified as the cause of pseudoxanthoma elasticum.⁴²² The exact biologic function of ABCC6 protein and the mechanism(s) by which mutations in ABCC6 cause pseudoxanthoma elasticum are unknown.

Osteogenesis Imperfecta

Osteogenesis imperfecta comprises a heterogeneous class of connective tissue disorders characterized by bone fragility. Bone fragility results from defective collagen synthesis, which leads to decreased bone mass, disturbed organization of bone tissue, and altered bone geometry (size and shape). Cardiovascular abnormalities include valvular lesions, such as mitral and aortic regurgitation, and an increased fragility of the blood vessels.⁴²¹

Mutations in one of the two genes encoding type I collagen, namely COL1A1 and COL1A2, have been identified as the cause. Over 1000 mutations in collagen genes have been described.⁴²¹ In addition, mutations in over a dozen genes have been reported in patients with various types of osteogenesis imperfecta.⁴²¹