GENETICS

Understanding of human genetics is growing exponentially. Genes determine basic human form and function, and it is becoming increasingly evident that genetic abnormalities contribute at least in part to most disorders. Genetic information and knowledge are leading to earlier disease identification and improved treatment capabilities. Advances have allowed detection with a single blood test of diseases once identified by painstaking evaluation of family trees and subsequent chromosome analysis. Familiarity with and improved understanding of normal and abnormal human genetics have become critical to the practice of medicine because genetics plays an ever-increasing role in diagnosis, prevention, and management of disease. This chapter is an overview of human genetics. Described are patterns of genetic transmission and the molecular basis of genetic disorders. Principles of genetic counseling are presented. Common genetic malformations in otolaryngology are discussed as are the genetics of hereditary hearing loss and head and neck cancer.
DNA AND CHROMOSOMES
All genetic information is encoded by four bases composing molecules of deoxyribonucleic acid (DNA) (Fig. 2.1). Sequences of DNA, called genes, are transcribed into corresponding sequences of ribonucleic acid (RNA), which are translated into sequences of amino acids that constitute specific proteins. The proteins serve as structural components of cells and tissues or as enzymes that catalyze chemical reactions. The entire developmental program of an organism is effected by the precisely timed and orderly expression of genes. The human total DNA content, or genome, contains about 3 billion base pairs of DNA, which encode more than 80,000 functional RNA molecules and proteins (1).
Most of the human genome is contained within the nucleus of the cell, packaged into structures called chromosomes. Each somatic cell contains 46 chromosomes, the size and structure of which are the same for every person (Fig. 2.2). The nonsex chromosomes (autosomes) are represented in pairs, one inherited from each parent. The sex chromosomes are two X chromosomes in women and girls or an X and a Y in men and boys. Each chromosome consists of a single continuous DNA molecule. In addition to the nuclear genome, each mitochondrion contains several copies of a circular DNA molecule of approximately 16,500 base pairs. The mitochondrial DNA encodes some of the proteins involved in oxidative phosphorylation, transfer RNA (tRNA), and ribosomal RNA (rRNA). Mutations in mitochondrial DNA account for a set of diseases with a distinctive pattern of maternal transmission.
A preliminary map of the human genome has been completed through the coordinated efforts of the Human Genome Project. The ability to identify genes responsible for specific human disorders and provide tools for diagnosis is of great clinical importance (1). Ethical issues regarding discrimination, privacy, and ownership of genetic information have been raised.
PATTERNS OF GENETIC TRANSMISSION
Genetic disorders occur sporadically, as is typical of chromosomal anomalies and mutations that are genetically lethal, or they are transmitted from generation to generation. The classic patterns of genetic transmission include autosomal recessive, autosomal dominant, and X-linked.
Several terms are used to describe patterns of genetic transmission. The constitution of a particular genetic locus is called the genotype of the cell or organism. The particular forms of a gene on each chromosome are alleles. Persons who carry two identical alleles at a gene locus are said to be homozygous. Those who carry two different alleles (e.g., one normal and one mutant) are heterozygous. The observable characteristics of a cell or organism that are controlled by a particular genetic locus are called the phenotype.
Autosomal Recessive Inheritance
A prototypic pedigree for a family with an autosomal recessive disorder is illustrated in Fig. 2.3A. Autosomal recessive traits are expressed only if both copies of a gene are affected by mutation, that is, they are homozygous. Both parents usually are heterozygous, each having one normal copy of the gene and one mutant copy. Because each carrier parent has a chance of one in two (50%) of passing the mutant gene to any child, the risk of a child’s receiving the mutant gene from both parents and having the recessive trait is one in four (25%). Siblings of affected persons have a 67% risk of being carriers, but their risk of having affected offspring themselves generally depends on the frequency of the mutant gene in the population. Heterozygotes usually display no phenotypic effects of carrying the mutant allele. Autosomal recessive traits that are lethal before childbearing age, such as Tay-Sachs disease, generally occur within sets of siblings, and there is no evidence of the disorder in previous generations. Such traits also tend to occur more commonly in families in which the parents are consanguineous.
Many autosomal recessive disorders are more frequent in specific racial or ethnic populations. This frequency represents the expression of a mutation present in one or a few of the founding members of a group (founder effect) and a tendency for marriages to occur within the same group. Autosomal recessive deafness, which is entirely consistent with a normal life span and normal reproduction, can be present in two generations if both deaf parents are homozygous for the same recessive form of deafness. Such marriages would cause deafness among all the children. It is more typical, however, for all children of such couples to have normal hearing, because there are many genetic causes of deafness, and most such couples have different genetic forms of deafness.
Hurler syndrome, or mucopolysaccharidosis (MPS) type I, is an example of an autosomal recessive trait for which the gene is well understood (2). Hurler syndrome is caused by a mutation in the a-L-iduronidase gene (IDUA) that causes production of a protein with absence of normal enzymatic activity. Heterozygotes retain sufficient enzyme activity to be healthy, but homozygotes cannot metabolize mucopolysaccharides, and these compounds are stored in many tissues. The Hurler phenotype is a progressive disease that includes corneal opacity, hearing loss, enlarged tongue, hepatosplenomegaly, joint contractures, and life-threatening respiratory, cardiac, and gastrointestinal complications. A different mutation in the same gene is the cause of Scheie syndrome, a less severe form of MPS (type V) without mental retardation, corneal opacity, or stiff joints. Because the enzymatic activity of the protein is low but not absent, the metabolic block is incomplete. Persons who are compound heterozygotes, with one MPSI allele and one MPSV allele, have an intermediate phenotype, called Hurler-Scheie syndrome. There are many different mutations within the DNA sequence of the gene. Each is associated with a specific reduced activity of the enzyme. Today it is understood that most affected persons who are not the product of consanguineous matings are compound heterozygotes for two different mutations of the IDUA gene and that the phenotype (Hurler, Scheie, or Hurler-Scheie) depends on the combined enzymatic activity of the enzyme products of the two IDUA alleles.
Autosomal Dominant Inheritance
Unlike autosomal recessive inheritance, autosomal dominant traits are expressed in heterozygotes, who have a 50% chance of passing the mutant gene from generation to generation (Fig. 2.3B). For this reason, such traits are expressed by members of each successive generation. Both sexes can be affected, and men and women have an equal chance of transmitting the gene to the next generation. Both parents of a child with an autosomal dominant disorder should be examined for signs of the condition. If neither is affected, the recurrence risk depends on the rate of penetrance of the disorder. In a disorder with complete penetrance, a child may be affected sporadically because of new mutation, which conveys a low risk of recurrence for the parents.
Classic autosomal dominant traits, such as neurofibromatosis type 1 (NF1), may exist in a family for many generations. In some instances, the person affected by the autosomal dominant trait appears to be the only affected member of the family. One explanation is nonpenetrance, which is the apparent lack of phenotypic effect of a gene in a known carrier. NF1 is typical of many autosomal dominant traits in that its expression varies among individuals. About 60% of persons older than 20 years who carry the NF1 gene have cutaneous neurofibromas, 17% have scoliosis, 13% have optic glioma, and 97% express five or more cafe-au-lait spots by 20 years of age. Many other features are associated with this syndrome. Therefore the penetrance of the gene approaches 100% if the entire spectrum of the phenotype is considered. Nevertheless, about 50% of patients with NF1 appear to have no family history of the disease and probably do have new mutations. This apparently high rate of mutation in the NF1 gene may be accounted for partly because the gene is large and perhaps because segments of the gene sequence are predisposed to mutation, especially during spermatogenesis.
Sex-linked Inheritance
Traits expressed by genes on the X or Y chromosome are called sex-linked. Most sex-linked disorders among humans involve genes on the X chromosome. Because a father transmits a Y chromosome to his son, X-linked traits are not passed among male family members. A woman who carries an X-linked recessive disorder has a 50% chance of transmitting the condition to each son and a 50% risk of transmitting carrier status to each daughter (Fig. 2.3C). Rare X-linked recessive disorders, such as Duchenne muscular dystrophy, occur mostly among male carriers, who express only the mutant gene because they have only one X chromosome. Female carriers are heterozygous and usually do not manifest the disorder.
X-linked Kallmann syndrome is caused by a mutation that results in a deficiency of hypothalamic gonadotropin-releasing hormone. The responsible gene, KAL, was localized to the distal short arm of the X chromosome (band Xp22.3). The protein product of the gene appears to act during embryogenesis as a cell adhesion molecule responsible for normal neuronal cell migration (3). Boys with Kallmann syndrome have hypogonadotropic hypogonadism, micropenis, cryptorchidism, unilateral renal agenesis, and other defects. They also have complete, or nearly complete, anosmia due to agenesis of the olfactory lobes, so a simple test for this disorder is the University of Pennsylvania Smell Identification Test balanced with ammonia testing, which is sensed through cranial nerve V, not cranial nerve I. Effective hormonal therapy for hypogonadism and infertility is available. There also are autosomal dominant and autosomal recessive forms of Kallmann syndrome with variable expression of the anosmia, and there is variable expression of the phenotype among women and girls. X-linked dominant traits can be expressed by both sexes. In some cases, the effects among boys are so severe as to be incompatible with survival.
Multifactorial Inheritance
Many traits appear to cluster in families but do not display the transmission expected for a single-gene dominant or recessive trait. An example is cleft lip, which is a relatively rare, sporadic trait. Although recurrence in families occurs far less often than would be expected by the rules of mendelian inheritance, the risk of recurrence in a family with one affected child is greater than the population risk. Monozygous (identical) twins are more frequently concordant for the trait than dizygous (fraternal) twins, suggesting that inheritance plays some role in the disorder. Such traits are said to be subject to multifactorial inheritance, meaning that many factors, including several distinct genes and environmental factors, contribute to the trait. Quantitative attributes, such as height, are believed to be controlled by multifactorial inheritance. Congenital malformations, such as cleft lip and neural tube defects, are explained by the threshold model of multifactorial inheritance, in which a combination of genetic and nongenetic factors add to create a “liability” for the malformation (Fig. 2.4). The malformation occurs only if the liability exceeds a threshold value. The theory of multifactorial inheritance explains many sporadic congenital malformations. Families in which a known multifactorial disorder has occurred usually are counseled with empiric recurrence risk data.
Chromosomal Anomalies
The disorders discussed earlier are caused by mutations of single genes. Other disorders simultaneously involve several genes. The most extreme examples are disorders in which there are extra copies of an entire chromosome. The best known is Down syndrome, produced by an extra copy of chromosome 21. Other chromosomal abnormalities compatible with live birth include trisomy of chromosomes 13, 18, X, and Y (4). Chromosome loss is less well tolerated and usually causes death in utero. The apparent exception is the 45,X karyotype (Turner syndrome), but even here, most 45,X conceptions do not survive to term. The population incidence of Turner syndrome is approximately 1/3,000, and many of these individuals are mosaic, having a 46,XX or other cell population in addition to the 45,X cell population. Most abnormalities of chromosome number occur sporadically as a result of errors of chromosome segregation during meiosis.
Another mechanism of chromosomal anomaly is rearrangement, which includes deletion, duplication, translocation, or inversion. Deletions and duplications tend to have widespread phenotypic effects, including mental retardation, growth retardation, and multiple unrelated congenital malformations. The phenotypic effect depends on the specific genetic material gained or lost and the extent of gain or loss. In general, a gain of 1% or a loss of 0.5% of the genome is consistent with viability and can occur in a balanced configuration, in which there is no net loss or gain of material. Persons with balanced translocations or inversions usually are phenotypically normal but have a risk of transmitting an unbalanced chromosomal constitution to offspring. The result is familial transmission of a disorder of multiple congenital anomalies.
A microdeletion syndrome is caused by duplication or deletion of a small segment of chromosome material that contains a small number of few genes, which are functionally unrelated but by chance are linked together on the chromosome. The phenotype may vary because of different breakpoints, making delineation of a clear syndrome difficult (5). Affected persons usually appear sporadically, but familial clusters are known and typically are caused by a balanced chromosome rearrangement that runs in the family. Before the chromosomal basis was understood, such conditions were thought to represent autosomal recessive or new dominant mutations (6).
An example of a microdeletion syndrome of interest to otolaryngologists is velocardiofacial syndrome. This is a heterogeneous disorder usually caused by a microdeletion in chromosome 22, band 22q11.2 (Fig. 2.5) (7). Other dysmorphology syndromes known to be associated with the same microdeletion include DiGeorge and, at least occasionally, CHARGE (coloboma of the iris, heart disease, atresia of the choanae, retarded growth and development, genital and ear anomalies) and 3C (craniocerebellocardiac), Bernard-Soulier, Opitz G and BBB, and Cayler cardiofacial syndromes. The phenotype is variable, but the diagnosis usually is made when the constellation of a conotruncal heart defect and other malformations, such as hypocalcemic seizures and hypoplasia of the thymus and parathyroid glands of DiGeorge syndrome, is present. Hearing loss is a frequent associated finding. The deletion sometimes is familial, appearing in pedigree analysis as an autosomal dominant disorder with variable expression. The proband typically is the more severely affected person, and some family members have only one feature of the phenotype, such as ventricular outflow or aortic arch malformation. It has been estimated that 15% to 20% of such cases are associated with this microdeletion of chromosome 22 (7).
The velocardiofacial syndrome deletion usually is not detected with classic chromosome analysis methods. For this reason, it is important to involve a dysmorphologist familiar with such conditions and who can request the specific cytogenetic test to establish the diagnosis.
NONTRADITIONAL PATTERNS OF INHERITANCE
Several nontraditional patterns of genetic transmission have been recognized in recent years. One is maternal inheritance due to transmission of genes in mitochondrial DNA. The entire mitochondrial DNA complement of a child is inherited from the mother. Mutations in mitochondrial genes therefore are transmitted from a mother to all her children. Because mitochondria are excluded from sperm, the father does not pass a mitochondrial mutation to any child. Characterized mitochondrial disorders include Leber hereditary optic atrophy and encephalomyopathy ( and at least two genetic forms of isolated deafness (9).
Another nontraditional inheritance pattern is dynamic mutation, or unstable trinucleotide repeats, which is responsible for several forms of neurodegenerative ataxia and a few other disorders (Table 2.1) (10). Some segments of DNA normally comprise 20 or more repetitions of a three-base sequence, such as CAG or CGG. Some persons have more copies, which can cause instability during gametogenesis and production of offspring who carry very long repeats. In some conditions, the number of repeats in an affected person is about double the normal number, and in other conditions, the number of repeats exceeds 1,000. Although the inheritance pattern of these conditions behaves in many respects as a simple mendelian trait, mostly autosomal dominant, several features set them apart. Expansion of the DNA repeat usually is more likely in either spermatogenesis or oogenesis, leading to differences in phenotype, depending on which parent carries the mutation. The severe congenital form of myotonic dystrophy occurs only among offspring of affected mothers because expansion occurs preferentially during oogenesis. In fragile X syndrome, expansion likewise is limited to oogenesis, so daughters of unaffected men who transmit fragile X syndrome never are affected, whereas some sons and daughters of carrier women are affected. In Huntington disease, large expansions are transmitted preferentially by the man during spermatogenesis. The exact mechanism for this parent-of-origin effect in dynamic mutations is not well understood but has been postulated to reflect selection against expanded repeats during gametogenesis (11). Once a DNA repeat begins to expand in length, expansion can continue with each succeeding generation. This dynamism of the trinucleotide repeat sequence explains a phenomenon known as genetic anticipation, whereby the age at onset of the disorder is younger with each succeeding generation and is accompanied by a more severe phenotype.
A third type of genetic disorder is associated with the phenomenon of genetic imprinting (12). For most genes, both copies are genetically active. For other genes, however, only the maternally or the paternally inherited copy is genetically active. Each of these genes is imprinted during either paternal or maternal gametogenesis. Maternal and paternal imprints can be recognized from their DNA methylation patterns, usually involving a cytosine-guanine (CpG) DNA base sequence. Once it is set, the methylation imprint can be faithfully maintained during mitosis by means of postreplication methylation. The imprint is erased only when someone of the opposite sex transmits the gene.
The classic examples of syndromes that exhibit imprinting are Prader-Willi and Angelman syndromes. Both Prader-Willi syndrome and Angelman syndrome usually are associated with microdeletions of the region 15q11 to 15q13. Specific segments within this region are responsible for each syndrome, but an imprinting defect is demonstrated by the fact that a deletion of paternal origin causes Prader-Willi syndrome, whereas an otherwise identical deletion of maternal origin causes Angelman syndrome. Studies of DNA methylation patterns show that some genes in region 15q11-q13 are imprinted during oogenesis and that other genes are imprinted during spermatogenesis. A person must have one paternally and one maternally inherited chromosome 15 to ensure normal expression of all genes in this chromosome region. A microdeletion of paternal chromosome 15 causes Prader-Willi syndrome, and microdeletion of maternal chromosome 15 causes Angelman syndrome. Some patients with Prader-Willi syndrome and Angelman syndrome do not have a microdeletion. Some of these persons demonstrate uniparental disomy, whereby both chromosome 15s are inherited from the same parent. This has the same effect as deletion of one of the genetic imprints. Errors in the imprinting process also can cause lack of normal gene expression. In rare instances, persons with familial Angelman syndrome have normal methylation patterns. These patterns probably involve a point mutation in the imprinting control region of the gene or genes responsible for the Angelman syndrome phenotype, and the inheritance pattern is autosomal recessive (13).
GENETIC DISEASE MAPPING
The profound effect of molecular genetics on clinical practice is the result of the ability to identify genes responsible for specific disorders among humans and to provide tools for diagnosis. Completion of a first draft of the human genome sequence is vital to accelerating the process of gene identification. The basic technology is gene cloning (Fig. 2.6). Segments of human DNA, which may represent random fragments of the genome or may correspond to specific expressed genes, can be inserted (cloned) into bacterial, viral, or yeast DNA and grown in culture. A single bacterial or yeast colony that has incorporated a piece of human DNA of interest can be isolated. Gene libraries that incorporate fragments of the human genome or that incorporate DNA copies of messenger RNA (cDNA) can be made. The cDNA libraries represent only these genes expressed in a particular cell type and are useful for cloning functional genes.
Identification of genes involved in clinical disorders has proceeded in several phases. The first genes to be cloned were those for which the protein product was already known, including globin genes (e.g., sickle cell disease, thalassemia) and enzymes (e.g., phenylketonuria or Hurler syndrome). As genomic tools became further refined, a second approach, positional cloning, became possible. This began with mapping the disease gene, usually by means of tracking the segregation of the disease gene through a family in relation to genetic markers that had already been mapped (linkage analysis). DNA was isolated in this region and examined for mutations among affected persons. This approach has seen major success in identification of genes for conditions such as Duchenne muscular dystrophy, cystic fibrosis, NF1, and NF2.
With the gene sequence available, a candidate gene approach is increasingly used. The location of the disease gene is determined, usually by means of linkage analysis. Then genes known to reside in the region are examined for mutation. The procedure starts with the most plausible candidates based on patterns of expression or the physiologic roles of the gene products.
Identification of a gene responsible for a clinical disorder is a scientific advance. Diagnostic tools can be developed, and gene therapy becomes possible in some cases. Most important, the physiologic basis of the disorder becomes amenable to study, and the way is opened to develop pharmacologic approaches to treatment.
MOLECULAR DIAGNOSIS OF HUMAN GENETIC DISORDERS
In keeping with the interesting variety of gene expression, a diversity of mutations, such as base substitutions, insertions, deletions, and chromosomal rearrangements, can produce genetic disorders among humans. A single disorder typically is caused by many different kinds of mutations of the same gene in different persons (Fig. 2.7). A mutation classically is thought to involve the coding region of a gene in which mutations of one or a few nucleotides produce an abnormal protein or loss of the protein. Mutations in the noncoding regions of the gene, such as the promoter region, splice sites, and termination and polyadenylation signals, also can produce abnormal proteins or reduced levels of normal proteins. Structural rearrangements (insertions or deletions) involving several nucleotides to thousands of nucleotides can produce aberrant proteins or result in absence of proteins (14).

Because the causes of many genetic disorders have been elucidated and the genetic map is more complete, it has become increasingly possible to use molecular genetic approaches for diagnosis. Such approaches allow precise definition of whether a person has inherited a mutant gene, often before appearance of the disease phenotype, and can provide the basis of prenatal diagnosis. Two approaches are used—direct detection and indirect detection (linkage analysis) of abnormal genes.
Direct detection of a mutation is possible if the responsible gene is cloned and a limited number of mutations are known to cause the disease. Most molecular diagnostic tests rely on amplification of a target sequence with a polymerase chain reaction (PCR) (Fig. 2.8). Short sequences of single-stranded DNA (15 to 30 bases called oligonucleotides) homologous to sequences on the opposite strands of genomic DNA serve as the flanking regions for amplification of a DNA fragment of interest. The genomic DNA is denatured into single strands by means of heating, and the synthetic oligonucleotides are allowed to anneal and serve as starting points for a DNA synthesis reaction. This process is repeated sequentially. The result is exponential synthesis of new DNA that corresponds to the region bounded by the two oligonucleotides. PCR technology has revolutionized study of the human genome, including typing of genetic markers, mutation screening, detection of point mutations, cDNA and genomic DNA cloning, genome walking, DNA sequencing, and in vitro mutagenesis (14). A variety of strategies are used to identify mutations. They range from complete gene sequencing to approaches that target a specific single-base change (Table 2.2).

If the disease gene is not cloned but linkage has been established between the disease locus and marker genes, or if the gene has too many defects, making direct analysis impractical, linkage studies can be used for diagnosis in some families. Minor variations of base sequence, polymorphisms, are common among humans. Most of these variants are not located in coding regions of genes and herefore are not responsible for phenotypic effects; however, they can be detected by means of PCR or Southern blotting and constitute heritable markers that can be tracked with a disease trait within a family. The DNA markers do not correspond to molecular defects within a gene but merely tag a disease-bearing chromosome as it is inherited. The use of linkage analysis in the diagnosis of a genetic disorder is illustrated in Fig. 2.9. Although it is a powerful approach, genetic linkage is subject to diagnostic error and is not possible for all families. Diagnostic error is caused by genetic recombination during meiosis, which can change the association of a particular marker allele with a disease allele, by misattributed paternity, or by genetic heterogeneity (diseases that look phenotypically identical but are caused by defects in different genes, such as Alzheimer disease). The technique can be used only by families in which there is a clear pattern of mendelian transmission, for which the clinical diagnosis implicates the linked disease gene unequivocally, and in which the persons who carry the disease gene are heterozygous for the marker, allowing the two chromosomes to be differentiated.
GENE THERAPY
Knowledge of the structure of a gene responsible for a genetic disorder can lead to improved therapy through earlier diagnosis. In addition, the disorder can be managed prospectively with classic treatments or new knowledge of pathogenesis gained from study of the gene product. The long-term hope for genetic therapy is to replace a defective copy of a gene in the cell and reverse the effects of the mutation. There are a number of ways of introducing foreign genes into cells and obtaining stable expression of these inserted sequences. Human gene therapy is being attempted in experimental settings for diseases such as cystic fibrosis, Duchenne muscular dystrophy, familial hypercholesterolemia, and cancer. There are many biologic obstacles to overcome, including obtaining sustained levels of expression comparable to those of normal cells, and targeting the inserted genes to the appropriate cell type. Currently more than 200 gene therapy protocols have been approved for treatment of patients with cancer, acquired immunodeficiency syndrome, and genetic diseases (15).
There are four general strategies in gene targeting. In one method used when there is a loss of normal gene function, extra copies of the normal gene are introduced into cells. This method can be effective in the management of autosomal recessive disorders, such as cystic fibrosis. In one study, investigators used adenovirus vector to introduce wild type p53 tumor suppressor gene into malignant tumors of the head and neck. A favorable response occurred with regression or stabilization of disease for 3.5 months among one half of the patients tested (16).
Targeted killing of specific cells is a second approach often used in gene therapy for cancer. In this approach, the inserted gene produces a lethal toxin that kills the target cells, encodes a gene that is sensitive to a certain administered drug, or stimulates the immune system to kill the target cells. Studies are under way in which a plasmid vector is used to introduce major histocompatibility complex HLA-B7 into squamous cell carcinoma of the head and neck and thereby induce immunologic killing of the cancer. Preliminary studies show a 20% to 25% response rate (15).
A third approach to gene therapy is to correct the mutation either at the DNA level (homologous recombination) or at the RNA level (ribozymes). The final approach is targeted inhibition of gene expression at the DNA, RNA, or protein level, such as use of antisense genes. This fourth approach may be yield important techniques for managing many types of cancer and infectious diseases (14).
TREATMENT OF PATIENTS WITH GENETIC DISORDERS
The American Society of Human Genetics has defined genetic counseling as a “communication process which deals with . . . the occurrence or risk of occurrence of a genetic disorder in a family” (17). The process includes helping the family to (a) understand the medical facts, including the diagnosis and course of the disorder, (b) appreciate the role of heredity in the disorder, including knowledge of the risk of recurrence, (c) understand the options available to deal with the risk of recurrence, (d) choose a course of action, and (e) make the best possible adjustment to dealing with the disorder and the choices they make.
Correct diagnosis is key to caring for patients with inherited disorders. Many disorders can appear similar clinically and yet be distinct genetically. Moreover, many genetic disorders have pleiotropic effects. A pedigree should be constructed for the family, similar to that shown in Fig. 2.3 to help recognize the pattern of transmission or determine that a disorder is sporadic.
A difficult challenge in genetic counseling is posed if a child has a problem for which a specific diagnosis cannot be made. Most often, this occurs when a child has congenital malformations that do not fit a particular syndrome. It is always wise to examine these children for chromosomal anomalies. At most institutions, an experienced dysmorphologist or medical geneticist is available for consultation. If the chromosomes are normal, it may be impossible to establish a specific cause. In this case, counseling must take into account the possibility that the disorder has a genetic basis and that recurrence is possible. Although the recurrence risk cannot be quantified in this situation, empiric recurrence risk may be available. The family needs to understand that the lack of a specific diagnosis or family history of the disorder does not preclude that the problem is genetic and that recurrence is possible.
There are several options for dealing with risk of recurrence. Options likely to be agreeable to a family depend on factors such as their perception of the severity of the disorder, whether the disorder can be managed or be diagnosed prenatally, and the ethical and religious beliefs of the family. Prenatal diagnosis, usually by means of amniocentesis or chorionic villus sampling, is widely accepted and is available for cytogenetic disorders and for an ever-increasing number of mendelian and nontraditionally inherited conditions. Modalities for prenatal diagnosis includes level III ultrasound examination, which can be used to detect some of the congenital malformations associated with many syndromes.
Prenatal diagnosis is undertaken for several reasons. The most obvious is consideration of pregnancy termination if a fetal anomaly is found. Other couples choose prenatal diagnosis to allow planning for the medical needs if a child has an inherited disorder. Prenatal surgical procedures can be performed for some anatomic defects. A major canon of genetic counseling is to be nondirective. The counselor must present options in a neutral manner, not allowing his or her opinions to influence a couple’s decision. Having a child with an inherited disorder imposes many practical and emotional stresses on a family. Complex medical decisions may have to be made and a substantial financial burden sustained. The disorder may be life threatening or may lead to chronic illness and developmental impairment. On the other hand, a familial condition to which adaptation is readily achieved, such as deafness, may not be considered a burden by the family. The correct diagnosis may be of interest only in terms of the likelihood of having children who share the trait of deafness and in terms of monitoring for associated problems such as nephritis.
The medical issues can dominate all other issues for a family. Health professionals can ease the burden by providing competent care, providing information in an understandable manner, and maintaining open communication with the family and the other health care professionals. The geneticist can play a special role by addressing the question of cause. It is especially important to take a thorough family and pregnancy history. It is rare to identify a specific environmental exposure that caused a child’s problems, but it is routine to learn that a family member is worried that an insignificant event might have contributed. These concerns usually are disclosed only after directed questioning. After recognizing the source of anxiety, the counselor or physician can reassure a family and help them deal more directly with the medical and emotional issues at hand.
GENETICS OF OTOLARYNGOLOGIC DISORDERS
A large number of otolaryngologic disorders are known to be familial or to have a genetic component. Although the details of these disorders are discussed elsewhere, some of the more important are described herein to highlight the genetic issues raised.
Congenital Malformations
A variety of congenital malformations affect the ear, nose, and throat. Some occur as a component of multiple congenital malformation syndromes, which may be caused by single-gene mutations or chromosomal abnormalities. Others occur sporadically, displaying multifactorial inheritance. A classic example is cleft lip, which occurs in isolation or with cleft palate. Although facial clefts usually occur sporadically, they can be a component of mendelian disorders or occur in association with multiple congenital anomalies. This is also true of anomalies such as preauricular pits or malformation of the external ear.
Kartagener Syndrome
Kartagener syndrome is an autosomal recessively inherited condition of dextrocardia, situs inversus, immotile sperm, anosmia, bronchiectasis, and chronic cough, all secondary to absence or malformation of the dynein arm structures of cilia.
Down Syndrome
Down syndrome, or trisomy 21, is the most common malformation syndrome and occurs in increasing frequency with increasing maternal age. Common otolaryngologic manifestations include upslanting palpebral fissure, low nasal bridge, macroglossia, narrow palate, protruding tongue, and atlantoaxial instability.
Craniosynostosis
Apert syndrome involves craniosynostosis, syndactyly and midfacial malformations, and mental retardation. Most cases are sporadic mutations. Crouzon syndrome involves craniosynostosis, maxillary hypoplasia, and proptosis. Intelligence is normal. These syndromes are transmitted in an autosomal dominant pattern.
Treacher Collins Syndrome
Treacher Collins syndrome or mandibulofacial dysostosis is characterized by a hypoplastic mandible and maxilla, hypoplastic supraorbital rims, narrow face, depressed cheek bones, bizarre inferiorly placed pinnae, downslanting palpebral fissures, and normal intelligence. The inheritance pattern is autosomal dominant.
Hereditary Deafness
Both congenital and acquired forms of deafness can have a genetic basis (9). One in every 1,000 infants is born with hearing loss. By the age of 80 years, more than 50% of persons have some degree of hearing loss (18). As environmental and infectious causes of hearing loss are being controlled, it is estimated that more than 60% of congenital hearing impairment is genetic and that one third to one half of all deafness has an inherited component (18). Hereditary hearing impairment (HHI) can be classified as syndromic or nonsyndromic (9,18). Syndromic hearing loss (30% of all HHI) occurs in association with other anomalous phenotypic features. Craniofacial malformations, renal abnormalities, skeletal dysplasia, and pigmentary anomalies are some common examples. Nonsyndromic hearing loss (70% of all HHI) occurs as an isolated deficit. Since approximately 1990, positional cloning techniques have allowed rapid identification of a number of single genes that cause nonsyndromic hearing loss.
Syndromic Hearing Loss
Hundreds of autosomal dominant, recessive, and X-linked forms of syndromic hearing loss have been described. Molecular genetic studies have revealed the locations of genes responsible for some of these syndromes. Waardenburg syndrome is a dominantly transmitted disorder that includes sensorineural hearing loss, dystopia canthorum (widely spaced inner canthi), and a disorder of neural crest migration. Intrafamilial heterogeneity in the phenotype (called variable expressivity) exists. PAX3, a gene on the long arm of chromosome 2, has been found to be mutated in many persons with Waardenburg syndrome (19). About 83% of persons who carry the gene have penetrance, that is they have physical manifestations. Dystopia canthorum, heterochromia, or white forelock occurs among approximately one third of gene carriers and deafness among approximately 25% (20).
Another genetically heterogeneous disorder associated with hearing loss is Usher syndrome. In Usher syndrome type I (USH1), partial sensorineural hearing loss occurs with development of retinitis pigmentosa after the first decade of life (21). Families with Usher syndrome type II (USH2) have congenital deafness and early onset of retinitis pigmentosa. Five distinct Usher syndrome genes have been mapped to chromosome bands 1q32 (USH2), 14q32 (USH1), 3q21 (USH3, the Finland variety), 11p15 (USH1C, the Acadian variety), and 11q13 (USH1B) (22).
Pendred syndrome is an autosomal recessive disorder associated with goiter and an increased risk of thyroid carcinoma. It has been mapped to the long arm of chromosome 7. There is variability in expression of the gene defect within families. Some members have near-normal hearing, and some have unilateral deafness. Another mutation in the gene responsible for Pendred syndrome may be responsible for one form of autosomal recessively inherited nonsyndromic deafness (23).
Deafness and nephritis occur together in many conditions. Alport syndrome, for example, is characterized by nephritis, deafness, and cataracts and can be inherited in an X-linked or an autosomal pattern. Mutations in various collagen protein genes are responsible for most forms of Alport syndrome, including the a-5-collagen gene, which is located in the short arm of the X chromosome (24), and genes that code for collagen IV subunits, which have been mapped to chromosome 2 (25).
Nonsyndromic Hearing Loss
Since approximately 1995, linkage analysis has accelerated mapping of genes that cause monogenic nonsyndromic hearing loss. Monogenic prelingual hearing loss is approximately 75% autosomal recessive, 20% autosomal dominant, 5% X-linked, and less than 1% mitochondrial (9). Two parents with autosomal recessive deafness can have children with normal hearing because the mutant gene may be different in each parent. By 1999, 44 gene loci on 19 chromosomes and 15 genes, including two mitochondrial genes, for nonsyndromic hearing loss had been identified (Table 2.3) (9). These loci are labeled DFN (for deafness). DFNA designates autosomal dominant loci; DFNB designates autosomal recessive loci; DFN are the X-linked recessive loci (9). Protein products of these deafness genes include transcription factors, myosin, connexin (involved in formation of cell membrane channels), formin (involved in maintenance of the inner ear cytoskeleton), ion transporters, structural proteins of the organ of Corti, and extracellular matrix proteins. Mitochondrial genes responsible for monogenic nonsyndromic hearing loss include the gene for 12S rRNA and the serine tRNA (9,18).
DFNB1 mutation (connexin-26) on chromosome 13q12 was the first cloned gene to be implicated in and is the most common cause of nonsyndromic hearing loss. It accounts for approximately one half of all cases of autosomal recessive prelingual hearing loss and 20% of all prelingual deafness (9). The hearing loss is moderate to profound but stable. Connexin encodes cell plasma membrane channels involved in intercellular exchange of ions and small molecules. Approximately 1% to 3% of white persons carry a DFNB1 mutation. Molecular screening is possible for connexin 26 abnormalities as part of the diagnosis of hereditary prelingual deafness (9,18).
DFNA9 (the COCH gene) on chromosome 14q11-13 is the gene most frequently involved in autosomal dominant postlingual hearing loss. Clinical features include progressive hearing loss starting in the high frequencies and variable vestibular symptoms. COCH is believed to encode an extracellular matrix protein. Deposits of mucopolysaccharides have been found in the inner ears of patients with the DFNA9 mutations.
DFN3 (encoding the POU3F4 gene), located on chromosome Xq21, is the most common X-linked locus involved in nonsyndromic hearing loss. Hearing loss is progressive with fixation of the stapes footplate. POU genes encode transcription factors. POU3F4 is expressed in the mesenchyme of the inner and middle ear, where it is involved in maturation of bone (9).
The most frequent form of mitochondrial hearing loss involves mutation of the 12S rRNA gene. This mutation occurs in association with aminoglycoside ototoxicity. The mutated rRNA is likely more similar to bacterial rRNA, the target of aminoglycoside antibiotics. Progress in identifying nonsyndromic HHI genes will clarify the underlying molecular mechanisms of hearing and hearing loss, improve genetic counseling, and lead to development of specific therapies for hearing loss.
Genetics of Head and Neck Cancer
Cancer of the head and neck accounts for about 5% of all deaths of cancer in the United States (26). Since approximately 1980, innovations in standard surgical treatment, radiation therapy, and chemotherapy have resulted in only modest improvements in survival from squamous cell carcinoma of the head and neck. The goal of research directed at understanding the basic genetic mechanisms of head and neck cancer is to increase the survival rate among persons with cancer of the head and neck.
Most malignant tumors among humans develop in a complex interaction between genetic and environmental factors. At the most basic level, all cancers are genetic in that development and progression occur because of accumulation of chromosomal and genetic mutations. Four basic relationships can be identified: persons with genetic predisposition for cancer but no environmental exposure, persons with environmental exposure but no genetic predisposition, spontaneous mutations among persons who have neither genetic predisposition nor environmental exposure, and persons with both genetic predisposition and environmental exposure (27). There has long been evidence that squamous cell carcinoma of the head and neck may have a genetic basis despite the existence of known and multifactorial environmental influences (27,28 and 29). In families with smoking-related malignant disease, genetic analysis supports an autosomal dominant inheritance pattern (30). This genetic susceptibility also may explain why some persons with only mild tobacco or alcohol exposure have squamous cell carcinoma of the head and neck, whereas others with many times more use never do (27).
Loss or alteration of cell-cycle control is an intrinsic factor in the development of cancer. Tumor suppressor genes are genes that have been identified as having regulatory control of the cell cycle. When such regulatory forces are altered or lost because of mutational events, cell-cycle control is changed. Poorly regulated or prolific cell growth can occur, and cancers can develop.
Oncogenes are genes that have been causally identified with the development of cancer. An example is the RET oncogene. Germline mutations in RET, located on chromosome 10q11.2, have been identified in families that manifest hereditary medullary carcinoma of the thyroid (31). Identification of RET mutations is used as screening for multiple endocrine neoplasia type 2b and familial medullary carcinoma of the thyroid. Because early identification and management of medullary carcinoma of the thyroid markedly affect outcome and survival, genetic screening of patients and their close relatives has become a critical part of the diagnosis and management of medullary carcinoma of the thyroid. In most cancer types, both loss of tumor suppressor genes and oncogene activation occur. The former is believed to be more important than the latter for squamous cell carcinoma of the head and neck (32). Many of the known oncogenes and tumor suppressor genes are common to many cancers, and identification of a genetic abnormality in one tumor type often is relevant to others.
Some persons are more susceptible to cancer because they are heterozygous for a tumor suppressor or oncogene mutation. Because a single, inherited altered gene already is present in a diploid cell (one hit), only the remaining normal gene copy has to mutate for cancer to develop (two hits). Without the original hereditary abnormality, development of cancer is less likely because two acquired mutations would have to occur. This premise has been shown to be true for some cancers. Retinoblastoma occurs in both heritable and sporadic patterns. Sporadic retinoblastoma is unilateral (30). Persons with the hereditary form have loss or mutation and inactivation of a tumor suppressor gene called Rb1. These persons have a hereditarily determined single hit. The likelihood that retinoblastoma will develop is nearly 50%, and these lesions often are bilateral. Fifty percent of offspring are susceptible to the cancer. The RB1 tumor suppressor gene helps to regulate transcription of other genes and thus is involved in regulation of the cell cycle. Insertion of a normal RB1 gene can result in a return of normal cell regulation (33).
Cytogenetics has been used in the study of squamous cell carcinoma of the head and neck (34,35). Several chromosomal abnormalities have been identified. Oncogenes and tumor suppressor genes are presumed to be located at the breakpoints of these deletions, amplifications, and translocations. Common chromosomal abnormalities identified in squamous cell carcinoma of the head and neck include 3p-, believed to be an early chromosomal change in squamous cell carcinoma of the head and neck; 11q13 rearrangements, the location of the cyclin D1 oncogene (36); and 9p21-22, the site of cell cycle gene p16 (32). Loss of 18q is likely related to advanced disease and carries a poor prognosis (36,37). Other chromosomal losses include 5q, 8p, and 13q,17p. Amplifications include 3q, 5q, and 11q. Cancer cells can be haploid (half the normal DNA content), diploid (two times the normal DNA content), or tetraploid (four times the normal DNA content). Aneuploidy (abnormal DNA content) is a feature of many cancer cells. It is believed to be caused by altered proliferation of tumor cells and to reflect aggressive clinical behavior (38). Ploidy analysis, however, has not shown any prognostic factors and has done little to help identify the nature of head and neck cancer (38).
The cell-cycle gene most widely studied in relation to cancer among humans is the tumor suppressor gene p53, found on chromosome 17p. The p53 protein helps to control the cell cycle by binding with cyclin-dependent kinins to arrest cell replication in G1 (39). This allows the cell to repair any DNA damage or mutations that have occurred. If DNA repair fails, p53 can induce apoptosis or programmed cell death (36). Loss of activity of p53 results in an increase in the number of chromosomal abnormalities (40). This loss of p53 occurs in more than half of instances of squamous cell carcinoma of the head and neck. For patients with a p53 abnormality in the index tumor, p53 can be evaluated at the margins of the tumor at the time of resection. The presence of mutant p53 at the margins is predictive of recurrence, even if the margins appear normal at light microscopic examination (41). It is likely that the presence of p53 is related to early genetic changes in squamous cell carcinoma, such as the conversion of normal mucosa to dysplastic mucosa. Although p53 overexpression has been found to be predictive of a favorable response to chemotherapy and organ preservation protocols, p53 expression has not been found to be predictive of survival from squamous cell carcinoma of the head and neck (36).
Cyclin D1, located at chromosome 11q13, is the most frequently amplified protooncogene in squamous cell carcinoma of the head and neck. This oncogene product accelerates cell cycle progression. Overexpression correlates with advanced disease and reduced overall and disease-free survival rates (32). The p16 gene product is an inhibitor of cyclin pathways and cyclin D1 and therefore is involved in cell cycle regulation. Inactivation of p16 is one of the earliest genetic events in squamous cell carcinoma of the head and neck (32).
The bcl family gene products are involved in cell cycle regulation and apoptosis. The bcl-2 gene product inhibits apoptosis by blocking p53 dependent pathways. Overexpression of bcl-2 has been linked to resistance to chemotherapy. The bax gene encodes an inhibitor of bcl-2. Bcl-xL prevents apoptosis; bcl-xs promotes apoptosis (36).
Epidermal growth factor, epidermal growth factor receptor, and transforming growth factor a are growth factor gene products frequently overexpressed in squamous cell carcinoma of the head and neck. None has been shown to be a reliable prognostic indicator or tumor marker for recurrence (32).
Squamous cell carcinoma of the head and neck arises from a clonal population of cells that have acquired many genetic alterations in a several-step process (34). Unlike the colon cancer model, in which an orderly sequence of genetic events leads from adenoma to metastatic carcinoma (42), it is likely that the genetic mutations in squamous cell carcinoma of the head and neck do not necessarily follow one rigid sequence. Nonetheless, certain genetic changes are believed to occur early and can be found in dysplastic tissue. Others occur late and reflect invasive squamous cell carcinoma. A possible progression of genetic changes for squamous cell carcinoma of the head and neck is depicted in Fig. 2.10.
CONCLUSION
Advances in understanding human genetics allow more precise diagnosis of medical disorders. Molecular methods increasingly allow diagnosis before the appearance of symptoms or even prenatally. The possibility of genetic recurrence should be addressed in the management of any inherited disorder, and counseling should be offered. As knowledge of pathogenesis is gained, many genetic disorders may become amenable to therapy, either medically, through gene product replacement, or through novel means of gene manipulation.