Genes in Families
Genetic counselling is the process by which patients and relatives at risk of a disorder that may be hereditary are advised of the consequences of the disorder, the probability of developing and transmitting it, and of the ways in which this may be prevented, avoided, or ameliorated. To achieve these aims, an accurate diagnosis and detailed information regarding the family history are essential. The basis for establishing a diagnosis depends on medical history, examination, and investigation. The diagnostic information is combined with the information obtained from the family pedigree to determine the mode of inheritance of the disorder and to calculate the risk of recurrence so that family members can be appropriately counselled.
A family history of a genetic condition may be due to the following:
- A mutation within a single nuclear gene
- A mutation within a mitochondrial gene
- A contiguous gene deletion or duplication involving anything from a few to a large number of genes
- A chromosomal rearrangement resulting in unbalanced products at meiosis
- Multifactorial inheritance
Accurate documentation of the family history is an essential part of genetic assessment, and the best method of recording this information is by constructing a family pedigree. Pedigrees are universally used in patients’ genetic records, journal articles, and textbooks as the means of relaying information in an easily interpreted visual format. Pedigrees also provide the basis for calculations required for both recurrence risk estimation in individual families and linkage analysis in gene-mapping studies.
In a pedigree, squares are used to represent males and circles to represent females. Generations are indicated by Roman numerals and individuals within each generation by Arabic numbers. Despite the universal use of the pedigree as a method of recording information and as an analytical tool, there is still considerable variation in the use of symbols relating to both routine medical information (pregnancy, spontaneous abortion, and termination of pregnancy) and new reproductive technologies (artificial insemination by donor semen, donor ovum, and surrogate motherhood).
While drawing a pedigree, it is usually simplest to start with the person seeking advice (the consultand). In some cases, the consultand will be an apparently healthy relative seeking information about how the condition may affect him/her or his/her offspring. Pedigree details are completed for both sides of the family, including previous and subsequent generations.
Example of a family pedigree for Duchenne muscular dystrophy.
Examples of symbols commonly used in drawing a pedigree.
Conventionally, the paternal lineage is placed on the left and the maternal lineage on the right. Within sibships, individuals are listed from left to right in birth order. The affected person (proband), through whom the family has been ascertained, or the relative seeking advice (the consultand), are normally indicated in the pedigree by an arrow. Name, date of birth, and a summary of relevant medical details should be recorded for all family members. Further information, including medical records, may be required for affected individuals to confirm actual diagnosis. Not all relevant information may be volunteered by the consultand, indeed certain sensitive information—such as a previous termination of pregnancy or illegitimacy—may be deliberately withheld if the accompanying partner is not aware of it. Important details that should be asked about include assisted conception, previous miscarriages, stillbirths, terminations, children adopted into or out of the family, and consanguinity. Ethnic origin should be recorded, as some genetic conditions are more prevalent in particular ethnic groups, and this information may provide a clue to the likely diagnosis. The family history should be completed on both sides of the family for a consultand couple, even if the presenting condition is clearly traced to one side, as unrelated pregnancy losses or handicapping disorders may have occurred on the other side of the family that could have greater reproductive impact for the couple than the disorder for which they are seeking information.
Unifactorial inheritance refers to those disorders that are due to the inheritance of a single mutant gene. The first descriptions of unifactorial inheritance were made by Mendel in 1865, when he published the results of his experiments on the garden pea in his paper “Versuche uber Pflanzen Hybriden” (“Experiments on Plant Hybrids”). His work was largely ignored until it was republished by Bateson in 1901, from which time the term Mendelian inheritance became synonymous with unifactorial inheritance.
From the ratios that Mendel described in his experiments on the garden pea, and the work of subsequent researchers, including Bateson, four main conclusions were drawn:
- Genes come in pairs (Mendel termed them factors), one inherited from each parent.
- Individual genes can have different alleles, some of which (dominant traits) exert their effects over others (recessive traits)—the principle of dominance. In Mendel’s own words, “those characters which are transmitted entire, or almost unchanged in the hybridization, and therefore in themselves constitute the characters of the hybrid, are termed the dominant, and those which become latent in the process, recessive.”
- At meiosis, alleles segregate from each other with each gamete receiving only one allele—the principle of segregation, or Mendel’s first law.
- The segregation of different pairs of alleles is independent—the principle of independent assortment, or Mendel’s second law.
With time, these principles have had to be modified. For example, although most genes come in pairs, for genes on the sex chromosomes males have only one allele; that is, they are termed hemizygous. Also, although alleles of a gene on different chromosomes show independent assortment, genes that are physically close together on the same chromosome do not—a phenomenon that has allowed mapping of genes in the human genome through linkage studies. These principles, however, still form a useful set of rules designed to explain the inheritance of many inherited characteristics and disorders.
Diseases inherited in a Mendelian fashion are categorized according to whether the gene is on an autosome or a sex chromosome and whether the trait is dominant or recessive.
Fundamental to the understanding of Mendelian inheritance are the concepts of dominance and recessiveness. Dominance is not a property intrinsic to a particular allele, but describes the relationship between it and the corresponding allele on the homologous chromosome. If the phenotypes associated with the genotypes AA and AB are the same but differ from the phenotype of BB, allele A is dominant to allele B and, conversely, allele B is recessive to allele A. Therefore, allele A manifests in the heterozygous state. An example of a dominant disease allele is that of Huntington disease, with most individuals affected with the disease being heterozygous for a mutant allele. However, individuals have been identified who have been shown by molecular techniques to be homozygous for the mutant allele by virtue of both parents being affected with Huntington disease. Such individuals do not appear different phenotypically from heterozygotes for the disorder.
If the phenotype of the heterozygous state, AB, is intermediate between the phenotypes of AA and BB, allele A is said to be incompletely dominant or semidominant to allele B. The skeletal dysplasia achondroplasia causes rhizomelic shortening of the limbs, a characteristic facies with midface hypoplasia, exaggerated lumbar lordosis, limitation of hip and elbow extension, genu varum, and trident hand. It was conventionally thought to be due to a dominant allele, but homozygotes for the mutant gene have a much more severe skeletal dysplasia, resulting in early death from respiratory obstruction due to a small thoracic cage and neurologic deficit due to hydrocephalus. Therefore, achondroplasia is an example of incomplete or semidominance. Homozygotes for most dominant mutant alleles causing human genetic diseases occur so rarely that it is not known whether they exhibit complete or incomplete dominance.
If the phenotype of AB displays the phenotypic features of both the homozygotic states, then alleles A and B are said to be codominant. The human ABO blood group system exhibits codominance. The system consists of three alleles: A, B, and O. Both A and B are dominant in relation to O, and therefore blood group A can have the genotype AA or AO. Blood group B can have the genotype BB or BO. However, neither A nor B shows dominance over the other, and therefore individuals with the genotype AB have the phenotypic characteristics of both blood group A and blood group B.
Most mutations result in an allele that is recessive to the wild-type allele; the phenotype is therefore only expressed in the homozygous state. This is because most mutations result in an inactive gene product, but the reduced level of activity due to the remaining wild-type allele is sufficient to achieve the effects of that gene product. An example is a gene for an enzyme that is only required in small amounts as a catalyst for a metabolic pathway. Although in some recessive inborn errors of metabolism it is possible to identify heterozygotes, more often than not, the only way to identify carriers is by direct mutation analysis, using molecular genetic techniques.
There are several mechanisms by which a mutation can lead to a dominant mutant allele whose phenotype is expressed in the heterozygous state.
For most mutant alleles, loss of function will usually exhibit recessive behaviour. Where a reduced amount or reduced activity of the gene product results in the phenotypic features, this is termed haploinsufficiency (e.g. in a critical rate-limiting step of a metabolic pathway). Any reduction in the amount of gene product will result in that pathway not being able to function at full activity. The same appears to apply to regulatory genes that could have a threshold level of activity. PAX3 is a gene coding for a DNA-binding protein, and point mutations in the gene result in Waardenburg syndrome type 1, characterized by deafness and pigmentary disturbances. Certain mutations in PAX3 have been shown to abolish all protein functions of that allele, so the phenotype must be due to a dosage effect as it manifests in the heterozygous state.
Another example in which the quantitative amount of a gene product is important is the genes that produce proteins in large quantities. An example is the gene for C1 esterase inhibitor, mutations of which cause the disorder hereditary angioneurotic edema. C1 esterase inhibitor is removed rapidly from the circulation at a rate independent of its concentration. Therefore, although heterozygotes produce 50% of the normal amount, they have only 15–20% of the normal amount in the circulation, leading to the clinical manifestations of the disorder.
This mechanism involves an excess of gene product leading to a disease phenotype. Although gene dosage of critical regions or genes has been invoked as the cause for the phenotypic features associated with autosomal trisomies, there are a few examples involving single-gene disorders. One example involves the PMP22 gene, which codes for the peripheral myelin protein 22. Duplication of the DNA sequence of one allele is associated with hereditary motor and sensory neuropathy type 1A.
Ectopic or temporally altered messenger RNA is expressed when a mutation occurs that affects the time or place of gene expression and usually involves a regulatory part of the gene. For example, during development in erythroid precursor cells, there is a switch from the production of γ-globin to the production of the δ-globin and β-globin. This switch is controlled, at least in part, by the binding of transcription factors to the γ-globin promoter. Point mutations in the globin-promoter region prevent the normal switch, resulting in the disorder of hereditary persistence of foetal haemoglobin.
Mutations can lead to proteins with a prolonged half-life or proteins that have lost their normal constitutive inhibitory regulatory activity. If a mutation occurs in a part of a gene that codes for the protein sequence acting as the recognition site for proteolytic degradation, this will not take place, with the protein remaining active. Many proteins possess domains that allow their activity to be reversibly inhibited. For example, skeletal muscle sodium channels undergo voltage-sensitive regulation, and mutations in the geneSCN4A that codes for the α subunit of the sodium channels result in the disorder hyperkalaemic periodic paralysis, characterized by muscle myotonia and paralysis due to loss of regulatory inactivation of the sodium channel.
If a mutant allele interferes with the wild-type allele, this is termed a dominant-negative mutation. This could occur in a multimeric protein in which a mutant subunit has an intact binding domain but altered catalytic activity, affecting the function of the entire multimer. If a protein is a dimer, one mutant and one wild-type allele would result in only 25% normal dimers, with up to a 75% reduction in activity.
Many structural proteins are multimers (e.g. the various types of collagen proteins). Each of the collagen subunit genes has a central portion coding for repeating tripeptide units that are essential for the assembly of the collagen molecule. The disease osteogenesis imperfecta is caused by point mutations in the central portion of one of the collagen subunit genes COL1A1 or COL1A2 leading to a structural deformation that causes disruption of the whole collagen protein.
Toxic protein alterations are mutations that cause structural alterations in proteins, thus disrupting normal function and leading to toxic products that poison the cell. An example is hereditary amyloidosis, in which mutations in the transthyretin gene lead to resistance to proteolysis, and hence to increased stability of the protein. The protein then undergoes multimerization and accumulates in the cell as fibrils, causing disruption of the cell.
Some mutations have been found to confer a new function on a gene product. For example, a fatal bleeding disorder was found to be caused by a missense mutation in the α 1 -antitrypsin gene, in which methionine was replaced by arginine at position 358, the effect of which was to convert α 1 -antitrypsin, normally an inhibitor of elastase, into an inhibitor of thrombin. This thrombin inhibitory activity was not compensated for by an increase in endogenous coagulant production, resulting in a severe bleeding disorder.
The mechanisms described so far show how mutations can cause dominant effects at a cellular level by the effects on the proteins produced. It is possible to have mutations that show a dominant pattern of inheritance in families, yet are recessive at the cellular or molecular level; that is, the gene is inactivated but has no other effect. The classic example of this is the retinoblastoma gene RB1, inactivation of which can lead to the formation of the developmental eye tumour retinoblastoma. Families can show a dominant mode of inheritance for this disorder, yet cells heterozygous for the mutation are completely normal, the mutation itself being recessive.
The dominant pattern of inheritance of familial retinoblastoma is the result of transmission of a first mutation with a second somatic mutation occurring in the normal allele of at least one retinal cell during a critical period of development, leading to the formation of a retinoblastoma—the “two-hit” hypothesis. There are several ways in which the normal allele in somatic cells can be inactivated. These include point mutations, deletions, translocations, and mitotic nondisjunction, resulting in the loss of a whole chromosome. It is now known that the two-hit hypothesis applies to most of the dominantly inherited familial cancer syndromes in which the germ line mutation in a tumour suppressor gene is recessive and a mutation in a somatic cell in the corresponding allele leads to the development of a tumour.
Autosomal-dominant inheritance refers to disorders caused by genes located on the autosomes, thereby affecting both males and females. The disease or mutant alleles are dominant to the wild-type alleles, so the disorder is manifest in the heterozygote (i.e. an individual who possesses both the wild-type and the mutant allele).
CHARACTERISTICS OF AN AUTOSOMAL DOMINANT INHERITED DISORDER
|Successive or multiple generations in a family are affected.|
|Males and females are both affected in approximately equal proportions.|
|Males and females can both be responsible for transmission.|
|There is at least one instance of male-to-male transmission.|
Pedigree consistent with autosomal-dominant inheritance.
Since individuals with autosomal-dominant disorders are heterozygous for a mutant and a normal allele, there is a 1-in-2 (50%) chance a gamete will carry the normal allele and a 1-in-2 (50%) chance a gamete will carry the mutant allele. Assuming that the individual’s partner will contribute a normal allele, there is a 1-in-2 (50%) chance that the offspring, regardless of sex, will inherit the disorder with each pregnancy.
Recurrence risks in autosomal-dominant inheritance.
There can be marked variability in the clinical manifestations of autosomal-dominant disorders, and they can demonstrate reduced penetrance (i.e. not every person with the mutant allele shows features of the disorder). The penetrance of a disorder is an index of the proportion of individuals with a mutant allele who manifest the disorder. An allele is said to be nonpenetrant if an individual known to be heterozygous for the allele, either by pedigree analysis or by molecular investigation, shows no signs of the disorder when subjected to appropriate clinical investigation. The penetrance of some genes is dependent on the age of the individual, as in Huntington disease, in which the penetrance is age dependent or is said to show delayed penetrance.
The penetrance (P) of a disorder is usually expressed as a proportion or percentage of the individuals carrying the gene who develop the disorder. If the P value is known for a particular condition, the risk to the offspring of an apparently unaffected individual can be calculated. In practice, the risk is usually less than 10%, as an unaffected relative is not likely to carry the gene if the penetrance is high, and a gene carrier is not likely to develop the disorder if the penetrance is low.
The expressivity of a gene is the degree to which a particular phenotype is expressed in an individual. Many autosomal-dominant diseases show variable expressivity such that individuals in the same family who carry an identical mutation can vary considerably in the severity of their disorder. For example, in the autosomal-dominant disorder neurofibromatosis type 1, the number of neurofibromas that an individual develops can vary dramatically from a few to many hundreds even within the same family. The variability seen in autosomal-dominant disorders may present as both inter- and intrafamilial differences. Intrafamilial variability may reflect the action of modifying genes, but interfamilial variability is more likely to be due to allelic heterogeneity at a single locus. A problem encountered in genetic counselling is that a mildly affected individual, such as a parent with only skin manifestations of tuberous sclerosis, may have a severely affected child. This situation is seen in many autosomal-dominant disorders, in that the mildly affected individuals are more likely to reproduce than the severely affected individuals.
A disorder is said to demonstrate anticipation if the phenotype of the mutant allele increases in severity as it is passed down the generations. An example of a disorder that demonstrates anticipation is myotonic dystrophy. A typical three-generation family with myotonic dystrophy showing anticipation is shown below. Another example of anticipation is seen in Huntington disease, in which the onset of symptoms is often seen to occur earlier with each succeeding generation.
Pedigree for a family with myotonic dystrophy demonstrating anticipation.
Sex influence involves the expression of an autosomal allele that occurs more frequently in one sex than the other. An example in humans is gout, with males affected more frequently than females until after the menopause, an effect probably mediated by hormonal differences.
Some traits are manifested only in individuals of one sex, an extreme situation known as sex limitation. This can occur when a gene affects an organ only possessed by one of the sexes (e.g. unicornuate uterus or ovarian cancer).
Pleiotropy refers to the phenomenon in which a single gene is responsible for a number of distinct and seemingly unrelated phenotypic effects. For example, the allele causing neurofibromatosis type 1 can produce abnormalities of skin pigmentation, neurofibromas of the peripheral nerves, short stature, macrocephaly, skeletal abnormalities, and fits. Each of the pleiotropic effects of an allele can show reduced or nonpenetrance and variable expressivity.
Some of the underlying mechanisms accounting for reduced penetrance and variable expressivity have been known for some time, while others are only now becoming apparent with the advent of modern molecular genetic techniques.
Environmental influences can affect the expression of genes. This can involve factors in the internal environment, such as hormones, or in the external environment, such as the effect of certain drugs and as barbiturates in acute intermittent porphyria. This disorder is characterized by attacks of abdominal pain, constipation, and psychiatric disturbances and is due to a mutation in a gene coding for an enzyme involved in heme biosynthesis. Attacks can be precipitated by certain drugs, including phenobarbital and the sulphonamides. Avoidance of these precipitating factors will result in nonpenetrance of the disorder.
Retinoblastoma has already been mentioned as an example of one of the familial cancer genes in which a “second-hit” somatic mutation needs to occur in order for the disorder to manifest.
There are a group of dominant genetic disorders in which the mutation is an unstable DNA triplet repeat sequence that expands in successive meioses and whose size correlates with the severity of the disorder. This accounts for the anticipation seen in such disorders as myotonic dystrophy and Huntington disease.
Information from studies of inbred animals suggests that it is likely that the genetic background of an individual (i.e. the nature of particular alleles at other loci that interact with a mutant allele) will influence the penetrance and expression of a disorder. Analysis of the effect of this type of interaction is complex and is poorly understood at present, although examples are beginning to be delineated. For example, a study suggests that the alleles of the SMAD3 gene, which encodes a key regulatory protein in the transforming growth factor beta signalling pathway and is known to interact indirectly with the breast/ovarian cancer gene BRCA2 may contribute to an increased risk of breast cancer in BRCA2 mutation carriers.
While nonpenetrance can be a possible cause of a dominant disorder arising in the offspring of completely normal parents, an alternative explanation is that a mutation has arisen during the transmission of the gene; that is, it represents a new or de novo mutation. This appears to be more common for certain disorders than others. For example, in achondroplasia both parents are of normal stature in 80% of families. This also reflects the reduced reproductive fitness of adults with achondroplasia. The observation that achondroplasia occurred more frequently in last-born children of a sibship was suggested by Penrose as attributable to increased paternal age associated with new mutations, on the basis that “older germ-cells, possessed by older parents, might be more likely to show deterioration in the form of genetical changes”.
Some dominant mutations are universally lethal before the affected individual reaches reproductive age, and therefore are always seen as new dominant mutations. Several lethal disorders that were previously considered to be recessive are now known in many instances to be due to new dominant mutations, such as the perinatal lethal form (type II) of osteogenesis imperfecta .
New dominant mutations that occur during gametogenesis are associated with a negligible recurrence risk for future siblings. However, if the disorder is compatible with survival to reproductive age (and the possibility of reproduction), the recurrence risk for the offspring of the affected individual is 50%.
In the case of certain new dominant mutations, there is a small but significant risk that a second child will be affected despite both parents being clinically normal. The ability to determine the molecular basis of many of these disorders has shown that this finding can be explained by the phenomenon of somatic mosaicism. This is possible when there are two genetically different types of cell in an individual, one carrying the mutant allele and the other not. If two genetically different types of cells occur within the gonads, this is referred to as gonadal mosaicism. However, there may be no mosaicism present in the gonads, where all cells may carry the mutant gene, and the mosaicism may be present in other tissues. This situation would more correctly be referred to as somatic mosaicism.
The mutational event occurs after fertilization, with the degree and tissue specificity of the mosaicism being dependent on the time when it occurred. There is evidence that this usually occurs early in development, as commitment of primordial cells to the germ line occurs before tissue allocation, and in studies of individuals with gonadal mosaicism, up to 50% also have the mutation in a somatic cell line.
The frequency of gonadal mosaicism differs between disorders. At present, it is unclear why it is a frequent finding; however, it is now known that genetic heterogeneity with several newly discovered auto-recessive forms of IO can account for a good deal of the increased risks to unaffected parents. In some disorders such as fascioscapulohumeral dystrophy and osteogenesis imperfecta, mosaicism is found in approximately 19% and 15% of all cases, respectively, while in others such as achondroplasia, there have only been a few rare case reports. This is an important point to remember while providing recurrence risk advice in genetic counselling.
Somatic mosaicism may also be present in an individual who manifests the phenotype of an autosomal-dominant disorder. A study looking at individuals who were the first members of their families to develop the signs of neurofibromatosis type 2, a disorder characterized by bilateral vestibular schwannomas, estimated that 24.8% of their study cohort were mosaic for the NF2 mutation. Therefore, a new dominant mutation may arise during meiosis as a germ line mutation or postconceptually as a somatic mutation that is then transmitted through the germ line.
Autosomal recessive inheritance refers to disorders due to genes located on the autosomes, but in which the disease alleles are recessive to the wild-type alleles and are therefore not evident in the heterozygous state, only being manifest in the homozygous state. The parents of an individual with an autosomal recessive disorder are heterozygous for the disease allele and are usually referred to as being carriers for the disorder.
CHARACTERISTICS OF AN AUTOSOMAL RECESSIVE INHERITED DISORDER
|Both males and females are affected.|
|The disorder normally occurs in only one generation, usually within a single sibship.|
|The parents can be consanguineous.|
Pedigree consistent with autosomal recessive inheritance.
If a couple are consanguineous, they have at least one ancestor in common in the preceding few generations. This means that they are more likely to carry identical alleles inherited from this common ancestor and could both transmit an identical allele to their offspring, who would then be homozygous for that allele. A consanguineous couple have an increased risk that their offspring will be affected with a recessive disorder. The rarer a particular disease is in a population, the more likely the parents are to be consanguineous. For example, cystic fibrosis is a common autosomal recessive disorder in whites in Western Europe, with an incidence of ~1 in 2000. The incidence of consanguinity in the parents of children with cystic fibrosis is not appreciably greater than that in the general population. By contrast, with very rare autosomal recessive disorders such as alkaptonuria, eight of the first 19 families originally described by Garrod were consanguineous.
When two parents carrying the same disease allele reproduce, there is an equal chance that gametes will contain the disease or the wild-type allele. There are four possible combinations of these gametes, resulting in a 1-in-4 (25%) chance of having a homozygous affected offspring, a 1-in-2 (50%) chance of having a heterozygous unaffected carrier offspring, and a 1-in-4 (25%) chance of having a homozygous unaffected offspring.
Recurrence risks in autosomal recessive inheritance.
When an individual with an autosomal recessive disorder has children, they will only produce gametes containing the disease allele. Since it is most likely that their partners will be homozygous for the wild-type allele, the partners will always contribute a normal allele and therefore all the children will be heterozygous carriers and unaffected. If, however, an affected individual has children with a partner who happens to be heterozygous for the disease allele, there will be a 50% chance of transmitting the disorder, depending on whether the partner contributes a disease or a wild-type allele. Such a pedigree is said to exhibit pseudodominance.
Recurrence risks for an individual with an autosomal recessive disorder and a normal partner.
Recurrence risks for an individual with an autosomal recessive disorder and a carrier partner.
Pedigree of an autosomal recessive disorder showing pseudodominance.
In autosomal recessive disorders, the difficulty lies not with risk estimation, but in determining the underlying mode of inheritance, as these disorders usually present as isolated cases with little contributory information to be gleaned from the pedigree. Carrier risks to other relatives can be calculated from the pedigree, and carrier testing may be appropriate for disorders with a high gene frequency or when consanguineous marriages are planned. The risk to members of the extended family for having an affected child will depend on their own risk calculated from pedigree data and the population-based carrier risk appropriate to their spouse. Higher risks in consanguineous marriages are dependent on the degree of relationship between the partners. In disorders of unknown genetic aetiology, consanguinity suggests, but does not prove, an autosomal recessive mode of inheritance.
It is not unusual in some recessive disorders, such as sensorineural deafness, for two affected individuals to have children. Assortative mating occurs in such instances because of social circumstances in which individuals with the same disability, such as deafness or visual impairment, are often educated together or share the same social facilities. If their disorder was due to a mutation in the same autosomal recessive gene, all their offspring would be affected. In a number of studies involving the offspring of parents with inherited sensorineural deafness, however, a significant proportion of such matings led to offspring with normal hearing. Although in some instances this could be due to other causes (e.g. acquired causes being mistaken for inherited deafness), in most instances, the gene causing the deafness in the two parents is different, a phenomenon known as genetic heterogeneity. Each parent will transmit the mutant allele for their own deafness, but a wild-type allele of the gene involved in their partner’s deafness. Therefore, the child is heterozygous for the two mutant alleles, what is known as a double heterozygote. This type of genetic heterogeneity involving different genes is known as locus heterogeneity.
Different modes of inheritance have also been documented for a number of clinically defined disorders with similar phenotypes. For example, autosomal dominant, autosomal recessive, and X-linked recessive inheritance have all been documented in hereditary spastic paraplegia, hereditary motor and sensory neuropathy, and retinitis pigmentosa depending on the causative gene. Locus heterogeneity makes it difficult to determine risks of recurrence for phenotypes that follow both dominant and recessive inheritance if the mode of inheritance is not clearly defined by the family pedigree. Laboratory analysis of families is also complicated by locus heterogeneity, as direct mutation analysis may not be practical for conditions in which one of the several/many genes is implicated and the suggestion of linkage between a polymorphic marker and the disease in a small family does not necessarily prove that a mutation at that particular locus is responsible.
Heterogeneity can also exist at the same locus; thus, an individual affected with a recessively inherited disorder can have two different mutations in the two alleles of the gene and is often called a compound heterozygote. Most individuals with recessive disorders are compound heterozygotes unless a specific mutation is especially prevalent in a particular population or the affected individual is the offspring of a consanguineous relationship, in which case the allele is likely to be identical by descent. This is known as allelic or mutational heterogeneity. The specific mutations that an affected individual possesses can, in fact, determine the severity of the disorder, as in cystic fibrosis, in which individuals who are homozygous for the most common mutation in the cystic fibrosis gene, ΔF508, have a higher incidence of pancreatic insufficiency. As the underlying mutations causing disease are identified, it is becoming increasingly apparent that in many cases the exact nature of the mutation will determine the phenotype—a phenomenon known as genotype–phenotype correlation.
In some cases, different classes of mutations in a particular gene may act in a dominant or in a recessive manner. For example, a wide variety of phenotypes have been associated with mutations in the LaminA/C gene. These include mutations acting in an autosomal-dominant fashion, causing Emery–Dreifuss muscular dystrophy, limb girdle muscular dystrophy type 1B, dilated cardiomyopathy, familial partial lipodystrophy, and Hutchinson–Gilford progeria syndrome, as well as mutations acting in an autosomal recessive fashion, such as Emery–Dreifuss muscular dystrophy, type 2B1 axonal neuropathy, and mandibuloacral dysplasia.
Uniparental disomy (UPD) refers to the presence of both homologs of a chromosome pair or chromosomal region in a diploid offspring being derived from a single parent. If the two homologs are identical due to an error in meiosis II, this is known as uniparental isodisomy, while if the two homologs are different but still from the same parent due to a meiosis I error, this is known as uniparental heterodisomy. UPD has been reported as a rare cause of the autosomal recessive disorder cystic fibrosis in the offspring of a couple in whom only one parent was a heterozygote carrier of the mutant allele. The affected offspring received both chromosome 7 homologs with the mutant allele from that parent. The recurrence risk in this situation would be negligible.
An autosomal recessive disorder may potentially be due to the inheritance of a mutation from one parent, with a de novo mutation occurring at the same locus on the chromosome inherited from the other parent. This is likely to be a very rare phenomenon, but accounts for some cases of spinal muscular atrophy due to a predisposition to generate deletions in the 5q13 region involving the SMN1 gene. The recurrence risk in this situation would relate to the risk of repeated de novo mutation and therefore negligible.
If a male affected with an X-linked recessive disorder survives to reproduce, he will always transmit his X chromosome with the mutant allele to his daughters, who will be obligate carriers. An affected male will always transmit his Y chromosome to a son, and therefore none of his sons will be affected. A carrier female has one X chromosome with the wild-type allele and one X chromosome with the disease allele; therefore, her sons have a 1-in-2 (50%) chance of being affected, while her daughters have a 1-in-2 (50%) chance of being carriers.
Recurrence risks for a male with an X-linked recessive disorder.
Recurrence risks for a female carrier of an X-linked recessive disorder.
For a female to be affected with an X-linked recessive disorder, her mother would have to be a carrier and her father affected with the disorder. Obviously, this situation is encountered only very rarely. Another possibility is that her mother is a carrier and the X chromosome transmitted by her father undergoes a new mutation. A female can also be affected by an X-linked recessive disorder if she has a single X chromosome (i.e. Turner syndrome), in which case she will be hemizygous for alleles on the X chromosome, like a male.
Early in embryonic development, one of the X chromosomes in females is inactivated in each cell with the result that the female, like the male, has a single functional X chromosome. In individuals with X chromosome aneuploidies, all but one of the X chromosomes are inactivated in each cell. The process of X-inactivation is controlled by a region of the X chromosome, the X-inactivation centre, situated on the proximal portion of the long arm. X-inactivation usually occurs as a random process such that, in approximately 50% of cells in a female, the maternally derived X chromosome is active, and in the other 50%, the paternally derived X chromosome is active. In females who are carriers of an X-linked recessive mutation, one half of the cells will actively express the disease allele. Occasionally, this can be demonstrated clinically. In X-linked retinitis pigmentosa, for example, careful fundoscopic examination of a female carrier can show a mosaic pattern of pigmentation.
A number of mechanisms can lead to nonrandom X-inactivation:
- Chance : Skewed X-inactivation can occur by chance.
- Monozygotic twinning : There have been several reports of monozygotic female twins, both heterozygous for a dystrophin gene deletion, one of whom was a manifesting carrier for Duchenne-type muscular dystrophy and the other an unaffected carrier. It was demonstrated that, in lymphocytes and fibroblasts of the affected twin, the majority of active X chromosomes had the deletion, while in the unaffected twin, most active X chromosomes possessed the intact gene. It has been postulated that the twinning process could have been a consequence of cell surface differences as a result of X chromosome inactivation, leading to the separation of the two cell masses.
- Cytogenetic abnormalities : In females with an X-autosome translocation, the normal X chromosome will be preferentially inactivated, maintaining the diploid state for the autosome involved in the translocation. If the translocation disrupts or interferes with the expression of a gene on the X chromosome, or if the X chromosome involved in the translocation carries a disease allele, that female will manifest the disorder. The finding of X-autosome translocations in manifesting female carriers of Duchenne-type muscular dystrophy was instrumental in the mapping and cloning of the dystrophin gene.
- Elimination of cells expressing the mutant allele : If a gene on the X chromosome is required for cell survival, the normal gene will always be found on the active X chromosome and the defective gene on the inactive X chromosome in the mature cell population, even though X-inactivation occurred as a random process. This has been demonstrated in a number of disorders, including X-linked severe combined immunodeficiency, and can be used in the determination of carrier status for females at risk.
As with autosomal-dominant disorders, gonadal mosaicism is an important phenomenon in X-linked recessive disorders, occurring particularly frequently in Duchenne-type muscular dystrophy, in which it has been shown to occur in both male and female gametogenesis. It is important to take this into account when advising mothers of apparently sporadically affected males with Duchenne-type muscular dystrophy of recurrence risks.
Mutations arising during meiosis may occur in male or female germ-cells. If a particular mutation arises largely in male germ-cells, as in Lesch–Nyhan syndrome and haemophilia A, the majority of mothers of affected boys will be carriers, and risks to sisters will be 50% regardless of how many unaffected brothers they have. In Duchenne muscular dystrophy, the overall mutation rate appears to be equal in males and females, but it has been suggested that most point mutations occur in spermatogenesis, while most deletions arise in oogenesis. In isolated cases, the recurrence risk might therefore be higher in nondeletion cases, as these mothers would be more likely to be carriers.
Offspring of either sex have a 1-in-2 (50%) chance of inheriting the disorder from affected females. The situation is different for males affected by X-linked dominant disorders, whose daughters will always inherit the gene and whose sons cannot inherit the gene. An example of an X-linked dominant disorder is vitamin D–resistant rickets. X-inactivation results in females with this disorder having less severe skeletal changes than those that occur in affected males.
Recurrence risks for a female affected with an X-linked dominant disorder.
Recurrence risks for a male affected with an X-linked dominant disorder.
An exception to the rule that females are less severely affected than males is craniofrontonasal dysplasia, in which heterozygous females have a coronal craniosynostosis and affected hemizygous males do not have a craniosynostosis. The condition is caused by mutations in the ephrin B1 gene, and it has been proposed that, in heterozygous females, patchwork loss of ephrin B1 disturbs tissue boundary formation at the developing coronal suture, whereas in males deficient in ephrin B1, an alternative mechanism maintains the normal boundary.
Genomic imprinting is a phenomenon in which gene expression depends on parental origin. Imprinting is due to an epigenetic mechanism, in which the primary DNA sequence of the gene is not changed, but transcriptional regulation is affected by mechanisms such as DNA methylation, histone acetylation, and histone methylation. In contrast to the biallelic expression of most genes, imprinted genes demonstrate monoallelic expression. This process modifies the transmission and expression of certain genetic diseases and should be borne in mind as a possible mechanism underlying disorders that do not follow typical Mendelian inheritance. Many disorders due to defects affecting imprinted genes arise de novo, but they can also be familial. For example, the familial paraganglioma syndrome is due to mutations in the succinate dehydrogenase subunits B, C, and D. SDHD is an imprinted gene with the paternal allele active in each cell and the maternal allele inactive. The mutation is dominant and an affected parent (male or female) has a 1-in-2 chance of passing it on to each child, but the child is at increased risk to develop paragangliomas if only the mutation is inherited from the father. Conversely, the Angelmans gene UBE3A is imprinted in the opposite way so that only the maternal allele is active.
Pedigree showing familial paragangliomas and the effects of imprinting of the SDHD gene.
Prader–Willi syndrome (PWS) and Angelman syndrome are probably the best known examples of disorders due to imprinted genes. These disorders affect different genes in the imprinted region at 15q11-13. Loss of expression of paternal candidate genes in this region leads to PWS, and the loss of expression of the maternal UBE3A (ubiquitin protein ligase E3A) gene leads to Angelman syndrome. The underlying mechanism is often a de novo deletion or point mutation within the imprinted gene itself, but some cases are due to epigenetic mechanisms such as UPD or the consequence of an imprinting centre defect. Maternal UPD results in PWS and paternal UPD in Angelman syndrome. Some familial cases of Angelman syndrome are due to mutations in the gene UBE3A. A mutation inherited from a mother will cause Angelman syndrome, but when inherited from the father the condition will not manifest. The imprint is reset during male and female gametogenesis. Therefore, in familial Angelman’s, if an unaffected female carries a mutation in UBE3A on her paternal allele it is reset at gametogenesis so that it is now on the active allele and the children who inherit it will have Angelman syndrome.
Some genes show tissue-specific imprinting; for example, the GNAS gene, which causes Albright hereditary osteodystrophy (AHO). AHO is due to G(s)alpha inactivating mutations, imprinted in a tissue-specific manner, with expression in the proximal renal tubules, thyroid, pituitary, and ovaries being from the maternal allele. Maternally inherited mutations lead to AHO with endocrine involvement (pseudohypoparathyroidism type 1A), whereas paternally inherited mutations lead to AHO alone. Pseudohypoparathyroidism type 1B (parathormone resistance without AHO) can be caused by a deletion of the imprinted promotor regions of the gene that control gene expression.
An increased incidence of imprinted disorders, notably Beckwith–Wiedemann Syndrome, has been reported in children conceived by in vitro fertilization or ICSI techniques, suggesting that the process of resetting the parental imprint may be perturbed by specific elements of assisted reproductive methodology.
Genetic or locus heterogeneity by which different genes can cause clinically identical disorders has been discussed above. However, these cases are considered to be monogenic in that, in any one family, only one locus is thought to be defective. Reports of families with retinitis pigmentosa in which the affected individuals are heterozygous for mutations in two different recessive genes (i.e. double heterozygotes) suggested the possibility of a previously undescribed mode of inheritance, known as digenic inheritance.
Although the families described were initially thought to be compatible with autosomal dominant retinitis pigmentosa with reduced penetrance, the families showed a number of unusual features:
- In each family, the disease originated in the offspring of unaffected individuals.
- Affected individuals transmitted the disorder statistically significantly to fewer than 50% of their offspring.
On molecular testing, it was found that both the affected and the unaffected individuals carried a mutation in the peripherin/RDS gene. Affected individuals were also heterozygous for a mutation in the ROM1 gene. These genes encode two of the polypeptide subunits of an oligomeric transmembrane protein complex present at the photoreceptor outer segment disc rims. Mutant peripherin/RDS protein can assemble with wild-type ROM1 to form structurally normal complexes, but cannot assemble with mutant ROM1 protein. Therefore, only the combination of the two heterozygous mutations is pathogenic.
Digenic inheritance has also been suggested in inherited sensorineural deafness and in arrhythmogenic right ventricular cardiomyopathy. Whether this is a true phenomenon and how common is it is still to be fully elucidated.
Bardet–Biedl syndrome (BBS) is a multisystem disorder characterized by obesity, retinal degeneration, polydactyly, gonadal, and renal malformations, and behavioural and developmental problems, with a population incidence of 1 in 14,000 to 1 in 16,000. Initially, segregation studies suggested that the inheritance pattern was autosomal recessive. BBS is a genetically heterogeneous condition and so far 14 genes have been identified accounting for about 70% of cases. BBS families demonstrate great intra- and interfamilial variation in the phenotype. A more complex inheritance pattern has emerged with the report that about 5% of cases have three mutations, two in one BBS gene, and a third in another BBS gene. There are a number of possible explanations for this finding. One is triallelic inheritance with the third allele affecting the variability in the phenotype that is seen in BBS. Alternatively, the third allele may represent a heterozygous recessive mutation which has no effect on the phenotype or a rare polymorphism.
The nuclear chromosomes are not the only source of coding DNA sequences within the cell. Mitochondria possess their own DNA, which, as well as coding for mitochondrial transfer RNA and ribosomal RNA, also carries the genes for 12 structural proteins that are all mitochondrial enzyme subunits. Mutations within these genes have been shown to cause disease (e.g. Leber’s hereditary optic neuropathy, LHON). The inheritance pattern of mitochondrial DNA is, however, very different from that of nuclear DNA. Mitochondria are exclusively maternally inherited. Therefore, mitochondrial mutations can only be transmitted through females, although they can affect both sexes equally. Genetic assessment of mitochondrial diseases is complicated by the great variability of these disorders and a pedigree that is seldom conclusive of maternal transmission, as many cases are due to sporadic mitochondrial DNA mutations or are determined by nuclear genes.
In Leber hereditary optic neuropathy, the pattern of maternal inheritance is well documented. The commonest mitochondrial DNA mutation is a point mutation in base pair 1178 of the ND4 gene of complex I. Two other common mutations have also been described (G3460A and T14484C), which also involve genes encoding complex I subunits of the respiratory chain. More than 95% of the cases are the result of one of these three mutations. Women with the mutation will transmit it to all offspring, but only around 1 in 2 males and 1 in 10 females with the mutation develop loss of vision. Affected and carrier males do not transmit the mutation to their offspring.
Pedigree showing Leber’s Hereditary Neuropathy due to a mitochondrial mutation.
A single cell contains many copies of the mitochondrial genome, and heteroplasmy for a mitochondrial mutation is usual. Heteroplasmy is the presence in the cell of both the wild-type and mutated copies of the gene. The proportion of mutant mitochondrial DNA in leukocytes is not a reliable indicator of which individuals will develop symptoms. As with other mitochondrial disorders, this presents problems in counselling asymptomatic individuals known to carry the mutation. The recurrence risks for mitochondrial disorders are difficult to determine. Large-scale mitochondrial DNA deletions, as in Kearns–Sayre syndrome, are usually sporadic, while point mutations, as in Leber hereditary optic neuropathy, are more likely to be maternally transmitted.
Factors that indicate the possibility of a familial chromosomal disorder in a pedigree include a family history of infertility, multiple spontaneous abortions, and malformed stillbirths or live-born infants with multiple congenital malformations occurring in a pattern that does not conform to that of Mendelian inheritance. The possibility of a subtle rearrangement not detected by routine cytogenetic analysis but leading to recurrent chromosome imbalance in offspring should be borne in mind and a number of techniques can be used to identify this situation, such as fluorescence in situ hybridization (FISH), telomere FISH, comparative genomic hybridization, and microarray analysis.
Pedigree illustrating the segregation of a balanced and unbalanced reciprocal chromosome translocation.
It is difficult to give precise reproductive risks to a person known to carry a balanced reciprocal translocation. The risk of viable chromosome imbalance will depend on the size and origin of the unbalanced segment generated, and there is an association between the mode of ascertainment and the risk of recurrence although this is not absolute. Ascertainment through the live birth of an abnormal baby is associated with a higher risk than ascertainment through miscarriage or infertility. Information from an individual pedigree is seldom sufficient to derive a family-specific risk based on reproductive outcomes in known carriers, and the estimation of risk is mainly based on the likely segregation pattern at meiosis and the chance of an imbalance being viable.
The majority of chromosomal disorders present as sporadic cases. Although very low, the recurrence risk may be increased above the normal age-related risk. After the birth of a child with Down syndrome due to trisomy 21 to a mother under the age of 35 years, the recurrence risk is 0.5% for trisomic Down syndrome and 1% for all chromosomal abnormalities. For mothers over the age of 35 years, the birth of a child with Down syndrome does not appear to increase recurrence risk significantly compared to the population-related risks. The risk for other family members is not increased.
A large group of relatively common disorders have a considerable genetic predisposition, but do not follow clear-cut patterns of inheritance within families. These include many birth defects and chronic diseases of later life. Liability to these disorders appears to be due to the interaction of several genetic and environmental influences. Clusters of cases within a family may simulate a Mendelian pattern of inheritance. It is often difficult to be precise about recurrence risks in multifactorial disorders. The risk is greatest among first-degree relatives, is usually small for second-degree relatives, and often does not exceed the general population risk for third-degree relatives. The risk increases when multiple family members are affected. Risks are also affected by the incidence of the disorder in the general population and the sex of the patient and relatives in disorders that have unequal sex incidence, such as pyloric stenosis and Hirschsprung disease. In these disorders, recurrence is higher in the siblings of the less affected sex probably reflecting a greater genetic load to cause the disease in that sex. The severity of the disorder also influences risk; for example, the recurrence risk is greater for bilateral cleft lip and palate than for unilateral cleft lip alone.
In many disorders, empirical risks have been derived from family studies that provide data on observed recurrences. Risks derived in this way will be less accurate in disorders that are genetically heterogeneous and may not be applicable to populations different from those in which the family studies were performed. Nevertheless, empirical risks provide a useful basis for discussing levels of risk during genetic counselling. Risk tables for some of the common congenital malformations, such as cleft lip, with or without cleft palate, pyloric stenosis, and neural tube defects have been published. Considerable heterogeneity occurs within groups of disorders traditionally considered to be multifactorial or polygenic, such as diabetes, epilepsy, and congenital heart disease, with a proportion of cases attributable to single-gene defects. Identification of a specific genetic aetiology, such as the presence of a submicroscopic deletion at chromosome 22q11 in some cases of nonsyndromic congenital heart disease, is important in providing genetic advice appropriate to a particular case. Hirschsprung disease provides a good example of a polygenic disorder in which involvement of several single loci have now been identified in a proportion of families. Data from initial family studies suggested sex-modified polygenic inheritance, and empirical recurrence risks have been produced for genetic counselling based on the sex of the index case and relative and the length of the aganglionic segment involved. A mode of inheritance compatible with an incompletely penetrant autosomal-dominant gene was suggested in some families. Linkage to a gene on chromosome 10 was subsequently demonstrated and mutations in the RET oncogene demonstrated. Mutations in other genes, notably the endothelin receptor type B gene, have also been implicated. Thus in a subset of families with Hirschsprung disease, there is a major unifactorial predisposition, a situation likely to be reflected in other multifactorial disorders.
Many patients presenting to the genetics clinic represent isolated cases within the family, and pedigree information does not contribute to defining the mode of inheritance. Genetic advice in such cases depends on reaching a diagnosis in the affected individual and identifying the underlying aetiology. Any of the following situations may apply:
- The disorder may be due to an autosomal-dominant gene defect, arising by new mutation, transmitted through a nonpenetrant or very mildly affected parent, or by a clinically unaffected parent who carries a mosaic germ line mutation. The situation may also represent nonpaternity. Risk to siblings varies between 0% and 50%, depending on the origin of the mutation, while risk to the offspring of the affected individual would be 50%.
- The disorder may be caused by an autosomal recessive gene defect with a 25% recurrence risk for siblings unless due to UPD or new mutation. If the carrier state can be confirmed by molecular or biochemical analysis, cascade screening of other family members may be appropriate when the population carrier frequency is high or consanguineous marriages are planned.
- The disorder may be due to an X-linked gene, usually presenting in a hemizygous male but affecting females if the X-inactivation pattern is skewed or the gene acts dominantly. Isolated cases may represent new mutations, which are frequent in lethal X-linked recessive disorders, or may be transmitted by asymptomatic mothers who are carriers or who have germ line mutations.
- The disorder may be due to a mitochondrial DNA mutation representing a sporadic case or maternal transmission.
- The disorder may be due to a chromosomal abnormality. Many of these, including the common trisomies due to nondisjunction, have a low risk of recurrence, but unbalanced karyotypes due to familial chromosomal rearrangements may carry high risks of recurrence, and investigation of the relatives is required.
- The disorder may be polygenic and recurrence risks depend on the disorder. These are based on empirical data derived from family studies.
- The disorder may have a nongenetic aetiology with no increase in the risk of recurrence, unless due to a teratogenic agent that further pregnancies will also be exposed to.