Clinical Genetics [M.Tevfik DORAK]

Genetics HLA MHC Genetic Epidemiology Population Genetics Glossary Homepage

CLINICAL GENETICS

M.Tevfik Dorak, MD, PhD

Please update your bookmark: http://www.dorak.info/genetics/notes05.html

It is correct that more than 4500 inherited genetic diseases are known (see Mendelian Inheritance Website and NCBI Genes & Diseases Online) but still only 2% of human diseases can be attributed to primarily genetic causes.

Each of us inherits hundreds of genetic mutations from our parents, as they did from their forebears. In addition, the DNA in our own cells undergoes an estimated 30 new mutations during our lifetime, either through mistakes during DNA copying or cell division or, more often, because of damage from the environment. Because most of these are in somatic cells, they are not passed on to the next generation. It is also estimated that each human being is a carrier of around five recessive lethal genes and perhaps even more recessive disease genes, which may pose risk for the offspring of related individuals (such as cousin marriages).

Examples of genetic disorders

I. Single-gene disorders (autosomal or sex-linked): Cystic fibrosis (in the most frequent mutation, the 1480 amino acid-long wild protein is missing a phenylalanine at position 508 and becomes 1479 amino acid-long, however, more than 30 mutations causing a defective protein have been identified), hereditary hemochromatosis (C282Y missense mutation: cysteine at position 282 is replaced by a tyrosine), myotonic dystrophy and Huntington disease (trinucleotide repeat polymorphism with parental origin effect: stronger when it is inherited from the father; it also causes genetic anticipation due to changes in the number of repeats), sickle cell anemia (a point missense mutation), thalassemia (alpha, beta; point ‘stop’ mutations), hemophilia (A, B), phenylketonuria-PKU (point mutation), Duchenne muscular dystrophy (DMD), adenosine deaminase (ADA) deficiency causing severe combined immune deficiency (SCID), Tay-Sachs disease (hexosaminidase A deficiency). In hypermobility type of Ehlers-Danlos syndrome, haploinsufficiency (where one copy is unable to produce the protein in sufficient quantity) due to a 30-kb deletion of tenascin-X (TNXB) gene is responsible for the disease. In Cri-Du-Chat syndrome (5p deletion), the genetic basis of the phenotype is haploinsufficiency for the telomerase reverse transcriptase gene (TERT), which is included in the deleted part of chromosome 5. In the rare disease erythropoietic protoporphyria, haploinsufficiency for ferrochelatase (FECH) contributes to the clinical phenotype but is not the only reason for the disease expression. Dominant negative mutations (where mutation on one copy renders the other copy inactive) are involved in osteogenesis imperfecta type I and autosomal dominant nephrogenic diabetes insidipus. Most single gene disorders can be investigated by prenatal diagnosis using DNA extracted from cells obtained from amniocentesis at 16-18 weeks' gestation or chorionic villus sampling (CVS) at about 10-12 weeks' gestation.

mtDNA disorders are another group of single-gene diseases: Leber hereditary optic neuropathy (LHON), neurologically-associated retinitis pigmentosa (NARP), myoclonic epilepsy and ragged red-fiber disease (MERRF), maternally inherited myopathy and cardiomyopathy (MMC) (See Taylor & Turnbull, 2005).

II. Diseases with multifactorial etiology (also called complex genetic disorders): Insulin-dependent diabetes mellitus (type 1 diabetes/T1D or IDDM), type 2 diabetes (T2D or NIDDM), cardiovascular disorders, multiple sclerosis (MS), rheumatoid arthritis (RA), cancer, autism, and schizophrenia are examples. These diseases show familial aggregation but not strong familial segregation. See Complex Diseases in Human Molecular Genetics.

III. Chromosomal disorders (numerical or structural) are another group of genetic disorders (see also Merck Manual):

a) numerical (aneuploidy): Trisomy 21 (Down syndrome; 47, +21), Trisomy 13 (Patau syndrome; 47, +13), Trisomy 18 (Edwards syndrome; 47, +18), Monosomy X (Turner syndrome; 45, X), Klinefelter syndrome (47, XXY)

b) structural: deletion (Di George syndrome: del 22; Cri-du-chat syndrome: 5p-)

c) others: uniparental disomy (uniparental copy of chromosome 15q: Prader-Willi syndrome ‘maternal’ or Angelman syndrome ‘paternal’, see below); Mosaicism (skewed X-chromosome inactivation: color blindness)

Mendelian segregation patterns (mode of transmission) in single gene disorders:

In single gene disorders (as opposed to multifactorial-complex disorders), the mutation's population frequency is low, its penetrance is high, and the contribution of environment is lower with notable exceptions of PKU and few others. Genetic counseling is important for personal-decision making involving reproductive issues. In assigning the inheritance pattern of a genetic disease, important features of pedigrees to consider are: sex ratio in affected individuals, male-to-male transmission, mother-to-son transmission (XLR) - mother-to-son/daughter transmission (mtDNA); proportion of offspring (proportion of siblings) affected and consanguinity. Situations that can confound the interpretation include: quasi-dominant inheritance (a homozygote and heterozygote (carrier) mating for a recessive disease); autosomal dominant sex-limited inheritance, de novo mutations (including new mutation on the second X-chromosome in a woman carrier for a X-chromosome mutation), phenocopy, skewed X-chromosome inactivation (atypical Lyonisation; one of the reasons for recurrent miscarriages-see Lanasa, 2001 and Brown, 1999) and mosaicism.

Autosomal recessive (AR): In general, two unaffected (carrier) parents produce diseased children of both sexes (the risk is 25%); some degree of consanguinity is usually involved; there is a horizontal pattern in the pedigree. There does not have to be a diseased individual in the pedigree. It is even possible that one of the parents is genetically normal despite having a homozygous child for a recessive allele. This may be due to de novo mutations or uniparental disomy (Zlotogora, 2004). Autosomal recessive disorders are usually common in populations with high level of inbreeding (restricted gene pool). Examples are Tangier disease in Tangier Island off the coast of Virginia, USA; many genetic disorders in Ashkenazi Jews (Tay-Sachs Disease, Gaucher disease, Fanconi anemia, Niemann-Pick Disease); congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency in Yupik Eskimos; CAH due to 11-beta hydroxylase deficiency in Moroccan Jews; and thalassemias (beta & alpha) in Cyprus and Sardinia. Heterozygotes for the recessive disease genes (carriers) are usually clinically unaffected but some biochemical evidence for carrier status may be found: heterozygotes for cystinosis are always asymptomatic, but intracellular cystine levels are up to 10 times greater than the normal amounts (homozygotes have levels 100-fold greater than normal). Similarly, heterozygotes for 21-hydroxylase gene (CYP21A2) mutations are usually asymptomatic but their adrenal sex steroid hormone levels may be found higher than normal after stimulation (see Figure 2 in an online review). AR diseases do not usually show variable expressivity within a given family.

Autosomal dominant (AD): Every patient has an affected parent; the risk of transmission of the disease to the offspring is 50%; no sex preference; two affected parents may have a healthy child (25% chance); homozygotes may have a more severe disease or may not exist (due to early, including embryonic, lethality, as in acute intermittent porphyria); normally there is no generation skipping resulting in a vertical pattern in the pedigree. If both parents appear to be normal for an AD disease but the offspring has it, the possibilities are as follows: biologic parents are different; incomplete penetrance in (affected) parent; phenocopy (the disorder is not genetically determined but mimicking one) or genocopy (having a phenotype caused by an AD disease but caused by a different genetic mechanism); de novo mutation; gonadal mosaicism for the disease mutation in one parent. Examples of AD diseases are adult polycystic kidney disease, Huntington chorea, and von Willebrand disease. AD diseases are usually due to mutations in receptor proteins (familial hypercholesterolemia) or structural proteins (hemoglobin C, procollagen). Expressivity may show gradual variation in successive generations.

X-linked recessive (XLR): Affected males transmit the gene (not necessarily the disease) to all daughters but not to sons (and to half of their grandsons); hemizygous males and homozygous females are affected, thus, it is more frequent in males; all affected females have an affected father and their mother is either an obligate carrier or affected. Female carriers of X-linked recessive diseases are generally asymptomatic but exceptions occur (manifesting carriers) as reported in hereditary hypophosphatemia with vitamin D-resistant rickets, Duchenne muscular dystrophy (DMD) and Wiskott-Aldrich syndrome; in other words, XLR diseases are more severe in males. Transmission is usually from carrier females to sons, which results in a criss-cross pedigree. Sporadic XLR disease is due to spontaneous mutation (1/3) or carrier mother (2/3). If an XLR disease occurs in a female, possible explanations include: atypical Lyonisation (Puck & Willard, 1998), new mutation on the X-chromosome of a carrier woman; 45,X; Xp deletion; translocation involving the mutation on the X-chromosome and an autosomal chromosome (no inactivation of the translocated X-chromosome part).

X-linked dominant (XLD): All daughters but no son of affected males are affected (no male-to-male transmission); a heterozygous affected woman transmits the disease to half of her children with males and females at equal risk; on average, twice as many females as males will be affected. One reason contributing to this ratio is the in utero lethality in males carrying X-linked mutations (see below). If the sex ratio in affected offspring is ignored, the pedigree has a vertical appearance (like AD transmission).

Mitochondrial transmission: In this exceptional situation, mitochondrial diseases are transmitted only from mothers to both sexes equally (a review on Clinical Mitochondrial Genetics by Chinnery et al; Spectrum of Mitochondrial Disease by Naviaux; see also Taylor & Turnbull, 2005). Leber optic atrophy, Kearns-Sayre syndrome, diabetes-deafness syndrome, MELAS syndrome, and Wolfram syndrome (mitochondrial form) are examples of mitochondrial DNA (mtDNA) disorders.

Common genetic disorders

Autosomal recessive (AR) diseases: Cystic fibrosis (CF), oculocutaneous albinism (OCA), hereditary hemochromatosis (HH), congenital adrenal hyperplasia (multiple mutations), sickle cell disease, beta thalassemia (multiple mutations), AT, ADA deficiency / SCID. The important characteristics of AR diseases, ethnic group specificity and increased risk with consanguinity, are seen in most of these diseases. In general, most common recessive disease genes are those encoding metabolic enzymes (Jimenez-Sanchez, 2001).

Autosomal dominant (AD) diseases: Brachydactyly (the first Mendelian trait identified in humans), Huntington disease, familial hypercholesterolemia, familial polycystic disease, one type of Alzheimer disease, neurofibromatosis type 1 and achondroplasia. Acute intermittent porphyria is an interesting autosomal dominant disease in that a heterozygous mutation causes a total enzyme deficiency. In general, most common autosomal dominant diseases are due to mutations in transcription factor genes (Jimenez-Sanchez, 2001).

X-linked diseases: Many conditions - including hemophilia A and B, G6PD deficiency, red-green color blindness, hereditary myopia, night blindness and ichthyosis - are sex-linked traits in humans. The most common XLR disease is DMD (the DMD gene is 2.4 Mbp with 80 exons. Average exon size is 150 bp). Some other common XLR diseases are hemophilia A, hemophilia B, fragile X syndrome, non-specific X-chromosomal mental retardation (MRXS3) and X-linked Bruton agammaglobulinemia (BTK). Testicular feminization / androgen insensitivity syndrome is another X-linked trait that causes XY individuals to develop into phenotypic females. Prostate cancer has an X-linked form (HPCX).

X-linked dominant (XLD) diseases / traits: Xg(a) blood group (usually shown as X_g), Vitamin D-resistant rickets (X-linked), incontinentia pigmenti and Rett syndrome (the last two are X-linked dominant; therefore, lethal in hemizygous males and only seen in females -see below for more examples-). The Rett syndrome gene (RTT) on Xq28 encodes MeCP2. It appears that 99.5% of the mutations are new ones (de novo), therefore the likelihood for a second girl having the disease in the same family is no more than 0.5%. The MeCP2 mutation affects brain development in such important ways that boys die either before or shortly after birth and never have the chance to develop actual Rett syndrome. The severity of the syndrome in girls is a function of the percentage of cells with a normal copy of MeCP2 that are left to function after random X inactivation. If X inactivation happens to turn off the X chromosome carrying the mutated gene in a large proportion of cells, the symptoms will be mild. If, instead, a larger percentage of cells has the healthy X chromosome turned off, the onset could be earlier and the symptoms more severe.

Cytoplasmic DNA (mtDNA) disorders: Discussed above (a review on Clinical Mitochondrial Genetics by Chinnery et al; Spectrum of Mitochondrial Disease by Naviaux; see also Taylor & Turnbull, 2005).

Nonclassical genetic phenomena

* Some autosomal dominant disorders may show sex-limitation. This happens when an affected male is infertile and the females do not express the disease. Then, the pedigree pattern will be identical to an X-linked recessive trait where males do not reproduce. Male-limited precocious puberty is such a disease. Females transmit it to half of their sons. The pedigree suggests a sex-linked inheritance.

* In some diseases, it has been documented that mutation rate is higher in males than in females: Duchenne muscular dystrophy (DMD), Lesch-Nyhan syndrome and hemophilia. In these diseases, increased paternal age effect will increase the risk of the disease in the offspring (see Chandley, 1991 for a review of parental origins of de novo mutations).

* When two affected parents have a child:

1. In autosomal recessive disorders, all children will be affected,

2. In autosomal dominant disorders, they may have a healthy child (if both are heterozygous, there is 25% chance of a healthy offspring).

* Uniparental disomy (UPD) is the presence of a pair of chromosomes originated from the same parent. The first step is usually the formation of a trisomy and then the loss of a chromosome (due to postzygotic nondisjunction) but gametic complementation (fusion of a gamete with two copies of the same chromosome with a gamete with none of the same chromosome) is also possible. The end result may be a homozygote child from a carrier parent. One particular UPD results in an interesting situation: Maternal UPD for chromosome 15 (15q11-q13) will result in Prader-Willi syndrome, while paternal UPD for the same chromosome will cause Angelman syndrome. UPD does not have to occur for the whole chromosome. Due to chromosomal rearrangements, UPD can occur for a portion of the chromosome. Beckwith-Wiedemann syndrome, for example, is sometimes due to partial UPD for chromosome 11p15 with paternal imprinting. Complete hydatidiform mole is a UPD, i.e., its genetic origin is completely paternal (even if the karyotype is 46,XX, which represents duplication of a haploid (23,X) sperm).

* Premature centromere separation or heterochromatin repulsion is another mechanism, which causes genetic disease: Roberts syndrome is an example.

* Genetic anticipation is seen mainly in autosomal dominant diseases (Huntington disease, congenital myotonic dystrophy, many cerebral ataxias) but also in Fragile X-syndrome, and in a single autosomal recessive disease (Friedreich ataxia). Curiously, genetic anticipation may also show sex-limitation. In Huntington disease, it only occurs when a male transmits the disease while the opposite is true for congenital myotonic dystrophy (maternal effect). Anticipation is due to trinucleotide repeat expansions, which may be in the coding region (Huntington disease) or in the untranslated region (Fragile X syndrome). Opposite of anticipation is also true and is reported in congenital myotonic dystrophy. This is called reverse mutation, negative expansion or contraction (see review by Brook (1993)). A sex-bias in meiotic instability has been reported in Friedreich ataxia: paternally transmitted alleles tend to decrease in a linear way that depends on the paternal expansion size, whereas maternal alleles can either increase or decrease (Munros, 1997).

* When a disease occurs first time in a family:

1. This is the first -de novo- mutation (in a disease for which the mutation has a very high rate as in Rett syndrome; also occurs in congenital adrenal hyperplasia, achondroplasia - seven-eights of mutations are new due to methylated CG dinucleotides in FGFR3). De novo microdeletion or point mutations in the transcriptional coactivator CREB-binding protein (CBP) on 16p13.3 cause Rubinstein-Taybi syndrome with a recurrence risk of 0.1% in sibs. The mutation rate in the NF1 gene is one of the highest known in humans, with approximately 50% of all NF1 patients presenting with novel mutations. About one-quarter of affected individuals with Marfan syndrome have new mutations; thus, a paternal age effect is present in sporadic cases (Chandley, 1991) as in all genetic syndromes with a high rate of new mutations (mutation rate is expressed as the number of new mutations per locus per generation; it is estimated as the incidence of new, sporadic cases of an autosomal dominant or X-linked disease that is fully penetrant such as achondroplasia. The new mutation rate ranges between 10^-4 to 10^-7 with a median 10^-6).

2. Both parents are healthy carriers of a recessive mutation (consanguinity or frequent mutations).

3. If the gene has been shown to be segregating in previous generations with no diseased individuals, the penetrance may have been low in earlier generations (acute intermittent porphyria is an example of low penetrance genetic disease). If it is an X-linked recessive disease and all the children so far are girls, the disease appears in the first male child.

4. It may be an adult-onset disease with genetic anticipation but the carriers of the gene in the previous generations did not live long enough to express the disease. Huntington disease is an example of late-onset genetic disease with anticipation. A similar situation may arise due to parent of origin effect. Differences in gene expression according to the parent from whom the gene was derived occur in Huntington disease and in myotonic dystrophy, and might be due to a difference in methylation of the genes in the two sexes (see Marx, 1988) or unknown modifier genes acting in a sex-limited fashion.

* Multifactorial inheritance: This pattern applies to complex diseases (like diabetes, hypertension, schizophrenia, cancer) where multiple genes and environmental factors play a role in the development of the disease. It has been best worked out in pyloric stenosis and orofacial cleft syndrome. These disorders are presumed to result from additive effects of multiple susceptibility genes with low penetrance. Individual mutations may not have any particular phenotype, but when act in concert and in the presence of the necessary environmental conditions, they may produce a disease phenotype. Under a model of multiple interacting loci, no single locus could account for more than a 5-fold increase in the risk of first-degree relatives. The disease shows increased incidence in families but with no recognizable inheritance pattern. The features of multifactorial inheritance are:

- The more severe the condition, the greater the risk to sibs,

- Carter effect: the sibs or offspring of a patient in less commonly affected sex have higher susceptibility to the disease,

- If it is a rare disease, the frequency of the disease among relatives is higher,

- If more than one individual in a family is affected, recurrence risk is higher,

- The risk falls rapidly as one passes from 1st to 2nd degree relatives,

There are two models proposed to explain multifactorial inheritance: multifactorial threshold model (a combined effect of multiple genes interacting with environmental factors; i.e. several or many genes, each of small effect, combine additively with the effects of non-inherited factors) and mixed model (a multifactorial liability with the involvement of a major gene). Multifactorial threshold model is based on a discontinued binary distribution for a quantitative trait meaning that the diseases are present or not in an individual but their inheritance is as if they were quantitative characters. This is due to a threshold effect that makes them appear as discontinued. For many diseases, there are at least two thresholds -differing by sex or causing different severity- (Reich, 1972). Examples include pyloric stenosis (sex dimorphism for liability) (Chakraborty, 1986) and orofacial cleft syndrome (two thresholds for fetal mortality and disease) (Dronamraju, 1982 & 1983). The latter model proposes a lower threshold level of liability resulting in a cleft formation and a higher level causing a fetal death (preferentially males). See also Introductory Statistical Genetics; See also Falconer's polygenic threshold model for dichotomous nonmendelian characters in Human Molecular Genetics and an example by Wanstrat & Wakeland, 2001. In multifactorial inheritance, the presence of even a major-disease causing allele is only predisposing rather than predictive. See also Introduction to Genetic Epidemiology.

From genotype to phenotype

Recessive traits usually result in defective enzymes / proteins (CAH, albinism, CF, SCID), dominants often alter structural, carrier or receptor proteins (familial hypercholesterolemia (LDL receptor), hemoglobin C, procollagen, fibrillin, elastin).

Albinism: In autosomal recessive oculocutaneous albinism, melanocytes cannot produce enough melanin. In heterozygotes there is enough of it.

Sickle cell anemia: In heterozygotes, there is enough HbA and no symptoms or signs of sickle cell disease.

Cystic fibrosis: In heterozygotes although the mutant CFTR (CF transmembrane conductance regulator) protein is produced, there is still enough wild type (not mutated) protein functioning properly. In homozygotes, the normal protein is totally missing.

In autosomal dominant diseases, heterozygosity and homozygosity may affect the disease expression. In Huntington disease, heterozygotes and homozygotes have the same clinical phenotype whereas in achondroplasia, homozygotes have a much more severe clinical course. Most X-linked dominant diseases cause in utero lethality in males (Rett syndrome, incontinentia pigmenti, Goltz syndrome / focal dermal hypoplasia, chondrodysplasia punctata type B, orofaciodigital syndrome 1) resulting in a higher female-to-male ratio in diseases individuals.

In diseases caused by haploinsufficiency and dominant-negative mutations, a mutation in one copy of the gene may cause the disease phenotype despite having one copy still functional (see above for examples). In haploinsufficiency, remaining amount of normal protein is insufficient for physiological function; in dominant-negative mutation, the normal copy is rendered nonfunctional by the mutant copy. These mutations will cause an autosomal dominant inheritance pattern. Osteogenesis imperfecta is due to dominant-negative mutations in type I collagen gene (COL1A1) mutations. Similarly, aquaporin-2 gene (AQP2) has a dominant-negative effect in causing type II nephrogenic diabetes insidipus. Diseases caused by haploinsufficiency are characterized by the lack of a correlation between the severity of the genetic defect (for example, size of a deletion) and the phenotype. One example is Digeorge syndrome. Histone H2AFX is a genomic caretaker that requires the function of both gene alleles for optimal protection against tumorigenesis. Therefore disruption of one allele causes genomic instability and tumor susceptibility due to haploinsufficiency (Celeste, 2003). Other examples of haploinsufficiency include PAX6 gene on 11p causing aniridia type II / WAGR syndrome; PTEN gene on 10q23 Cowden Disease and several cancers; SHOX (short stature homeobox containinggene) causing short stature as a dominant phenotype (Ogata, 2001); GLI3 gene on 7p13 causing Greig cephalopolysyndactyly syndrome (GCPS), elastin (ELN) gene on 7q11.2 causing Williams-Beuren syndrome, and transferring receptor gene TFRC. The original Knudson's two-hit hypothesis (Knudson, 1975) suggesting loss of heterozygosity or homozygous deletion are the two hits required for the loss of tumor suppressor gene activity has now been extended to include haploinsufficiency (see Balmain, 2003 and Paige, 2003) as well as transcriptional silencing by DNA methylation of promoters that can disable tumor-suppressor genes (Jones & Laird, 2000).

In many genetic diseases, the disease has a variable clinical spectrum from mild to severe and even lethal (variable expressivity). This phenotypic heterogeneity is usually due to involvement of different alleles (allelic heterogeneity) or genes (locus heterogeneity) in disease pathogenesis. Gaucher disease, heritable collagen disorders, muscle diseases, dominantly inherited ataxias, neurofibromatosis type 1 and type 2, congenital adrenal hyperplasia, and alpha-1-antitrypsin deficiency are examples of diseases showing phenotypic heterogeneity. Examples of locus heterogeneity include hypertrophic cardiomyopathy (CMH, IHSS) which may be caused by mutations in cardiac troponin I (TNNI3), cardiac troponin T2 (TNNT2), alpha-tropomyosin (TPM1), cardiac myosin binding protein C (MYBPC3), cardiac myosin heavy chain-alpha (MYH6) etc; congenital adrenal hyperplasia may be caused by mutations in CYP21A2 or CYP11B1. Although most complex diseases (such as essential hypertension, noninsulin-dependent diabetes mellitus, schizophrenia, prostate cancer, familial glioma and inflammatory bowel diseases) appear to be examples of locus heterogeneity because many loci are involved, this is not the case because ‘combinations’ of susceptibility alleles of various loci contribute to disease pathogenesis rather than different alleles of various loci causing the same disease.

Just like Beadle’s one gene-one enzyme (protein) idea is no longer strictly true, one gene can also cause different diseases through different imprinting patterns depending on the parent of origin, or different mutations. Confusingly, this is also called allelic heterogeneity. Different mutations in cardiac sodium channel gene SCN5A cause either long QT syndrome-3 (LQT3) or Brugada syndrome (both may cause sudden death in young healthy individuals). Different mutations in WAS gene cause either Wiskott-Aldrich syndrome or X-linked congenital thrombocytopenia (Zhu, 1995). Jervell and Lange-Nielsen syndrome (JLNS) is due to homozygosity for a mutation in the KVLQT1 gene, heterozygosity for a mutation in the same gene causes long QT syndrome-1 (LQT1). Dystrophin gene mutations cause Duchenne muscular dystrophy, Becker muscular dystrophy or X-linked dilated cardiomyopathy depending on the mutation. Myosin VIIA gene (MYO7A) mutations result in autosomal recessive sensorineural deafness type 2 (DFNB2), autosomal dominant nonsyndromic sensorineural deafness type 11 (DFNA11) and Usher type 1B syndrome (USH1B). It is interesting that different mutations in the DFNB2 gene may result in either dominant or recessive hearing loss (Tamagawa, 1996). Other examples: fibroblast growth factor receptor 2 (FGFR2) mutations cause Crouzon, Pfeiffer or Apert syndrome; mutations in L1 cell adhesion molecule gene (L1CAM / CD171) cause X-linked hydrocephalus or MASA syndrome; peripheral myelin protein 22 (PMP22) mutations cause Charcot-Marie-Tooth disease type 1A or hereditary neuropathy with liability to pressure palsies; cystic fibrosis gene (CFTR) mutations do not only cause cystic fibrosis but also congenital bilateral aplasia of the vas deferens; heterozygosity for CFTR mutations is associated with 'idiopathic' chronic pancreatitis.

Lack of perfect genotype-phenotype correlation may be due to the following:

* Variable expressivity

* Incomplete penetrance

* Environmental interaction

* Gene interaction (epistasis)

* Allelic or locus heterogeneity

* Mitochondrial inheritance

* Epigenetic influences

* Sex-influence

Population Genetics

However rare a genetic disease may be, the carrier state for the disease gene is likely to be very common. Phenylalanine hydroxylase deficiency (classic PKU) has a frequency of 1 in 10,000 but a carrier frequency of 1 in 50. Similarly, the mutations of CYP21A2 (congenital adrenal hyperplasia) may be present in 1 in 3 Ashkenazi Jews (Speiser, 1985), CFTR mutations in 1 in 20, and the C282Y mutation of HFE in 1 in 10 Northern Europeans. This is one reason why eugenics approaches would never work in elimination of a disease. For each affected individual (for a recessive disease), there may be up to 100 carriers showing no symptoms of the disease.

Because of their relatively homogenous origins (founder effects), genetic drift and isolation, all reducing genetic heterogeneity (the 'noise' in association studies); large families providing very extensive pedigrees; and high coefficient of kinship, populations like Finland, Iceland, Sardinia and Newfoundland are useful for the study of complex diseases. They exhibit an increased prevalence of rare recessive diseases (congenital nephrotic syndrome of the Finnish type and Newfoundland rod-cone dystrophy). Linkage disequilibrium mapping of common diseases is also a more feasible approach in isolated / homogeneous populations compared with admixed ones. Disease gene mapping by linkage disequilibrium analyses can be achieved at the population rather than the pedigree level in such populations.

Heterozygosity for an allele that causes an autosomal recessive disease in homozygotes is usually asymptomatic but detectable by laboratory tests. Heterozygosity for mutations causing congenital adrenal hyperplasia, hereditary hemochromatosis, alpha-1-antitrypsin deficiency, cystic fibrosis, ataxia telangiectasia, homocystinuria, and hereditary fructose intolerance results in biochemically detectable changes.

To explain the propagation of disease genes, compensating heterozygote advantage has been suggested for a few of them: Phenylketonuria (PKU) mutation (Woolf, 1975; Woolf, 1986; Woo, 1989); cystic fibrosis mutation (CFTR) (Pier, 1998); 21-hydroxylase mutations (CYP21A2) (Witchel, 1997); HFE mutations (Datz, 1998); alpha-thalassemia mutation (HBA1) (Flint, 1986); beta-E-thalassemia (hemoglobin E) (Chotivanich, 2002); sickle cell disease mutations (HbF) (Friedman & Trager, 1981); and glucose-6-phosphate dehydrogenase mutations (Luzzatto, 1969). The concept of compensating heterozygote advantage (or balancing selection) in this context was first proposed by JBS Haldane in 1949. An increased frequency of heterozygosity for alpha-1-antitrypsin deficiency in twins and parents of twins has been noted, which led to the suggestion that 'increased' fertility and twinning may be the mechanism for heterozygous advantage for antitrypsin deficiency (Liberman, 1979). It is believed that the mutant may increase ovulation rate and enhance the success of multiple pregnancies in heterozygotes (Clark & Martin, 1982; Boomsma, 1992). Heterozygote advantage via lower miscarriage rate has been suggested in phenylketonuria (Woolf, 1975; Woolf, 1986). See glossary for heterozygote advantage, and reviews on maintenance of disease-causing mutations: Rotter & Diamond, 1987; Carter & Nguyen, 2011.