ArticlePDF AvailableLiterature Review

Abstract

Recombinant DNA technology has made possible the localization and isolation of disease-related genes, the tracking of disease-related alleles through family pedigrees, the direct detection of the pathological lesion itself and the in vitro expression of both normal and mutant genetic information at the mRNA and protein levels. Undoubtedly the most immediate practical spin-off from recombinant DNA technology in medical genetics has been in the sphere of improved disease diagnosis and prediction, where advances have been dramatic. We review the nature of inherited disease, current approaches to its analysis, diagnosis and prediction, mechanisms of gene mutation and the available techniques for mutation detection. Also examined are the various genetic factors that can alter the relationship between genotype and clinical phenotype. Finally, the genetics of selected adult-onset disorders are explored in the context of considering the accuracy and reliability of disease prediction.
Q J Med 1997; 90:83–103
Review
QJM
Adult-onset genetic disease: mechanisms, analysis and
prediction
D. RAVINE and D.N. COOPER
From the Institute of Medical Genetics, University of Wales College of Medicine, Cardiff, UK
Summary
Recombinant DNA technology has made possible advances have been dramatic. We review the nature
of inherited disease, current approaches to its anal-the localization and isolation of disease-related
genes, the tracking of disease-related alleles through ysis, diagnosis and prediction, mechanisms of gene
mutation and the available techniques for mutationfamily pedigrees, the direct detection of the patholo-
gical lesion itself and the in vitro expression of both detection. Also examined are the various genetic
factors that can alter the relationship between geno-normal and mutant genetic information at the mRNA
and protein levels. Undoubtedly the most immediate type and clinical phenotype. Finally, the genetics of
selected adult-onset disorders are explored in thepractical spin-off from recombinant DNA techno-
logy in medical genetics has been in the sphere of context of considering the accuracy and reliability
of disease prediction.improved disease diagnosis and prediction, where
Introduction
Although molecular genetic analysis provides pre- and Alzheimer’s disease. However, immediate
therapy is not the only goal of predictive testing.dictive medicine with a potentially very powerful
and versatile tool, predictive evaluation is not There are many reasons why at-risk individuals opt
for testing for a disease for which no cure currentlydependent solely upon modern molecular tech-
niques. High blood pressure, obesity, serum choles- exists. Prominent among these include being able
to readjust lives to the unavoidable prospect of aterol levels, family history, behavioural traits and
many other features have long been recognized as shortened period of health and/or to be able to
plan social and reproductive lives. The hope ofrisk factors and hence predictors of future disease in
otherwise apparently healthy individuals. However, being found not to have the disease in question
features prominently for many, but counselleesthe application of recombinant DNA methodology
has greatly increased the power and scope of predict- often find themselves trapped in a complex mesh
of psychological and social pressures to undergoive (and preventive) medicine.
In the absence of possibilities for therapeutic predictive testing. Well-informed pre-test counselling
is an essential component of the testing process,intervention, such advances in predictive tests can
be expected to generate all kinds of ethical but it must be recognized that for many adult
genetic conditions, particularly those which areproblems in their wake. Indeed, for an increasing
number of adult-onset disorders, recently developed currently incurable, some people consciously
choose not to be tested so as to preserve uncertainlypredictive tests confer little or no direct therapeutic
benefit upon the patient, e.g. Huntington’s disease and so the possibility of hope.
Address correspondence to Professor D.N. Cooper, Institute of Medical Genetics, University of Wales College of Medicine,
Heath Park, Cardiff CF4 4XN
© Oxford University Press 1997
D. Ravine and D.N. Cooper84
and males, although the absence of a normal allele
Inheritance: models and mechanisms
in the hemizygous state often results in males being
more severely affected than females.
Patterns of inheritance
To understand the spectrum of human genetic dis-
Variable penetrance and expressivity
ease, it is first necessary to consider the way in
which genes may be inherited. Most inherited dis-
For some diseases, all individuals who carry the
orders follow a simple Mendelian form of transmis-
defective gene manifest a clinical phenotype. In such
sion through families, that is the inheritance patterns
cases, penetrance of the disorder is said to be
obey Mendel’s two laws of inheritance. First, genes
complete. However, the inheritance of certain gene
are units that segregate; members of the same pair
mutations does not invariably give rise to a disease
of genes, alleles, are never present in the same
phenotype. Diseases in which the inherited disease
gamete. Second, genes assort independently; mem-
allele is not invariably expressed as a clinical pheno-
bers of different pairs of genes move to gametes
type are said to exhibit partial or incomplete penetr-
independently of each other. Almost all human genes
ance. This represents a potentially important source
are located within 23 pairs of chromosomes, 22 of
of error in the ascertainment of whether or not an
which are called autosomes. The other pair are the
individual is actually a carrier of the defect in
sex chromosomes; females have two X chromosomes
question. An allied concept is that of variable
whereas males have an X and a Y chromosome.
expressivity which is used to describe the variation
Inheritance patterns are determined by whether a
in pathological symptoms in those individuals in
gene is located on an autosome or a sex chromosome
whom the presence of the gene is evident from its
and whether or not a mutant allele is able to affect
clinical effects. Variable penetrance and expressivity
the phenotype in the presence of a normal or
result from a subtle interplay of both genetic and
wild-type allele. In addition, a few diseases follow a
environmental influences which can together be
cytoplasmic mode of inheritance that is independent
termed the genotype-phenotype relationship (discus-
of the inheritance of chromosomes. Mitochondria,
sed below).
almost all of which are inherited maternally in the
cytoplasm of the ovum, have their own genome and
Genomic imprinting and its relevance to
a limited number of human genetic diseases result
from inherited abnormalities of mitochondrial DNA.
disease
Monk2 loosely defined genomic imprinting as the
‘differential modification of the maternal and paternal
Dominance and recessiveness
contributions to the zygote, resulting in the differen-
tial expression of parental alleles during developmentA disorder is considered dominant if clinical symp-
toms appear in the heterozygote in whom only one and in the adult’. This differential modification
appears to be essential for normal development,allele is defective. In such cases, the disease allele
is said to be dominant over the wild-type allele. since parthenogenetic embryos (whether diploid
paternal or diploid maternal) do not survive to term:Dominant mutations often result in a clinical pheno-
type by giving rise to reduced or abnormal expression in diploid maternal embryos, fetal development is
normal but development of the extra-embryonicof a gene product. This gene product is often a
receptor or a structural, carrier or regulatory protein membranes is abnormal. In diploid paternal embryos,
it is the other way around (reviewed in Monk).2rather than an enzyme (reviewed in Wilkie1). For the
clinical phenotype to become apparent in a recessive Clearly, maternal and paternal chromosomes must
differ epigenetically, and in such a way that differentdisorder, both alleles must be mutant (homozygous
state). A disease allele may be said to be recessive developmental programmes are followed.
In principle, imprinting must be established eitherto the wild-type allele in cases where production of
50% of the normal level of the gene product in the before, or early in, gametogenesis, must be stable
enough to be retained in somatic cells, and must beheterozygote is sufficient to avoid clinical symptoms.
This gene product is often an enzyme but may also capable of being erased in the germline at every
generation prior to the fresh establishment of thebe a regulatory molecule (e.g. tumour suppressor
protein; see below). In the case of an X-linked imprinting pattern. The idea that imprinting might
involve DNA methylation was proposed by Monkrecessive disease, affected males who carry the
disease lesion are said to be hemizygous since they et al.3 on the grounds that this post-synthetic modi-
fication was heritable, reversible and known to playdo not carry a wild-type homologous X chromosome
which serves to mask clinical expression of the a role in the control of gene expression. This view
soon received solid support from the results of studiesdisorder in females. By contrast, in X-linked dominant
disorders, the mutation is evident in both females on transgenic mice.4,5 The regulation of the process
Adult-onset genetic disease 85
of imprinting both spatially and temporally is, how- (e.g. skewed X inactivation found in manifesting
carriers of Duchenne muscular dystrophy). By con-ever, not yet fully understood (reviewed in Razin
and Cedar6; Efstratiadis7; Barlow8). trast, X inactivation may be protective, as in cases
where multiple X chromosomes are present (e.g. XXY,Genomic imprinting appears to contribute to the
occurrence of insulin-dependent diabetes mellitus XXX) and all but one are inactivated. DNA methyl-
ation may be involved in this process, and so X(IDDM). In HLA-DR4-positive diabetics, there is
evidence of the existence of a gene or genes affecting inactivation may be regarded as a special case of
genomic imprinting.HLA-DR4 IDDM susceptibility positioned in the
region of 11p15 where the genes for insulin (INS)
and insulin-like growth factor II (IGF2) are located.
IDDM-associated alleles for polymorphisms in this
Disease analysis
region have been shown to be transmitted preferen-
Analysis, diagnosis and prediction of human
tially to HLA-DR4-positive diabetic offspring. This
genetic disease
effect is strongest in paternal meioses, suggesting a
possible role for maternal imprinting.9 Imprinting is
Introduction
also evident in some tumours, including Wilms’
The diagnostic setting in medical genetics is basically
tumour and hereditary non-chromaffin paragangli-
of two distinct kinds: (i) confirmation/exclusion of a
omas. The predisposition to paragangliomas, also
particular disease in symptomatic patients and (ii)
referred to as glomus tumours or chemodectomas,
prediction in asymptomatic people. In the first, a
is inherited as an autosomal dominant trait.
disease may already be suspected in an individual
Interestingly, all familial patients appear to have
through clinical observation, and a DNA test may
inherited the trait from their father, whereas offspring
then help to confirm or exclude the diagnosis of the
from obligate female carriers do not develop tumours.
disease in question. Ultimate confirmation will how-
Linkage studies support the existence of genetic
ever only be obtained once the pathological lesion
heterogeneity in the predisposition to these tumours,
in the affected gene has been detected in the patient.
with two genes being located in different regions of
If this is not possible, indirect methods of genotype
chromosome 11q.10,11 On the basis of these observa-
analysis (see below) can provide a very powerful
tions, an interesting model for tumour initiation and
diagnostic tool either on their own or in combination
progression has been proposed by Mariman et al.11
with independent phenotypic investigations. The
As patients often develop more than one tumour
greater the clinical heterogeneity manifested by a
affecting different glomus bodies, these authors sug-
disease, the greater will be the differential diagnostic
gested that one or both of the predisposing genes on
uncertainty, and the more often will objective
11q encode tumour-suppressor proteins. Assuming
confirmation be required. Of enormous importance
that both genes code for functionally equivalent
in genetic counselling is the ability to distinguish
tumour suppressors, the Knudson hypothesis would
between heritable and non-heritable types
predict that all four alleles from the two genes have
( phenocopies) of a distinct clinical entity. In this
to be turned off before glomus tumours can develop.
review, we concentrate upon disease prediction in
In this respect, genomic imprinting would account
hitherto asymptomatic individuals.
for the silencing of the two maternal copies, whereas
a paternally inherited mutation in one of the genes
Locus heterogeneity
would represent the third hit. A single somatic
mutation in the remaining active paternal copy Gene mapping has already shown its importance to
clinical medicine through the locus (non-allelic)would subsequently suffice to initiate tumour growth.
heterogeneity it has revealed.12 Locus heterogeneity
is said to occur in cases where a similar disease
X chromosome inactivation
phenotype arises as a consequence of defects in
different genes. Locus heterogeneity has extremelySince females have two X chromosomes whilst males
have only one, some means of equalizing the expres- important implications for disease diagnosis, predic-
tion and treatment, since the nature of the clinicalsion of X-linked genes between the sexes must be
available. This is achieved by inactivating one X phenotype or the age of onset and severity may
differ between non-allelic forms of the disorder(s).chromosome at random in every female cell. This
inactivation is controlled by an X-linked gene which Thus mutations in at least seven different loci have
been implicated in the causation of Charcot-Marie-is only active on otherwise inactive X chromosomes.
Occasionally, in the presence of an abnormality on Tooth disease (reviewed in Harding13). RFLP tracking
(see below) has permitted the confirmation ofone X chromosome, skewed X inactivation may
occur, with preferential inactivation of the normal X locus heterogeneity in a variety of other conditions.
Over 20 loci where mutations cause retinitischromosome resulting in manifestation of disease
D. Ravine and D.N. Cooper86
pigmentosa have been mapped. There are at least
Multifactorial inheritance
eight autosomal loci and three X-chromosomal loci
A large number of people living in societies with a
responsible for causing non-syndromic retinitis pig-
high individual life expectancy will sooner or later
mentosa in different kindreds (reviewed in Wolf14).
suffer from an adult-onset disease in which genes
Some of the genes involved in these disorders have
related to the disease have probably played a major
now been isolated and characterized. Some have
causative role. Since these diseases are not generally
been chromosomally localized, whilst the existence
associated with a detectable chromosome aberration
of the remainder is inferred from the absence of
or a clear Mendelian pattern of inheritance, many
co-segregation with known loci in certain families.
of them have been called polygenic or multifactorial,
Linkage studies can of course also provide evidence
even though it is not usually known how many
for locus homogeneity. Thus the clinically variable
genetic loci are involved or how they interact with
conditions Huntington’s disease (see below) and
environmental factors.
Friedreich’s ataxia have shown no sign of locus
Many diseases such as diabetes and hypertension
heterogeneity and probably result from the malfunc-
are ill-defined with respect to their mode of inherit-
tion of a single gene in each case.
ance, and a considerable proportion may prove to
have a multifactorial basis. Undoubtedly, many
Allelic heterogeneity
Mendelian traits remain to be recognized, having
Another cause of phenotypic and clinical variation
been obscured so far by their rarity or by other
in a given disorder is allelic heterogeneity. This term factors such as late onset or variable penetrance. For
is used to describe the ability of different mutations others, the terms predisposition, susceptibility or
penetrance serve perhaps only to cloak our ignorancein the same gene to give rise to very different clinical
in acceptable garb rather than to illuminate our
phenotypes. Thus mutations in the COL1A1 gene
thinking. This notwithstanding, such diseases may
are responsible for several different conditions, viz.
still prove amenable to the same analytical tech-
osteogenesis imperfecta types I-IV, Ehlers-Danlos
niques that have been applied to the study and
syndrome type VII, and osteoporosis. Different muta-
diagnosis of single gene defects.17–20
tions in the COL2A1 gene are known to cause at
least seven distinct disorders: achondrogenesis, hypo-
chondrogenesis, spondylo-epiphyseal dysplasia con-
The genotype-phenotype relationship
genita, spondylo-epimetaphyseal dysplasia, Kneist
dysplasia, Stickler syndrome and osteoarthritis with
For many ‘monogenic disorders’, it is probable that,
mild skeletal dysplasia. Similarly, mutations of the
even if one could somehow set aside the influence
RET proto-oncogene have been identified in indi-
of environmental variables for a moment, precise
viduals with Hirschsprung’s disease, multiple endo-
knowledge of the pathological lesion alone would
crine neoplasia types 2A and 2B and both sporadic
still provide insufficient information to be able to
and familial medullary thyroid carcinoma.
predict the clinical phenotype exactly. Even a disease
An example of the effect of linked polymorphisms
with no allelic heterogeneity, sickle-cell anaemia, in
is the case of fatal familial insomnia (FFI) and a
which all patients are homozygous for the same b-
subtype of Creutzfeldt-Jakob disease (CJD). These
globin (HBB) gene lesion, is no exception to this
conditions are both associated with possession of
rule. The clinical severity of sickle-cell disease is
the same mutation (Asp178Asn) in the PRPN gene
highly variable, and may be ameliorated by the
but are distinguished by the presence of different
effects of genetic variation either within the HBB
alleles of a common polymorphism on the same
gene regulatory region, in far upstream regions
chromosome: Met129 specifies FFI whilst Val129
controlling the expression of the linked G-c (HBG2)
specifies CJD.15 The effect of the polymorphic alleles
and A-c (HBG1) globin genes or at the unlinked a-
when they occur on the other chromosome is to
globin (HBA) locus (reviewed in Cooper and
alter the duration and age of onset of both FFI
Krawczak21). Thus, since no gene can exist in splen-
and CJD.
did isolation, no gene (or gene product) can be
For some disorders, a large number of allelic
completely insulated from other genetic influences,
variants are known. Haemophilia B with at least 500
either from its immediate sequence environment or
different lesions documented in the factor IX (F9)
from other gene loci. In a sense therefore, all
gene is one of the most dramatic examples.16 In
inherited disease states, including so-called ‘single
haemophilia B, the bleeding severity manifested by
gene defects’, must be regarded in some way as
patients is usually predictable from the nature of the
multifactorial. This truism should not, however,
underlying mutational lesion although it may still obscure the fact that those genetic factors which
occasionally vary between individuals bearing the influence the genotype-phenotype relationship in a
given disorder can often be studied individually andsame mutation.
Adult-onset genetic disease 87
may even be quantified, thereby improving the gation. In principle, any two genetic loci on the
accuracy of prognostic predictions. same chromosome should be transmitted together
A number of different factors (of both genetic and
from one generation to the next. This implies that
environmental origin) can account for the fact that
polymorphic loci in the vicinity of a mutated gene
two individuals with identical genetic lesions at a
can serve as markers for the presence of a given
given locus can experience quite different clinical
gene lesion. The pathway of the mutation through
symptoms. These factors determine the relationship
family pedigrees may thus be tracked by studying
between genotype and phenotype.14 Among the
the segregation of marker alleles instead of disease-
genetic modifying factors are the proportion of cells
related alleles. However, in practice, homologous
bearing the lesion (mosaicism), the activity state of
chromosomes recombine and exchange pieces
the chromosome carrying it (X-inactivation and
of DNA during meiosis in a process termed crossing-
imprinting—see above), the effects of other loci and
over. Thus with respect to any pair of loci on
their protein products (epistasis), the status of the
homologous chromosomes, germ cells may be
other allele and the variable expression of the wild-
divided into two groups: recombinants and non-
type or mutant allele as a consequence either of
recombinants.
alternative mRNA processing or the influence of
The observed proportion of recombinants with
genetic variation in the gene promoter region.
respect to any two loci is termed their recombination
Genetic imprinting and X-inactivation are epi-
fraction,(h). h equals exactly half the probability of
genetic mechanisms for phenotype modification (see
at least one cross-over event taking place and is
above). They are temporal and/or tissue-specific
therefore a function of the physical distance between
phenomena and are (at least in principle) reversible,
the two loci. The closer the loci are, the smaller the
although the sequelae of an imprinted gene may
recombination fraction, with a value of zero attained
persist, particularly if it has shaped early morphogen-
when both loci coincide. The upper limit of h is 0.5,
esis or organogenesis.
corresponding theoretically to the recombination
Somatic mosaicism is an important cause of
fraction between loci located on non-homologous
phenotype modification resulting in variation in the
chromosomes. Clearly, the study of recombination
clinical expression of an inherited trait or disorder
fractions represents a useful means to analyse and
(reviewed in Hall22). The proportion of affected cells
localize new disease loci.
in the expressing tissue(s) of the individual concerned
Major goals of linkage analyses are (i) to prove
is a prime determinant of phenotypic severity.
that the marker and mutated gene are linked and (ii)
Mosaicism for a specific gene mutation may be
to estimate h. Evidence in favour of, or against, a
confined to somatic cells or to the germline, or be
given recombination fraction is expressed by the
present in both, depending upon the developmental
quantity z(h), termed the lod score.23 z(h)isa
stage at which the lesion occurred. If the germline
logarithmic value, and a positive value of argues in
of that individual is affected, then subsequent pro-
favour of linkage. If h is considerably smaller than
geny will be at risk of inheriting a much more
0.5, then a lod score exceeding 3 (chance of greater
severe phenotype. Germline mosaicism provides an
than 1000:1) is usually regarded as sufficient evid-
explanation for the inheritance pattern in cases where
ence for linkage. If z(h) is less than –2 (chance of
multiple affected offspring are born to clinically and
less than 1:100), then recombination fractions smaller
phenotypically normal parents. It arises through the
than h are excluded from being the true value. h is
occurrence of a de novo mutation in a germline cell
estimated by that recombination fraction which yields
or one of its precursors during the early embryonic
the maximum lod score, i.e. the maximum likelihood.
development of the parent. Although the frequency
The greater the number of informative families
with which germline mosaicism occurs may well
studied, the more reliable and convincing are the
differ quite markedly between different disease states,
results generated from linkage studies. The overall
it does provide an important explanation for the
lod score is calculated as the sum of the lod scores
recurrence of rare mutations within a single family,
obtained in separate families.
and should always be considered when counselling
Once the linkage between a marker and a disease
such cases.
gene has been established, and when the recombina-
tion fraction, h, between them is known with suffi-
cient accuracy, then the marker can be employed in
Indirect disease diagnosis and
the diagnostic testing. To determine whether an
prediction
individual carries one or more mutant alleles, and if
phenotypic information regarding carriership is lack-
Linkage analysis
ing, chromosomes carrying the mutation(s) must be
identified in other family members, preferably includ-
Genes are organized in chromosomes and hence do
not always obey Mendel’s law of independent segre- ing the parents. If the marker allele characterizing a
D. Ravine and D.N. Cooper88
mutant chromosome is also present in the proband, protein [GNAS1 gene] in McCune-Albright syn-
drome). Different point mutations in the adult skeletalthen this may indicate that they have also inherited
the mutation. However, if marker and disease gene muscle sodium channel a subunit (SCN4A) gene
cause hyperkalaemic periodic paralysis anddo not coincide (i.e. if h>0), then the identification
of mutants by marker alleles generates probabilistic paramyotonia congenita, by interfering with normal
voltage-sensitive inactivation of the sodium current.results. The efficiency and accuracy of risk assess-
ment can be improved in several ways, e.g. by the Dominant negative mutations, in which product
from the mutant allele inactivates the wild-typeuse of flanking markers and the creation of haplo-
types, phenotypic markers (e.g. serum creatine kinase product, usually occur in genes that are involved
with protein subunit activity or assembly (e.g. somelevels in Duchenne muscular dystrophy), and linkage
disequilibrium data. somatic TP53 gene lesions responsible for malig-
nancy). Toxic precursors or waste products mayWhile linkage analysis remains important for some
diseases, increasingly, direct testing for disease- result from dominant mutations in a number of
proteins including b-amyloid precursor protein (APP)causing mutations is taking its place. The major
consequences of this change are a greatly improved and prion protein (PRNP). Mutations causing altered
structural proteins may result in a more severespeed and reliability of diagnosis, and elimination of
the need to involve other family members in predict- phenotype than loss-of-function mutations (e.g. muta-
tions of fibrillin (FBN1) in Marfan syndrome). Finally,ive gene testing. Nevertheless, linkage analysis
remains useful for some diseases, even when the some dominant mutations may result in new protein
functions that are deleterious (e.g. the c-ABL/BCRdisease gene is known. Exclusion testing using
linkage analysis has sometimes used in conditions fusion products in the 9;22 Philadelphia translo-
cation).such as Huntington’s disease, enabling fetal risk to
be refined (e.g. to determine whether a 25% risk Characterized gene mutations have been found
within coding sequences, promoter regions, in splicecan be adjusted towards 50% or 0%) without altering
the parental 50% risk. Linkage analysis is still com- junctions, within introns and in polyadenylation sites
(see below and Figure 1). Indeed, they may interferemonly used for families in cases where the disease
gene has been isolated but mutation screening would with any stage in the pathway of expression from
gene to protein product. Table 1 presents a basicbe very labour-intensive, or if the pathological lesions
have proven elusive. classificatory system of mutation types by reference
to the nature and position of the gene lesion and
the stage in the expression pathway with which it
interferes. Mutations are firstly divided on the basis
Direct disease diagnosis and
of whether they result in the reduced synthesis of a
prediction
gene product or the synthesis of a gene product
which is structurally or functionally abnormal.
Genes, mutations and disease
Mutations are then secondarily divided into four
categories; promoter function, gene structure, RNAThere are a variety of mechanisms by which muta-
tions can give rise to an inherited disease. Mutations processing and translation. Some gene lesions may
be placed into more than one category. For example,may lead to a loss or gain of gene function (reviewed
in Wilkie1). In dominant disorders, haploinsuffici- missense mutations with drastic effects on protein
structure and stability, or which served to activateency, or reduced gene function may result from
inactivation of one of a pair of alleles. This may an exonic cryptic splice site, can fall into both
categories. Similarly, a missense mutation close tocause disease through the reduced synthesis of tissue-
specific proteins synthesized in large quantities (e.g. an intron/exon splice junction could affect mRNA
splicing efficiency as well as protein structure. Theglobins), by interference with a rate-limiting step of
a metabolic pathway (e.g. low-density lipoprotein effects of specific amino acid substitutions on protein
structure are the subject of several reviews.24–26receptor), or by reduced activity of a regulatory gene
working close to a threshold value (e.g. RDS/per-
ipherin in retinitis pigmentosa). In recessive disorders,
Methods of detecting mutations
loss of function is an invariable outcome of the
mutant homozygous state. Predictive testing for adult-onset genetic disease will
often involve ascertaining whether or not a particularGain-of-function mutations responsible for domin-
antly inherited disorders may result from a variety of pathological gene lesion is present in a given relative
of a patient already known to suffer from the diseasemolecular mechanisms, including increased gene
dosage (e.g. the PMP22 gene duplication in Charcot- in question. As we have seen above, even although
the disease-related gene may be unknown, if aMarie-Tooth disease 1A), increased or constitutive
protein activity (e.g. activating mutations in the G
sa
linked marker is available, linkage analyses can be
Adult-onset genetic disease 89
Figure 1. Schematic drawing of an archetypal human gene containing three exons. The horizontal arrow denotes the
direction of mRNA transcription in a 5’3’ direction from the transcriptional initiation site (cap). The poly(A) addition site
is located immediately downstream of the AATAAA motif.
Table 1 A classification of types of mutation found to cause human single-gene defects, through either reduced synthesis
of a normal protein or normal synthesis of an abnormal protein
Phenomenon Cause
Reduced synthesis of a normal gene product
Defect in promoter function Binding of positive regulatory protein reduced or abolished
Binding of negative regulatory protein increased
Gene structure Deletions (frameshift)
Insertions, duplications, inversions (frameshift).
RNA processing/stability Mutations in transcriptional initiation site causing failure
to initiate transcription
Splice junction mutations resulting in exon skipping
and/or cryptic splice site utilization
Activation of cryptic splice sites
Polyadenylation/cleavage signal mutations
Mutations in 3’ untranslated region
Translation Initiation and termination codon mutations
Mutations in 5’ untranslated region
Nonsense mutations
Synthesis of structurally/functionally abnormal gene product
Shortened gene product Deletions (in-frame), nonsense mutations
Fusion genes Deletions involving two linked genes
Elongated gene product Insertions, duplications (in-frame)
Termination codon mutations
Defective post-translational modification/processing
Instability of protein product
Impaired assembly/secretion
Altered substrate/cofactor/receptor affinity Missense mutations
attempted to track the disease gene through the routinely employed in mutation detection and
screening are outlined below.family pedigree. Such studies are inherently proba-
bilistic in nature, and their reliability is critically
dependent on the mode of transmission and penetr-
Southern and Northern blotting
ance of the disease under study, the informativity of
the marker, the distance between marker and disease With the discovery of restriction enzymes, site-
specific endonucleases that cleave large DNA frag-loci, the pedigree structure and availability of family
members. In the ideal situation, linkage studies will ments at defined sites, and the development of
techniques to mass-produce these fragments (DNAbe unnecessary; both the identity of the defective
gene and the nature of the gene mutation will already cloning), the way was open to analyse the fine
structure of human genes and to begin to characterizehave been established, and the task simply reduces
to testing family members directly for possession of the genetic lesions underlying inherited disease. The
first technique to make this possible was Southernthat lesion. In the majority of cases, however, whilst
the identity of the defective gene may be known, blotting.27 Total genomic DNA is cleaved with a
restriction enzyme and transferred from the agarosethe nature of the pathological lesion within it is not.
One of a number of mutation screening methods gel upon which it has been size-fractionated to a
porous membrane which serves as a solid support.can then be used to detect the lesion to ensure that
subsequent predictive tests will be fast, accurate and It may then be challenged with a gene radiolabelled
probe to reveal the structure of the homologousreliable. Some of the approaches and strategies
D. Ravine and D.N. Cooper90
region of DNA in the genome of both patients and heteroduplexes as a result of mutation may also be
detected by denaturing gradient gel electrophoresiscontrols. Northern blotting, based upon the same
principle but adapted for use on RNA,28 is also an (DGGE) (reviewed in Myers et al. 39 and Fodde &
Losekoot40). The basic principle is simply that DNAimportant means of studying the phenotypic sequelae
(whether qualitative or quantitative) of a given duplexes migrate through a gradient of denaturant
until they reach a position where the strands meltdisease lesion at the RNA level.
and no further migration can occur. The melting
behaviour and rate of electrophoretic migration differ
Polymerase chain reaction
for the mutant and wild-type DNA heteroduplexes.
In 1985, Saiki et al.29 described an elegant new
Similarly, single-strand conformation polymorphism
method which, by circumventing the need to clone
(SSCP) analysis relies upon the electrophoretic mobil-
specific DNA fragments, has revolutionized the prac-
ity of a single-stranded DNA molecule being a
tice of molecular diagnostics. Now termed the poly-
function not only of its size but also of its nucleotide
merase chain reaction (PCR), this technique involves
sequence.41–43
the primer-mediated enzymic amplification in vitro
The above mutation-screening techniques provide
of specific target sequences in genomic DNA by
little or no positional information on the mutation
repeated cycles of (i) heat denaturation of the
detected. Confirmatory DNA sequencing is therefore
double-stranded template, (ii) primer annealing and
always required,33,36,44 but there are many alternat-
(iii) extension of the annealed primers with a
ives for use in screening large numbers of patients
thermostable DNA polymerase. Target specificity is
for mutations.45,46 Restriction enzyme cleavage or
determined by the choice of short (~20 bp) oligonu-
allele-specific oligonucleotide discriminant hybrid-
cleotide primers which are designed to hybridize to
ization47 performed on PCR-amplified DNA samples
opposite DNA strands flanking the sequence to be
are two other options. In the future, mutation detec-
amplified, with their 3’ ends facing inwards.
tion promises to become quicker, easier and cheaper.
Successive cycles of amplification result in a continu-
Indeed, automation may include the production of
ous doubling of, and exponential increase in, the
miniature hand-held diagnostic devices48 which
sequence copy number as newly synthesized copies
could potentially be used in the doctor’s surgery.
become available for primer binding. This extremely
versatile technique is typically capable of amplifying
The nature of human gene mutation
a single copy of a DNA sequence from the human
genome approximately 100-fold in a few hours. The One goal of predictive testing in adult-onset disorders
is to be able to infer the likely clinical phenotypetheory and practice of PCR have been well reviewed
elsewhere.30–33 (e.g. age of onset, severity) from knowledge of the
nature of the underlying gene mutation. The nature
of the genetic lesion involved will also determine
PCR-based methods of mutation detection
the type of test which can be used in the predictive
and analysis
testing/screening of hitherto asymptomatic relatives.
Point mutations and deletions are the commonest
The introduction of PCR has made possible the rapid
genetic lesions in the human genome. The remainder
detection and characterization of a wide variety of
comprise a mixed assortment of insertions, duplica-
different gene lesions underlying human genetic
tions, inversions and complex rearrangements.
disease.21 A number of mutation-screening methods
Characterized mutations occur not only in coding
have been devised to reduce the considerable
sequences but also in promoter regions, in splice
amount of labour involved in detecting and charac-
junctions, within introns and in polyadenylation sites.
terizing gene lesions causing human genetic disease
Mutations may interfere with any stage in the path-
(reviewed in Cotton;34 Forrest and Cotton;35 Rossiter
way of expression from gene to protein product.
and Caskey36). The most frequently used are briefly
The interested reader is referred to Cooper and
mentioned below. None detect all potentially detect-
Krawczak21 for in-depth coverage of this topic. These
able lesions within a gene and they vary with respect
different types of lesion will nevertheless be briefly
to such matters as ease of use, equipment require-
reviewed here.
ments and safety considerations.
The chemical mismatch cleavage technique37
serves to detect mismatched bases in hybrid duplexes
Single base-pair substitutions within the
formed between wild-type and mutant DNAs, and
coding region
was first applied diagnostically to the detection of
single base-pair substitutions in the factor IX (F9) Roughly 25% of point mutations are CGTG or
CGCA transitions, representing a 8-fold highergene causing haemophilia B using genomic DNA
from the patients concerned.38 Mismatches within frequency for this dinucleotide than that predicted
Adult-onset genetic disease 91
from random expectation.21,49 This is thought to be scriptional initiation and serve to decrease (or less
often, increase) the level of mRNA/protein productdue to the hypermutability of the methylated
dinucleotide CpG; spontaneous deamination of synthesized. What these lesions have in common is
their ability to alter or abolish the binding capacity5-methylcytosine (5mC) to thymidine in this doublet
gives rise to CTorGA transitions depending of cis-acting DNA sequence motifs for the trans-
acting protein factors that normally interact withupon the strand in which the 5mC is mutated. CpG
hypermutability in inherited disease implies that the them.
CpG sites in question are methylated in the germline
and thereby rendered prone to 5mC deamination.
Gross gene deletions
The spectrum of point mutations not in CpG
Gross gene deletions may arise through a number of
dinucleotides is also non-random.21,49 In principle,
different recombinational mechanisms including
the non-randomness of the initial mutation event,
homologous unequal recombination (occurring
the non-randomness of the DNA sequences under
between either related gene sequences or repetitive
study, differences in the relative efficiency with
elements). Alu sequences flanking deletion
which certain mutations are repaired, differences in
breakpoints have been noted in a considerable
phenotypic effect (and hence selection), or a bias in
number of human genetic conditions and may repres-
the clinical detection of such variants, may all play
ent hotspots for gene deletions.21
a role in determining the observed mutational
spectrum.
The majority of single base-pair substitutions caus-
Short gene deletions
ing human genetic disease alter the amino acid
Deletion breakpoint junctions flanking short
encoded (missense mutations), but a sizable propor-
(<20 bp) human gene deletions are non-random
tion result in the introduction of a termination codon
both at the nucleotide and dinucleotide levels, an
(nonsense mutations). The likelihood of clinical
observation consistent with a sequence-directed
detection is estimated to be about three times as
mechanism of mutagenesis.21 Direct repeats flanking
high for nonsense mutations as for missense muta-
the deleted sequence are a common finding, consist-
tions.21 Using a multi-domain molecular model of
ent with a model of slipped mispairing at the
the human factor IX protein, Wacey et al.26 have
replication fork. Two specific types of sequence have
shown that the likelihood that a factor IX mutation
been found at high frequency in the vicinity of short
(causing haemophilia B) will come to clinical atten-
gene deletions: polypyrimidine runs of at least 5 bp
tion is a complex function of the sequence character-
(YYYYY) and a ‘deletion hotspot consensus sequence’
istics of the factor IX (F9) gene, the nature of the
(TGRRKM).21
amino acid substitution, its precise location and
immediate environment within the protein molecule,
Short insertions
and its resulting effects on the structure and function
of the protein.
That insertional mutagenesis might be as intrinsically
non-random as point mutations and gene deletions
was strongly suggested by the findings of Fearon
Single base-pair substitutions within splice
et al.,51 who reported ten independent examples
sites
of DNA insertion within the same 170 bp intronic
Splicing defects have been estimated to make up
region of the DCC gene (a locus which has been
between 8% and 15% of all single base-pair substitu-
proposed to play an important role in human colorec-
tions causing human genetic disease.50 Mutations
tal neoplasia; see below). Insertional mutations invol-
appear to occur disproportionately at the most evolu-
ving the introduction of <10 bp DNA sequence into
tionarily conserved positions within the splice site.50
a gene coding region are (i) non-random and appear
This illusion results from a detection bias, caused by
to be highly dependent upon the local DNA
the relative phenotypic severity of these lesions.
sequence context and (ii) may be explained by those
Phenotypic consequences of splice site mutations
mechanisms held to be responsible for gene
include exon skipping and cryptic splice site util-
deletions.21
ization.
Inversions
Single base-pair substitutions within
Inversions are a highly unusual mutational mechan-
promoter regions
ism. The best known example is that found recently
in the factor VIII (F8) gene causing haemophilia A:A number of mutations are now known which occur
within gene promoter regions.21 These lesions disrupt this rearrangement occurs in about 40% of severely
affected patients and recurs at high frequency.52,53the normal processes of gene activation and tran-
D. Ravine and D.N. Cooper92
The mechanism responsible is thought to be homo- cerebellar ataxia type 1, fragile X E mental retarda-
tion, dentatorubral pallidoluysian atrophy, Machado-logous intra-chromosomal recombination between
a gene (F8A) located in intron 22 of the F8 gene Joseph disease and Freidriech’s ataxia.61 Fragile sites
FRAXF and FRA16A also contain unstable CGGand two additional homologues of the F8A gene
situated 500 kb upstream of the F8 gene. repeats but are not known to be associated with a
clinical abnormality. We explore the molecular gen-
etics of adult-onset Huntington’s disease in more
Expansion of unstable repeat sequences
detail below.
A recently recognized mutational mechanism
involves the instability of certain trinucleotide repeat
sequences (reviewed in Caskey et al.54 and Rousseau
Molecular genetics of adult-onset
et al.55). This mechanism was first reported as a
cause of the fragile X syndrome, the most frequent
disorders and their predictive
type of inherited mental retardation, which is associ-
importance
ated with the presence of a fragile site on the X
chromosome (FRAXA). The brain-expressed FMR1 With the chromosomal localization, isolation and
characterization of the underlying genes, predictivegene responsible was found to contain an ususual
(CGG)
n
repeat which exhibited copy number vari- testing for adult-onset disease is becoming increas-
ingly possible. Molecular diagnostics may howeveration of between 6 and 54 in normal healthy controls,
between 52 and >200 in phenotypically normal be a double-edged sword, and new ethical questions
concerning counselling before and after predictivetransmitting individuals (the ‘premutation’) and
between 300 and >1000 in affected males and testing, together with issues of confidentiality and
possible stigmatization of those found to be predis-females (the ‘full mutation’).56–59 Thus a continuum
exists between a copy number polymorphism present posed to genetic disease, are having to be addressed.
In principle, adult-onset disorders may arise as ain the general population, the asymptomatic premuta-
tion which involves limited expansion of (CGG)
n
result of the slow accumulation of a toxic substance,
slow tissue death or through inability to repair DNAcopy number, and the full mutation which appears
to require copy number expansion beyond a certain damage. In practice, the reasons for late onset are
usually unknown. With the new molecular geneticthreshold value. It would appear that those alleles
with a repeat copy number of <46 exhibit no techniques, however, an understanding of the
molecular pathology of the disease is not requiredmeiotic instability. The premutation represents a
comparatively small increase in (CGG)
n
copy number for the development of a diagnostic or predictive test.
In adult-onset disorders, it has often been difficultbut is not associated with mental impairment (at
least not in 97% of cases, a proportion comparable to distinguish low penetrance (see above) from the
absence of a disease state in a given individual.to that in the general population). However, those
individuals bearing the premutation exhibit a high Depending on the disease, the penetrance may
approach 100% if the individuals carrying the dis-probability of having either affected children or
affected grandchildren. Expansion of premutations to ease-related allele live long enough. On the other
hand, there may be individuals who carry thefull mutations only occurs during female meiotic
transmission. For alleles with 52113 repeat copies, disease-related allele but who will never develop
symptoms. Conventionally, this problem has beenthe premutation expands to the full mutation in 70%
of transmissions whereas the corresponding figure approached by means of life tables (age-of-onset
curves) which permit the empirical estimation of thefor alleles with >90 repeat copies is 100%. The
probability of repeat expansion thus correlates with probability that an at-risk yet asymptomatic indi-
vidual will develop the disease in question (e.g.the repeat copy number in the premutation allele,
consistent with a mechanism of slipped mispairing Huntington’s disease62). The advent of genetic
methods to allow predictive testing has provided anduring replication. Expansion of a sequence can thus
itself lead to further expansion, a process termed escape from the limitations inherent in the use of
life tables, and has at least in principle made possibledynamic mutation by Richards et al.60
The discovery of this novel mutational mechanism an accurate assessment of individual risk based upon
objective disease-specific criteria. From the patients’led to the recognition that the expansion of unstable
repeats was responsible for other human inherited point of view, the decision to opt for predictive
testing will depend very much upon their perceiveddiseases, often manifesting a wide range of clinical
severity and exhibiting inheritance patterns which risk, the severity of the disease in question, the
accuracy and reliability of predictive tests and theappear to be non-Mendelian. The other diseases
are spinal and bulbar muscular atrophy, myotonic benefits accruing to them from early presymptomatic
risk determination (e.g. prognostic information, pro-dystrophy, Huntington’s disease (see below), spino-
Adult-onset genetic disease 93
phylactic treatment). Clearly, the availability and Among those in whom misdiagnoses, sample mix-
up or clerical error have been excluded, it remainseffectiveness of therapies will be major factors in the
decision-making process. unclear whether this is due to allelic or locus
heterogeneity.The accuracy and reliability of new genetic tests
will in turn be determined by many factors including The association between unstable CAG repeat
expansion and HD is only the beginning of the tasklocus heterogeneity, the nature of any underlying
mutation or mutational mechanism, the type of test to unravel the molecular pathology of the defect and
to design appropriate therapies. Thus expansion of(i.e. linkage analysis vs. direct gene analysis) and the
influence of both environmental and epistatic factors. the CAG repeat may well be influenced by other
sequence variation at the HD locus78,79 or other loci.Various examples of common adult-onset disorders
are now briefly reviewed in order to illustrate the What is the phenotypic effect of the CAG repeat
expansion? Since HD is a dominant disorder, itimportance of such factors as allelic heterogeneity,
locus heterogeneity, variable penetrance, anticipa- probably results from a gain-of-function of the protein
product, possibly associated with transglutamination-tion, epistasis and the genotype-phenotype relation-
ship for accurate prediction of disease and effective mediated80 or glutamine zipper-mediated81 formation
of intramolecular complexes.genetic counselling.
Predictive testing for HD had been available for
some time prior to the cloning of the HD gene. Such
Huntington’s disease
procedures involved the use of polymorphic markers
flanking, and located at some distance from, the HDHuntington’s disease (HD) is a neurodegenerative
disorder, affecting about 1/10 000 people, which gene to track the disease allele through affected
pedigrees.82,83 Since linkage analysis is indirect, thecan be said to be inherited as a true autosomal
dominant trait. This is because HD homozygotes are results were inherently probabilistic. In the absence
of a therapy, many ethical issues concerningno more severely affected than heterozygotes. It is
now known that, in the vast majority of cases, HD informed consent, the testing of children, exclusion
testing during pregnancy and confidentiality haveis caused by the expansion of an unstable CAG
repeat (encoding a polyglutamine tract) within the been encountered.84 Now that direct mutation ana-
lysis of the HD gene is possible, diagnosis andHD gene which encodes the protein huntingtin.63
Normal individuals exhibit between 9 and 30 copies prediction of occurrence of HD will be that much
more accurate. However, the test should still beof this repeat whereas HD patients almost invariably
possess between 40 and 120 repeats.63–67 De novo interpreted with caution, because its specificity and
sensitivity remain to be determined.expansion of a CAG repeat in sporadic HD is
associated with the possession of an already large
repeat of 3438 units.68 Thus an expansion of >40
Familial Alzheimer’s disease
repeats is strongly diagnostic of HD, 3040 repeats
are difficult to interpret and <30 repeats makes HD Alzheimer’s disease (AD), an aetiologically hetero-
geneous disorder, is the most common cause ofunlikely but does not exclude it.69
HD exhibits anticipation in that a progressively dementia. It occurs with an incidence of 0.0035
between the age of 65 and 69 and 0.073 betweenearlier age of onset is often apparent in succeeding
generations. The basis for this appears to be a the age of 85 and 89.85 Three different loci of major
effect have so far been identified as being involvedcorrelation between the length of the CAG repeat
and the age of onset. The possession of longer in the aetiology of familial Alzheimer’s disease (FAD;
reviewed in Schellenberg.86 In the vast majorityrepeats (>55) is associated with an earlier age of
onset.70,71 However, below 50 repeats, there is little (70%) of cases of early-onset FAD (<50 years of
age), the disease is caused by mutations in a geneor no relationship with age of onset. Also, with a
late age of onset, the effect of repeat length appears (AD3) on the long arm of chromosome 14.87–90 This
gene encodes a protein (S182) with multiple trans-to diminish.72 Thus repeat number has little predictive
value for individuals.69 Repeat expansion appears to membrane domains but it is still unclear how per-
turbation of this protein results in the pathogenesisbe greater upon paternal transmission63,73–75 provid-
ing an explanation for the observed preponderance of AD.
A second Alzheimer’s susceptibility gene has beenof affected fathers of patients with early-onset HD.
HD patients exhibit significant levels of somatic identified on the long arm of chromosome 191 and
a number of different pathological lesions have beeninstability, resulting in repeat length mosaicism.76
This mosaicism appears to occur in a tissue-specific identified.92,93 This gene encodes a protein (STM2)
which is highly homologous to the chromosome-fashion; the cerebellum possesses smaller alleles than
other regions of the brain.76 On rare occasions, HD 14-encoded S182, raising the possibility that both
may play a role in the same biochemical pathway.patients do not exhibit CAG repeat expansion.77
D. Ravine and D.N. Cooper94
At least seven different point mutations in the
Venous thrombosis
amyloid precursor protein (APP) gene on the long
A hereditary predisposition to venous thrombosis
arm of chromosome 21 have also been characterized
may arise as a consequence of a lesion in any one
in AD patients. A fragment of this protein, b-amyloid,
of at least eleven different genes encoding
is an important constituent of the senile plaques
antithrombin III (AT 3 ), factor V (F5), factor XII (F12),
which characterize AD. These lesions are located
three chains of fibrinogen (FBA, FBB, FBG), heparin
within, or flank, the b-amyloid-encoding portion of
cofactor II (HCF2), plasminogen (PLG), protein C
the protein and appear to alter the cleavage of the
(PROC), protein S (PROS) and thrombomodulin
amyloid precursor protein and the secretion of the
(THBD).111 Some 300 different gene lesions (single
b-amyloid peptide.94,95 Mutations at this locus are
base-pair substitutions, deletions and insertions) have
associated with the development of late-onset FAD
so far been reported at these loci in patients with
(age 5060) but only account for about 2–3% of
thrombotic disease.
cases of early-onset FAD.96,97
However, an inherited deficiency state does not
In addition to these major gene effects, a fourth
automatically imply that the individual concerned
locus, the apolipoprotein E (APOE) gene on
will come to clinical attention. Although homozy-
chromosome 19q,98,99 is known to be involved in
gous or compound heterozygous individuals will
conferring susceptibility to late onset FAD (>60
almost certainly experience some thrombotic symp-
years). Specifically, there is an association between
toms (from relatively mild to neonatally fatal), most
the APOE e-4 allele (Cys112Arg) and late-onset
heterozygotes probably go undetected. Thus clinic-
AD100–104 the frequency of the e-4 allele is around
ally symptomatic AT3 deficiency in the general
0.40 in AD patients, compared to about 0.16 in
population occurs at an 80-fold lower frequency
Caucasian controls.105 Further, the risk of AD
(f=2.1×10–5) than does the clinically asympto-
increases and the mean age of onset decreases in
matic deficiency state (f=1.65× 10–3).112 Similarly,
proportion to the number of APOE e-4 alleles.99 This
clinically symptomatic protein C (PROC) deficiency
translates into an odds ratio of about 6 for those
occurs at a 2352-fold lower frequency (f=
carrying one or more e-4 alleles. There are however
2.8–6.2×10–5) than its asymptomatic counterpart
two problems with the direct use of these association
(f=1.4×10–3).113–115 One gene lesion (factor V
data for predictive purposes. First, since some 40%
Leiden, Arg506Gln), implicated in the aetiology
of AD patients do not possess the e-4 allele, its
of perhaps 30% of all cases of familial thrombotic
presence is neither a necessary nor a sufficient
disease, occurs with a prevalence of between 2%
condition for disease pathogenesis. It may be that
and 7% in the general population,116–118 but only a
the e-4 variant does not so much increase the risk
small fraction of carriers are clinically affected.
of AD as lower the age of onset and possibly
Thus, a defect in any one of the above mentioned
accelerate disease progression in those already pre-
genes is on its own probably insufficient to cause
disposed. Secondly, for it to have any predictive
thrombosis.
value at all, it will be necessary to perform epidemiol-
Other predisposing factors such as pregnancy,
ogical studies of APOE allele frequencies in patients
surgery, trauma and contraceptive pill usage appear
and controls from different populations and ethnic
to be required to precipitate a thrombotic epis-
groups. Finally, APOE interacts with both b-amyloid
ode.119,120 Genetic factors are clearly also involved.
and microtubule-associated proteins.106 Whilst the
For example, genotypic variation in the PROC gene
APOE e-4 variant is thought to promote the deposi-
promoter influences transcriptional efficiency of the
tion of b-amyloid in senile plaques and cerebral
PROC gene, the plasma protein C concentration and
vessels,107 the precise mechanism by which the e-4
ultimately thrombotic risk.121,122 Epistatic effects may
variant contributes to the pathogenesis of AD remains
also play a role. Thus in the Dutch population,
elusive. To make predictive testing even more diffi-
coinheritance of the factor V Leiden mutation with
cult, it may be that the influence of the APOE
protein C deficiency has been shown to increase the
modifier locus is itself modified by specific alleles at
risk of an individual coming to clinical attention by
unlinked loci e.g. a1-antichymotrypsin (ACT )108 and
some six-fold,123,124 although this risk may vary
very-low-density lipoprotein receptor (VLDLR).109
between different populations.125 Similarly, coinherit-
There are still several families who do not exhibit
ance of the factor V Leiden mutation with protein S
mutations in the known AD susceptibility genes,92,93
deficiency126 and antithrombin III deficiency127 also
suggesting that even more AD loci remain to be
increases the risk of an individual coming to clinical
discovered. The combination of variable penetrance
attention. The latter study has provided a potential
and locus heterogeneity exhibited by FAD has made
explanation for the particularly severe form of AT3
genetic risk analysis difficult, and the genetic deficiency segregating in some families. This is
because the AT 3 (chromosome 1q23–q25.1) and thecounselling of at-risk individuals hazardous.110
Adult-onset genetic disease 95
F5 (1q21q25) genes are both present in the same is considerable intrafamilial variability with respect
to clinical severity and age of onset,141 the identifica-region of the long arm of chromosome 1 and
therefore have the potential to cosegregate through tion of the particular gene involved in a given family
will be of some prognostic value, and accuratefamily pedigrees. Such combinatorial effects of differ-
ent loci will not be rare, since many of the deficiency predictive testing for PKD may aid early clinical
management of those proven to have a mutation.states giving rise to thrombosis are individually fairly
frequent in the general population (e.g. AT3, 0.0016;
PROC, 0.0014; F5, 0.020.07; F12, 0.023; HCF2,
Familial hypercholesterolaemia
0.009).
ATIII deficiency and protein C deficiency are Individuals with familial hypercholesterolaemia
(FHC) manifest defective catabolism of low-densityunusual in that they are recessive disorders with
respect to both mortality and reproductive fitness, lipoprotein (LDL), resulting in a 23-fold increase in
circulating plasma LDL cholesterol, and excessivebut dominant disorders with respect to morbidity,
albeit with reduced penetrance.119,128,129 Prophylactic deposition of cholesterol in the arterial wall, leading
to premature atherosclerosis and coronary heartanticoagulation is an effective treatment for those
who have already exhibited thrombotic manifesta- disease. Heterozygous FHC is one of the most
common single-gene disorders, and affects about 1tions. However, it is not normally recommended for
relatives of the clinically affected individual, even if in 500 individuals. The much rarer homozygotes are
more severely affected and exhibit a much earlierthey are proven carriers of a predisposing gene
lesion, unless there are clear mitigating circumstances onset of disease; in these individuals, angina pectoris,
myocardial infarction and sudden death usuallye.g. trauma, surgery. This is because, as indicated
earlier, the penetrance of predisposing gene lesions manifest between the ages of 5 and 30. Predictive
identification of those most at risk is important, sinceis usually low and many other factors contribute to
the eventual likelihood of a thrombotic episode. prophylactic treatment with cholesterol-lowering
agents can help to reduce mortality and morbidity.142Detection of clinically asymptomatic heterozygous
individuals by molecular genetic means may there- Mutations in the low-density lipoprotein receptor
(LDLR) gene are a major cause of familial hyperchole-fore be of limited predictive utility.
sterolaemia, and over 100 different point mutations
and short deletions have been characterized to date.
Autosomal dominant polycystic kidney
Extensive variation in clinical severity has been noted
disease
in FHC and this may be attributed, at least in part,
to allelic heterogeneity.143,144 Perhaps the bestAutosomal dominant polycystic kidney disease
(PKD), usually a late-onset disorder, affects about 1 example of the effect of allelic heterogeneity is the
comparison of patients homozygous for two distinctin 1000 individuals. It causes hypertension, urinary
tract infection, nephrolithiasis and ultimately renal LDLR gene lesions.145 individuals with >10 kb dele-
tion of the LDLR gene exhibited higher mean plasmafailure, as well as extra-renal manifestations in the
cardiovascular system (e.g. cerebral arterial aneur- cholesterol levels, an earlier age of onset of coronary
heart disease and a higher frequency of coronaryysms and heart valve prolapse) and gastrointestinal
tract (e.g. hepatic cysts and bowel diverticu- deaths than individuals with a Trp66Gly missense
mutation. Although allelic heterogeneity is verylosis).130 PKD is now known to exhibit considerable
locus heterogeneity, which greatly complicates important, intra-familial variability in clinical pheno-
type among carriers of the same lesion also occurs.146genetic counselling in affected families. The majority
of cases of autosomal dominant PKD have been One explanation for intra-familial variability in clin-
ical phenotype may be the cholesterol-loweringascribed to mutations in the PKD1 gene now identi-
fied on the short arm of chromosome 16.131 The effect of the E2 isoform of apolipoprotein E (APOE)
encoded by the e2 allele.147 APOE plays an importantmean age of onset of end-stage renal disease in PKD1
patients is about 56 years whereas it is about 70 in role in the regulation of plasma LDL-cholesterol
levels. The E2 isoform of APOE binds poorly tonon-PKD1 families.133 A second gene locus, PKD2,
is present on chromosome 4134,135 and the PKD2 LDLR, and it is thought that in its presence, ineffi-
cient delivery of cholesterol to hepatocytes maygene has recently been cloned.136 There is also
evidence for a third gene (PKD3) unlinked to the up-regulate the LDLR.
The LDLR gene is not the only locus known toother two.137,138
A European study has estimated the prevalence be defective in hypercholesterolaemia. Thus between
2% and 5% of FHC patients possess a substitutionof PKD1 among patients with ADPKD to be 85%.139
However, it is now known that PKD2 has a less of Arg3500 by Gln in the putative LDLR-binding
domain of apolipoprotein B-100, the main compon-severe phenotype,140 making it possible that PKD2
families have been under-ascertained. Although there ent of LDL.148,149 This APOB gene mutation reduces
D. Ravine and D.N. Cooper96
the affinity of LDL for LDLR, although it results in a usually of the order of 4550%.152 One function of
the encoded protein, p53, is to act as a recessiveless severe hypercholesterolaemia than found in FHC.
Even in the absence of an LDLR defect, the clinical tumour suppressor which, when inactivated, no
longer exerts its protective influence through a DNAseverity associated with the heterozygous possession
of the Arg3500Gln APOB gene mutation varies damage/cell cycle control pathway. Interestingly,
some TP53 mutations can also activate p53 as anenormously from person to person.150 A second,
much rarer, APOB gene lesion, which substitutes oncogene. Cancer, however, is clearly a multi-step
process with many more genes involved in theCys for Arg3531 in the same region, also
decreases LDLR binding affinity and results progression toward malignancy. The complexity of
this process is illustrated briefly by describing ourin FHC.151
current understanding of two common cancers, those
of the colon and breast:
Familial cancers
Cancer may arise as a result of chromosome loss,
Colon cancer
chromosome rearrangement or via more subtle muta-
tions in one or more of a large number of genes. Colorectal cancer is the most common cause of early
death in non-smokers in the Western world. FamilialCancer is a multi-step process and most cancers
arise as a consequence of somatic mutations which colon cancer falls into two main categories, familial
adenomatous polyposis (FAP) and hereditary non-by definition are not inherited. Whereas inherited
mutations may predispose to cancer, somatic muta- polyposis colon cancer (HNPCC) (reviewed in Burt
et al. 153). FAP is an autosomal dominant conditiontions serve to determine the site of the tumour.
Cellular genes in which mutations can give rise to characterized by polyps (adenomas) which, although
not initially malignant, may evolve into an invasivetumours are basically of three types. (i) Proto-
oncogenes. The products of these genes normally carcinoma. A large number of mutations, both germ-
line and somatic, have now been reported in thepromote cell proliferation, and the dominantly-acting
mutant products are excessively or inappropriately underlying APC tumour suppressor gene.154 HNPCC
is also inherited in an autosomal dominant fashionactive (i.e. they exhibit a ‘gain-of-function’).
Activation of a proto-oncogene may occur by single and results from mutations in at least four mismatch
repair genes: MLH1 (chromosome 3p), MSH2 (2p),base-pair substitution (e.g. RAS causing a variety of
tumours), gene amplification (e.g. N-MYC in PMS1 (2q) and PMS2 (7q). These lesions cause a
genome-wide instability of microsatellite DNAneuroblastoma) or translocation (e.g. the fusion
of the chromosome 9 ABL oncogene to the sequences, thereby resulting in a gain of alleles for
a high proportion of markers tested, regardless ofchromosome-22-encoded BCR gene in chronic mye-
loid leukaemia). (ii) Tumour suppressor genes (‘anti- their chromosomal location (reviewed in Dunlop155).
The development of colon cancer can be envisagedoncogenes’). These are genes whose wild-type prod-
ucts inhibit cell proliferation and whose mutant as the steady accumulation of non-allelic mutations
each of which confers a growth advantage upon theproducts have lost their function. Since both alleles
must be inactivated to alter cell function, tumorigen- cell. Since inherited mutation of the APC gene is
sufficient for polyposis to occur, loss of the secondesis requires a ‘two-hit’ mechanism. Usually, the first
mutation is inherited whilst the second involves the APC allele may be an early event in colon cancer.
Some adenomas have K-RAS (12p) oncogene muta-loss of all or a part of a specific chromosome by
deletion, non-disjunction, mitotic recombination etc tions which may speed the hitherto benign growths
along the path to malignancy. Additional inactivatingleading to a constitutional loss of heterozygosity for
genetic markers in the region of the gene in tumor mutations in the DCC (18q) and the TP53 (17p)
tumour suppressor genes probably lead to a furthertissue (e.g. at 5q21–22 near the APC gene in colon
cancer, see below). (iii) DNA repair genes. The loss of growth control. Since 1520% of colon
tumours have mutations in mismatch repair genes, itproducts of these genes are involved in maintaining
the integrity of the genome and the fidelity of may be that lesions in these ‘mutator genes’ dramatic-
ally increase the probability of mutation in the K-information transfer. Several mismatch repair genes
are known whose protein products are involved in RAS, DCC and TP53 genes, thereby greatly speeding
up progression toward malignancy. An understandingchecking the DNA for mismatched base-pairs. The
loss of both alleles leads to inefficient DNA replica- of the molecular genetics of this condition should
potentiate the early identification of at-risk indi-tion and repair.
The quest for the ‘Rosetta stone’ of cancer briefly viduals, allow the stage-specific assessment of tumour
progression in affected individuals and aid thestopped at the door of TP53, the most widely mutated
gene in human tumorigenesis; the frequency of development of new therapeutic approaches to
colorectal cancer.mutation varies from one cancer to another, but is
Adult-onset genetic disease 97
role in the aetiology of schizophrenia: a concordance
Breast cancer
of 45% for identical twins as compared to 15% for
Only a small proportion (5–10%) of cases of breast
fraternal twins.166
cancer is attributable to inherited factors, but this
Early genetic linkage studies suggested a locus of
proportion is higher in families with onset before the
major effect on chromosome 5, but this has not been
age of 45. It nevertheless manifests an autosomal
substantiated.167 Linkage has been repeatedly
dominant pattern of inheritance with high penetrance
reported for some families using markers on chromo-
(i.e. a lifetime risk of tumour development close to
some 6p22-p24,168–171 whilst other reports suggest
100%). The major locus responsible (BRCA1) has
the possible existence of susceptibility loci on chro-
been localized to chromosome 17q21, and inherited
mosomes 3p and 8p.172 Such results should be
lesions are thought to be responsible for development
treated with caution until confirmatory evidence is
of breast cancer in 8090% of families with both
forthcoming. This is because of (i) the inherent
breast and ovarian cancer but only about 4050%
problems of disease diagnosis and ascertainment in
of families with breast cancer alone (reviewed in
this condition, (ii) the relatively low lod scores
Bishop156). Interestingly, some 86% of lesions
generated so far, bearing in mind the number of
detected are predicted to result in the production of
markers tested, (iii) the difficulty in defining the
a truncated protein.157 Further, mutations occurring
model of inheritance for linkage studies (not a
in the 3’ third of the gene appear to be associated
problem when non-parametric methods such as
with a lower proportion of ovarian cancer.158 One
affected sib-pair analysis are used) and (iv) the fact
caveat to bear in mind when attempting to assess
that many studies have not scanned the entire
the significance of mutations found is that some
genome before linkage data are reported.
coding region variants are found in control popula-
More promising has been the recent finding of
tions of women.159
microdeletions of chromosome 22q11 in schizo-
Inherited mutations in a second breast cancer
phrenic patients with signs of velo-cardio-facial syn-
susceptibility locus, BRCA2, localized to chromo-
drome.173,174 These deletions have helped to define
some 13q12, appear to account for an additional
a 2 Mb region of the chromosome which may
40–50% of cases of familial breast cancer, but a
harbour a schizophrenia susceptibility gene. The
much lower proportion of cases with breast and
shortest region of overlap between the deletions
ovarian cancer.160–162 BRCA2 lesions are almost
is 300 kb and includes the catechol-O-methyltrans-
exlusively microdeletions which, as in the case of
ferase (COMT ) gene.174 The mechanism by which
BRCA1, are predicted to result in the production of
22q11 microdeletions increase the risk of schizophre-
a truncated protein.
nia is unknown. It is also unclear why only a
The genetics of breast cancer are as yet poorly
proportion of the known 22q11 microdeletions are
understood but there is already evidence for the
associated with schizophrenia/psychosis. Linkage
involvement of additional loci (H-RAS,163 TP53 and
studies (e.g. Vallada et al.175) are necessary to confirm
the ataxia telangiectasia gene, AT M ) in hereditary
these initial findings. The observation of Morris
susceptibility to breast cancer. Such changes may be
et al.176 that the genomes of schizophrenic subjects
important prognostically; thus the presence of a TP53
exhibit a significantly higher frequency of CAG triplet
mutation is associated with lower patient survival.
repeat expansion at undefined loci remains to be
satisfactorily explained.
The study of bipolar affective disorder (manic-
Schizophrenia and bipolar affective
depressive illness) has followed a similar path. Early
disorder
over-enthusiastic claims to have found loci of major
effect on chromosomes 11 and X have not beenCompared to some of the other disorders considered
here, schizophrenia and bipolar affective disorder subsequently confirmed (reviewed in Gelernter177),
with the possible exception of an associationare at a very early stage in their genetic analysis
(reviewed in Propping and No
¨
then,164 McGuffin between alleles at the monoamine oxidase A (MAOA)
locus.178 Some evidence has emerged regarding lociet al.165) This is a consequence of their multifactorial
nature, the absence of an obvious Mendelian mode on chromosomes 4,179 18, 180,181 and 21.182 Future
studies promise to determine the genetic basis forof inheritance, incomplete penetrance, and difficulty
in establishing reliable diagnostic criteria. This not- the twin clinical observations that mothers transmit
bipolar affective disorder more often than fathers183withstanding, schizophrenia has been known for
some time to run in families: the risk for offspring of and that offspring of affected parents tend to experi-
ence a more severe illness with earlier onset.184schizophrenic parents is 13%, considerably higher
than that in the general population of approximately While possible explanations such as imprinting and
anticipation spring to mind, it will be some consider-1% (reviewed in Plomin et al.166). Adoption studies
support the view that genetic factors play a major able time before gene cloning and characterization
D. Ravine and D.N. Cooper98
7. Efstratiadis A. Parental imprinting of autosomal
puts bipolar affective disorder on the same predictive
mammalian genes. Curr Op Genet Dev 1994; 4:265–80.
footing as some of the other common diseases
8. Barlow DP. Imprinting: a gamete’s point of view. Trends
considered here.
Genet 1994; 10:194–9.
9. Julier C, Hyer RN, Davies J, et al. Insulin-IGF2 region on
chromosome 11p encodes a gene implicated in HLA-
DR4-dependent diabetes susceptibility. Nature 1991;
Conclusions
354:1559.
In predictive medicine, identification of the affected
10. Heutink P, van der Mey AGL, Sandkuijl LA, et al. A gene
locus, and where possible the exact mutation, is
subject to genomic imprinting and responsible for
critically important. Novel molecular genetic tech-
hereditary paragangliomas maps to chromosome
11q23-qter. Hum Molec Genet 1992; 1:7–10.
niques have recently become available that have
potentiated both the localization and isolation of
11. Mariman ECM, van Beersum SEC, Cremers CWRJ,
Struycken PM, Ropers HH. Fine mapping of a putatively
disease-associated genes from the human genome.
imprinted gene for familial non-chromaffin paragangliomas
In several adult-onset disorders, the predictive testing
to chromosome 11q13.1: evidence for genetic
of at-risk individuals went through an initial phase
heterogeneity. Hum Genet 1995; 95:56–62.
of gene tracking using linked polymorphic markers.
12. Schinzel A, McKusick VA, Francomano C, Pearson PL.
Genetic counselling in these cases required a risk
Report of the committee for clinical disorders and
calculation based upon empirical data. The accuracy
chromosome aberrations. In Cuticchia AJ, Pearson PL, eds.
and reliability of these tests was, however, very
Human Gene Mapping 1993. A Compendium. Baltimore,
Johns Hopkins University Press, 1993:73572.
dependent upon marker informativity and pedigree
structure. After this first phase, tests then became
13. Harding AE. From the syndrome of Charcot, Marie and
Tooth to disorders of peripheral myelin proteins. Brain
more direct as the disease-associated genes were
1995; 118:809–18.
cloned. In many cases, predictive testing can now
14. Wolf U. The genetic contribution to the phenotype. Hum
be performed directly by screening for the possession
Genet 1995; 95:127–48.
of specific gene lesions. Acceptance of new predict-
15. Monari L, Chen SG, Brown P, et al. Fatal familial insomnia
ive tests will, however, depend on the perceived
and familial Creutzfeldt-Jakob disease: different prion
pre-test risk to the individual concerned, the typical
proteins determined by a DNA polymorphism. Proc Natl
clinical severity of the disease in question, the
Acad Sci USA 1994; 91:283942.
benefits of early diagnosis, and of course the accu-
16. Giannelli F, Green PM, Sommer S, et al. Haemophilia B:
racy and reliability of the test and its prognostic
database of point mutations and short additions and
implications. Diagnostic and prognostic accuracy
deletionsfifth edition. Nucleic Acids Res 1994;
will depend largely on the relationship between
22:35346.
genotype and clinical phenotype. Many factors influ-
17. Cooper DN, Clayton JF. DNA polymorphism and the study
ence the genotype-phenotype relationship, including
of disease associations. Hum Genet 1988; 78:299312.
the inheritance pattern of the disease, the nature of
18. Humphries SE. DNA polymorphisms of the apolipoprotein
the mutational mechanism, allelic heterogeneity,
genes—their use in the investigation of the genetic
component of hyperlipidaemia and atherosclerosis.
locus heterogeneity, variable penetrance and epist-
Atherosclerosis 1988; 72:89108.
asis as well as environmental variables. These factors
19. Sobell JL, Heston LL, Sommer SS Delineation of genetic
should be taken into account when counselling and
predisposition to multifactorial disease: a general approach
testing at-risk individuals.
on the threshold of feasibility. Genomics 1992; 12:1–6.
20. Lander ES, Schork NJ. Genetic dissection of complex traits.
Science 1994; 265:2037–47.
References
21. Cooper DN, Krawczak M. Human Gene Mutation. Oxford,
BIOS Scientific Publishers, 1993.
1. Wilkie AOM. The molecular basis of genetic dominance.
J Med Genet 1994; 31:89–98.
22. Hall JG. Somatic mosaicism: observations related to
clinical genetics. Am J Hum Genet 1988; 43:355–63.
2. Monk M. Genomic imprinting. Genes Dev 1988; 2:9215.
23. Morton N. Sequential test for the detection of linkage. Am
3. Monk M, Boubelik M, Lehnert S. Temporal and regional
J Hum Genet 1955; 7:266318.
changes in DNA methylation in the embryonic,
extraembryonic and germ cell lineages during mouse
24. Pakula AA, Sauer RT. Genetic analysis of protein stability
embryo development. Development 1987; 99:371–82.
and function. Ann Rev Genet 1989; 23:289–310.
4. Reik W. Genome imprinting. Chap. 5 in Grosveld F,
25. Alber T. Mutational effects on protein stability. Ann Rev
Kollias G, eds. Transgenic Animals. London, Academic
Biochem 1989; 58:76598.
Press, 1992:99–126.
26. Wacey AI, Krawczak M, Kakkar VV, Cooper DN.
5. Li E, Beard C, Jaenisch R. Role for DNA methylation in
Determinants of the factor IX mutational spectrum in
genomic imprinting. Nature 1993; 366:362–5.
haemophilia B: an analysis of missense mutations using a
multi-domain molecular model of the activated protein.6. Razin A, Cedar H. DNA methylation and genomic
imprinting. Cell 1994; 77:473–6. Hum Genet 1994; 94:594608.
Adult-onset genetic disease 99
27. Southern EM. Detection of specific sequences among DNA Wallace RB. Detection of sickle cell bS-globin allele by
hybridization with synthetic oligonucleotides. Proc Natlfragments separated by gel electrophoresis. J Mol Biol
1975; 98:503–17. Acad Sci USA 1983; 80:27882.
28. Alwine JC, Kemp DJ, Stark GR. Method for detection of 48. Gershon D. DNA diagnostic tools for the 21st century.
specific RNAs in gels by transfer to diazobenzyloxymethyl- Nature Medicine 1995; 1:1023.
paper and hybridization with DNA probes. Proc Natl Acad
49. Cooper DN, Krawczak M. The mutational spectrum of
Sci USA 1977; 74:5350–4.
single base-pair substitutions causing human genetic
29. Saiki RD, Scharf S, Faloona F, et al. Enzymatic
disease: patterns and predictions. Hum Genet 1990;
amplification of b-globin genomic sequences and
85:5574.
restriction site analysis for diagnosis of sickle-cell anemia.
50. Krawczak M, Reiss J, Cooper DN. The mutational
Science 1985; 230:13504.
spectrum of single base-pair substitutions in mRNA splice
30. Saiki RK, Gyllensten UB, Erlich HA. The polymerase chain
junctions of human genes: causes and consequences. Hum
reaction. Chap. 6 in Davies KE, ed. Genome Analysis: A
Genet 1992; 90:4154.
Practical Approach. Oxford, IRL Press, 1988.
51. Fearon ER, Cho KR, Nigro JM, et al. Identification of a
31. Williams JF. Optimization strategies for the polymerase
chromosome 18q gene that is altered in colorectal cancers.
chain reaction. Bio/Techniques 1989; 7:7628.
Science 1990; 247:4956.
32. Vosberg HP. The polymerase chain reaction: an improved
52. Lakich D, Kazazian HH, Antonarakis SE, Gitschier
method for the analysis of nucleic acids. Hum Genet 1989;
J. Inversions disrupting the factor VIII gene are a common
83:1–15.
cause of severe haemophilia A. Nature Genet 1993;
5:23641.
33. Bloch W. A biochemical perspective of the polymerase
chain reaction. Biochemistry 1991; 30:2735–47.
53. Naylor JA, Green PM, Rizza CR, Giannelli F. Analysis of
factor VIII mRNA reveals defects in everyone of 28
34. Cotton RGH. Detection of single base changes in nucleic
haemophilia A patients. Hum Molec Genet 1993; 2:117.
acids. Biochem J 1989; 263:1–10.
54. Caskey CT, Pizzuti A, Fu Y-H, Fenwick RG, Nelson DL.
35. Forrest S, Cotton RGH. Methods of detection of single base
Triplet repeat mutations in human disease. Science 1992;
substitutions in clinical genetic practice. Mol Biol Med
256:7849.
1990; 7:4519.
55. Rousseau F, Heitz D, Mandel J-L. The unstable and
36. Rossiter BJF, Caskey CT. Molecular scanning methods of
methylatable mutations causing the fragile X syndrome.
mutation detection. J Biol Chem 1990; 265:127536.
Hum Mutation 1992; 1:916.
37. Cotton RGH, Rodrigues NR, Campbell RD. Reactivity of
56. Rousseau F, Heitz D, Biancalana V, et al. Direct diagnosis
cytosine and thymine in single base-pair mismatches with
by DNA analysis of the fragile X syndrome of mental
hydroxylamine and osmium tetroxide and its application to
retardation. N Engl J Med 1991; 325:167381.
the study of mutations. Proc Natl Acad Sci USA 1988;
85:4397401.
57. Kremer EJ, Pritchard M, Lynch M, et al. Mapping of DNA
instability at the fragile X to a trinucleotide repeat
38. Montandon AJ, Green PM, Giannelli F, Bentley DR. Direct
sequence p(CCG)
n
. Science 1991; 252:1711–4.
detection of point mutations by mismatch analysis:
application to haemophilia B. Nucleic Acids Res 1989;
58. Verkerk AJMH, Pieretti M, Sutcliffe JS, et al. Identification
17:334757.
of a gene (FMR-1) containing a CGG repeat coincident
with a breakpoint cluster region exhibiting length variation
39. Myers RM, Sheffield VC, Cox DR. Detection of single-base
in fragile X syndrome. Cell 1991; 65:905–14.
changes in DNA: Ribonuclease cleavage and denaturing
gradient gel electrophoresis. Chap.5 in Davies KE, ed.
59. Fu Y-H, Kuhl DPA, Pizzuti A, et al. Variation of the CGG
Genome Analysis: A Practical Approach, Oxford, IRL
repeat at the fragile X site results in genetic instability:
Press, 1988.
resolution of Sherman paradox. Cell 1991; 67:104758.
40. Fodde R, Losekoot M. Mutation detection by denaturing
60. Richards RI, Holman K, Friend K, et al. Evidence of
gradient gel electrophoresis (DGGE). Hum Mutation 1994;
founder chromosomes in fragile X syndrome. Nature Genet
3:83–94.
1992; 1:25760.
41. Orita M, Suzuki Y, Sekiya T, Hayashi K. Rapid and
61. Willems PJ. Dynamic mutations hit double figures. Nature
sensitive detection of point mutations and DNA
Genet 1994; 8:2135.
polymorphisms using the polymerase chain reaction.
62. Harper PS, Newcombe PG. Age of onset and lifetable risks
Genomics 1989a; 5:874–9.
in genetic counselling for Huntington’s disease. J Med
42. Orita M, Iwahana H, Kanazawa H, Hayashi K, Sekiya T.
Genet 1992; 29:239–42.
Detection of polymorphisms of human DNA by gel
63. Huntington’s Disease Collaborative Research Group. A
electrophoresis as single-strand conformation
novel gene containing a trinucleotide repeat that is
polmorphisms. Proc Natl Acad Sci USA 1989b;
expanded and unstable on Huntington’s disease
86:276670.
chromosomes. Cell 1993; 72:97183.
43. Hayashi K, Yandell DW. How sensitive is PCR-SSCP? Hum
64. Duyao M, Ambrose C, Myers R, et al. Trinucleotide repeat
Mutation 1993; 2:338–46.
length instability and age of onset in Huntington’s disease.
44. Gyllensten UB. PCR and DNA sequencing. Biotechniques
Nature Genet 1993; 4:387–92.
1989; 7:7008.
65. Snell RG, MacMillan JC, Cheadle JP, et al. Relationship
45. Chehab FF. Molecular diagnostics: past, present and
between trinucleotide repeat expansion and phenotypic
future. Hum Mutation 1993; 2:3317.
variation in Huntington’s disease. Nature Genet 1993;
4:3937.46. Syvanen A-C, Landegren U. Detection of point mutations
by solid-phase methods. Hum Mutation 1994; 3:172–9.
66. Telenius H, Kremer HPH, Theilmann J, et al. Molecular
analysis of juvenile Huntington disease: the major47. Connor BJ, Reyes AA, Morin C, Hakura K, Teplitz RL,
D. Ravine and D.N. Cooper100
influence on (CAG)
n
repeat length is the sex of the affected 83. Lazarou LP, Meredith AL, Myring JM, et al. Huntington’s
disease: predictive testing and the molecular geneticsparent. Hum Molec Genet 1993; 2:153540.
laboratory. Clin Genet 1993; 43:1506.
67. Kremer B, Goldberg P, Andrew SE, et al. A worldwide
84. Ball DM, Harper PS. Presymptomatic testing for late-onset
study of the Huntington’s disease mutation. The sensitivity
genetic disorders: lessons from Huntington’ disease. FASEB
and specificity of measuring CAG repeats. N Engl J Med
J 1992; 6:281819.
1994; 330:14016.
85. Bachman DL, Wolf PA, Linn RT, et al. Incidence of
68. Myers RH, MacDonald ME, Koroshetz WJ, et al. De novo
dementia and probable Alzheimer’s disease in a general
expansion of a (CAG)
n
repeat in sporadic Huntington’s
population: the Framingham study. Neurology 1993;
disease. Nature Genet1993; 5:168–72.
43:51519.
69. Read AP. Huntington’s disease: testing the test. Nature
86. Schellenberg GD. Genetic dissection of Alzheimer disease,
Genet 1993; 4:329–30.
a heterogeneous disorder. Proc Natl Acad Sci U.S.A. 1995;
92:85529.
70. Craufurd D, Dodge A. Mutation size and age at onset in
87. Mullan M, Houlden H, Crawford F, Kennedy A, Rogues P,
Huntington’s disease. J Med Genet 1993; 30:100811.
Rossor M. Age of onset in familial early onset Alzheimer’s
71. Stine OC, Pleasant N, Franz ML, Abbott MH, Folstein SE,
disease correlates with genetic aetiology. Am J Med Genet
Ross CA. Correlation between the onset age of
1993; 48:12930.
Huntington’s disease and length of the trinucleotide repeat
88. Sherrington R, Rogaev EI, Liang Y, et al. Cloning of a gene
in IT-15. Hum Molec Genet 1993; 2:1547–9.
bearing missense mutations in early-onset familial
Alzheimer’s disease. Nature 1995; 375:75460.
72. Kremer B, Squitieri F, Telenius H, et al. Molecular analysis
of late onset Huntington’s disease. J Med Genet 1993;
89. Cruts M, Backhovens H, Theuns J, et al. Genetic and
30:991–5.
physical characterization of the early-onset Alzheimer’s
disease AD3 locus on chromosome 14q24.3. Hum Molec
73. MacDonald ME, Barnes G, Srinidhi J, et al. Gametic but
Genet 1995; 4:1355–64.
not somatic instability of CAG repeat length in
90. Alzheimer’s Disease Collaborative Group. The structure of
Huntington’s disease. J Med Genet 1993; 30:9826.
the presenilin 1 (S182) gene and identification of six novel
74. Goldberg YP, Kremer B, Andrew SE, et al. Molecular
mutations in early onset AD families. Nature Genet 1995;
analysis of new mutations for Huntington’s disease:
11:21922.
intermediate alleles and sex of origin effects. Nature Genet
91. Levy-Lahad E, Wijsman EM, Nemens E, et al. A familial
1993; 5:1749.
Alzheimer’s disease locus on chromosome 1. Science
1995a; 269:970–3.
75. Trottier Y, Biancalana V, Mandel JL. Instability of CAG
repeats in Huntington’s disease: relation to parental
92. Levy-Lahad E, Wasco W, Poorkaj P, et al. Candidate gene
transmission and age of onset. J Med Genet 1994;
for the chromosome 1 familial Alzheimer’s disease locus.
31:377–82.
Science 1995b; 269:973–7.
93. Rogaev EI, Sherrington R, Rogaeva EA, et al. Familial
76. Telenius H, Kremer B, Goldberg YP, et al. Somatic and
Alzheimer’s disease in kindreds with missense mutations in
gonadal mosaicism of the Huntington disease gene CAG
a gene on chromosome 1 related to the Alzheimer’s
repeat in brain and sperm. Nature Genet 1994;
disease type 3 gene. Nature 1995; 376:7758.
6:409414.
94. Felsenstein KM, Hunihan LW, Roberts SB. Altered cleavage
77. Andrew SE, Goldberg YP, Kremer B, et al. Huntington
and secretion of a recombinant beta-APP bearing the
disease without CAG expansion: phenocopies or errors in
Swedish familial Alzheimer’s disease mutation. Nature
assignment? Am J Hum Genet 1994; 54:852–63.
Genet 1994; 6:2515.
78. Telenius H, Almqvist E, Kremer B, et al. Somatic
95. Haass C, Hung AY, Selkoe DJ, Teplow DB. Mutations
mosaicism in sperm is associated with intergenerational
associated with a locus for familial Alzheimer’s disease
(CAG)
n
changes in Huntington disease. Hum Molec Genet
result in alternative processsing of amyloid beta-protein
1995; 4:18995.
precursor. J Biol Chem 1994; 269:17741–8.
96. Goate A, Chartier-Harlin M-C, Mullan M, et al.
79. Almqvist E, Spence N, Nichol K, et al. Ancestral
Segregation of a missense mutation in the amyloid
differences in the distribution of the d2642 glutamic acid
precursor protein gene with familial Alzheimer’s disease.
polymorphism is associated with varying CAG repeat
Nature 1991; 349:704–6.
lengths on normal chromosomes: insights into the genetic
evolution of Huntington disease. Hum Molec Genet 1995; 97. Murrell J, Farlow M, Ghetti B, Benson MD. A mutation in
4:20714. the amyloid precursor protein associated with hereditary
Alzheimer’s disease. Science 1991; 254:979.
80. Green H. Human genetic diseases due to codon
98. Pericak-Vance MA, Bebout JL, Gaskell PJ, et al. Linkage
reiteration: relationship to an evolutionary mechanism.
studies in familial Alzheimer disease: evidence for
Cell 1993; 74:955–6.
chromosome 19 linkage. Am J Hum Genet 1991;
81. Perutz MF, Johnson T, Suzuki M, Finch JT. Glutamine
48:103450.
repeats as polar zippers: their possible role in inherited
99. Corder EH, Saunders AM, Strittmatter WJ, et al. Gene dose
neurodegenerative diseases. Proc Natl Acad Sci USA 1994;
of apolipoprotein E type 4 allele and the risk of
91:53558.
Alzheimer’s disease in late onset families. Science 1993;
261:9213.
82. Skraastad MI, Verwest A, Bakker E, et al. Presymptomatic,
prenatal and exclusion testing for Huntington disease using 100. Strittmatter WJ, Saunders AM, Schmechel D, et al.
Apolipoprotein E: high activity binding to b-amyloid andseven closely linked DNA markers. Am J Med Genet 1991;
39:217–22. increased frequency of type 4 allele in late-onset familial
Adult-onset genetic disease 101
Alzheimer disease. Proc Natl Acad Sci USA 1993; 118. Beauchamp NJ, Daly ME, Hamptom KK, Cooper PC,
Preston FE, Peake IR. High prevalence of a mutation in the90:197781.
factor V gene within the U.K. population: relationship to
101. Tsai MS, Tangalos EG, Petersen RC, et al. Apolipoprotein
activated protein C resistance and familial thrombosis. Br
E: risk factor for Alzheimer disease. Am J Hum Genet
J Haematol 1994; 88:21922.
1994; 54:643–9.
119. Allaart CF, Poort SW, Rosendaal FR, Reitsma PH, Bertina
102. Chartier-Harlin MC, Parfitt M, Legrain S, et al.
RM, Brit E. Increased risk of venous thrombosis in carriers
Apolipoprotein E, epsilon 4 allele as a major risk factor for
of hereditary protein C deficiency defect. Lancet 1993;
sporadic early and late-onset forms of Alzheimer’s disease:
341:1348.
analysis of the 19q13.2 chromosomal region. Hum Molec
120. Henkens CMA, van der Meer J, Hillege JL, van der Schaaf
Genet 1994; 3:569–74.
W, Bom VJJ, Halie MR. The clinical expression of
103. Polvikoski T, Sulkava R, Haltia M et al. Apolipoprotein E,
hereditary protein C and protein S deficiency: a relation to
dementia and cortical deposition of b-amyloid protein. N
clinical thrombotic risk factors and to levels of protein C
Engl J Med 1995; 333:1242–7.
and protein S? Blood Coag Fibrinol 1993; 4:55562.
104. St Clair D, Rennie M, Slorach E, Norrman J, Yates C et al.
121. Spek CA, Koster T, Rosendaal FR, Bertina RM, Reitsma PH.
Apolipoprotein E e4 allele is a risk factor for familial and
Genotypic variation in the promoter region of the protein
sporadic presenile Alzheimer’s disease in both
C gene is associated with plasma protein C levels and
homozygote and heterozygote carriers. J Med Genet 1995;
thrombotic risk. Arterioscler Thromb Vasc Biol 1995;
32:642–4.
15:21418.
105. Saunders AM, Strittmatter WJ, Schmechel D, St George-
122. Scopes D, Berg L-P, Krawczak M, Kakkar VV, Cooper DN.
Hyslop PH, Pericak-Vance MA, Joo SH. Association of
Polymorphic variation in the human protein C (PROC)
apolipoprotein e-4 allele with late onset familial and
gene promoter can influence transcriptional efficiency in
sporadic Alzheimer’s disease. Neurology 1993;
vitro. Blood Coag Fibrinol. 1995; 6:317–21.
43:146772.
123. Koeleman BPC, Reitsma PH, Allaart CF, Bertina RM.
106. Strittmatter WJ, Roses AD. Apolipoprotein E and Alzheimer
Activated protein C resistance as an additional risk factor
disease. Proc Natl Acad Sci USA 1995; 92:4725–7.
for thrombosis in protein C-deficient families. Blood 1994;
107. Schmechel DE, Saunders AM, Strittmatter WJ, et al.
84:10315.
Increased amyloid beta-peptide deposition in cerebral
124. Gandrille S, Greengard JS, Alhenc-Gelas M, et al.
cortex as a consequence of apolipoprotein E genotype in
Incidence of activated protein C resistance caused by the
late-onset Alzheimer disease. Proc Natl Acad Sci USA
ARG 506 GLN mutation in factor V in 113 unrelated
1993; 90:9649–53.
symptomatic protein C-deficient patients. Blood 1995;
108. Kamboh MI, Sanghera DK, Ferrell RE, DeKosky ST.
86:21924.
APOE*4-associated Alzheimer’s disease risk is modified by
125. Hallam PH, Millar DS, Krawczak M, Kakkar VV,
a1-antichymotrypsin polymorphism. Nature Genet 1995;
Cooper DN. Population differences in the frequency of the
10:486–8.
factor V Leiden variant among individuals with clinically
109. Okuizumi K, Onodera O, Namba Y, et al. Genetic
symptomatic protein C deficiency. J Med Genet 1997; in
association of the very low density lipoprotein (VLDL)
press.
receptor gene with sporadic Alzheimer’s disease. Nature
126. Zller B, Berntsdottir A, Garcia de Frutos P, Dahlbck B.
Genet 1995; 11:207–9.
Resistance to activated protein C as an additional genetic
110. Mullan M, Crawford F, Buchanan J. Technical feasibility of
risk factor in hereditary deficiency of protein S. Blood
genetic testing for Alzheimer’s disease. Alzheimer Disease
1995; 85:3518–23.
& Associated Disorders 1994; 8:10215.
127. van Boven HH, Reitsma PH, Rosendaal FR, et al. Factor V
Leiden (FV R506Q) in families with inherited antithrombin111. Tuddenham EGD, Cooper DN. The Molecular Genetics of
deficiency. Thromb Haemost 1996; 75:417–21.Haemostasis and its Inherited Disorders. Oxford University
Press, Oxford 1994.
128. Rosendaal FR, Heijboer H, Brit E, et al. Mortality in
hereditary antithrombin-III deficiency—1830 to 1989.
112. Tait RC, Walker ID, Perry DJ, et al. Prevalence of
Lancet 1991; 337:2602.
antithrombin deficiency in the healthy population. Br
J Haematol 1994; 87:106–12.
129. Allaart CF. Hereditary deficiency of protein S and protein
C. PhD Thesis. University of Leiden, The Netherlands,
113. Gladson CL, Scharrer I, Hach V, Beck KH, Griffin JH. The
1994.
frequency of type I heterozygous protein S and protein C
deficiency in 141 unrelated young patients with venous
130. Lieske JC, Toback FG. Autosomal dominant polycystic
thrombosis. Thromb Haemost 1988; 59:1822.
kidney disease. J Am Soc Nephrol 1993; 3:1442–50.
114. Miletich J, Sherman L, Broze G Absence of thrombosis in
131. European Polycystic Kidney Disease Consortium. The
subjects with heterozygous protein C deficiency. N Engl
polycystic kidney disease 1 gene encodes a 14kb transcript
J Med 1987; 17:991–6.
and lies within a duplicated region on chromosome 16.
Cell 1994; 77:881–94.
115. Tait RC, Walker ID, Reitsma PH, et al. Prevalence of
protein C deficiency in the healthy population. Thromb
132. Parfrey PS, Bear JC, Morgan J, et al. The diagnosis and
Haemost 1995; 73:87–93.
prognosis of autosomal dominant polycystic kidney
disease. N Engl J Med 1990; 323:1085–90.
116. Bertina RM, Koeleman BPC, Koster T, et al. Mutation in
blood coagulation factor V associated with resistance to
133. Ravine D, Walker RG, Gibson RN, et al. Phenotype and
activated protein C. Nature 1994; 369:64–7.
genotype heterogeneity in autosomal dominant polycystic
kidney disease. Lancet 1992; 340:1330–3.
117. Svensson PJ, Dahlbck B. Resistance to activated protein C
as a basis for venous thrombosis. N Engl J Med 1994; 134. Kimberling WJ, Kumar S, Gabow PA, Kenyon JB, Connolly
CJ, Somlo S. Autosomal dominant polycystic kidney330:51722.
D. Ravine and D.N. Cooper102
disease: localization of the second gene to chromosome 152. Soussi T. The p53 tumor suppressor gene: from molecular
biology to clinical investigation. Chap. 7 in Cowell JK, ed.4q13-q23. Genomics 1993;18:46772.
Molecular Genetics of Cancer. Oxford, BIOS Scientific
135. Peters DJM, Spruit L, Saris JJ, et al. Chromosome 4
Publishers 1995:13578.
localization of a second gene for autosomal dominant
polycystic kidney disease. Nature Genet 1993; 5:35962.
153. Burt RW. Genetics of colon cancer: impact of inheritance
on colon cancer risk. Annu Rev Med 1995; 46:371–9.
136. Mochizuki T, Wu G, Hayashi T, et al. PKD2, a gene for
polycystic kidney disease that encodes an integral
154. Nagase H, Nakamura Y. Mutations of the APC
membrane protein. Science 1996; 272:133942.
(Adenomatous Polyposis Coli) gene. Hum Mutation 1993;
2:42534.
137. de Almeida S , de Almeida E , Peters D, et al. Autosomal
dominant polycystic kidney disease: evidence for the
155. Dunlop MG. Molecular genetics of colon cancer. Chap. 6
existence of a third locus in a Portuguese family. Hum
in Cowell JK, ed. Molecular Genetics of Cancer. Oxford,
Genet 1995; 96:83–8.
BIOS Scientific Publishers, 1995:11334.
138. Daoust MC, Reynaud DM, Bichet DG, Somlo S. Evidence
156. Bishop DT. Genetics of breast and ovarian cancer. Chap. 5
for a third genetic locus for autosomal dominant polycystic
in Molecular Genetics of Cancer. Cowell JK, ed. Molecular
kidney disease. Genomics 1995; 25:7336.
Genetics of Cancer. Oxford, BIOS Scientific Publishers,
1995:93111.
139. Peters DJM, Sandkuijl LA, Snijdewint FGM, et al. Genetic
heterogeneity of polycystic kidney disease in Europe.
157. Shattuck-Eidens D, McClure M, Simard J et al. A
Contrib Nephrol 1992; 97:128–39.
collaborative survey of 80 mutations in the BRCA1 breast
and ovarian cancer susceptibility gene. JAMA 1995;
140. Constantinou-Deltas CD, Papageorgiou E, Boteva K, et al.
273:53541.
Genetic heterogeneity in adult dominant polycystic kidney
disease in Cypriot families. Hum Genet 1995; 95:416–23.
158. Gayther SA, Warren W, Mazoyer S, et al. Germline
mutations of the BRCA1 gene in breast and ovarian cancer
141. Milutinovic J, Rust PF, Fialkow PJ, et al. Intrafamilial
families provide evidence for a genotype-phenotype
phenotypic expression of autosomal dominant polycystic
correlation. Nature Genet 1995; 1:428–33.
kidney disease Am J Kid Dis 1992; 5:46572.
142. Levine GN, Keaney JF, Vita JA. Cholesterol reduction in 159. Durocher F, Shattuck-Eidens D, McClure M, et al.
cardiovascular disease. Clinical benefits and possible Comparison of BRCA1 polymorphisms, rare sequence
mechanisms. N Engl J Med 1995; 332:512–21. variants and/or missense mutations in unaffected and
breast /ovarian cancer populations. Hum Molec Genet
143. Hobbs HH, Brown MS, Goldstein JL, Russell DW. Deletion
1996; 5:83542.
of exon encoding cysteine-rich repeat of low-density
lipoprotein receptor alters its binding specificity in a
160. Wooster R, Neuhausen SL, Mangion J, et al. Localization
subject with familial hypercholesterolemia. J Biol Chem
of a breast cancer susceptibility gene, BRCA2,to
1986; 261:1311420.
chromosome 13q12–13. Science 1994; 265:208890.
144. Soutar AK. Familial hypercholesterolaemia and LDL
161. Wooster R, Bignell G, Lancaster J et al. Identification of the
receptor mutations. J Intern Med 1992; 231:633–41.
breast cancer susceptibility gene BRCA2. Nature 1995;
378:78992.
145. Moorjani S, Roy M, Torres A, et al. Mutations of low-
density-lipoprotein-receptor gene, variation in plasma
162. Tavtigian SV, Simard J, Rommens J, et al. The complete
cholesterol, and expression of coronary heart disease in
BRCA2 gene and mutations in chromosome 13q-linked
homozygous familial hypercholesterolaemia. Lancet 1993;
kindreds. Nature Genet 1996;12:333–7.
341:1303–6.
163. Phelan CM, Rebbeck TR, Weber BL, et al. Ovarian cancer
146. Kotze MJ, Davis HJ, Bissbort S, Langenhoven E, Brusnicky
risk in BRCA1 carriers is modified by the HRAS1 variable
J, Oosthuizen CJ. Intrafamilial variability in the clinical
number of tandem repeat (VNTR) locus. Nature Genet
expression of familial hypercholesterolemia: importance of
1996; 12:30911.
risk factor determination for genetic counselling. Clin
164. Propping P, No
¨
then MM. Schizophrenia: genetic tools for
Genet 1993; 43:295–9.
unraveling the nature of a complex disorder. Proc Natl
147. Sass C, Giroux L-M, Ma Y, et al. Evidence for a
Acad Sci USA 1995; 92:76078.
cholesterol-lowering gene in a French-Canadian kindred
165. McGuffin P, Owen MJ, Farmer AE. Genetic basis of
with familial hypercholesterolemia. Hum Genet 1995;
schizophrenia. Lancet 1995; 346:678–82.
96:216.
166. Plomin R, Owen MJ, McGuffin P. The genetic basis of
148. Tybjaerg-Hansen A, Humphries SE. Familial defective
complex human behaviors. Science 1994; 264:17339.
apolipoprotein B-100: a single mutation that causes
167. Owen M, McGuffin P. The molecular genetics of
hypercholesterolemia and premature coronary artery
schizophrenia; blind alleys, acts of faith, and difficult
disease. Athersclerosis 1992; 96:91107.
science. Br Med J 1992; 305:664–5.
149. Myant NB. Familial defective apolipoprotein B-100: a
168. Wang S, Sun C, Walczak CA, et al. Evidence for a
review, including some comparisons with familial
susceptibility locus for schizophrenia on chromosome
hypercholesterolaemia. Atherosclerosis 1994; 104:1–18.
6pter-p22. Nature Genet 1995; 10:41–6.
150. Davignon J, Dufour R, Roy M, et al. Phenotypic
169. Straub RE, MacLean CJ, O’Neill FA, et al. A potential
heterogeneity associated with defective apolpoprotein
vulnerability locus for schizophrenia on chromosome
B-1000 and occurrence of the familial
6p2422: evidence for genetic heterogeneity. Nature
hypercholesterolemia phenotype in the absence of an LDL-
Genet 1995; 11:287–93.receptor defect within a Canadian kindred. Eur J Epidemiol
1992; 8(S1):10–17.
170. Moises HW, Yang L, Krisbjarnarson H, et al. An
international two-stage genome-wide search for151. Pullinger CR, Hennessy LK, Chatterton JE, et al. Familial
ligand-defective apolipoprotein B. J Clin Invest 1995; schizophrenia susceptibility genes. Nature Genet 1995;
11:3214.95:122534.
Adult-onset genetic disease 103
171. Schwab SG, Albus M, Hallmayer J, et al. Evaluation of a 178. Lim LCC, Powell J, Sham P, et al. Evidence for a genetic
association between alleles of monoamine oxidase A genesusceptibility gene for schizophrenia on chromosome 6p
by multipoint affected sib-pair linkage analysis. Nature and bipolar affective disorder. Am J Med Genet 1995;
60:32531.Genet 1995; 11:325–7.
172. Pulver AE, Lasseter VK, Kasch L et al. Schizophrenia: a 179. Blackwood DHR, Lin He, Morris SW, et al. A locus for
bipolar affective disorder on chromosome 4p. Naturegenome scan targets chromosomes 3p and 8p as potential
sites of susceptibility genes. Am J Med Genet 1995; Genet 1996; 12:427–30.
60:252–60.
180. Berrettini WH, Ferraro TN, Goldin LR, et al. Chromosome
18 DNA markers and manic-depressive illness: evidence173. Lindsay EA, Morris MA, Gos A, et al. Schizophrenia and
chromosomal deletions within 22q11.2. Am J Hum Genet for a susceptibility gene. Proc Natl Acad Sci USA 1994;
91:591821.1995; 56:1502–3.
174. Karayiorgou M, Morris MA, Morrow B, et al. 181. Freimer NB, Reus VI, Escamilla MA, et al. Genetic
mapping using haplotype, association and linkage methodsSchizophrenia susceptibility associated with interstitial
deletions of chromosome 22q11. Proc Natl Acad Sci USA suggests a locus for severe bipolar disorder (BPI) at
18q22-q23. Nature Genet 1996; 12:436–41.1995; 92:7612–16.
175. Vallada HP, Gill M, Sham P. Linkage studies on 182. Straub RE, Lehner T, Luo Y, et al. A possible vulnerability
locus for bipolar affective disorder on chromosomechromosome 22 in familial schizophrenia. Am J Med
Genet 1995; 60:139–46. 21q22.3. Nature Genet 1994; 8:291–6.
183. McMahon FJ, Stine OC, Meyers DA, Simpson SG, DePaulo176. Morris AG, Gaitonde E, McKenna PJ, Mollon JD,
Hunt DM. CAG repeat expansions and schizophrenia: JR. Patterns of maternal transmission in bipolar affective
disorder. Am J Hum Genet 1995; 56:127786.association with disease in females and with early age-at-
onset. Hum Molec Genet 1995; 4:195761
184. McInnis MG, McMahon FJ, Chase GA, Simpson SG,
Ross CA, DePaulo JR. Anticipation in bipolar affective177. Gelernter J. Genetics of bipolar affective disorder: time for
another reinvention? Am J Hum Genet 1995; 56:1262–6. disorder. Am J Hum Genet 1993; 53:385–90.
... Although most single-gene disorders proba bly have a strong genotype-phenotype rela tionship, the genes are not insulated from others nor can they be influenced by other genetic influences such as their immediate sequence environment or other gene loci . For example, precise genetic abnormalities infor mation in sickle-cell anemia (homozygous parents with p-globin gene lesion) fails to predict the exact clinical phenotype (25). The variability in severity of the disease is due to the effects of genetic variation either with in the regulatory regions that are linked to the gene or the far upstream regions. ...
Article
Full-text available
The human genome project was initiated in 1990 to study the structure and characteristics of human DNA that are important for understanding gene functions and their relation to diseases. The large-scale genome research has driven the technology advancement in genetic testing, drug design, gene therapy, and other genetic related areas such as pharmacogenetics. Although the project reveals potential benefits, it raises ethical, legal, and social issues. The outcomes of individuals' genetic information disclosure may lead to confidentiality and genetic discrimination issues. In addition, clinical relevance of genetic testing and psychological effect from the results are debatable. This article discusses the potential benefits and risks from the human genome project.
... PKD1 may component), in autosomal dominant polycystic kidney require the presence of PKD2 for stable expression disease (ADPKD) and in many other adult-onset [5]. It should be clear, however, that late and slow genetic diseases [1]. In these diseases, the first clinical progression may be found in some PKD1 families and manifestations often occur in adulthood, after 40 years in some members of PKD1 families [2]. ...
Article
A 5‐year‐old, smooth‐haired miniature dachshund exhibited a progressive history of frequent episodes of aggression towards objects, head pressing, circling and intermittent pelvic limb ataxia over a period of 2 weeks. Visual impairment was also noticed several months prior. Neurolocalisation was consistent with forebrain disease. Blood and urine analysis, magnetic resonance imaging of the brain and cerebrospinal fluid analysis were all unremarkable. The patient was euthanased due to disease progression, and postmortem histopathological analysis and electronic microscopy identified the accumulation of intracellular granular material in a number of neurons, mainly located in the thalamus and hypothalamus. This supported a diagnosis of neuronal ceroid lipofuscinosis. Genetic testing showed no mutations in the causative genes reported for this breed. This case reports the occurrence of a potentially novel subtype of neuronal ceroid lipofuscinosis, due to a currently unidentified genetic mutation, leading to an adult onset of clinical signs in the dachshund.
Article
Carrier screening can identify people at risk of conceiving pregnancies affected with inherited genetic disorders or who have a genetic disorder with late or variable onset. Carrier screening based on whole exome sequencing (WES) data can offer more comprehensive assessment than on-target carrier screening tests. A total of 224 Chinese adult patients WES data was analyzed, except positive variants associated with the patients' major complaint, 378 pathogenic (P) and "likely pathogenic" (LP) variants from 175 adult patients were identified. Whole exome-wide frequency of carriers for Mendelian disorders in Chinese adult patients was about 78.13% in this study, which was lower than the previously reported carrier frequency in healthy population. Contrary to expectations, the number of P or LP variants did not increase with larger chromosome size or decrease with smaller chromosome size. Totally 83 novel P or LP variants were identified which could further expand the carrier variants spectrum of the Chinese population. GJB2: NM_004004.6:c.299_300delAT:p.His100fs*14 and C6:NM_000065.4:c.654T>A:p.Cys218* were found in two or more patients, which might be two underestimated carrier variants in Chinese population. We also found 9 late-onset or atypical symptoms autosomal/X-linked dominant Mendelian disorders causative genes, which were easily overlooked during pathogenicity analysis. These results can provide a strong basis for preventing and avoiding the prevalence rates of birth defects and reducing social and family burdens. By comparing with three different expanded carrier screening gene panels, we further confirmed carrier screening based on WES could offer more comprehensive assessment and WES was applicable for carrier screening.
Article
Background: Paramyotonia congenita (PMC) is an autosomal dominant disorder characterized by cold- or exercise-induced myotonia. PMC is caused by a mutation in SCN4A which encodes the α-subunit of the skeletal muscle sodium channel. Methods: The patient was an 11-year-old Japanese girl who was diagnosed as having PMC. To confirm the diagnosis, an orbital ice-pack test and blinking tests were performed. Next, to identify the mutation, genetic analysis of SCN4A was performed. Finally, to evaluate the mutation effect on the protein structure, in silico protein modeling analysis was performed. Results: Cold- and exercise-induced myotonia was reproduced in the patient with non-invasive bedside tests: ice-pack and blinking tests. In the genetic analysis, a missense mutation, c.4343G>A in SCN4A, was identified, which may result in an arginine to histidine substitution at 1448 in the protein sequence (p.Arg1448His). According to the protein modeling analysis, the mutation neutralized the positive electrostatic charge at 1448 in the DIV/S4 segment and disrupted the beginning of the helical structure in the DIV/S3-S4 linker of the SCN4A protein. Conclusions: Diagnostic physical interventions in the patient confirmed the phenotype presentation consistent with PMC, and the in silico protein modeling analysis of p.Arg1448His predicted structural changes which can affect function of the protein. All the data confirmed the diagnosis of PMC in the patient and added to existing literature emphasizing the important role of arginine residue at 1448.
Article
The purpose of this review is to give a brief summary of the techniques of single point mutation detection by gene array analysis in a manner which will illustrate the current state-of-the-art, as well as the strengths and limitations, of the technology. This provides a framework for future development of the technology by focusing our efforts on the areas in which clinical gain is most substantial and direct.
Article
Genetic risk assessment for cardiovascular disease is less advanced and less widely performed to date than it is for cancer. Yet it is no less important. Alert clinicians should "think genetically" and follow up appropriately when confronted with a client having a family history of heart disease, early heart disease themselves, a known genetic disorder in which cardiac problems may be a component, or signs and symptoms indicative of a familial component to the heart problem observed. It is important for the clinician to know how, when, and to whom referral for further genetic evaluation and counseling should be made. Genetic testing and screening in children or adolescents for conditions such as hypertrophic cardiomyopathy (HCM), familial hypercholesterolemia (FH), and long QT (LQT) syndrome, when indicated, can help to save lives through preventive treatment and therapeutic interventions. Preparticipation sports physicals are one means of providing such screening and are important to conduct properly under guidelines recommended by the American Heart Association. Genetic testing for relatives of persons already identified to have heritable cardiac conditions is becoming more and more integral to mainstream primary health care but engender controversy when testing of children is involved. Clinicians must know how to interpret the results of such tests. Appropriate genetic counseling must accompany risk assessment, genetic testing, and screening for cardiovascular disease.
Article
Full-text available
A candidate gene for the chromosome 1 Alzheimer's disease (AD) locus was identified (STM2). The predicted amino acid sequence for STM2 is homologous to that of the recently cloned chromosome 14 AD gene (S182). A point mutation in STM2, resulting in the substitution of an isoleucine for an asparagine (N141l), was identified in affected people from Volga German AD kindreds. This N141l mutation occurs at an amino acid residue that is conserved in human S182 and in the mouse S182 homolog. The presence of missense mutations in AD subjects in two highly similar genes strongly supports the hypothesis that mutations in both are pathogenic.
Article
Full-text available
The initial observation of an expanded and unstable trinucleotide repeat in the Huntington's disease gene has now been confirmed and extended in 150 independent Huntington's disease families. HD chromosomes contained 37−86 repeat units, whereas normal chromosomes displayed 11−34 repeats. The HD repeat length was inversely correlated with the age of onset of the disorder. The HD repeat was unstable in more than 80% of meiotic transmissions showing both increases and decreases in size with the largest increases occurring in paternal transmissions. The targeting of spermatogenesis as a particular source of repeat instability is reflected in the repeat distribution of HD sperm DNA. The analysis of the length and instability of individual repeats in members of these families has profound implications for presymptomatic diagnosis.
Chapter
The low-density lipoprotein (LDL) receptor is a prototype for a class of cell surface receptors that bind lipoproteins and other circulating macromolecules and carry them into cells by receptor-mediated endocytosis. The human LDL receptor is a single-chain transmembrane glycoprotein of 839 amino acids. It specifically binds lipoproteins that contain apo B-100 and the active form of apo E. The ligand-binding domain comprises the 292 amino-terminal amino acids and is composed of a cysteine-rich sequence of 40 amino acids that is repeated seven times with minor variations. The cytoplasmic domain, composed of 50 amino acids at the carboxyl-terminal end of the protein, serves to direct the receptor to coated pits, where the bound LDL is rapidly internalized. Between these two ends of the protein there is a 400-amino-acid region that is homologous to the precursor for epidermal growth factor, a 58-amino-acid region that contains up to 18 carbohydrate chains linked to serine or threonine, and a 22-amino-acid membrane-spanning region.
Article
In Western Europe and the United States approximately 1 in 12 women develop breast cancer. A small proportion of breast cancer cases, in particular those arising at a young age, are attributable to a highly penetrant, autosomal dominant predisposition to the disease. The breast cancer susceptibility gene, BRCA2, was recently localized to chromosome 13q12-q13. Here we report the identification of a gene in which we have detected six different germline mutations in breast cancer families that are likely to be due to BRCA2. Each mutation causes serious disruption to the open reading frame of the transcriptional unit. The results indicate that this is the BRCA2 gene.
Article
DNA diagnostics was chosen last year by the US Department of Commerce as of strategic importance for US competitiveness. Thirteen companies and joint ventures received government support for technology development.
Article
Using a systematically ascertained sample of 57 families, each having 2 or more members with a consensus diagnosis of schizophrenia (DSM-III-R criteria), we have carried out linkage studies of 520 loci, covering approximately 70% of the genome for susceptibility loci for schizophrenia. A two-stage strategy based on lod score thresholds from simulation studies of our sample identified regions for further exploration. In each region, a dense map of highly informative dinucleotide repeat polymorphisms (heterozygosity greater than .70) was analyzed using dominant, recessive, and “affected only” models and nonparametric sib pair identity-by-descent methods. For one region, 8p22-p21, affected sib-pair analyses gave a P value = .0001, corresponding to a lod score approximately equal to 3.00. For 8p22–p21, the maximum two-point lod score occurred using the “affected only” recessive model (ZMAX = 2.35; θM = θF); allowing for a constant sex difference in recombination fractions found in reference pedigrees, ZMAX = 2.78 (θM/θF = 3). For a second region, 3p26–p24, the maximum two-point lod score was 2.34 (“affected only” dominant model), and the affected sib-pair P value was .01. These two regions are worthy of further exploration as potential sites of susceptibility genes for schizophrenia. © 1995 Wiley-Liss, Inc.
Article
Molecular diagnostics (MDx) is currently a clinical reality that has its roots deep in the study of gene function, structure, and regulation. The multitude of human mutations identified in the various genetic diseases can now he assayed in the clinical molecular diagnostics laboratory. The polymerase chain reaction (PCR) has facilitated the transition from the research to the clinical laboratory, however, many methods which scan and identify known mutations may not be applicable in a clinical environment. A few of these methods are discussed and one technology that is well suited for clinical use is suggested. Well-trained personnel, regulation of MDx laboratories, and automation are a few of the requirements that will carry us into the promising future of molecular diagnosis. © 1993 Wiley-Liss, Inc.