Transcript mol. medicine 2

Molecular medicine - 2
Example of identifying a monogenic
condition by positional cloning
cystic fibrosis caused by mutations in the
CF gene
Most common severe autosomal recessive condition
among Caucasians.
About 5% of white Caucasians of European descent
are asymptomatic carriers.
Frequency of 1 / 2,500 affecting approximately
30,000 people
In 1985, CF locus was localized on the long arm of
chromosome 7
In 1989, the gene implicated in CF was isolated
(Kerem 1989; Riordan 1989; Rommens 1989).
Pathology
Woe to that child which
when kissed on the
forehead tastes salty. He
is bewitched and soon
must die
CF from gene to product
CF encodes a Cl- channel
(is caused by defects in
the CF gene which results
in either a decrease in its
Cl- transport capacity or
its level of cell surface
expression
CF gene encodes a cystic fibrosis
transmembrane conductance regulator
The genetic analysis showed that this gene,
which is responsible for this disorder, contains
24 exons spreading over 250 kb of
chromosome 7 (7q31) and encodes an mRNA
of 6.5 kb.
CFTR – cystic fibrosis
transmembrane
conductance regulator
member of ATP binding cassette (ABC) membrane transporter
superfamily
2 homologous halves -1480 amino acids long
each half has 6 transmembrane domains (M1-12) &
1 nucleotide binding domain (NBD) which are linked by
a cytoplasmic regulatory domain (R-domain) that contains
phosphorylation sites
 epithelial Cltransport Cltransport rate
determined by
activation of CFTR
which in turn
depends on its state
of phosphorylation.
 Acts as a regulator
of other channels &
transporters e.g
CFTR mediates
cAMP regulation of
amiloride sensitive
epithelial Na+
channels (EnaCs)
CFTR function
http://www.infobiogen.fr/services/chromcancer/IntroItems/Images/CFTREnglFig2.jpg
CFTR channel
Minimum channel diameter
– 5.3A
Maximum channel diameter
- 10-13A
Charge selectivity:
R352, cytoplasmic end of M6
Overall structure:
Channel with a large
extracellular vestibule which
narrows towards the
cytoplasmic end where the
anion selectivity filter is
located.
Channel lining is formed by M1,
M3, M6 & M12 segments.
J Biol Chem (2000) vol 275 No 6 pp 3729 by MH Akabas
Regulation of CFTR gating
2 processes control Cl- movement
phosphorylation: necessary to activate the channel. The R domain
contains phosphorylation sites for cAMP-dependent protein kinase
A (PKA), C (PKC) and type II cGMP dependent protein kinases.
CFTR deactivation mediated by phosphatases PP2C & PP2A.
ATP binding & hydrolysis:
Opening / closing of channel controlled by ATP binding & hydrolysis
which occurs in the NBD segment.
The R domain interacts with NBD & regulates their ATP affinity.
Spectrum of CF mutations that affect function
F508
70% of CF patients show a specific
deletion F508
single amino acid (F) deletion in exon
10 which codes for first portion of
NBD-1 of the CFTR protein.
This leads to misfolding of CFTR in
the endoplasmic reticulum(ER).
These immature CFTR proteins are
then polyubiquinated & targeted for
proteosome degradation
Mapping of CFTR
1985 gene for CF linked to enzyme paraoxanase (PON)
PON mapped to chromosome 7 and CF mapped to 7q31-32 (random
DNA marker D7S15)
2 flanking markers established (~2x106bp apart)
proximal MET oncogene and distal D7S8
extensive mapping and characterisation around the candidate region
by chromosome walking, chromosome jumping and microdissection
(~300kbp cloned)
CFTR candidate region
Mapping of CFTR
2 new markers identified – KM19 and XV2c – which showed
strong linkage disequilibrium
5’ end of gene located
Bovine equivalent of candidate gene isolated from genomic
library
7 cDNA libraries screened with human clone. 1 cDNA clone
identified. Northern blots show 6.5 kb mRNA
Rest of the gene obtained by screening and PCR
1989 CFTR gene eventually isolated by mutation screening
Letter to Dr. Collins. Courtesy of the
National Human Genome Research Institute
The spectrum of human diseases
Cystic fibrosis
thalassemia
<5%
cancer
Huntington’s
‘Mendelian’ diseases (<5%)
Autosomal dominant inheritance:
e.g huntington’s disease
Autosomal codominant inheritance
e.g Hb-S sickle cell disease
Autosomal recessive inheritance:
e.g cystic fibrosis, a & b thalassemias
X-linked inheritance:
e.g Duchenne muscular dystrophy (DMD)
Mapping complex loci
PAF – population attributable factor:
Fraction of the disease that would be eliminated if the risk
factor were removed
High PAF for single gene conditions (>50% for CF)
Low PAF for complex disease (< 5% for Alzheimer’s)
Identifying genes involved in complex
diseases
Steps
Perform family, twin or adoption studies
- check for genetic component
Segregation analysis
- estimate type and frequency of susceptibility alleles
Linkage analysis
- map susceptibility loci
Population association
- identify candidate region
Identify DNA sequence variants conferring
susceptibility
Linkage versus Assocciation
Association studies compare the
allele frequency of a
polymorphic marker, or a set of
markers, in unrelated patients
(cases) and healthy controls to
identify markers that differ
significantly between the two
groups.
Used to identify common modestrisk disease variants
Higher density of markers needed
e.g. HapMap uses association
data
Linkage analyses search for
regions of the genome with a
higher-than-expected number of
shared alleles among affected
individuals within a family.
Used to identify rare high-risk
disease alleles
<500 markers needed for initial
genome scan
Haplotype analysis
• specific combination of 2 or more DNA marker
alleles situated close together on the same
chromosome (cis markers)
• SNPs most commonly used markers in haplotypes.
• series of closely linked mutations accumulate
over time in the surviving generation derived
from a common ancestor.
• powerful genetic tool for identifying ancient
genetic relationships.
• Alleles at separate loci that are associated with
each other at a frequency that is significantly
higher than that expected by chance, are said to
be in linkage disequilibrium
Direct versus indirect association analysis.
a, In direct association analysis,all functional variants (red arrows) are catalogued and tested for association with disease. A
GeneSNPs image of the CSF2 gene is shown. Genomic features are shown as boxes along the horizontal axis (for example,
blue boxes indicate exons). Polymorphisms are shown as vertical bars below the axis, with the length of the line indicating allele
frequency and colour indicating context (for example, red indicates coding SNPs that change amino acids).
b, For indirect association analysis, all common SNPs are tested for function by assaying a subset of tagSNPs in each gene
(yellow arrows), such that all unassayed SNPs (green arrows) are correlated with one or more tagSNPs. Effects at unassayed
SNPs (green arrows) would be detected through linkage disequilibrium with tagSNPs. Images adapted from GeneSNPs
(http://www.genome.utah.edu/genesnps).
Formation of haplotypes over time
Ancient disease loci are
associated with haplotypes
• Start with population genetically isolated for a long
time such as Icelanders or Amish
• Collect DNA samples from subgroup with disease
• Also collect from equal number of people without
disease
• Genotype each individual in subgroups for
haplotypes throughout entire genome
• Look for association between haplotype and disease
phenotype
• Association represents linkage disequilibrium
• If successful, provides high resolution to narrow
parts of chromosomes
Haplotype analysis provides high
resolution gene mapping
Why is it still so difficult?
Genetic heterogeneity
Mutations at more than one locus cause same
phenotype
e.g. thalassemias
– Caused by mutations in
either the a or b-globin
genes.
– Linkage analysis studies
therefore always
combine data from
multiple families
Variable expressivity - Expression of a mutant
trait differs from person to person
• Phenocopy
– Disease phenotype is not caused by any
inherited predisposing mutation
– e.g. BRCA1 mutations
• 33% of women who do not carry BRCA1 mutation
develop breast cancer by age 55
Incomplete penetrance
– when a mutant genotype does not
always cause a mutant phenotype
• No environmental factor associated with
likelihood of breast cancer
• Positional cloning identified BRCA1 as one
gene causing breast cancer.
– Only 66% of women who carry BRCA1 mutation
develop breast cancer by age 55
• Incomplete penetrance hampers linkage
mapping and positional cloning
– Solution – exclude all nondisease individuals form
analysis
– Requires many more families for study
• Polygenic inheritance
– Two or more genes interact in the expression of
phenotype
• QTLs, or quantitative trait loci
– Unlimited number of transmission patterns for QTLs
» Discrete traits – penetrance may increase with number
of mutant loci
» Expressivity may vary with number of loci
– Many other factors complicate analysis
» Some mutant genes may have large effect
» Mutations at some loci may be recessive while others
are dominant or codominant
Polygenic inheritance
E.g heart attacks or cholesterol levels
Sudden cardiac death (SCD)
Breast cancer
Common condition – familial or sporadic forms
Although a genetic basis for familial BC identified, the
causes of sporadic disease still unknown
Mutations in 2 loci account for 20-25% of early onset
(<45 years) breast cancer cases due to inherited
factors
– BRCA1: mutations found in 80-90% of families
with both breast and ovarian cancer
– BRCA2: mutations mainly in male breast cancer
familiesSudden cardiac death (SCD)
Alzheimer’s disease
Affects 5% of people >65 years and 20% of people over 80
has familial (early-onset) or sporadic (late-onset) forms, although
pathologically both are similar
Aetiology of sporadic forms unknown
familial AD – mutations in APP, presenilin-1 and 2
Sporadic AD – strong association with APOe4, Apolipoprotein e4,
which affects age of onset rather than susceptibility
3 major alleles (APO E2, E3, and E4)
Position
112
158
ApoE2
Cys
Cys
ApoE3
Arg
Cys
Sudden
cardiac
death
(SCD)
ApoE4
Arg
Arg
Epigenetics – differential imprinting
Prader-Willi syndrome
failure to thrive during infancy,
hyperphagia and obesity during
early childhood, mental retardation,
and behavioural problems
molecular defect involves a ~2 Mb
imprinted domain at 15q11–q13 that
contains both paternally and
maternally expressed genes
Angelman syndrome
births and characteristics include
mental retardation, speech
impairment and behavioural
abnormalities
AS defect lies within the imprinted
domain at 15q11–q13
Genetic causes
Prader-Willi syndrome
Angelman syndrome
70% have a deletion of the
PWS/AS region on their
paternal chromosome 15
70% have a deletion of the PWS/AS
region on their maternal
chromosome 15
25% have maternal
uniparental disomy for
chromosome 15 (the
individual inherited both
chromosomes from the
mother, and none from the
father)
7% have paternal uniparental
disomy for chromosome 15 (the
individual inherited both
chromosomes from the father, and
none from the mother)
5% have an imprinting
defect
<1% have a chromosome
abnormality including the
PWS/AS region
3% have an imprinting defect
11% have a mutation in UBE3A
1% have a chromosome
rearrangement
11% have a unknown genetic cause
Reading
HMG3 by T Strachan & AP Read : Chapter 15
AND/OR
Genetics by Hartwell (2e) chapter 11
References on Cystic fibrosis:
Science (1989) vol 245 pg 1059 by JM Rommens et al (CF mapping)
J. Biol Chem (2000) vol 275 No 6 pp 3729 by MH Akabas (CFTR)
Optional Reading on Molecular medicine
Nature (May2004) Vol 429 Insight series
•
human genomics and medicine pp439 (editorial)
•
Mapping complex disease loci in whole genome
studies by CS Carlson et al pp446-452