Transcript DNA and Gene Expression - Department of Psychology

Cognitive Disabilities
General Cognitive Disability
• Behaviourally diagnosed
• IQ
– Average = 100, standard deviation = 15
– Only 5% of population has cognitive disabilities
• Levels
–
–
–
–
Mild (IQ 50-70); about 85% of cases
Moderate (IQ 35-50)
Severe (IQ 20-35)
Profound (IQ below 20)
General Cognitive Disabilities
• Genetics relatively well studied
– Especially for serious (IQ<50) cognitive
disabilities
– Can now investigate specific disabilities at gene
level
• Mild disabilities (candidates for QTLs) are
harder
Quantitative Genetics
• Sibs of mildly cognitive disabled have
below average IQ scores
• Sibs of severely cognitively disabled do not
– Why?
Mild Disabilities: Genetic or
Environmental?
• Genetics seems to be important
– Concordances for mild cognitive disabilities:
MZ 75%, DZ 46%
• High/low IQ equally heritable across age
(infancy to middle age)
– But few scores in these studies go below IQ 70
• ~50% mildly disabled children also have
behavioural problems
– Quantitative range of normal distribution?
Single Gene Disorders
• Over 250 known disorders that show low IQ
(also other, often primary, symptoms)
• Generally quite rare
PKU
• Single gene autosomal recessive
– Defect in PAH gene (chr. 12) for enzyme phenylalanine
hydroxylase
– Converts amino acid phenylalanine to other essential
compounds
– Over 400 disease causing mutations of PAH
• About 1 in 10,000-15,000 births
• Screening of infants at birth since 1961
• Environmental fix; restrict access to proteins (and
some other foods) and supplement amino acids
lacking from no protein diet
Neurofibromatosis Type 1
• Member of the neurocutaneous syndromes,
all producing neurologic and dermatologic
lesions
• Autosomal dominant disorder from
mutation of gene NF1 (chr. 17)
– NF1 encodes neurofibromin protein, involved
in intracellular signaling
• Cognitive and learning disabilities also
common
X-Linked
• Genes on X chromosome
• Tend to effect males
• Females: X-inactivation --> mosaicism
– Females with disorder, often show highly
variable severity depending on how
many/which cells affected
X-inactivation
• One X copy “silenced” in early blastocyst
– Packaged into transcriptionally inactive
heterochromatin
– Prevents twice as many X chromosome gene
products being produced as in the male
• In placental mammals, which X
chromosome inactivated is random
• Default state is inactivation for X chr.
– In XXX individuals, still only one active X
• Autosomally-encoded “blocking factor”
hypothesized; binds to one X and prevents
its silencing
• X-inactivation centre (XIC) necessary and
sufficient for silencing
– Translocating XIC to autosomal chromosome
causes inactivation of that autosome; X lacking
XIC not inactivated
Fragile X Syndrome
• About 1 in 3600 males, 1 in 4000-6000 females
• Mutation of the FMR1 gene on the X chr.
– Normal FMR1: 6 to 55 repeats of CGG codon.
– Fragile X: 230+ repeats
• Expansion of CGG --> methylation of that portion
of DNA --> silences expression of FMR1 protein
• Methylation constricts the chromosome; “fragile”
appearance under the microscope
Fragile X
• Males have one copy of X chromosome; those
with expanded FMR1 are symptomatic
– Cognitive disabilities and physical features (elongated
face, large ears, low muscle tone)
• Females’ two Xs doubles chances of having
functioning FMR1 allele
• With X-inactivation females with 1 expanded
FMR1 allele can have some symptoms
Rett Syndrome
• Deceleration of growth of head, hands, and feet,
cognitive impairment, GI disorders, serious verbal
deficits
• Sporadic (de novo) mutations to gene MECP2 on
X chromosome
• Almost always affects females; male fetuses rarely
survive to term or die at very early age
– With one functioning gene, female fetuses can produce
enough proteins to survive to birth, but symptoms
manifest between 6-18 months post-birth
Duchenne Muscular Dystrophy
• Mutation of gene on X chromosome that codes for
dystrophin protein
– Necessary structural component of muscle tissue
• Progressive muscle weakness and loss of muscle
mass; some cognitive disfunction
• Affects males
– Females can be carriers; rarely manifest symptoms
• Fatal by age 30
Lesch-Nyhan Syndrome
• X-linked recessive; rare, 1 in 380,000 live births
• Deficiency of enzyme hypoxanthine-guanine
phosphoribosyltransferase (HGPRT)
– Causes build up of uric acid in all body fluids
• Severe mental and physical deficits; selfmutilating behaviour common
• Female carriers typically asymptomatic; may have
some difficulties from increased uric acid
excretion
Chromosomal Abnormalities
• More common
• Generally moderate to severe disability
– 7% of children with unexplained moderate or severe
disabilities have detectable deletions vs. 0.5% of
children with mild disabilities
• Additional chromosomes
– More frequent; easily detectable…
• Deletions of parts of chromosomes
– Use of microarrays identifying more of these
– Might ultimately be more common
Williams Syndrome
• Small deletion (chr. 7)
– Affects about 20 genes; elastin production, LIM
kinase, and others
• Usually spontaneous mutations
• Connective tissue defects, multiple medical
problems, generally moderate cognitive
disfunction; some similarities to autism
Angelman Syndrome
• Genetic imprinting; inherited from mother
• Deletion of genes at 15q11-13
– E.g., defect/deletion in gene UBE3A, coding for an
ubiquitin ligase; absence results in defects in
hippocampus and cerebellum
• Usually novel in formation of gametes
• Risk to sibs of proband is ~1%
• Moderate cognitive disfunction, motor and speech
impairments, seizures, inappropriate happy
demenor
Prader-Willi Syndrome
• Genetic imprinting; from father
• Deletion or defect of up to 7 genes at
15q11-13
– Effect is on some non-coding RNA; modifies
other non-coding RNAs and mRNA processing
• Low IQ, compulsive overeating
Down Syndrome
• Most common cause of cognitive disability
(about 1/1000 births)
• Trisomy 21
• Over 300 symptoms; physical, behavioural,
and cognitive
• Mean IQ ~55; dementia commonly sets in
by age 45
Sex Chromosome Abnormalities
• XXY male
– 1/750; low testosterone in
adolescence
– Small testes, enlarged breast
tissue, infertility, low IQ, poor
speech and language
• XXX
– 1/1000 females
– Mean IQ ~85, poor on verbal
tasks
• XYY
– 1/1000 males
– Speech deficits, language and
reading difficulties; tendency
to be tall
• X0
– 1/2500 females
– 99% miscarry; accounts for
10% of spontaneous abortions
– Short stature, abnormal sexual
development, verbal IQ normal
but performance IQ ~90 after
adolescence
XYY Males
• 1961: first described case, karyotype of man whose son had Down
Syndrome
• 1965: 7 of 197 inmates in a Scottish maximum security prison are
XYY; unusually tall and aggressive. “Super-male Syndrome”
• 1968: Murderer of Parisian prostitutes found to be XYY
• 1968: Richard Speck (murdered 8 nursing students in Chicago)
reported to be XYY; subsequently rejected
• 1968: 23 studies published examining rates of XYY in prisons, 0
studying XYY in the general population
• 1969: study screens inmates for XYY (but only karyotype tall, violent
ones); finds 1 of the 9 tested is XYY
• 1969: first good study on base rates of chromosomal abnormalities
conduced (about 1/1000)
• 1970: Newsweek article on XYY “Congenital Criminals”, Y becomes
the “Crime Chromosome”
Dale & Harley (1980)
Mental Institute Penal Institute Mental-Penal
Men karyotyped
5100
6983
5479
XYY
13
31
104
XYY/1000
2.5
4.4
20.0
Nanko et al. (1979)
Sample Size
Number of Studies
Karyotypes
XYYs found
XYY/1000
<100 101-200
6
9
326
1287
4
10
12.3
7.8
201-500
3
1199
6
5.0
>500
5
2838
7
2.4
Guesstimate
• If 1/1000 males is XYY and 5/1000 inmates
is XYY, then what proportion of XYY
males are in jail?
• Rough guess = 0.5%
– Based on knowing proportion of population
which is incarcerated in Western countries
Why?
• Some hypotheses:
– 1. They are violent
– 2. They are tall
– 3. They have cognitive disabilities (lower
intelligence)
Violent
• No personality or crime differences between XYY
and XY inmates (Clark et al., 1970)
• XYY actually seem to be less violent than XY
inmates
– More likely incarcerated for property than violent crime
• Also, XXY are disproportionately incarcerated,
and their crime pattern is like that of XYY (Hook,
1979)
Tall
• Tall XY are not jailed disproportionately to
other XY (Hook & Kim, 1971)
Lower Intelligence
•
•
•
•
•
31,436 men born in Copenhagen in 1944-47
IQ tested during mandatory military service
Studied all tall (>184 cm) men
Karyotyped 4,139 of them
Used criminal history database and school
exam records
Findings
• Karyotype finds 12 XYY and 16 XXY
• Base crime conviction rate of 9.3%
• 5 XYY had records (non-violent offences), 3 XXY
had records
• Intelligence test scores were low for XYY and
XXY; convicted XYY and XXY even lower
• XY males with records also had lower IQ test
scores
Inference
• Cognitive deficits are important
• XYY not necessarily any more violent or
dangerous
• Cognitive deficits limit employment options
• May be “poor” criminals - more likely to
get caught, miss plea bargain opportunities,
etc., possibly due to lower IQ
Quantitative Genetics
• Mild cognitive disability likely quantitative,
not qualitative
• Polygenic
• QTL hypothesis
– Many genes making small contributions
• Many disorders, then, may simply be the
“low end” of the normal distribution of the
population’s traits
DF Extremes Analysis
• Method for estimating heritability of
disorders that are defined qualitatively (i.e.,
either have or don’t have the disorder)
• But disorders depend on an underlying trait
that varies quantitatively along a continuum
DF Extremes Analysis
• Developed by DeFries and Fulker
• A regression-based method for analyzing
twin data
• Designed for proband-selected data
– At least one twin has an extreme score
• Based on regression to the mean
Regression to the Mean
• 1000 individuals tested on some measure
Regression to the Mean
• Test same 1000 individuals one month later;
gives a similar profile
Regression to the Mean
• Focus on people who scored very low at
time 1 (extreme group)
• Can we predict how this group should score
at time 2?
Regression to the Mean
• Mean at time 2 for extreme group from time 1
depends on the correlation between the two time
points
• Imagine correlations of +1, 0, and something in
intermediate
Regression to the Mean
Time 1
Correlation = +1
Time 2
Time 1
Correlation = 0
Time 2
Time 1
Intermediate correlation
Time 2
Pop. mean
DF Extremes Analysis
• Considers the differential regression
towards the mean between MZ and DZ
twins
• To parallel the example, time 1 is the
proband score and time 2 is the co-twin
score
Implementation
• E.g., select individuals in low 5% of a trait
• Look at the scores of their co-twins
• If a trait is influenced by genes, MZ cotwins should not regress towards the mean
as much as DZ co-twins
• Due to higher correlation between MZ
twins
Visualization
Probands
MZ co-twins
DZ co-twins
Proband MZ
mean co-twin
mean
DZ
Pop.
co-twin mean
mean
Heritability and Non-shared
Environment
• Regression to the mean reflects info about the
correlation between twins.
• Standardize means so population mean becomes 0
and the proband mean is 1
• Then transformed MZ and DZ co-twin means can
be implemented just as MZ and DZ correlations
are.
• 2 x diff. b/t MZ & DZ co-twin mean estimates
heritability and 1 minus the MZ co-twin mean
estimates non-shared environment
Graphically
Proband
mean
MZ co-twin
mean
e2
1
DZ co-twin
mean
h2/2
Population
mean
h2/2 + c2
0
e = unique environment
c = shared environment
Learning Disorders
• Dyslexia
– ~10% of children have difficulty learning to read
• 80% of diagnosed learning disorders have difficulty
reading
• Also, math disorder (moderate heritability from
twin studies) co-occurs with reading disorder
• Sibs and parents of reading-disabled probands do
worse on tests of reading ability
• Study of 250 twins with one reading disabled: 66%
concordance for MZ, 36% for DZ
Verbal Ability is Heritable
1.0
0.8
Unique envir.
0.6
Shared envir.
0.4
Heredity
0.2
0.0
Early
Middle Adolescence Adulthood
childhood childhood
Old
Age
Meta-Analysis of recent twin studies (Adapted from: Price (2002))
Specific Language Impairment
(SLI)
1.0
0.8
MZ twins
Risk 0.6
DZ twins
0.4
1st degree relatives
0.2
Controls
0.0
Relations
Data from Stromswold (1998) and Stromswold (2001) (Figure adapted from Price (2002))
Vernes et al. (2008)
• Rare mutations of FOXP2 transcription factor
cause some forms of SLI
• Genome screened for regions bound by FOXP2
• Tested for SNPs in set of 184 families with SLI
• FOXP2 binds to and dramatically down-regulates
CNTNAP2 gene
– One of largest genes in human genome (~1.5% of chr.
7)
– Role in cell adhesion molecules in developing human
cortex and axon differentiation
• Significant quantitative association with nonsenseword repetition
Reading Disability
• Causes of individual variation throughout the
population may be different from the causes of
differences between a group with extreme scores
and the rest of the population
• Genetic contribution to variation in reading ability
does not mean illiterates are genetically different
from the rest of the population
• MZ twins correlate 0.9, DZ twins correlate 0.65
• h2 ~ 0.5 (group heritability)
QTLs for Reading Disability
• Can reject various single gene hypotheses
– E.g., autosomal dominant, X-linked
• Sib-paired QTL linkage analysis
– Sib of reading disabled had lower reading
ability when the two shared the same version of
6p21
QTL Linkage Map for 6p21
Index of
statistical
significance
DNA markers
D6S105 sig. at p = 0.05 for
siblings and 0.01 for DZ twins
Developmental Dyslexia
• Galaburda et al. (2006)
• Developmental dyslexia: severe and specific
difficulty in reading acquisition unrelated to other
cognitive abilities and education circumstances
– Phonological deficits with mental representations and
processing of speech sounds
• Ectopias, nests of neurons, in cortical layer 1 and
focal microgyria affecting language areas
– Neural migration and axon growth, especially in first
year of life
Candidate Genes
• Long history of familial occurrence and twin
studies
• DYX1C1, KIAA0319, DCDC2 and ROBO1 are
dyslexia candidate susceptibility genes
• Proteins from these genes diverse; may be
functionally linked either directly or by similarity
to other proteins involved in neuronal migration
and axon growth
– Coordinating changes in cell adhesion and cytoskeletal
restructuring
Protein Function
• ROBO1 and KIAA0319
– Proteins for transmembrane adhesion molecules and
receptors guiding axons to proper targets
– In vitro assays of KIAA0319 allele linked to dyslexia
show 40% decrease in expression; in vivo effects not
yet known
• DCDC2 and DYX1C1
– Act as downstream targets that modulate changes in
cytoskeletal processes involved in motility of
developing neurons
Communication Disorders
• DSM-IV: four types
–
–
–
–
Expressive language disorder
Mixed receptive and expressive disorder
Phonological disorder
Stuttering
• 25% of 1st degree relatives report similar
disorders
• MZ concordance 90%, DZ 50%; high heritability
at 2 years
KE Pedigree
• 3 generation pedigree of family with severe
speech and language disorder
• Autosomal-dominant monogenic trait
• Linkage pedigree points to responsible
locus, SPCH1, on 7q31 and FOXP2 gene in
particular
– FOXP2 encodes a transcription factor containing a
polyglutamine tract
– Disrupted by point mutation in affected members of KE
family
KE Pedigree
I
Adapted from Lai et al. (2001)
II
III
Affected male
Affected female
Unaffected male
Unaffected female
Deceased