Transcript Update on Genetics of Alzheimer Disease

Insights into Basic and Clinical Neurobiology
Derived from the Analysis of Genetic causes of
Neurodegenerative Disease
P. St George-Hyslop
Centre for Research in Neurodegenerative Diseases,
Toronto Western Hospital Research Institute,
University of Toronto,
Toronto, Ontario, CANADA
Overview
• Genetics and Biology of Dementias
– Alzheimer Disease:
• APP, PS1, PS2, APOE ε4
• Other unidentified genes
– Fronto-temporal Dementia (& PSP , CBD)
• Tau
– Dementia with Lewy Bodies
• APOE ε4
• Current knowledge of known disease causing pathways;
• Application of current knowledge
– Prediction of future risks, pharmacogenomics
– Design of rational therapeutics
Emerging Concept: neurotoxic intra- or extra-cellular
deposition of insoluble proteins (-sheet conformation) is
the cause of many neurodegenerative diseases
Disease
Protein
Enabling event
Alzheimer Disease
Frontotemporal Dementia
Creutzfeldt-Jacob
Familial Encephalopathy
Familial British Dementia
Parkinson’s Disease
Aβ (βAPP)
Tau
PrPSc (PrPc)
Neuroserpin
ABri (BRI)
α-synuclein
β- /γ-secretase
?
?
?
Furin cleavage
?
What causes Alzheimer Disease?
• Genetic Factors (40% of attributable population risk):
– Mutations in genes:
•
•
•
•
•
Amyloid Precursor Protein (APP);
Presenilin 1 (PS1);
Presenilin 2 (PS2);
Apolipoprotein E (APOE ε4);
Other genes on other chromosomes.
• Environmental Factors (± genetic predispositions):
– Evidence for specific environmental factors is not robust
•
•
•
•
Lower childhood education
Head Injury
Cerebrovascular disease
?Aluminium
Genetic and “non-genetic” cases are
indistinguishable
• Genetic and non-genetic cases have identical:
– Clinical features;
– Brain pathology;
– Brain biochemistry (increased brain levels of
Amyloid β-peptide (Aβ) and tau);
– Mortality.
Genetic Determinants of
Alzheimer’s Disease
Presenile
Familial AD
Senile
Familial AD
Presenilin 1 APP Presenilin 2
gene
gene
gene
(chr 14) (chr 21) (chr 1)
age: 25–60 yrs 40–65 yrs 45–84 yrs
Sporadic
AD
APOE 4 allele
(chr 19)
>50 yrs
Other genes yet to be identified
The APP gene encodes a Type 1 membrane protien, a
fragment of which accumulates in AD brain
Citron et al. Nature Med. 3: 67-72, 1997
Aβ peptide
domain
APP
Cell
membrane
Physiological Endo-proteolytic Processing of APP
Citron et al. Nature Med. 3: 67-72, 1997
Pardossi-Piquard R et al. Neuron 46:541-554, 2005.
Aβ40 >> Aβ42
-secretase
Uptake, chaperoning, &
degradation of Aβ by
neprilysin, IDE, others
g-secretase
AICD
(?Signalling)

a
g
APP
Cell
membrane
a-secretase
Transcriptional
induction
Mutations Causing Alzheimer Disease cause
mis-processing of APP
Citron et al. Nature Med. 3: 67-72, 1997
APP
mutations
Aβ
-secretase
Uptake, chaperoning, &
degradation of Aβ by
neprilysin, IDE, others
g-secretase
AICD
(?Signalling)
Extracellular

a
TM domain
g
Intracellular
APP
a-secretase
FAD-causing mutations in APP are
localized in/around the Aβ peptide domain.
Codon Mutation
670/671 Lys-Met/
Asn-Leu
692
Ala->Gly
693
Glu->Gln
Glu->Gly
694
Asp->Asn
713
Ala->Thr
714
Thr->Ile
715
Val->Met
716
Ile->Val
717
Val->Ile/Phe
Phenotype
FAD
Effect
β-secretase cleavage
FAD
Haemorrhage
Haemorrhage
Haemorrhage
FAD
FAD
FAD
FAD
FAD
Fibrillogenesis/toxicity
Fibrillogenesis/toxicity
Fibrillogenesis/toxicity
Fibrillogenesis/toxicity
N-truncated Aβ42

Extracellular
a
N-truncated Aβ42
TM domain
g
Aβ42
Intracellular
Aβ42
APP
/Gly
723
Leu->Pro
FAD
Aβ42
FAD-causing mutations in APP are alter the
amount or the fibrillogenic potential of Aβ peptide
Codon Mutation
670/671 Lys-Met/
Asn-Leu
692
Ala->Gly
693
Glu->Gln
Glu->Gly
694
Asp->Asn
713
Ala->Thr
714
Thr->Ile
715
Val->Met
716
Ile->Val
717
Val->Ile/Phe
Phenotype
FAD
Effect
β-secretase cleavage
FAD
Haemorrhage
Haemorrhage
Haemorrhage
FAD
FAD
FAD
FAD
FAD
Fibrillogenesis/toxicity
Fibrillogenesis/toxicity
Fibrillogenesis/toxicity
Fibrillogenesis/toxicity
N-truncated Aβ42
N-truncated Aβ42
Aβ42
Aβ42
/Gly
723
Leu->Pro
FAD
Aβ42
Mutations Causing Alzheimer Disease cause
mis-processing of APP
Citron et al. Nature Med. 3: 67-72, 1997
APP
mutations
PS1/PS2
mutations
-secretase
A
Uptake, chaperoning, &
degradation of Aβ by
neprilysin, IDE, others
g-secretase
AICD
(?Signalling)
Extracellular

a
TM domain
g
Intracellular
APP
a-secretase
Naturally Occurring Mutations in Presenilins Alter APP Processing
• Predicted to encode homologous
polytopic transmembrane
proteins (PS1 and PS2).
• Contain conserved aspartate
residues in transmembrane
domains (protease active site).
• >100 missense/in-frame splicing
mutations in PS1 scattered
throughout PS1 molecule;
Cytoplasm
Membrane
Lumen
XD
XGXGD
• > 12 mutations in PS2;
• Mutations in PS1 and PS2 often
affect orthologous residues.
• PS1 and PS2 mutations all alter
Aβ production – increase Aβ42.
Sherrington et al. Nature 375: 754-760, 1995
Rogaev et al Nature 376: 775-778, 1995
Citron et al. Nature Med. 3: 67-72, 1997
Presenilin Proteins Form a Complex With Nicastrin APH-1 and PEN-2 To
Cleave Amyloid Precursor Protein (APP) and generate neurotoxic Aβ peptide.
AICD
Golgi/ER
ε-site
Cytoplasm
Membrane
Lumen/
Cell surface
D
D
Presenilin
PEN-2
APH-1
Nicastrin
g-site
Alzheimer
Disease
Sherrington, Nature, 1995
Rogaev, Nature, 1995
Katayama, Nature Cell Biol, 1999
Yu, Nature, 2000
Chen, Nature Cell Biol, 2002
Sisodia, Nature Neurosci, 2002
Pardossi-Piquard Neuron, 2005
A
_
Similar presenilin-dependent
intramembranous cleavages for:
•Notch
•Delta
•p75
•LRP1
•SorLA
•Others...
Presenilin Mutations Cause Alzheimer Disease
by altering γ-secretase cleavage of APP
Citron et al. Nature Med. 3: 67-72, 1997
APP
mutations
PS1/PS2
mutations
-secretase
A42
g-secretase-42
AICD
(?Signalling)
Extracellular

a
TM domain
g
Intracellular
APP
a-secretase
Uptake, chaperoning, &
degradation of Aβ by
neprilysin, IDE, others
Apolipoprotein E and
Alzheimer’s Disease
• APOE has 3 variants: 2, 3, 4;
• APOE 2 increased frequency in normal elderly,
reduced frequency in AD;
• APOE 4 associated with Sporadic/familial AD
(dose-dependent relationship with age of onset);
• APOE 4 association not specific to AD, and not
all APOE 4 carriers will succumb to disease.
• APOE ε4 appears to block removal of Aβ via LRP
receptors, causing accumulation of Aβ.
Mutations Causing Alzheimer Disease cause
mis-processing of APP
Citron et al. Nature Med. 3: 67-72, 1997
APP
mutations
PS1/PS2
mutations
-secretase
APOE 4
A
g-secretase
X
↓ Uptake, chaperoning, &
degradation of Aβ
AICD
(?Signalling)
Extracellular

a
TM domain
g
Intracellular
APP
a-secretase
A accumulates
A aggregates
into neurotoxic
protofibrils
What’s the evidence for this linear
pathway?
Enhancer and suppressor interactions amongst
genes causing Alzheimer Disease
Gene interactions in human patients with AD:
– APP717 mutation + APOE 4 allele = earlier onset (enhancer);
– APP717 mutation + APOE 2 allele = delayed onset (suppressor);
– PS1E280A + APOE ε4 = earlier disease (enhancer)
– PS2N141V + APOE ε4 = earlier disease (enhancer) .
Gene interactions In animal models
– APP717 mutation + PS10/0 = no disease (suppressor);
– APP717 mutation + PS1mutations = enhanced disease (enhancer).
St George-Hyslop et al Science 263:536-537, 1994
Pastor, P. et al. Ann Neurol 54, 163-9 (2003)
Suppressor
WT/WT
ε2/ε3
A717/WT A717/WT
ε2/ε3 ε4/ε3
A717/WT
ε4/ε3
APP
Elderly
+ APOEold)
ε2 carrier
V717I(>65yrs
eventually
asymptomatic
developed
carrier AD,
of APP
butV717I
at
>2
mutation
SD beyond mean age-of-onset.
A717/WT
ε3/ε3
APP genotype (A= APP717)
APOE Genotype
Enhancer and suppressor interactions amongst
genes causing Alzheimer Disease
Gene interactions in human patients with AD:
– APP717 mutation + APOE 4 allele = earlier onset (enhancer);
– APP717 mutation + APOE 2 allele = delayed onset (suppressor);
– PS1E280A + APOE ε4 = earlier disease (enhancer)
– PS2N141V + APOE ε4 = earlier disease (enhancer) .
Gene interactions In animal models
– APP717 mutation + PS10/0 = no disease (suppressor);
– APP717 mutation + PS1mutations = enhanced disease (enhancer).
St George-Hyslop et al Science 263:536-537, 1994
Pastor, P. et al. Ann Neurol 54, 163-9 (2003)
Enhancer effect of cross-breeding
mutant PS1 and mutant APP mice
APP mice – 2 months
PS1 mice - 2 months
APP x PS1 mice - 2 months
Enhancer and suppressor interactions
amongst genes causing Alzheimer Disease
• Confirms that the known AD genes really do
act in the same biochemical pathway
affecting APP processing.
St George-Hyslop et al Science 263:536-537, 1994
Pastor, P. et al. Ann Neurol 54, 163-9 (2003)
What are the other genes?
General Paradigms for Gene Discovery
LINKAGE BASED
•Difficult to collect families
•Expensive
•Relatively few assumptions
•Robust directly observable results
CASE
:
CONTROL ASSOCIATION
•Easy to collect sporadic cases
•Cheap, quick
•Easy to mess up
•Requires assumption that cases and controls
are from same founder population..
What are the other AD genes?
Case:Control
> 100 candidate genes reported to be associated with AD;
Generally had poor track-record of replication (NB: one or two
‘independent replications’ in the face of many non-replications =
non-replication);
Family linkage-based method
Confirmed localization of an AD-gene to broad region of
chromosome 10 containing several hundred genes (the specific
gene remains to be found);
Confirmed localization of an AD-gene to broad region of
chromosome 12 containing several hundred genes (the specific
gene remains to be found)
What is the role for the
microtubule associated protein Tau
and
neurofibrillary tangles?
Fronto-temporal dementia:
molecular genetics
• Mutations in Tau gene on chromosome 17q in ~10-40% of
FTD cases;
• Mutations disturb binding of tau protein to microtubules,
causing accumulation of free unbound tau;
• Free unbound tau aggregates into fibrils and these then
coalesce into paired helical filaments as the neurofibrillary
tangle;
• The tau fibrils then injure cells (but mechanism is
unclear).
Conclusions to Be Drawn From the Discovery of
Pathogenic Mutations in Tau in FTD
• Disturbed tau/microtubule homeostasis,
regardless of cause, is toxic to neurons
A accumulation initiates a biochemical
cascade leading to neuronal death
Dementia
Cause:
(eg gene
defect)
A peptide
accumulation
Neuronal
injury
Altered Tau
metabolism
Neuronal
dysfunction
and death
How is this knowledge applied for
patients?
• Adjunctive Diagnostics
• Therapeutic Targets
Prediction of future risk for AD?
•
Testing and genetic counselling feasible for:
–
–
–
Highly penetrant forms, with
Clear patterns of inheritance, and
Relatively predictable age-of-onset:
•
•
•
•
PS1
APP
Tau
Testing and genetic counselling not presently feasible/useful for:
–
Incompletely penetrant forms with variable age-of-onset:
•
•
•
–
PS2
APOE
Putative genes on chromosomes 10, 12 etc
NB: Advent of future therapies may make even fuzzy-risk data from such
genes useful
Can Genetics Predict Conversion From
MCI To AD?
•
Intuitive expectation:
– Carrier of AD risk allele with MCI would be more likely to convert to
AD.
•
Actual data available only for ApoE
•
ApoE ε4 predictive:
– Petersen et al, JAMA 274: 538,1995
– Bartrez-Faz et al, JAGS 49: 485, 2001
•
ApoE ε4 not predictive:
– Marquis et al, Arch. Neurol. 59: 601, 2002
– Tierney et al, Neurol. 46: 149, 1996.
Prediction of therapeutic response
• Theoretically reasonable;
• Remains to be validated.
Gene 1
Step 1
Gene 2
Step 2
Rx 1
Environment factor 1
Step 3
Step 4
Rx 2
AD
Using A accumulation pathways as a
target for therapies
Dementia
Cause:
(eg gene
defect)
A peptide
accumulation
Neuronal
injury
Altered Tau
metabolism
Neuronal
dysfunction
and death
Exploiting Knowledge Gained to Create New
Diagnostics and Therapeutics
•Anti-A antibodies to remove A;
•Block enzymes;
•Block aggregation.
Cause:
(eg gene
defect)
X
A peptide
accumulation
Dementia
Neuronal
injury
Neuronal
dysfunction
and death
Altered Tau
metabolism
Janus et al Nature. 408: 979-982, 2000,
McLaurin et al, Nature submitted, 2004
How can the amyloid cascade be blocked?
Citron et al. Nature Med. 3: 67-72, 1997
Pharma:
Pharma:
A
-secretase
Vaccine:
toxic
g-secretase
Uptake, chaperone, or
degradation (by neprilysin).
AICD
(?Signalling)

a
Pharma
g
X
APP
Cell
membrane
A accumulates
A aggregates
into neurotoxic
protofibrils
Conclusions:
• All known genes causing AD modulate APP
and Aβ processing;
• Neurodegeneration from mutations in tau
prove that tau accumulation is also a toxic
event (regardless of whether caused by
mutation in tau or due to Aβ accumulation)
• Knowledge of pathway will provide targets
for disease-modifying therapies.
Acknowledgements
S. Arawaka
F. Chen
L. Farrer,
P. Fraser
YJ. Gu
H. Hasegawa
M. Ikeda
T. Katayama
T. Kawarai
G. Levesque
M. Nishimura
A. Petit
E. Rogaeva
N. Sanjo
P. St George-Hyslop
D. Westaway
Canadian Institutes of Health Research
Howard Hughes Medical Institute
Alzheimer Society of Ontario,
Canadian Genetic Diseases Network
A. Bruni,
F. Checler
JF Foncin,
G. Marcon,
M. Mortilla,
A. Orlacchio,
E. Paitel
S. Piacentini,
L. Pinessi,
I. Rainero,
S. Sorbi,
R. Tupler,
G. Vaula
CONTACT INFORMATION
• Analysis of familial cases:
P. St George-Hyslop, University of Toronto
tel: 416-978-7460
[email protected]
• Animal models (transgenic mice etc):
David Westaway
[email protected]
• Reagents (clones, cell lines, antibodies, etc)
P. St George-Hyslop, University of Toronto
[email protected]