вход по аккаунту


Apolipoprotein E genotypes and age of onset in early-onset familial Alzheimer's disease.

код для вставкиСкачать
10. Lisak RP, Bealmear B. Antibodies to interleukin-1 inhibit cytokine-induced proliferation of neonatal rat Schwann cells in vitro.
J Neuroimmunol 1791;31: 123--132
11. Lisak RP, Bealmear B. Antibodies to interleukin-6 inhibit
Schwann cell proliferation induced by unfractionated cytokines.
J Neuroimmunol 1974;50:127-132
12. Scarpini E, Lisak RP, Beretta S, et al. Quantitative assessment
of class 11 molecules in normal and pathological nerves. Brain
. Apolipoprotein b
Genotypes and Age of
Onset in Early-Onset
Familial Alzheimer’s
Ephrat Levy-Lahad, MD,* Amnon Lahad, MD,I
Ellen M. Wijsman, PhD,*$ Thomas D. Bird, MD,*lI
and Gerard D. Schellenberg, PbD“
The effect of the apolipoprotein E (APOE) ~4 allele on
age of onset was analyzed i n two groups of families with
early-onset Alzheimer’s disease (AD), (1) Volga German
(VG) kindreds, in which AD is caused by an unknown
locus and (2) early-onset non-VG families showing evidence of linkage to chromosome 14. The ~4 allele did
not have a detectable effect on age of onset in either
group. This finding suggests some forms of early-onset
familial AD are not modifiable by APOE status, or that
AD can be caused by APOE-independent pathways.
Levy-Lahad E, Lahad A, Wijsman EM, Bird T D ,
Schellenberg G D . Apolipoprotein E genotypes and
age of onset in early-onset familial Alzheimer’s
disease. Ann Neurol 1995;38:678-680
Alzheimer’s disease (AD) is genetically heterogeneous.
There are two known causative loci for early-onset
(age, <60 years) autosomal dominant familial AD
(FAD), the amyloid precursor protein (APP) gene on
chromosome 21 [l, 21 and an unidentified gene on
chromosome 14 13). A third locus, the apolipoprotein
E (APOE) gene on chromosome 19, confers genetic
From the Divisions of “Medical Genetics, rGerontology and Geriatric Medicine, and Department of Medicine; $Departmentof Biostatistics; and tSchool of Public Health, University of Washington, and
“Department of Neurology and ‘Geriatric Research Education and
Clinical Center, Seattle Veterans Affairs Medical Center, Seattle,
Received Nov 28, 1974, and in revised form Apr 24 and Jun 16,
1995. Accepted for publication Jun 20, 1995.
Address correspondence to Dr Schellenberg, Geriatric Research Education and Clinical Center (182b), Seattle Veterans Affairs Medical
Center, 1660 South Columbian Way, Seattle, WA 98108-1577.
risk. Of the three alleles of the APOE gene ( ~ 2 ~, 3 ,
and ~ 4 )the
, ~4 allele is a risk factor for late-onset
(age, >60 years) FAD [4-6} and for both early- and
late-onset “sporadic” cases [S-8). It appears to act as
a dose-dependent age-of-onset modifier; each copy of
the ~4 allele may lower age of onset by as much as 7
to 9 years [ 4 , 9 , 107. Recent evidence suggests that the
~4 allele may also modify age of onset in subjects with
APP mutations { l l , 12). In one analysis of early-onset
FAD families linked to the chromosome 14 locus,
APOE genotypes did not appear to modify age of onset
113). However, most cases analyzed were from two
families with an extremely early age of onset (family
means, 33 and 34 years) [141, which could suggest
that the lack of effect observed is related to a unique
mutation at the chromosome 14 locus.
To determine whether the ~4 allele modifies age of
onset in other forms of heritable AD, we examined
two groups of well-characterized families with genetically distinct, autosomal dominant early-onset FAD [3,
Subjects and Methods
The Volga German (VG) kindreds (family mean ages of onset, 50-65 years) are eight families that do not have APP
mutations and in which linkage analyses exclude the chromosome 14 locus and the APOE region of chromosome 19 as
the loci responsible for A D 13, 6, 161. A D in the V G is
probably the result of a common genetic founder 1151.
The early-onset non-VG (ENVG) kindreds (family mean
ages of onset, 42-52 years) are 12 early-onset kindreds in
which A D is caused by the chromosome 14 locus in most,
if not all, families 13, 61.
APOE genotypes and allele frequencies (Table 1 ) [6] were
determined as previously described 2171.
Statistical Analysis
Analysis of mean age of onset versus presence or absence of
~4 was done by t test. Mean age of onset versus €4dose was
tested by multiple regression controlled for family of origin
(VG or ENVG). Kaplan-Meier survival analysis curves for
age of onset given presence or absence of ~4 were constructed for both V G and ENVG groups. Onset distributions
were based on known age of onset in affected individuals
and age at last examination in unaffected blood relatives.
Related individuals are not statistically independent for allele frequencies or age of onset. Calculating means in related
subjects gives unbiased estimation of the parameter mean
but underestimates standard error (SE) 16, 181, which could
result in a type I error. However, since no difference was
found in our analysis (see Results), other methods for estimating SE were not necessary.
To circumvent the problem of related individuals being
nonindependent, we also randomly selected from each family
one pair consisting of 1 case with ~4and 1 case without ~4
(using a random digits table). These pairs are truly indepen-
Copyright 0 1995 by the American Neurological Association
Table 1. Apolipoprotein E Genotypes and Allele Frequencies
ENVG Disease Status
VG Disease Status
Total (n)
0 (0%)
0 (0)
2 (2.5%)
14 (47%)
0 (0%)
3 (8.5%)
23 (66%)
0 (0%)
9 (25.5%)
0 (0%)
48 (58%)
2 (2.5%)
27 (32%)
4 (5%)
0 (0%)
9 (45%)
8 (40%)
0 (0%)
9 (8%)
81 (71%)
0 (0%)
21 (18.5%)
3 (2.5%)
0 (0%)
2 (8%)
18 (72%)
0 (0%)
4 (16%)
0.017 (0.017)
0.65 (0.062)
0.33 (0.061)
0.024 (0.012)
0.75 (0.033)
0.22 (0.032)
0.050 (0.034)
0.68 (0.074)
0.28 (0.071)
0.043 (0.024)
0.829 (0.045)
0.129 (0.040)
0.040 (0.013)
0.842 (0.024)
0.118 (0.021)
0.040 (0.028)
0.840 (0.052)
0.120 (0.046)
0 (0%)
10 (33%)
5 (17%)
Allele frequencies (SD)
a = affected; r
at risk (blood relatives of affected subjects); s
dent of each other and were compared by paired t test. Since
these are same-family pairs, ENVG and VG pairs were analyzed together. Selecting a single pair per family obviously
results in loss of power.
In the VG group, ~4 dose did not have a significant
effect on age of onset (Table 2). Also, for presence
versus absence of ~ 4mean
age of onset in cases with
at least one ~4 allele was slightly lower than in cases
with no e4 allele, but the difference was not significant.
In the ENVG group, mean age of onset of cases with
no ~4 alleles was actually younger than that of cases
with one e4 allele (see Table 2), and no cases were
homozygous for ~ 4The
. e4 allele also had no significant effect on age of onset using multiple regression
controlled for family origin (ENVG vs VG), both for
analysis by presence or absence of ~4 alleles and by
gene dosage of ~ 4 In
. both groups, e4 status did not
change age at onset Kaplan-Meier survival curves
(data not shown).
In combined random pair analysis, mean age of onset
of cases with no ~4 allele was 54.1 years (SE, ? 13.0;
n = lo), younger than cases with at least one ~4 allele,
58.7 years (SE, k18.9; n = 10). Eight famiiies could
not be used for this analysis; in six families (two VG,
four ENVG) there were no affected subjects with e4
alleles, in one ENVG family both affected individuals
Table 2. Mean Age of Onset and Apolipoprotein E Genotype
Mean Age of Onset
‘i SD (n)
APOE ~4 Dose
62.2 t 10.0 ( 1 5 )
58.7 ? 11.0 (10)
60.6 t 6.0 ( 5 )
5.3 (26)
45.1 t 3.7 ( 8 )
APOE = apolipoprotein E; VG = Volga German; ENVG = earlyonset non-VG.
were ~ 3 1 heterozygotes,
~ 4
and in one family there was
only one affected with a known APOE genotype.
In the VG and ENVG families, we found that APOE
genotype does not influence age of onset, even though
this study had 80% power (for presence vs absence of
~ 4 to) detect a 3.8-year difference (two-tailed, p <
0.05), well below the APOE age-modifying effect previously described t4, 9, 10).
We previously examined the association of APOE
and AD in these two groups [6} (see Table 1). In the
ENVG group, the ~4 allele is clearly not associated
with A D ( ~ frequency
in cases, 0.129). Presumably,
the autosomal dominant mutation(s) segregating in the
ENVG families is the major determinant of AD. Still,
APOE could be expected to modify age of onset if it
interacted with these mutation(s). However, we found
no support for this hypothesis.
In the VG group, no association between e4 and
AD was observed when spouses were used as controls
[b). This lack of association was based on a high frequency of ~4 in both VG affected (0.33) and VG
spouses (0.28). Selection of appropriate controls is critical in allelic association studies to avoid spurious positive results due to sample admixture. Because the VG
kindreds have a unique ethnic background and are descendant from a small founding population, spouses
are the most appropriate controls. The ~4 frequency
in VG spouses is higher than in most white populations
where the ~4 frequency is typically 0.15 {19}. The following are several explanations for a higher ~4 frequency in spouses: (1) The VG population may have
an elevated ~4 frequency similar to Scandinavian (0.24)
controls 120). This is possibly caused by genetic drift
since the V G population was founded by a relatively
small group of Germans who emigrated to Russia in
the 1760s {lS}. (2) The VG kindreds were ascertained
Brief Communication: Levy-Lahad et al: APOE and Alzheimer’s Disease 679
for a high load of A D cases and, in retrospect, this may
have also meant ascertaining for a high load of ~ 4 ,
especially since age of onset in the V G overlaps with
age of onset in the general population. ( 3 ) e4 frequency
could be an overestimate due to the limited number
of spouse samples available; in this case, there could
be an association between e4 and AD. In the present
study, we compared ages of’ onset between affected
individuals with different ApoE genotypes. Thus, the
selection of controls is not relevant to the conclusion
that there is no APOE ~4 age-of-onset effect in the
V G group. Where ~4 has been found to be associated
with AD, an age-of-onset effect has been observed [4,
9, 101, and the lack of such an effect in the VG provides additional evidence against an association of ~4
and A D in the V G group.
An analogous situation was observed in hereditary
cerebral hemorrhage with amyloidosis Dutch-type
(HCHWA-D), which is caused by a mutation in the
APP gene C21). In these pedigrees there is also a high
e4 frequency (0.34) in 3 1 affected individuals, including 5 affected ~ 4 1 homozygotes,
yet ApoE obviously
segregates independently of the APP mutation, and
ApoE genotypes d o not correlate with dementia or age
of onset. The high ~4 frequency in these pedigrees is
a result of founder effect and unrelated to the APP
In late-onset FAD [4-61, sporadic A D [S-81, and
possibly in A D caused by APP mutations [ I 1, 121, the
APOE ~4 allele is associated with decreased age of
onset. This study suggests that in at least two other
forms of FAD, the e4 allele does not appear to modulate age of onset, i.e., the (1) V G and ( 2 ) ENVG pedigrees linked to the chromosome 14 locus. This lack of
effect was reported in one study based mainly on two
chomosome 14 FAD pedigrees with an extremely early
age of onset [13]. Results presented here for the
ENVG kindreds extend the lack of e4 effect to chromosome 14-linked pedigrees with a later age of onset.
The A D mutations in the VG and the chromosome
14 loci could be so deleterious that even though APOE
is part of the same pathogenic pathway, the APOE
genotype does not modify disease course. APOE might
also affect penetrance of A D in some way without
changing age of onset. Another possibility, suggested
by the fact that even in late-onset AD, 40 to 60% of
affected subjects do not have the e4 allele, is that there
are at least two distinct mechanisms that converge to
the common phenotype of AD, one of which does not
involve APOE.
Supported by NIA grant AGO5 136 (Alzheimer’s Disease Research
Center at the University of Washington) and an NRSA post-doctoral
fellowship (F32 AG05635) (E.L.-L.). Also supported by G M 15253
Veterans Administration Research Funds (T.D.B), the American
Health Assistance Foundation (T.D.B., G.D.S.), and the American
Physicians Fellowship (E.L.-L.).
680 Annals of Neurology Vol 38 No 4 October 1995
W e thank D. Nochlan and M. Sumi for neuropathologic characterization of many of the families. Informed consent was obtained from
each subject or next of kin with approval of the University of Washington Human Subjects Review Committee.
1. Goate A, Chartier-Harlin M-C, Mullan M, et al. Segregation of
a missense mutation in the amyloid precursor protein gene with
familial Alzheimer’s disease. Nature 1991;349:704-706
2. Chartier-Harlin MH, Crawford F, Houlden H , et al. Early-onset
Alzheimer’s disease caused by mutations at codon 717 of the
bera-amyloid precursor protein. Nature 1991;353:844-846
3. Schellenberg G D , Bird TD, Wijsman EM, et al. Genetic linkage
evidence for a familial Alzheimer’s disease locus on chromosome 14. Science 1993;258:668-671
4. Corder EH, Saunders AM, Strittmatter WJ, et al. Gene dose of
apolipoprotein-E type-4 allele and the risk of Alzheimer’s disease in late onset families. Science 1993;261:921-923
5. Saunders AM, Strittmatter WJ, Schmechel D, e t al. Association
of apolipoprotein-E allele E-4 with late-onset familial and sporadic Alzheimer’s disease. Neurology 1993;43: 1467-1472
6. Yu C-E, Payami H , Olson JM, et al. The apolipoprotein E/CI/
CII gene cluster and late-onset Alzheimer’s disease. Am J H u m
Genet 1994;54:631-642
7. Van Duijn CM, de Kniff P, Cruts M, et al. Apolipoprotein E4
allele in a population based study of early-onset Alzheimer’s
disease. Nature Genet 1994;7:74-78
8. Poirier J, Davignon J, Bouthillier D, et al. Apolipoprotein-E polymorphism and Alzheimer’s disease. Lancet 1993;342:697-699
9. Tsai MS, Tangalos EG, Petersen RC, e t al. Apolipoprotein E:
risk factor for Alzheimer disease. Am J Hum Genet 1994;54:
10. Payami H , Montee KR, Kaye JA, et al. The apolipoprotein E
E4 allele and sex-specific risk of Alzheimer’s disease. JAMA
11. Hardy J, Houlden H, Collinge J, e t al. Apolipoprotein-E genotype and Alzheimer’s disease. Lancet 1993;342:737-738
12. St George-Hyslop P, Crapper McLachlan D, Tuda D , et al.
Alzheimer’s disease and possible gene interaction. Science 1994;
13. Van Broeckhoven C, Backhovens H , Cruts M, et al. APOE
genotype does not modulate age of onset in families with chromosome 14 encoded Alzheimer’s disease. Neurosci Lett 1994;
169: 179-180
14. Van Broeckhoven C, Genthe AM, Vandenberghe A, e t al. Faiure of familial Alzheimer’s disease to segregate with the A4amyloid gene in several European families. Nature 1987;329:
15. Bird TD, Lampe T H , Nemens EJ, er al. Familial Alzheimer’s
disease in American descendants of the Volga Germans: probable genetic founder effect. Ann Neurol 1988;23:25-31
16. Schellenberg G D , Payami H , Wijsman EM, et al. Chromosome-I4 and late-onset familial Alzheimer disease (FAD). Am
J Hum Genet 1993;53:619-628
17. Hixson JE, Vernier DT. Restriction isotyping of human lipoprotein E by gene amplification and cleavage with Hha I. J Lipid
Res 1990;31:545-548
18. Boehnke M. Allele frequency estimation from data on relatives.
Am J Hum Genet 1991;48:22-25
19. Davignon J, Gregg RE, Sing CF. Apolipoprotein E polymorphism and arteriosclerosis. Arteriosclerosis 1988;8:1-2 1
20. Hallman DM, Boerwinkle E, Saha N , et al. The apolipoprotein
E polymorphism: a comparison of allele frequencies and effects
in nine population. Am J Hum Genet 1991;49:338-349
21. Haan J , Van Broeckhoven C, van Duijn CMC, et al. The apolipoprotein E €4 allele does not influence the clinical expression
of the amyloid precursor protein gene codon 693 or 692 mutations. Ann Neurol 1994;34:434-437
Без категории
Размер файла
339 Кб
familiar, apolipoprotein, disease, age, alzheimers, genotypes, early, onset
Пожаловаться на содержимое документа