Neuron, Vol. 15, 219-228, July, 1995, Copyright 0 1995 by Cell Press
in Alzheimer’s Disease Brain
of Stable Complexes Involving
E and the Amyloid B Peptide
Lars 0. Tjernberg,’
Anders Ft. Karlstrom,
Samuel E. Gandy,# Lars Lannfelt,ll
Lars Terenius,’ and Christer Nordstedt’
Section of Experimental Alcohol and
Drug Addiction Research
Department of Clinical Neuroscience
tDepartment of Cell and Molecular Biology
Medical Nobel Institute
S-171 77 Stockholm
*Ludwig Institute for Cancer Research
S-751 24 Uppsala
BDepartment of Biochemistry
S-l 12 87 Stockholm
IIDepartment of Geriatric Medicine
S-l 41 86 Stockholm
#Department of Neurology and Neuroscience
Cornell University Medical College
New York. New York 10021
l Johan Thyberg,t
Genetic evidence suggests a role for apolipoprotein
(apoE) in Alzheimer’s disease (AD) amyloidogenesis.
was purified and characterized.
amyloid-associated apoE apparently exists not as free
molecules but as complexes with polymers of the amy-
loid g peptide (Af3). Brain Af%apoE complexes
detected irrespective of the apoE genotype, and simi-
lar complexes could be mimicked
structure of purified A3-apoE complexes was fibrillar,
and immunogold labeling revealed apoE immunoreac-
tivity along the fibrils. Thus, we conclude that AP-apoE
complexes are principal components of AD-associated
brain amyloid and that the data presented here support
a role for apoE in the pathogenesis
apoE from 32 AD patients
We found that brain
in vitro. The fine
Alzheimer’s disease (AD) is an aging-related dementia
without known treatments. One invariant neuropathologi-
cal feature of AD is deposition of intracerebral and menin-
geal amyloid (reviewed in Selkoe, 1991). The predominant
component of AD-associated amyloid is the amyloid 6 pep-
tide (A& which is mainly 40-42 amino acid residues in
length (Glenner and Wong, 1984; Masters et al., 1985;
Roher et al., 1993; Naslund et al., 1994). A8 is derived by
proteolytic processing of an integral membrane protein,
the j3-amyloid precursor protein (Kang et al., 1987). The
A8 peptide is detectable in soluble form in plasma and
cerebrospinal fluid (Seubert et al., 1992) and in medium
from cell cultures (Haass et al., 1992) whereas in amyloid
deposits, the A6 peptide forms an insoluble and protease-
resistant core (Masters et al., 1985). The physicochemical
properties of A6 enable it to form polymers and fibrils spon-
taneously (Kirschner et al., 1987; Nordstedt et al., 1994).
It has, however, been suggested that other proteins pres-
ent in, or in the vicinity of, the amyloid may act as”patholog-
ical chaperones,” facilitating A8 fibrillogenesis and depo-
sition (Wisniewski and Frangione, 1992). A number of
proteins unrelated to A8 have been detected immunohis-
tochemically in AD-associated
heparan sulfate proteoglycans (Snow et al., 1990), al-anti-
chymotrypsin (ACT; Abraham et al., 1988), amyloid P com-
ponent (Coria et al., 1988) complement proteins (Roze-
muller et al., 1989), and, most notably, apolipoprotein E
(apoE; Namba et al., 1991; Wisniewski and Frangione,
apoE is a 299 residue protein with a molecular mass of
approximately 34 kDa. In plasma, apoE associates with
lipoproteins and mediates transport of lipids and choles-
terol into the cells by binding to the low density lipoprotein
receptor or the low density lipoprotein receptor-related
protein/az-macroglobulin receptor (for review, see Weis-
graber, 1994). apoE exists in three major isoforms, apoE2,
apoE and apoE4, which are products of the ~2, ~3, and
~4 alleles, respectively. The isoforms differ in arginine or
cysteine content at positions 112 and 158. Recent genetic
studies suggest a role for apoE in the pathogenesis of AD.
The frequency of the ~4 allele is significantly higher in
sporadic (Mayeux et al., 1993; Rebeck et al., 1993; Saun-
ders et al., 1993) and familial late onset AD (Corder et
al., 1993; Strittmatter et al., 1993a) than in the general
population. In addition, brain tissue from AD patients car-
rying the ~4 allele contains more amyloid than brain tissue
from AD patients with other apoE genotypes (Rebeck et
al., 1993; Schmechel et al., 1993; Berr et al., 1994) im-
plying a direct role for apoE in cerebral amyloidogenesis.
How apoE might promote accumulation of amyloid and
development of AD is still not understood. Recent in vitro
experiments suggest that both apoE and apoE bind A8
peptide variants in a f3-mercaptoethanol-sensitive
(Strittmatter et al., 1993b; LaDu et al., 1994). The com-
plexes formed are stable to boiling in SDS, implying strong
intermolecular binding. Electron microscopic analysis of
these in vitro complexes demonstrated that both apoE
and apoE interact with A8 to form monofibrillar struc-
tures, distinct from the twisted ribbons formed by synthetic
A8 alone (Sanan et al., 1994). It has also recently been
shown that apoE (Maet al., 1994; Wisniewski et al., 1994)
and ACT (Ma et al., 1994) enhance polymerization of A8
in vitro, thereby providing a direct link between these pos-
tulated pathological chaperones and amyloid formation.
amyloid. These include
Figure 1. lmmunoblot
from AD Patients and Nondemented
mented controls, respectively,
troblotted onto nitrocellulose membranes, and probed with A6-specific
antibody 6ElO or apoE-specific antibody 2El. The total amount of
protein in each lane was 60 pg. Positions of the molecular mass mark-
ers in kilodaltons are shown at left.
Analysis of Amyloid-Enriched HFIP Extracts
HFIP extracts from 2 AD patients
were separated by SDS-PAGE,
and 2 nonde-
Amyloid-associated apoE purified from human brain tis-
sue has not previously been studied biochemically.
have extracted AD-associated amyloid deposits from pa-
renchymal brain tissue and report here that a C-terminal
fragment of apoE apparently exists in a complex with A6
polymers of varying sizes. Considerably more AD-apoE-
immunoreactive complexes were recovered
brains than from brains of elderly, nondemented controls.
The A6-apoE complexes were detectable in brain extracts
from all 32 AD patients examined, without apparent rela-
tionship to the apoE genotype (~31~3, .53/~4, or ~41~4). The
purified AD-apoE complexes were resistant to various
agents previously shown to dissociate A8 polymers, im-
plying strong intermolecular association. When viewed by
negative stain electron microscopy, the AD-apoE com-
plexes appeared as bundles of tightly packed amyloid fi-
brils morphologically similar to those of noncomplexed A8
purified from AD brain. lmmunogold labeling of the purified
complexes demonstrated that apoE immunoreactivity was
distributed along the fibrils and fibril bundles. Complexes
formed in vitro between synthetic A6 and recombinant
apoE showed chromatographic,
morphological features indistinguishable
apoE complexes purified from AD brain.
from the Aj3-
Detection of Highly Insoluble AP-apoE Complexes
in AD Brain Tissue
Amyloid-enriched brain tissue fractions from 32 AD pa-
tients and 4 nondemented controls were extracted with
hexafluoro-Ppropanol (HFIP). Extracted proteins from 2
representative patients in each group were analyzed by
immunoblotting using the A8-specific antibody 6E10 or the
apoE-specific antibody2El (Figure 1). In the samples from
the AD patients, but not from the nondemented controls,
a prominent band with an apparent molecular mass of
- 4.5 kDa was labeled by antibody 6ElO. This species was
identified as a mixture of intact and N-terminally truncated
A8(1-40) and A8(1-42) variants by microsequencing and
electrospray mass spectrometry (Naslund et al., 1994). In
addition to small quantities of A8 oligomers, the AD sam-
ples contained high molecular mass forms of A8 ranging
in size from 22 to 24 kDa to the separation limit of the gel,
corresponding to A3 polymers of varying sizes. Immu-
noblotting of the same extracts using the apoE-specific
antibody 2El yielded a staining pattern very similar to that
obtained with 6E10, except that 2El did not label mono-
meric and oligomeric A6 (Figure 1). Hence, 2El-immune
reactive material was apparently comigrating with A8 poly-
mers in SDS-PAGE. The immunoreactivity
antibodies could be abolished by preabsorption with their
respective antigens. The similar appearances of the AD-
and apoE-immunoreactive material indicate that the two
antibodies may have bound to the same structures, i.e.,
SDS-resistant complexes formed by apoE and polymers
of A6. Hereafter, these structures will be referred to as
A6-apoE complexes. As seen in Figure 1, the apparent
mass of the smallest A8-apoE complex is 22-24 kDa,
which is below the nominal 34 kDa mass of apoE, sug-
gesting that the apoE component of the complexes is trun-
cated (see below).
To enable biochemical and morphological studies of the
AD-apoE complexes, a protocol for partial purification was
developed. Proteins from an amyloid-enriched AD brain
tissue fraction were first separated by gel filtration on a
Superose 12 column (Figure 2A). Using this method, pro-
teins smaller than 22 kDa (including monomeric and oligo-
merit A6 peptide not associated with apoE) were effi-
ciently separated from the remainder of the material. The
amount of A8 immunoreactivity
apoE-complexed form varied, but was densitometrically
determined to constitute at least 50% of the total A6 immu-
noreactivity in each case (Figure 28). Fractions containing
the Aj3-apoE complexes (fractions 2-3) were pooled and
then subjected to reverse-phase liquid chromatography
(RPLC; Figure 2C). Previous experiments had shown that
noncomplexed A6 elutes at - 18% acetonitrile (ACN) in
this RPLC system (data not shown). The Af3-apoE com-
plexes eluted as a broad peak between 24% and 29%
ACN (Figure 2D). It is noteworthy that all fractions con-
taining A8 polymers also displayed apoE immunoreactiv-
ity. Fractions 7-l 0 from the RPLC sessions were pooled
and used in the subsequent characterization and dissocia-
of the Ap-apoE Complexes from AD
apparently residing in
of the Afl-apoE Complexes
A set of different antibodies was used to investigate
whether the A6-apoE complexes harbored molecules
other than A6 and apoE and to rule out cross-reaction
between antibodies. As seen in Table 1, the only antibod-
ies labeling the complexes were those raised against A8 or
and Biochemical Characterization
Purified from Human Brain
Figure 2. Fractionation
Filtration on a Superose 12 Column
(A) Absorbance profile at 280 nm of the HFIP extract eluted in 70%
formic acid. Fractionation started after an elution volume of 6 ml. Mono-
meric, noncomplexed A6 eluted at - 13.5 ml (fraction 9) in this system.
(6) lmmunoblot analysisof the Superose 12fractions. Aliquots of each
fraction were separated by SDS-PAGE,
lose membranes, and probed with Af3-specific antibody 6ElO (closed
circles) or apoE-specific antibody 2El (open circles). The resulting
autoradiograms were quantified by densitometry.
formed over the full length of each lane, and therefore optical density
values refer to the total immunoreactivity
(C) AB-apoE complexes recovered from the Superose 12 column were
further purified by RPLC on a Vydac C4 column and developed with
a linear gradient of buffer B (see Experimental
(D) Each RPLC fraction was probed with AD-specific antibody 6E10
(closed circles) or apoE-specific antibody 2El (open circles) and sub-
jected to the same analysis as in (E).
of an Amyloid-Enriched HFIP Extract by Gel
electroblotted onto nitrocellu-
Scanning was per-
in each sample.
Procedures for details).
apoE. Interestingly, antibody 1 D7, recognizing an epitope
overlapping the LDL receptor-binding domain located be-
tween residues 142-158 of apoE, did not label the com-
plexes. In control immunoblot experiments, 1 D7 intensely
labeled recombinant apoE. This, together with the positive
labeling obtained with antibody 3Hl (recognizing an epi-
Table 1. lmmunoreactivity
Determined by lmmunoblotting
of the AD-apoE Complexes As
a Similar staining as with 6ElO.
’ Similar staining as with 2El.
Kunitz protease inhibitor; P-tau, phosphorylated
APP, 6-amyloid precursor
tau; TTFI, transthyretin.
tope between residues 220-272)
apoE complexes are composed of the C-terminal part of
apoE. Although material reactive with antibodies to the
f3-amyloid precursor protein, ACT, ubiquitin, and phos-
phorylated tau could be recovered from AD brain using
the same extraction protocol, none was associated (i.e.,
copurified) with the AD-apoE complexes.
Microsequencing confirmed that A6 was a component
of the complexes. A Lys-C digest contained two peptides
with A6 sequences, one in the 0.1% trifluoroacetic acid
(TFA)-soluble fraction and one in the formic acid-soluble
fraction (see Experimental Procedures). The sequence ob-
tained in the 0.1% TFA-soluble fraction was XXIIGLMV
(eluting at 32% ACN), corresponding to residues 29-36
of Af3. In the formic acid-soluble
XAEFRHDSGYEVH (eluting at 35% ACN), corresponding
to residues l-l 3 of A6, was obtained. We were unable to
obtain any sequence corresponding to apoE, either by
trypsin or by Lys-C digestion. This was interpreted as evi-
dence either that apoE was present in levels too low to
allow microsequencing or that the apoE moiety of the com-
plex adopts a conformation too dense to allow penetration
of the proteases used.
suggests that the Af3-
fraction, the sequence
It was hypothesized that the A6 and apoE components in
the complexes might be separated if the material were
exposed to strong dissociating agents. The A@apoE com-
plexes were therefore treated with acids, bases, deter-
gents, chaotropic or denaturing agents, or organic sol-
vents (see Experimental Procedures), some of which have
previously been shown to dissociate A6 polymers purified
from AD brain (Masters et al., 1985). The incubates were
separated by SDS-PAGE, electroblotted onto nitrocellu-
of Ap-apoE Complexes
Figure 3. Congo
Purified AD-apoE complexes
taining 1% Congo red, smeared onto glass slides, and viewed under
polarized light at a magnification of 200 x
When Viewed under Polarized Light
AP-apoE Complexes Demonstrate
in 50% ethanol con-
lose membranes, and analyzed with antibody 6ElO. The
criteria used when determining whether the complexes
had dissociated were reduced amounts of AD-apoE com-
plex immunoreactivity with a molecular mass higher than
22-24 kDa and appearance of detectable amounts of A6
immunoreactivity within the 3-5 kDa region. Either crite-
rion would indicate that the A6 peptide had been released
from the complexes. None of the agents employed in-
duced any measurable dissociation.
AJI-apoE Complex Formation
Af3-apoE complexes were detected in brain extracts from
all 32 AD patients studied. In 21 of the patients, the apoE
genotype was known, showing the following distribution:
(n = 8) Es/Ed
(n = 12), and E4/E4 (fl = 1). No AD
patient with an ~2 allele was present in this study, probably
owing to the reported low frequency of this allele among
AD patients (4.4%; Peacock and Fink, 1994).
and apoE lsoforms
The tinctorial properties of the purified Af3-apoE com-
of Afi-apoE Complexes
Figure 4. Negative
(A) The purified AD-apoE complexes were incubated in TBS for 1 hr,
negativelystained with2% uranyl acetate, andexaminedinanelectron
(6) Less dense fibril matrix found in another region of the same speci-
men as in (A).
(C) Noncomplexed A!3 purified from AD brain tissue incubated and
stained as in (A).
Bars, 100 nm.
of Aggregated AB-
plexes were indistinguishable from those typical of tissue
amyloid (Glenner, 1980); i.e., when stained with Congo
red, they manifested green birefringence under polarized
light (Figure 3). The fine structure of the Af3-apoE com-
plexes was investigated further by electron microscopy.
After negative staining with uranyl acetate, tight bundles
of fibrils were found (Figure 4A). The strictly parallel ar-
rangement of the fibrils resembles that previously ob-
served in AD brain tissue (Davies and Mann, 1993). In
some regions, the bundles had apparently separated, and
finer substructures became evident (Figure 48). Thus, a
web-like matrix could be seen, which in its finest exten-
Complexes Purified from Human Bram
Figure 5. lmmunogold
(A) Labeling of Ab-apoE complexes with antibody against apoE (2El),
followed by negative staining.
(B) Labeling of the same specimen as in (A) with normal mouse IgG
as a control for nonspecrfic binding of the primary antibody.
Bars, 100 nm.
Labeling of the Ab-apoE Complexes
sions consisted of one or a few fibrils twisted around each
other. The fibrils had a diameter of 6-8 nm, in good accor-
dance with published data on the dimensions of A8 fibrils
(Sanan et al., 1994). The fine structure of noncomplexed
A8 purified from AD brain (see Figure 2A, fraction 9) is
shown in Figure 4C. As seen: the general morphology of
noncomplexed A8 was very similar to that of the Af3-apoE
complexes. Hence, the morphology of A8 fibrils is similar
in the presence or absence of apoE.
lmmunogold labeling of the A6-apoE complexes with
antibody 2El revealed that both the large fibril bundles
and the dispersed fibrils were strongly immunoreactive
(Figure 5A). This suggests that apoE was present not only
on the surface of the bundles but also between the individ-
Figure 6. lmmunoblot
lmmunoblot analysis of coincubates
either with (B) or without (A) 10 mM DTT. The products were collected
by centrifugation and treated with HFIP and 9 M urea before electro-
phoresis. Positions of the molecular mass markers in kilodaltons are
shown at left.
Analysis of AD-apoE Complexes Formed In
between Ab and apoE and corre-
were performed for 4 days at 37% Incubations
ual fibrils. Some scattered gold particles could also be
seen in close proximity to the fibril bundles. These particles
probably represent labeling of material disrupted from the
original structures during sonication, since the overall
background labeling was very low (Figure 5B).
To establish further that the Af3-apoE complexes are dis-
tinct chemical entities and not artifacts attributable to the
extraction or purification procedures, the formation of Af3-
apoE complexes was also characterized in vitro. Recombi-
nant apoE and synthetic Af3( l-40) or A/3( 1-42) were coin-
cubated under the conditions described in Experimental
Procedures. Without addition of reducing agents to the
incubation medium, apoE and apoE interacted with both
A6 variants and formed heterologous complexes with a
molecular mass of -40 kDa (Figure 6A). In agreement
with previous reports, the complexes were SDS-stable
(Strittmatter et al., 1993b; LaDu et al., 1994) and, in addi-
tion, were resistant to the HFIP and 9 M urea used in the
present protocol. It was further evident that only a fraction
of the total apoE reacted with A6 to form these avid com-
plexes (varying between 5% and 20010, as determined by
of Afl-apoE Complexes In Vitro
Figure 7. Electron Micrographs of At%-apoE Complexes and High Mo-
lecular Mass Ab Aggregates Generated
(A) Ab-apoE complexes generated in vitro by coincubation
40) and apoE in the presence of 10 mM DTT. The complexes were
purified by gel filtration, polymerized
(B) lmmunogold labeling with 2El of the same specimen
followed by negative staining.
(C) High molecular mass aggregates of Ab(l-40)
the sample in (A).
Bars, 100 nm.
in TBS for 1 hr, and negatively
as in (A),
that were treated as
densitometry). If, however, the incubation was performed
in the presence of a reducing agent (10 mM dithiothreitol
[OTT]), the complex formation proceeded beyond the stoi-
chiometric 1 :l complex, and higher molecular mass com-
plexes labeled with both 6ElO and 2El were formed (Fig-
ure 6B). In the case of A6(1-40), it is interesting to note
the stepwise 4 kDa mass increments of the complexes,
suggesting the successive incorporation of A6 monomers.
The A@apoE complex bands became less distinct and
more similar to the in vivo material when using synthetic
A6(1-42) indicating that the higher hydrophobicity of the
bound A8 induced aberrant, smear-like migration of the
complexes in SDS-PAGE. No AB-apoE complexes were
detected at zero-time incubations (data not shown), con-
firming that the complexes formed are not artifacts caused
by postincubation sample treatment.
The AD-apoE complexes formed under reducing condi-
tions in vitro were separated from residual noncomplexed
apoE and A6 by gel filtration in 70% formic acid and exam-
ined by electron microscopy after negative staining and
immunogold labeling. As for the A6-apoE
found in vivo, the high molecular mass products from the
in vitro coincubations principally formed dense aggregates
of parallel fibrils. However, in certain regions, a lattice of
fibrils with extensive crossover characteristics could be
found (Figure 7A). In addition, immunogold labeling of the
in vitro coincubates using anti-apoE antibody demon-
strated the same pattern of gold particle decoration, albeit
at a lower density, as that observed for the Af3-apoE com-
plexes formed in vivo (Figure 78). The lower density of
immunogold labeling of the in vitro Af3-apoE complexes
might be an effect of the limited time of incubation, or it may
additional, unknown factor. Differences in A6 fibril mor-
phology attributable to either apoE isoform were not ob-
served. When A8(1-40) or Af3(1-42) was incubated alone,
both peptides formed high molecular mass products that,
when subjected to negative staining, displayed the same
kind of fibril bundles as the in vivo and in vitro AP-apoE
complexes (Figure 7C). These fibrils were not immunore-
active to the apoE-specific antibody 2El. Thus, the influ-
ence of apoE on the final pattern of A6 fibril morphology
under these experimental conditions seems to be negli-
Here we present evidence that complexes between A8
and apoE are components of AD amyloid. The complex
formation occurs with either the apoE or the apoE iso-
form, as determined by apoE genotype analyses of the
AD patients and by in vitro reconstitution studies. The A8-
apoE complexes are joined by uncommonly strong inter-
molecular bonds, since they were resistant to treatment
with a number of potent dissociating agents, some of
which have previously been shown to dissociate A6 poly-
mers from AD brain (Masters et al., 1985). lmmunoblot
analysis suggests that the complexes are composed of
the C-terminal part of apoE, a finding supported by recent
in vivo studies (Wisniewski et al., 1995) and in vitro binding
studies (Strittmatter et al., 1993b). The relatively broad
retention profile of the Af3-apoE complexes in the RPLC
system used is probably attributable to heterogeneity of
the complexes, with the smallest complexes being com-
posed of one apoE and one A6 molecule and the large
complexes of severalfold more A6 than apoE molecules.
The molar excess of A6 may also explain why micro-
sequencing of the complexes yielded A8 fragments but
no apoE fragments. The retention times of the Af3-apoE
complexes were consistently longer compared with that
of noncomplexed A(3 purified from AD brain tissue. These
longer retention times indicate that the Af3-apoE com-
plexes are more aggregable than A8 alone. Furthermore,
Complexes Purified from Human Brain
all A3 eluting as polymers in the present chromatography
system is associated with apoE, suggesting the coexis-
tence of two populations of A6 in AD amyloid: one that
is dissociable in HFIP and formic acid and another, the
apoE-bound, that is resistant to dissociation. This conclu-
sion is in part supported by immunohistochemical
where antibodies against apoE stain some, but not all,
Af3-immunoreactive deposits (Strittmatter et al., 1993a;
Kida et al., 1994).
It has been suggested that pathological chaperones
may mediate amyloid formation of polypeptide fragments
(Wisniewski and Frangione, 1992). The Af3-apoE interac-
tion described here isof such apparent avidity that dissoci-
ation of the components under physiological conditions is
unlikely. The presence of apoE within or at the surface of
the A/3 fibrils is therefore not surprising. However, another
candidate pathological chaperone protein, ACT(Ma et al.,
1994) which by immunohistochemistry
to colocalize with Af3 in AD amyloid (Abraham et al., 1988),
was apparently not bound to A[3 using the present extrac-
tion and purification protocol. This may reflect a more tran-
sient mode of ACT interaction with AD.
The fine structure of the aggregated Af3-apoE com-
plexes was similar to that of noncomplexed AD, as deter-
mined by negative stain electron microscopy. The similar
morphology suggests that the A(3 components of the in
vivo complexes might be composed primarily of Af3 vari-
ants ending at residue 42, since At3 variants ending at
residue 42 are principal components of AD-associated am-
yloid (Roher et al., 1993; Naslund et al., 1994) and since
coincubations between Af3(1-42) and apoE in vitro yielded
a product with an immunoblot appearance nearly identical
to that of the complexes recovered from AD brain. The
in vitro-generated, high molecular mass polymers of Ab
showed the same pattern of tightly aligned fibrils, irrespec-
tive of the presence or absence of apoE. These results,
and other recent findings of Lansbury and coworkers (Ev-
ans et al., 1995), are indicative of a minor role for apoE in
determining the morphology of A6 fibrils and fibril bundles.
Rather than influencing the fibril morphology per se, apoE
may affect the rate of Af3 fibril formation, as in vitro data
suggest (Ma et al., 1994; Sanan et al., 1994; Wisniewski
et al., 1994; S. E. G., unpublished data).
It is particularly interesting that we have detected highly
stable Af3-apoE complexes in light of the genetic evidence
that apoE isoforms are an important determinant of an
individual’s risk for AD. In our study, we observe that these
complexes can include either apoE or apoE
Moreover, the apoE part of the AB-apoE complexes ap-
pears to be composed of a C-terminal fragment truncated
at a site C-terminal to residue 158. This complexed apoE
fragment is therefore identical in primary structure in all
three apoE isoforms. A direct interaction between Ab and
the polymorphic sites in apoE may not be critical, since
amino acid substitutions at residue 112, the single site at
which apoE and apoE differ, have been shown to affect
the structural properties of the C-terminal domain of apoE
(Dong et al., 1994). The Af3-binding capacity of apoE may
therefore be governed by domain interactions, similar to
what has previously been shown to determine the isoform-
has been shown
specific lipoprotein preferences
(Dong et al., 1994). It is not clear, though, whether cleav-
age of apoE occurs before or after A3 binding.
Since the extraction yield of the AD-apoE complexes is
not known, and possibly fluctuates between preparations,
no quantitative interindividual comparison in the amount
of the complexes was made. However, examination of the
immunoblot data from the 32 AD patients does not indicate
that there was a correlation between the apoE genotype
and the amount of Af3-apoE complexes extracted. It will
now be important to determine whether quantitative differ-
ences can be observed when different isoforms are com-
pared, orwhether apoE isoforms have adifferential affinity
for associating secondarily with deposited Af3. Of possible
relevance to this issue, differential association with soluble
Ab has been observed for denatured, delipidated apoE
isoforms (Strittmatter et al., 1993b; Ma et al., 1994; Sanan
et al., 1994; Wisniewski et al., 1994) and physiological
apoE isoforms (LaDu et al., 1994). Such differences be-
tween apoE isoforms in their association with A9 may lead
to differential proficiency in catalyzing fibril formation (Ma
et al., 1994; Sanan et al., 1994; Wisniewski et al., 1994)
thus possibly accounting for the overrepresentation of the
apoE isoform seen in both sporadic (Mayeux et al., 1993;
Saunders et al., 1993) and familial AD (Corder et al., 1993;
Strittmatter et al., 1993a). Alternatively, the isoforms could
differ in their ability to stabilize the 6 structure of A6 in
amyloid, since the deposition of Af3 in amyloid has been
shown to be a dynamic, reversible process (Maggio et al.,
The effect of inclusion of DlT in order to obtain in vitro
A&apoE complexes with similar chromatographic and im-
munoreactive characteristics as the complexes formed in
vivo superficially contradicts the results reported by Stritt-
matter and colleagues, who observed that inclusion of re-
ducing agents abolished AD-apoE complex formation
(Strittmatter et al., 1993b; Sanan et al., 1994). The discrep-
ancy in results is most likely explained by differences in
the protocols used for complex solubilization prior to SDS-
PAGE. Since A6 and apoE
DTT effect seen in the in vitro coincubates cannot be ac-
counted for by keeping sulfhydryl groups in a reduced
state. DlT may instead inhibit the formation of methionine
sulfoxides and so enhance the ability of apoE to bind oligo-
mers and polymers of A(3.
Complex formation between Ab and apoE in vivo might
have functional consequences
apoE to bind glycosaminoglycans.
tinct heparin-binding sites, one of which is located in the
C-terminal part of apoE between residues 202 and 243
(Cardin et al., 1966). This site does not overlap the postu-
lated AD-binding site in apoE, located between residues
244 and 272 (Strittmatter et al., 1993b). The apoE moiety
of the AD-apoE complexes may therefore provide an “an-
chor”forA(3 polymers, binding them to the glycosaminogly-
can-rich extracellular matrix of the brain tissue. These an-
chored Ab polymers can in turn serve as nuclei for further
A(1 polymerization and growth of amyloid deposits (Jarret
and Lansbury, 1993).
Since the AD-apoE complexes were found in significant
of apoE and apoE
lack cysteine residues, the
related to the ability of
ApoE contains two dis-
levels in all 32 AD brains examined and virtually identical
results were obtained when analyzing amyloid derived
from leptomeningeal AD tissue (J. N., unpublished data),
it is proposed that the formation of AD-apoE complexes
is an invariable feature of AD amyloidogenesis.
vivo observations may be relevant to an elucidation of the
molecular events that underlie the pathophysiology of ce-
rebral amyloidogenesis and to improving our understand-
ing of the role of apoE in this process.
Human Brain Tissue and Antibodies
Autopsy material from AD individuals (n = 32) was obtained from the
Huddinge Brain Bank, Huddinge Hospital, Sweden. AD diagnosis was
confirmed according to standard protocols
The AD cases had an average age of 78 years (range = 57-94 years).
The control cases (n = 4) were obtained from the Department of Foren-
sic Medicine, Karolinska Institute, Stockholm,
pathological examination nor clinical records revealed any sign of de-
mentia or other neurological disorders,
control cases was 75 years (range = 60-86 years). Occipital cortex
(IO-40 g) from both AD and control cases was removed at autopsy
and stored at -70°C until further use. Occipital cortex was used for
reasons of availability. ApoE genotypes were determined
tissue as described (Wenham et al., 1991). The antibodies
this study were either from commercial
exception, been described previously.
used were 6ElO (Kim et al., 1990), 4G8 (Kim et al., 1988), 2El (Boeh-
rmger), 1048(Chemicon), lD7and 3Hl (Milneetal..
et al., 1983), 22Cll (Boehringer), Alz 90 (Boehringer),
lich et al., 1992), tau-l (Boehringer),
following antisera were also used: G504 (see below), anti-AD (Boeh-
ringer), 369A (Buxbaum et al., 1990), anti-ACT (Calbiochem),
ubiquitin (Boehringer), and anti-TTR (Sigma). The general specificities
of these antibodies are given in Table 1. Polyclonal antiserum G504.
a gift from Dr. Toshiharu Suzuki, Rockefeller
in rabbits against AP(l-40) and purified by affinity chromatography
using Ae(l-40) Immobilized on a Hi-Trap NHS column (Pharmacia).
(McKhann et al., 1984).
Sweden. Neither gross
and the average age of the
sources or have, with one
The monoclonal antibodies
and AT-8 (Innogenetics). The
University, was raised
Brain tissue was extracted as described (Ntislund et al., 1994). Bnefly,
repeated extraction of the tissue in a buffer containing
mM phenylmethylsulfonyl fluoride, 150 mM NaCI, and 50 mM Tns-
HCI (pH 7.4) yielded a SDS-insoluble
with 1 ,I ,1,3,3.3-hexafluoro-2.propanol
and delipidation, the HFIP extract wasdried under a stream of nitrogen
Protein determination was performed
assay (Pierce). Samples were initially dissolved in 9 M urea, 50 mM
Tris-HCI (pH 7.4), boiled for 5 min, and diluted to 3 M urea with water
prior to protein determination.
1% SDS, 0.1
pellet that In turn was extracted
(HFIP). After removal of debris
using the bicinchonlnic acid
Samples were dissolved in 9 M urea and sonicated (Branson Sonifier
250) before addition of 3 x SDS-PAGE
were boiled and resolved on a 54/o-18% Tris-Tricine
tem. After electrophoresis, the proteins were electroblotted
cellulose membranes (0.2 urn; Schleicher & Schuell) or polyvinylidene
difluoride membranes (0.45 pm; Millipore). Membranes were probed
with monoclonal antibodies or antisera for 3-12 hr at room tempera-
ture, and antibody binding was detected by adding horseradish perox,-
dase-linked sheep anti-mouse, donkey anti-rabbit (both supplied by
Amersham), or rabbit anti-goat (Sigma) immunoglobulin,
on the host animal providing the primary antibody. The blots were
visualized usmg the enhanced chemiluminescence
sham) according to the instructions of the manufacturer.
of the blots was performed using a Bio-Rad Model GS-670 densitome-
ter and the accompanying Molecular Analyst software.
sample buffer. The samples
gradient gel sys-
The dry HFIP extracts were dissolved in 70% (v/v) distilled formic acid,
centrifuged at 10,000 x g for 2 min (to remove insoluble material),
and purified by gel filtration on a 1 x 30 cm Superose 12 column
(Pharmacia) developed with 70% formic acid at a flow rate of 200 pII
min (Roher et al., 1993). Absorbance
aliquots of the collected fractions were dried by vacuum centrifugation
after addition of SDS (0.1% final concentration).
were subjected to immunoblotting,
apoE complexes and noncomplexed
the solvent was evaporated. The Afl-apoE
solved in 70% formic acid and further purified by RPLC using a 0.46
x 15 cm Vydac 214 TP C4 column developed over 30 min with a
linear gradient of O%-80% buffer B, where buffer A was 60% formic
acid in 0.1% TFA and buffer B was buffer A containing 40% ACN
(Heukeshoven and Dernick, 1985). Flow rate was set at 500 jdlmin,
and absorbance was monitored at 280 nm. Aliquots of the collected
fractions were evaporated after addition of SDS (0.1% final concentra-
tion) and analyzed by immunoblotting.
complexes were pooled, evaporated,
characterization, digestion, dissociationstudies,
microscopy (see below).
of the Afl-apoE Complexes
was monitored at 280 nm, and
The dried aiiquots
the fractions containing the AD-
AD were pooled separately, and
complexes were redis-
Fractions containing Ap-apoE
and used for immunological
of the Ap-apoE
The partially purified AD-apoE complexes were alkylated with Cvinyl-
pyridine according to standard procedures. Desalting of the alkylated
proteins was performed on a 0.32 x 10 cm FastDesalting
(Pharmacia) developed with 20% ACN, 0.1% TFA at a flow rate of
100 pllmin, while monitoring absorbance
void fraction was evaporated to dryness, and the alkylated AD-apoE
complexes were resuspended in 10 pl of 9 M urea. After dilution to 1
M urea with 0.1 M ammonium bicarbonate buffer (pH 7.8), digestions
were performed with endoproteinase
trypsin(Promega) for 18 hrat 37OCat anenzyme:substrate
mass) of approximately 1:20. The incubations were terminated by the
addition of TFA (0.1% final concentration).
gests were resuspended in 0.1% TFA and centrifuged
for 2 min. The resulting supernatant and pellet were treated separately.
The peptides in the supernatant were separated
Pharmacia SMART system equipped with a 0.21 x 10 cm C2-Cl8
mRPC column (Pharmacia) developed over 40 min with a linear gradi-
ent of O%-50% ACN in 0.1% TFA at a flow rate of 100 pllmin. The
pellet was dissolved in 70% formic acid and diluted with 0.1% TFA
to a final formic acid concentration
0.21 x 15 cm Vydac 214 TP C4 column developed with the same
solvent system as for the supernatant peptides. N-terminal sequencing
of the peptide fragments was performed on an Applied Biosystems
Model 477 A gas-phase microsequencer
to the manufacturer.
Digestion and Microsequencing
at 254 and 280 nm. The
Lys-C (Boehringer) or modified
After evaporation, the di-
at 10,000 x g
by RPLC using a
of 10% before separation on a
using procedures according
One milliliter of each of the agents listed below was added to the
purified AD-apoE complexes. The complexes were incubated in poly-
propylene tubes with each agent for 12-48 hr. The samples were
then diluted with water and dialyzed, using benzoylated dialysis tubing
(Sigma), against 0.1 M ammonium bicarbonate (pH 7.8), before evapo-
ration and analysis by immunoblotting.
(lo%), urea (10 M), guanidine-HCI
ACN (lOO%), NH,OH (25%). NaOH (1 M), TFA (30%), and formic acid
Studies of the Ap-apoE Complexes
The agents used were SDS
(6 M), phenol (90%), HFIP (1 OO%),
Light and Electron
taining 1% Congo
smeared onto glass slides and viewed under polarized light. For elec-
tron microscopy, the AD-apoE complexes and noncomplexed
fied from AD brain (see Figure 2A, fraction 9) were dissolved in HFIP
before dilution in Tris-buffered saline (TBS) to a final concentration
5% HFIP. The samples were incubated in TBS (pH 7.4) at 37OC for
1 hr and pelleted by centrifugation
red (Bush et al., 1994). The suspension
were suspended in 50% ethanol con-
at 20,000 x g for 20 min using a
Purifted from Human Brain
Beckman TLA 45 rotor. The pellets were washed twice with water and
resuspended in a final volume of 50 aI of water. Following sonication
(Branson Sonifier 250) 5 ul aliquots were placed on copper grids
covered by a carbon-stabilized Formvar film, and 5 ul of freshly pre-
pared 2% uranyl acetate in water was added. After l-2 min, excess
fluid was removed with a filter paper and the grids were air-dried. The
specimens were examined and photographed
at 60 kV. For immunogold staining, A6-apoE
sorbed to nickel grids covered by a carbon-stabilized
air-dried. After washing in TBS for 2 min, nonspecific
blocked by incubation in TBS with 3% bovine serum albumin (BSA)
for 30 min. The grids were then placed on a droplet of either the
apoE-specific antibody 2E1 (1:50) or normal mouse IgG for 60 min
(both diluted in TBS, 0.1% BSA), passed over six droplets of washing
solution for 2 min each (TBS., 0.1% BSA), placed on a droplet of anti-
mouse IgG conjugated to 10 nm colloidal gold particles for 30 min
(Sigma; diluted 1:20 in TBS, 0.1% BSA). and passed over another six
droplets of washing solution. The grids were fixed in 2% glutaralde-
hyde in 0.1 M sodium cacodylate-HCI
under a jet of water from a Pasteur pipette for 15-20 s, and air-dried.
Before the final examination in the electron microscope, the specimens
were negatively stained with 2% uranyl acetate in water.
in a JEOL EM 1OOCX
Formvar film and
buffer (pH 7.3) for 5 min, washed
A6(1-40) and A6(1-42) were dissolved at 200 uM in 20 mM HEPES,
150 mM NaCI, 0.05% CHAPS (pH 6.90) by sonication. To create reduc-
ing conditions, 10 mM DTT was included in the buffer. Recombinant
apoE (PanVeraCorp., Madison, Wisconsin) at 1 mglml in 0.7 M ammo-
nium bicarbonate was added to a final concentration
KM), which adjusted the pH of the incubation solution to 7.4. The
incubation volume was 50 ul, and the amount of each reactant was
corroborated by amino acid analysis (apoE) and RPLC (A6). After 4
days of incubation at 37% in a shaking water bath, the A6-apoE
complexes were pelleted by centrifugation
in a Beckman TLA 45 rotor (4%). The pellets were resuspended
HFIP for 30 min at room temperature
immunoblotting. To remove noncomplexed
tron microscopy, the HFIP-treated and dried pellets were dissolved in
70% formic acid. The AD-apoE complexes were purified by gel filtra-
tion in 70% formic acid on a 0.32 x 30 cm Superose
(Pharmacia) using the SMART system. The fractions containing the
A6-apoE complexes, as determined by immunoblotting,
dried, and preparedforelectron microscopyasdescribedforthe
samples. Samples in which A6 was incubated alone were subjected to
of Ap-apoE Complexes In Vitro
of 50 ug/ml (1.47
at 20,000 x g for 20 min
and dried, before analysis by
apoE and A5 before elec-
We thank Drs. T. Suzuki and T. Ramabhadran for providing antiserum
G504 and monoclonal antibody 56-1, respectively; Drs. K. S. Kim and
H. M. Wisniewski for providing ascites 6ElO and 4G6; and Dr. Yves
L. Marcel, Lipoproteinand AtherosclerosisGroup,
Hear-l Institute, Ottawa, Ontario, Canada, for providing
and 3Hl. We also thank lnga Volkmann, Lena Lilius, and Karin Blom-
gren for expert technical assistance. AD brain tissue was obtained from
the Huddinge Brain Bank, Huddinge Hospital, Sweden. This study was
supported by National Institutes of Health grants HL49571-01
and AG-11506 (S. E. G.), The Swedish Medical Research
(J. T. and L.T.), The C. V. Starr Foundation and The Starr Program
for Neurogeriatric Research (S. E. G.), The Swedish Heart Lung Foun-
dation (J. T.), and the King Gustaf V 60th Birthday Fund (J. T.). J. N.
is supported by a fellowship from The Swedish Board of Industrial
Development. C. N. is a recipient of fellowships from The Berth von
Kantzow Foundation, The Swedish Society for Medical Research, and
The Nicholson Foundation.
The costs of publication of this article were defrayed
the payment of page charges. This article must therefore be hereby
marked “advertisement” in accordance
solely to indicate this fact.
ascites 1 D7
in part by
with 16 USC Section 1734
Received February 2, 1995; revised April 6, 1995
Abraham, C. R., Selkoe, D. J.. and Potter, H. (1986). lmmunochemical
identification of the serine protease inhibitor al-antichymotrypsin
the brain amyloid deposits of Alzheimer’s
Berr, C., Hauw, J.J., Delaere, P., Duyckaerts,
(1994). Apolipoprotein E allele ~4 is linked to increased deposition of
the amyloid 6-peptide (A-6) in cases with or without Alzheimer’s dis-
ease. Neurosci. Lett. 178, 221-224.
Bush, A. I., Pettingell, W. H., Multhaup, G., d. Paradis, M., Vonsattel,
J.-P., Gusella, J. F., Beyreuther, K., Masters, C. L., and Tanzi. R. L.
(1994). Rapid induction of Alzheimer
Science 265, 1464-1467.
Buxbaum, J. D., Gandy, S. E., Cicchetti,
A. J., Fracasso, R. P., Ramabhadran,
Greengard, P. (1990). Processing of Alzheimer B/A4 amyloid precursor
protein: modulation by agents that regulate protein phosphorylation.
Proc. Natl. Acad. Sci. USA 87, 6003-6006.
Cardin, A. D., Hirose, N., Blankenship,
J. A., Sparrow, D. A., and Sparrow, J. T. (1966). Binding of a high
reactive heparin to human apolipoprotein
rin-binding domains. Biochem. Biophys. Res. Commun. 734,763-769.
Corder, E. H., Saunders, A. M., Strittmatter,
D. E., Gaskell, P. C., Small, G. W., Roses, A. D., Haines, J. L., and
Pericak-Vance, M. A. (1993). Gene dose of apolipoprotein
allele and the risk of Alzheimer’s disease in lateonset families. Science
Coria, F., Castano, E., Prelli, F., Larrondo, L. M., van Duinen, S.,
Shelanski, M. L., and Frangione, B. (1966). Isolation and characteriza-
tion of amyloid P component from Alzheimer’s disease and other types
of cerebral amyloidosis. Lab. Invest. 58, 454-456.
Davies, C. A., and Mann, D. M. A. (1993). Is the “preamyloid” of diffuse
plaques in Alzheimer’s disease really nonfibrillar? Am. J. Pathol. 143,
Dong, L.-M., Wilson, C., Wardell, M. R., Simmons, T., Mahley, R. W.,
Weisgraber, K. H., and Agard, D. A. (1994). Human apolipoprotein
J. Biol. Chem. 269, 22356-22365.
Evans, K. C., Berger, E. P.. Cho, C.-G., Weisgraber,
bury, P. T., Jr. (1995). Apolipoprotein
namic inhibitor of amyloid formation: implications for the pathogenesis
and treatment of Alzheimer disease. Proc. Natl. Acad. Sci. USA 92,
Glenner, G. G. (1960). Amyloid
f3-fibrilloses (first of two parts). N. Engl. J. Med. 302, 1263-1292.
Glenner, G. G., and Wong, C. W. (1964). Alzheimer’s disease: initial
report of the purification and characterization
lar amyloid protein. Biochem. Biophys. Res. Commun. 720,885-690.
Haass, C., Schlossmacher, M. G., Hung, A. Y., Vigo-Pelfrey, C., Mel-
lon, A., Ostaszewski, 8. L., Lieberburg,
Teplow, D. B., and Selkoe, D. J. (1992). Amyloid 6-peptide is produced
by cultured cells during normal metabolism.
Heukeshoven, J., and Dernick, R. (1965). Characterization
system for separation of water-insoluble
matogr. 326, 91-l 01.
Jarret, J. T., and Lansbury , P. T., Jr. (1993). Seeding “one-dimensional
crystallization” of amyloid: a pathogenic
disease and scrapie? Cell 73, 1055-1056.
Kang, J., Lemaire, H. G., Unterbeck,
C. L., Grzeschik, K. H., Multhaup, G., Beyreuther, K., and Muller-Hill,
8. (1967). The precursor of Alzheimer’s
resembles a cell-surface receptor. Nature 325, 733-736.
Kida, E., Golabek, A. A.. Wisniewski,
Regional differences in apolipoprotein
plaques in Alzheimer’s disease brain. Neurosci. Lett. 767, 73-76.
disease. Cell 52, 467-501.
C., and Amouyel, P.
A6 amyloid formation by zinc.
P., Ehrlich, M. E., Czernik,
T. V., Unterbeck, A. J., and
D. T., Jackson, R. L., Harmony,
E: identification of two hepa-
W. J., Schmechel,
E type 4
K. H., and Lans-
E is a kinetic but not a thermody-
deposits and amyloidosis: the
of a novel cerebrovascu-
I., Koo, E. H., Schenk, D.,
Nature 359, 322-325.
of a solvent
Poliovirus proteins by re-
liquid chromatography. J. Chro-
mechanism in Alzheimer’s
A., Salbaum, J. M., Masters,
disease amyloid A4 protein
T., and Wisniewski, K. E. (1994).
E immunoreactivity in diffuse
Kim, K. S., Miller, D. L., Sapienza,
Grundke-lqbal, I., Currie, J. R., and Wisniewski, H. M. (1966). Produc-
tion and characterization of monoclonal
V. J., Chen, C.-M. J., Bai, C.,
antibodies reactive to syn-
Neuron Download full-text
Kim, K. S., Wen, G. Y., Bancher, C., Chen, C. J. M., Sapienza. V. J.,
Hong, H., and Wismewski, H. M. (1990). Detection and quantitation
of amyloid 6-peptide with two monoclonal
Commun. 7, 113-122.
Kirschner, D. A., Inouye, H., Duffy, L. K., Sinclair, A., Lind, M., and
Selkoe, D. J. (1967). Synthetic peptide homologous
from Alzheimer disease forms amyloid-like
Acad. Sci. USA 84, 6953-6957.
LaDu, M. J., Falduto, M. T., Manelli, A. M., Reardon, C. A., Getz.
G. S., and Frail, D. E. (1994). Isoform-specific
E to b-amyloid. J. Biol. Chem. 269, 23403-23406.
Ma, J., Yee, A., Brewer, H. B., Jr., Das, S.. and Potter, H. (1994).
Amyloid-associated proteins al-antichymotrypsin
E promote assembly of Alzheimer p-protein into filaments. Nature 372,
Maggio, J. E., Stimson, E. R., Ghilardi, J. R., Allen, C. J., Dahl, C. E.,
Whitcomb, D. C., Vigna, S. R., Vinters, H. V., Labenski, M. E., and
Mantyh, P. W. (1992). Reversible in vitro growth of Alzheimer disease
6-amyloid plaques by deposition of labeled amyloid peptide. Proc.
Natl. Acad. Sci. USA 89, 5462-5466.
Masters, C. L., Simms, G., Weinman, N. A., Multhaup, G., McDonald,
B. L., and Beyreuther, K. (1965). Amyloid plaque core protein in Alzhei-
mer disease and Down syndrome.
Mayeux, R., Stern, Y., Ottman, R., Tatemichi,
Maestre, G., Ngai, C., Tycko, B., and Ginsberg, H. (1993). The apolipo-
protein E epsilon 4 allele in patients with Alzheimer’s
Neural. 34, 752-754.
McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D., and
Stadlan, E. M (1964). Clinical diagnosis of Alzheimer’s disease: report
of the NINCDS-ADRDA Work Group under the auspices of Department
of Health and Human Services Task Force on Alzheimer’s
Neurology 34, 939-944.
Mime, R. W., Douste-Blazy, P. H., Marcel, Y. L.,and Retegui, L.(l961).
Characterization of monoclonal antibodies against human apolipopro-
tern E. J. Clin. Invest 68, 111-117.
Namba. Y., Tomonaga, M., Kawasaki,
(1991). Apolipoprotein E immunoreactivity
and neurofibrillary tangles in Alzheimer’s
amyloid in CreutzfeldtJakob disease. Brain. Res. 547, 163-166.
Naslund, J., Schierhorn, A., Hellman, U., Lannfelt, L., Roses, A. D.,
Tjernberg, L. O., Silberring, J., Gandy, S. E., Winblad, B., Greengard,
P., et al. (1994). Relative abundance of Alzheimer A6 amyloid peptide
variants in Alzheimer disease and normal aging. Proc. Natl. Acad. Sci.
USA 91, 0378-6362.
Nordstedt, C., Naslund, J., Tjernberg, L. O., Karlstrom. A. R., Thyberg,
J., and Terenius, L. (1994). The Alzheimer A6 peptide develops prote-
ase resistance in association with its polymerization
Chem. 269, 30773-30776.
Peacock, M. L., and Fink, J. K. (1994). ApoE a4 allelic association
with Alzheimer’s disease: independent confirmation
gradient gel electrophoresis. Neurology 44, 339-341.
Hebeck, G. W., Reiter, J. S., Strickland, D. K., and Hyman, B. T. (1993).
Apolipoprotein E in sporadic Alzheimer’s disease: allelic variation and
receptor interactions. Neuron 7 7, 575-560.
Roher, A. E., Lowenson, J. D., Clarke, S., Wolkow, C., Wang, R.,
Cotter, R. J., Reardon, I. M., Ziircher-Neely,
Ball, M. J., and Greenberg, B. D. (1993). Structural alterations in the
peptide backbone of b-amyloid core protein may account for its deposi-
tion and stability in Alzheimer’s disease. J. Biol. Chem. 268, 3071-
Rozemuller, J. M.. Eikelenboom, P., Stam, F. C., Beyreuther, K., and
Masters, C. L. (1989). A4 protein in Alzheimer’s disease: primary and
secondary cellular events in extracellular
pathol. Exp. Neurol. 48, 674-691.
amyloid peptide. Neurosci. Res. Commun. 2,
antibodies, Neurosci. Res.
to beta protein
fibrils in vifro. Proc. Nab.
bmding of apolipoprotein
Proc. Natl. Acad. Sci. USA 82,
T. K., Tang, M. X.,
H., Otomo, E., and Ikeda, K.
in cerebral amyloid deposits
disease and kuru plaque
into fibrils. J. Biol.
H. A., Heinrikson. R. L.,
amyloid deposition. J. Neuro-
Sanan, D.A., Weisgraber, K. H., Russel, S. J., Mahley, R. W., Huang,
Roses, A. D., and Strittmatter,
ates with 6 amyloid peptide of Alzhemer’s disease to form novel monofi-
brils. J. Clin. Invest. 94, 660-669.
Saunders, A.M., Strittmatter, W. J., Schmechel, D., St George-Hyslop,
P. H., Pericak-Vance, M. A., Joo, S. H., ROSI, B. L., Gusella, J. F.,
Crapper-MacLachlan, D. R., Alberts, M. J., et al. (1993). Association
of apolipoprotein E allele ~4 with late-onset familial and sporadic Alz-
heimer’s disease. Neurology 43, 1467-1472.
Schmechel, D. E., Saunders, A. M., Strittmatter, W. J., Crain, B. J.,
Hulette, C. M., Joo, S. H., Pericak-Vance,
Roses, A. D. (1993). Increased amyloid 6-peptide deposition in cere-
bral cortex as a consequence of apolipoprotein
onset Alzheimer disease. Proc. Natl. Acad. Sci. USA 90,9649-9653.
Selkoe, D. J. (1991). The molecular pathology of Alzheimer’s disease.
Neuron 6, 467-490.
Seubert, P., Vigo-Pelfrey, C., Esch, F., Lee, M., Dovey, H., Davis, D.,
Sinha, S., Schlossmacher, M., Whaley, J., Swindlehurst,
(1992). Isolation and quantitation of soluble Alzheimer’s 6-peptide from
biological fluids. Nature 359, 325-327.
Snow, A. D., Mar, H., Nochlin, D., Sekiguchi, R. T., Kimata, K.. Koike,
Y., and Wight. T. N. (1990). Early accumulation
neurons and in the 6-amyloid protein-containing
disease and Down’s syndrome. Am. J. Pathol. 737, 1253-1270.
A., Schmechel, D., Wisniewski,
W. J. (1994). Apolipoprotein
T., Frangione, B.,
M. A., Goldgaber, D., and
E genotype in late-
C., et al.
of heparin sulfate in
lesions of Alzheimer’s
M., Enghild, J., Salvesen, G. S., and Roses, A. D. (1993a). Apolipopro-
tein E: high avidity binding to 6-amyloid and increased frequency of
type 4 allele in late-onset familial Alzheimer’s
Acad. Sci. USA 90, 1977-1981.
Strittmatter, W. J., Weisgraber, K. H.. Huang, D. Y., Dong, L.-M..
Salvesen, G. S., Pericak-Vance, M., Schmechel, D., Saunders, A. M.,
Goldgaber, D., and Roses, A. D. (1993b). Binding of human apolipo-
protein E to synthetic amyloid b peptide: isoform-specific
implications for late-onset Alzheimer disease. Proc. Natl. Acad. Sci.
USA 90, 6096-8102.
Weisgraber, K. H. (1994). Apolipoprotein
tionships. Adv. Prot. Chem. 45, 249-302.
Weisgraber, K. H., Innerarity, T. L., Harder, K. J., Mahley, R. W., Milne,
R. W., Marcel, Y. L., and Sparrow, J. T. (1963). The receptor-binding
domain of human apolipoprotein E. J. Biol. Chem. 258,12346-12354.
Wenham, P. R., Price, W. H., and Blundell, G. (1991). Apolipoprotein
E genotyping by one-stage PCR. Lancet 337, 1156-l 159.
Wisniewski, T., and Frangione, B. (1992). Apolipoprotein
ical chaperone protein in patients with cerebral and systemic amyloid.
Neurosci. Lett. 735, 235-238.
Wisniewski, T., Castano, E. M.. Golabek, A., Vogel, T., and Frangione,
B. (1994). Acceleration of Alzheimer’s fibril formation by apolipoprotein
E in vitro. Am. J. Pathol. 145, 1030-1035.
Wisniewski, T.. Lalowski. M., Golabek, A., and Frangione, B. (1995).
Is Alzheimer’s disease an apolipoprotein
Wunderlich, D., Lee, A., Fracasso, R. P., Mierz, D. V., Bayney, R. M.,
and Ramabhadran, T. V. (1992). Use of recombinant
for generation and rapid characterization
Application to the Kunitz domain of human p amyloid precursor protein.
J lmmunol Meth. 747. l-11
W. J., Saunders, A. M., Schmechel, D., Pericak-Vance,
disease. Proc. Natl.
E: structure-function rela-
E: a patholog-
E amyloidosis? Lancet 345.
of monoclonal antibodies.