close

Вход

Забыли?

вход по аккаунту

?

Amyloid -protein deposition in the leptomeninges and cerebral cortex.

код для вставкиСкачать
:ein Deposition in the
Amyloid P-Prot
.
Leptomeninges an d Cerebral Cortex
.
T
Yasuhisa Shinkai, MD,* Masahiro Yoshimura, MD,t Maho Morishima-Kawashima, PhD,* Yuji Ito, MD,$
Hiroyuki Shimada, MD,5 Katsuhiko Yanagisawa, MD," and Yasuo Ihara, MD*
To further investigate the process of amyloid P-protein (AP) deposition, we determined, using sensitive enzyme immunoassays, the levels of AP40 and AP42 (ADS) in the soluble and insoluble fractions of the leptomeninges (containing
arachnoid mater and leptomeningeal vessels) and cerebral cortices from elderly control subjects showing various stages of
AP deposition and from patients affected by Alzheimer's disease (AD). In both locations, insoluble AP levels were higher
by orders of magnitude than soluble AP levels. Soluble AP levels in cortices were much lower than those in leptomeninges. In insoluble AP in the cortex, AP42 was by far the predominant species, and AP42 in AD cortices was characterized by the highest degree of modifications in the amino terminus. In contrast, this AP42 predominance was not
observed in insoluble AP in the leptomeninges, which were found to be able to accumulate APs to an extent similar to
that in the cortex, on a weight basis. The levels of insoluble AP in the leptomeninges or cortex generally correlated with
the degree of cerebral amyloid angiopathy or the abundance of senile plaque, respectively. However, the presence of
plaque-free cortical samples showing significant levels of insoluble A042 suggests that biochemically detectable AP accumulation precedes immunocytochemically detectable AP deposition in the cortex.
Shinkai Y, Yoshimura M, Morishima-Kawashima M, Ito Y, Shimada H, Yanagisawa K, Ihara Y. Amyloid P-protein
deposition in the leptomeninges and cerebral cortex. Ann Neurol 1997;42:899-908
Alzheimer's disease (AD), the most common form of
dementia among elderly individuals, is characterized
pathologically by the presence of innumerable senile
plaques (SPs) and neurofibrillary tangles throughout
the cerebral cortex [I].In addition, AD is often complicated by amyloid deposition in the leptomeningeal
vessel, referred to as cerebral amyloid angiopathy
(CAA) [2]. The major component of SP and CAA is
amyloid P-protein (AP), a 39- to 43-residue protein
[3-51, which is derived from a larger membranespanning glycoprotein [3].
A two-residue difference in the carboxyl terminus of
AP, one terminating at Val4' (AP40) and the other
terminating at Ala4' (AP42), has recently been highlighted as an important factor involved in Pamyloidogenesis [6]. Using well-characterized enzyme
immunoassays (EIAs) for the two AP species (APs), we
measured the AP levels in soluble fractions from leptomeninges containing arachnoid mater and leptomeningeal vessels (together referred to as the leptomeninges, unless otherwise indicated) from autopsy cases [7].
The study showed that (I) AP40 and AP42 are detectable in the soluble fraction of the leptomeninges;
(2) the AP levels in leptomeninges remain very low un-
From the *Department of Neuropathology, Faculty of Medicine,
University of Tokyo, ?Tokyo Medical Examiner's Office, $Department of Pathology, Tokyo Medical College, $Department of Clinical Pathology, Tokyo Metropolitan Tama Geriatric Hospital, Tokyo, and "Department of Dementia Research, National Institute for
Longevity Sciences, Aichi, Japan.
til age 50, and increase steeply thereafter; (3) the level
of AP42 is almost always severalfold higher than that
of AP40; and (4) in some cases with no immunocytochemically detectable CAA, significant levels of APs are
present [7].These findings should help clarify a part of
the characteristics of AP deposition in the leptomeninges during aging. In the present study we have undertaken the quantitation of both soluble and insoluble
Aps in carefully dissected leptomeninges and cerebral
cortices from elderly subjects who showed various
stages of AP deposition, and from AD patients, to further understand the processes of AP deposition in these
two locations in elderly control and AD brains.
Materials and Methods
Autopsy
The present study is based on autopsy cases at Tokyo Metropolitan Geriatric Hospital, Itabashi, and Tokyo Metropolitan Tama Geriatric Hospital, Higashimurayama, Tokyo.
Postmortem delay ranged from 4 to 12 hours. These subjects, age 80 to 90 years (mean, 85.4 years), were examined
thoroughly by general pathologists and neuropathologists,
and none of them were diagnosed as having had AD. Five
AD cases (68-84 years; mean, 76.8 years) were neuropatho-
Received Apr 11, 1996, and in revised form Jun 20, 1996, and May
20 and Aug 19, 1997. Accepted for publication Aug 19, 1997.
Address correspondence to Dr Ihara, Department of Neuropathology, Faculty of Medicine, University of Tokyo, 7-3-1 Hongo,
Bunkyo-ku, Tokyo 113, Japan.
Copyright 0 1997 by the American Neurological Association
899
logically investigated, all of which were complicated by some
degrees of CAA (as judged by Congo red birefringence). The
diagnosis of AD was made based on both clinical and neuropathological criteria [S].
Leptomeninges and Cerebral Cortex
About 30 mg of leptomeninges (containing leptomeningeal
vessels) were taken from the cortical surface (-2 X 2 cm) at
the occipital pole and in several cases at the other sites. The
leptomeninges consisted mainly of arachnoid mater and leptomeningeal vessels, because it was difficult to completely
strip pia mater from the cortical surface. About 80 mg of
cortical pieces were sampled at the underlying area (Brodmann area 17) and in several cases also from frontal cortices;
attached leptomeninges that penetrated deeply into the cortex and white matter were carefully dissected out. Specimens
for AP quantitation were rinsed immediately in ice-cold saline, weighed, and stored at - 80°C until use. Cortical blocks
from adjacent sites or from the same locations on the contralateral side and in several cases from other cortical areas
including frontal lobes were fixed in 10Yo formalin and subjected to histological and inimunocytochemical examinations
(see later).
Tissue Extraction
Sampled leptomeninges (-30 mg wet weight) were cut into
small pieces with a sharp blade and homogenized in 19 volumes
of Tris-saline (50 mM Tris-HCI, pH 7.6, 0.15 M NaCI) containing 1 rnM EGTA, 0.5 mM diisopropyl fluorophosphate,
0.5 mM phenylmethylsulfonyl fluoride, 1 p&nl Nu-ptosyl-Llysine chloromethyl ketone, 1 pg/ml antipain, 0.1 pg/ml pepstatin, and 1 pg/ml leupeptin, with a Teflodglass motor-driven
homogenizer (20 strokes). Sampled cortical pieces (-80 mg wet
weight) were similarly homogenized (10 strokes) in 4 volumes
of the above buffer. The leptomeningeal and cortical homogenates were centrifuged at 265,000 g for 15 min on a TL 100.3
rotor in a TLX centrifuge (Beckman). The supernatants were
diluted more than fivefold with 20 mM sodium phosphate
buffer (pH 7.0) containing 0.4 M NaCI, 2 mM EDTA, 10%
Block Ace (Dainippon, Tokyo, Japan), 0.2% bovine serum albumin, 0.075% 3-[(3-cholamidopropyI)dimethylammonio]-lpropanesulfonate (CHAPS),and 0.05% NaN, (buffer EC), and
applied to the two-site EIAs [9, 101. T o quantitate A@ in
the insoluble fraction, the pellets, after washed once more
with Tris-saline, were further extracted with more than 100
volumes (to the initial tissue volume) of 70% formic acid.
The homogenates were centrifuged on a T L 100.3 rotor as
outlined above. The resultant supernatants were neutralized
with NaOH and Trizma base and applied to EM.
spectively, on western blot of insoluble fractions from AD
and elderly brains, 4 3 4 3 is almost undetectable, as shown
with BC65, a monoclonal antibody specific for AP43 [14],
and thus BC05-based values can be considered to represent
largely AP42 (this assumption does not rule out the presence
of minute amounts of AP43 in insoluble fractions) [ 151.
In the present study, BAN50 or 4G8, each a capture antibody, was coated on a multiwell plate (Immunoplate I,
Nunc, Roskilde, Denmark), whereas BA27 or BC05 was
used as a detection antibody after conjugation with horseradish peroxidase.
Enzyme Immunoassays
EIAs for Aps were performed as described previously [9, lo].
In brief, aliquots (100 pl) from appropriately diluted Trissaline extracts of tissue homogenates or (neutralized) formic
acid extracts of pellets, as well as an authentic peptide,
Apl-40 or Apl-42 (Bachem, Torrance, CA), dissolved in
dimethyl sulfoxide, were applied to a BANSO- or 4G8-coated
multiwell plate, and the loaded plate was incubated at 4°C
overnight. After being rinsed with phosphate-buffered saline,
loaded wells were incubated with horseradish peroxidaseconjugated BA27 or BC05 at room temperature for 6 hours.
Bound enzyme activity was measured with the TMB Microwell Peroxidase Substrate System (Kirkegaard and Perry
Laboratories, Gaithersburg, MD). Each point in Figure 1
(A-D) represents the mean of three (or two) values obtained
by three (or two) different dilutions of the sample.
The values obtained by the 4G8-based EIA (4GNBA27 or
4GSlBCO5) were compared with the values obtained by using the other combination, 4G8 as a detection antibody with
BA27 or BC05 being a capture antibody (BA27/4G8 or
BC05/4G8) [ 121. Both combinations showed no differences
in the EIA values [y = -2,265.6 + 1.3989x, 2 = 1.000,
for AP40, where x = log(4GUBA27 value) and y =
log(BA27/4G8 value); y = 7,108.7 + 0.86847~, = 0.985,
for AP42, where x = log(4G8/BC05 value) and y =
log(BC05/4G8 value)]. This indicates that no specificity
changes are caused by using the above antibodies as either
capture or detection antibodies (also see Reference 16).
T o examine whether the AP levels were determined
accurately in leptomeningeal samples, the two EM systems
with distinct specificities, 4G8-based and BAN50-based
ones, were compared with each other. The 4G8-based EIA
values were found to have an excellent correlation with the
BANSO-based values in those samples (also see Reference 7).
In the present leptomeningeal samples, y = 476.50
1.0154x, 2 = 0.935, for soluble AP40, y = 493.31 +
1.5683x, 2 = 0.737, for soluble AP42, y = 5,738.4
0.99718x, 2 = 0.962, for insoluble AP40, y = 17,420
2 . 9 6 9 7 ~ ,12 = 0.764, for insoluble AP42, where x = log(BAN50 value) and y = log(4G8 value).
T o further confirm the data by EIA, Aps in several cortical and leptomeningeal samples (n = 5) were quantitated by
western blotting according to a recently developed sensitive
procedure [17]. Small aliquots of the formic acid extracts
(see earlier) were dried by Speed Vac (Savant). For each sample, one part of the aliquot was solubilized with 2% hot sodium dodecyl sulfate (SDS) solution for protein determination with bicinchoninic acid (BCA) protein assay kit (Pierce,
a
+
+
Antibodies
AP monoclonal antibodies used in rhe present study were
BAN50 (raised against AP1-16), BA27 (specific for AP40),
BC05 (specific for Af342), and 4G8 (specific for AP17-24),
whose specificities were described in detail elsewhere [7, 9111. 4G8 was purchased from Senetek PLC.
BC05 is known to have a low affinity for AP43 (1/5-1/10
that for AP42) and thus BCO5-based values were previously
documented as AP42(43) [7, 10, 12, 131. However, whereas
AP40 and 42 are readily detected with BA27 or BC05, re-
900 Annals of Neurology
Vol 42
No 6
December 1997
+
Rockford, IL), and the other part was solubilized with the
sample buffer (50 mM Tris-HCl, p H 6.8, 12% glycerol, 2%
SDS, 2.5% mercaptoethanol, 4 M urea), and subjected to
Tridtricine gel electrophoresis. Various amounts of authentic
Apl-40 and 1-42 were similarly treated and subjected to
the same gel system. Separated proteins were blotted onto
nitrocellulose membrane (pore size, 0.22 p m , Schleicher &
Schuell, Dassel, Germany). The blots, after heat treatment
[17], were incubated with BA27 or BC05. After rinse they
were further incubated with horseradish peroxidaseconjugated goat anti-mouse IgG (Transduction Laboratories,
Lexington, KY). Bound antibodies were visualized with the
enhanced chemiluminescence (ECL) system (Amersham,
Buckinghamshire, UK). In our hands, the above protocol
made it possible to clearly detect up to 25 fmol of (formic
acid treated) APs. For each sample, western blot quantitation
was performed four to six times. Only the AP species that
migrated at about 4 kd, probably an AP monomer, was
quantitated with a densitometer (Model GS-700 imaging
densitometer, Bio-Rad) with synthetic APs used as standard;
AP species at about 8 kd and smear were not quantitated.
The data by western blot and EIA quantitation were found
to correlate well with each other; y = - 106,923.152
3.952s 2 = 0.742, for AP40; and y = -10,200.169 +
0.683x, ? = 0.845, for AP42, where y = western blot value
expressed as picomoles and x = EIA value expressed as picomoles. Thus, the values obtained by the two methods
agreed fairly well with each other, suggesting that EIA quantitate AP monomer exclusively, but further confirmation is
required for this point.
The above observations indicate that the amounts of APs
in the insoluble fractions, as determined by EIA, should be
considered as the lowest estimate. Furthermore, we do not
know whether the extraction with concentrated formic acid
is complete (no better solvents are currently known), and it
is quite possible that substantial amounts of insoluble APs
are tightly bound to other components and/or readily reaggregate into amyloid fibrils during neutralization [ 181, thus
precluding the detection by the two-site EIA. In fact, EIA
cannot quantitate highly aggregated forms of APs (data not
shown).
The amount of AP deposition in some AD brains determined in this study differed severalfold from that reported
by another group using the same E M system [12]. This may
in part be attributed to differences in the extraction method.
In our experience, if a brain contains large amounts of APs,
a large volume of formic acid is required for complete extraction of the APs; substantial amounts of APs (10-20% of
the initial amount) were left in the residues, when 10, instead of 100, volumes of formic acid were used (data not
shown). The use of a capture antibody as a detection antibody, and vice versa, may be expected to create problems
[ 161, but we found no discrepancy in the values between the
two systems (see above). In a few cases, multiple cortical sites
were sampled and measured for APs. The AP contents were
found to vary greatly depending on the sampling locations,
with the occipital pole usually showing the highest levels of
APs in a given brain; the differences in the AP levels among
cortical regions sometimes amounted to more than one order
of magnitude in a given brain (data not shown). This may
+
also explain in part much higher AP levels observed in our
series compared with data reported by other groups [12].
Choice of the weight basis or protein basis was also examined with regard to the above cases. There was an excellent
correlation between the ratio of leptomeningeal AP42 to cortical AP42 on a protein basis and that on a weight basis (y =
46.288
1.064x, 2 = 0.93, where y = leptomeningeal
AP40/cortical AP40 on a protein basis, x = that on a weight
1.256x, 2 = 0.923, where y = leptobasis; y = 0.383
meningeal AP42/cortical AP42 on a protein basis, x = that
on a weight basis). Thus, when the amount of leptomeningeal AP exceeds that of cortical AP on a weight basis, the
same can be applied to the protein basis ratio.
+
+
Semiquantitative Immunocytocbemistry of CAA
and SP
The formalin-fixed cortical blocks were dehydrated and embedded in paraffin, and cut into 6-pm-thick sections. Sections were immunostained with 4G8 or anti-human tau by
the avidin-biotin method (Vectastain Elite, Vector Laboratories, Burlingame, CA), as described previously [13]. For enhancement of AP immunostaining, deparaffinized tissue sections were treated with concentrated formic acid for 5
minutes at room temperature. The degree of CAA and the
abundance of SPs were semiquantitatively assessed on APimmunostained sections.
The degree of CAA was rated according to the number of
vascular amyloid deposits along a given leptomeningeal
length (-2 mm) in three nonselected microscopic XlO0
fields: 0, none; 1, 1 to 2; 2, 3 to 4; 3, 5 to 6; and 4, more
than 6/2 mm. The numbers of SPs were counted and averaged in three nonselected microscopic X 100 fields (-3.14
mm2). Small dot form plaques ( < l o p m in diameter) were
not counted. The abundance of SPs was rated as follows: 0,
none; 1, 1 to 50; 2, 51 to 100; and 3, more than 100/3.14
mm2.
Regarding tangles, only AD cortices showed abundant
neuropil threads detected with anti-human tau.
Results
Soluble Aps in Leptomeninges and Cortex
The specificities and sensitivities of the EIA were repeatedly confirmed [9, 10, 121. BANSO-based EIA can
be used to quantitate intact (full-length) Apl-40 and
Apl-42 [7],whereas 4G8-based EIA is assumed to
measure all AP species (total A@) including intact one
and variously amino-terminally modified or ragged species truncated up to position 10 [18-201. We first
quantitated total Aps in the soluble fractions of leptomeninges or cortices using a 4G8-based EIA (Fig 1A
and B). The levels of soluble APs in the cortex were
orders of magnitude lower than those in the leptomeninges, per gram wet weight (see Fig 1A and B; p <
0.01, Mann-Whitney U test). The levels of soluble
Aps in the leptomeninges appeared to correlate with
the degree of CAA (see Fig 1A). In fact, Spearman correlation analysis showed that the rating score for CAA
correlated well with the levels of soluble AP40 (gP =
Shinkai et al: AP Deposition in Leptomeninges and Cortex
901
soluble fraction of leptorneninges
103 t
0
0
Em
.g
t
. .
102r
:
10’;
r
0 :
10’
0
:
:
.
0.
k
v
0
0
e
O
0.
0
.
a
0 .
0 .
g:
9
0
100’
0
0
0
:
..
0
-
103
i
102
;.
0.
.
el
8 -
0
0.
0
8
._ .
0
e
0
..
.
.
0
0
0 .
-_ _ --_ ---
------
_ -_ _ -
100
0
1
2
3
4
AD
Degree of CAA
A
10
106
.
10
105
r
..
A
E
.-m
.tt.
1041
.
F
-K
.
10J:
e
a
0
102
10
1
2
3
4
AD
0
Degree of CAA
8
-
.
8
%_-.
0
0 .
101
.
0
0
$
j !:
0
0
L_ _ _ _ _ _ _ _ _ - - _ - _ :
10
..
.
..
.
.
0
0
0
0
0
0
0
0
1
0
2
3
Abundance of SP
AD
D
C
Fig I . Levels of amyloid p-protein rpecies (APs) in the leptomeninges and cerebral cortex and the degree of cerebral amyloid angiopathy (CAA) or the abundance o f senile plaques (SPs). AP40 (0)
and AP42 (a)were quantitated by enzyme immunoassay, using
4G8 as a capture antibody and BA27 or BC05 as a detection antibody. The ordinate is a logarithmic scale. The right-most colurnn in each $gure shows the levels of APs in Alzheimer? disease (AD) leptomeninges and cortices (5 cases). (A) APs in the soluble
fractions of leptomeninges and the degree of CAA, which was graded as 0, 1, 2, 3, and 4 (see Materials and Methods). The reliable range of values was above 4 pmol AP/g wet weight, as indicated by the broken line. (B) APs in the soluble fraction o f the
cortex and the abundance o f SPs, which was graded as 0, I , 2, and 3 (see Materials and Method). The reliable range of values
was above 0.2 pmol AP/g wet weight. (C) Aps in the insoluble fractions of leptomeninges and the degree of CAA.The reliable
range of values was above 80 pmol AP/g wet weight. (0)APs in the insoluble fractions of cortices and the abundance of SPs. The
reliable range of values was above 80 pmol AP/g wet weight.
0.72, p < 0.0001) and soluble AP42 ( P = 0.61, p =
0.0002) in the leptomeninges. Soluble APs were
present at readily detectable levels i n the leptomeninges, even from those specimens showing no CAA in
the adjacent areas (see Fig IA). This confirmed our
previous observation [7].
The ratio of soluble AP40/AP42 in leptomeninges
correlated only very weakly with the severity of CAA
902 Annals of Neurology
Vol 42
No 6
December 1997
(gP = 0.379, p = 0.0476), but it was significantly elevated in AD brains compared with that in control
brains, if the two groups are compared irrespective of
the degree of CAA ( p < 0.01, Mann-Whitney Utest).
This is again consistent with our previous observation
showing that AD leptomeninges are characterized by
higher levels of AP40 [7].
In a similar manner, there was a significant correla-
tion between the abundance of SPs and the levels of
soluble AP40 (r”P = 0.688, p = 0.0002) and soluble
AP42 (I”p = 0.53, p = 0.001) in the cortex (see Fig
la). The ratio of soluble AP40/AP42 was significantly
elevated in AD cortices ( p < 0.05, Mann-Whitney U
test).
Imoluble Aps in Leptomeninges and Cortex
We quantitated the levels of Aps in the insoluble fractions of the leptomeninges and cortex and correlated
those with the extent of CAA and the abundance of
SPs. The insoluble fraction of the leptomeninges was
found to contain unusually high levels of APs; the
AP42 contents appeared to be, indeed, more than
those in the cortex per gram wet weight (statistically
not significant; see Fig 1C). The levels of Aps in each
insoluble fraction appeared to be proportional to the
degree of CAA or the abundance of SPs (see Fig 1C
and D ; also see Reference 22). The Spearman correlation analysis confirmed that the degree of CAA correlated with the levels of AP40 (fP = 0.64, p = 0.0003)
and AP42 ( P = 0.61, p = 0.0004) in the leptomeninges and that the abundance of SPs correlated well
with the levels of AP40
= 0.76, p < O.OOO1) and
AP42 ( r ” P = 0.86, p < 0.0001) in the cortex.
It is noteworthy that there was remarkable predominance of A642 over AP40 in the insoluble fractions
from cortices irrespective of the stage of SP formation
(AP42/AP40; mean = 109 in cortex vs mean = 47 in
leptomeninges, p < 0.01, Mann-Whitney U test), but
there was no such AP42 predominance in the leptomeninges (see Fig 1C and D ) .
The AP42 levels in AD cortices were significantly
higher than those in controls at stages 0, 1, and 2 ( p <
0.05, Mann-Whitney Utest) but not those in controls
at stage 3. The ratios of insoluble AP401AP42 were
significantly elevated in the leptomeninges of AD
brains, compared with those of control brains ( p <
0.05, Mann-Whitney U test), but not statistically significant in the cortices.
We also determined the ratios of soluble AP40 or 42
to insoluble AP40 or 42, respectively, in the leptomeninges and cortex, parameters that may represent the
aggregation states in these two locations. In the leptomeninges, the ratios of soluble AP40 to insoluble
AP40 and soluble AP42 to insoluble AP42 were 8.5%
and 3.5%, respectively. In the cortex, the corresponding figures were 6.8% and 0.1%. The ratios for soluble
to insoluble AP40 were not significantly different ( p >
0.05, Student’s t test), whereas those for soluble to insoluble AP42 were significantly different between the
leptomeninges and cortex ( p < 0.05, Welch’s t test).
Because occipital cortex is known as the area rich in
cored plaques, one may argue that this sampling may
be a cause for unusually high levels of insoluble AP,
and the data on the frontal cortex, an area relatively
poor in cored plaques, are quite different; the levels of
insoluble AP may not be so high. Thus, we selected
several cases showing varying abundance of plaques,
following the criteria used in this study. The results
obtained have been essentially the same as those from
the occipital cortex; frontal cortices also contained very
high levels of insoluble A @ ,in parallel with the abundance of SPs (data not shown). Although we cannot
completely exclude the possibility that other areas in
the brain may show different characteristics, the
present results suggest that the major characteristic uncovered in the occipital pole, a large amount of insoluble AP in the cortex, can be found anywhere
throughout the cortex.
AP Modijcations in Leptomeninges and Cortex
Because BANSO- and 4G8-based EIAs most likely
measure intact, unmodified AP and all AP species, respectively, the ratio of 4G8-based values to BANSObased values represents the extent of modifications; a
larger ratio indicates more modifications of AP (mostly
due to truncation). In the leptomeninges, the ratio for
AP42 was larger than that for AP40 in both soluble
( p < 0.01, Mann-Whitney Utest) and insoluble fractions ( p < 0.01, Mann-Whitney U test). In cortices a
similar tendency was noted only in the insoluble AP
( p < 0.01, Mann-Whitney U test).
Thus, AP42 was more modified than AP40 in both
the leptomeninges and the cortex. The most modified
among the species of Aps from all the fractions was
insoluble AP42 in the cortex (Fig 2, and data not
shown). According to the Spearman correlation analysis, the extent of the modifications of insoluble AP42
in the cortex correlated with the abundance of SPs ( P
= 0.628, p = 0.003) (see Fig 2). In leptomeninges the
extent of the modifications was significantly higher in
AD brain than that in controls ( p < 0.05, MannWhitney U test).
Discussion
We have been concerned primarily with AP deposition
in the cortex, because this deposition may eventually
lead to the formation of neurofibrillary tangles, and to
extensive neuronal loss that is closely related to dementia [l]. AP deposition in the leptomeninges may bring
about the extreme fragility of meningeal vessels, resulring in nonhypertensive cerebral hemorrhage among elderly individuals [2].AP deposition in the cortex and
that in the leptomeningeal vessels often coexist in the
elderly brain, but this does not necessarily mean that
the two pathological processes are interdependent; the
rare conditions, extensive SP formation without CAA
(Cases 14, 15, 16, and 17 in Tables 1 and 2) and severe CAA without SP formation (Cases 33 and 34 in
Tables 1 and 2), are well documented (see References 7
and 3). Accordingly, it is most likely that these two
Shinkai et al: AP Deposition in Leptomeninges and Cortex
903
0
1
3
2
AD
Abundance of SP
Fig 2. Amino-terminal rnodz$cations of arnyloid P-protein 42
(AP42) in the insolublefractions o f cortices and the abundance of senile plaques (SPs). The ratios qf AP42 quantitated
by 4G8-based enzyme immunoassay (EIA) to that by BAN50based EIA are plotted against the abundance of SPs. Cases
with EIA values bdow the reliable range of AP values were
excluded JFom calculation. A D = Alzbeimer j disease.
processes are independent of each other in nature.
Thus, to understand P-amyloidogenesis, it is essential
to isolate one process from the other and characterize
each separately.
The accurate quantitation of APs in the cortex has
been hampered by the almost inevitable contamination
of cortical samples with attached leptomeninges (containing meningeal vessels), which penetrate deeply into
the cortex. T o circumvent this problem, we first quantitated APs in leptomeninges that are dissected out of
the cortex easily and observed relatively high levels of
soluble APs and very high levels of insoluble APs, especially in cases having some degree of CAA (see Fig
1A and C). It is surprising to learn how much AP can
accumulate in the leptomeninges per gram wet weight
(see below).
Some brains show advanced pathologies in the leptomeninges (CAA) but not in the cortex (SP). In such
cases even slight contamination of cortical samples with
leptomeningeal tissues would obscure the true levels of
soluble and insoluble APs, especially AP40, in the cortex. The present data (see Fig 1A and C) further suggest that even if there is no immunocytochemically detectable CAA, contamination of leptomeninges could
potentially distort the result on the AP levels in the
cortex (see below). This probably explains our puzzling
previous observations that control cortical pieces accompanying very little leptomeningeal tissues occasionally showed unusually high levels of APs (also see Reference 9). After completing AP quantitation in
904
Annals of Neurology
Vol 42
No 6
December 1997
leptomeninges, we quantitated APs in the cortex from
brains exhibiting various stages of SP formation. In
fact, very low levels of soluble APs in cortical samples
may indicate the least, if any, contamination of the
samples with leptomeninges. This somewhat roundabout approach uncovered the true characteristics of
A@ deposition in the cottex.
AP deposition in the cortex, compared with that in
the leptomeninges, is characterized by remarkable predominance and greater insolubility of AP42. AP42 was
recovered from the cortex almost exclusively into the
insoluble fraction, whereas very small amounts of APs
were released into the soluble fraction. In contrast, in
the leptomeninges, much larger proportions of APs
were soluble. As described above, the levels of soluble
APs in the cortex were probably overestimated because
leptomeninges contaminating the samples contained
large amounts of soluble APs (see Fig 1A). In fact, we
cannot exclude the possibility that the soluble APs in
the cortex were derived entirely from very small
amounts of contaminating leptomeningeal tissues.
Marked predominance of AP42 in the insoluble
fraction of the cortex was noted not only at the early
stage of SP formation but also in its advanced stage in
elderly controls and in AD cortices (see Fig 1D). This
result complements a series of recent immunocytochemical observations, strongly suggesting that AP42 is
the predominant species and the one initially deposited
[13, 231. With very careful dissection used in this
study, the marked AP42 predominance in the cortex
throughout all stages of SP formation is now firmly
established (also see Reference 21).
AP deposition in the leptomeninges appears to differ
from that in the cortex in the higher solubility and
abundance of AP40. No predominance of AP42 over
AP40, as seen in the cortex, was found in the leptomeninges (see Fig 1A and C). Much larger proportions
of APs were recovered in the soluble fraction of the
leptomeninges compared with those in the soluble fraction of the cortex; a solubility index for AP [soluble
AP/(soluble AP
insoluble Ap)] in the leptomeninges
is larger than that in the cortex. The index for AP40 in
the leptomeninges is 7.20%, whereas that for AP40 in
the cortex is 3.14%. The index for AP42 in the leptomeninges is 3.38%, but that for AP42 in the cortex is
0.169%. These solubility characteristics were already
suggested by two previous studies [24, 251.
The most striking characteristic revealed by the
present study is that the leptomeninges can accumulate
APs to an extent barely seen in the cortex on a weight
basis. First, it is possible that the amounts of APs in
the cortex are more underestimated by EIA than those
in the leptomeninges; APs in the cortex may have a
stronger tendency to reaggregate during neutralization
of formic acid, thus reducing the EIA values. More importantly, this may suggest questions regarding the va-
+
Table 1. Amyloid P-Protein Deposition in Leptomeninges
~
Soluble Fraction
Case
Age
No.
(yr)/Sex
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
82/F
861F
84/F
891M
811F
80/F
82/F
83/F
81/M
81/F
811F
821M
961F
821F
87/M
941M
98/M
81/F
88/M
891F
80/M
83/M
911F
85/M
86IF
80/F
80/F
87/M
90/F
89/F
801F
85/F
85/M
89/M
88/F
81/M
931F
84/F
751F
761F
811F
68/M
Apl-40
Insoluble Fraction
Apx-42”
Apx-40”
Apl-42
<4
<4
18
<4
<4
23
430
6.1
0.9
8.5
2.1
1.4
26
2.3
4,250
60
<4
<4
6.6
<4
27
9.6
6,500
54
0.49
<4
<4
0.7
45
43
1.8
<4
8.5
3.1
130
53
2.4
<4
650
1,300
<4
1.3
2
<4
2.9
6.7
42
56
120
190
7.5
9.6
<4
2.5
14
18
140
56
11
7.2
1,100
770
130
93
57
75
9,000
7,000
32,000
33,000
61
63
700
830
6,300
5,100
680
1,100
14,000
8,500
9,200
6,250
160
59
11
30
7,600
4,125
12,000
6,100
1.7
12
900
3.4
12
21
15
1,300
0.7
2.4
54
100
25
230
2
210
540
180
620
120
180
126
4,800
4,250
1,750
2,100
3,400
880
380
20
4
580
610
27
50
2,000
16
42
75
41
5,500
<4
9.4
110
237
170
590
<4
1,050
1,600
330
2,200
400
590
1,200
7,600
3,000
2,800
6,900
8,800
2,900
1,000
150
22
2,400
2,900
Apl-40
Apx-4Ob
Apl-42
Apx-42‘
Stage“
of CAA
-
1.3
0.8
12
1.5
1.4
27
500
5.6
16
62
<80
<80
25
40
<80
<80
410
3,100
76
24
140
320
2,100
<80
<80
280
42
70,000
770
700
930
80
500,000
1,700
1,600
1,430
56
69
1,700
400
4,300
20
130
80
390
690
380
48
70
660
280
17,500
590
1,200
2,300,000
980,000
850
18,000
<80
148
2,400
400
9,000
<80
<80
<80
660
660
260
<80
<80
2,900
220
14,000
1,900
1,100
2,400,000
9 10,000
1,250
23,000
76
1,200
790
520
19,000
22
26
1,200
2,400
710
6,000
26
7,300
7,200
8,000
175,000
3,300
6,000
48,000
130,000
500,000
>500,000
1,000
4,100
4,200
950
140,000
<80
80
4,200
8,800
3,000
9,800
<80
30,000
33,000
10,000
400,000
15,000
15,000
180,000
310,000
500,000
>500,000
8,200
630,000
800,000
2,400
56
35,000
140,000
14,000
360,000
750,000
1,800
148
45,000
540,000
55,000
89,000
70,000
200
68
9,000
92,000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
2
2
2
3
3
3
3
3
4
4
4
210,000 4
240,000 4
210,000 AD
2,350 AD
230 AD
85,000 AD
490,000 AD
-
”The values below 4 pmol/g wet weight were determined at the low end of the standard curve and may not be very accurate.
“The values below 0.2 pmol/g wet weight were determined at the low end of the standard curve and may not be very accurate.
‘The degree of CAA was rated according to the number of vascular amyloid deposits along a given leptomeningeal length (-2 mm) in three
nonselected microscopic XlOO fields: 0, none; 1, 1-2; 2, 3-4; 3, 5-6; 4,>6/2 mm.
AP
=
amyloid P-protein; CAA = cerebral amyloid angiopathy.
lidity of the two-site EIA; there could be some proteins
in the leptomeninges cross-reactive with the antibodies
used in this study, thus providing unusually large false
values. However, this appears to be unlikely because
(1) the values obtained using both BANSO- and 4G8based EIAs (having different specificities) correlate very
well with each other in the leptomeningeal samples (see
Materials and Methods); and (2) the EIA data were in
parallel with those obtained by western blot quantita-
tion. The other concern would be that, as a more accurate denominator, one should use the protein
amount in the tissue rather than the wet weight, which
may cause a relatively large error. However, the AP
amounts expressed on a protein basis provided the
same relative abundance of AP in leptomeninges (see
Materials and Methods). Thus, the present data
strongly suggest that AP deposition in the leptomeninges occurs to an extent similar to that in the cortex.
Shinkai et al: AP Deposition in Leptomeninges and Cortex 905
Table 2. Amyloid /%Protein Deposition in Cerebral Cortices
Soluble Fraction
Case
No.
1
2
3
4
5
6
25
33
34
7
8
9
28
29
10
11
12
13
20
30
35
14
15
16
17
21
22
38
39
40
41
42
Insoluble Fraction
Aee
&)/Sex
Apl-40
82lF
86lF
841F
89lM
811F
80lF
86lF
851M
89lM
821F
83/F
81lM
87lM
901F
811F
811F
82lM
96lF
891F
891F
881F
811F
87lM
94lM
98lM
801M
831M
84lF
751F
76lF
811F
68lM
0.04
0.05
0.1
0.12
0.1
<0.04
0.3
0.83
3.9
11
0.2
2.8
1.6
5.3
0.56
0.25
2.2
6.8
0.25
0.91
40
4.5
2.3
0.36
17
1.7
1
175
2.7
0.43
69
4.5
Apx-40“
Apl-42
Apx-42”
Apl-40
Apx-40b
<0.2
<0.2
<0.2
0.35
0.2
<0.2
0.68
<0.04
<0.04
<0.04
<0.04
C0.04
C0.04
C0.04
6.4
3.6
3.8
0.11
7.8
1.6
4.5
0.09
1.7
1.7
5.1
0.41
5.9
6.7
<0.2
<0.2
<0.2
0.4
0.5
<0.2
<0.2
3.6
7.1
13
0.35
16
3.1
9
0.43
10
8.8
30
0.65
16
11
38
3.3
0.38
50
30
8.9
9.8
38
1.4
24
4.7
<16
(16
<16
<16
< 16
< 16
<16
<16
110
44
< 16
22
< 16
46
64
<I6
28
< 16
<16
25
5,200
38
48
260
360
38
45
12,500
240
2,300
740
1,300
<80
1.1
6.6
9.2
0.35
3.7
3.8
6.4
1.5
1.3
11
12
0.55
2.8
67
9.3
8.4
0.93
159
3.7
2.6
300
18
1.1
0.15
14
12.5
4
0.38
2
0.25
3.5
0.7
13
2.3
110
18
<80
<80
<80
<80
<80
<80
<80
200
<80
<80
<80
<80
<80
128
<80
<80
<80
<80
<80
3,700
<80
210
170
1,000
95
180
13,000
1,130
1,800
1,680
1,600
Apl-42
30
<16
44
<16
<16
<16
70
670
1,100
460
160
2,900
450
5,000
920
1,400
880
830
1,200
3,100
9,500
>15,000
3,300
240
4,500
3,600
5,000
5,600
2,300
7,900
1,900
15,000
Stage‘
A p ~ 4 2 ~of SPs
<80
<80
<80
<80
<so
<80
<80
520
2,300
1,200
240
6,800
2,000
17,000
2,300
15,000
10,500
7,400
4,200
19,000
45,000
>100,000
19,000
400
30,000
12,000
34,000
39,000
90,000
37,000
33,000
95,000
0
0
0
0
0
0
0
0
0
1
1
1
1
1
2
2
2
2
2
2
2
3
3
3
3
3
3
AD
AD
AD
AD
AD
“The values below 80 pniol/g wet weight were determined at the low end of the standard curve and may not be very accurate.
‘The values below 80 pniol/g wet weight were determined ar rhe low end of rhe standard curve and may not be very accurate.
T h e numbers of SPs were counted and averaged in three nonsekcted microscopic X 100 fields (-3.14 mm’). The abundance of SPs was rated
as follows: 0, none; 1, 1-50; 2, 51-100; 3, >100/3.14 mm2.
AP
=
amyloid P-protein; SPs
=
senile plaques.
One may argue that the unusually high levels of AP
detected are attributed to the area sampled, because the
occipital lobe is known as the area rich in CAA. It is
possible that the present data represent the characteristics specific for occipital leptomeninges and that those
of the other sites may be quite different; such high levels of AP may not be found. However, leptomeninges
at other sites showed, depending on CAA scores, similar levels of soluble and insoluble AP, as observed in
the occipital samples [7, 261. Thus, this strongly suggests that the accumulation of large amounts of AP is
not specific for the occipital leptomeninges but is characteristic of Ieptomeninges throughout the cortex.
The leptomeninges are defined here as a complex of
arachnoid mater and leptomeningeal vessels. If leptomeningeal vessels alone accumulate A@, as commonly
thought based on immunocytochemistry, they should
accumulate severalfold more A(3s on a weight basis. In
906 Annals of Neurology Vol 42
No 6
December 1997
this regard, our recent observations are particularly intriguing; leptomeningeal layers from which vessels are
carefully dissected out contain soluble and insoluble
APs at levels comparable with those in the accompanying vessels on a weight basis [26]. Thus, it is likely
that the leptomeningeal layers themselves are a large
reservoir of APs thus far unrecognized [26].
The extent of modifications in the amino terminus
of AP42 in insoluble fractions of the cortex correlated
with the abundance of SPs, the more AP42 is modified, the more advanced is SP formation (see Fig 2). A
straightforward interpretation of this finding is that the
extent of modifications represents the “in vivo aging”
of AP42 in SPs in the cortex. The highest modifications seen in AD cortices (and leptomeninges) may indicate that AP deposition started much earlier in AD
brains than in elderly control brains and that AP thus
underwent higher degrees of the processing. In general,
in any fraction, AP42 was modified to a higher degree
than AP40. This may be consistent with the view that
AP42 deposition precedes and/or dominates AP40
deposition.
At present, we do not know from which cellular
compartments soluble and insoluble APs originate. Soluble APs may simply be generated due to mechanical
disruption of amyloid aggregates in the tissue and subsequent soiubilization into the buffer during homogenization. Vascular amyloid fibrils are known to have
somewhat different structures [27] and to be more soluble than SP cores [24]. An alternate possibility is that
soluble APs, in particular those in leptomeninges, may
come from the cytosolic compartment and/or from an
extracellular nonfibrillar reservoir of APs [9, 281. Insoluble APs may represent immunocytochemically detectable AP deposits in the cortex (SPs) or those in the
leptomeningeal vessels (CAA), as suggested by the correlations between the amounts of APs and these lesions
(see Fig 1C and D).
However, it should be noted that some SP-free or
CAA-free samples contain high levels of soluble and
insoluble APs (see Fig 1B and D). Although we cannot
exclude the possibility that the cortical areas in which
APs were biochemically quantitated actually have a
very small number of SPs, and those in which AP deposits were immunocytochemically examined do not,
the presence of several such cases having no SPs in adjacent sections, but containing high levels of APs,
points to the possibility that up to a certain stage APs
accumulate in the tissue without forming immunocytochemically detectable SPs, just as previously suggested in leptomeningeal AP deposition [7].
This assumption may also be consistent with the recent
observation that brains of neonatal and young Down’s
syndrome patients, apparently free from any pathology,
contain unusually high levels of soluble AP [29]. Thus,
in view of much higher levels of insoluble AP in leptomeninges and cortex, at present, the most important
question to be answered is, Which comes first, soluble
AP or insoluble AS, in the period when no pathologies
are found?
This study was supported by a Grant-in-Aid for Scientific Research
from the Ministry of Education, Science and Culture (07408025)
and by a Grant-in-Aid for Scientific Research from the Ministry of
Health and Welfare, Japan.
We are grateful to D. J. Selkoe for providing AD brains used in the
present study, J. Saishoji for cechnical assistance, C. Hamada for
statistical analysis, and M. Anzai for typing the manuscript.
References
1 . Selkoe DJ. Alzheimer’s disease: a cencral role for amyloid.
J Neuropathol Exp Neurol 1994;53:438-447
2. Vinters HV. Cerebral amyloid angiopathy. A critical review.
Stroke 1987;18:311-324
3. Kang J, Lemaire H-G, Unterbeck A, et al. The precursor of
Alzheimer’s disease amyloid A4 protein resembles a cell-surface
receptor. Nature 1987;325:733-736
4 . Mori H, Takio K, Ogawara M, et al. Mass spectrometry of
purified amyloid P protein in Alzheimer’s disease. J Biol Chem
1992;267:17082-17086
5. Roher AE, Lowenson JD, Clarke S, ec al. Structural alteration
in the peptide backbone of P-amyloid core protein may account
for its deposition and stability in Alzheimer’s disease. J Biol
Chem 1993;268:3072-3083
6. Jarrett JT, Berger EP, Lansbury PT Jr. The carboxyl terminus
of the 6 amyloid protein is critical for the seeding of amyloid
formation: implications for the pathogenesis of Alzheimer’s disease. Biochemistry 1993;32:4693-4697
7. Shinkai Y, Yoshimura M , Ito Y, er al. Amyloid 6-proteins 1-40
and 1-42(43) in the soluble fraction of extra- and intracranial
blood vessels. Ann Neurol 1995;38:421-428
8. Khachaturian 2s.Diagnosis of Alzheimer’s disease. Arch Neurol 1985;42:1097-1104
9. Suzuki N , Iwatsubo T, Odaka A, et al. High tissue content of
soluble 61-40 is linked to cerebral amyloid angiopathy. Am J
Pathol 1994;145:452-460
10. Suzuki N , Cheung TT, Cai X-D, et al. An increased percentage
of long amyloid P-protein is secreted by familial amyloid
P-protein precursor (PAPP717) mutants. Science 1994;264:
1336 -1340
1 1 . Kim KS, Miller DL, Sapienza VJ, et al. Production and characterization of monoclonal antibodies reactive to synthetic cerebrovascular amyloid peptide. Neurosci Res Commun 1988;2:
121-130
12. Gravina SA, H o L, Eckman CB, et al. Amyloid p protein (A@)
in Alzheimer’s disease brain. Biochemical and immunocytochemical analysis with ancibodies specific for forms ending at
AP40 or AP42(43). J Biol Chem 1995;270:7013-7016
13. Iwatsubo T, Odaka A, Suzuki N, ec al. Visualization of
AP42(43) and AP40 in senile plaques with end-specific A6
monoclonals: evidence that an initially deposited species is
AP42(43). Neuron 1994;13:45-53
14. Yanagisawa K, Odaka A, Suzuki N, et al. G M l gangliosidebound amyloid P-protein (AP):a possible form of preamyloid
in Alzheimer’s disease. Nat Med 1995;1:1062-1066
15. Iizuka T , Shoji M, Harigaya Y, et al. Amyloid P-protein ending
at Thr43 is a minor component of some diffuse plaques in the
Alzheimer’s disease brain, but is not found in cerebrovascular
amyloid. Brain Res 1995;702:275-278
16. Pirttila T, Kim KS, Mehta PD, et al. Soluble amyloid 6-protein
in the cerebrospinal fluid from patients with Alzheimer’s disease, vascular dementia and controls. J Neurol Sci 1994;127:
90-95
17. Ida N , Hartmann T , Pantel J, et al. Analysis of heterogeneous
PA4 peptides in human cerebrospinal fluid and blood by a
newly developed sensitive western blot assay. J Biol Chem
1996;271:22908-229 14
18. Naslund J, Thyberg J, Tjernberg LO, et al. Characterization of
stable complexes involving apolipoprotein E and the amyloid 6
peptide in Alzheimer’s disease brain. Neuron 1995;15:219-228
19. Miller DL, Papayannopoulos IA, Styles J, et al. Peptide compositions of the cerebrovascular and senile plaque deposits of
Alzheimer’s disease. Arch Biochem Biophys 1993;301:41-52
20. Tamaoka A, Odaka A, Ishibashi Y, et al. APP 717 missense
mutation affects the ratio of amyloid P protein species (AP142(43) and A p l - 4 0 ) in familial Alzheimer’s disease brain.
J Biol Chem 1995;269:32721-32724
21. Naslund J, Schierhorn A, Hellman U, et al. Relacive abundance
of Alzheimer AP amyloid peptide variancs in Alzheimer disease
and normal aging. Proc Natl Acad Sci USA 1994;91:83788382
Shinkai et al: AP Deposition in Leptomeninges and Cortex
907
22. Frucht SJ, Koo EH. P-Amyloid protein is higher in Alzheimer’s
disease brains: description of a quantitative biochemical assay.
J Neuropathol Exp Neurol 1993;52:640-647
23. Iwatsubo T, Mann DMA, Odaka A, er al. Amyloid p-protein
(A@) deposition: Ap42(43) precedes Ap40 in Down’s syndrome. Ann Neurol 1995;37:294-299
24. Joachim CL, Duffy LK, Morris J H , Selkoe DJ. Protein chemical and immunocytochemical scudies of meningovascular pamyloid protein in Alzheimer’s disease and normal aging. Brain
Res 1988;474:100-111
25. Roher AE, Lowenson JD, Clarke S, et al. p-Amyloid-(l-42) is
a major component of cerebrovascular amyloid deposits: implication for the pathology of Alzheimer’s disease. Proc Natl Acad
Sci USA 1993;90:10836-10840
908
Annals of Neurology
Vol 42
No 6
December 1997
26. Hamano T, Shinkai Y, Yanagisawa K, et al. Arnyloid P-protein
(Ap) accumulation in the leptomeninges during aging and in
Alzheimer’s disease. J Neuropathol Exp Neurol 1997;56:922932
27. Kanernaru K, Hasegawa M, Shimada H, Ihara Y. The presence
of a novel protein in calf serum that recognizes (3-amyloid in
the formalin-fixed section. Am J Pathol 1990;137:677-687
28. Frackowiak J, Zoltowska A, Wisniewski HM. Non-fibrillar pamyloid protein is associated with smooth muscle cells of vessel
walls in Alzheimer disease. J Neuropathol Exp Neurol 1994;53:
637-645
29. Teller JK, Russo C, DeBusk LM, et al. Presence of soluble
amyloid p-peptide precede amyloid plaque formation in
Down’s syndrome. Nar Med 1996;2:93-95
Документ
Категория
Без категории
Просмотров
0
Размер файла
950 Кб
Теги
deposition, leptomeningeal, protein, amyloid, corte, cerebral
1/--страниц
Пожаловаться на содержимое документа