| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Short Communication |
From the Department of Neuropathology,*
Faculty of
Medicine, University of Tokyo, the Japan Society for the Promotion of
Science,
the Department of
Neuropathology,
Medical Research Institute,
Tokyo Medical and Dental University, and the Tokyo Medical Examiner's
Office,§
Tokyo, and Core Research for
Evolutional Science and Technology (CREST),¶
Japan Science and Technology Corporation, Kawaguchi, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The hypothesis is supported by several findings: 1) in the brains from Down's syndrome patients of various ages, the earliest lesion in the neocortex is found to be diffuse plaques consisting exclusively of Aß42,5 2) rare forms of familial AD are caused by mutations of ß-amyloid precursor protein (APP),1 and 3) mutations of presenilin 1 and 2 as well as of APP cause enhanced secretion of Aß42.1,2 Along this line of investigation, there is now a large collection of reports describing the neurotoxicity of Aß. Its neurotoxicity is observed in vitro under well controlled conditions; fibrillar, but not amorphous, Aß aggregates are believed to exert toxic effects on cultured neurons.6,7 These toxic effects may be mediated through oxidative stress,8 although other routes of mediation are also possible.
However, there are some important, but often neglected, observations refuting in vivo neurotoxicity of Aß. The most remarkable is that in the hippocampus CA1 and entorhinal cortex, a great number of NFTs often occur without SPs during normal aging9-12 and presumably in the initial stage of AD,13 as has been repeatedly confirmed by several groups using silver staining or immunocytochemistry. In addition, two lines of transgenic mice exhibiting Aß deposits did not show NFT formation or neuronal loss, suggesting that Aß accumulation detected as extracellular Aß deposits is not sufficient for NFT formation or neuronal loss.14,15 These findings led us to believe that the amyloid cascade hypothesis of AD is not valid in CA1 and thus may not be so in the neocortex. However, neuropathological findings alone cannot invalidate the amyloid cascade hypothesis, because one cannot exclude the possibility that Aß can accumulate to significant levels in CA1 without forming immunocytochemically detectable SPs, and that this invisible but significant accumulation could lead to NFT formation in a subset of neurons and/or their loss in CA1.
In this study, we sought to examine 1) whether NFTs appear in CA1/T4 (medial occipitotemporal cortex) without the accumulation of SDS-dissociable Aß42 as quantitated by the sensitive two-site enzyme immunoassay (EIA)16 and 2) whether there is a similar temporal relationship of NFT with SDS-stable Aß dimers that are detected by Western blotting but not EIA, using specimens obtained from consecutive autopsy cases with ages ranging from 24 to 92 years.17
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The present study is based on autopsies performed (n = 74; 56 men and 18 women) at the Tokyo Medical Examiner's Office (Otsuka, Tokyo).17 The ages at death of the 74 subjects ranged from 24 to 92 years (3 at 20 to 29 years of age, 4 at 30 to 39 years of age, 17 at 40 to 49 years of age, 18 at 50 to 59 years of age, 13 at 60 to 69 years of age, 10 at 70 to 79 years of age, 8 at 80 to 89 years of age, and 1 at 92 years of age). Postmortem delay ranged from 2 to 24 hours. AD was diagnosed based on both clinical and neuropathological criteria.18,19
Tissue Preparation and Extraction
Cortical pieces, ~80 to 110 mg each, of CA1 and T4 at the level of the lateral geniculate body were sampled from fresh brains at autopsy and stored at -80°C until use. The attached leptomeninges and vessels were carefully dissected out. Cortical blocks from adjacent sites and/or from the same locations on the contralateral side were fixed in 10% buffered formalin and processed for histological and immunocytochemical examination.
Each of the sampled pieces was homogenized as described elsewhere.17 Five microliters from each homogenate was smeared on polylysine-coated glass slides for tau immunostaining. The remaining part of each homogenate was further homogenized and centrifuged at 100,000 x g for 15 minutes. The supernatant was used for quantitation of Aß (soluble Aß); the pellet (insoluble fraction) was homogenized in 70% formic acid and the resultant suspension was centrifuged as described above. The supernatant was neutralized with NaOH and trizma base and subjected to the EIA (insoluble Aß).20
Enzyme Immunoassay
The two-site EIA for Aß consisted of Aß monoclonal antibodies BNT77, BA27, and BC05. BNT77 was used as the capture antibody, and BA27 and BC05 were used as the detector antibodies.20 BNT77, the epitope of which is thought to be located in Aß11-16, is considered to capture all Aß species truncated up to position 10, but not p3, which starts at Aß17.20 The BNT77-based EIA employed in this study detects only Aß species that run as monomers on SDS-polyacrylamide gel electrophoresis.16
Western Blotting
Each aliquot of the formic acid extract of insoluble fraction derived from 2.5 mg of tissue was dried by Speed Vac (Savant Instruments, Framingdale, NY) and solubilized with the SDS sample buffer (50 mmol/L Tris HCl (pH 6.8), 12% glycerol, 2% SDS, 2.5% mercaptoethanol, 4 mol/L urea). These samples were subjected to Tris/tricine gel electrophoresis, and the separated proteins were blotted onto a nitrocellulose membrane (pore size 0.22 µm; Schleicher & Schuell, Dassel, Germany). The blot, after heat treatment,21 was incubated with appropriately diluted BA27 or BC05. After washing with Tris/saline-based buffer, the blot was further incubated with horseradish-peroxidase-conjugated goat anti-mouse IgG (Transduction Laboratories, Lexington, KY). Bound antibodies were visualized using the enhanced chemiluminescence (ECL) system (Amersham Pharmacia Biotech, Little Chalfont, UK). This modified version of Western blotting reproducibly detected as little as 20 pg (5 fmol) of Aß1-42 or Aß1-40 per lane, which is equal to 2 pmol/g wet weight.
Besides specimens, synthetic Aß1-40 or 1-42 (10, 20, 50, and 100 pg; Bachem, Torrance, CA) was also loaded onto each gel for Western blot quantitation of Aß. SDS-stable Aß dimers were quantitated using a standard curve for synthetic Aß40 or 42 (SDS-dissociable Aß40 or 42), and the concentration was expressed as the Aß40 or 42 monomer equivalent. Thus, it was postulated that the blotting efficiency and reactivity of SDS-stable dimers with BA27 or BC05 are the same as those of SDS-dissociable Aß40 or 42.16 It should be also noted that Aß oligomers other than Aß dimers are not taken into account.
Semiquantitative Immunocytochemistry for the Abundance of NFTs
The formalin-fixed, paraffin-embedded tissue blocks were cut into 6-µm-thick sections. These sections and smeared aliquots from the brain homogenates were immunostained with anti-human tau by the avidin-biotin method (Vectastain Elite, Vector Laboratories, Burlingame, CA). The abundance of NFTs was assessed by manually counting them on anti-human tau-immunostained sections in five unselected fields under x200 magnification, and the numbers were averaged. The abundance was rated as follows: -, none; +, 1 to 5; ++, 6 to 10; +++, more than 10 per one x200 field. The NFT abundance estimated on smears excellently agreed with that obtained from immunostained tissue sections.
Apolipoprotein E Genotyping
Typing of the apolipoprotein E (ApoE) genotype was performed using the polymerase chain reaction (PCR) as described previously.17
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the present series, there were 27 cases showing NFTs but no SPs
in CA1, as judged by immunocytochemistry, a result that is in good
agreement with the data reported by other groups.9-13
In
contrast, there was no case showing SPs but not NFTs in CA1. We
examined whether there was a temporal relationship between the
appearance of NFTs and the accumulation of SDS-dissociable Aß that is
quantitated by BNT77-based EIA.16
Consistent with the
immunocytochemical observations, NFTs occurred without accumulation of
SDS-dissociable Aß42 in the insoluble fraction of CA1 homogenates in
22 cases (Figure 1, A and B)
. Because EIA
is approximately 30-fold more sensitive than immunocytochemical
detection of SPs,17
it seemed that the EIA quantitation may
have further confirmed the immunocytochemical data, that the formation
of NFTs precedes that of SPs in CA1.
|
;
Mann-Whitney's U test, not significant). Likewise, the appearance of
NFTs was independent of the levels of SDS-dissociable Aß42 in the
soluble fraction or those of SDS-dissociable Aß40 in the soluble and
insoluble fractions of CA1 homogenates (data not shown). However, the
formation of NFTs appeared to be related to the age at death of the
subjects: none of the patients (24 cases) who died under the age of 50
years exhibited NFTs in CA1. For ages of 50 to 64 years and the age of
65 years and above, the incidence of NFTs increased to 9 of 22 (41%)
and 25 of 28 (89%), respectively. NFTs were found in 22 EIA-negative
cases in which SDS-dissociable Aß42 levels were below the detection
limit (<12 pmol/g; Figure 1A
). Accumulation of SDS-Dissociable Aß42 Precedes the Appearance of NFTs in T4
In contrast to CA1, there were only a small number of cases
showing NFT formation in T4 (Figure 1, C and D)
. There were nine cases
showing SPs but no NFTs, whereas three cases showed NFTs but no SPs in
T4 (data not shown). Twenty-one cases contained detectable levels of
SDS-dissociable Aß42, but no NFTs (Figure 1D)
. The incidence of NFTs
was significantly correlated with the extent of Aß42 accumulation in
T4 (Figure 1D
; Mann-Whitney's U test, P < 0.05).
Furthermore, there was no apparent relationship between the incidence
of NFTs and the age of death of the subjects in T4 (Figure 1C)
.
Presence of SDS-Stable Aß Dimers in the Presence or Absence of NFTs in CA1
We further assessed by an improved version of Western
blotting21
the levels of SDS-stable Aß dimer that cannot
be detected by EIA.16
Representative Western blots of CA1
specimens with BC05, together with BNT77-based EIA values and
immunocytochemical findings for SPs and NFTs, are given in Figure 2
. Lanes 1, 2, and 3 are CA1 specimens
from a 59-year-old man (ApoE genotype: 
3/3), a 61-year-old man
(2/3), and a 56-year-old man (3/3), respectively, in all of
whom accumulation of Aß42 dimers but not SDS-dissociable Aß was
revealed by Western blotting, whereas negligible levels of
SDS-dissociable Aß42 were detected by EIA. Whereas the first two were
free from NFTs, the last showed some NFT formation.
|
.
Eight cases showed accumulation of SDS-stable Aß dimers (Aß42
dimers, 4 cases; Aß40 dimers, 3 cases; both Aß42 and Aß40 dimers,
1 case) but no NFTs. Fifteen cases showed neither SDS-stable Aß
dimers nor NFTs. Only one case showed NFT formation without SDS-stable
Aß dimer accumulation. Thus, SDS-stable Aß dimers accumulated
without NFT formation in CA1 (Figure 3
, Fisher's exact method,
P < 0.01), suggesting that SDS-stable Aß dimers
accumulated before NFT formation.
|
;
lanes 5 and 6 are specimens from a 44-year-old man (3/3) and
another 44-year-old man (3/3), respectively, both of whom
exhibited no SPs. One showed no accumulation of Aß monomers or Aß
dimers (lane 5), whereas the other showed accumulation of Aß dimers
but not Aß monomers (lane 6). The third case (66-year-old man;
3/3) showed significant accumulation of both dissociable Aß42
and SDS-stable Aß42 dimers (lane 7) and SPs but not NFTs.
Among 26 cases showing negligible levels of dissociable Aß by EIA in
T4, 17 showed accumulation of SDS-stable Aß dimers (Aß42 dimers, 4
cases; Aß40 dimers, 9 cases; both Aß42 and Aß40 dimers, 4 cases)
but no NFTs (Figure 3)
.16
There was no statistically
significant difference in the incidence of SDS-stable dimers between
CA1 and T4 (
2
test). Eight cases showed neither
accumulation of SDS-stable Aß dimers nor NFTs (Figure 3)
. Only one
case showed both SDS-stable Aß dimer accumulation (Aß42 dimers) and
NFTs (Figure 3)
. Thus, SDS-stable Aß dimers accumulated without
accompanying NFTs in T4, as in CA1.
With regard to the effect of ApoE alleles on NFT formation, 9 of 13
3/4 carriers showed NFTs in CA1, whereas 5 of them showed NFTs in
T4. Of 58 3/3 carriers, 25 showed NFTs in CA1, and 6 of them
showed NFTs in T4. There was a higher incidence of NFT in T4 of
3/4 carriers than 3/3 carriers (Fisher's exact method,
P < 0.05, data not shown), but not in CA1 (Fisher's
exact method, P = 0.13, data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because of the undetectability of SDS-stable Aß dimers by the most commonly used EIA,16 we undertook the assessment of Aß monomers (SDS-dissociable Aß) and dimers in these specimens by an improved version of Western blotting. Most unexpectedly, many EIA-negative specimens showed some accumulation of SDS-stable Aß40 or Aß42 dimers but not of SDS-dissociable Aß42.16 This may suggest that SDS-stable Aß dimers appear earlier than SDS-dissociable Aß42 in CA1 and T4.16 These SDS-stable Aß dimers were detected both in CA1 and T4 in approximately one-half of the cases at the age of 40 to 60 years showing negligible Aß42 levels by EIA.16 Thus, it is likely that a much larger amount of Aß dimers relative to SDS-dissociable Aß42 exists in the earliest stage of ß-amyloidogenesis.
There were significant differences in the extent of SDS-dissociable
Aß42 accumulation between CA1 and T4; even when certain cases showed
significant accumulation of SDS-dissociable Aß42 in T4, they often
showed negligible levels of such Aß42 in CA1.17
Thus, it
was quite unexpected that the majority of CA1 specimens showing NFTs
contained Aß dimers, and further that many cases without NFTs also
contained some Aß dimers (Figure 3)
. The incidence of Aß
dimer-positive cases among EIA-negative cases in CA1 is similar to that
in T4 (not significantly different, 2
test). These
results suggest that 1) SDS-stable Aß dimers may accumulate earlier
than NFTs in CA1 (Figures 1 and 3)
and 2) such Aß dimers start to
accumulate in CA1 and T4 at a similar time, whereas accumulation of
SDS-dissociable Aß42 appears to be significantly delayed in CA1
compared with T4.17
Taken together, the temporal profile of
Aß accumulation and NFTs in CA1 may be as follows: SDS-stable Aß
dimers start to accumulate, followed by NFTs, and SDS-dissociable Aß
accumulation follows, and SPs appear at the end of this sequence. In
contrast to CA1, there were many cases showing accumulation of
SDS-dissociable Aß42 and SPs but no NFTs in T4 (Figure 1)
. Thus, in
T4, SDS-stable Aß dimers may accumulate first, followed by
SDS-dissociable Aß and SPs. NFTs appear at the end of the sequence.
These findings suggest that CA1 may be particularly resistant to the
accumulation of SDS-dissociable Aß42 for unknown reasons and,
furthermore, that SDS-stable Aß dimers have a distinct metabolic
pathway from SDS-dissociable Aß in CA1.
Given that the mechanism underlying NFT formation during aging is the same in both CA1 and T4, SDS-stable Aß dimers may be a candidate. More than 2 pmol/g wet weight of SDS-stable Aß dimers (expressed by Aß monomer equivalent; see Materials and Methods) may have already accumulated in CA1 showing no NFTs. It is reported that the stratum pyramidale of CA1 exhibits an exceptionally low extracellular volume fraction of 0.12 in rat brain.24 If 1) this can be applied to human CA1, 2) the specific gravity of gray matter is assumed to be 1.0356,25 and 3) SDS-stable Aß dimers accumulate exclusively in the extracellular space, 17 nmol/L SDS-stable Aß dimers would accumulate in the extracellular space of CA1 showing no NFTs. Thus, the concentrations of SDS-stable Aß dimers in CA1 without NFTs are comparable to those employed for in vitro neurotoxicity experiments; diffusible, nonfibrillar Aß1-42 oligomers exert neurotoxicity at nanomolar concentrations through an as yet unidentified cell surface receptor and the activation of fyn.26 One of the most notable characteristics of SDS-stable Aß dimers or oligomers is its inability to form fibrils.26,27 This presents a sharp contrast to SDS-dissociable Aß molecules that easily polymerize into fibrils and the amounts are well correlated with amyloid burden.17 SDS-dissociable Aß may be essential for SP formation, but not NFT formation.
Currently, we do not know whether the detected SDS-stable Aß dimers exist in the extracellular space as postulated above or represent the Aß species in the intracellular compartment. It is possible that these dimers are generated from secreted SDS-dissociable Aß in the extracellular space.28-30 Another possibility is that SDS-stable Aß dimers accumulate in a certain intracellular compartment of neurons. Consistent with this, SDS-stable Aß dimers appear to be located in the detergent-insoluble fraction of neuroblastoma cells.31 Furthermore, our preliminary data show that the detergent-insoluble, low-density membrane fraction obtained from human brains contains these Aß dimers. Thus, it is also possible that SDS-stable Aß dimers detected in EIA-negative specimens represent the Aß species bound to this particular membrane domain of neural cells.
The origin of the dimers and putative oligomers (see Figure 2
) remains
to be elucidated. SDS-stable Aß dimers migrate usually at ~6 kd, a
little faster than the dimers generated from synthetic Aß1-40 or
1-42. A fraction of the dimers is labeled with BAN50 (the epitope in
Aß1-10), suggesting that their amino termini are not largely deleted.
Currently we are not certain as to whether faster migration of the
SDS-stable dimers (Figure 2)
represent amino-terminal truncation or an
aberrant conformation by misfolding, as suggested by
others.30
We also do not know whether these dimers and
oligomers are covalently linked or noncovalently linked. These
molecules are not dissociated into monomers with SDS or other harsh
denaturants, including urea and guanidine hydrochloride, raising the
possibility of nondisulfide cross-linking. However, these properties do
not necessarily indicate that these molecules are covalently linked.
For example, the microtubule-binding domain of the tau in paired
helical filaments shows similar aggregation that is highly resistant to
SDS and guanidine hydrochloride, but appears not to be covalently
linked.32
In this case, deamidation and isoaspartate
formation in the selected Asn and Asp residues likely have a
significant role in SDS-stable oligomerization.32
Possibly,
similar post-translational modifications may be at work for the
formation of SDS-stable Aß dimers and oligomers.
One important question remains: which of the Aß dimers, Aß40 or 42, is more important in the initial phase of Aß dimer deposition? We cannot completely exclude the possibility that Aß40 dimers detected in autopsied brains are generated from Aß42 dimers by partial proteolysis during the postmortem period. This may be possible because among brains with short postmortem delays there was no Aß40 dimer-only brain.16 However, to confirm this, careful examination of many autopsied cases with short postmortem delays is required. From the previous data alone, the appearance of Aß42 dimers, but not Aß40 dimers, is age dependent in CA1 and T4.16 This raises the possibility that Aß42 dimers are the species mainly involved in the deposition, although we cannot exclude the possibility that Aß40 dimers have some important role in the initial stage of Aß deposition.
In summary, SDS-stable Aß dimers may be a missing link that can resolve an apparent contradiction of the amyloid cascade hypothesis in CA1.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Accepted for publication March 20, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. J. Garzon and M. Fahnestock Oligomeric Amyloid Decreases Basal Levels of Brain-Derived Neurotrophic factor (BDNF) mRNA via Specific Downregulation of BDNF Transcripts IV and V in Differentiated Human Neuroblastoma Cells J. Neurosci., March 7, 2007; 27(10): 2628 - 2635. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. H. Chong, Y. J. Shin, E. O. Lee, R. Kayed, C. G. Glabe, and A. J. Tenner ERK1/2 Activation Mediates Abeta Oligomer-induced Neurotoxicity via Caspase-3 Activation and Tau Cleavage in Rat Organotypic Hippocampal Slice Cultures J. Biol. Chem., July 21, 2006; 281(29): 20315 - 20325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Urbanc, L. Cruz, F. Ding, D. Sammond, S. Khare, S. V. Buldyrev, H. E. Stanley, and N. V. Dokholyan Molecular Dynamics Simulation of Amyloid {beta} Dimer Formation Biophys. J., October 1, 2004; 87(4): 2310 - 2321. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Delacourte, N. Sergeant, D. Champain, A. Wattez, C.-A. Maurage, F. Lebert, F. Pasquier, and J.-P. David Nonoverlapping but synergetic tau and APP pathologies in sporadic Alzheimer's disease Neurology, August 13, 2002; 59(3): 398 - 407. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and 29 cases
did (
) in CA1
(A); and 63 cases did not show NFTs
(
, non-AD cases
in B and D. P < 0.05 for indicated
comparisons in B and D.
