вход по аккаунту


Axonal disruption and aberrant localization of tau protein characterize the neuropil pathology of Alzheimer's disease.

код для вставкиСкачать
Axonal Disruption and Aberrant
Localrzation of Tau Protein Characterize the
Neuropil Pathology of Alzheimer’s Disease
Neil W. Kowall, MD,” and Kenneth S. Kosik, MDt
The microtubule-associated protein tau, a major antigenic component of paired helical filaments, has been demonstrated in neurofibrillary tangles and in neurites of senile plaques. With optimal fixation and histochemical methods,
we show the normal axonal location of tau protein in human cerebral cortex and the striking alterations of tau
distribution that affect the cortical neuropil in Alzheimer’s disease. Normally, cortical tau-immunoreactive fiber
bundles form a pattern resembling that seen with myelin stains. The prominence of white matter staining suggests that
tau may be especially enriched in projection systems. Alzheimer’s disease causes massive axonal disruption and the
dislocation of tau protein from its usual axonal domain into neuronal cell bodies, dendrites, and presynaptic regions.
The normal pattern of axonal staining in cortex is disrupted and white matter staining is reduced. Prominent abnormal
tau-immunoreactive neuropil fibers are densely present even in cortical regions without classical neurofibrillary tangle
and senile plaque formation. The striking neuropil abnormalities, revealed by the aberrant localization of tau protein,
are likely to contribute to neuronal dysfunction in Alzheimer’s disease.
Kowall NW, Kosik KS: Axonal disruption and aberrant localization of tau protein characterize the
neuropil pathology of Alzheimer’s disease. Ann Neurol 22:639-643, 1987
Defining the molecular nature of senile plaques (SP)
and neurofibrillary tangles ( N R ) is a major goal of
research into Alzheimer’s disease (AD). The deposition of amyloid beta protein within the core of SP and
in cerebral vasculature, a phenomenon described in
both human and subhuman species {I), is postulated
to be of fundamental importance [2-61; however,
its uniquely human association with paired helical
filaments (PHF) remains to be explained. PHF have
previously been shown to contain the microtubuleassociated protein (MAP) tau as a major antigenic
component 17-12}. MAPs are thought to confer stability on microtubules so that their inherent dynamic
instability is skewed toward polymerization { 131. Discrete cellular and subcellular localizations have been
described for some MAPs, particularly those in the
nervous system. MAP2, one of the high-molecularweight MAPs, is restricted to the somatodendritic
compartment of neurons {14-181. Tau protein has a
complementary distribution in that it is present only
in axons {19-21). The sorting of microtubular populations into these highly segregated axonal and
somatodendritic domains is established early in development t22) and is maintained throughout the life of
the organism. The optimized fivation and immunocy-
tochemical techniques used here permit us to define
the distribution of tau both in normal human cerebral
cortex and in AD, in which the cortical neuropil is
disrupted by dystrophic neurites that are more prominent than the more fixation-resistant NET.
From the +Neurology Service, Massachusetts General Hospital, and
the tCenter for Neurologic Diseases, Brigham and Womens Hospital, Harvard Medical School, Boston, MA.
Received Mar 18, 1987, and in revised form May 19. Accepted for
publication May 20, 1987.
Postmortem specimens were obtained from 6 patients with
pathologically verified AD (mean age 72, range 61-81) and
6 age-matched controls (mean age 74, range 60-84). Cortical
blocks were selected from superior frontal (Brodmann Areas
{A] 24, 32, and 8), inferior frontal (A 10, 11, and 12),
rolandic (A 4 , 3, 1, and 2), temporal (A 28, 36, 22, 21, and
20), and occipital (A 17 and 18) regions. Tissue was fixed at
4°C for 24 to 36 hours in paraformaldehyde-lysine-periodate
fixative {23]. After cryoprotection in 20% sucrose, 50-c~
free-floating sections were stained using a monoclonal antibody (5E2) specific for tau on immunoblots 1241. Sections
were incubated overnight at room temperature in supernatant (clone 5E2) diluted l : 10 in 0. l M phosphate-buffered
saline (PBS), pH 7.4, containing 0.3% Triton X-100. After
three 10-minute washes in PBS, sections were incubated in
biotinylated horse antimouse antibody for 2 hours (Vector
Labs; 1:200 in PBS). After further PBS washes, sections
were placed in avidin-biotin peroxidase complex for 90 minutes (Vector labs; 1:lOO in PBS). Following more PBS
Address reprint requests to Dr. Kowall, Neurology Service, Massachusetts General Hospital, Boston, MA 02114.
Fig 1. Tau histochmistry in normul human cerebral cortex.
Deep layers of (A)primary motor cortex, (B) primary sensory
cortex, and (C) primary visual cortex showing dense tau staining
in white matter (asterisk) and radial bundles of tau-immunoreactive fibers. Axonal bundles are less distinct and thicker
in motor cortex than in sensoy or visual cortex. (A-C x 160
before 17% reduction.)(0)Layer IV is densely stained in primary visual cortex (arrows). ( X 40 before 17% reduction.)(E)
Scattered senile plaques were traversed 6.i radialIy oriented tauimmunoreactivefiber bundles in 2 control subjects (arrows).
( x 400 before 17% reduction.)
washes, the peroxidase complex was disclosed with 3,3’diaminobenzidine tetrahydrochloride 1251. Sections incubated in parallel without primary antibody failed to develop specific staining. Some sections were counterstained
with thioflavine S (to visualize amyloid senile plaque cores
under ultraviolet fluorescence) or cresyl violet (to delineate
cortical architecture) before mounting.
In normal human cerebral cortex we observed an
axonal pattern of tau immunoreactivity. Positively
640 Annals of Neurology Vol 22
No 5
stained fibers, 0.2 to 5 p, in diameter, formed discrete,
15- to 30-pthick vertical bundles that arose from the
densely immunoreactive white matter and coursed radially in a regular 30- to 50-p, array perpendicular to
the pial surface (Fig 1). The bundles were thicker and
more loosely packed in the motor cortex (Fig 1A) than
in the primary sensory and visual cortices (Fig lB, C).
They extended into Layer IV in all regions; in primary
visual cortex they penetrated to Layer I1 (Fig 1D). In
supragranular layers the bundles thinned out as individual fibers appeared either to terminate abruptly or
veer off on new trajectories. Horizontal fiber bands of
variable intensity were most prominent in Layers I and
IV. In Layer I they occasionally formed a dense mat of
large, often varicose, subpial fibers (Fig 2A, C ) . In
primary visual cortex, a 250-p,-thick band of dense
fiber staining was seen in Layer IV (line of Gennari,
see Fig 1D). This band was less prominent in other
cortical areas and. absent in motor cortex. The fiber
patterns seen with tau staining resemble the myeloarchitectonic axonal patterns described by Vogt using
November 1987
Fig 2. Tau histochemistry in normal and Alzheimer's disease
(AD) cortex. The normally sparse superficial tau pattern in
Layer I of Brodmann Area 8 (A)is replaced by a striking network of dystrophic neurites in A D (B). (A,B x 160 before
38% reduction.) SimiIurIy, the dense tau-reactive axon plexus in
Layw I of Area 20 (C) is total& disrupted and replaced Sy
distorted and abruptIy terminating neurites in A D (0).
neurofibrillary tangles (NFT) were contiguous with curb fibers
in Andrites (E) (arrowhead). Note the striking, dense background of abnormalfibers that obscures the NFT and senile
plaques (SP). SP contained tau-immunoreactive neurites ( F , arrows and with thiojavine S countwstainfor amyloid, G,
arrows). Some neurons in A D corttx contained diffuse tauimmunoreactive material (H) (arrow). (C-H x 400 before
38% reduction.)
myelin stains 1261. The preponderance of tau immunoreactivity in white matter and larger axons suggests that the axons of cortical projection neurons may
be particularly enriched in this protein. Neuronal cell
bodies or dendrites were not tau immunoreactive in
normal control subjects. In 2 control subjects occasional neurites within SP and rare N l T were seen in
the temporal cortex (see Fig 1E).
The distribution of tau immunoreactivity was strikingly altered in AD (see Fig 2B, D, E-H). A dense
pattern of dystrophic neurites, which often obscured
SP and NFT, was seen throughout the neuropil in all
cortical layers (see Fig 2E, F). Unlike normal immunoreactive fibers, these neurites were short, kinked,
and curled without any distinct orientation (see Fig 2B,
D-H). Neurite density was enhanced in Layers I11 and
V where NJT were prominent in pyramidal neurons
(see Fig 2E) but did not correlate with the presence of
SP. Neuropil abnormalities were a consistent feature
in patients with AD that were not seen in any of the
control subjects. Neurofibrillary tangles consisted of
perinuclear tau-reactive fibers that often extended into
proximal dendrites and occasionally terminated as fine
dystrophic neurites (see Fig 2E). Occasional cells in
AD-affected cortex contained diffuse tau-immunoreactive material (see Fig 2H). The nature of this material is currently being studied at the ultrastructural
level. The neuritic portion of the SP was also tau immunoreactive. Reactive filiform, bullrush-like neurites
were observed around the perimeter of amyloid cores
(see Fig 2F, G). The normal axonal tau staining pattern
was markedly diminished in AD cortex. The intensity
of white matter staining was reduced and the pattern of
cortical fiber bundles was either greatly diminished or
Kowall and Kosik: Tau Histochemistry in AD
eliminated. The vertical bundles that are normally
prominent in the infragranular cortical layers (see Fig
1) were replaced with dystrophic neuropil fibers (see
Fig 2). This pattern of axonal disruption contrasts with
the relative preservation of dendritic anatomy we observed in AD cortex using histochemical methods to
detect the distribution of MAP2 1251. The massive
axonal disruption in patients with AD did not permit a
clear determination of the relationship of SP to axonal
distribution. However in control subjects with few SP
and a normal axonal staining pattern, SP were located
within radially deployed axonal bundles (see Fig 1E).
Most immunoreactive fibers coursed uninterrupted
through the SP and others appeared to terminate in
the SP neurites. This observation complements the
finding of SP within the interfascicular region of apical
dendritic bundles. In contrast to the tau-immunoreactive axons, the apical dendritic shafts of pyramidal
cells deviate around SP E251. Dystrophic neuropil
neurites were not observed in the control subjects.
Classical histopathological descriptions of AD have
emphasized the occurrence of SP and NFT [271. More
widespread neuropil abnormhties have been reported
by others using silver stains and antibodies raised
against PHF [28, 291, but not to the extent that we
have found. Dystrophic cortical neurites are more
prominent and widespread than either SP or NET.
Improved fixation and histochemical methods may allow the detection of the lower concentrations of tau
that occur in neuropil fibers and normal axonal structures. Alternatively, various molecular configurations
of tau-immunoreactive structures may contribute to
their differential sensitivity to fixation. Preliminary
electron microscopic observations show that filamentous bundles within many dystrophic neurites are composed of straight .lo- to 15-nm filaments rather than
PHF, which may be more resistant to fixation effects
(Kosik KS, Kowall N W , unpublished observations).
Using the Golgi method, Probst and co-workers
{30) found long filiform dendritic outgrowths in the
vicinity of SP. We show here that tau-immunoreactive
neurites are present throughout the neuropil. For example, in the subpial zone of the molecular layer
(Layer I), where SP are sparse and NET do not occur,
the normal axonal staining pattern of tau is eliminated
and replaced by myriad short irregular neurites (see
Fig 2A-D). Whether these dystrophic neurites are retracted axons and/or highly dislocated tau in distal apical dendrites is not yet clear. The observed loss of
axonal architecture suggests that the disruption of microtubules containing tau may cause a more generalized collapse of cortical axonal systems, particularly
those of long projecting neurons. The concomitant
shift of tau immunoreactivity into highly aberrant re-
gions within the neuron may result in the accumulation
of filamentous aggregates that eventually become
PHF. These tau-immunoreactive inclusions are generally present in neiironal perikarya and the proximal
portion of apical dendrites, regions in which tau is normally not detectable. In this aberrant locale tau may
become an abnormal substrate for certain MA:
associated kinases [ 3 1, 32). The phosphorylated tau
protein may direcltly contribute to the formation of
PHF 133, 34).
Neurons in general, but particularly those with long
axons, require highly stable microtubule populations
to support transport over long time periods [351. To
the extent that tau. functions to stabilize microtubules
in these axons, its defective function or dislocation in
AD would lead to enhanced vulnerability of long projection systems. The breakdown of axonal microtubule
systems would lead to the multiple neurotransmitter
defects that have been described in A D {for example,
N. W. K. is supported by the MRC of Canada and NIA grant
AGO5 134, Massachusetts Alzheimer Disease Research Center. K.
S. K. is supported by Teacher Investigator award 5K07 NS00835
and NIA grant AGO61 72. We would like to thank Michael Perrotti
and Bernard Quigley fix expert technical assistance, Lawrence Cherkas for photography, and Robert J. Ferrante for reviewing the manuscript.
I. Selkoe DJ, Bell DS, Podlisny MB, et al. Conservation of brain
amyloid proteins in aged mammals and humans with Alzheimer’s disease. Ijcience 1987;235:873-877
2. St. George-Hyslop PH, Tanzi RE, Polinski RJ, et al. The genetic defect causing familial Alzheimer’s disease maps on
chromosome 21. Science 1987;235:885-890
3. Tanzi RE, Gusella.JM, Watkins PC, et al. Amyloid beta protein
gene: cDNA, mRNA distribution, and generic linkage near the
Alzheimer locus. Science 1987;235:880-884
4. Goldgaber D, Lerman MI, McBride OW, et al. Characterization
and chromosomal localization of a cDNA encoding brain amyloid of Alzheimer’s disease. Science 1987;235:877-879
5. Robakis NK, Wimiewski HM, Jenkins EC, et al. Chromosome
2 lq21 sublocaliza.tion of gene encoding beta-amyloid peptide in
cerebral vessels .md neuritic (senile) plaques of people with
Alzheimer’s disease and Down’s syndrome. Lancet 1987;1:384385
6. Kang J, Lemair 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
7. Brion JP, van den Bosch de Aguilar Ph, Flament-Duranci J.
Senile dementia ‘Df the Alzheimer type: morphological and inimunocytochemical studies. In: Ttaber J, Gispens WH, eds. Advances in Applied Neurological Science: Senile Dementia of the
Alzheimer Type. Berlin: Springer, 1985;164-174
8. Nukina N, Ihara Y . One of the antigenic determinants of paired
helical filaments is related to tau protein. J Biochem 1986;
9. Kosik KS, Joachim CL, Selkoe DJ. Microtubule-associated ptotein tau is a major antigenic component of paired helical
642 Annals of Neurology Vol 22 No 5 November 1987
filaments in Alzheimer’s disease. Proc Natl Acad Sci USA
10. Wood JG, Mirra SS, Pollock NJ, Binder LI. Neurofibrillary
tangles of Alzheimer’s disease share antigenic determinants with
the axonal microtubule-associated protein tau. Proc Natl Acad
Sci USA 1986;83:4040-4043
11. Grundke-Iqbal I, Iqbal K, Quinlan M, et al. Microtubuleassociated protein tau: a component of Alzheimer paired helical
filaments. J Biol Chem 1986;26 1:6084-6089
12. Yen S-H, Dickson DW, Crowe A, et al. Alzheimer’s
neurofibrillary tangles contain unique epitopes and epitopes in
common with the heat-stable microtubule-associated proteins,
tau and MAP2. Am J Pathol 1987;12681-91
13. Drubin DG, Kirschner MW. Tau protein function in living cells.
J Cell Biol 1986;103:2739-2746
14. Matus A, Bernhardt R, Hugh-Jones T. High molecular weight
microtubule-associated proteins are preferentially associated
with dendritic microtubules in brain. Proc Natl Acad Sci USA
15. Caceres A, Banker G, Steward 0, et al. MAP2 is localized to
the dendrites of hippocampal neurons which develop in culture.
Neuroscience 1984;4:394-410
16. DeCamilli P, Miller P, Navone F, et al. Distribution of MAP2
in the nemous system of the rat studied by immunofluorescence. Neuroscience 1984;11:819-846
17. Huber G, Matus AJ. Differences in the cellular distribution of
two microtubule-associated proteins, MAP1 and MAP2, in rat
brain. Neuroscience 1984;4:151-160
18. Escobar MI, Pumienra H, Caviness VS Jr, et al. Architecture of
apical dendrites in the murine neocortex: dual apical dendritic
systems. Neuroscience 1986;17:975-989
19. Peng I, Binder LI, Black MM. Biochemical and immunological
analysis of cytoskeletal domains of neurons. J Cell Biol
20. Binder LI,Frankfuner A, Rebhun LI. The distribution of tau in
the mammalian central nervous system. J Cell Biol 1985;
21. Tytell M, Brady ST, Lasek RJ. Axonal transport of a subclass of
tau proteins. Evidence for the regional differentiation of microtubules in neurons. Proc Natl Acad Sci USA 1984;8l :15701574
22. Crandd JE, Jacobsen M, Kosik KS. Ontogenesis of microtubule-associated protein 2 (MAP2) in embryonic mouse
cortex. Dev Brain Res 1986;28:127-133
23. Mclean IW, Nakane PK. Periodate-lysine-paraformaldehyde
fixative for immunoelectron microscopy. J Histochem Cytochem 1974;22:1077-1083
24. Kosik KS, Nukina N, Ihara Y, Selkoe DJ. Tau specific monoclonal and polyclonal antibodies suggest that tau is a major anti-
3 1.
genic component of Alzheimer neurofibrillary tangle. J Neuropachol Exp Neurol 1986;45:378
Kosik KS, Rogers J, Kowall N W . Senile plaques are located
between apical dendritic clusters. J Neuropathol Exp Neurol
Vogt 0. Die myeloarchitektonische Felderung des menschlichen Stirnhirns. J Psycho1 Neurol 1910;15:221-232
Tomlinson BE, Corsellis JAN. Ageing and the demenrias. In:
Adams JH, Corsellis JAN, Duchen LW, eds. Greenfield’s
Neuropathology. London: Arnold, 1984;95 1- 1025
Tomlinson BE, Blessed G, Roth M. Observations on the brains
of demented old people. J Neurol Sci 1970;11:205-242
Braak H, Braak E, Grundke-Iqbal I, Iqbal K. Occurrence of
neuropil threads in the senile human brain and in Alzheimer’s
disease: a thud location of paired helical filaments outside of
neurofibrillary tangles and neuritic plaques. Neurosci Lett
1986;65:35 1-355
Probst A, Basler V, Bron B, Ulrich J. Neuritic plaques in senile
dementia of the Alzheimer type: a Golgi analysis of the hippocampal region. Brain Res 1983;268:249-254
Schulman HJ. Phosphorylation of microtubule-associated proteins by a Cdcalmodulin-dependent protein kinase. Cell Biol
Lindwall G, Cole RD. Phosphorylation affects the ability of tau
proteins to promote microtubule assembly. J Biol Chem 1984;
259:5 301-5 305
Grundke-lqbal I, Iqbal K, Tung Y-C, et al. Abnormal phosphorylation of the microtubule-associated protein, tau, in Alzheimer cytoskeletal pathology. Proc Natl Acad Sci USA 1986;
83:49 13-491 7
Ihara Y,Nukina N, Miura R, Ogawara M. Phosphorylated tau
protein is integrated into paired helical filaments in Alzheimer’s
disease. J Biochem 1986;99:1807-1810
Schapp BJ, ‘Vale RD, Sheetz MP, Reese TS. Single microtubules from squid axoplasm support bidirectional movement of organelles. Cell 1985;40:455-462
Rossor MN, Garrett NJ, Johnson AI. A postmortem study of
the cholinergic and GABA systems in senile dementia. Brain
Davies P, Kattman R, Terry RD. Reduced somatostatin-like
immunoreactivity in cerebral cortex from cases of Alzheimer’s
disease and Alzheimer senile dementia. Nature 1980;288:279280
Bissette G, Reynolds GP, Kilts CD, et al. Corticotrophin releasing factor-like immunoreactivity in senile dementia of the Alzheimer type. JAMA 1985;254:3067-3069
Crystal HA, Davies P. Cortical substance P-like immunoreactivity in cases of Alzheimer’s disease and senile dementia of the
Alzheimer type. J Neurochem 1982;38:1781-1784
Kowall and Kosik: Tau Histochemistry in AD
Без категории
Размер файла
1 814 Кб
aberrant, neuropil, tau, disruption, pathologic, characterized, axonal, protein, localization, disease, alzheimers
Пожаловаться на содержимое документа